Ada 088441

Y USARTL-TR-7- 22 / y
AIRCRAFT CRASH SURVIVAL DESIGN GUIDE I

VOLUME IV - AIRCRAFT SEATS, RESTRAINTS,
LITTERS, AND PADDING
F' f=.•:.- AIMULA INC. 48TH STREET

2223 SOUTH
TEMPE, ARIZONA 85282
FINAL REPORT
00AU
$r , I
'I2 1
PREPARED FOR
APPLIED TECHNOLOGY LABORATORY
U. S. ARMY RESEARCH AND TECHNOLOGY LABORATORIES (AVRADCOM)
~
FORT EUSTIS, VIRGINIA 23604
I0
Approved for public release; distribution unlimited.
808 21 036
DISCLAIMER NOTICE
THIS DOCUMENT IS BEST QUALITY

PRACTICABLE. THE COPY FURNISHED
TO DTIC CONTAINED A SIGNIFICANT
NUMBER OF PAGES WHICH DO NOT
REPRODUCE LEGIBLY.
PPLIED TECHNOLOGY LABORATORY POSITION STATEMENT . -
This revised edition of the Crash Survival Design Guide was

prepared for the Applied Technology Laboratory by Simula, Inc.,
under the terms of Contract DAAJ02-77-C-0021. The original
Crash Survival Design Guide was published in 1967 as USAAVLABS
Technical Report 67-22 and subsequent revisions were published
as USAAVLABS Technical Report 70-22 and USAAMRDL Technical Re-
port 71-22. This current edition consists of a consolidation
of design criteria, concepts, and analytical techniques devel-,.,
oped through research programs sponsored by this Laboratory
over the past 20 years into one report suitable for use as a
designer's guide by aircraft design engineers and other inter-
ested personnel.
This document has been coordinated with USAAVRADCOM, the U. S.

Army Safety Center, the U. S. Army Aeromedical Research Labor-
atory, and several other Government agencies active in aircraft
crashworthiness research and development.
The technical monitors for this program were Messrs. G. T.

Singley III, R. F. Bywaters, W. J. Nolan, and H. W. Holland of
the Safety and Survivability Technical Area, Aeronautical Sys-
tems Division, Applied Technology Laboratory.
Comments or suggestions pertaining to this Design Guide will

be welcomed by this Laboratory.
DISCLAIMERS
The findings in this report are not to be construed as an otffc,al Depter.ment of the Army position unless so
designated by other authorized documents.
VMen Government drawings, specifications, or other data are used for any purpose other titan in connection
with a definitely related Government procurement operation, the United States Government thereby incurs no
responsibility nor any obligation whatsoever; and the fact that the Government may have formulated, furnished,
or in any way supplied the said drawings, specificetions, or other data is not to be regarded by implication of
otherwise as in any manner licensing the holder or any other person or corporation, or conveying any riglhts
perrmision, to manufacture, usll, or sell any patented invention that may in any way be related tnereto.
Trade nomes cited in this repors do not constitute on official endorsement or apoloyal of the use of such
commercial hardware or software.
DISPOSITION INSTRUCTIONS
Destroy this report when no longer needed. Do not return It to the originator.
NOTE: This is a revised edition of the Crash Survival Desil 'S

Guide (formerly USAAMRDL Technical Report 71-22). A•,
previous editions are obsolete and should be destroye. -
.. . ....-. ,-.- .n ,..L.,.,,s.•.•diI1•4 -

SCCURITY CLASSIFICATION OF 1HIS PAGE (Whon Data Eftlm.e0
~'EPORT DOCMENTATION PAGE BEFORE COMPLETING FORM
!Ufj 40 4
E.SMED
-*C
e
/0 S.. P.?ISTO SjATEMEnT ofgl.Rf-
Appove
fr pbli rleae;n
disribtio unlimited.021
P. SJPERFO
MING TANýAy O NAEADADRS 0 PORMELMNPOJC.TS
Vomlum Incfafievl.erpr
C22aS.worthiStrets 2 9eA
Sy6etri
Energy,Absrptiona82204
ofcrahwoortyLaotry
Appled U. S. ajr~~f
Army uCntane heei ae o
onlyrc an TolectinolofyvaLaboratioriaeio- a"nd aapriett
(airAcraft FrsortEusinsbu surggeste desig4 codtin7nd1i
MOuIO
D14ig AtheE
deNcoverY oRIS
f illmtfowiCng trlnOpfics):.SCRTYCAS * tl oof
DO~~ ~ ~ 1JN517 ~
DPlNVPIVIIOSE6CCLASSIFIEDIWOWGROn ~ I*
W STATEMENTIO(of thISPAopoff)at ECURITSYIO Rtee
Approved ~ ~
for pulcrees;/itiutonlmtd
611CUt1TY CLAWaFICATIONM or THIS PAgg(Ubw Oat* bdh,,
20. (Continued)
" Volume I - Design Criteria and Checklists*
Volume II - Aircraft Crash Environment and Human Tolerance-'

Volume III- Aircraft Structural Crashworthiness8
Volume IV - Aircraft Seats, Restraints, Litters, and Padding,
Volume V - Aircraft Postcrash' Survival.
This volume (Volume IV) contains information on aircraft seats,
litters, personnel restraint systems, and hazards in the occu-
pant's immediate environment. Requirements for design of seats,
litters, and restraint systems are discussed, as well as design
principles for meeting these requirements and testing for verifi-
cation that the systems perform as desired. Energy-absorbing de-
vices for use in seats are described, as are various types of
cushions. Delethalization of cockpit and cabin interiors is dis-
cussed, including the use of protective padding and the design of
controls for prevention of injury. Finally, computerized methods
of analysis for evaluation of seats, restraints, and the occupant's
immediate environment are presented.
Acesion For
MC TAB
Un~mounced
r)tI
UNCLASSIFIED
6ICUSITY CLASSIFICAIION OF ThIS PAOEf(IVh DA1. 5IMOP.)
PREFACE
This report was prepared for the Safety and Survivability Tech-
nical Area of the Applied Technology Laboratory, U. S. Army
Research and Technology Laboratories (AVRADCOM), Fort Eustis,
Virginia, by Simula Inc. under Contract DAAJ02-77-C-0021, ini-
tiated in September 1977. The Department of the Army Project
Number is IL162209AH76. This guide is a revision of USAAMRDL
Technical Report 71-22, Crash Survival Design Guide, published
October 1971.
A major portion of the data contained herein was taken from

U. S. Army-sponsored research in aircraft crashworthiness con-
ducted from 1960 to 1979. Acknowledgment is extended to the
U. S. Air Force, the Federal Aviation Administration, NASA,
and the U. S. Navy for their research in crash survival. Ap-
preciation is extended to the following organizations for pro-
viding accident case histories leading to the establishment of
the impact conditions in aircraft accidents:
9 U. S. Army Safety Center (USASC), Fort Rucker, Ala-
bama.
D. C.
"
e Civil Aeronautics Board, Washington,
S. Naval Safety Center, Norfolk,
U. Virginia.
Norton
* U. S. Air Force Inspection and Safety Center,
Air Force Base, California.
Additional credit is due the many authors, individual compa-

nies, and organizations listed in the bibliographies for their
contributions to the field. The contributions of the follow-
ing authors to previous ediions of the Crash Survival Design
Guide are most noteworthy:
D. F. Carroll, R. L. Cook, S. P. Desjardins, J. K. Drum-
mond, J. L. Haley, Jr., A. D. Harper, H. G. C. Henneberger,
N. B. Johnson, G. Kourouklis, W. H. Reed, S. H. Robertson,
"L. M. Shaw, Dr. J. W. Turnbow, and L. W. T. Weinberg.
This volume has been prepared by S. P. Desjardins and Dr. D. H.
Laananen of Simula Inc. M. J. Reilly of the Boeing Vertol Com-
pany provided recommendations for sections dealing with troop
and gunner seats, and P. A. Rakszawski contributed to the chap-
ter on protective padding.
-- ~~- - ----
•-__-
,L - ,, ,I II II !I ' I I
TABLE OF CONTENTS
Section Title Pa.e

PREFACE .
LIST OF ILLUSTRATIONS ................

LIST OF TABLES . .. .. .. .. .. .. .. .. . .. 15
INTRODUCTION. . . . . . . . . . . . . . . . . . . ... 17
CHAPTER 1. BACKGROUND DISCUSSION. .......... . 20
CHAPTER 2. DEFINITIONS. . .......... . . 25
2.1 AIRCRAFT COORDINATE SYSTEMS AND

ATTITUDE PARAMETERS. ........... 25
2.2 ACCELERATION-RELATED TERMS . ...... 26
2.3 VELOCITY-RELATED TERMS . ........ .. 26
2.4 FORCE TERMS. . . . ....... .. .. 27
2.5 DYNAMICS TERMS ............. . . 29
2.6 CRASH SURVIVABILITY TERMS ............. 30
2.7 OCCUPANT-RELATED TERMS ......... 30
2.8 SEATING GEOMETRY . . . . . ...... 33
S2.9 STRUCTURAL TERMS ......... .... 35
t
CHAPTER 3. PRIMARY DESIGN CONSIDERATIONS ...... 37
3.1 INTRODUCTION . . . ....... ... 37

3.2 OPERATIONAL ENVIRONMENT ........ 37
3.2.1 Comfort ..... ............ .... 37
3.2.2 Seat Adjustments ........ ... 38
3.2.3 Vibration Damping.. ............... 39
3.3 CRASH ENVIRONMENT. ......... . 40
3.3.1 Dynamics and Kinematics:. ... 40
3.3.2 Design Conditions and Envelopes . . . . 44
3.3.3 Structural Distortion. ............. 45
3.4 APPLICABILITY OF CRITERIA ........ 46
3.5 ACCEPTANCE CRITERIA ............. .. 47
3.6- SELECTION CRITERIA . . .... 47
CHAPTER 4. DESIGN PRINCIPLES FOR SEATS AND

LITTERS...... . . . . . . . . . . . . 48
4.1 INTRODUCTION . . . . . ....... 48

4.2 LITTERS AND THEIR ORIENTATION . .... 49
4.3 MATERIALS ........... ......... . 49
4.4 STRUCTURAL CONNECTIONS ...... ......... 50
4.4.1 Bolted Connections. . . . ...... 50
||5
TABLE OF CONTENTS (CONTD)
Section Title Page
4.4.2 Riveted Connections . . . . . 51

4.4.3 Welded Connections . . . . . . . . .*. . 51
4.4.4 Seat Attachment...... . . . ... . 51
4.4.5 Joint Deformation. . . . . . . . . .. . 52
4.5 STRENGTH ..................... .. 62
4.5.1 General. . ............. 62
4.5.2 Limit Analysi Concepts. ....... 65
4.5.3 Large Deformation Analysis . . . . ... 66 "
4.5.4 Strain Concentrations ............ ... 66
4.6 RESTRAINT SYSTEM ANCHORAGE . . . . ... 67
4.7 CRASH ENERGY ABSORPTION. . . . . . ... 67
4.7.1 General. . ...... . .... . . . 67
4.7.2 Principle of Energy Absorption -
Illustration .... ............. . 70
4.7.3 Dynamic Response. .*... .. 82
4.8 COMPUTERIZED METHODS OF ANALYSIS . ... 93
4.8.1 General. . . . . . . . ........... 93
4.8.2 Program SOM-LA . ... ........ 95
4.8.3 Calspan Corporation -'CVS .. . ... 99
4.8.4 PROMETHEUS .... ............. 100
4.8.5 Air Force Head-Spine Model . . . . . . . 102
4.8.6 One-Dimensional Seat/Occupant Models . 103
CHAPTER 5. ENERGY-ABSORBING DEVICES . . . . . . . . 106
5.1 INTRODUCTION .............. ... 106

5.2 WIRE OR STRAP BENDING ............ . .. 109
5.3 INVERSION TUBE ..... ......... . . . 112
5.4 ROLLING TORUS. . . . . . . . . . . ... 114
5.5 CRUSHING HONEYCOMB ........... .. 118
5.6 EXTENSION OF BASXC METAL TUBE OR ---
FLAT STRAP ....... ............ .. 119
5.7 ELONGATION OF BASIC STRANDED CABLE . . . 121
5.8 ROD PULL-THROUGH TUBE. . . . . . . ... 121
5.9 TUBE FLARING . . ............ . 121
5.10 TENSION PULLEY ......... ............. 123
5.11 SELECTION OF AN OPTIMUM LOAD LIMITER . . 123
CHAPTER 6. SEAT CUSHIONS .... .............. ... 125
6.1 INTRODUCTION .... ......... ... 125

6.2 REQUIREMENTS . . . .......... 126
6.3 ENERGY-ABSORBING CUSHIONS. ........ 127
6.4 NET-TYPE CUSHIONS ............... ... 128
6.5 OTHER CUSHIONS ................ .. 128
6.6 HEADRESTS ...... ............. ... 128
S. . . . . . ... - .. w .. ..-
'9-- -,;• ... . --.- '•' • •" q- ,, • -_

Section Title
CHAPTER 7. DESIGN PRINCIPLES FOR PERSONNEL

RESTRAINT SYSTEMS. . . ...... ........... 129
7.1 INTRODUCTION .... ........ . . ... .. 129

7.2 TYPES OF SYSTEMS . . . . . . . ..... 130
7.2.1 Aircrew Systems ..... ............ ... 130
7.2.2 Troop Systems. ......... .... 133
7.2.3 Crew Chief and Door/Window Gunner
Systems.. .. ..... . ...... 134
7.2.4 Inflatable Systems ...... .137
7.3 GENERAL DESIGN CRITERIA...... ........ 137
7.3.1 Comfort ................. ....... 137
7.3.2 Emergency Release Requirements .. 139
7.3.3 Lap Belt Anchorage . . ........ 140
7.3.4 Shoulder Harness Anchorage ..... 144
7.3.5 Lap Belt Tiedown Strap Anchorage 145
7.3.6 Adjustment Hardware ........ ........ 146
7.3.7 Location of Adjustment and Release
Hardware . . . . . . . . . . 146
7.3.8 Webbing Width and Thickness
Requirements ..... ............. ... 146
"7.3.9 Hardware Materials. . . . . . ..... 146
7.3.10 Structural Connections . . . . ..... 147
7.4 WEBBING AND ATTACHMENTS ........... ... 148
7.4.1 Properties ............. .. ....... 148
7.4.2 -Width and Thickness Requirements .... 151
7.4.3 Webbing Attachment Methods ....... ... 151
7.4.4 Energy-Absorbing Webbing .......... .. 159
7.5 RESTRAINT SYSTEM HARDWARE .......... .... lo
7.5.1 General .................. .... 160
7.5.2 Buckles and Emergency Release. ..... 163
7.5.3 Adjustment Hardware. . . ..... .
... 163
7.5.4 Inertia Reels, Control, and
Installation ............. 164
CHAPTER 8. SEAT STRENGTH AND DEFORMATION -

REQUIREMENTS . 166
8.1 INTRODUCTION ............ 166

8.2 RECOMMENDED OCCUPANT WEIGHTS FOR
SEAT DESIGN. ............... . 166
8.2.1 Crewseats. ................. 167
8.2.2 Troop and Gunner SeatB ........... 168
8.3 STRENGTH AND DEFORMATION . . . . .... 168
8.3.1 Forward Loads. . . . . . ........... 168
8.3.2 Use of Design Curves ............ .. 171
4
7
Section Title•e Page

8.3.3 Aftward Loads. . . . . . . ...... . 172
8.3.4 Downward Loads .......... ...... .172
8.3.5 Upward Loads . . . . . . .............. 178

8.3.6 Lateral Loads and Deformation. .... 178
8.4 OTHER SEATS . . ... 180
8.5 PERSONNEL RESTRAINTWHAýRESS'TESTIN. 180
8.6 STRUCTURAL TEST REQUIREMENTS . . . . 180
8.6.1 Static Test Requirements . . . . . . .*. 181
8.6.2 Dynamic Test Requirements ........ ... 188
8.6.3 Data Acquisition and Reduction . . . . . 193
8.6.4 Seat Component Attachment. . . . . . . . 195
CHAPTER 9.
9.1
9.2
LITTER STRENGTH AND DEFORMATION
REQUIREMENTS ....................
INTRODUCTION ...........
RECOMMENDED OCCUPANT WEIGHTS FOR
LITTER DESIGN .. .. .. ................
.. ....
196
196
19
A
9.3 VERTICAL LOADS ...... .... . . . . . 197
9.3.1 Downward Loads ..... .............. . 197
9.3.2 Upward Loads .... ............. 199
9.4 LATERAL AND LONGITUDINAL LOADS .... 199
9.5 LITTER RESTRAINT HARNESS TESTING . . . . 200
9.6 LITTER SYSTEM TEST REQUIREMENTS. . . .. 200
9.6.1 Static Test Requirements.. ........ ... 200
9.6.2 Litter System Dynamic Test
Requirements ........... . . . . 202
CHAPTER 10. DELETHALIZATION OF, COCKPIT AND CABIN

INTERIORS ..... ................ .... 203
10.1 INTRODUCTION ... .............. .... 203

10.2 OCCUPANT STRIKE ENVELOPES ........ ... 203
10.2.1 Full Restraint... ........ 203
10.2.2 Lap Belt-Only Restraint ........ .... 205
10.2.3 Seat Orientation .... ......... .... 206
10.3 ENVIRONMENTAL HAZARDS. .......... .... 206
10.3.1 Primary Hazards .... .............. . 206
10.3.2 Secondary Hazards. ............. , .. . 207
10.3.3 Tertiary Hazards . . . . . . . . . . .. 21C
10.4 HEAD IMPACT HAZARDS .... . ...... .... 21C
10.4.1 Geometry of Probable Head Impact
Surfaces . . . .. . . ..... .......... 21C
10.4.2 Tolerance to Head Impacts. . . . . . 210
Test Procedures ........ .. .. .... 21]
a.
10.4.3
....... )
10.5 INSTRUMENT PANEL STRUCTURE PROXIMITY . 21;
Section Title
10.6 RUDDER PEDAL CONFIGURATION .... . . . 213
10.7 CONTROL COLUMNS. . . .... . 213
S10.9 10.8 SIGHTING AND VISIONIC
ENERGY-ABSORBING SYSTEMS. FOR ....
REQUIREMENTS 214 '
I" 10.9.1 COCKPIT AND CABIN INTERIORS .......
General. .219 219
10.9.2. Types of Padding Ma~e~i;l; ;n• ..
•10.9.3 Properties . . . .. .. . .
. . Methods•
Standard Test ." . . . . . 222
227"
0.9.14 Research on Materials for Energy-
Absorbing Applications . ....... .. 230
10.9.5 Application of Padding Material. . . . . 243
10.9.6 Ductile Materials . .......... 244
REFERENCES .. ................. .. 246
BIBLIOGRAPHY. . . ......... ........... 259

INDEX . . . . . . . . . . . .............. 264
LIST OF ILLUSTRATIONS
Figure
1 Aircraft coordinates and attitude

directions . . . ............... 25
2 Typical aircraft floor acceleration

pulse... . . . . . . . . ....... 28
3 Terminology for directions of forces
on the body ................... 31
4 Seating geometry ... .............. 34
S5 Sketch illustrating buckling or

"dishing" formation. . . ......... 53
"6 Static test floor warpage requirement

to ensure seat retention in crashes. .. 55
7 Concepts for release of floor-
distortion-induced moments . . . . . .. 56
._-,9
LIS O ILLUSTRTIONSJCONTDI
8 Aft seat leg casting attachment. . . . . 57

9 Aft seat leg casting attachment
modification . .*. . . . . . . . . . . . 57
30 Torsional release of joints. . . . . . . 58
11 Fully released joint . . . . . . . . .. 59
12 Bulkhead in-plane warping. . . . . . . . 60
13 Universal release of a joint . . . . . . 61
14 Pin joint releases oriented to allow

rotation around an aircraft roll 62
axis . .... . . .. . . . . . . . . .
15 Comparison of analysis methods for
simple beams ................. ..... 63
16 Seat leg anchorage to floor track ... 65
17 Deceleration-time, velocity-time, and

distance-time curves used in analysis
of seat/occupant displacement with
respect to the airframe ..... ......... 77
18 Airframe velocity-time curve ...... .. 79
19 Deceleration versus time for various
components of seat and occupant. ..... 84
20 Typical seat pan, dummy chest, and
dummy pelvis response to vertical
crash loading ........... .. . .... 86
21 Initial condition, no load ....... ... 88
22 Onset of deceleration load wherein
pelvic area is responding to decel-
eration load, but the upper torso
and head are not ............... 88
23 High-deceleration load, springs
compressed ........ ............... 88
10
LIST OF ILLUSTRATIONS (CONTD)
*Figure Eno-
24 Duration and magnitude of headward
acceleration endured by various
subjects . .. . . . . . . . . . . . . . 91
25 Twelve-segment occupant model. . . . . . 97
26 Occupant model contact surfaces ..... 98
27 SOM-LA energy-absorbiing seat model. . . 98

28 CVS graphics display model . . . . . . . 100
29 PROMETHEUS occupant model. . . . . . .. 101
30 Three-dimensional head-spine model . . . 103
31 Lumped-parameter model of seat, seat

cushion, and occupant. . . . . . . . . . 105
32 Crashworthy troop seat . . . . . . . . . 110
33 Troop seat tension energy absorber

including characteristics for two
wire diameters ............. 11i
34 Tubular strut wire-bending energy
absorber with force-deflection
curves ..... . . . . . . . . . . 113
35 Inversion tube concept with typical

force-deflection characteristic ..... 115
36 Comparison of dynamic and static load-
deformation curves for inversion
~ . . . . . . . . . . . . . . *... .
~tubes 116
IS
37 - Rolling -torus energy absorber. . . . . . 117
38 Illustration of corrugated aluminum

foil formed into annular column. . . . . 119
elongation curves for stainless steel
tubes . . . . . . . . . . . . . . . . . . 120
LIST OF ILLUSTRATIONS (COND)
Figure page
deformation curves for compression
tubes. . . . . . . . . . . ........... 122
41 Illustration of fragmentation and

rolling processes in tube-flaring
device . . ..... . . ............. . . 123
42 Tension-pulley load limiter. . . . . . . 124
43 Basic aircrew restraint system . . . .. 131
44 Aircrew restraint system, including

reflected shoulder straps. . . . . . .. 132
45 Aircraft troop/passenger restraint
systems. . . . . . . ........... ...... 135
46 Gunner restraint system. .......... 136
47 Inflatable body and head restraint . . 138
48 Buckle fitting attachment and motion

angles . . . . . . . . . . . . . . . . . 140
49 Pelvic rotation and submarining caused

by high longitudinal forces combined
with moderate vertical forces . .. . . . 142
50 Lap belt anchorage geometry. . . . . . . 143
51 Shoulder harness anchorage geometry. . . 145

52 Load elongation characteristics for
MIL-W-25361 (Type II) polyester
webbing
rates . . for
. . static
. . . and
. . rapid
. . . loading
. . . . . . 15.1,
53 Stitch pattern and cord size . . . . . . 153
54 Stitch patterns tested . . . . .... 155
55 Wrap radius for webbing joints . . . . . 158

56 Webbing fold at metal hardware
attachment ...... . .... .... 159
12
j• LIST OF ILLUSTRATIONS (CONTD)
Figure Eae
57 Aircrew restraint system . . . . . . . . 161
58 Reflected shoulder strap restraint

system . . . . . . . .. . . . 162
59 Seat forward load and deflection

requirements for all types of Army
aircraft (forward design pulse). . . . . 170
60 Lateral seat load and deformation

requirements for all types of Army
aircraft ........ 179
61 Suggested method of applying floor

warping for static testing of seats. . . 184
62 Suggested method of applying bulkhead

warping for static testing of seats. . . 185
63 Static load application point and

critical body block pelvis geometry. . . 186
64 Requirements of dynamic tests if

Ssubstituted for static tests . . . . .*. 189
65 Dynamic test requirements for
qualification and for research/
development testing. . . . . . . . . . . 190
66 Example of input pulse for seats having

less than 12 in. of stroke . . . . . . . 192
67 Graphic approximation example. . . . . . 194

68 Litter downward load and deflection
requirements . 198
69 Litter forward or lateral load and
deflection requirements for all
types of Army aircraft . . .. .. . . . 199
70 Full-restraint extremity strike

envelope - side view . . . . ..... 204
71 Full-restraint extremity strike

envelope -top view ..... ..... 205
S.. 13
LIST OF ILLUSTRATIONS (CONTD)
Figure
72 Pull-restraint extremity strike

envelope - front view. . . . .. .# . o 206
73 Lap belt-only extremity strike

envelope - side view. . . . . . . ... 207

envelope - top view ........... 208

envelope - front view. . . . . . . . . . 209
76 Wayne State Tolerance Curve for the

human brain in forehead impacts against
plane, unyielding surfaces . . . . . . . 211
77 Measured head velocities in sled tests

with anthropomorphic dummies and
cadavers ................ 212
78 Antitorque, or rudder, pedal geometry
to prevent entrapment of feet. . . . . . 214
79 Pilot/gunner station occupant strike
in-board profile 4-G impact. . . . . . . 215
so Crushable eyepiece concept ....... 218
81 Impact behavior (headform deceleration

versus speed) of three padding
materials. . . . . . . . . . . . . . . . 220
82 Impact behavior (average dynamic _

stress versus speed) of three padding
materials. .
83 Minimum tensile strength versus product

density for polyethylene sheet and
plank . .. . . . . . . . . .0 a 0 * , 223
84 Recommended stress-strain properties

for padding material for head contact,
with cushion thickness up to 1.5 in. . . 230
85 Stress-strain curves for polyethylene
foam • 232
14
LIST OF ZLLUSTA TZONS ICONTD)
86 Effect of density on stress-strain

curves for polyurethane-foamed
plastic. . . . . . . . . . . . . . . . 233
87 Evaluation criteria for energy-
absorbing material .
234
"88 Load-distributing material evaluation
criteria . . . . . . . . . . . . . . . . 236
89 Response of composite foam, compared
with urethane, and polystyrene . . . . . 238
90 Typical response of plastic
compression test ............ foam to
239
91 Specific energy absorbed to
maximum
strain versus crosshead velocity
. .. . 242
92 Dynamic crushing pressure versus
impact velocity for tests at
temperatures .............. two
244
LIST OF TABLES
Table
I Crashworthiness criteria for
preliminary design process the
. . . . . . 24
2 Summary of design conditions
for
rotary- and light fixed-wing
aircraft . . . . . . . . . . . . . . . . 44
3 Comparison rof load-limiting
devices for
lO00-to-4000Olbloads. . . .
. . . . . . 108
4 Restraint harness components
load-
elongation design and test
ments. . . . . . . . . require-
. . . . . . . . .
149
5 Restraint webbing characteristics
. . . 150
6 Minimum webbing width requirements
. . . 152
7.4
*U
LIST OF TABLES (CONTD) .
Table e
7 Breaking strength of stitch patterns
(test series one) . . . . . . . . . .. 156
8 Breaking strength of stitch patterns
(test series two). . . . . . . . . . . . 157
9 Typical aviator weights .... ........ 167 *
10 Troop and gunner weights . . . .... 169
11 Seat design and static test
requirements .................. .... 182
12 Litter systems static test
requirements .............. 201
13 Potential optical relay tube crash
hazards ................ . . . . . . 216 I
14 Energy-absorbing plastic foams and
some typical applications .......... ... 224
15 Properties of selected flexible
cellular polymers ............ . . . . 226
16 Summary of ASTM test methods and

specifications for flexible cellular
plastics . . . . . . . . . . . . . . . . 228
I
17 Material summary ... ............ .... 240
18 Test results summary .......... 241
16 *
p INTRODUCTION
For many years, emphasis in aircraft accident investigation
was placed on determining the cause of the accident. Very
little effort was expended on the crash survival aspects of
aviation safety. However, it became apparent through detailed
studies of accident investigation reports that significant im-
provements in crash survival could be made if consideration
were given in the initial aircraft design to the following fac-
tors that influence survivability:
S•1. Crashworthiness of Aircraft Structure - The ability
of the aircraft structure to maintain living space
for occupants throughout a crash.
2. Tiedown Chain Strength - The strength of the linkage
preventing occupant, cargo, or equipment from break-
ing free and becoming missiles during a crash se-
quence.
Z
3. Occupant Acceleration Environment - The intensity

and duration of accelerations experienced by occu-
pants (with tiedown assumed intact) during a crash.
4. Occupant Environment Hazards - Barriers, projections,
and loose equipment in the immediate vicinity of the
occupant that may cause contact injuries.
5. Postcrash Hazards - The threat to occupant survival
posed by fire, drowning, exposure, etc., following
the impact sequence.
Early in 1960, the U. S. Army Transportation Research Command*
initiated a long-range program to study all aspects of aircraft
safety and survivability. Through a series of contracts with
the Aviation Safety Engineering and Research Division (AvSER)
of the Flight Safety Foundation, the problems associated with
occupant survival in aircraft crashes were studied to deter-
mine specific relationships among crash forces, structural
failures, crash fires, and injuries. A series of reports coy-
ering this effort was prepared and distributed by the U. S.
Army, beginning in 1960. In October 1965, a special project
initiated by the U. S. Army consolidated the design criteria
presented in these reports into one technical document suit-
able for use as a designer's guide by aircraft design engineers
'Now the Applied Technology Laboratory, Research and Technology

Laboratories of tht. U. S. Army Aviation Research and Develop-
ment Command (AVRADCOM).
17
and other interested personnel. The document was to be a sum-
mary of the current state of the art in crash survival design,
using not only data generated under Army contracts, but also
information collected from other agencies and organizations.
The Crash Survival Design Guide, first published in 1967, real-
ized this goal.
Since its initial publication, the Design Guide has been re-
vised several times to incorporate the results of continuing
research in crashworthiness technology. The last revision,
published in 1971, was the basis for the criteria contained in
the Army's military standard dealing with aircraft craahworthi-
ness, MIL-STD-1290(AV), "Light Fixed- and Rotary-Wing Aircraft
Crashworthiness" (Reference 1). This current revision, the
"fourth, contains the most comprehensive treatment of all as-
pects of aircraft crash survival now documented. It can be
used as a general text to establish a basic understanding of
the crash environment and the techniques that can be employed
to improve chances for survival. It also contains design cri-
teria and checklists on many aspects of crash survival and thus
can be used as a source of design requirements.
The current edition of the Crash Survival Design Guide is pub-
lished in five volumes. Volume titles and general subjects
included in each volume are as follows:
Volume I - Design Criteria and Checklists

i Pertinent criteria extracted from Volumes II through V,
presented in the same order in which they appear in those
volumes.
Volume II - Aircraft Crash Environment and Human Tolerance
Crash environment, human tolerance to impact, military

anthropometric data, occupant environment, test dummies.
Volume III - Aircraft Structural Crashworthiness
Crash load estimation, structural response, fuselage and

landing gear requirements, rotor-- requirements, ancillary
equipment, cargo restraints, structural modeling.
1. Military Standard, MIL-STD-1290(AV), LIGHT FIXED- AND

ROTARY-WING AIRCRAFT CRASHWORTHINESS( Department of De-
fense, Washington, D. C., 25 January 1974.
18
Volume IV - Aircraft Seats, Restraints, Litters, and Padding
Operational and crash environment, energy absorption, seat
design, litter requirements, restraint system design,
occupant/restraint system/seat modeling, delethalization
of cockpit and cabin interiors.
Volume V - Aircraft Postcrash Survival
Postcrash fire, ditching, emergency escape, crash locator
beacons, retrieval of accident information.
This volume (Volume IV) contains information on aircraft seats,
litters, personnel restraint systems, and hazards in the occu-
pant's immediate environment. Following a general discussion
of aircraft crashworthiness in Chapter 1, a number of terms
commonly used in discussing the crash environment, seats, and
occupant protection are defined in Chapter 2. Chapter 3 pre-
sents design considerations for aircraft seats, and Chapter 4,
principles for crashworthy seac design. Energy absorption is
discussed in Chapter 5. Principles for cushion and restraint
system design are presented in Chapters 6 and 7, and strength
and deformation requirements for seats and litters are stated
in Chapters 8 and 9, respectively. Cockpit delethalization,
including protective padding, is discussed in Chapter 10.
19
19
1. BACKGROUND DISCUSSION
The overall objective of designing for crashworthiness is to

eliminate unnecessary injuries and fatalities in relatively
mild impacts. A crashworthy aircraft alfo reduces aircraft
crash impact damage. By minimizing personnel and material
losses due to crash impact, crashworthiness conserves re-
sources, is a positive morale factor, and improves the combat
effectiveness of the fleet. Results from analyses and research
during the past several years have shown that the relatively
small cost in dollars and weight of including crashworthy fea-
tures is *a wise investment (References 2 through 13). Conse-
quently, new generation aircraft are being procured to strin-
gent, yet practical requirements for crashworthiness.
To provide as much occupant protection as possible, a systems
approach to crashworthiness must be followed. Every available
subsystem must be considered in order to maximize the protec-.
tion afforded to vehicle occupants. When an aircraft impacts
2. ENGINEERING ANALYSIS OF CRASH INJURY IN ARMY OH-58 AIR-

CRAFT, USASC Technical Report, U. S. Army Safety Center,
Fort Rucker, Alabama, to be published.
3. ENGINEERING ANALYSIS OF CRASH INJURY IN ARMY CH-47 AIR-

CRAFT, USAAAVS Technical Report 78-4, U. S. Army Agency
for Aviation Safety, Fort Rucker, Alabama, June 1978.-
4. ENGINEERING ANALYSIS OF CRASH INJURY IN ARMY AH-l AIR-

for Aviation Safety, Fort Rucker, Alabama, March 1978.
5. Carnell, B. L., CRASHWORTHINESS DESIGN FEATURES FOR AD-

VANCED UTILITY HELICOPTERS, in Aircraft Crashworthiness,
K. Saczalski, et al., eds., University Press of Virginia,
Charlottesville, Virginia, 1975, pp. 51-64.
6. Bainbridge, M. E., Reilly, M. J., and Gonsalves, J. E.,

CRASHWORTHINESS OF THE BOEING VERTOL UTTAS, in Aircraft
Crashworthiness, K. Saczalski, et al., eds., University
Press of Virginia, Charlottesville, Virginia, 1975,
pp. 65-82.
7. Rich, M. J., INVESTIGATION OF ADVANCED HELICOPTER STRUC-

TURAL DESIGNS, Volume I, ADVANCED STRUCTURAL COMPONENT DE-
SIGN CONCEPT STUDY, Sikorsky Aircraft, Division of United
Technology Corporation; USAAMRDL Technical Report 75-59A,
Eustis Directorate, U. S. Army Air Mobility Research and
Development Laboratory, Fort Eustis, Virginia, May 1976,
AD A026246.
20
4-• r l . . i i" .. f ' i "•" -- -"• "•FP -"

the ground, deformation of the ground absorbs some energy.
"This is an uncontrolled variable since the quality of the im-
pacted surface usually cannct be selected by the pilot. If
the aircraft lands on an appropriate surface in an appropriate
attitude, the landing gear can be used to absorb a significant
amount of the impact energy. After stroking of the gear, crush-
ing of the fuselage contributes to the total energy-absorption
process. The fuselage must also maintain a protective shell
around the occupant, so the crushing must take place outside
the protective shell. The functions of the seat and restraint
system are to restrain the occupant within the protective shell
during the crash sequence and to provide additional energy-
absorbing stroke to further reduce occupant decelerative load-
ing to within human tolerance limits. The structure and com-
ponents immediately surrounding the occupant must also be
considered. Weapon sights, cyclic controls, glare shields,
instrument panels, armor panels, and aircraft structure must
be delethalized if they lie within the strike envelope of the
occupant.
8. Hoffstedt, D. J., and Swatton, S., ADVANCED HELICOPTER
STRUCTURAL DESIGN INVESTIGATION, The Boeing Vertol Com-
pany; USAAMRDL Technical Report 75-56A, Eustis Directorate,
U. S. Army Air Mobility Research and Development Labora-
tory, Fort Eustis, Virginia, March 1976, AD A024662.
9. Hicks, J. E., AN ANALYSIS OF LIFECYCLE ACCIDENT COSTS FOR
THE ADVANCED SCOUT HELICOPTER, U. S. Army Agency for Avia-
tion Safety, Fort Rucker, Alabama, January 1977.
10. McDermott, J. M., and Vega, E., THE EFFECTS OF LATEST
MILITARY CRITERIA ON THE STRUCTURAL WEIGHT OF THE HUGHES
ADVANCED ATTACK HELICOPTER YAH-64, Journal of the American
Helicopter Society, Vol. 23, No. 4, October 1978, pp. 2-9.
11. Haley, J. L., Jr., CRASHWORTHINESS VERSUS COST: A STUDY
OF ARMY ROTARY WING AIRCRAFT ACCIDENTS IN PERIOD JANUARY
1970 THROUGH DECEMBER 1971, paper presented at the Air-
craft Crashworthiness Symposium, University of Cincinnati,
Cincinnati, Ohio, October 1975.
12. Hicks, J. E., ECONOMIC BENEFITS OF UTILITY AIRCRAFT CRASH-
WORTHINESS, USAAAVS Technical Report 76-2, U. S. Army
Agency for Aviation Safety, Fort Rucker, Alabama, July
1976.
13. THE ECONOMIC BENEFITS OF CRASHWORTHINESS AND FLIGHT SAFETY
DESIGN FEATURES IN ATTACK HELICOPTERS, USAAAVS Technical
Report 77-2, U. S. Army Agency for Aviaticn Safety, Fort
Rucker, Alabama, June 1977.
1 21
Ideally, it would seem most efficient to simply specify human
tolerance requirements and an array of vehicle crash impact
conditions and then develop the helicopter as a crashworthy
system with a mixture of those crashworthy features that are
most efficient for the particular helicopter being designed.
Unfortunately, the validated structural and/or human tolerance
analytical techniques needed to perform and evaluate such a
maximum design freedom approach to achieving crashworthiness
are not available. Furthermore, testing complete aircraft suf-
ficiently early in the development cycle to permit evaluation
of system concepts in time to permit design changes based on
the test results is not practical. The systems approach dic-
tates that the designer consider probable crash conditions
wherein all subsystems cannot perform their desired functions;
for example, an impact situation in which the landing gear can-
I
not absorb its share of the impact crash energy because of air-
craft attitude at impact. Therefore, to achieve the overall
*
goal, minimum levels of crash protection are recommended for
the various individual subsystems.
between the two extremes of:
A balance must be struck
(1) defining necessary perform-
I
ance on a component level only, and (2) requiring that the air-
4
craft system be designed for an array of impact conditions
with no component design and test criteria.
Current helicopter crashworthiness criteria require that a new
aircraft be designed as a system to meet the vehicle impact
design conditions recommended in Volume II; however, minimum
criteria are also specified for a few crash critical components.
For example, strengths and minimum crash energy-absorption re-
quirements for seats and restraint systems are specified. All
strength requirements presented in this volume are based on the
crash environments described in Volume II. Testing require-
ments are based on ensuring compliance with strength and defor-
mation requirements. Mandatory minimum crashworthiness design
criteria for U. S. Army light fixed- and rotary-wing aircraft
are stated in MIL-STD-1290(AV) (Reference I). All pilot, co-
pilot, observer, and student seats in either rotary- or light
fixed-wing aircraft should conf-orm to the re-quirements of
MIL-S-58095(AV) (Reference 14).
Although much higher levels of crashworthiness can be achieved
in completely new aircraft designs, the crashworthiness of
existing aircraft can be significantly improved through retro-
fitting these aircraft with crashworthy components adhering to
the design principles of this design guide. This can even be
1'4. Military Specification, MIL-S-58095(AV), SEAT SYSTEM:
CRASHWORTHY, NON-EJECTION, AIRCREW, GENERAL SPECIFICATION
FOR, Department of Defense, Washington, D. C., 27 August
1971.
22
achieved while expanding the combat effectiveness of the air-
* craft. Examples of this are the successful program to retro-
fit all U. S. Army helicopters with crashworthy fuel systems
(Reference 15), and the U. S. Navy program to retrofit the
CH-46 with crashworthy armored crewseats (Reference 16).
In an initial assessment, the definition of an adequate crash-
worthy structure may appear to be a relatively simple matter.
In fact, many influencing parameters must be considered before
an optimum design can be finalized. A complete systems ap-
proach must be employed to include all influencing parameters
concerned with the design, manufacture, overall performance,
and economic restraint on the aircraft in meeting mission re-
quirements. Trade-offs between the affecting parameters must
be made in order to arrive at a final design that most closely
meets the customer's specified requirements. It must be re-
membered that for each type cf aircraft, different emphasis
will be placed in the parameter mix. Table 1 summarizes major
crashworthiness criteria that must be considered during the
preliminary design definition phase.
15. Cook, R. L., and Goebel, D. E., EVALUATION OF THE UH-lD/H

HELICOPTER CRASHWORTHY FUEL SYSTEM IN A CRASH ENVIRONMENT,
Dynamic Sciencer,. Division of Marshall Industries; USAAMRDL
Technical Report 71-47, U. S. Army Air Mobility Research
and Development Laboratory, Fort Eustis, Virginia, Novem-
ber 1971.
16. Domzalski, L. P., et al., U. S. NAVY DEVELOPMENTS IN CRASH-
WORTHY SEAIING, Naval Air Development Center; Proceedings
1978 SAFE Symposium, Survival and Flight Equipment Asso-
ciation, Canoga Park, California, October 1978.
23
--
to 4) f,-4 = 0)
0 3-4 r4
-040 0 4 1
(0 1U m V wc *-A
(D . 1 0 F A' 4-.i r 00
03(
E4 0 a -1 41 .04 4.1 0-.I
4)
.1. > .a a r 1 4 0 w
U) k U M00 C3 IA H 0(a 4 )I34IA4( 0 J
U) 02 4) 2( *. 1 r4 &0 (A' P.40 0.r Q -440
1: 14 34 -44
L) 43i0p w 0~ (U' 04 k -H
~ 1 4 (04. 0 c 0 9 w 4 u043 z0 a~ 40 41
04 (03 03 020 0 1 33uc

0
F40
a1 C
.
w too ~40V 1
4. (IS ck 0> 24 0 Iv )P
w. 0. adO
0 h U.4
ra to5(43
0) 4) o C 0 0 4.
to * 4 * u IV to
4j 0 0 A V
0 k 9 tl4 0E C 41~ 14
ra 0 ) u 11 h IJr t 41. (0004
w c3 40 3 >4 r P-04 C0034.3 4 e4 414
> -1.4 3 4F-40 A IMWV ' MM4J4 A 0 u4
4J 01 CO 42' 4 00 0434 0 0 A
00 :3- ca w
0
r., 01
w 4 430 03041 A03iO( 040

Eq w t 1.4 0412 Ec~z Q- .
0 4
VC IV . 0 cl ov 0 44 pf
1.)~A 34 0
to 14 04 W 44 1.444 4 -
cn 4 0 ai 03 4 r. ~0 0 c 4
43 4U4) 0 cv wU 03430 u 1
o '4
4U 4J 044 r.3 ~41 U)-
U :K
H 1 10 4 1-4 0 -400 0)4
434
0 3 U44 t0 ) Ad00
U) 4 0op 00 0H u 4 )0
4
U)w U P 4r VCm 0
to4 w. 0 ) w
0 i 0 i v
0 40 t )
0 Qir 04-4 A4
f-I E
040 ) V.0 0 (V Q 0. -P4 LOA 4 (A4
* 00 0 0 0
Ui 43
-
* 0
- 0
6C ~ 44 0"- to H-
00 E
iv 04
1. V4 .. j
C 2 4) .lJQ0
U) 0 O0) 9: H4 F4 U )>0~-4
$4(yea -'4 0 P4i V IV>0 a,
3 0 .404
4 049 X rq to 0 eli 040.4
'C u I F.0 0d4A- U 4J V M 2-40U0
E
U)
a- -4
-,.~4 (a 'r 4 4u k. *d.0-4 41 4f
P4 4 030 43 U0)d 03 41 0r4 0
. 1 F)434r. H 4)h 00 43 0404
W) I.44) r. V 0 02 EU
u v -V4U 42
to U * *
24
4-
S2. DEFINITIONS
2.1 AIRCRAFT COORDINATE SYSTEMS AND ATTITUDE PARAMETERS
Positive directions for velocity, acceleration, and force com-

ponents and for pitch, roll, and yaw are illustrated in Fig-
-: ure 1. When referring to an aircraft in any flight attitude,
it is standard practice to use a basic set of orthogonal axes
as shown in Figure 1, with x, y, and z referring to the longi-
tudinal, lateral, and vertical directions, respectively.
+z
Yaw
Pitch •
Roll
Figure 1. Aircraft coordinates and attitude directions.
25
• -. •. . .. . . ... • "• -1j

2.2 ACCELERATION-RELATED TERMS
* Acceleration
The rate of change of velocity. An acceleration is
in magnitude or in direction. Acceleration may pro-
duce either an increase or a decrease in velocity.I

There are two basic types of acceleration: linear,
which changes translational velocity, and angular (or
rotational), which changes angular (or rotational)
velocity. With respect to the crash environment, un-
less otherwise specified, all acceleration values are
those at a point approximately at the center of the
floor of the fuselage.
* Deceleration
Acceleration which produces a decrease in velocity.
* Abrupt Accelerations
Accelerations of short duration primarily associated
with crash impacts, ejection seat shocks, capsule
impacts, etc. One second is generally accepted as
the dividing point between abrupt and prolonged ac-
celerations. Within the extremely short duration
range of abrupt accelerations (0.2 sec and below),
the effects on the human body are limited to mechan-
ical overloading (skeletal and soft tissue stresses),
there being insufficient time for functional distur-
Sbances due to fluid shifts.
•The Term G
The ratio of a particular acceleration to the accel-

eration due ýo gravitational attraction at sea level
(32.2 ft/sec ). in accordance with common practice,f
this report will refer to accelerations measured -
in G. To illustrate, it is customarily understood
that 5 G represents an acceleration of 5 x 32.2, or
161 ft/sec
2.3 VELOCITY-RELATED TERMS
e Velocity Change in Major Impact (Av)
The decrease in velocity of the airframe during the
major impact, expressed in feet per second. The
.ajorimpact is the one in which the highest forces
26
are incurred, not necessarily the initial impact.
For the acceleration pulse shown in Figure 2, the
major impact should be considered ended at time t2 .
Elastic recovery in the structure will tend to re-
verse the direction of aircraft velocity prior to
t Should the velocity actually reverse, its di-
rection must be considered in computing the velocity
change. For example, an aircraft impacting downward
with a vertical velocity component of 30 ft/sec and
rebounding with an upward component of 5 ft/sec
should be considered to experience a velocity change
Av = 30- (-5) = 35 ft/sec
during the major impact. The velocity change during

impact is further explained in Section 7.2 of Volume
III.
* Longitudinal Velocity Change

The decrease in velocity during the major impact mea-
sured along the longitudinal (roll) axis of the air-
craft. The velocity may or may not reach zero during
the major impact. For example, an aircraft impacting
the ground at a forward velocity of 100 ft/sec and
slowing to 35 ft/sec before rebounding into the air
would experience a longitudinal velocity change of
65 ft/sec during this impact.
* Vertical Velocity Change

sured along the vertical (yaw) axis of the aircraft.
The vertical velocity generally reaches zero during
the major impact.
a Lateral Velocity Change -

sured along the lateral (pitch) axis of the aircraft.
2.4 FORCE TERMS
* Load Factor
A crash force can be expressed as a multiple of the
weight of an object being accelerated. A load fac-
tor, when multiplied by a weight, produces a force
27
Peak
Averaget --
4.0
I
Figure 2. Typical aircraft floor acceleration pulse.
-which can be used to establisý_ static strength (see

. . a'_... Static
_._ ... Strength). Load factor is expressed in uni~ts
of G.
9 Forward Loado
Loading in a direction toward the nose of the Iroll)

longituinal air- ji
craft, parallel to the aircraft
28
9 Aftward Load
Loading in a direction toward the tail of the air-
craft, parallel to the aircraft longitudinal (roll)
axis.
* Downward Load
Loading in a downward direction parallel to the ver-

tical (yaw) axis of the aircraft.
Upward Load
Loading in an upward direction parallel to the verti-
cal (yaw) axis of the aircraft.
* Lateral Load
Loading in a direction parallel to the lateral (pitch)
axis of the aircraft.
* Combined Load
Loading consisting of components in snore than one of
the directions described in Section 2.2.
2.5 DYNAMICS TERMS
* Rebound
Rapid return toward the original position upon re-

lease or rapid reduction of the deforming load, uou-
ally associated with elastic deformation.
*Dynamc Overshoot
The amplification of decelerative force on cargo or
personnel above the input decelerative force (ratio
of output to input). This amplification is a result
of the dynamic response of the system.
e Transmissibility
The amplification of a steady state vibrational input
amplitude (ratio of output to input). Transmissibil-
ities maximize at resonant frequencies and may pro-
duce motion and acceleration amplification similar to
dynamic overshoot.
29
* - - ----. . ~ - -* - - - - - -
2.6 CRASH SURVIVABILITY TERMS
* Survivable Accident
An accident in which the forces transmitted to the
occupant through the seat and restraint system do not
exceed the limits of human tolerance to abrupt accel-
erations and in which the structure in the occupant's
immediate environment remains substantially intact,
to the extent that a livable volume is provided for
the occupants throughout the crash sequence.
* Survival Envelope
The range of impact conditions--including magnitude
occur-
and direction of pulses and duration of forces
ring in an aircraft accident--wherein the occupiable
area of the aircraft remains substantially intact,
both during and following the impact, and the forces
transmitted to the occupants do not exceed the limits
of human tolerance when current state-of-the-art re-
straint systems are used.
It should be noted that, where the occupiable volume
is altered appreciably through elastic deformation
during the impact phase, survivable conditions may
not have existed in an accident that, from postcrash
inspection, outwardly appeared to be survivable.
2.7 OCCUPANT-RELATED TERMS
a Human Body Coordinates
In order to minimize the confusion sometimes created
by the terminology used to describe the directions
of forces applied to the body, a group of NATO scien-
tists compiled the accelerative terminology table of
equivalents shown in Figure 3 (Reference 17). Termi-
nology used throughout this guide is compatible with -
the NATO terms as illustrated.-
17. Gell, C. F., TABLE OF EQUIVALENTS FOR ACCELERATION TERMI-

NOLOGY, Aerospace Medicine, Vol. 32, No. 12, December
1961, pp. 1109-I111.
30
I.-' ____*--____- ____

-. ___.
Headward
(+G Z) Direction of
accelerative force
Vertical
Back to chest Headward - Eyeballs-down

(sternumward) Lateral rightEyeballs-up
Later (+Gh) Transverse
y Lateral right - Eyeballs-
left
Lateral left - Eyeballs-
right
Back to chest - Eyeballs-
in
Lateral left Chest Chest to back - Eyeballs-
to back out
(-G
y (spineward) Note:
Tailward (-Gx The accelerative force on
(-G the body acts in the same
direction as the arrows.
Figure 3. Terminology for directions of forces on

the body. (From Reference 17)
e Anthropomorphic Dummy
A device designed and fabricated to represent not
only the appearance of humans but also the mass dis-
tribution, joint locations, motions, geometrical sim-
ilarities such as flesh thickness and load/deflection
properties, and relevant skeletal configurations such
as iliac crests, ischial tuberosities, rib cages,
etc. Attempts are also made to simulate human re-
sponse of major bcructural assemblages such as
- - thorax, spinal column, neck, etc. - The dummy is
strapped into seats or litters and used to bimulate
a human occupant in dynamic tests.
* Human Tolerance
For the purposes of this document, human tolerance

is defined as a selected array of parameters that
describe a condition of decelerative loading for
which it is b,!lieved there is a reasonable probabil-
ity for survival without major injury. As used in
S'31
this volume, designing for the limits of human toler-
ance refers to providing design features that will
maintain these conditions at or below their tolerable
levels to enable the occupant to survive the given
crash environment.
Obviously, the tolerance of the human body to crash
environments is a function of many variables includ-
ing the unique characteristics of each person as well
as the loading variables. The loads applied to the
body include decelerative loads imposed by seats and
restraint systems as well as localized forces due to
impact with surrounding structures. Tolerable magni-
tudes of the decelerative loads depend on the direc-
tion of the load, the orientation of the body, and
the means of applying the load. For example, the
critical nature of loads parallel to the occupant's
spine manifests itself in any of a number of types of
spinal fractures, but typically the fracture is an
anterior wedge, or compressive failure of the front
surface of a vertebra. Forces perpendicular to the
occupant's spine can produce spinal fracture through
shear failures or from hyperflexion resulting, for
example, from jackknife bending over a lap belt-only
restraint. The lap belt might inflict injuries to
the internal organs if it is not retained on the pel-
vic girdle but is allowed to exert its force above
the iliac crests in the soft stomach region. Exces-
sive rotational or linear acceleration of the head
can produce concussion. Further, skull fracture can
result from head impact with surrounding structure.
Therefore, tolerance is a function of the method of
occupant restraint as well as the characteristics of
the specific occupant. Refer to Chapter 4 of Vol-
ume 1I for a more detailed discussion of human tol-
erance.
9 Submarining
Rotation of the hips under and about the lap belt as r
a result of a forward inertial load exerted by decel-
eration of the thighs and lower legs, accompanied by
lap belt slippage up and over the iliac crests. Lap
belt slippage up and over the iliac crests can be a
direct result of the upward loading of the shoulder
harness straps at the center of the lap belt.
* Effective Weight
The portion of occupant weight supported by the seat
with the occupant seated in a normal flight position.
22
1-*-
___
__
__
___
__
___
__
__
This is considered to be 80 percent of the occupant
weight since the weight of the feet, lower legs, and
part of the thighs is carried directly by the floor
through the feet.
9 Iliac Crest Bone
The upper, anterior portion of the pelvic (hip) bone.
These "inverted saddlem bones are spaced laterally
about 1 ft apart; the lower abdomen rests between
these crest bones.
. Lap Belt Tiedown Strap (also Negative-G Strap,

r Crotch Strap)
Strap used to prevent the tensile force in shoulder

straps from pulling the lap belt up when the re-
strained subject is exposed to -G (eyeballs-out)
acceleration.
2.8 SEATING GEOMETRY (SEE FIGURE 4 FROM REFERENCE 18)
e Design Eye Position
A reference datum point based on the eye location
that permits the specified vision envelope required
by MIL-STD-850 (Reference 19), allows for slouch,
and is the datum point from which the aircraft sta-
tion geometry is constructed. The design eye posi-
tion is a fixed point in the crew station, and re-
mains constant for pilots of all stature via appro-
priate seat adjustment.
* Horizontal Vision Line
A reference line passing through the design eye po-
sition parallel to the true horizontal and normal
cruise position.
"18. Military Standard, MIL-STD-1333, AIRCREW STATION GEOMETRY

FOR MILITARY AIRCRAFT, Department of Defense, Washington,
D. C.
19. Military Standard, MIL-STD-850, AIRCREW STATION VISION
REQUIREMENTS FOR MILITARY AIRCRAFT, Department of Defense,
Washington, D. C.
33
i 1.
Design eye position
Horizontal 13 in.
vision line
S~130 desired
i minimum
back angle
Thigh tangent 31.5 in.

line / •
Back tangent
I >/l line
90o0'/-Buttock
J• reference
; line
/ / --Seat reernc
100 minimum I
/20" maximum! .5.75 in.
,f,, or heli- H
"copters, Buttock reference point planes
IN 51 minimum
"/ for others Heel rest line
(Not necessarily the floor)
Figure 4. Seating geometry. (From Reference 18)
• Back Tangent Line

A straight line in the midplane of the seat passing
tangent to the curvatures of a seat occupant's back
-when leaning back and naturally compressing the back
cushion. The seat back tangent line is positioned
13 in. behind the design eye position measured along
a perpendicular to the seat back tangent line.
• Buttock Reference Line
A line in the midplane of the seat parallel to the
horizontal vision line and tangent to the lowermost
natural the seat of
protrusion
sitting on a selected size of occupant
cushion.
34•
L.i
0 Seat Reference Point (SRP)
The intersection of the back tangent line and the
buttock reference line. The seat geometry and loca-
tion are based on the SRP.
9 Buttock Reference Point

A point 5.75 in. forward of the seat reference point
on the buttock reference line. This point defines
. the approximate bottom of an ischial tuberosity,
thus, representing the lowest point on the pelvic
structure and the point that will support the most
load during downward vertical loading.
o Heel Rest Line
The reference line parallel to the horizontal vision

line passing under the tangent to the lowest point
on the heel in the normal operational position, not
necessarily coincidental with the floor line.
STRUCTURAL TERMS
* Airframe Structural Crashworthiness
The ability of an airframe structure to maintain a

protective shell around occupants during a crash and
to minimize magnitudes of accelerations applied to
the occupiable portion of the aircraft during crash
impacts.
* Structural Integrity
The ability of a structure to sustain crash loads
without collapse, failure, or deformation of suffi-
cient magnitude to: (1) cause injury to personnel,
or (2) prevent the structure from performing as in-
tended.
o Static Strength --
The maximum static load that can be sustained by a

structure, often expressed as a load factor in terms
of G (see Load Factor, Section 2.4).
@ Strain
The ratio of change in length to the original length
of a loaded component.
35
• Collapse
Plastic deformation of structure to the point of loss
of useful loJ.-4-carrying ability. Although normally
considered detrimental, in certain cases collapse can
prove beneficial as a significant energy-absorbing
process, maintaining structural integrity.
* Failure
Loss of load-carrying capability, usually referring
to structural linkage rupture.
* Limit Load
In a structure, limit load refers to the load the
structure will carry before yielding. Similarly, in
an energy-absorbing device, it represents the load
at which the device deforms in performing its func-
tion.
"* Load Limiter, Load-Limiting Device, or Energy
Absorber
These are interchangeable names of devices used to
limit the load in a structure to a preselected value.
These devices absorb energy by providing a resistive
force applied over a deformation distance without
significant elastic rebound.
"* Specific Energy Absorbed (SEA)
The energy absorbed by an energy-obsorbing device or
structure divided by its weight. SEA is usually pre-
sented in inch-pounds per pound.
* Bottoming
The exhaustion of available-stroking distance -accom-
panied by an increase in force, e.g., a seat strok-
ing in the vertical direction exhausts the available
distance and impacts the floor.
* Bulkhead
A structural partition extending upwards from the
floor and dividing the aircraft into separate com-
partments. Seats can be mounted to bulkheads in-
stead of the floor if sufficient strength is pro-
vided.
36
iLi ~iii
3. PRIMARY DESIGN CONSIDERATIONS
3.1 INTRODUCTION
Occupant protection and survival in aircraft accidents should
be a primary consideration in the design, development, and
testing of aircraft seats and litters. All operational re-
quirements as specified in other design guides should also be
met. Adequate occupant protection requires that both seats
and litters be retained generally in their original positions
within the aircraft throughout any survivable accident. In
addition, the seat should provide an integral means of crash
load attenuation, and the occupant's strike envelope should be
delethalized.
3.2 OPERATIONAL ENVIRONMENT

Several environmental and operational factors other than those
associated with crashworthiness affect the design of an ade-
quate seating system. Because of their importance in overall
design, these factors are mentioned briefly prior to the more
detailed presentation of in'ormation concerning crashworthi-
ness.
3.2.1 Comfort
The comfort of an aircraft seat is a safety-of-flight factor
rather than a crash-safety-design factor. An uncomfortable
seat can induce pilot fatigue in a short period of time. Pilot
fatigue is an indirect cause of aircraft accidents. Comfort
is thus of primary concern and must not be unduly compromised
to achieve crash safety.
Comfort is influenced by several factors, including the vi'bra-
tional environment. Adequate comfort also involves maintenance
of adequate body angles and load distributions. Therefore,
thigh tangent angles and seat back angles are influential in
body comfort. If the back angle is less than 13 degrees, the
occupant's back will be required to counteract too much forward
moment resulting from the weight of the body acting through
centers of gravity forward of the spinal column. As the back
angle is increased beyond 13 degrees, the center of gravity is
moved back and the moment is reduced, which provides for much
greater comfort. If the thigh tangent angle is too low, too
much effort will be required to maintain the lateral orienta-
tion of the legs. If the cushion supports the lateral position
of the legs, comfort will be improved. Also, increasing the
thigh tangent angle seems to rotate the pelvis to the rear,
effectively moving the center of gravity aft and providing a
rearward moment in the pelvis that reduces the forward moment
1 37
4- -.- - -- -. - _ __ _ __ _ _ __ _ _ - .-
I
on the spine. A thigh tangent angle of 5 to 20 degrees is re-
quired by MIL-STD-1333 (Reference 17); however, it is recom-
mended here that tangent angles greater than 10 degrees be
used to maximize comfort and to reduce submarining tendencies.
Another aspect of comfort includes the width of the seat. Too
narrow a seat can exert lateral forces on the sides of the body
or force the body to be held forward out of the constraints of
the seat bucket, again increasing discomfort. Maximum seat
widths should be provided consistent with the space availabl.
in the aircraft, including consideration for the volume around
the seat needed for lateral deflection during crash stroking
and for items such as the collective control. Typically, the
seat pan is required to be at least 18 in. wide with human fac-
tors specialists requesting 20 and compromising at 19 in. This
dimension need not be restrictive if proper consideration is
given to providing sufficient room for the seat during the ini-
tial design of the surrounding aircraft structure.
The surface upon which the occupant sits has a major influence
on comfort. The function of this surface is to spread the con-
tact load over the largest possible area, thereby decreasing
high pressure points and preventing restriction of blood flow
in these areas. In the past, this has been accomplished by
nets or by extremely thick, soft cushions. Although such solu-
tions provided comfort for prolonged flights, this practice is
no longer acceptable since the low spring rates of these nets
or cushions make them extremely hazardous in crash situations.
The low spring rates allow large relative velocities to build
up between the occupant and the airframe or seat during the
imposition of decelerative loads and increase the hazard to
the occupant. Thus, the cushion must provide adequate distri-
bution of loads but not allow excessive motion during crash
loading.
Another aspect of comfort is thermal ventilation. The thermal
ventilation requirement for seat cushions is particularly im-
portant in hot, humid climates. The close contact between the
buttocks or the back and the interfacing cushions can result
in an elevation of temperatdresýcoincident with collection-of
moisture through perspiration. Provisions should be made for
air circulation to carry the hot, humid air out of this inter-
face area, or thermal comfort will be inadequate.
3.2.2 Seat Adjustments
Passenger seats are not usually adjustable; however, in most
cases, adjustment is mandatory for crewseats. First, the cock-
pit and crew station have been designed for a particular eye
position. This eye position is associated with the size of a
38
* • 50th-percentile male occupant; consequently, occupants of
smaller or larger stature may not be located efficiently if
seat adjustment is not provided. Theoretically, the seat ad-
justment enables each occupant to adjust his eye position to
the optimum point. Typically, a ±2.5-in, vertical adjustment
from the neutral seat reference position is required tc account
for this variation in occupant size. Plus or minus 2.5 in. of
fore-and-aft adjustment is also required to permit the desired
repositioning of the eye and for locating the occupant at the
proper distance from controls, pedals, etc. Of course, human
factors should be considered in the design of adjustments.
Adjustment mechanisms should be easily found, easy to use, and
required adjustment motions should be precise, allowing the
occupant to easily get into the most comfortable position with-
out a great deal of distraction. Further, there should be an
efficient verification that the seat is firmly locked into the
chosen position.
3.2.3 Vibration Damping

By its basic nature, the helicopter includes a great number of
vibration sources, primarily as a result of the relatively
great number of moving parts. Typical critical frequencies
include the multiples associated with numbers of blades and
rotor speed. Critical conditions are located at multiples of
the main rotor speed; for example, one, two, four, and eight
per revolution. On four-bladed main rotors, the four-per-
revolution frequency is typically between 18 and 20 Hz. This
driving frequency will be present constantly during cruise;
therefore, it is highly desirable that the resonant frequency
of the seat, both empty and occupied, fall outside the 18- to
20-Hz frequency range. Other frequencies, such as eight per
revolution, can also be a problem. For startup and shutdown
conditions, the resonant frequency of the seat should be high
I
(not lie in the range of 2 to 25 Hz), and considering the
eight-per-revolution frequency it would be desirable to keep
the natural frequency above 40 Hz.
Seat vibrational problems are often difficult to solve because
the predetermined size and general structure of the seat seem
to control the occupied seat natural frequency rather than-the
design options that lie withi-r the limits of weight and cost.
However, the occupied seat natural frequency must be considered
since seat vibration can be very distracting to the occupant,
for example, in the lateral direction where the thighs touch
the sides of the bucket.
Stiffening of the structure is extremely costly in weight; how-
ever, in certain situations it may be the only viable solution
39
-. - - * - - - . - - - - - - - *
to the problem. Dampers that can be added to the seating sys-
tem normally consist of sprung and damped masses. Usually
these mechanisms are very heavy and resorting to their use is
not acceptable in a production aircraft. Isolation of the seat
components by dash pots or elastomeric bearings may provide
possible solutions to this problem.
To summarize, consideration must be given to the vibrational
characteristics of the seat as associated with the vibrational
environment produced in the specific aircraft for which the
seat is-being designed.
3.3 CRASH ENVIRONMENT
3.3.1 DXnamics and Kinematics
When an aircraft crashes, any number of loading combinations

can be imposed on the seat. This is true for rotary- or fixed-
wing aircraft. It would not be useful to try to identify each
and every loading combination; however, studies that have been
completed indicate combinations of loadings that must be dealt
with in the design of the seat and restraint system. For ex-
ample, the stall-spin accident typical of light fixed-wing air-
craft can produce high lateral loadings, the resultant of which
can be oriented in any direction in the longitudinal-lateral
or yaw plane. Studies of helicopter crashes show very high
incidences of side impacts or rollover after impact for some
classes of helicopt,.- (Reference 20).
As an example of the dynamics and kinematics of an aircraft

crash, consider one of the new generation helicopters crashing
in a nose-up, or flare orientation. The tail boom may strike
the ground first, followed by rotation of the aircraft around
a pitch axis, then, by impact of the fuselage. The gear will
strike the ground, and, if it is a wheeled landing gear, the
tires will begin to flatten, absorbing a small amount of en-
ergy. When the rim contacts the ground, the wheel may fail as
the lower oleo strut begins stroking. After completion of the
lower oleo stroke, the second stage will begin, and energy-
absorbing stroke will continue until the fuselage impacts the
ground. If the ground is relatively soft, the ground will de-
form under the loading of the wheels and absorb some energy.
As the fuselage impacts, the softer ground will deform again
while the fuselage structure is deforming. As the fuselage
structure deforms, additional energy is absorbed. At this
20. Haley, J. L., ANALYSIS OF EXISTING HELICOPTER STRUCTURES

TO DETERMINE DIRECT IMPACT SURVIVAL PROBLEMS, U. S. Army
Board for Aviation Accident Research, Fort Rucker, Ala-
bama, 1971.
40
4 - __________
- -
point in the sequence, the loads can achieve the significant
magnitudes required to initiate energy-absorbing stroke of the
seat. The landing gear are designed to stroke at a lower load
than that required to activate the vertical energy-absorbing
system in the seats; thus, stroking of the gear will occur
prior to vertical stroking of the seat. This will typically
result in energy-absorbing stroke of the gear followed by an
increase in fuselage loading when the fuselage impacts the
ground and begins to crush. During some part of the crash se-
quence, the seat and fuselage may be stroking together. The
decelerative loads may increase and the fuselage will even-
tually be stopped. Depending on the conditions of the parti-
cular crash, the seat may go on stroking, until it either ab-
sorbs the residual energy of the supported mass or bottoms at
the end of its stroke. Thus, the seat may be the last item in
the load path of interest to remain in motion during the crash
sequence.
One important point here can be used to advantage by the seat

designer. In a crash with combined loading, extremely high
longitudinal or lateral loads can be applied to the seat after
stroking of the energy-absorbing gear and during fuselage
crushing. However, once the fuselage has come to a stop, crash
loading is no longer exerted on the seat, and it may continue
its stroke until either the residual energy or the seat stroke
is expended. This can be important to the designer. For ex-
ample, consider a seat design that includes only vertical
energy-absorbing stroke. The seat is not required to with-
stand the high combined loads throughout its complete vertical
stroke, only that portion of the stroke while the lateral/
longitudinal crash loading is applied.
For those aircraft using wells, or depressions, in the floor
under the seat to provide for increased stroke distance beyond
the distance available between the seat pan and the floor, the
seat must be guided sufficiently to clear the sidewalls of the
well to utilize that additional distance. In a seat with a
low lateral spring rate, the seat may move laterally to the
point where it no longer lines up with the well under the seat
pan during the application of the longitudinal/lateral loading.
If the longitudinal/lateral loading is removed soon enough,
the seat may be able to return to alignment and still stroke
into the well under the seat. However, this occurs only if
the longitudinal/lateral loading (in certain cases) has pro-
duced elastic, rather than plastic, deformation. If the de-
formation has been plastic, removal of the load will not cause
the seat to return to its original over-the-well position but
will allow it to continue its vertical stroke in the deflected
configuration. On the other hand, if the elastic deformation
is not damped sufficiently, or if the distance above the well
41
is not sufficient, the rebound of the seat may carry it beyond
the well on the other side without sufficient time to return
to center as it goes through the floor plane. These motions
can be considered during seat design and development phases to
minimize a seat's weight while providing the cra-hworthy per-
formance desired.
Several factors should be considered during the design of a

seat that uses a well to increase available stroke. First, as
much clearance as possible should be left between the outside
of the seat pan and the inside of the well. This will allow
for considerable deflection from the no-load position without
creating impact or interference hazards. The next considera-
tion is that the seat be made as stiff as possible in the la-
teral direction to limit the extent of deflection without
imposing too high a weight penalty. Designing a seat with
energy-absorbing stroke in the lateral direction may compro-
mise the all-important vertical stroke. Usually, the confines
around a pilot/copilot seat, which consist primarily of collec-
tive controls and consoles, do not permit sufficient lateral
motion of the seat to avoid hazardous interference with the
vertical stroke. Since the vertical stroke is the only re-
quired energy-absorbing stroke, its blockage will significantly
degrade the degree of seat crashworthiness. Additionally, re-
cent studies indicate a high frequency of thorax and head in-
juries (Reference 21). Allowing the seat to move either later-
ally or longitudinally any more than necessary could increase
the risk of head or chest impact on surrounding structure.
One could infer from the above discussion that energy-absorbing
strokes in the lateral or longitudinal directions are not de-
sirable and serve to increase the overall hazard to the occu-
pant. However, this general statement cannot be made, as the
degree of hazard or benefit will depend on the configuration
of the specific aircraft and the location of the seat within
the aircraft. In certain aircraft, space will be available
for seats that stroke in more than just the vertical direction,
and, when it is available, it may be advantageous to include
it in the system design.
Because of the cabin location of troop seats, -they typically _
have a less lethal area surroundiig, them than crewseats, and
21. Singley, G. T., III, and Desjardins, S. P., CRASHWORTHY

HELICOPTER SEATS AND OCCUPANT RESTRAINT SYSTEMS, in 0pr&
tional Helicopter Aviation Medicine, AGARD Conference ro-
ceedings No. 255, North Atlantic Treaty Organization, Ad-
visory Group for Aerospace Research and Development,
Neuilly sur Seine, France, May 1978.
42
do not have to stroke into wells. Troop seats are typically
load limiting in the
S'- longitudinal and lateral as well as verti-
cal directions. This three-dimensional load limiting reduces
occupant decelerative loading and the crash loads on the seat
structure in the transverse direction in comparison to a
vertical-only load-limiting seat. Lower loading of the seat
allows a lighter seat design. In the case of a side-facing
seat, load limiting along the seat's lateral axis is necessary
if the occupant decelerative loading during the specified air-
craft forward crash impact conditions of Volume II is to be
kept within human tolerance limits for lateral decelerations.
in reviewing the dynamics and kinematics of crashing aircraft,
it beoomes quite apparent that all combinations of orientations,
loading, and load directions can exist. (Volume II presents
a detailed discussion of crash impact dynamics and kinematics.)
It should also be remembered that the seat is designed to ab-
sorb only a portion of the crash energy required to decelerate
the occupant in a tolerable environment. There are numerous
crash orientations in which the aircraft has a lateral compo-
nent of impact velocity, whether it results from a lateral
drift of the aircraft or from its attitude at impact. These
components of velocity can produce high lateral loading of gear,
which, in some cases, may simply break off before absorbing
significant energy. Consider, for example, the case of an air-
craft impacting the ground with a high roll angle. Loss of
gear will result in the aircraft fuselage impacting the ground
without the reduction in energy normally attributed to stroking
of the gear. Therefore, systems analyses must take this factor
into account. As an example of the possible dangers, it might
be decided that landing gear should absorb all the crash energy
associated with the 42-ft/sec vertical impact; therefore, seat
stroking would not be required. The results of applying this
logic to hardware would seriously reduce the overall crash-
worthiness of the aircraft in those crashes where the full en-
ergy absorption of the gear could not be realized. Therefore,
seats should contain the minimum energy-absorbing stroke de-
fined in this document, regardless of the energy-absorption
capacity of the gear.
After- a helicopter crashes, the rotating main rotor may strike
the -ground or other obstacles and roll the helicopter onto its
side. Because of the high center of gravity, the helicopter
may roll over without any added lateral impulse from the main
rotor blades after gear failure. In any case, the kinematics
of crashed helicopters can be quite complex and violent, and
the helicopter may come to rest in any orientation. Because
of these kinematics, loads are specified in all directions for
seats. This subject will be covered in more depth later in
this volumey however, the crash kinematics of these aircraft
I 43
t TF•-'i
•' i• r-•-at I I iil.1
demand strength requirements in all directions, including up-
ward and aftward. In this regard, it should be remembered that
the seat may have used a significant portion of its available
vertical stroking distance durinc, the major impact. If the
aircraft should then follow through with a flip, or land on
its back, the system should maintain the seat near its final
stroked position rather than allowing the seat to return to
its original position. Upward travel could be extremely haz-
ardous if the roof of the fuselage were severely crushed and
the occupant were free to travel unrestrained back toward his
initial position. Severe head and/or neck injuries could re-
suit.
In summary, it must be remembered that, to produce a crash-

worthy design, systems analyses must consider likely combina-
tions of loadings, including potential losses of energy-
absorbing structure, such as landing gear.
3.3.2 Design Conditions and EnveloP2s

The design impact conditions for light fixed- and rotary-wing
aircraft are presented in Volume II and are repeated here in
Table 2. All seats, restraint systems, and litters should be
designed for these impact velocities and provide the desired
performance in the design crash environments.
TABLE 2. SUMMARY OF DESIGN CONDITIONS

FOR ROTARY- AND LIGHT FIXED-
WING AIRCRAFT
Velocity
Impact change
Direction (ft/sec)
Longitudinal 50
Vertical 42
Lateral* 25
Lateral** 30
*Light fixed-wing, attack, and cargo
helicopters.
**Other helicopters.
44
3.3.3 Structural Distortion
4-
Structural distortion of the airframe and its resulting load-
ing of the seat must be considered in the design stages. For
example, a ceiling-mounted seat may experience lower loads than
a floor-mounted seat because of the distortion or deflection
of the roof and supporting walls. However, additional stroke
distance may be required due to the inefficiency of the stroke
provided by distortion of the airframe as compared to that pro-
vided by a load-limited seat. The effective, stroke of a seat
considered to be rigidly attached (no energy absorbers between
the seat and roof) to the roof also must be ýconsidered. If
the seat pan is 12 in. from the floor of the aircraft and the
roof of the aircraft is expected to distort downward on the
order of 12 in., careful consideration must be given to elimi-
nating rebound rather than increasing total stroke, which could
result in bottoming. In the practical case, the roof probably
distorts something less than the distance between the seat pan
and the floor of the aircraft; therefore, energy-absorbing
stroke should be provided in the seat to maximize usage of the
available space. A systems analysis should be applied to this
situation to establish the correct combination of variables.
A considerable amount of the downward motion of an aircraft
ceiling may be elastic. It would be advantageous to eliminate
the rebound from this elastic distortion from the occupant and
seat. Consideration could be given to a device that allowed
vertical downward motion of the seat but restrained it from
following the roof during its elastic rebound. It is possible
to think in terms of an energy-absorbing device under the seat,
connected to the floor, that resists motion in both the down-
ward and upward directions. This would allow only partial re-
turn of the seat resisted by the energy-absorbing system.
Another alternative would be to provide energy absorbers be-
tween the seat and the roof that would stroke when the roof
returned to its equilibrium height. Again, a device that locks
the seat in its lowest position would be required with this
concept. Adequate support of the ceiling to support the ap-
plied loads with low deflections eliminates the prpblems men-
tioned. Efficient use of ceiling-mounted seats can be achieved
in these aircraft.
A major consideration in providing crashworthy seating systems

is the possibility of a local distortion in the part of the
aircraft to which the seat is attached. For example, a floor-
mounted seat may have to withstand severe distortions as a re-
sult of underfloor and floor deformations caused by impact
forces. If the aircraft crashes on uneven ground or encounters
rocks or stumps, distortions of the underfloor structure can
occur. The seat structure or seat attachment to the floor
2* 45
&
should be adequate to permit these distortions without produc-
ing failure of the seat structure or its attaching mechanisms.
It should be noted that the forces causing this distortion can-
not be resisted by the seat structure. In other words, it is
not feasible to build a seat strong enough, if rigid, to main-
tain the attachment to the aircraft in these situations. The
crash loads causing the distortion will, in most cases, exceed
any strength that can be designed into the seat, thus, produc-
ing failures if not adequately accounted for in the design.
Likewise, distortion of bulkheads in bulkhead-mounted seats

presents the same type of problem. It is apparent that local
distortion of a bulkhead usually will not be of the magnitude
of the distortions that can occur in the floor structure of an
aircraft. Rocks and stumps can produce extremely large local
deformations of structures upon which floor-mounted seats are
mounted. However, rocks and stumps normally will not be in-
volved in distortion of bulkheads and bulkhead structure. Con-
sequently, the distortion requirements for seat mountings on
bulkheads are less severe. A recent search of USASC crash rec-
ords identified no known cases of bulkhead-mounted seat loss
due to bulkhead distortion or fracture of attaching structure.
It is expected that sidewalls will deform more than transverse
bulkheads, although they would not be as susceptible to rocks
and stumps as floors. The deformation would usually be one of
the wall buckling outward near the floor and changing the la-
teral and vertical relationships between attachment points.
However, it should be remembered that in helicopters, sidewall-
mounted seats are not usually pilot or copilot seats and there-
fore, are usually not of the stiffness that would create a
problem in the environment described. For some fixed-wing air-
craft, though, this may not be the case. For these cases, the
aircraft/seat interface should be designed to be compatible
by allowing flexibility in the seat, in the attachments, in
stiffening the sidewall of the aircraft, or by simply not at-
taching rigid seats to sidewalls. Floor, bulkhead, and side-
ward warpage requirements are presented in Section 4.4.5, Joint
Deformation.
3.4 APPLICABILITY OF CRITERIA
The recommendations in this volume will apply to all categories
of U. S. Army aircraft. Those recommendations having applica-
tion to a specific class or category of aircraft only will be
so indicated by the text.
46
I
S3.5 ACCEPTANCE CRITERIA

In addition to operational requirements specified in other de-
sign guide documents, seats and litter systems should be de-
signed to provide occupant protection under crash conditions
as specified in Volume II. Appropriate stress analyses, tests,
and operational requirements outlined in this volume should be
met by every seat, restraint, litter system, and by the cock-
pit and cabin interior prior to acceptance.
3.6 SELECTION CRITERIA
Crashworthy seats, restraint systems, litter systems, and cock-
pit and cabin materials should be evaluated on the basis of
the occupant protection provided, and on their anticipated re-
liability and serviceability under the oierational and poten-
tial crash conditions expected.
47.
II
~.I
O 47
_.•°•.
1 4 .' ~ ~ ~ . *~l... "
4. DESIGN PRINCIPLES FOR SEATS AND LITTERS

4.1 INTRODUCTION
There are several types of Army aircraft seating systems: pi-
lot, copilot, crew chief, gunner, observer, student, medical
attendant, troop, and passenger. Cockpit seats are typically
forward-facing; however, cabin seats may face in any direction.
Most are single-place seats, but in a few aircraft, two-,
three-, and four-occupant cabin seats are provided. A single
occupant seat is the preferred configuration in order to avoid
situations where the energy-absorbing systems of multi-unit
seats are rendered ineffective due to less than full occupancy
(insufficient weight to activate the energy-absorbing mechan-
isms at loads within human tolerance limits). To the maximum
extent practical, seats should be interchangeable to enable
standardization. It is desirable that all seats face in the
same direction so that the seat backs protect occupants from
loose equipment which can become flying projectiles during.
crash impact.
The rearward-facing seat is optimal for providing maximum sup-
port and contact area in longitudinal impacts. The only criti-
cal impact sequence for the rearward-facing seat is one that
involves a severe lateral component that allows sideward move-
ment of the occupant prior to application of the longitudinal
or vertical pulse. However, lateral torso movement can be pre-
vented by use of an adequate restraint system of much lighter
weight than that required for other seat orientations. When
practical, the rearward-facing seat should be used.
Those crew members required to face forward in the conduct of
their duties can be afforded adequate protection by the use of
a restraint system consisting of shoulder straps, a lap belt,
and a lap belt tiedown strap as discussed in Chapter 7. The
lap-belt-only restraint is undesirable, as noted in the human
tolerance section of Volume II. If all forward-facing pas-
sengers are provided with adequate upper- and lower-torso re-
straint, forward-facing seats -re acceptable as a second choice
to rearward-facing seats. If a single, diag6nal, upper-torso
restraint is used, it should be placed over the outboard shoul-
der of the occupants to provide restraint against lateral pro-
trusion of the occupant outside the aircraft or impact with the
sidewall.
Previously, side-facing seats have been provided with lap belt
restraint only. This arrangement is considered completely in-
adequate for providing crash protection. Even with the addi-
tion of a shoulder harness or diagonal chest strap, the toler-
ance to abrupt acceleration is minimal. The use of side-facing
48
seats is least desirable from the crash safety standpoint;
however, when no reasonable alternative to their use exists,
adequate restraint must be provided. If a single, diagonal,
upper-torso restraint is used, it should be placed over the
forward-facing shoulder (relative to the aircraft).
4.2 LITTERS AND THEIR ORIENTATION

The supine position of a litter patient is ideal for resisting
vertical impacts. The contact arep is the maximum possible,
and the decelerative forces act transversely to the body. tor
existing litters, the major problem occurc as a result of im-
pact forces in the lateral/longitudinal plane. The relatively
flat litter surface makes it difficult to provide an adequate
restraint harness to resist these loads. The current practice
of wrapping two lengths of webbing around the litter offers a
degree of restraint oriented transversely to the body, but only
frictional forces prevent the body from sliding off the litter
in the lengthwise direction.
Litters should be installed laterally to provide more positive

restraint for expected combined crash forces.. A lateral litter
orientation also will prevent the litter from becoming com-
pletely detached from its current supports as occurs in a lon-
gitudinal orientation explained in Reference 22. The litter
must withstand all of the environments previously described
for the seats.
4.3 MATERIALS
Designers should select materials that offer the best strength-

to-weight ratios while still maintaining sufficient ductility
to prevent brittle failures. The guidelines in this section
will alert the designer to certain material properties that
can contribute to improved structural designs. These proper-
ties include ultimate strength, elongation, and energy-
absorbing capabilities. The standard method for selecting ma-
terials using elastic analysis is adequate for most conditions
in the working lifeof an article. For crashworthiness, how-
ever, only one applicat-i-on of the maximum load is expected,
and the behavior of the material beyond the yield point gen-
erally is important.
22. Weinberg, L. W. T., AIRCRAFT LITTER RETENTION SYSTEM DE-

SIGN CRITERIA, Aviation Crash Injury Research (AvCIR), Di-
vision of Flight Safety Foundation, Inc.; USAAVLABS Tech-
nical Report 66-27, U. S. Army Aviation Materiel Labora-
tories, Fort Eustis, Virginia, April 1966, AD 632457.
49
S_...t.. . . .
The degree of ductility needed in a seat's basic structural
parts is highly dependent upon whether the seat structure is
designed to absorb energy by the use of a separate Ilcd-
limiting device or whether large plastic deflections of the
basic structure are required. As a general rule, a value of
10-percent elongation is a rough dividing line between ductile
and nonductile materials. The 10-percent value is recommended
as a minimum for use on all critical structural members of
nonload-limited seats because the exact peak load is unpredict-
able due to pulse shape, dynamic response of the system, and
velocity change. A minimum elongation of 5 percent in the
principal loading direction is suggested for use on critical
members of load-limited seats because the loads and strains
are more predictable.
Castings are not recommended for use in primary load paths.
In general, their quality is more difficult to verify and re-
produce, and their ductility and fracture toughness are less
than for forgings.
The effects of stress corrosion, (for example, selection of
7075 aluminum alloy in a T73 condition rather than T6), must
be considered, as well as hydrogen embrittlement due to heat
treating or various processing steps such as pickling (for ex-
ample, 17-4PH stainless steel). In short, adherence to all
the normal engineering design principles must prevail.
Flammability and toxicity retardation requirements are discus-

sed in Volume V. Upholstery padding and other materials used
in seats should meet the specified requirements.
4.4 STRUCTURAL CONNECTIONS
4.4.1 Bolted Connections

For the manufacture of basic aircraft structure, most aircraft
companies recommend 15- and 25-percent margins of safety for
shear and tensile bolts, respectively. These factors are in-
tended to allow for misalignment of holes, stress concentra-
tions, and fatigue. Fatigue is not generally a factor in the
design of a seat or litter system fitting, since high loading
of the fitting would be a one-time situation. Therefore, the
safety factor for shear and tensile bolts located in load-
limited portions of the seat where loads can be predicted ac-
curately, can be reduced to 5 and 10 percent, respectively.
Also, good aircraft engineering practice dictates that bolts
less than 0.25 in. in diameter should not be used in tensile
applications because of the ease with which these smaller bolts
can be overtorqued. Because of the obvious advantages of
structure being able to distort while maintaining load-carrying
50
ability, fasteners of maximum ductility for the application
should always be selected. Where possible, fasteners such as
bolts and pins should have a.minimum elongation of 10 percent
in the longitudinal and transverse directions.
4.4.2 Riveted Connections
The guidelines for riveted joints are presented in MIL-HDBK-5,
and it is recommended that these guidelines be followed (Refer-
ence 23).
4.4.3 Welded Connections
Welded joints can be 100 percent efficient; however, the ac-
tual efficiency is dependent upon the skill of the welder, the
process used, and the inspection procedures followed. Welded
joints can be completely acceptable and even superior to bolted
or riveted joints. However, strict inspection procedures
should be used to ensure that welded joints are of good qual-
ity. Welded joints may result in stress concentrations and
misaligned parts in a manner similar to bolted joints; there-
fore, the cross-sectional area of the basic material in the
vicinity of a welded joint should be 10 percent greater than
the area needed to sustain the design load. Welding processes
are discussed in Military Specifications MIL-W-8604, -6873,
-45205, anrl -8611; these specifications should be used as
guides to ensure quality welding.
4.4.4 Seat Attachment

Cockpit seats are either bulkhead or floor mounted. Acceptable
means of attaching seats to the cabin interior are listed below
(refer to Section 3.3.3 for a discussion of ceiling-mounted
seats and ceiling support stiffness):
1. Suspended from the ceiling with energy absorbers,
and wall stabilized.
2. Suspended from the ceiling with energy absorbers,
and floor stabilized.
3. Wall mounted with energy absorbers.
23. Military Handbook, MIL-HDBK-5, METALLIC MATERIALS AND

ELEMENTS FOR ASROSPACE VEHICLE STRUCTURES, Department of
Defense, Washington, D. C.
51
4. Floor mounted with energy absorbers.
5. Ceiling and floor mounted (vertical energy absorbers
above and below seat).
Suspension or mounting of all seats should not interfere with
rapid ingress or egress. Braces, legs, cables, straps, and
other structures should be designed to prevent snagging or
tripping. Loops should not be formed when the restraint system
is in the unbuckled position. Cabin seats must often be de-
signed so that they may be quickly removed or folded and se-
cured. Tools should not be required for this operation. The
time required by one person to disconnect each single occupant
seat should not exceed 20 sec. The time required by one person
to disconnect multi-occupant seats should not exceed 20 sec *
multiplied by the number of occupants. All foldable seats

should be capable of being folded, stowed, and secured or un-
stowed quickly and easily by one person in a period not to ex-
ceed 20 sec multiplied by the number of occupants.
4.4.5 Joint Deformation
Floor distortions as a result of impact can cause failure of
the seat structure or tiedown connections in an aircraft crash
(see Figure 5). A floor distortion can take the form of a
bulge or dish in the floor surface between the seat tiedown
connections. This produces a rotation of the seat relative to
the floor surface, resulting in a connection failure if the
deflection limits for the attachments are exceeded. A twist-
ing or warping of the floor surface can also take place, pro-
ducing distortion loads in the seat structure. Seat or connec-
tion failure can result from the additional loads imposed.
The seat designer must anticipate possible floor bulging or
I
warping and take appropriate measures in seat structural de-
sign to minimize the adverse effects.
For basically rigid seat structures that are distorted the cri-
tical design parameter appears to be the torsional rigidity of
the seat pan, bucket, and/or structural members. If the tor-
sional rigidity is low, only small forces-are introduced. How-
ever, for stiff seat members, the warpage forces may produce a
structural failure or impose a preload that, when coupled with
-crash inertial loads, results in failure. A high torsional
rigidity in the seat pan may arise from integrating stiff la-
teral cross tubes between side trusses so that the tubes must
also twist with the seat pan. Consequently, it may be desir-
able to connect the cross tubes to the seat pan in such a way
that the seat pan is free to twist independently of the cross
tubes or to design the crossmembers to be soft in torsion. In-
tegrally armored crewseats are stiff and difficult to release
52
r.r
* i
Distortion in Final position

lateral direction
(around roll
Seat mount locations
, •%,. Distortion in.
S "'" ' fore/aft direction

(around pitch axis)
Rock, eart
or other
obstacle Initial position
Figure 5. Sketch illustrating buckling

or "dishing" formation.
from the support structure in order to permit distortion. One

method used successfully to solve this problem has been a three-
legged seat. The three support points can follow the floor
movement without distorting the seat structure because the seat
is free to tip (Reference 24).
24. Desjardins, S. P., and Harrison, H., THE DESIGN, FABRICA-
TION, AND TESTING OF AV INTEGRALLY ARMORED CRASHWORTHY
CREWSEAT, Dynamic Science, Division of Marshall Indus-
tries; USAAMRDL Technical Report 71-54, Eustis Director-
ate, U. S. Army Air Mobility Research and Development
Laboratory, Fort Eustis, Virginia, January 1972, AD742733.
1 53
To prevent seat connection failures induced by floor dist ion,
structural joints should be capable of large angular displace-
ments in all directions without failure. A seat designed prop-
erly for structurally integral load limiting would also satis-
factorily accommodate floor buckling and warping under crash
conditions. Figure 6 illustrates recommended limits of floor
warping or buckling that must be withstood by all floor-mounted
seat designs. The mounts should be capable of withstanding a
')-degree warp of the floor, as well as a ±10-degree rotat.on
"o. ut a roll axis of a single track. The angles are based on
Jistortions that have been noted in potentially survivable ac-
cidents.
With respect to the floor surface and to accommodate rotations
that result from floor bulging, several design configurations
may be considered. Two of these are presented below and are
illustrated in Figure 7.
• A deliberate plastic hinge of sufficiently ductile
material may be incorporated into the tiedown connec-
tion design. This plastic hinge would be required
to permit yielding without failure up to a rotation
angle that exceeds the maximum anticipated as a re-
sult of floor bulging. The hinge also would be re-
quired to carry the associated compressive, tensile,
and shear loads in order to retain the seat while
yielding in bending.
* A structural release such as a ball-and-socket joint
may be used to pcmit relative rotation.
Other methods, such as a combination of a plastic hinge about
one axis and rotation about an axle or pin oriented along a
perpendicular axis, are acceptable also. The joint must be
capable of sustaining large tension, compression, and shear
forces during and after rotation.
The effect of not providing for relative seat leg-to-floor ro-
tation can be illustrated by an actual example. The rear-.legs_-
of a crewseat on early models of a7 U. S. Army helicopter were
attached to a base frame with castings-as illustrated in Fig-
ure 8. These castings failed repeatedly in accidents as a re-
sult of combined axial and bending stresses acting at the re-
gion of stress concentration. Studies showed that the seat
could sustain a longitudinal decelerative force nearly twice
as great when the bending moment at the juncture between the
rear leg and the track fitting was removed.
54
4 . - - - _____ , __ - - . . ...-- • •-
.__________________-
I
S- Thi+-i0* mi5
4oiiaini lutae nFgr . Temmn a
°z
_Y
Figure 6. Static test floor warpage requirement to improve

the probability of seat retention in crashes.
....... in
This modification is illustrated in Figure 9. The moment was

relieved by cutting the corners off the casting so that only
the section around the center bolt remained. The joint was
55.
nw :
S
I'~.
, j
Hinge,
0 i
I
I
(a) Plastic hinge (b) Ball-and-socket

Joint
Figure 7. Concepts for release of floor-distortion-
induced moments.
thereby changed from a fixed- to a pinned-end
SubsequenIt tests showed improved load-carrying configuration.
capacity.
Other methods of relieving torsion and
spherical bearings and slotting holes moments include using
through which -bolts pass.
For example, if a crossmember is required
during floor warping, slots that to move torsionally
relieve the leads can be pro-
vided for fasteners at end fittings.
Figure 10. Figure 11 illustrates This is illustrated in
an
released joint acted on by two torsional example of a fully-
loads and a moment.
The same general principles that apply
also apply for bulkhead-mounted for floor-mounted seats
seats except
tion and degree of warping of the bulkhead that the deflec-
than that of the floor. This is probably appear to be less
being less vulnerable to local planar due to the bulkhead
distortion caused by
I
56
Occupant
O load
Z Aft seat leg

• (steel tube)
A4
Castng
Floor tracks
Ar-Floor
Figure 8. Aft seat leg casting attachment.
IN Pi
Failure point
(stress con- M
centration)
Fixed joint 'Pinned joint

Floor Floor
track track
View ABefore modification view A After modification

Figure 9. Aft seat leg casting attachment modification.
-57
ti..-
Ti
Nut
TorsionT
release-\
Torsion release
Figure 10. Torsional release of joints.
items such as rocks and stumps impacted by the underfloor

structure. -A possible bulkhead distortion configuration is
shown in Figure 12. The recommended angular deflection re-
quirement for bulkhead-mounted seats is a 5-degree rotation in
the plane of the bulkhead. To accommodate local deformation,
each attachment of the seat to the bulkhead should be released
to permit t10-degree rotations in any direction. One technique
for accomplishing this is with spherical bearings, as illus-
trated in Figure 13.
Sidewall-mounted seats require the same considerations as
bulkhead-mounted seats. As mentioned previously, the sidewalls
58
_ i
6TorTorsion
release
release
release
Figure 11. Fully released joint.
of aircraft tend to bow outboard during impacts-with high ver-

tical loading. Therefore, it is advisable that these seats be
designed to accept relatively large distortions without fail-
ure. Although the angles are not known, it is expected that
they may reach 25 degrees.
Seats that are mounted totally on the sidewall should not
create a problem, as they will simply move with the sidewall.
Extremely flexible seats also should be inherently immune from
these problems. However, rigid seats mounted to both the floor
59
• .........-......... ..
'--',-...a• ". •
Bulkhead
Seat attach
-[points (4 places)
Lower fuselage
(a) Initial configuration
Bulkhead
(b) Postcrash configuration
Figure 12. Bulkhead in-plane warping.
60
•lilI I 11 II'TII'IT'I
Clevis
I I
ug LugClevis
Figure 13. Universal release of a joint.
and the sidewall will require special design considerations.

One way to provide the flexibility needed is to include re-
leases such as pin joints, oriented to allow rotation around
an aircraft roll axis. An example is shown in Figure 14. The
attachments should be designed to permit the angle 0 to reach
25 degrees at the maximum dynamic deflection.
The underfloor, bulkhead, or sidewall structure must be de-
signed to be compatible with the seat. For example, the design
of structural releases between the seat and the track may en-
able the seat to maintain its attachment during large floor de-
formations but may add to the torsional responsibilities of un-
derfloor beams. If a large downward load is applied to the
floor structure through a joint that does not carry moment (re-
leased), then the underfloor beams must resist any moment that
may be developed without assistance from the seat structure.
To illustrate, take the case of a seat strut attached through
a release to the front-floor tracK. During longitudinal load-
ing in the forward direction, the strut is loaded in compres-
sion and applies a large downward load at the release. Any
eccentricity between the load vector and the centroid of the
underfloor beam will produce torsional loading around the
beam's long axis. The beam must possess the capability to re-
sist this torsional load through either its own torsional
strength or that of its supporting structure.
61
Initial fuselage
configuration
Deflected
fuselage
configuration
• A
\ ~/,
Floor l
Figure 14. Pin joint releases oriented to allow rotation

around an aircraft roll axis.
4.5 STRENGTH
4.5.1 General
An elastic stress analysis, as used in the design of airframes
and aircraft components subjected to normal flight loads, is
inadequate for the study of all the structure in a crash situa-
tion. For normal flight loads, keeping the stresses well below
the material yield stress to avoid permanent deformation is
necessary because of fatigue problems and, perhaps, other con-
siderations. In a crash situation, however, where only one
62
application of maximum load is expected, fatigue is not a fac-
tor, and the final configuration of a structural component or
its subsequent operational use need not be considered. Conse-
quently, the load-carrying capacity of components deformed be-
yond the elastic limit should be considered in determining the
ultimate seat strength. As a matter of fact, it is advisable
for certain items in the load path to use the rupture strength
as listed for many materials in MIL-HDBK-5 (Reference 23).
The concepts of limit analysis or, in some circumstances, large
deformation analysis, may be employed to make the best use of
* materials in certain components.
It may appear that the only difference between an elastic
stress analysis and an ultimate strength analysis is that the
former is more conservative. However, a more significant dis-
tinction is demonstrated by a comparison of two designs having
the same maximum stresses for elastic behavior but decidedly
different load-carrying capacities when the loads exceed the
elastic limits. For example, consider the following two simi-
lar designs: (a) two simple beams spanning three supports and
(b) a continuous beam spanning the same three supports, as il-
lustrated in Figure 15.
Mw12
(a) Two simple beams (b) Continuous beam
Figure 15. Comparison of analysis methods

for simple beams.
For a uniformly distributed load, w, the bending moment dia-

grams are as shown (assuming elastic behavior). It is noted
that in each case the maximum bending moment is wl /8 and each
design has the same stress. There is a temptation to equate
the designs from a strength viewpoint. However, considering
design (a), if the load is gradually increased, the bending
moment at the center of each span will eventually equal the
moment resistance capability of the beam. For a ductile mater-
ial, a yield hinge would fcrm then at these maximum moment
* 63
points. Additional load could not be accepted without a me-
chanical collapse. This critical load would represent a real-
istic ultimate capacity for the beams. On the other hand, when
a yield hinge occurs in design (b) under similar circumstances,
it would occur at the middle support and, hence, not produce a
collapsing mechanism. The load, w, could be further increased
without collapse until a second set of yield hinges forms be-
tween the supports. Only then would collapse occur. It is
intuitively evident, and may be demonstrated by analysis, that
design (b) sustains a much greater ultimate load than does de-
sign (a), yet the difference is not discernible from elastic
analysis. The design of an entire occupant retention system,
ignoring inelastic postyield behavior, would result in compo-
nents of varying ultimate strengths, some much stronger than
others. The overdesigned components do not increase the
strength of the system. It is desirable that all components
work at the same allowable strength level just before failure.
A 1963 study of the restraint system used in three U. S. Army
aircraft indicated that the strengthening of a few weak links
in the tiedown chain improved the crash strength of these sys-
tems by a factor of 2 with only minor weight increases (Ref-
erences 25 through 27). A simple example of the benefit of
strength analysis beyond the elastic limit is the improvement
in the tiedown strength of the crewseat floor track in one of
the three aircraft. In the existing arrangement, the seat leg
may be positioned directly above a pair of seat track tiedown
bolts (Figure 16). The elongation of the bolts prior to their
failure would not be sufficient to permit bending in the floor
track; thus, no appreciable load could be transmitted to the
adjacent pair of bolts. To improve the ultimate strength of
this connection, it was suggested that aluminum collars, which
compress at a load slightly less than the breaking strength of
25. Haley, J. L., Jr., and Avery, J. P., Ph.D., PERSONAL RE-
STRAINT SYSTEMS STUDY - HC-lB VERTOL CHINOOK, AvCIR 62-26,
Aviation Crash Injury Research (AvCIR), Division of Flight
Safety Foundation, Inc., Phoenix, Arizona, November 1962.
26. Haley, J. L., Jr., and Avery J. P., Ph.D., PERSONAL R-E-
STRAINT SYSTEMS STUDY - HU-IA AND HU-lB BELL IROQUOIS,
AvCIR 62-27, Aviation Crash Injury Research (AvCIR), Divi-
sion of Flight Safety Foundation, Inc., Phoenix, Arizona,
December 1962.
STRAINT SYSTEM STUDY - CV-2 DE HAVILLAND CARIBOU, AvCIR
62-16, Aviation Crash Injury Research (AVCIR), Division of
Flight Safety Foundation, Inc., Phoenix, Arizona, April
1964.
Present attachment
Seat reaction Leg stud
|
0.50 in.J Floor track
i
Proposed attachment
I
• ;•':7• •0.25 in.
Aluminum collars added I

Figure 16. Seat leg anchorage to floor track.
the bolt, be added beneath the nut. Thus, the collars would
yield prior to failure of the center bolts and permit the track
to bend and transmit some load to the adjacent bolts. This
arrangement approximately doubled the ultimate tiedown strength
of the floor track while adding a negligible amount of weight.
4.5.2 Limit Analysis Concepts

Where ductile materials are used, strain concentrations do not
produce rupture prior to significant plastic deformation. If
the geometric configuration of the structure permits only small
elastic deflections, a rigid-plastic mathematical model may
be used. This permits the-use of a limit analysis, which as-
sumes no deformation of structUre until sufficient plastic
hinges, plastic extensors, etc., exist to permit a geometri-
cally admissible collapse mode.
Limit analysis is concerned with finding the critical load suf-

ficient to cause plastic collapse with the physical require-
ments of static equilibrium, yield conditions for the materials,
65
and consistent geometry considerations. The principles of
limit analysis are well developed by a number of authors (Ref-
erences 28 and 29, for example). Two useful principles are
mentioned here: the upper and lower bound theorems. The upper
bound theorem for the limit load (collapse load for a rigid-
plastic structure) states that the load associated with the
energy dissipated in plastic deformation will form an upper
bound for the limit load. The lower bound theorem, on the
other hand, states that the load associated with a statically
admissible stress distribution, which at no poiat exceeds the
yield conditions, forms a lower bound for the limit load. Use
of the upper and lower bound theorems to bracket the limit load
for a given structure makes it possible to obtain a realistic
evaluation of the structure's load-carrying capacity.
4.5.3 Lar.e Deformation Analysis
If a structure contains elements that will permit large, sta-
ble elastic deformations when under load, the equilibrium of
the deformed state must be considered in evaluating ultimate
*strength. For example, if a suitable attachment is made to a
thin flat sheet rigidly fixed at the edges so as to load the
sheet normal to the surface, a diaphragming action will occur.
The equilibrium and stress-strain (e)astic-plastic) relations
for the deformed state would determine the load-carrying capac-
ity. An example of this situation is a seat pan in which mem-
brane rather than flexural stresses are important.
4.5.4 Strain Concentrations

Handbook stress concentration factors provide sufficiently ac-
curate data to allow the designer to modify the strizcture in
the vicinity of stress concentrations. When large deforma-
tions at high load-carrying capacity are desired, as in energy-
absorbing seats, these areas frequently become strain concen-
tration points and rupture occurs, due to excessive strain, in
areas with little deformation and energy input. Large amounts
of energy can be absorbed in the structure only if large volumes
28. Beedle, L., PLASTIC DESIGN OF STEEL FRAMES, John Wiley

and Sons, New York, 1958.
29. Hodge, P. G., Jr., PLASTIC ANALYSIS OF STRUCTURES, McGraw-
Hill Book Company, New York, 1959.
66
-A ____________________________-_
of material are strained uniformly. For further information on
the subject, see pages 69-73 of Reference 30.
4.6 RESTRAINT SYSTEM ANCHORAGE

- The design requirements for occupant restraint systems are pre-
sented in Chapter 7; however, the seat designer must consider
the effect of the anchorage of the restraint system on the char-
acteristics of the seat design. The restraint system should be
anchored to the seat rdther than to basic aircraft structure.
If the restraint system is anchored to basic aircraft struc-

ture, a desirable reduction of loads on the seat frame results;
however, the restraint system must be designed to permit the
energy-absorbing deformation of the seat during an impact.
For example, if a load-limited seat strokes vertically and the
seat belt is anchored to the floor, the loosening of the belt
would permit the occupant to "submarine" under the belt or to
move laterally. When the harness is anchored to the seat struc-
ture, the problem of maintaining a snug harness is reduced.
An advantage of attaching the shoulder harness to basic air-
craft structure is the large reduction in overturning moment
on the seat. To make this attachment acceptable, a simple
load-limiting device might be incorporated into the shoulder
harness anchorage to allow for longitudinal or vertical move-
ment of the seat. On some aircraft, where room allows it,
another option is to locate the anchor point far enough to the
rear of the seat to allow vertical energy-absorbing stroke of
the seat with only a rotation of the strap about the anchor
point. If the distance is sufficiently large, the fore-and-aft
motion resulting from the strap swinging in an arc can also be
insignificant.
4.7 CRASH ENERGY ABSORPTION
4.7.1 General
The magnitude of a crash force is a function of the input ve-
locity and the stopping distance. The stopping distance is
controlled basically by the crushing of the airframe and land-
ing gear in a given direction coupled with the gouging of the
impact surface. The magnitude of the average deceleration of
30. Turnbow, J. W., et al., AIRCRAFT PASSENGER-SEAT-SYSTEM
RESPONSE TO IMPULSIVE LOADS, Aviation Safety Engineering
and Research (AvSER), Division of Flight Safety Founda-
tion, Inc.; USAAVLABS Technical Report 67-17, U. S. Army
Aviation Materiel Laboratories, Fort Eustis, Vir ;iria,
August 1967, AD 661088.
* 67
L~..
a given point of the aircraft may be calculated from the fol-
lowing equation:
i.v a 2 -vf 2 v 2 -vf 2

i;2S or = 2gs
where a = average deceleration, ft/sec2
G a average deceleration, G
vO a initial velocity, ft/sec
vf = final velocity, ft/sec
g = acceleration due to gravity, 32.2 ft/sec2
S = total displacement of the point of the aircraft

with respect to the ground, ft
It can be seen from the equation that t ' magnitude of the de-
celeration is inversely proportional to the stopping distance.
In the ca!- of a rigid structure impacting a nonyielding sur-
"face, the deceleration would be infinite. Some crushing of
structure and soil educes or attenuates the deceleration to
finite levels. Often, however, there is insufficient crush-
ing to attenuate deceleration magnitudes to human tolerance
levels. Tolerable levels can be achieved by increasing the
stopping distance. The extra stopping distance may be pro-
vided by using: (I) additional crushable airframe structure,
(2) energy-absorbing landing gear, (3) a seat design that
possesses an energy-absorption mechanism(s) (load-limiting or
controlled seat collapse), or (4) a combination of methods (1),
(2), and (3).
The energy-absorption capability of a seat structure is also of

considerable importance in evaluating the seat dynamic strength.
Due to extension of the restraint harness, compressibility of
the soft human tissue under the harness, penetration into the
seat cushion, and relative movement of body parts, the occu-
pant's center of gravity acquires a velocity relative to the
airframe during an abrupt deceleration.
Depending upon the magnitude and duration uf the deceleration
pulse, as well as the nature of the connection between the oc-
cupant and the seat structure, '-he maximum relative velocity
may be large. The seat structure, in order to perform its in-
tended retention function, must then either M) possess the
capability of sustaining the maximum inertial force imposed by
68
the deceleration of the occupant and the seat without collapse,
"or (2) possess sufficient energy-absorption capacity to reduce
the occupant's relative velocity to zero before structural fail-
ure occurs. The first alternative may result in an excessive
strength requirement because the input pulse shape and elastic-
ity of the restraint system and cushion can result in signifi-
cant dynamic overshoot. Computer simulation and experimental
observation have shown that overshoot factors range from 1.2
to 2.0, necessitating a seat design strength requirement of
24 G to 40 G to accommodate an input floor pulse of 20 G.
The second alternative of using collapse behavior (load limit-
ing) appears to offer the more practical approach to seat de-
sign. With this option, the seat structure would begin plastic
deformation when the acceleration of the occupant and seat mass
reaches a level corresponding to the critical structural load.
The seat must absorb enough energy without failure to stop the
motion of the occupant relative to the aircraft. Of course,
this energy must be absorbed at force levels within human tol-
erance limits to provide the intended protective function.
In an attempt to eliminate common misconceptions regarding the
role of energy-absorbing seats, a few introductory comments
are made:
* The seat energy-absorbing system does not absorb all
the -gy associated with the impactce--nocty. The
ser
mo g aeriences
.the energy theistotal velocity
absorbed change; however,
by deforming earth,
stroKing landing gear, and deforming structure.
o The absorption of energy by the above processes pro-
duces the triangular-shaped deceleration versus time
pulse used as the design input to the seat.
* The seat energy-absorbing stroke simply lengthens
the stopping distance of the occupant by allowing
energy-absorbing stroke of the seat to occur as the
other energy-absorbing processes are nearing comple-
tion. In a crash in which the aircraft comes to
rest in the major impactr-much of the seat stroke |
can occur after complete deceleration of the aircraft
fuselage. Thus, after the fuselage stops, the seat
may continue to stroke until its kinetic energy has
been exhausted.
* Disregarding dynamic response differences, the same
stroke distance is required to decelerate any mass at
a given deceleration magnitude. Therefore, lighter
people do not require shorter strokes than heavier
69
people for the same deceleration magnitudes.
course, loads required to decelerate occupants Of
of
different weights at equal deceleration magnitudes
must vary with occupant weight.
* Consideration of the first comment explains why it

is detrimental to allow slack to develop in the re-
straint system or seat attachments. If the occupant
is allowed to continue to move with little or no
restraint through any significant portion of the
energy-absorbing process anywhere in the system (not
just in the seat and restraint system), a great deal
more stroke or, a much higher load, will be required
to decelerate the occupant. If the occupant moves
with little restriction until the fuselage stops mov-
ing, the occupant will then require the same stop-
ping distance as the fuselage to experience the same
G loads as the fuselage. Since this stroke is not
available, the loads would be high.
Aside from the seat structure, there are other areas within
the aircraft where energy absorption may find application.
Protective padding, generally a plastic foam, should be used
where structure is likely to be impacted by the occupant, par-
ticularly where head impact is likely. Deforming structure
such as sheet metal behind the foam also is helpful in such
items as instrument panels, glare screens, etc. Characteris-
tics that should aid in the selection of foams for such appli-
cations are discussed in Section 10.9. Also, energy-absorbing
webbing for restraint systems and litters is discussed in Sec-
tion 7.4.4.
4.7.2 Principle of Energy Absorption - Illustration
As an example of the energy-absorption allocations, rewrite
equation (1) for stopping distance as followsi
SNVo02 _ vf 2
s 0 29- (2i
Assuming that v a 42 ft/sec, ve 0, and the average decelera-

tion produced b? deforming terrAin, flattening tires, stroking
energy-absorbing gear, and crushing fuselage is 10 G:
422
(2)(32.2)(l0) " 2.73 ft - 32.87 in.
70
This stroke is 2.73 times the minimum required for the seat;
however, the loads are well within human tolerance limits. If
the entire cumulative stroke could be accomplished et 11.5 G,
which is assumed to produce a deceleration environment toler-
able to humans in this direction, the total distance is
422
S (2)(32.2)(il.5) a 2.38 ft a 28.60 in.
Obviously, 28.60 in. of stroke is impractical for a seat, so

the crash energy-absorption function must be a combination of
energy-absorbing landing gear, crushable airframe structure,
and seat energy absorption. The following example illustrates
how the seat and airframe (including the landing gear) combine
to limit decelerative loading of the occupant, assuming rigid
body mechanics, a triangular deceleration input pulse, and a
seat energy absorber load-deflection curve with the same rise
time as the input pulse and a constant limit load.
The triangular deceleration-time plot is an assumed, idealized

input to the system. In actual practice, the dynamic response
of the system as measured on any individual component does not
match this form because of the differing dynamic properties of
the components as discussed in Sectivn 4.7.3.2. The displace-
ment of the seat/occupant system relative
computed using the following notation: to the airframe is
Let Gm a maximum airframe deceleration in the vicinity of
mthe seat attachment, G
GL a maximum seat/occupant system deceleration, G
K - GL/Gm, tL/tm (limited to 0.5 or less)

t a time at maximum airframe deceleration (one-half in-
put pulse duration), sec
-'tL - time to reach-maximum system deceleration, sec
t a time, sec
v a velocity, ft/sec
va = velocity of airframe at any time t, ft/sec

vs x velocity of seat/occupant system, ft/sec
.'
71
r*
VL common airframe and system velocity at t t
ft/sec
vo - initial impact velocity, ft/sec
g s acceleration due to gravity, ft/sec2
2
a - airframe acceleration, ft/sec
a. = seat/occupant acceleration, ft/sec2
S = displacement, ft
Sa - airframe displacement, ft
S8 a seat/occupant system displacement, ft
The airframe acceleration in the interval 0 : t • tm is given
by
a = -Ggt/tm (3)
where the minus sign indicates a deceleration. The velocity

during the same interval, starting from an initial value of vo,
can be found by integration of Equation (3)
va v+
a ÷ adt
avO JGmg(L)dt
Ggt 2
'V 0o "2 (4)
The airframe displacement at time t is then
Sa= Vadt
72
i ,•T ]i ,
I-____- ___- ____ ___ ____ ___
*,T -. -,..,i t. n. u ...l, .n . n t
-. - - -- -.------
• • .:",[
!m (V• G m2 dt2
- Gmgtm22rtm
/6 (5)
For the interval tm < t < 2tM, the airframe acceleration is
a - -Grg + Gmg(t - tm)/tm (6)
and the velocity,
Va nv 0 + Iadt
II
t2
Sv0 + Gmg(tm - 2t + 2-t) (7)
30 that, at t = 2 tm
4 tm 2 tm)
va = v 0 + G g(tm - +
M V -GMgtm (8)
Since the peak deceleration G is that required to bring the

aircraft to rest at time 2 tmI*
Va 0 v0 0- G
Gtm
73
--L..-
and
V0 ~Gm tM
The airframe displacement at 2tm is then
-' 2
v 0t M tm/6
+ I v + G9( 2t + t )d
m
2v t~ - G ,gt 2 (10)
Substituting Equation (9) into Equation (10) the total airframe

di~splacemnent is
Sa ~VtM *Ggtm2 (
omm0
The acceleration of the seat/occupant system matches that of

the airframe for 0 5 t i•t ~where t is determined by the
limiting deceleration G 1IýsingEquations(4an-5,th
veloityanddisplacement of the seat at. tLcan be-found as
follows:
GmgtL
5e 0 -2t I
74
"",GmgtL
VotL (12)
For t S
systak t
comes t where tff is
ýdrest the time when the seat/occupant
as a -G-g (13)
and the system velocity in this interval is given by
V GmgtL2 jasdt
Gm tL 2
Vo - 2 t. - GLg(t - tL) (14)
Since v = 0 at t = tf, Equation (14) can be used to find the

final tfme tf
0V GmgtL2 G
Ou•vO - 2t - GLgtf * GLgtL
Introducing the variable
K - GL/Gm - tL/tm (15)
75
--- --- - --- -.

-- -- -- -- .. -.-- -
the time tf can be written
tf 1( K (16)
Using Equation 415) to substitute for t and GL in Equations

(12) and (14), the seat/occupant system %isplacement at t f is
found by
G gK 3t 2 fi K2G gtm
Ss =VKtm - m + (vo + K 2 m m KGmgt)dt

m
F K2 K32 Kt -2
s mGmgi + 2-)ttf -6
=mgtm + K) 17
The stroke distance required by the seat is the displacement

of Equation (17) less that of the airframe, which is given by
Equation (11):
-t
G 2. 1 K K3
Stroke, S = m tm 4 +18) - j-l)
The above result also can be obtained geometrically, using the

velocity and displacement curves shown in Figure 17. Pcr fur-
ther clarification, this somewhat simpler procedure is pre-
sented below. -
The velocity of the airframe at time t is equal to the initial

velocity plus the change in velocity from t 0 to t -t,
vt -vo + (a). 19 !
£ 0 a2 (19)
76
. . . . . - .
-. ______________________ __________ *:.-, i
GM Input to seat
V ~energy- absorption
system
~Seat/occupant
0 system deceler-
41 ation
o,GL
ý- t __Oý2t
-Airframe velocityi
>4
Seat/occupant
~system velocity
0 'r1V II
tL
L
Total seat/occupant
2t
-
-I
Stroke of seat
4J system displacement -- energy-absorbing
44 system
Airframe displacement
Time Wt, sec
Figure 17. Deceleration-time, velocity-time, and distance-

time curves used in analysis of seat/occupant
displacement with respect to the airframe.I
* 77
Substituting the value of a from Equation (3),
vV
t S~ y 0 + [G~]t
t (2
fGmgt
m1220
V Vt [ m: (20)
Now, assuming that the airframe comes to rest so that v(t -

2 tm) = 0, the total velocity change can be said to equal
the
initial velocity. Since this corresponds to the total area
under the deceleration versus time curve,
Vo Gmgtm (21)
Substituting Equation (21) into Equation (20) yields
Gg t 22
vt = Ggt -GMg
2 tm
t =Gmm (22)
Using Equation 142), we can now compute the common velocity of

the airframe and the system at time tLI
L Gmgtm 2tm - - 23)
The change in velocity in the time interval tL is '
6v • vo - VL (24)
"i
78 "
Substituting Equations (21) and (23) into Equation (24) yields
GmgtL2
Av Gmgtm - (Gmgtm - 2tmL
2
GmgtL
2t m (25)
The areas of interest in the velocity-versus-time graph in Fig-

* ure 17 can be calculated now using the relationships just de-
rived together with geometrical considerations.
Recognizing that the curve describing the velocity of the air-
frame consists of the two parabolic segments shown in Fig-
ure 18, connected at time t , it can be seen that A is the
area under a parabola of bale t. and height tv. Thhrefore,
AI =•2 AGmtL2) gtL2

2t. / tL -- 3tm (26)
VO
4o
0
2 tm
tm
Time
Figure 18. Airframe velocity-time curve.
79
-. 4 m -. - •
Area Al is simply a rectangle of base tL and height vL, so
that
(G GigtL2 t G gtL (
ttL gtm /2 GmgtLtm " (27)Lt
Si
Since the system is undergoing a constant deceleration begin-

ning at tL, area AV can be represented by the relationship
22
Av VL
•j i
Substituting from Equation (23) and noting that GL - KGm,
2 Gg
=~
A(v n G tL2)
2KGm-g (28)
The area sought as representing the energy-absorption stroke

of the seat is A In order to solve for this area, A
must first be established. A can be determined by n ig
that, due to the triangular sAie of the acceleration pulse,
the airframe velocity curve consists of two parabolic segments
meeting at the midpoint of the curve, as shown in Figure 18.
If a straight line is constructed joining v and 2t , the two
shaded areas bounded by the curve and the l0e can shown to
be equal since they are both areas between parabolic curves
described by the same basic equation and a secant. The total
area under the curve can then be said to be the same as the
area of the triangle formed by the coordinate axes and line
connecting v0 and 2 tm. Therefore, -
A1 + A1 +A 111 = (.-) 2tm = Votm = gt2 (29)
and
ANV = (AI + AIX AV) - (AI + Ail + AIII) (30)
80
Substituting from Equations (26), (27), (28), and (29) yields
G gtL3 GgtL3
A IV 3t + GmgtLtm 2t
r2 1 4-
(G- g Gmgt L2 Gmgtm 2

m2gt t 2 2KG g
Simplification and substitution of Xtm for tL yields
A =S 2 K3
"I"mm 2 1 (31)
which is the same as Equation (18).

I. As an example, consider a triangular pulse representing a
change in velocity of 42 ft/sec with
Gm = 48 G
t = 0.02/ sec
m
K = 1 = 0.24
48
The required stroke is then
0,
•024 1 - '0.4)3 I
Stroke = (48)(3•6)(0.027,2
0 + 2(0.24)
4"-:L20.24 244
= 16.25 in.
Test data show this stroke to be less than that required. Much
of this difference can be attributed to system inefficiencies.
It has been found in tests that an efficiency of approximately
t 81
4. - -. ___ . . ____
. -.- ___ -.-- ___ ___
80 percent can be expected from a rod-bending sled decelera-
tor and a wire-bending seat load limiter (References 31 and
32). Therefore, correcting the calculated distance yields
16.25/0.8 - 20.31 in. It must be realized that 20.31 in. is
probably a valid stroke for systems with little or no friction,
such as ceiling-mounted troop seats. For seats guided by slid-
ing or rolling components, friction adds to the resistive force,
thus producing an apparent increase in efficiency. However, in
general, large frictional resistance is not desirable because
of the variation of the net resistive force and hence occupant
decelerative loading as a function of loading direction. Re-
view of the above indicates that the 12-in. minimum seat stroke
required for the design pulse (used in the above calculations)
is hardly adequate and should not be compromised unless other
provisions are included to reduce the residual energy that the
seat is required to absorb.
4.7.3 Dynamic Response
4.7.3.1 Effective Weight: The .cacept of effective weight

has been used to account for ma4jcs supported by components
other than the stroking portion of the seat; e.g., the seat
occupant's lower legs supported by the floor during vertical
loading. The effective weight of the occupant plus the weight
of the movable portion of the seat is multiplied by the limit-
load factor (G) during calculation of limit loads. The tech-
nique is not completely accurate physically because rigid
bodies cannot adequately simulate the dynamic response of the
real system. Seat designs should be dynamically analyzed and
tested to establish their dynamic response and to demonstrate
that they will provide the desired degree of occupant protec-
tion.
4.7.3.2 Theoretical System Response: A major design factor

influencing the seat response is the movable seat mass. For
very light seats, the gross response of the occupant can be
estimated using the approximate mass of the occupant acting on
the seat (80 perceizt when considering the vertical direction
as discussed later in this chapter). However, when the seat
31. Reilly, M. J., CRASHWORTHY, TROOP SEAT TESTING PROGRAM,
The Boeing Vertol Company; USAAMRDL Technical Report 4
77-13, Eustis Directorate, U. S. Army Air Mobility Re-
search and Development Laboratory, Fort Eustis, Virginia,
August 1977, AD A048975.
32. Reilly, M. J., CRASHWORTHY, HELICOPTER GUNNER SEAT TESTING

PROGRAM, The Boeing Vertol Company; USARTL Technical Re-
port 78-7, U. S. Army Research and Technology Laboratories,
Fort Eustis, Virginia, February 1978, AD A054970.
82
mass increases to values typical of integrally armored crew-
seats, interaction between the mass and spring properties of
the seat and occupant can become significant. The occupant
S~and seat components then realize sharp deceleration excursions,
• i.e., spikes.
The dynamics of the problem are illustrated in Figure 19, which

presents the theoretical response of an integrally armored crew-
seat and occupant to an input crash pulse as calculated by a
digital computer analysis (described in Reference 24) and sum-
marized in Section 4.8.6. The analysis simulates the occupant
by three lumped masses representing the head, chest, and pel-
vis. The cushion and seat are represented by two additional
masses. The five masses are connected by damped springs in
the model.
The response curves for the seat structure, occupant pelvis,

and chest are shown as functions of time for the indicated in-
put excitation. The seat used was an energy-absorbing, in-
tegrally armored model set to stroke at 18 G (18 times the ef-
fective weight of the occupant plus movable seat). The armored
seat bucket weighed 40.6 lb, and the energy absorber provided
a trapezoidal force-versus-deformation characteristic. It can
be seen that the dynamic response of the seat and segments of
the body are not independent of one another and vary as the
model springs load and unload.
Initially, the seat pan deceleration lags the input pulse as
the springs representing the flesh and the cushion as well as
the elastic spring of the seat structure are loaded. The
stroking force of the energy absorber was sized for a deceler-
ation of a particular mass, and -he effective mass is not yet
being applied to the seat structure because of the incomplete
spring compression. Therefore, the seat pan deceleration ex-
ceeds the deceleration required to effect the force necessary
to stroke the energy absorber. The seat pan deceleration ap-
proaches 43 G before the cushion and flesh springs compress to
the point that significant deceleration of the pelvis begins.
As deceleration of the pelvic mass increases, an increasing
reaction force is applied in the downward direction ern the seat
pan. Thp seat pan deceleration decreases from A3 G to approxi-
mately 27 C as the effective mass is increased.
Because the input decelerative loading is still increasing and
the chest inertial load has not yet been applied to the system,
both the seat pan and the pelvic decelerations increase. As
the spring representing the buttocks flesh and cushion bottoms
out, the pelvic deceleration continues to increase, further
83
50 Initial spike m Input pulse (heavy solid)
\- Vertical seat pan (solid)
-00-Vertical dummy chest (dotted)
--..Vertical dummy pelvis (dashed)
40.
,0 I !
econdary spike
30"
I0
20 0
0)
U)I 0/ , /
0 //
o I
10 /0
i0I / /
//
/ 0 /C0/
/ 0
/Initial notch
-10 Secondary notch
0 26 40 60 80 100
Time, msec
Figure 19. Deceleration versus time for various
components of seat and occupant.
(From Reference 24).
loading the seat pan and decreasing its deceleration. It can

be seen that the Beat pan experiences a small acceleration
under the combined loading of tne occupant's pelvis and chest.
As the chest deceleration increases, the decelerations of the

seat pan and the pelvis tend to normalize near the G level cor-
responding to the limit-load factor of the energy-absorbing
system.
84
"In summary, the limit load must be set at a load factor consid-
erably below the tolerable level in order to limit the occu-
pant response to a tolerable level, particularly for seats of
high movable mass.
4.7.3.3 Empirical System Response: In the past few years,

several programs were conducted in which crashworthy armored
seats were dynamically tested (References 16, 24, and 33
through 35). These programs included drop tests in which the
seats' response to decelerative loading in the vertical direc-
tion was measured. Two types of tests were conducted: in the
first, the impact velocity vector was parallel, but in the op-
posite direction, to the loading and along the vertical axis of
the seat, the yaw axis related to the aircraft (upward and per-
pendicular to the aircraft longitudinal axis), and in the sec-
ond, the seat was pitched forward 30 degrees and rolled 10 de-
grees relative to the aircraft axis system. These dynamic
tests demonstrated a characteristic deceleration-time history
very similar to that theoretically predicted (see Figure 201.
The characteristic shape has been evident in essentially all
tests to date; however, the magnitudes of the spikes and
notches vary. The characteristic shape of the seat pan
deceleration-versus-time history includes a high initial spike
followed by a deep notch that sometimes passes through zero
and actually becomes an acceleration rather than a decelera-
tion. This notch is followed by a second high spike followed
by various waveforms, damping out and usually centering around
the load factor used in sizing the energy-absorption system
loads.
33. Mazelsky, B., A CRASHWORTHY ARMORED PILOT SEAT FOR HELI-

COPTERS, ARA Inc.1 USAAVSCOM Technical Report 73-34 and
Report No. NADC-74018-40, Joint Report issued by Naval
Air Systems Command, Washington, D. C., and U. S. Army
Aviation Systems Command, St. Louis, Missouri, January
1974, AD A007551.
34. -Domzalski, L. P.rjand Singley, G. T., III, JOINT APMY/NAVY
TEST PROGRAM FOR UTTAS SEATING SYSTEMS, NADC-79229-60, -
Naval Air Development Center, Warminster, Pennsylvania, to
be published.
35. Dummer, R. J., QUALIFICATION TEST REPORT 613-1787 COOL-
QUALIFICATION TESTING OF ARMORED CRASHWORTHY AIRCREW SEAT
RA-305 2 5 -1 (FOR SIKORSKY AIRCRAFT CONTRACT 576344), U. S.
Army Contract No. DAAJ01-77-C-0001, Norton Company, Indus-
trial Ceramics Division, Worcester, Massachusetts, revised
January 1979.
85
00
to 0i
0 00
* L)
00 0
0)
>1 0
4)0 0
0 r0
D'U0TIV28TOOQ
IjI
The explanation of the characteristic waveform is associated

• •
¶ with the inherent dynamic response of the seating system and
its occupant. As explained previously, total coupling of the
seat and its occupant is not achieveO since the occupant con-
sists of masses connected by body members, such as the spinal
column and neck, which are not rigid.
4
Further, because the dummy is seated on simulated buttocks
flesh and a comfort cushion, it is not rigidly connected to
the seat pan. Since the energy-absorbing mechanism of the seat
must be set for a given load (calculated by multiplying the
effective weight of the occupant and movable part of the seat
by the desired limit-load factor of 11.5 G), the actual decel-
eration measured on the seat pan will vary inversely to the
Scoupled weight (ht) according to the relationship a = FP/wt
where a represents the deceleration in G units, F, the load in
pounds resisting the stroke of the seat, and w , the coupled
weight in pounds. The term coupled, as used here, simply in-
dicates that the applicable connecting springs are compressed
sufficiently to result in the body segment being decelerated
in phase at approximately the same rate as the seat pan (as
would a rigidly attached mass).
A deceleration applied to the seat pan initially decelerates

the movable seat mass only. Consequently, deceleration of the
seat pan reaches a large magnitude as indicated by the initial
spike in Figure 20. As the cushion and the simulated flesh on
the buttocks compress, the deceleration of the pelvic mass in-
creases. As the spinal column compresses, the deceleration
of the chest increases. The deceleration of these masses in-
creases as a result of the increased load in the connecting
members. There is a spring constant involved when the connect-
ing members are represented by springs. This constant defines
the relationship between deflection and load; i.e., the further
a spring is compressed, the larger the load required to com-
press it. The body segments react in the same way. There-
fore, the greater the compression, the higher the load, and
the higher the deceleration of the body segments.
I
As an illustration, consider Figures 21, 22, and 23. Comparing
the figure of a seat occupant with a system of springs and
masses, it can be seen that when the initial deceleration of
the seat pan commences, the springs in the body are not com-
pressed. The body and seat system have been under a 0-G envi-
ronment during the drop. Therefore, the springs are totally
unloaded as illustrated in Figure 21. This simply means that
since the springs are not compressed when the deceleration
first commences, large loads cannot immediately be applied to
the body segments. As the pulse continues, the body segments
* 87
- ...--.---
-----.
- -- . .-.--- i
- 0 -Head
Figure 21. Initial condition, no load.
area
is omprespodspingtodclrinla,
a4 Seat
Crash load
Figure 23. Onigh-fdeceleration load wpringn

coprelsedc
arasrepodngtodceert88lod
bu4- pertroadhedaent
continue to move undur the resistive load of the partially com-
pressed springs, thus decelerating more slowly than the seat
and building up a velocity relative to the seat pan. Even-
tually the velocities of the body segments and the seat pan
must all approach a common value. This usually occurs later in
the sequence, after the secondary spike. In the interval, the
deceleration of the seat pan responds as a function of energy
absorber force, input pulse, seat and dummy weight, and spring
and damping characteristics.
Initially, the seet pan deceleration reaches a high value (ini-
tial spike). This ccc'"r because the resistive force in the
energy-absorbing systn tq sot at R given value considering
the weight of the movai.'. aortion o' the seat and the occupant.
The seat pan is decelerat.ed initialJy at a magnitude consistent
with the force of the energy-absorAing mechanism divided by
only the weight of the movable part of the seat, which is con-
siderably less than the design weight (the weight of the mov-
able portion of the seat and the effective weight of the occu-
pant). Thus, it is expected that the magnitude of initial seat
pan deceleration will always exceed the limit-load factor for
which the composite energy-absorbing system was designed.
Eventually, the cushion and buttocks springs are compressed,

and the pelvic mass loads into the seat pan (see Figure 22).
The increase of the coupled mass decreases the deceleration of L
the seat pan from its initial peak. The seat pan deceleration
then decreases drastically as evidenced by the initial notch
in the deceleration-time history. At times, when the decelera-
tion actually turns into an acceleration, it simply means that
the mass of the pelvis is receiving a relatively high decelera-
tion and the reaction load is high enough to accelerate the
seat pan towards the aircraft floor. It is apparent that the
magnitude of this notch is a strong function of the spring rate
of the seat. Since the spine normally is still not compressed
significantly, it is not carrying high loads. This is evi-
denced by the small decelerations measured in the chest, which
is being supported by the spine.
Since this is a dynamically loaded spring system, the springs

asoociated with the buttocks and the cushion can overshoot as
they bottom out during the sequence and then unload again. The
unloading permits the seat pan deceleration to rise again to
the secondary spike on the trace. As the pelvis unloads, the
reaction load on the seat pan decreases and the seat pan decel-
eration spike can be extremely high. Note that the high decel-
eration of the seat pan does not necessarily correlate with the
high deceleration of the pelvis or chest. From the data re-
viewed, both analytical and empirical, it is generally the op-
posite; i.e., the unloading of the pelvis and/or the chest pro-
duces the spike in the seat pan deceleration.
89
il l-IN.i ..11I-ll __________ - . ...

As the cushion and buttocks again load up and the pelvis de-
celeration increases, the high seat pan deceleration of the
second spike is decreased. Also, the two characteristic decel-
eration spikes are usually followed by an increased compressive
load in the spine and a buildup of deceleration of the chest.
Eventually, the phasing of the decelerations of the various
system segments begins to converge toward the average load fac-
tor for which the limit load of the energy-absorbing system
was designed, as illustrated in Figure 23.
It is informative to note (see Figure 20), that the peak decel-
erations of the seat pan do not necessarily coincide with peak
decelerations of the human occupant and, thus, are not neces-
sarily hazardous to occupant safety. The Eiband human toler-
ance data of Section 4.3.3, Volume II, repeated here in Fig-
ure 24 for ease of reference, do not present information on the
seat pan deceleration excursions from the average, or uniform
acceleration experienced by the vehicle, and are therefore not
informative on the subject. Again, this is an area presently
being investigated under U. S. Army sponsorship. These cri-
teria will be updated when additional data are available.
The response phenomena described above comprise the predictable
response of the occupant/seat system to the input pulse. The
high decelerations measured on the seat pan are not necessarily
correlated with high decelerations of the occupantl however,
this does not imply that the seat will provide the required
protection. The entire deceleration history to which the occu-
pant is exposed must be considered. As pointed out, low decel-
erations of the seat pan may be accompanied, and caused by,
high loads imposed on the occupant. Thus, it is imperative
that additional information relative to human tolerance to
transient loading in the vertical direction be obtained and
that the criteria for designing vertical energy attenuating
systems for seats be refined and made more comprehensive. Sev-
eral programs are now underway to expand knowledge in this area.
As the data become available, criteria contained in this guide
will be updated,
4.7.3.4 Tailoring of Energy Absorber: Results of analyses
conducted under a U. S. Navy-sponsored program (Reference 36)
indicated that the force-versus-deformation characteristic of
the energy-absorbing system can be shaped to enable more effi-
cient use of the stroke distance available. As yet, test data
have not substantiated the analyses. However, proper tailoring
36. Carr, R. W., and Phillips, N. S., DEFINITION OF DESIGN
CRITERIA FOR ENERGY ABSORPTION SYSTEMS, Beta Industries
Incorporated; Report No. NADC-AC-7007, Naval Air Develop-
ment Center, Warminster, Pennsylvania, 11 June 1970,
AD 871040.
90
- -~ -. 4
4-
4J C
1. t41
914
of the waveform should produce the potential for maximization
of the efficiency of a particular system. Research efforts
considering this approach are presently being sponsored by the
U. S. Army. Results should be available by 1981.
4.7.3.5 Adjustable Multiple Limit-Load Devices: Since stroke

in aircraft cockpits, it is extremely important to maximize
the efficient use of the available distance. Energy absorbers

that stroke at a given limit load are sized for the effective
'.,iygit -f the 50th-percentile occupant (by weight). This Im-
plies that the majority of occupants will stroke at or near
the optimum load. Heavier people, however, will stroke at
a lower deceleration level and will require a longer stroke
than the average for which the system was designed, while the
lighter occupant will stroke at a higher deceleration and use
less of the available stroke. Obviously, the energy-absorption
load settings are not optimum for occupants whose weights lie
in the tails of the occupant weight distribution. If adjust-
ment of the limit load for the weight of the specific occupant
is possible, then the system can be optimized for the total
occupant weight distribution.
Variable limit-load energy absorbers can be controlled either
passively (requiring no action by the occupant) or actively
(requiring a conscious action by the occupant). The lassive
device is more complicated and would require a considerably
more sophisticated system to control. It would have to be a
force/time integrating system, since the limit load of the en-
ergy absorber could not be a function of the dynamic loading
of the seat associated with occupants simply sitting down hard.
There is no doubt that this type of device could be developed;
but the cost and weight may be prohibitive, while an actively
controlled device should be neither overly complex nor costly.
To achieve most of the advantages offered by the system, the
load would not have to be infinitely adjustable but could be
applied in several increments. The occupant would simply turn
a dial or move a lever to a weight range best fitting his own
weight.
Previous studies (Reference 24) have indicated that a total
excursion of approximately 6 G results-from using a single
limit load set for the 95th-percentile occupant weight. The
6-G excursion could be essentially eliminated with either type
of variable-load energy absorber. This would allow decelera-
tions of all occupants to be nearly identical and enable use
of essentially the same stroke distance for the same decelera-
tive loading and input crash severity for all occupant weights.
92
4.i
Variable limit-load energy absorbers should therefore be incor-.
porated in the vertical direction in all new crashworthy seat-
ing systems and retrofit should be considered for seating sys-
tems now in use that include stroking capabilities together
with replaceable energy absorbers.
4.8 COMPUTERIZED METHODS OF ANALYSIS

4.8.1 General
Prediction of occupant and seat structure response to dynamic
loading presents a complex engineering problem. The use of
computer-aided design in these cases is essential, since the
dynamic interaction of the occupant and the seat/restraint sys-
tem is much too complex for analysis by manual techniques.
A number of dynamic models of the human body have been devel-
oped for use in crash survivability analysis. These models
vary in complexity and possess from I to 40 degrees of freedom
(References 37 through 54). One-dimensional models have been
37. Bacchetti, A. C., and Maltha, J., MADYMO - A GENERAL PUR-
POSE MATHEMATICAL DYNAMICAL MODEL FOR CRASH VICTIM SIMULA-
TION, Report No. 753012-C, Instituut voor Wegtransportmid-
delen, Netherlands 1978.
38. Bartz, J. L., DEVELOPMENT AND VALIDATION OF A COMPUTER
SIMULATION OF A CRASH VICTIM IN THREE DIMENSIONS, Pro-
ceedings, Sixteenth Stapp Car Crash Conference, Society
of Automotive Engineers, Inc., New York, 1972, pp. 105-127.
39. Danforth, J. P., and Randall, C. D., MODIFIED ROS OCCUPANT
DYNAMICS SIMULATION USER MANUAL, Publication No. GMR-1254,
General Motors Corporation Research Laboratory, Warren,
Michigan, 1972.
40. Fleck, J. T., Butler, F. E., and Vogel, S. L., AN IMPROVED
THREE DIMENSIONAL COMPUTER SIMULATION OF MOTOR VEHICLE
CRASH VICTIMS, Final Technical Report No. ZO-5180-L-l (in
four volumes), Calspan Corporation, Buffalo, New York,
- 1974.
41. Furosho, H., Yokoya, K., and Fujiki, S., ANALYSIS OF OCCU-
PANT MOVEMENTS IN REAR-END COLLISION, Paper No. 13, in
Safety Research Tour in the U.S.A. from the Viewpoint of
Vehicle Dynamics, 1969.
42. Furosho, H., and Yokoya, K., ANALYSIS OF OCCUPANT'S MOVE-
MENT IN HEAD-ON COLLISION, Transactions of the Society of
Automotive Engineers of Japan, No. 1, Tokyo, Japan, 1970,
Spp.
145-155.
used in prediction of human body response to an ejection seat
firing (References 55 through 57), which, if the body is tightly
restrained, can be approximated as a one-dimensional phenomenon.
However, a vehicle crash generally involves a horizontal compo-
nent of deceleration, which forces rotation of body segments
with respect to each other. If no lateral component of decel-
eration is present, a two-dimensional model will suffice, pro-
vided the restraint system is symmetrical. However, lateral
43. Glancy, J. J., and Larsen, S. E., USERS GUIDE FOR PROGRAM
SIMULA, Report Inc.,
TDR No. Dynamic Science,
72-23, Arizona, 1972. Division of
Ultrasystems, Phoenix,
44. Huston, R. L., Hessel, R., and Passerello, C., A THREE-

DIMENSIONAL VEHICLE-MAN MODEL FOR COLLISION AND HIGH AC-
CELERATION STUDIES, Paper No. 740275, presented at Auto-
mobile Engineering Conference, Society of Automotive
Engineers, Inc., Detroit, Michigan, 25 February - 1 March
1974.
45. McHenxy, R. R., ANALYSIS OF THE DYNAMICS OF AUTOMOBILE
PASSENGER-RESTRAINT SYSTEMS, Proceedings, Seventh Stapp
Car Conference, Society of Automotive Engineers, Inc.,
New York, 1963, pp. 207-249.
46. Robbins, D. H., THREE-DIMENSIONAL SIMULATION OF ADVANCED
AUTOMOTIVE RESTRAINT SYSTEMS, Paper No. 700421, In 1970
International Automotive Safety Conference Compendium
P-30, Society of Automotive Engineers, Inc., New York,
97 0.
47. Robbins, D. H., Bennett, R. 0., Jr., and Bowman, B. M.,

USER-ORIENTED MATHEMATICAL CRASH VICTIM SIMULATOR, Pro-
ceedings, Sixteenth Stapp Car Crash Conference, Society of
Automotive Engineers, Inc., New York, 1972, pp. 128-148.
48. Robbins, D. H., Bennett, R. 0., and Roberts, R. L., HSRI
TWO-DIMENSIONAL CRASH VICTIM SIMULATOR: ANALYSIS, VERI-
FICATION, AND USER'S MANUAL, Final Report, Report No.
HSRI-Bio-M-70-8, Highway Safety Research Institute, Uni-
versity of Michigan, Ann Arbor, Michigan, 1970.
49. Robbins, D. H., Bowman, B. M., and Bennett, R. 0., THE

MVMA TWO-DIMENSIONAL CRASH VICTIM SIMULATIONS, Proceed-
ings, Eighteenth Stapp Car Crash Conference, Society of
50. Segal, D. J., REVISED COMPUTER SIMULATION OF THE AUTOMO-

BILE CRASH VICTIM, Report No. VJ-2759-V-2, Cornell Aero-
nautical Laboratory,, Inc., Buffalo, New York, 1971.
94
loading is common in helicopter accidents. Also, the diagonal
S• shoulder belt used in some troop restraints is asymmetrical
and may cause lateral motion.of the occupant even in the ab-
sence of a lateral deceleration. Therefore, a model that is
generally useful in restraint system evaluation must be capable
of predicting three-dimensional motion, and several three-
dimensional kinematic models, made up of interconnected rigid
links, have been developed (References 38, 44, 47 and 53).
Subsequent sections of this chapter describe the models of pos-
sible use to designers of seats and restraint systems.
4.8.2 Program SOM-LA
In 1972, the FAA initiated a program to provide a practical
engineering tool for use in the design and evaluation of seats
and restraint systems for light aircraft. This program in-
corporated a dynamic model of the human body combined with a
51. Segal, 0. J., and McHenry, R. R., COMPUTER SIMULATION OF
AUTOMOBILE CRASH VICTIM - REVISION, Report No. VJ-2492-I,
Cornell Aeronautical Laboratory, Inc., Buffalo, New York,
1967.
52. Twigg, D. W., and Karnes, R. N., PROMETHEUS, A USER-
ORIENTED PROGRAM FOR HUMAN CRASH DYNAMICS, Boeing Computer
Services, Report No. BCS 40038, Department of the Navy,
Office of Naval Research, Waahington, D. C., 1974.
53. Young, R. D., THREE-DIMENSIONAL MATHEMATICAL MODEL OF AN

AUTOMOBILS PASSENGER, Research Report 140-2, Texas Trans-
Sportation institute, College Station, Texas, 1970.
54. Young, R. D., Ross, H. E., and Lammert, W. F., SIMULATION
OF THE PEDESTRIAN DURING VEHICLE IMPACT, Paper No. 27,
Proceedings, Third International Congress on Automotive
Safety, Vol. II, Society of Automotive Engineers, Inc.,
New York, 1974.
55. Kroeger, W. J., INTERNAL VIBRATIONS EXCITED IN THE OPERA-

TION OF PERSONNEL EMERGENCY ESCAPE CATAPULTS, Memorandum
Report 340, Frankfort Arsenal Laboratory Division, Phila-
delphia, Pennsylvania, 1946.
56. Latham, W. F., A STT')Y IN BODY BALLISTICS: SEAT EJEC-
TION, Proceedings of Royal Society, London, England, 1957,
B 147: 121-139.
57. Stech, E. L., and Payne, P. R., DYNAMIC MODELS OF THE HU-
MAN BODY, Frost Engineering Development Corp., AMRL Tech-
nical Report 66-157, Aerospace Medical Research Laboratory,
Wright-Patterson Air Force Base, Ohio, 1969.
95
finite element model of the seat structure. The program is
intended to provide the designer with a tool with which he
can analyze the structural elements of the seat as well as
evaluate the dynamic response of the occupant during a crash.
The digital computer program based on this model is called
SOM-LA (Seat/Occupant Model: Light Aircraft).
The aircraft occupant is modeled by twelve rigid mass segments
with rotational springs and dampers at the joints. The seg-
ments represent the head, neck, upper torso, lower torso, upper
arms, forearms, thighs, and lower legs, as shown in Figure 25.
Each of the torso joints possesses three rotational degrees of
freedom. The elbows, the knees, and the head-neck joint are
hinge-type connections, each of which is allowed one additional
degree of freedom. In total, the occupant model possesses 29
degrees of freedom.
Rotations at the body joints are resisted by torsional springs
and dampers, whose characteristics depend on user selection of
human or dummy occupant.
External forces are applied to the body segments by the seat
cushions, the floor, the belt restraint system, and an optional
inflatable restraint. The four available restraint system con-
figurations consist of a lap belt alone or combined with a
single diagonal belt over either shoulder, or a double shoul-
der belt. A lap belt tiedown strap may be used with the double
shoulder belt system. The restraint loads are transmitted to
I
the occupant through ellipsoidal surfaces to the upper and
Slower torso segments; and, the points of application depend on
current belt geometry. The capability of the belts to move
relative to the torso surfaces allows simulation of submarining
under the lap belt as well as prediction of the lateral motion
which may result with a single diagonal shoulder belt.
For calculation of external forces exerted on the occupant by
the seat cushions and restraint system, and for prediction of
impact between the occupant and the aircraft interior, 24 sur-
faces are defined on the body. These surfaces are ellipsoids,
spheres,- and cylinders, as shown in Figure 26. .
The user may select either a finite element model of the seat
structure or a rigid seat representation. The latter can in-
clude nonlinear translational and rotational stiffness elements
for simulation of an energy-absorbing seat as shown in Fig-
ure 27. The bucket is assumed rigid, and frame elasticity is
modeled by the torsional spring.
The program has been run on CDC 6600, Univac 1108, and IBM 370
systems. Input data include force-deflection information for
96
B.-7
j •
R#$.
4
• -. . e -- • ••.....
v-
I12
SSegment mass center
e25. Twel -egm t occupant model.
97
1
20
4 241 7 1 13 11
Figure 26. Occupant model contact . ; KI

surfaces.
KT: Translational
stiffness
(energy absorber)
KRIt: Rotational stittngg
8
I
KT
Rigid
frame
R Rigid bucket
Figure 27. SOM-LA energy-absorbing

seat model.
98
the cushions and belts; crash conditions, in terms of intial
velocity and attitude and time variations of six acceleration
components; occupant description; seat design data; and, if
the prediction of impact with the aircraft interior is desired,
a description of the cabin surfaces. Output data include time
histories of occupant segment positions, velocities, and ac-
celerations; restraint system loads; seat deflections and
forces; details of contact between the occupant and the air-
craft interior; and several measures of injury severity. The
injury criteria used in the program are all computed from seg-
ment accelerations. The dynamic response index (DRI) provides
an indication of the probability of spinal injury due to a ver-
tical acceleration parallel to the spine. The Severity Index
is calculated for the chest and head, and the Head injury Cri-
terion of Federal Motor Vehicle Safety Standard 208 is also
computed.
This original model is described in Reference 58. Since its

publication, a number of modifications to the model have been
made to improve simulation quality and to provide increased
capability and additional desirable output. In that interim,
an extensive testing program was initiated by the FAA Civil
Aeromedical Institute (CAMI) to provide data for validation of
the model. Work on model improvement and validation is contin-
uing, as described most recently in Reference 59.
4.8.3 Calspan Corporation - CVS
Probably the most sophisticated biomechanical model of the hu-

man body intended for crash simulation is the Calspan Corpora-
tion Crash Victim Simulator (CVS). Originally reported on in
1972 (Reference 38), the program includes a body dynamics model
with 40 degress of freedom and a contact model that generates
forces from contact with vehicle surfaces. The extensive vali-
dation effort has included the following experiments:
58.- Laananen, D. H., DEVELOPMENT-OF A SCIENTIFIC BASIS FOR

ANALYSIS OF AIRCRAFT SEATING SYSTEMS, Dynamic Science,
SDivision of Ultrasystems, Inc.; Report No. FAA-RD-74-130,
U. S. Department of Transportation, Federal Aviation Ad-
ministration, Washington, D. C., 1975, AD A004306.
59. Chandler, R. F., and Laananen, D. H., SEAT/OCCUPANT CRASH
DYNAMIC ANALYSIS VALIDATION TEST PROGRAM, Paper No. 790590,
presented at Business Aircraft Meeting, Society of Automo-
tive Engineers, Inc., Wichita, Kansas, April 1979.
99
- -. *.... . .. • t .. a-.--.
* Static bench tests with a spherical membrane and
spherical contact surfaces to validate the air bag
shape and contact force algorithm.
e Pendulum tests with a dummy torso form restrained
and decelerated with an air bag to further validate
* Tests with instrumented anthropomorphic dummies on

an impact sled at 20 and 30 mi/h with both belt and
air bag restraints, in which both planar and non-
planar dummy responses were produced.
* A head-on, laterally offset, car-to-car crash test,
with the primary vehicle containing two instrumented
anthropomorphic dummies.
A graphics display model provides rather sophisticated three-
dimensional views of occupant response, as shown in Figure 28.
Present capabilities of the program, a user manual, and a de-
"scription of its validation are presented in Reference 40.
Figure 28. CVS graphics display model.
4.8.4 PROMETHEUS
In 1572, Boeing Computer Services began work on modification of
a two-dimensional occupant model called SIMULA, which had been
developed earlier by Dynamic Science, Inc. and Arizona State
University. Their final product, which includes interactive,
user-oriented capabilities, is called PROMETHEUS (Reference 52).
100
- - - ' .- -- -- --- -• - "k "

W_'
PROMETHEUS simulates a crash victim with either a two-

dimensional, seven-link, side-facing mathematical model, shown
in Figure 29(a), restrained by a seat belt and shoulder har-
ness, or an eleven-link, forward-facing, unrestrained model,
shown in Figure 29(b). A nonlinear finite element model of
the impacting structure is incorporated. A new, fast differ-
ential equation solver was developed for the program to effi-
ciently compute the transient response of the finite element
vehicle structure and rigid-link occupant in a crash situa-
tion. The program is an interactive, user-controlled system
designed for the rapid analysis/data edit/reanalysis cycles
necessary for efficient parametric studies. PROMETHEUS input
aids include free-field data input and an on-line data edit
capability. Output provides user-selected time history and
occupant configuration plots, as well as abbreviated output
lists for rapid scan of results. The program operates on the
CDC 6600 computer in either a batch or an interactive mode.
;Joint 9 Joint
Point mass IDistributed mass
(a) Side-facing model (b) Forward-facinq model.
Figure 29. PROMETHEUS occupant model.
101
a.-n.n n- n I I I I I l l l ll-
4.8.5 Air Force Head-Spine Model
I* Under the sponsorship of the U. S. Air Force Aeromedical Re-

search Laboratory, a three-dimensional, discrete model of the
human spine, torso, and head was developed for the purpose of
evaluating mechanical response in pilot ejection. It was de-
veloped in sufficient generality to be applicable to other
body response problems, such as occupant response in aircraft
crash and arbitrary loads on the head-spine system. There are
no restrictions on the distribution of direction of applied
loads, so a wide variety of situations can be treated. The
model is described in Reference 60.
The anatomy is modeled by a collection of rigid bodies, which
represent skeletal segments such as the vertebrae, pelvis,
head, and ribs, interconnected by deformable elements, which
represent ligaments, cartilageneous joints, viscera, and con-
nective tissues. Techniques for representing other aspects of
the ejection environment, such as harnesses and the seat geom-
etry, are included. The model is valid for large displacements
of the spine and treats material nonlinearities. The elements
of the model are illustrated in Figure 30.
The basic model is modular in format, so that components may
be omitted or replaced by simplified representations. Thus,
while the complete model is rather complex and involves sub-
stantial computational effort, various simplified models that
are quite effective in duplicating the response of the com-
plete model within a range of conditions are available. Three
methods of solution are available for the analysis: direct
integration in time by either an explicit, central difference
method; by an implicit, trapezoidal method; or by a frequency
analysis method.
A variety of conditions have been simulated, including differ-
ent rates of onset, ejection at angles, effects of lumbar cur-
vature, and eccentric head loadings. It has been shown that
large initial curvatures and perfectly vertical acceleration
loadings result in substantial flexural resporse of the spine,
which cause large bending moments. It has been further shown
that the combination of the spine's low flexural stiffness,
initial curvature, and mass eccentricity are such that stabil-
ity cannot be maintained in a 10-G ejection without restraints
or spine-torso-musculature interaction.
60. Belytschko, T., Schirver, L., and Schultz, A., A MODEL FOR
ANALYTIC INVESTIGATION OF THREE-DIMENSIONAL HEAD-SPINE
DYNAMICS - FINAL REPORT, University of Illinois at Chicago
Circle; AMRL Technical Report 76-10, Aerospace Medical
Research Laboratory, Wright-Patterson Air Force Base,
Ohio, April 1976, AD A025911.
102
Figure 30. Three-dimensional head-spine
model. (From Reference 60)
The complete models were used mainly to study

the rib cage and viscera on spinal response. the effects of
stiffness of the torso is increased substantiallyThe flexural
by a visceral
model, even though it has no inherent flexural
addition, stiffness. In
the viscera provide significant reductions
axial loads. in the
4.8.6 One-Dimensional Seat/Occupant Models
Although a three-dimensional simulation should be
used for com-
plete prediction of aircraft occupant dynamics
in investigating
"4•. 103
restraint system properties or cockpit configurations to elimi-
nate secondary impact hazards, the more simple one-dimensional
models also may be useful in crashworthy seat analysis. For
example, a model such as that illustrated in Figure 31 provides
an economical means of optimizing energy absorber characteris-
tics, which would be simulated by spring X . Energy absorber
force-deflection characteristics might be ýkried while search-
ing for the most favorable occupant response, evidenced by a
minimum of spinal deflection, head acceleration, etc. The most
notable difficulty with the use of such a model lies in obtain-
ing valid occupant properties, i.e., masses and spring charac-
teristics. One such model that has been used in seat evalua-
tion is described in Reference 61.
Another widely known one-dimensional model is used to compute
the Dynamic Response Index (DRI). The DRI is a predictor of
spinal injury due to ÷G acceleration, and is based on the re-
sponse of a single-degree-of-freedom model as described in de-
tail in Volume II.
61. Auyer, W., and Turnbow, J., A STUDY OF THE DYNAMIC RE-
SPONSE OF A DAMPED, MULTI-DEGREE OF FREEDOM, SPRING-MASS
SYSTEM WHICH SIMULATES A SEAT, SEAT CUSHION, AND SEAT OC-
CUPANT SUBJECTED TO A VERTICAL IMPACT ACCELERATION, Avia-
tion Safety Engineering and Research (AvSER), Division of
Flight Safety Foundation, Inc. (unpublished report).
104
xs
He ad
M5
Nck C4
___
Spna
105
, .i,'. •-' • ", ' •,.' K . 4
sea Cushion,anocpat
[L••,. ~~~~pia ... .. .... K.y. .... n
I 105 M
*lo
_torso
. 5. ENERGY-ABSORBING DEVICES
5.1 INTRODUCTION
A multitude of devices for absorbing energy and limiting loads

have been proposed, developed, and tested. As demonstrated
earlier, the kinetic energy of a moving mass can be absorbed
by applying a force over a distance; this is the primary mech-
anism for absorbing crash energy. The larger the distance
through which the force acts, the lower the average load on
the mass. Energy-absorbing mechanisms in aircraft structures
which transmit crash forces to the occupant should stroke at
loads tolerable to humans and should provide stroke distances
consistent with these loads and with the energy to be absorbed.
Past experience has shown that plastic deformation of material,

primarily metal, results in a reasonably efficient energy-
absorbing process. Consequently, most load-limiting or energy-
absorbing devices use that principle.
Desirable features of energy absorbers are as follows:
* The device should provide a predictable force-versus-
deformation trace.
* The rapid loading rate expected in crashes should
not cause unexpected changes in the force-versus-
deformation characteristic of the device.
* The assembly in which the device is used should have
the ability to sustain tension and compression.
(This might be provided by one or more energy ab-
sorbers, or by the basic structure itself, depending
on the system design.)
"* The device should be as light and small as possible.
"* The specific energy absorption (SEA) should be high.
* Thc device should be economical.
"* The device should be capable of being relied upon to
perform satisfactorily throughout the life of the
aircraft (a minimum of 10 years or 8000 flight hours)
without requiring maintenance.
"* The device should not be affected by vibration, dust,
dirt, or other environmental effects. It should be
protected from corrosion.
106
e The device(s) should decelerate the occupant in the
most efficient manner possible while maintaining the
loading environment within the limits of human toler-
ance.
The discussion that follows refers to load limiters as sepa-
rate devices. This is not meant to imply that load limiters
must be separable devices at the exclusion of the integral de-
sign concept wherein the structure itself is designed to col-
lapse in a controlled and predictable fashion. Rather, the
discussion is present*1 in this way to simplify portrayal of
different methods of absorbing energy and limiting loads.
Research on simple, compact, load-limiting devices has been
conducted by the Government and by private industry. These
data are recorded in References 62 through 70. A brief discus-
sion of some of the more common energy-absorption devices and
concepts applicable to seats is presented in the following text
and in Table 3.
62. Ezra, A., and Fay, R. J., AN ASSESSMENT OF ENERGY ABSORB-

ING DEVICES FOR PROSPECTIVE USE IN AIRCRAFT IMPACT SITUA-
TIONS, in Dynamic Response of Structures, G. Herrmann and
N. Perrone, eds., Pergammon Press, Elmsford, New York,
1972, pp. 225-246.
63. Reilly, M. J., CRASHWORTHY TROO" S.ýAT INVESTIGATION, The
Boeing Vertol Company; USAAMRDL Technical Report 74-93,
Development Laboratory, Fort Eustis, Virginia, December
1974, AD/A-007090.
64. Kroell, C. K., A SIMPLE, EFFICIENT, ONE SHOT ENERGY AB-

SORBER, Reprint from Bulletin No. 30, Shock, Vibration,
and Associated Environments, Part III, General Motors Re-
search Laboratory, Warren, Michigan, February 1962.
65. Guist, L. R., and Marble, D. P., PREDICTION OF THE INVER-
SION LOAD OF A CIRCULAR TUBE, NASA Technical Note D-3622,
Ames Research Center, Moffett Field, California, June 16,
1966.
66. Haley, J. L-, Klemme, R. E., and Turnbow, J. W., TEST AND
EVALUATION OF 1000-4000 POUND LOAD-LIMITING DEVICES, Dy-
namic Science, AvSER Facility Report M69-2, for U.S. Army
Aviation Materiel Laboratories, Fort Eustis, Virginia,
February 1969.
107
i. . ... .
0
-o 4.6 4.4 a,4 4 a
I
04.
60.
C4 0
31460 4 0
4.A 4.O j
' 0
4.~ C1
C 9 9 -06 U
4,49 - a6 4
--
-~'0 - - - .91
'4" 15 x0 7
6. 4L
40 w0 a. -
C CC C-o 0. 96
~'0 ' 50 4.1 041
o w a o9 - 0 9 ,
a, 96 C ~ . 6.
cc c
go-
-4. 4- :.
9~~~~ N )9C r
'0
'4~ ~ v A U c O
'4~~~~ '0:99U 0.v .4

A" I. ft9
So. V . U 9-
c X,4~
9l1
4P c .I, cV AoU 41 9. m 6 V . 2
00 0 mu O
41~9. ACA 1
-'4C CAw 0 *1 . c6 4
>1 '0 C
C #.44 40- 640. # 0 #49 o 0 I-C
> o4. 0 9. C
C 94 0 5 av 0 .06 M*A V 4'
C C -969 A -108
In Table 3 Long-Term Reliability refers to the ability of the
device to perform its function without benefit of maintenance
throughout the life of the aircraft. The weight used in calcu-
lating SEA values includes the necessary end fittings required
to apply the load except as noted.
Pertinent characteristics of each device listed in Table 3 are
discussed in succeeding paragraphs. The concepts that have
found use in actual seat designs are presented first.
5.2 WIRE OR STRAP BENDING
This device uses the force required to bend a metal wire or
strap around a die or roller. It can be as simple as a steel
wire threaded through a perforated plate or a wire wound around
rollers. One characteristic that may be a problem with this
device (as with all devices affected by or utilizing friction
from metal-to-metal contact) is that an initial peak load
higher than the normal stroking load is induced. This initial
load increase can be reduced or eliminated by providing initial
slack in the wire when passing it over the rollers. These de-
vices, by themselves, do not have the ability to sustain com-
pressive loads. However, by anchoring both ends of the wire
and attaching the seat bucket to the rollers, compressive as
well as tensile loads can be sustained.
67. Rich, M. J., VULNERABILITY AND CRASHWORTHINESS IN THE DE-

SIGN OF ROTARY-WING VEHICLE STRUCTURES, Paper No. 680673,
presented at Aeronautic and Space Engineering and Manufac-
turing Meeting at Los Angeles, Society of Automotive En-
gineers, Inc., New York, October 1968.
68. Bendix Products Aerospace Division, ENERGY ABSORBING CHAR-
ACTERISTICS OF CRUSHABLE ALUMINUM STRUCTURES IN A SPACE
ENVIRONMENT, Report No. SPP-65-107 (NASA-CR-65096), pre-
pared for NASA Manned Spacecraft Center, Houston, Texas,
July 1965.
69. McGehee, J. R., A PRELIMINARY- EXPERIMENTAL INVESTIGATION
OF AN ENERGY-ABSORPTION PROCESS EMPLOYING FRANGIBLE METAL
TUBING, NASA Technical Note D-1477, National Aeronautics
and Space Administration, Washington, D. C., October 1962.
70. Schwartz, M., DYNAMIC TESTING OF ENERGY-ATTENUATING DE-
VICES, NADC Report No. AC-6905, Naval Air Development
Center, Warminster, Pennsylvania, October 1969.
)- 109
I-. - -
'.
Two variations of the wire-bending device have been developed

and used in the ceiling- and floor-mounted troop seat illus-
trated in Figure 32. The two tension-type devices at the top I
of the troop seat are shown in greater detail in Figure 33.
11
I. i
I
Aic Aft rcriaft flo IJ
Figure 32. Crashworthy troop seat. (From Reference 63)
In the analysis of energy absorbers for the troop seat, re-I

ported in Reference 63, wire of varying diameter was investi-
gated in order to produce a notched force-deflection curve as
recommended in Reference 36. It was concluded that the notched
110I
" ' • .. .... ... - -,,-_-- - - ' • I ;, hi'"

Support point Support point
Music wire
Rollers
Top
Housing of
j '- /seat
Support pir. back
Sre
Stroke ..... _ •mn
"_ 12 in.
mi~n.
op of
sect back
Seat back
Normal Pully stroked

Position Position
1600
1400
1200
1000
Design load 1,020 lb
800
600
400
200 (1) Initial vertical attenuator (heavier wire)

(2) Revised vertical attenuator (0.100-in. diameter wire)
0 1 2 3 4 5 6 7 $ 9 10 fll- 12 13 14
Deflection, in.
Figure 33. Troop seat tension energy absorber includ4 ng

characteristics for two wire diameters.
t -111
SC
.1
force-deflection curve was not suitable for lightweight troop *
seats due to the sensitivity of the system response to location
of the notch in the load-versus-deflection characteristic. A
fixed location for the notch was not compatible with the var-
ious dynamic response phasing resulting from the wide range of
troop and equipment weights. The trapezoidal force-deflection
curve produced by the constant limit-load device, although not
as efficient theoretically and ideally as the notched curve
for a specific dynamic condition, appeared to be more tolerant
of the wide range of seat occupant weights. Figure 33 shows
the force-deflection characteristics of the device that were
measured for two different wire diameters. I
I
The other variation of the wire-bending energy absorber used

in the above mentioned troop seat, and shown in Figure 34, is
capable of functioning in tension or compression. The deviceu
is contained in two telescoping aluminum tubes. A cap is
placed on the inner end of the inner tube. Music wire of
0.100-in. diameter, in the shape of a hairpin, is looped i
through the cap, and the two free ends are secured to a stud
in outer end
the rollers
three of the inner
sandwiched tube.
between A trolley
two plates bends the wire ofas'
consisting
the trolley moves back or forth on the wire. The trolley is
pinned to the outer tube, and a slot is provided in the inner
tube to allow passage of the pin connecting the two.
5.3 INVERSION TUBE
This device uses the force required to invert (to turn inside-
out or outside-in) a length of metal tubing. The concept
was developed by an American automobile manufacturing com-
pany for incorporation into steering columns to produce con-
trolled collapse loads (see Reference 64). No real disadvan-
tages have been noted in experimental tests to date except with
?hose loaded in compression. In dynamic tests of troop seats
(Reference 71) using these devices in compression, there was a
tendency for the outer and inner tubes to misalign, which re-
sulted in failure and crippling of the inner tube. However,
this problem can be solved by using an internal guide to keep-
the initial eccentricity from developi-ng. It is possible that
atmospheric corrosion could occur in the closed space between I
the inner and outer tube walls, especially in the bend radius.
It has been suggested that this potential problem might be
solved by injecting a low-density, closed-cell plastic foam
into the small volume between the inner and outer tube walls
71. Singley, G. T., III, FULL SCALE CRASH TESTING OF A CH-47C
HELICOPTER, paper presented at 32nd Annual National V/STOL
Forum, American Helicopter Society, Washington, D. C.,
May 1976.
112
6I -------------
*.-..
•Music wire
Outer alumiunum tube
TInner aluminum tube
1600*r "'"°° Slot in inner tube
1600
o4
400'-80 ; •
-eslcticn, in.
Figure 34. Tubular strut wire-bending energy absorber

force-deflection c orves. with
(From Reference 31)
113 5
4 V......- - -.
-
r 4 . -.- - -•:n•- ----- -.-. '. -lk- • .
.. -
...' - -. ". . . .- c -'_•' -. .
to prevent moisture penetration of this area. Also, the tubes
could be plated and/or coated to protect them from corrosion.
The materials used so far in inversion tubes have been 3003-H14

aluminum and mild steel, as described in References 64 through
66. It is possible that an annealed, higher strength alloy
steel, such as 4130 or stainless steel, could yield even higher
specific energy values than those shown in Table 3. However,
the aluminum devices that are in use are both compact and
lightweight.
Figure 35 illustrates a specific design concept of the inver-

sion tube energy absorbers (Reference 72). The load-deflection
curves are also shown for nine devices, three of which were
subjected to fatigue testing and three to both environmental
and fatigue testing. The traces are essentially flat for the
entire stroke distance after the initial peak.
Typical load-deformation information for tubes made from

3003-H14 aluminum is presented in Figure 36. These curves show
the variation in the load-deformation curve for a total of 12
dynamic test specimens and 2 static test specimens. Note that
the peak dynamic load is about 1300 lb while the minimum dy-
namic load is about 1010 lb. This 25-percent variation is
higher than the variation now being experienced at equivalant
loading rates in operational systems. Posttest examination of
the specimens showing 15- to 25-percent load increases indi-
cated that the inverting tube had been scrubbing the inner
wall of the loading tube due to asymmetric loading in the jig;
therefore, friction probably caused some of the load increase.
Figure 36 shows clearly that the inverting load remains essen-
tially constant during the stroke (Reference 64).
5.4 ROLLING TORUS
Early versions of this energy absorber consisted of a number

of torus elements located in the annular space between two
telescoping cylinders. Modification of this concept has re--
sulted in the substitution of a continuous helix of stainless
steel wire for the toroidal elements. The interference fit
between the cylinders and tori, or wire, causes the wire to
roll when axial loads are applied. The cyclic plastic defor-
mation of the rolling tori or *ire helix and elastic deforma-
tion of the tubes effect the energy absorption. The cylinders
remain intact and do not plastically deform when subjected to
S7 esjardins, S. P., et al., CRASHWORTHY ARMORED CREWSEAT

FOR THE UH-60A BLACX HAWK, paper presented at 35th Annual
National Forum, American Helicopter Society, Washington,
D. C., May 1979.
114
RMW
* Thin-walled aluminum
roll tube (3003-H14)
Unstroked
Outer tube_
Partially stroked Plasticall

deformingA
region I
1500j
100
o toThree devices
previously en- Three new
Three devices vironmentally devices used
previously and fatigue as control
fatigue tested tested group
50 --- S/N 1002 ---- &S/N 1005 u---S/N 1008
.......... SIN 1003 S/N 1006 ~- SIN 1009
.........S/N 1004 -- uS/N 1007 - S/ IN 1010
0
0 2 4 6 8 10 12 14 16 18 20I
Stroke, in.
Figure 35. Inversion tube concept with typical

force-deflection characteristic.
* 115
6061T-6, 0.035-in.
wall X 2.125-in. 10.25
OD loading tube
d 5.5 n- Stroke
2.0 in.- 3003-H14, 0.035-in.

wall X 1.75-in. OD
inverting tube
2000
Wt 0.35 lb each
Load discontinued
iooal
1500 : '.. -
1000 • -- ' • .. - •' C d-.. . • • •
S 500 --- Max dynamic Results based

Min dynamic on 2 static and
- Static 12 dynamic tests
0 1 2 3 4 5 6
Deformation (compression), in.

Figure 36. Comparison of dynamic and static load-
deformation curves for inversion tubes.
impact loading. The impact force is transmitted through the

tubes to the tori or wire helix. Dynamic testing of these de-
vices is reported in Reference 70.
The load !imiters using wire as the working medium (Figure 37)
are normally made with cylinders that range from 1 to 2 in. in
diameter with a wall thickness of approximately 0.035 in. The
wire ranges between 0.030 and 0.035 in. in diameter and is of
300 series stainless steel. These bidirectional devices may be-
used repeatedly several times until fatigbe failure of the wire
occurs. The recent investigation of a lighter weight aluminum
energy absorber of this type is documented in Reference 73.
73. Mazelsky, B., INVESTIGATION OF AN ALUMINUM ROLLING HELIX

CRASH ENERGY ABSORBER, ARA, Inc.1 USAAMRDL Technical Re-
port 77-8, Eustis Directorate, U. S. Army Air Mobility
Research and Development Laboratory, Fort Eustis, Vir-
ginia, May 1977, AD A042084.
116
W'_ I
i. 4000 "
3000"
2000-r MI
01 2 3 4
Deflection, in.
Section
Outside
Figure 37. Rolling torus energy absorber.
4.11
• 117
......- ,,..
Devices of this type can be single or multiple staged. The
multiple-staged energy absorbers include three tubes with
helices of wire between the walls of the outer tube and the
center tube, and between the center tube and the inner tube.
In operation, one helix of wire is rolled to the end of its
stroke and then the second stage is initiated and rolled.
Staged energy absorbers provide increased stroke distance
without an appreciable increase in prestroked envelope.
The device produces a somewhat jagged load-versus-deformation
characteristic as can be seen in Figure 37. Further, the in-
terference contact between the tori and the cylinders, the
closed spaces between the tube walls, and the spaces between
the wire wraps are prime areas for corrosion. This potential
should be considered during the development, test, and usage
of this device. The long-term effects on performance of the
interference fit between the wire and the tubes is another area
for concern. J
Seats with energy-absorbing mechanisms utilizing this device
are now in use in a modified U. S. Marine helicopter (Refer-
ence 16) and in a utility helicopter developed for Iran.
5.5 CRUSHING HONEYCOMB
This device uses the force required to crush or deform a column
of low-density material. In order to provide sufficient column
stability and transverse load resistance, it appears that most
applications will require a telescoping cover to give a d4i-
tional bending strength. Table 3 shows this device to be above
average in all categories with the exception of rebound load-.
could probably be added by the
of a load
ability. Rebound capacity
suitable mechanism that allows movement in
incorporation
only one direction.
This device, besides being used on seats, is used as a load
limiter in the main landing gears of some helicopters. In
these applications, the crushable material is installed above
the oleo piston as outlined in Reference 67. The energy-
absorption ability of these devices has been responsible for
preventing major structural damage to several aircraft in se-
vere accidents.
To date, the best crushable material for use in this type of

device appears to be corrugated aluminum foil backed by flat
foil, cemented at the nodal points as illustrated in Fig-
ure 38. Further research information on the development of
crushable aluminum columns may be found in Reference 68.
118
i• !im
i • • ld "....i...............i
• •-............. .- "-
Load
S~Loadl
Cement points
Start of
wrapping
Figure 38. Illustration of corrugated aluminum
foil formed into annular column.
5.6 EXTENSION OF BASIC METAL TUBE OR FLAT STRAP

This concept uses the inherent plasticity of certain ductile
metals which elongate under a relatively constant force. The
primary problem with this device is strain concentration at
the end connections. Research to date indicates that annealed
stainless steel in the AISI 300 series is least susceptible to
strain concentrations because of its excellent ductility (45 to
50 percent).
The flat strap device was evaluated for use as a vertical load
limiter for a pilot's seat by the U. S. Naval Aircraft and Crew
Systems Technology Directorate, now part of the U. S. Naval
Air Development Center and was found to perform satisfactorily
(Reference 74). Since a flat strap sustains only minimum com-
pressive loads, a separate rebound device would be necessary
for application in personnel seats-.-
The thin-walled tube will perform in much the same manner as
the flat strap, and it has the advantage of sustaining higher
74. Woodward, C. C., et al., INVESTIGATION, DESIGN AND DEVEL-
OPMENT OF AN F7U-3 EJECTION SEAT ENERGY ABSORPTION SYSTEM
FOR REDUCTION OF CRASH FORCE LOADS, NADC Report ACEL-335,
Naval Air Development Center, Warminster, Pennsylvania,
June 1957.
A; 119
*- - . . . . . . . . -... . * .
compressive loads; although this capab ity s still inadequate.
It is desirable that the tube elongate throughout its length
rather than locally; for example, at the end attachments. A
successful method of achieving nearle uniform elongation is the .
use of a low-modulus bonding agent between the tube and the ap-
propriate end fitting (see Reference 66).
Typical load-elongation characteristics of a 0.02-in. wall by
0.50-in.-diameter stainless steel tube are illustrated in Fig-
ure 39.
e
Epoxy bonTest length
0.02-in. All x 0.50-in. diameter tube

3000 W 0,21 1ý eacl , .- in . --- _____
/< • Failure
2000
S•.•_ - ... 'd
M 1500.
0 [/---Min
f Max dynamic
-Static
dynamic
Results based on
2 static and 12
I dynamfi(; tests
500 3044 ANNEALED STAINLESS STEEL ....
1 2 3 4 5 6 7
Deformation, in.
110 20 30 40 50 60
Deformation (elongation), percent
Figure 39. Comparison of dynamic and static load-elongation

curves for stainless steel tubes.
120
N_
ME
5.7 ELONGATION OF BASIC STRANDED CABLE
This device has the same characteristics as the basic metal

tube or flat strap; however, the flexibility of a cable obvi-
ously has advantages for some load-limiter applications. The
cable end fittings are capable of sustaining the ultimate load
of the cable under static and dynamic conditions. This device
appears to be most applicable to bracing lightweight seats,
such as troop and gunner seats.
5.8 ROD PULL-THROUGH TUBE
This device uses the force required to expand the diameter of

a tube as a hardened, oversized rod or tube is drawn through
it, or to compress an inner rod or tube as it is drawn through
a die. The force required to overcome friction also contrib-
utes to the ernergy absorbed by this device; however, dependence
on friction to maintain a uniform load is unpredictable. The
frictional resistance of the device tested in Reference 66 (a
compression tube device with a rigid outer cylinder) was re-
duced by lubrication, but the device exhibited an initial peak
load as indicated by point A in Figure 40.
It can be seen in Figure 40 that the stroke of this device was
limited to 4 in. and that the failure load was about three
times the stroking (sustained) load. Thus, the tested device
had a safety factor of at least 3 to 1 built into it, and this
fact partially accounted for the poor specific energy rating
shown in Table 3. It can be seen in the figure that the maxi-
mum variation in the stroking load was from 1300 to 1600 lb,
or about 21 percent.
5.9 TUBE FLARING
This device simultaneously uses the forces required to expand

the diameter of a tube to the failure point and to bend the
tube walls through 90 degrees. The tube wall either shatters
into fragments or rolls up into spirals around the periphery
of the tube, as illustrated in Figure 41. A review of Refer-
ence 69 indicates that the above processes are sefisitive to
the ratio of the wall thickness to the die -radius; and that
ratios of less than 0.3 are likely to result in a rolling proc-
ess, while ratios of greater than 0.4 are likely to result in
the fragmentation on the basis of tests using 2024-T3 aluminum
tubes.
This concept has been evaluated for an experimental crewseat

by the U. S. Naval Aircraft and Crew Systems Technology Direc-
torate, now part of tbI'U. S. Naval Air Development Center, as
described in Reference 70. The device was used as the vertical
1 121
- ~~
- - ----- -*-
- ---- - - - -- - - - - -
i.n
* .A
5000- 50Compression
- . tube
Wt 0.86 lb each Failure
4500 CODE
Max dynamic Results based on
4000 - . Min dynamic 1 static and 4_ _
-0Static dynamic tests
3500
S3000
2500 Load discontinued -

before bottoming
A occurred,
I' I
2000 ----
- I! I
4
.
1500- " - - •____

- - - -"
I.000
I
!
. _ _. __ _ __ _ _
I
_ __ _ _ _
.I
500ii I-- -___

0 3 .
0 2 3 4 5 6
Deformation (elongation), in.
Figure 40. Comparison of dynamic and static load-deformation
curves for compression tubes.
energy absorber in the seat. The device also was used as the
vertical load limiter for an experimental troop seat, as de-
scribed in Reference 63.
The device cannot sustain rebound forces because only a mini-
mum rebound resistance is provided by friction between the tube
and the forming die. However, a mechanism was installed in the
forming die to grip the tube against rebound movement.
122'
0
Forming die Fragmentation Rolling
Figure 41. Illustration of fragmentation and
rolling-
processes in tube-flaring device.
5.10 TENSION PULLEY

The tension-pulley load limiter
shown in Figure 42 requires
that the pulley casing be literally
rotates. The plastic deformation split open as the pulley
of the casing material ef-
fects the energy absorption. As the
is unidirectional and operates under name implies, the device
has been used in cargo restraint tensile loading only. It
systems and energy-absorbing
troop seats, as described in Reference
75.
5.11 SELECTION OF AN OPTIMUM LOAD LIMITER
An optimum load-limiting system
of the data presented above. Thecannot be selected on the basis
lines with due consideration to data should be used as guide-
the requirements for each spa-
cific application.
75. Turnbow, J. W., Robertson, S. H.,

and Carroll, D. F., DY-
NAMIC TEST OF AN EXPERIMENTAL TROOP
AN H-21 HELICOPTER, Aviation Crash SEAT INSTALLATION IN
Injury Research (AvCIR),
Division of Flight Safety Foundation,
nical Report 63-62, U. S. Army Inc.1 TRECOM Tech-
Aviation Materiel Labora-
tories, Fort Eustis, Virginia, November
1963.
123
'I
Deformed
housing
jihi
AA
I I
Pigure 42. Tension-pulley load limiter.
124
• ..__ ..._ .. ..
i= ,• ':
6. SEAT CUSHIONS
6.1 INTRODUCTION
The seat bottom and back with which the occupant is in constant
contact should be designed for comfort and durability. Suffi-
cient clearance between fabric backs and bottoms or sufficient
cushion thickness of the appropriate material stiffness should
be provided to preclude body contact with the seat structure
when subjected to either the specified operational or crash
loads. Seat bottoms made of fabric should be provided with
I: means of tightening to compensate for sagging in use. The con-
flicting requirements of long-term comfort-versus-crash safety
considerations have made this a difficult design area.
For seat cushions, the problem is one of developing a compro-
mise design that will provide both acceptable comfort and
safety. In the past, the comfort requirement was met by pro--
viding very thick, soft, foam cushions that allowed the oc-
cupant to sink in deeply, thereby producing a contour and
spreading the load around the person's buttocks so as to de-
crease local high pressure and eliminate point loading. This
approach provided both immediate and long-term comfort. A
method of providing thermal comfort was to force air through
4 the cushion, or to use stretched net cushions, which provided
contouring and load spreading as well as the free passage of
air. The passage of air allows the evaporation of sweat and,
thus, achieves the desired cooling effect.
Crash-safety considerations require a minimal thickness of foam
to minimize or eliminate vertical motion of the pelvis during
high vertical loadings. This requirement conflicts with the
method chosen for providing pressure comfort described in the
previous paragraph, and constitutes a problem that must be
solved to provide an acceptable cushion.
One approach producing the desired compromise between crash
safety- and comfort uses a cushion hase-with a contour that
matches the universal buttocks configuration as closely as pos-
sible. This wraparound configuration spreads the load and de-
creases localized pressure without resorting to soft foams.
Additional comfort layers of foam can then be added to the base,
and the cushion base can be equipped with slots or holes which
allow for fore-and-aft passage of air to provide the desired
cooling. A layer of rate-sensitive foam can be used on top of
the base to provide a contour transition softer than the base.
This layer must either be open celled, or holes must be pro-
vided to allow for vertical movement of air. A layer of soft,
open-celled foam can be used on top of the rate-sensitive foam
to provide the initial comfort material and to also provide
125
- 1
vertical and horizontal air motion. The entire cushion can be
covered with a fire-retardant, open, nylon material to provide
for wear and abrasion resistance.
The total thickness of the compressed cushion at the buttock
reference point should be minimized and can be limited to be-
tween 0.5 and 0.75 in. of thickness. This cushion probably
permits not more than 3/8 in. of vertical motion of the pelvic
structure (ischial tuberosities) from the 1-G loaded position
to the full vertical crash-loaded position.
Other methods of achieving the desired effect are available.
One is to include the basic provisions just described but to
achieve the thermal effect plus some loading comfort by the
use of special coverings such as lamb's wool. This type of
cover uses the lamb's skin with a small depth of combed and
clipped wool on the occupant interface surface. These covers
need holes cut through the leather to allow free passage of
air for cooling as previously discussed.
To meet the required crashworthy characteristics, the optimum
aircraft seat cushion will:
"* Be extremely lightweight.
"* Possess flotation capabilities.
"* Be nonflammable.
"• Be nontoxicl will not give off fumes when burned,
charred, or melted.
"* Be tough and wear resistant.
"* Be easily changeable.
"* Provide comfort by distributing the load and reducing
or eliminating load concentrations.
"* Provide thermal comfort through ventilation.
"* Provide little or no rebound under crash loading.
"* Allow an absolute minimum of motion during crash
loading.
6.2 REQUIREMENTS
For seats of light movable weight (less than 30 lb), cushions
should be used for comfort only. The maximum uncompressed
126
thickness for a properly contoured cushion should be 1-1/2 in.,
unless it can be shown through analysis or through dynamic
tests that the cushion design and material properties produce
a beneficial (reduced force transmissibility) result.
For seats of greater movable weight, such as integrally ar-
mored seats, every effort should be made to design a cushion
that minimizes relative motion between the occupant and the
seat and that acts as a shock damper between the occupant and
* the heavy seat mass. Viscoelastic and loading-rate-sensitive
materials, such as discussed previously, can be used to accom-
* plish this goal. Again, dynlamic analysis and/or testing should
be conducted to demonstrate that the cushion design produces a
desirable system result over the operational and crash environ-
mental range of interest.
6.3 ENERGY-ABSORBING CUSHIONS

The use of load-limiting cushions in lieu of load-limiting
seats is undesirable for two reasons:
a The downward movement of the torso into a crushable
seat cushion produces slack in the restraint harness.
This slack could allow injury during subsequent lon-
gitudinal or lateral acceleration in forward-facing
seats by contributing to dynamic overshoot and/or by
allowing the lap belt to move upward into the soft
portion of the abdomen. For an aft-facing seat, this
slack is not as significant for longitudinal acceler-
ations but applies to the lateral direction.
* A crushable cushion does not make optimum use of the
available stroke distance since space must be allowed
for the crushed material. A crushable cushion can
be only approximately 75 percent as efficient as a
mechanical load-limited system thAt allows the seat
to stroke completely to the floor.
Crushable cushions are impractical in rotary- and light fixed-

wing aircraft because of the long stroke distance required
to attenuate the high vertical - loads present in the 95th-
percentile crashes. The only justifiable use of energy-
absorbing cushions instead of load-limited seats might be in
retrofit circumstances where, because of limitations in exist-
ing aircraft, another alternative does not exist (see Refer-
ence 76 for further information on energy-absorbing cushions).
76. DYNAMIC TEST OF CRUSHABLE SEAT CUSHIONS, AvSER Report
M67-6, Aviation Safety Engineering and Research (AvSER),
Division of Flight Safety Foundation, Inc., Phoenix, Ari-
zona, August 1967.
127
low• -o. .. .,.N. ,..

.. 1ý7 "Wr
Recent research has indicated that foams can be used more eco
nomically than honeycombs without reduction in performance.
Foams are much easier to form and are less costly than metallic
honeycomb materials and are therefore recommended for this use.
6.4 NET-TYPE CUSHIONS
This type of cushion serves the same purpose as the filled
cushion; however, a net material is stretched over a contoured
seat frame, and the body is supported by diaphragm action in
the net rather than by deformation of a compressible material.
The net-type cushion might more properly be called a net su -
port. If a net support is used in the seat, its reboud ca-
acteristics should be capable of limiting the return movement
from the point of maximum deformation to 1-1/2 in. Net sup-
ports should not increase the probability of occupant submar-
ining or dynamic overshoot. The net elastic-stretch limita-
tion might be achieved by including a stiffer net, such as a
steel or aluminum woven material under the net support.
6.5 OTHER CUSHIONS

In most cases the back cushion will not play a significant role
in the crash dynamics; however, it will influence comfort and
can influence the injury tolerance of the spine. The cushior
should be of a lightweight foam material or net. The foam can
be a standard furniture type that meets the other requirements
listed in Section 6.2. Lumbar supports, particularly those
that are adjustable by the occupant, are desirable for comfort
and for safety reasons. A firm lumbar support that holds the
lumbar spine forward in extension increases the tolerance to
+Gz loading.
6.6 HEADRESTS
A headrest should be provided for occupant head/neck whiplash
protection. Headrest cushions are used only to cushion head
impact and prevent whiplash injury due to backward flexure of
the neck. The cushioning effect can be provided by a thin pad
and a deformable headrest or a thicker cushion on a more rigid
headrest. For the thicker cushion, the provisions of Section
10.9 should be applied and at least 1.5 in. of cushion should
be provided if possible within the space limitations of the
application.
128
.
..
....
I
7. DESIGN PRINCIPLES FOR PERSONNEL RESTRAINT SYSTEMS
7.1 INTRODUCTION
Crash injury accident statistics indicate that failure of per-
sonnel restraint harnesses has been a frequent cause of injur-
ies and fatalities in U. S. Army aircraft accidents. This is
unfortunate because body restraint is relatively easy to con-
trol. Adequate restraint in a crash can mean the difference
between life and death, since evacuation from a burning or
sinking aircraft is considerably improved if no prior injury
or debilitation has occurred. It is the intent of this sec-
tion to provide general criteria and quidelines for the design
of personnel restraint systems to reduce injury or debilitation
in a crash situation. Design criteria for cargo restraint sys-
tems are presented in Volume III.
Restraint harnesses for personnel should provide the restraint
necessary to prevent injuries to all aircraft occupants in
crash conditions approaching the upper limits of survivability.
Appropriate strength analysis and tests as described in Section
8.5 should be conducted to ensure that a restraint system is
acceptable.
*1 Numerous methods of restraining the human body have been pro-
posed, investigate&, and used. Some of these have proven to
be exceptionally good, and some have left much to be desired.
However, there are certain qualities that a harness should pos-
sess if it is to be used routinely for military flights. These
desirable qualities are listed below:
* Comfortable and light in weight.
* Easy for the occupant to put on and take off even in
- the dark.
* Contain a single-point release system, easy to oper-

ate with one (either) hand since a debilitated per-
son might have difficulty in releasing more than one
buckle with a specific hand. Also, it should be pro-
tected from inadvertent release, e.g., caused by the
buckle being struck by the cyclic control or by iner-
tial loading.
e Provide personnel with freedom of movement to operate

the aircraft controls. This requirement necessitates
the use of an inertia reel in conjunction with the
shoulder harness.
129
1 I
# Provide sufficient restraint in all directions to
prevent injury due to decelerative forces in a po-
tentially survivable crash.
* Webbing should provide a maximum area, consistent
with weight and comfort, for force distribution in
the upper torso and pelvic regions and should be of
low elongation under load to minimize dynamic over-
shoot.
7.2 TYPES OF SYSTEMS
7.2.1 Aircrew Systems
The existing military lap belt and shoulder harness configura-
tion with a center tiedown strap as shown in Figure 43 is the
minimum acceptable harness for use by U. S. Army pilots. The
lap belt tiedown strap resists the upward pull of the shoulder
straps and prevents the belt's displacement into abdominal tis-
sue. The tiedown strap is comfortable to wear since it does
not contact the pelvis, and it is narrow enough within limits
of acceptable strength so that little leg rubbing is encoun-
tered by the wearer during antitorque or rudder pedal opera-
tion.
The configuration shown in Figure 44 provides improved lateral
restraint due to the addition of the reflected shoulder straps.
This system, which resulted from the investigation reported in
Reference 77, consists of one dual-spool inertia reel or two
separate inertia reels with two reflected straps, a shoulder
harness collar assembly, a lap belt assembly including retrac-
tors, and a buckle assembly. The buckle assembly consists of
a single-point release buckle permanently attached to the tie-
down strap. The tiedown strap consists of a fixed-length strap
for any specific seat and cushion design, and an anchor fitting
that connects the strap to the seat pan beneath the seat cush-
ion. The left- and right-hand lap belts, connected at the
single-point release buckle, are attached to the seat or-air-
craft structure through automatic rock/unlock retractors.
The shoulder harness collar assembly consists of a pad in the
form of a collar fitting around the crewman's neck, over which
the shoulder harness straps are routed. The lower shoulder
77. Carr, R. W., and Desjardins, S. P., AIRCREW RESTRAINT SYS-
TEM - DESIGN CRITERIA EVALUATION, Dynamic Science, Divi-
sion of Ultrasystems, Inc.1 USAAMRDL Technical Report 75-2,
Development Laboratory, Fort Eustis, Virginia, February
1975, AD A009059.
130
- -!
I
Item identity.3
1. Buckle assembly
A. Single-point
release buckle 3B
B. Tiedown strap
C. Tiedown anchor
2. Lap belt assembly
A. Lap belt
Adjuster
3. Shoulder harness
assembly
A. Inertia reel
B. Inertia reel strap
C. Lower shoulder
strap
D. Adjuster 3C!
2I
1B A
4
ardI
Figure 43. Basic aircrew restraint system.

~rd
•131
""A•I'i
mm
• • "•"• •"I
item identity•
1. Suckle assembly
A. Single-point
B. Tiedown strap
C. Tiedown anchor B
2. Lap belt assembly
A. Lap belt
B. Retractor
3. Shoulder harness
collar assembly 3
A. Pad
B. Roller fitting
C. Adjuster
D. Lower shoulder strap i"
4. Inertia reel assembly
A. Reflected strap
B. Anchor D
I"
~rd
S1 !OWN
Figure 44. Aircrew restraint system, including

reflected shoulder straps.
132
straps connect to the bottom of the collar assembly through the
adjusters. The reflected straps pass through the roller fit-
tings at the top of the collar. Each reflected strap is ex-
tended forward from an inertia reel, looped through the roller
fitting, and then directed rearward to the opposite side of the
seat back. These straps are attached to the seat through an-
chor fittings on the reflected ends and through inertia reels
at the other end. The lap belt straps, tiedown strap, and
lower shoulder straps are all connected at the single-point
release buckle. Details of the hardware in these systems are
discussed in Section 7.5.
7.2.2 Troop Systems
Considerations in the selection of a troop or passenger seat
restraint system are different from those for an aircrew sys-
tem. First of all, the seat may face forward, sideward, or
aftward. Secondly, the restraint system must be capable of
being attached and removed quickly in an operational environ-
ment by troops encumbered by varying types and quantities of
equipment. Also, whereas a pilot probably u..es the restraint
system in his aircraft so frequently that its use becomes a
matter of habit, troops and passengers can be expected to be
unfamiliar with the system. The effects of this lack of famil-
iarity would probably become more pronounced in a combat situa-
tion when the risk involved in not using the restraint system
becomes even higher. Therefore, hardware should be uncompli-
cated and, if possible, resemble the familiar, such as automo-
tive hardware. Finally, the need to quickly remove and stow
the seats requires compact and lightweight restraint systems.
For the aft-facing passenger, the need for a tiedown strap is
negligible, since the seat back will provide the primary re-
straint; however, the shoulder harness should be retained as
part of the restraint system to provide adequate support in
crashes that produce significant lateral loads.
It is difficult to provide adequate restraint for stde-facing
passengers with a lap belt and shoulder harness alone. Leg
restraint would also be preferred; -however, leg restraints are
generally not practical because of operational requirements.
which necessitate the side-facing seats in Army aircraft. A
reflected shoulder strap and side belt strap offers a compro-
mise solution; however, they too have met with resistance be-
cause of weight and cost considerations. The belt side straps,
extending from the lap belt high on the thigh to the seat pan
forward of the lap belt anchor, help to hold the belt in place
over the pelvic region as well as provide more area to resist
the pressure from the pelvis. The reflected shoulder strap
provides improved upper torso restraint.
133
wI
Two systems that resulted from the investigation reported in
Reference 78 are shown in Figure 45. The Type II troop re-
straint system was designed to mount on a forward-facing or
aft-facing troop seat and consists of a two-strap shoulder har-
ness and a lap belt assembly. The two shoulder straps are at-
tached to two single inertia reels. They extend forward and
down over the occupant's upper torso and are connected into
the single-point release, lift-lever buckle. The lap belt as-
sembly includes left- and right-hand belts, with adjusters,
that are connected together at the lap belt buckle. The Type
I troop restraint system was designed to mount on a side-facing
troop seat and differs from the Type II restraint by having a
single shoulder strap that passes diagonally across the occu-
pant's upper torso. It should pass over the shoulder closest
to the nose of the aircraft. If the Type I system is used in
either a forward- or aft-facing seat, the diagonal shoulder
strap should pass over the outboard shoulder to restrain the
occupant from protruding outside the aircraft during lateral
loading.
7.2.3 Crew Chief and Door/Window Gunner •ystems
Restraint systems for crew chiefs and door/window gunners are
similar to troop systems; however, they must allow the crew-
member to move out of the seat to perform duties such as maneu-
vering the gun or observing tail rotor clearance while landing
in unprepared areas. The system should restrain the occupant
to the seat the instant he returns to the seat and provide ade-
quate restraint during a crash. The system should maintain
the lap belt buckle in the proper relationship to the gunner,
preventing the shoulder straps from pulling it up or the lap
belt from pulling it sideways. Such a system has been de-
scribed in Reference 79 and is shown in Figure 46. It consists
of a lap belt with inertia reels on each side of the seat and
two shoulder straps connected in an inverted-Y arrangment to a
single inertia reel strap. The lap belt with thigh strap at-
tachment is easy to put on and prevents the lap belt from rid-
ing up during operation of the gun. The lap belt is plugged
into the two seat pan inertia reels when the crewmember is:to
78. Carr, R. W., HELICOPTER TROOP/PASSENGER RESTRAINT SYSTEMS
DESIGN CRITERIA EVALUATION, Dynamic Science, Division of
Ultrasystems, Inc.; USAAMRDL Technical Report 75-10, Eus-
tis Directorate, U. S. Army Air Mobility Research and De-
velopment Laboratory, Fort Eustis, Virginia, June 1975,
AD A012270.
79. Reilly, M. J., CRASHWORTHY HELICOPTER GUNNER'S SEAT IN-
VESTIGATION, The Boeing Vertol Company; USAAMRDL Technical
Report 74-98, Eustis Directorate, U. S. Army Air Mobility
ginia, January 1975, AD A005563.
134
22
22
SI ',
2, / 4.._ /s
I" [ . I\- -3
66
Type I Type II
Item identity
1. Inertia reel
2. Shouldar strap
3. Lap belt anchor
4. Buckle with shoulder strap connection
5. Lap belt
6. Adjuster/fitting
Figure 45. Aircraft troop/passenger restraint systems.
135
;t, .- - --. - .:- .-. A, , . .-

. II
-i
4
5. La.bl
Itenm
identity
2. Shoulder strap
3. Shoulder strap adjuster
4. Attachment release buckle
6. Lap belt inertia reel

7. Thigh straps * I
8. Thigh strap adjuster
9. Lap belt plug-in fitting
Figure 46. Gunner restraint system. (Prom Reference 79)
136
______.....__ __.....
..... I
be seated or standing in front of the seat. The shoulder har-
ness and lap belt with thigh straps may serve as a "monkey har-
ness" when the crewmember disconnects the two lap belt plug-in
fittings from the inertia reels. The resultant configuration
permits the crewmember more extensive travel within the cabin
while still being connected to the shoulder harness inertia
reel, thereby restraining the crewmember from falling out of
the aircraft.
7.2.4 Inflatable Systems
An automatically inflatable body and head restraint system for
helicopter crewmen has been jointly developed and tested by
-. .the Naval Air Development Center and the Applied Technology
"Laboratory. As illustrated in, Figure 47, this system provides
increased crash protection because it provides automatic pre-
tensioning that forces the occupant back in his seat, thereby
reducing dynamic overshoot and reducing strap loading on the
wearer when the inflated restraint is compressed during-the
crash. The concentration of strap loads on the body are re-
duced because of the increased bearing surface provided when
the restraint is inflated, and both head rotation and the pos-
sibility of whiplash-induced trauma are also thus reduced.
Although more complex and costly than conventional belt sys-
tems, such a system may be justified because of its occupant
protection potential. Development of the system and results
of testing are documented in References 80 and 81.
7.3 GENERAL DESIGN CRITERIA

7.3.1 Comfort
Comfort must not be compromised by crash-survival requirements
for obvious reasons. For example, a lap belt with an adjust-
ment fitting located directly over the iliac crest bone would
provide a constant source of irritation that would result in
eventual fatigue to the wearer. The main comfort considera-
tion for restraint harnesses is-the absence of rigid hardware
80. Schulman, M., and McElhenney, J., INFLATABLE BODY AND HEAD
RESTRAINT, NADC-77176-40, Naval Air Systems Command, Pe-
artment of the Navy, Washington, D. C., September 1977,
D A046477.
81. Singley, G. T., III, TEST AND EVALUATION OF IMPROVED AIR-
CRAFT RESTRAINT SYSTEMS FOR COMBAT HELICOPTERS, Paper No.
A.18, presented at NATO/AGARD Aerospace Medical Panel,
Aerospace Specialist's Meeting on Aircrew and Survivabil-
ity, North Atlantic Treaty Organization, Bodo, Norway,
May 20-23, 1980.
137
-.
I -4
II
1381
0i
I
i*
located over bony portions of the torso. Also, webbing that
is too wide or too stiff could cause discomfort.
7.3.2 Emergency Release Requirements
From a crash survival point of view, it is mandatory that a
shoulder harness/lap belt combination have a single point of
release that can be operated by one (either) hand to make it
easier for debilitated occupants to quickly free themselves
from their harnessing in a severe crash because of the dangers
of postcrash fire or sinking in water. The force required to
release the harness with only one finger should fall between
20 and 30 lb on the basis of existing requirements for mili-
tary harnesses. An excessive force could hii.der rapid emer-
gency release, while a light force could cause inadvertent re-
lease. Further, the release should be possible with the weight
of the occupant hanging in the restraint system after exper-
iencing the full crash loads. This will guarantee that the
occupant can release the system after a severe survivable crash
even when inverted in a loaded restraint system. The release
forces for the inverted case should be minimized and, in any
case, should not exceed 50 lb applied with only one finger. It
should be possible to produce the torque necessary to release
rotary buckles by applying a load at a single point on the
handle as described above.
In restraint systems other than the Type I of Figure 45, if a
lift latch or similar type buckle is used, the restraint system
design should ensure that the latch lifts from left to right on
all installations. This will reduce the possibility of reverse
installations and their resultant hazard.
The release device must either have the capability to with-
stand the bending moments associated with doflections and mo-
tions during loading, or it should contain features that allow
the fittings to align themselves with the loads, thereby reduc-
ing or eliminating the moments. If belt loading direction is
such as to cause the strip to-bunch up in the end of a slot,
failure can occur through initiation of edge tear. As a re-
sult of an investigation of restraint system design criteria
reported in Reference 77, the fitting angles illustrated in
Figure 48 are recommended.
Eliminating fitting rotation in the flat plane of the buckle
during loading may prove to be difficult in lightweight sys-
tems. If the integrity of the attachment of the fitting within
the buckle can be compromized by rotation, then rotation must
be completely eliminated. Experience has shown that it is
better to design the attachment of the fitting within the
139
-it
:4
Figure 48. Buckle fitting attachment and motion angles.
buckle to be insensitive to rotation, i.e.,r a round pin in a

round hole, rather than a flat-faced dog which must seat on a
flat face of a slot, than to rely on restraining the fitting
against rotation. In the latter case, a small amount of rota-
tion can cause point loading of a corner of the dog against
one end of the slot. The point loading can easily increase
the stress applied at the contact point to i..s ultimate bear-
ing strength. This will result in metal deformation and the
formation of a sloped surface which then can act to cam open
the attachment mechanism.
Further, the release mechanism (buckle) should be protected

against accidental opening. Neither decelerative loading of
components nor contact with aircraft controls such as cyclic
controls should open the device. It was mentioned earlier in
this volume that required cockpit dimensions should be re-
viewed. It appears that the occupant can be placed too close
to the cyclic control in helicopters and that a fully retracted
cyclic head can contact the buckle. The buckle release mech-
anism should be protected against inadvertent release either
during operation or in-a crash. It should be emphasized that,
iof contact between the cyclic control and the buckle is pos-
sible in an-operational mode, a considerable overlap can exist
during crash loading when the restraint system is deformed for-
ward several inches.
7.3.3 Lap Belt Anchorage
The anchorage points for the lap belt can be locatei either on
the seat bucket or on the basic aircraft structure. if the an-
chorage is located on basic aircraft structure, the movement
of the seat under the action of load-limiting devices must be
140
,,I , , -. . '" . .. • .. " ----- - •-, ,•• : • f • III i•' --- d

It
considered to ensure that the lap belt restraint remains effec-
tive regardless of seat position. If the seat includes longi-
tudinal load limiting, attachment of the lap belt to the basic
structure will not be practical because then most of the for-.
ward load will be car.ied to the aircraft structure through the
belt rather than through the seat. In such a case careful con-
sideration must be given to the belt strength since the belt
must restrain the motion of the seat, as well as the occupant.
The lap belt should be anchored to provide optimum restraint
for the lower torso when subjected to eyeballs-out (-G )
forces. One of the anchorage variables which has an infl~ence
on restraint optimization is the location of the lap belt an-
chorage in the fore-and-aft direction. The important charac-
teristic is the angle in a vertical fore-and-aft plane between
I
a projection of the lap belt centerline and the buttock refer-
ence line, or plane. This angle defines the geometrical rela-
tionship between the longitudinal and vertical components of
the belt load. A small angle provides an efficient path for
supporting longitudinal loads while a large angle provides an
efficient system for supporting large vertical loads. Thus,
for supporting large forward-directed loads, a small angle
would be desirable, but for reacting the large vertical loads
imposed on the lap belt by the loaded shoulder harness a large
angle is required. The compromise for location of the an-
chorage must consider all the variables including the tendency
for the occupant to submarine under the lap belt. In an acci-
dent with high combined vertical and longitudinal impact forces
the restrained body will tend to sink down into the seat (where
the magnitude of the displacement depends on cushion proper-
ties) and almost simultaneously be forced forward. This move-
ment is illustrated in Figure 49. If the lap belt angle is too
small the belt can tend to slip over the iliac crests of the
pelvic bone, allowing the pelvis to rotate under the belt. The
inertial load of the hips and thighs tend to pull, or subma-
rine, the lower torso under the belt. Lower torso restraint
is then accomplished through lap belt loading of the soft ab-
dominal portions of the body, possibly causing visceral injury
in addition to the spinal injury illustrated in Figure 49.
To counteract the tendency for submarining the lap belt angle

can be increased; however, the load in the belt increases for
a given torso deceleration because of the smaller longitudinal
loading component available to restrain the occupant. Further,
additional forward motion is allowed because of the increased
deflection of the webbing caused by the increased loading and
the greater forward rotation of the lap belt. However, as the
webbing is loaded, it presses down into the thighs of the occu-
pant and minimizes the possibility of submarining by picking
141
--
P' Sioulder !,arno~'s r'all
-------.. . ... . . .. . .. .- • ' -
* d ~ lxn.ctuo1 torso dis'laccfrnt
Figure 49. Pelvic rotation and submarining caused by

high longitudinal forces combined with
moderate vertical forces.
up the longitudinal component of restraint load required to

bring the system into equilibrium.
In order to avoid the increased possibility of both spinal and

abdominal injury, a properly designed restraint system should
not allow submarining to occur. Still, an efficient angle
should be maintained to limit the forward motion of the occu-
pant.
Comfort is another concern in lap belt anchor location. A pi-

lot must raise and lower his thighs during operation of rudder
pedals or antitorque pedals. If the lap belt anchor is too far
forward, the lap belt will pass over the pilot's thighs forward
of the crease between the thighs and the pelvis and thus may
interfere with vertical leg motion. It is important, there-
fore, to position the lap belt anchorage so that it provides -
optimum restraint while not interfering-with the pilot's oper-

ational tasks. A forward location of the anchor does not neg-
atively influence the comfort of passengers since passengers
are not required to perform operations with their legs.
In order to accomplish these objectives, the vertical angle

between the lap belt centerline and the buttock reference line
as installed on the 50th-percentile occupant should not be less
than 45 degrees and should not exce 5 degrees, as shown in
Figure 50(a). Further. it is deF Ae to locate the anchor
point at or below the buttock r, Bnce line for comfort and
142
a) All systems b) Systems with lap belt c) Systems without tiedown
tiedown straps straps
4 5 - 5 5V4
Seat reference L
uttock point ;typical) "7
reference
line 2.0 + .25
- 0
Figure 50. Lap belt anchorage geometry.
performance. If the anchor point must be located above the

buttock reference line, as on most armored seats, the anchor
point should be positioned to ensure that the belt-angle lies
within the desired 45- to 55-degree range. For a system having
a lap belt tiedown strap to counteract the upward force of the
shoulder harness (e.g., in pilot seats), the lap belt anchors
should be positioned so that the centerline of the lap belt
passes through the seat reference point as shown in Figure
50(b). If the restraint system does not have a tiedown strap
(e.g., in passenger seats), the lap belt anchor should be po-
sitioned so that the belt centerline passes through the buttock
reference line 2 to 2-1/4 in. forward of the seat reference
point as shown in Figure 50(c). This position provides suffi-
cient vertical load components to help counteract the upward
force of the shoulder straps. For positioning anchors that do
not fall on the buttock reference line, the angle between the
lap belt centerline and the buttock reference line can be as-
sumed to be 45 degrees for systems with tiedown straps and 55
degrees for those without.
Submarining can be reduced by ensuring that the lap belt is
tight, as shown in studies reported in Reference 82. Thus,
82. Roberts, V. L., and Robbins, D. H., MULTIDIMENSIONAL
MATHEMATICAL MODELING OF OCCUPANT DYNAMICS UNDER CRASH
CONDITIONS, Paper No. 690248, Society of Automotive En-
gineers, Inc., New York, January 1969.
143
-. -,
!I
* -'
-,1
care should be taken to train occupants to tighten the lap belt

to the maximum consistent with comfort and to not loosen the
belt anytime during flight.
For seats that limit lateral motion of the occupant with struc-
ture, such as in armored seats, the anchorage point and hard-
ware should possess sufficient flexibility and strength to sus-
tain design belt loads when the belt is deflected laterally
toward the center of the seat through an angle of up to 60 de-
grees from a vcrtical position. The side motion of fittings on
other seats should also be capable of supporting design loads
with the lap belt deflected laterally away from the center of
the seat through an angle up to 45 degrees from the vertical.
These recommendations are made to ensure that lateral loading
on the torso will not result in lap belt anchorage failure.
7.3.4 Shoulder Harness Anchorage

The shoulder harness or inertia reel anchorage can be located
either on the seat back structure or on the basic aircraft
structure. In placing the inertia reel, strap routing and pos-
sible reel interference with structure during seat adjustment
or energy-absorbing stroke of the seat must be considered. Lo-
cation of the anchorage on the basic aircraft structure will
relieve a large portion of the overturning moment applied to
the seat in longitudinal loading; however, due consideration
must be given to the effect of seat bucket movement in load-
limited seats. Vertical movement of the seat pan can be pro-
vided for by placing the inertia reel aft of the seat back
shoulder strap guide a sufficient distance so that seat verti-
cal movement will change the horizontal position and the angle
of the straps very little.
Shoulder straps should pass over the shoulders in a plane per-
pendicular to the back tangent line or at any upward (from
shoulders to pull-off point) angle not to exceed 30-degrees,
as illustrated in the upper-left sketch in Figure 51.
Any installation that causes the straps to pass over the shoul-
ders at an angle below the horizontal adds additional compres-
sive force to the seat occupant's spine as shown in the lower
sketch of Figure 51. A shoulder harness pull-off point at
least 26 in. above the buttock reference line is needed to en-
sure that the straps do not apply an excessive downward load on
the spine of a 95th-percentile occupant.
The shoulder harness anchorage or guide at the top of the seat

back should permit no more than 0.5-in. lateral movement (slot
no more than 0.5 in. wider than strap) to ensure that the seat
occupant is properly restrained laterally. The guide should
provide smooth transitions to the slot. The transition contour
144
4.
Correct
Backtanent
ineTorso carries small
M~aximum vertical load
shoulder
strap
angle,
-. Shoulder
strap load
v, (Force diagram)
2 6.0 to
271.0 in.k
Seat referec
point
Buttock reference / b~ s
line -
0)
Incorrect i
Torso carriei-large vertical load
Figure 51. Shoulder harness anchorage geometry.

should be of a radius no less than 0.25 in. and should extend
completely around the periphery of the slot to minimize edge
wear on the strap and reduce the possibility of webbing failure
due to contact with sharp edges under high loading.
7.3.5 La etTeon ta nhrg
When the upper body is thrown forward against the shoulder
straps, an upward pull is exe' - on the lap belt. If a lap
..... ........
I
belt tiedown strap is not being used, the tendency is for the
belt to be pulled up off of the iliac crests and into the soft
solar plexus area, causing injury to abdominal viscera as pre-
viously shown in Figure 49. A tiedown strap attached to the
center of the lap belt will prevent the upward belt movement.
It is recommended that the tiedown strap be located on the seat
pan centerline at a point 14 to 15 in. forward of the seat back.
For shorter seat pans, the anchor must be placed as far forward
as possible.
7.3.6 Adjustment Hardware
Adjusters must carry the full design load of the restraint sys- i
tem subassembly, of which they are a part, without slipping or
crushing the webbing between items such as locking cams and
the opposite locking surface. In extremely highly loaded ap-
plications, this may require that the strap be doubled in a
manner that requires the adjuster to carry only half of the
strap assembly load. The force required to adjust the length
of webbing should not exceed 30 lb in accordance with existing
military requirements for harnesses. Insofar as possible, all 3
adjustments should be easily made with one (either) hand. Ad-
justment motions should be toward the single-point release
buckle.
7.3.7 Location of Adjustment and Release Hardware
This hardware must not be located directly over hard points of
the skeletal structure, such as the iliac crests of the-pelvis
or the collarbones. The lap belt length adjusters should be
located either at the center of the belt near the release
buckle or at the side of the hips below the iliac crests, pre-
ferably the latter. The shoulder strap adjusters should be
located as low on the chest as possible in order to avoid con-
centrated pressure on the collarbones.
7.3.8 Webbing Width and Thickness Requirements
Selection of the optimum webbing width for a lap belt and
shoulder harness must be based on two conflicting requirements:
(1) maximum width for lowest pressure and (2) minimum width for
maximum comfort and minimum hardware weight. Webbing require-
ments are discussed in detail in Section 7.4.
7.3.9 Hardware Materials
All materials used for the attachment of webbing (release
buckles, anchorages, and length adjusters) should be ductile
enough to deform locally, particularly at stress concentration
points. Ductility in restraint harness hardware is not as cri-
tical when energy-absorbing provisions are incorporated into
146
- -- ....- .- -... . .. .......--.. ..-- •- -

the seat, because the maximum loading of the system is limited.
Thus, it would be possible to specify low-ductility materials
on load-limited seats and to specify high-ductility, moderate-
strength materials on nonload-limited seats. Such a specifica-
tion could possibly lead to the inadvertent installation of
low-ductility harness fittings on rigid, nonload-limited seats.
For example, it is known that 20-G-strength shoulder straps
have been mistakenly installed in place of 40-G straps. To
prevent such a possibility, it is concluded that, wherever ap-
plicable, all harness fittings should be made of equivalent
high-ductility materials to ensure their interchangeability.
A minimum elongation value of 10 percent (as determined by
standard tensile test specimens) is recommended for all metal
Sharness-fitting materials. The 10-percent elongation value
can be achieved with copper-base aluminum alloys, low-carbon
steels, and stainless steel. There are obviously some compo-
nents that, for operational purposes, rely on hardness. These
components should be designed to perform their necessary func-
tion but be made from materials as nearly immune as possible
to brittle failures.
7.3.10 Structural Connections
7.3.10.1 Bolted Connections: Safety margins of 15 and 25 per-
cent for shear and tensile bolts, respectively, are recommended
by most aircraft companies for the manufacture of basic air-
craft structure. These margins are intended to allow for mis-
alignment of holes, stress concentrations, and fatigue strength
reductions; however, the bolt's fatigue strength is not a fac-
tor for a one-time maximum loading as occurs in a crash. Thus,
it is concluded that the safety margins for shear and tensile
bolts in restraint systems can be reduced to 5 and 10 percent,
respectively.
Good aircraft engineering practice also dictates that bolts
less than 0.25 in. in diameter should not be used in tensile
applications because of the ease with which these smaller bolts
can be overtorqued. Wherever possible the bolts should be de-
signed for shear rather than tension.
7.3.10.2 Riveted Connections: Riveted joint design guidelines
are presented in MIL-HDBK-5, "Metallic Materials-and Elements
for Aerospace Vehicle Structures" (Reference 23). This hand-
book i s recommended as a guide for restraint system hardware -
design.
7.3.10.3 Welded Connections: Welded joints can be 100 per-
cent efficient; however, they may be only 50 percent efficient,
depending upon the skill of the welder. Since welded joints
can be completely acceptable and in some cases superior to
147
S .. . . : - .. - ., :•? k---4L *'-:_

bolted or riveted joints, it is not reasonable to prevent the
use of this type of joint if strict inspection procedures are
used to ensure that all welded joints are adequate. Welding
processes are discussed in Military Specifications MiL-W-8604,
-6873, -45205, and -8611. These specifications should be used
as guides to ensure quality welding.
Welded joints may contain stress concentration points and mis-
aligned parts in a manner similar to bolted joints; therefore,
the cross-sectional area of the basic material in a welded
joint should be 10 percent greater than the area needed to sus-
tain the design ultimate load.
7.3.10.4 Plastic Strength Analysis: Plastic analysis methods
should be used for strength determination wherever applicable
in order to obtain maximum-strength hardware at the lowest pos-
sible weight. Plastic analysis makes maximum use of the strain
energy available in ductile metals. References 28 and 29 cover
this subject.
7.4 WEBBING AND ATTACHMENTS
7.4.1 Properties
The maximum load to be sustained by restraint harnesses can be
determined by a review of seat load-deflection requirements
(Chapter 8). The curves shown there include the effects of
dynamic overshoot loads. The maximum load shown is 35 G for
the cockpit seat, where the seat structure provides for little
elongation. The required load is reduced as the deformation
is increased. Although the restraint harness could be designed
to varying loads in accordance with the energy-absorber G level
used in the seat, it is believed to be more practical and fool
proof to design a single-strength restraint harness that can be
interchanged with all seats of similar configuration and orien-
tation. The main advantage of a single-strength harness would
be the assurance that it could be interchanged between load-
limited seats and nonload-limited seats without fear that an
understrength harness might be installed. On this premise,-the
design strength of all forward-facing and side-facing~restraint
harnesses should be equal to or greater than the strength of
the cockpit seats because the higher design strength of the
cockpit seats would govern. At first, this solution might
seem to be too conservative because of the lower load levels
required for cabin seats; however, closer scrutiny indicates
that the assymmetrical nature of the forces on the harness in
the side-facing seats could result in loads just as high as
those experienced in the forward-facing cockpit harness for a
more symmetrical loading.
148
A __________11__________nln________I_____i-n
The distribution of the total load on the various harness com-
ponents is not easily determined; however, these forces have
ueen fairly well approximated by theoretical calculations and
by experimental test data. The test data have been obtained
from tests on restrained 95th-percentile anthropomorphic dum-
mies under a variety of test conditions. T he maximum design
loads for the various harness components attached on the seat 7
are listed in Table 4. These loads may appear to be higher
than necessary to offer restraint on a 35-G seat for a 222-lb
"occupant; however, these loads allow for (1) torso variations,
(2) asymmetric loadings, and (3) a safety factor to ensure that
the harness does not fail before the seat fails.
TABLE 4. RESTRAINT HARNESS COMPONENTS LOAD-ELONGATION

DESIGN AND TEST REQUIREMENTS (MIL-S-58095(AV))
Maximum elongation
Minimum load (design goal)
Harness components (lb)(a) (in.)(b) -
Inertia reel strap(s) 6 0 0 0 (c) (e)

Shoulder harness strap(s) 4 0 0 0 (d) 1 .5
Lap belt 4000 2.0

Lap belt tiedown strap 4500 0.5
NOTES: (a) Applied in straight tension.
(b) Total length of harness component tested must be
the same as when installed on the seat and ad-
justed for a 95th-percentile clothed occupant.
(c) This represents the total load from all shoulder
straps. A single diagonal shoulder strap should
carry 6000 lb.
(d) This represents the minimum load that one of two
shoulder straps should carry.
(e) This applies only to the shoulder harness and in-
ertia reel strap outsiLde-the reel (exclusive of
the webbing wound on the spool of the inertia
reel).
It will be noticed that the inertia reel strap is required to

carry 6000 lb when the inertia reel itself is designed to carry
only 4000 lb. Since no inertia reel spool failures have oc-
curred, and since combinations of shoulder strap loads exceed-
ing 6000 lb have been measured, it must be assumed that the
loads in the straps are reduced thrnugh friction with the body
a and guide slots to no more than 4000 lb at the reel.
149
The elongation of all webbing used in the harness must be mini-
mized to decrease overshoot. Dynamic tests conducted with an-
thropomorphic dummies and several tests with cadavers indicate
that a total elongation greater than about 1.5 in. will result
in an overshoot iad that is more than 25 percent above input
peak loads at tY.o floor. The load and deformation requirements
indicated for st.is in Chapter 8 are based on the use of a re-
straint harness that elongates only 1.5 in. away from the seat
back. Torso compression and "normal" harness slack, however,
account for another 2 in. (at a 35-G load). Therefore, the
total torso movement away from the seat back is about 3.5 in. . I
The value of 1.5 in. for harness elongation is less than half
of the total torso movement. A lower value would not be rea-
aonable unless the normal harness slack and tissue deformation
are also reduced. Table 4 shows that the shoulder strap elon-
gation is restricted to 1.5 in., while the lap belt is re-
stricted to 2.0 in. of total end-to-end stretch or 1.0 in. of(
loop elongation. Restraint systems for the new generation of
Army helicopters use a low-elongation polyester webbing, the
characteristics of which are listed in Table 5.
TABLE 5. RESTRAINT WEBBING CHARACTERISTICS
Nominal Minimum
Restraint webbing Webbing brek.ing
system width thickness streigth Elongation*
component (in.) (in.) (lb) (percont)
Inertia
reel 1-3/4 0.057 6980 6.9 @ 3000 lb
Shoulder
straps 2 0.057 7800 7.6 @ 4000 lb
Lap belt 2-1/4 0.057 8880 7.8 @ 4000 lb

Lap belt
tiedown -
strap 1-3/4 0.057 6980 6.9 @ 3000 lb
*Based on 10-in. gage length.
150
S.. .. . .. _ _-
. - . -,
II
I%
. Dynamic testing of polyester webbing has demonstrated the dy-

namic elongation to be approximately 60 to 75 percent of the
static (References under
ure 52 elongation 83).same load, as illustrated in Fig-
63 andthe
7.4.2 Width and Thickness Requirements

Selection of the optimum webbing width for a lap belt and
shoulder harness must be based on two conflicting requirements:
*- (1) maximum width for lowest pressure and (2) minimum width
for maximum comfort and minimum hardware weight. The widths
* specified in Table 6 are believed to be a good compromise be-
tween these conflir[ing requirements. All webbing used for
restraint harnebs,. r st be thick enough to ensure that the
webbing does not'.A.. or crease to form a "rope" or present a
thin sharp edge under high loading that will cause damage to
soft tissue. Such damage is more likely to occur in the neck
region during a lateral loading or, in the pelvic region during
a forward loading. Although requirements based on early inves-
tigations using nylon webbing specified a minimum thickness of
0.090 in., it has since been determined that state-of-the-art
webbing materials must be thinner in order to achieve the de-
sired low elongation. No significant problem of injuries
caused by the thin webbing has been observed with this low-
elongation webbing which has seen extensive automotive use.
Therefore, based on currently available materials, a minimum
thickness of 0.055 in. is considered acceptable.
7.4.3 Webbing Attachment Methods
7.4,3.1 Stitched Joints: The strength and reliability of

stitched seams must be ensured by using the best known cord
sizes and stitch patterns for a specified webbing type. The
stitch patterns and cord sizes used in existing high-strength
military restraint webbings appear to provide good performance.
The basic stitch pattern used in these harnesses is a "W-W"
configuration for single-lapped joints. Research by the U. S.
Naval Aerospace Recovery Facility (NARF) at El Centro, Cali-
fornia, has reaffirmed the adequacy of basic "W-W" stitch pat-
terns; however, the research also revealed that a larger--size
cord with fewer stitches per inch gave superior performance to
the No. 4 MIL-T-7807 cord currently being used. On the basis
of this research, the 50-lb strength No. 6 ccrd at 4-1/2 to 5
83. Kourouklis, G., Glancy, J. L., and Desjardins, S. P., THE

DESIGN, DEVELOPMENT, AND TESTING OF AN AIRCRAFT RESTRAINT
SYSTEM FOR ARMY AIRCRAFT, Dynamic Science, Division of Ul-
trasystems, Inc.; USAAMRDL Technical Report 72-26, Eustis
Directorate, U. S. Army Air Mobility Research and Develop-
ment Laboratory, Fort Eustis, Virginia, June 1972,
AD 746631. 151
8000
7250
000Failure
000/ 6100
.0 Dynamic,'
700 -
5000 /
4000 -
S/ /I_,
100
0 5 10 15
Elongation, 04percent
Figure 52. Load elongation characteristics for

MIL-W-25361 (Type II) polyester
webbing for static and rapid loading
rates.
I
TABLE 6. MINIMUM WEBBING WIDTH REQUIREMENTS
Minimum width
Webbing identity Ain.)
Lap belt 2-1/4'

Shoulder strap 2.0
Tiedown strap 1.5
*A greater width (up to 4 in.) or pad is de-
sirable in the center abdominal area.
152
- . .-. . . . ., . "
stitches per inch is recommended, as illustrated in Figure 53,
for use on MIL-W-25361 webbings. Also, the heavier cord can
be expected to provide better resistance to sunlight degrada-
tion and abrasion. The use of the 50-lb cord and an 80-percent
efficiency results in a minimum strength of 160 lb/in. (4
stitches x 50 lb/stitch x 80 percent) for a single-lapped joint
or 320 lb/in, for a looped joint. Thus, the total stitch
length needed can be determined by the total required load.
4-1/2 to 5 stitches per inch

MIL-T-7807B, No. 6 nylon cord
(Type I or 1I, Class I, 50 ib)
Metal or equivalent seam strength
hardware with smaller diameter cord for
new thinner webbing.
wiThe
0.12 in. - . 0 o25-in.
spacing
S4-1. 5-in. [- -- 8One stitch
minimum minimum
Figure 53. Stitch pattern and cord size.
It has been shown recently that the heavier thread is not com-
patible with the new low-elongation polyester webbing (Refer-
ence 84). For these webbings, a smaller diameter cord offers
the advantages of reduced webbing fiber damage and the ability
to be used with automatic sewing machines and is therefore ac-
ceptable.
The strength of stitched joints can be expected to decrease
with age because of normal weather exposure and because of the
normal dust and grit collection between the webbing surfaces.
The grit and dust can gradually abrade the cords over a period
of time. The use of a 30-percent increase in the total stitch
length required is recommended to offset the normal aging
84. Farris, L., HIGH STRENGTH STITCHING FOR AIRCRAFT PERSONNEL
RESTRAINT SYSTEMS, Pacific Scientific Co.; Proceedings,
1978 SAFE Symposium, Survival and Flight Equipment Asso-
153
| ~I I
DI
strength decrease as well as the possible abrasion strength j

decrease. Covering the stitched joints with cloth to provide
wear protection for the cords is also recommended.
An example of establishing the total seam length is given:
Assume: A single-lapped joint, 50-lb cord strength, with a
4000-lb joint load.
Then, the minimum stitch strength is
(50)(4)(0.80) * 160 lb/in.
and the minimum seam length is
4000
4000 - 25 in.
160
Therefore, the total seam length is
(25)(1.3) - 32.5 in. I
The total seam length is achieved through placing many short

lengths in a rather small area. Several patterns have been
developed and tested; however, the W-W as described below is
still preferred. The size of the overlapped and stitched area
should be minimized to reduce weight, reduce the stiffened sec-
tion of the webbing, and provide more room between fittings for
adjustment.
Unpublished data from comparative tests of five stitch pat-

terns performed by NARF indicated better performance of two
new stitch patterns over the basic W-W pattern. The data from
this research are reported here with permission of NARF. -
The five stitch patterns tested are shown in Figure 54. These
patterns were sewn in Types XIII and XXII of MIL-W-4088 nylon
webbing used for parachutes. Three samples of each stitch pat-
tern were tested. Table 7 shows the results of the first test
series. Because of the low number of total stitches, the re-
sults were inconclusive, and a second test series was per-
formed. Patterns 2 and 5 were eliminated from the second
series. Table 8 shows the results of the second test series.
It relates the performance of the two stitch patterns, I and
154
I
• - -,'
- --.--- -_<,
<.---------------- "-"-- ------ -- ,-i
" ~~~Pattern (j-
5 Stitches/inch
_ HI
S ...... - .-- "
Pattern
Pattern I -• - - - -I
II
2 -- - -- - -- - - - II
7 Stitches/inch
Pattern --
5 Stitches/inch
I-- - - - --.----- ---
Pattern 5 S
L-- T•-- -- -- -- --- ~.w--

5 Stitches/inch--3
Pattern I . " --- I

I . .I -- -" I
5 Stitches/inch
Li
Figure 54. Stitch patterns tested.
'4
S' 155
- .....
• . . ............---. ,_
--. ."... , •
S0 0 0.1E
In Ln in Q
3,0 n V Ivi LA m EN
f
P) N &A00 0
t0 r- 0 0
-' 0
%D %D In
0 n In In In .
fn0 0 t0o
O 0
14
0 InM
o0 t-
n
'0
01
CID
n
~0
mIE'
4'l
In A N
.o 0N in 0
go 0 In In 'A In 0D E N
vd ýq 0~r4 ; to
A In 0 LA OD 0 o
vn In O 0 %0 Go 0
4 In _0 -0 4 4
m0 v OD 0%
.0 n 0 n 0
Ln 0 41
4~ ~~~~
0 0 0 * 04 0-
0 4 0
C4 4 E r.0
0 -0
00 co
v v r-4
%0 In OD 0 H
4 in IV In IV 9 4 EN m r
OO)
in0 ) In i
:
V qn 41 0 .NA4 N 0
M4 cc w
.4 4 4
0 .0o w4 A).go 0 w4
IA .14 -.4
156
"TABLE 8. BREAKING STRENGTH OF STITCH PATTERNS (TEST SERIES TWO)
Webbin and stitch

Sample (a) (a) (a) (a) (a) (a)
1 4400 4410 4 5 4 0 (b) 6340 6420 6215
Breaking 2 4710 4740 5080 6480 6490 6060

strength
(ib) 3 4820 4360 4870 7200 6500 6070
Average breaking
strength (ABS) (lb) 4643 4503 4830 6673 6470 6115
ASS/ABS for
pattern 3 1.03 1.00 1.07 1.03 1.00 0.945
Approximate
total stitches 260 270 270 260 270 270
ABS/stitch (lb) 17.86 16.68 17.89 25.67 23.96 22.65
ABS/stitch/ABS
for pattern 3 1.07 1.00 1.07 1.07 1.00 0.945
(a) A designates MIL-W-4088 Type XIII nylon webbing.
B designates MIL-W-4088 Type XXII nylon webbing.
Numerials 1, 3, and 4 designate stitch patterns as shown in
Figure 54.
(b) Jaw separation 20-in. minimum. All other tests at 2-in.
minimum.
4, to the performance of pattern 3, the W-W pattern, for the

two different types of webbing. Stitch patterns 1 and 4 ex-
hibited better strength properties than pattern 3 (W-W) when
Type XIII webbing was used. Pattern 4 did not perform as well
when Type XXII webbing was used, while pattern 1 again indi-
cated better strength characteristics than did pattern 3.
The W-W stitch pattern as shown-Ln Figure 53 is still recom-
mended until more conclusive information on these or other
stitch patterns becomes available.
157
- '
Ii
7.4.3.2 Webbing Wrap Radius: The wrap radius is the radius
of the fitting over which the webbing is wrapped at buckles,
anchorages, and adjusters, as illustrated in Figure 55. De-
tailed information on just how small this radius can be before
the strength of the webbing is affected is not availablel how-
ever, the 0.062-in. minimum radius shown is based upon the
geometry of existing high-strength restraint harnesses. This
radius should be carried around the ends of the slot as shown
in Figure 55 to preclude edge cutting of webbing if the webbing
should be loaded against the slot end.
A0
webbing
minimum
typical.
Wrap radius (0.062-in. minimum)
should be carried completely
around ends of slot as shown by
the heavy line above
DETAIL A
Figure 55. Wrap radius for webbing joints.
7.4.3.3 Hardware-to-Webbing Folds: A possible method of re-

ducing fitting width at anchorage, buckle, or adjuster fittings
is to fold the webbing as shown in Figure 56. This reduces
the weight and size of attachment fittings; however, it can
also cause premature webbing failure because of the compressive
force applied by the top layer of webbing to the lower against
the fitting slot edge. If this technique is to be used, tests
to demonstrate integrity are recommended. Also, for configura-
tions that require two load paths, such as lap belts, where an
adjuster cannot hold the required 4000-lb load, the webbing is
looped through a full-width slot which halves the load in each
strap. An adjuster is then included in one strap. Adjustment
requires that the webbing be freely drawn through the fitting,
a requirement that folded webbing cannot meet.
7.4.3.4 Surface Roughness of Fittings: A surface roughness
of no more than RMS-32 is recommended to prevent fraying of
the webbing due to frequency of movement over the metal.
158
'I
|H
fi "iii'' ii'
'°i i'l ii' i' ~i i'~i'F':• '- j .... _a,,-, ' -. . .
F
0
-El
Figure 56. Webbing fold at metal hardware attachment.
7.4.4 Energy-Absorbing Webbing

Energy-absorbing restraint system webbing has been considered
for limiting loads on the occupant. The potential advantages
of energy-absorbing webbing are: (1) reduction of maximum load
exerted by the webbing on the occupant and (2) reduction of the
amount of elastic energy stored in the webbing. Webbings of
this type have been developed and are described briefly here
for information purposes. They are not recommended for use in
seating systems for the reasons presented below.
The principle of energy absorption for the first webbing mater-
ial depends on a core wrap of fiberglass that breaks at a de-
sign loadl then, the outer cover of nylon wrap takes over the
loading, gripping the fiberglass until it breaks again. The
construction of the webbing varies, depending on the type of
force-versus-percent-of-elongation curve desired. For this
webbing, the general shape for the force-versus-elongation
curve includes a linear elastic region followed by a region of
constant force.
The construction of the second type of e iergy-absorbing web-
bing differs greatly from the first. It is made of polyester,
159
- ~ - - - --. ---
Ii-_______________________________________________ C
r . . .. ..... _- j - . . ..
• I I I I I I I I II I |
and the energy absorption is produced by the filaments them-

selves. The polyester filaments are heat shrunk from their
original sizes; and they do not return to the shrunk dimen-
I
sions after the load application. This has the effect of
plastic deformation; and this property provides the energy-
absorption capability of the material. The general shape for
the force-versus-elongation curve for this webbing is a con-
stant rate in pounds per inch which makes inefficient use of
stroke distance.
A third type of energy-absorbing webbing material has been
evaluated for parachute applications at the U. S. Naval Aero-
space Recovery Facility. The material is made by stitching
together two pieces of webbing. The two pieces of webbing
separate (peel) at a constant load by breaking the stitches
holding them together. The constant breaking force can be
varied by increasing or decreasing the number of stitches. Be-
cause of its construction, the material does not appear to be
suitable for use in aircrew restraint systems.
Because of other considerations, including primarily the in-
creased potential for secondary impacts of occupants, energy-
absorbing webbing is not recommended for use in seating systems.
The limited room available in aircraft requires that the strike
envelope be minimized. Therefore, the use of the lowest elonga-
tion available is specified.
7.5 RESTRAINT SYSTEM HARDWARE
7.5.1 General
The restraint system configured for use in a particular loca-
tion in an aircraft will include various hardware selected on
the basis of a trade-off among such factors as crashworthiness,
weight, and cost. An aircrew system meeting the requirements
of MIL-S-58095(AV) that has been developed is illustrated in
Figure 57. The system shown in Figure 58, which is defined by
a draft military specification (Reference 85), offers improved
1
protection but is heavier and more expensive. For example, it
includes two inertia reels for the reflected shouluer strap
system, which reduces both lateral and forward motion. Its
use may be warranted where space is a problem and strike envel-
opes need to be minimized. Also, this system's use of lap belt
retractors rather than adjusters provides greater convenience
in ingress, greater comfort by eliminating the adjuster, and
greater crash safety by eliminating slack (preload held on the
lap belt by torsional spring in retractor). The weight of the
85. Proposed Draft Military Specification,
RESTRAINT SYSTEM, AIRCREW, September 1974.
160
I
S • . .. ... w. _- .
Inertia reel
Inertia reel strap

'-l- -. 75-in. 'ide
webbing
2.00-in. wide webbing

Shoulder k I,
harness Adj uster
2.00-in. wide
webbing
!'• • •...Lower shoulder
Lap belt-/ N • • Lap belt•

2.25-in. wi e-- Ii,:, aho
webbing Fixed fitting on
tiedown strap
• Tiedown strap
'•-1.75-in, wide webbing
Tiedown anchor
Figure 57. Aircrew restraint system.
system shown in Figure 57 is 5.50 lb and that of the system in

Figure 58, 8.50 1b, with the difference due mostly to the addi-
tional inertia reel and the two lap belt retractors of the
latter system.
The various hardware components involved in a state-of-the-art
restraint system are described below. information on produc-
tion items is included where available.
161
-I - --- - --
1.75-in, wide Inertia reel
webbing
Reflected
10, .,Astrap
Ref lected strap
Roller fitting
'A djuster
shoulder 5
2 .00-in, wide
strapwebbing
!,I Plug-in fitting

Rtary buckle
Rtator .4vi Fixed fitting
wide
Tiedown strap webbing
1.75-in. wide webbing
~4-Tiedown anchor
Figure 58. Reflected shoulder strap restraint

system. (From Reference 85)
162
7.5.2 Buckles and Emergency Release
The buckle is of the single-point release type for all systems

and provides positive release of all strap fittings (with the
exception of the one to which it is permanently attached).
These capabilities should help prevent entrapment of a wounded
occupant.
7.5.2.1 Aircrew Restraint Buckle: To facilitate egress in
emergencies, a rotary-release buckle provides the advantage of
operation by a force applied in many directions. In one exist-
ing buckle, each fitting can be inserted and locked separately.
When the release handle is rotated, springs move the fittings
far enough so that none will reengage when the handle is re-
leased. This is an example of a desirable feature that will
protect against a potential hazard created by a fitting relock-
ing upon release of the handle. For example, if one lap belt
fitting relocks, it could partially restrain the occupant as he
attempts an emergency egress.
7.5.2.2 Troop/Passenger Restraint nuckle: The restraint sys-

tems recommend d for troop seat insta-llations, as shown in Fig-
ure 45, include a single-point, lift-lever release buckle that
is permanently attached to one of the lap belt straps. The
lift-lever release mechanism replaces the rotary release here
for the convenience of troops or passengers who, because of
infrequent system use, might find it easier to use in emergen-
cies since it resembles automotive hardware (References 63, 78,
and 86). The design of such a buckle is described in Refer-
ence 78.
7.5.3 Adjustment Hardware
The lap belt-length adjusters are located either at the center

of the lap belt near the attachment-release buckle or at the
side of the hips of the occupant below the iliac crests of the
pelvis. Shoulder strap adjusters are located as low on the
chest area as possible to avoid a concentrated pressure over
the collarbones of the seat occupant. It is possible for the
seat occupant to make strap adjustments easily -with either
hand. A downward pull on-the free end of the shoulder harness
straps tightens the shoulder harness, and an upward pull on the
* free end of the lap belt straps tightens the lap belt.
86. Reilly, M. J., ENERGY ATTENUATING TROOP SEAT DEVELOPMENT,

The Boeing Vertol Company; Report NADC-AC-7105 with Adden-
dum NADC-73121-40, U. S. Naval Air Development Center,
Aerospace Crew Equipment Department, Warminster, Pennsyl-
vania, May 1971.
163
7.5.4 Inertia Reels, Control, and Installation
7.5.4.1 Inertia Reels and Controls: Inertia reels currently
installed on the crewseats of U. S. Army aircraft are designed
in accordance with the requirements in Reference 87. The design
requirements specified in MIL-R-8236 are compatible with the
other restraint harness requirements listed in this chapter,
and it is recommended that the use of this specification be
continued.
Some discussion of inertia reel function is included here to

explain the two basic types of reels listed in MIL-R-8236. The
basic function of the inertia reel is to give the crewmember
full freedom of movement during normal operating conditions
while automatically locking the shoulder harness during an
abrupt deceleration.
The freedom of movemei'... is obtained by spring-loading the reel
cable or webbing to which the shoulder straps are attached.
.This allows the shoulder harnes to be extended without appar-
ent restraint of the shoulders (only 6 lb at maximum exten-
sion). The reel will be constantly taking up any slack.
There are two basic types of MIL-R-8236 reels. The impact -

sensitive type requires a 2- to 3-G impact on the rtia reel
housing itself to lock automatically. Normal flight loads,
including severe turbulence, will not activate this reel.
The rate-of-extension type reel, although mechanically differ-
ent, serves the same purpose. Its automatic operation depends
on the rate at which the inertia reel strap is reeled off,
which makes it a function of the rate of upper torso displace-
ment away from the seat back, regardless of direction. The au-
tomatic operation of this reel can be checked at any time by a
jerk on the shoulder straps. The shoulder harness, after being1
locked automatically, reels up the slack in the strap every
time the occupant bounces back toward the seat back. Even-
tually, the- occupant will find himself fftmly locked against
the seat back.
A few words about the manual control lever ,ill clear possible
misconceptions about its use. Both types ot reels have iden-
tical control levers, usually mounted under the seat pan, on
the seat side, or at some other convenient location. The lever
has two positions: manual and automatic. The manual position
I
87. Military Specification, MIL-R-8236, REEL, SHOULDER HAR-
NESS, INERTIA LOCK, Department of Defense, Washington,
D. C.
164
S. . ........ .4
.....
.4
permits the pilot to lock the reel if rough conditions are an-
ticipated, or at any other time warranted. Normally, the con-
tLol lever should be in the a'itomatic position to allow the
wearer to lean forward easily and reach all controls without
first having to release the control lever. MIL-R-8236 requires
that both reel types lock automatically before the occupant
travels more than 0.5 in. during an emergency deceleration.
In &ddition to the MIL-R-8236-type reel, which has the function
of preventing further strap extension, there are power-haulback
reels, which rapidly retract slack to apply a tensile load to
the belt, Generally, these systems, some of which use a basic
MIL-R-8236 inertia reel, are powered by a gas generator and
must be manually actuated prior to impact. Automatic actuation
by an acceleration sensor is not recommended because human tol-
erance considerations limit the haul-back velocity. By the
time the crash could be sensed, there would not be time to com-
plete the haulback within tolerable accelerative limits.
7.5.4.2 Inertia Reel Installation in Rotary- and Fixed-Wing
Aircraft: Accident statistics indicate that rotary-wing air-
craft frequently impact on their sides, or impact vertically
with little longitudinal de- -ration. Therefore, it is con-
cluded that all rotary-wir . VTOL aircraft should incorpo-
rate the rate-of-extensior reel, because a unidirectional
(-G ) acceleration (needec -o actuate the impact type reel)
might not :e present in all rotary-wing or VTOL aircraft acci-
dents.
On the other hand, the study of about 92 fixed-wing aircraft
accidents, described in Volume II, revealed that only one acci-
dent occurred in which no longitudinal (-G ) acceleration was
present. On this basis, it is concluded tiiat a unidirectional-
(impact) type reel would be adequate for fixed-wing aircraft.
However, it is recommended that the rate-of-extension type reel
be used on all aircraft types to assure locking regardless of
load direction.
The inertia reel may be anchored to the seat back structure or
to the basic aircraft structure. The shoulder straps must be
maintained at the correct angle with respect to the wearer's
shoulder at all times as described in Section 7.3.4. If an
anchorage to basic structure is used, consideration must be
given to the possible seat bucket motion so that the shoulder
strap angle or length does not change by a significant amount
during energy-absorbing stroke. The reel should be mounted
and the webbing routed so that the webbing does not bear on
the reel housing. Excessive webbing loading of the housing
can produce housing and/or webbing failure as the housing is
not designed as a contact surface for loaded webbing.
t 165
8. SEAT STRENGTH AND DEFORMATION REQUIREMENTS
8.1 INTRODUCTION
Previous sections of this volume have presented background in-
formation to aid in understanding the problems involved in de-
-
signing
presents crashworthy seats and
specific design and test
restraint systems.for This
requirements seat chapter
systems
and litter systems. Occupant sizes and weights to be used in
the design are defined, as are the required static design
strength-deformation relationships. Static tests to demon-
strate the adequacy of the system in all loading directions
are presented. Finally, dynamic test requirements, to demon-
strate that the seat systems, restraint systems, and litter
systems will provide the degree of protection desired, ar.; also
defined. Successful completion of all static tests and dynamic
tests are required to demonstrate acceptability of a design.
In this chapter, the direction of applied loads are referred
to in terms of forward or aftward, lateral or vertical, and
upward or downward. These terms, together with aircraft and
occupant axes, are defined in Chapter 2, Definitions, and refer
to seat loading in directions consistent with the aircraft
coordinate system. Thus, a forward load on a forward-facing
seat is in the positive x direction with respect to both the
seat and the aircraft. If the seat is a side-facing seat, the
forward load would be applied to the seat in the plus-or-minus
y direction, depending on whether the seat faces right or left
respectively in the aircraft. For an aft-facing seat, the for-
ward load would be applied in the negative (-x) direction (to-
wards the back of the seat).
8.2 RECOMMENDED OCCUPANT WEIGHTS FOR SEAT DESIGN
It is recommended that the upper and lower limits of occupant
weights to be considered in seat design be based on the 95th
and 5th percentiles. Ideally, seat stroke limits should be
sized for the 95th-percentile occupant while the occupant ac-
celeration limits should be determined for the 5th percentile. -
If this were done, the resistive forces would be tolerable
while the stroke lengths would also be adequate for all occu-
pants in the design range (5th through 95th percentile). How-
ever, in most situations sufficient stroke distance will not
be made available in the aircraft to permit using the ideal ap-
proach; therefore, compromises will have to be made. Specific
criteria for these cases are presented in this chapter and, for
the present, should be viewed as minimum state-of-the-art de-
sign goals.
166
.- I- -.----
. . ~. .--------- --- ~-- -. . .- - -i
4
8.2.1 Crewseats
The design weight should be based on the typical weight of the
seated occupant, not the extremes. Although the weight of a
95th-percentile, combat-equipped aviator can be as high as
250 lb, it is believed that a majority of the flight hours log-
ged in Army aircraft over the past 20 years have been noncombat
hours. Consequently, it is probable that crewmembers will be
lightly equipped. The restrictions placed on crewseats, in-
cluding stroke length, control access, and seat armor, limit
the flexibility of design options. If the crewseats were to
be designed to protect occupants over the full range of weights
(144 to 250 lb), a weight-sensitive energy-absorbing system
would be mandatory. Thus, the typical aviator weight recom-
mended for crewseat design should not include combat gear.
Based on data of Reference 88, typical aviator weights are pre-
sented in Table 9.
TABLE 9. TYPICAL AVIATOR WEIGHTS
95th- 50th- 5th-

percentile percentile pe,:zntile
weight weight weight
Item (lb) (lb)
Aviator
(Reference 88) 211.7 170.5 133.4
Clothing 3.1 3.1 3.1
Helmet 3.4 3.4 3.4
Boots 4.1 4.1 4.1
Total
weight 222.3 181.1 144.0
Vertical -
-effective
weight 175.2 142.3 112.6
88. Churchill, E., et al., AZ1THROPOMETRY OF U. S. ARMY AVIA-

TORS 1970, Anth'ropology Research Project; USANL Technical
Report 72-52-CE, U. S. Army Natick Laboratories, Natick,
Massachusetts, December 1971, AD 743528.
167
(it _ _
I
Variable-load energy-absorbing systems are highly desirable to

maximize efficiency and thus, protection in limited space.
Therefore, they should be incorporated in seat designs when-
ever possible.
8.2.2 Troop and Gunner Seats
The same percentile range of occupant sizes should be consid-
ered for troop and gunner seat designs. A greater variation
of clothing and equipment is used by troops than by aviators;
troop seats should be designed to accommodate them. The 95th-
percentile occupant should be considered heavily clothed and
equipped, while the 5th-percentile occupant should be consid-
ered lightly clothed and equipped. Based on data contained in
References 31, 32, 63, 79, and 89, the typical weights of
seated troops in aircraft are shown in Table 10.
8.3 STRENGTH AND DEFORMATION
8.3.1 Forward Loads
In Section 4.7, it was shown that for a load-limited system

there is a minimum displacement that must be achieved if the
system is to remain in place during a given deceleration pulse.
Actually, all systems are load limited, although not necessar-
ily through original intent. The inherent load-deflection
curve for any system imposes a definite limit on the system's
ability to resist impulsive loading. The objective of inten-
tionally load-limited seat systems is to make the best use of
the space available for relative displacement of the seat and
occupant with respect to the airframe, while maintaining loads
on the occupant consistent with the type of restraint system
used and the occupant's capacity to survive the loads imposed.
The basic data used in developing the seat design curves pre-
sented in Figure 59 were obtained through a computer simula-
tion of the seat/occupant system (see Reference 25) and from
the results of static and dynamic seat tests (References 31,
32, 63, and 79) using body blocks and eanthropomorphic dummies,
respectively. The computer simulation allowed the calculation
of the seat displacement for given load-limiting values. The
simulator included a realistic kinematic behavior of the occu-
pant and the nonlinear effects of the restraint system. It is
estimated that the requirements given in Figure 59 are not con-
servative for the input pulses selected for design purposes.
89. U. S. Army, THE BODY SIZE OF SOLDIERS - U. S. ARMY ANTHRO-
POMETRY - 1966, USANL Te-hnical Report 72-51-CE, U. S.
Army Natick Laboratories, Natick, Massachusetts, December
1971, AD 743465.
168
TABLE 10. TROOP AND GUNNER WEIGHTS
95th- 50th- 5th-

percentile percentile percentile
weight weight weight
Item (lb) (ib) (Ib)
Troop/Gunner
weight
(Reference 89) 201.9 156.3 126.3
Clothing
(less boots) 3.0 3.0 3.0
Boots 4.0 4.0 4.0

Equipment 33.3 33.3 33.3
Total
weight 242.2 196.6 166.6
Vertical
effective
weight
clothed 163.9 127.4 103.4
Vertical
effective
weight 136.7
equipped 197.2 160.7
These are a 30-G peak triangular pulse of 50-ft/sec velocity

change in the cockpit and a 24-G peak with 50-ft/sec velocity
change in the cabin area.
The static loads that the seat must withstand are obtained by
multiplying the load factors (G) shown in Figure 59 by the sum
of the total weight of the 95thzpercentile crewmember or pas-
senger plus the weight of the seat and any armor or equipment
attached to or carried in the seat. For crewseats, the weight
of combat gear is not included (see Section 8.2.1).
Longitudinal displacement of approximately 6 in. for cockpit
seats and 12 in. for cabin seats measured at the seat refer-
ence point (the seat reference point may be projected to the
outside of the seat pan for measurement convenience) is the
practical limit for seats in existing Army aircraft. Since
169
* __ .~ . .-
6L1
4 45
FW I
40 -Acceptable failure FD;
A area
~ "i AceptabExaplerformac
ý Ccki set fowad-acn set
30
a /SamlAcceptabes faiur areace,
W 25
~o 20 Base c___
S10
,:4~ ~Cockit seats

0 .....
A ~ A / ~Ž
mcepasured fatluseatreaeec on
:1.7
25-. - - _ __ _ _ _ _ _ __ _ _- -.-
there is typically more room available in cabins than in cock-
pits, the advantages o! longer energy-absorbing strokes can
usually be achieved. Longer strokes permit the absorption of
seat weight and increase theloads
equivalent energy a& lower levelandof thus can serve
protection to z Auce
offered over
a wider occupant weight range.
In viewing Figure 59, it can be seen that for cabin seats 12 in.
of stroke enables the minimum limit load to be reduced to 15 G;
whereas, for cockpit seats a 20-G minimum limit load is re-
quired with only 6 in. of stroke. The 15-G and 20-G minimum
limit loads fix the G levels of the base curves for the cabin
and cockpit seat respectively. The available stroke will be
unique for each specific aircraft, and the energy-absorbing
mechanisms in the seats should be compatible with the available
stroke distances. If forward or sideward motion threatens to
limit the effectiveness of the vertical energy attenuating sys-
tem or increase the possibility of severe injury caused by sec-
ondary impact of the occupant with items in the aircraft, then
energy-absorbing stroke in directions other than vertical
should not be used. The 6 in. and 12 in. allowed by the curves
of Figure 59 should be viewed as maximum distances which are
subject to limitations of available space in each specific air-
craft and location in the aircraft.
The initial slope of the cockpit seat base curve to 0.75 in.
of deflection allows for elastic deformation consistent with a
relatively rigid crewseat while the lighter weight and more 4
flexible troop/gunner seat requires a lesser slope. The 30-G
and 35-G upper cutoffs reflect consideration of human toler-
ance limits, load variations between cockpit and cabin loca-
tions, and practical limitations of seat weight and excessive
airframe loading.
I
8.3.2 Use of Design Curves
To be acceptable, a seat design must have a characteristic
load-deflection curve that rises to the left and above the
base curves of Figure 59 and extends into the region beyond
the upper curve. This discussion also appliesýto the lateral
strength and deformation requirements discussed in Section
8.3.6. In Figure 59, curves A, C, and E are acceptable curves,
but curve B is unacceptable because it does not reach the re-
quired ultimate strength. Curve D reveals inefficient use of
seat deflection by intruding into the base area. The seat is
deflecting at too low a load, thus absorbing less energy than
it could.
171
wi W ~
8.3.3 Aftward Loads
Large aftward loads seldom occur in fixed-wing aircraft acci-

dents but may occur in rotary-wing accidents. A capability to
withstand 12 G is recommended for aftward loads for all seats.
This value will usually be automatically met by all seats meet-
ing the forward load requirements. Occupant weight should be
the total weight of the 95th-percentile crewmember or trooper
as presented in Section 8.2.
8.3.4 Downward Loads
Human tolerance to vertical impact limits the acceptable forces

in the vertical direction for all aircraft seats. The maximum
allowable headward acceleration (parallel to the back tangent
line) for seated occupants is on the order of 23 G for dura-
tions up to approximately 0.006 sec. Since most back tangent
lines are oriented at a backward leaning angle of 13 degrees
from the vertical aircraft axis, tolerance to vertical impact
loads should be somewhat increased over the stated criteria.
In spite of this, however, the 48-G design pulse imposes the
requirement for energy absorption in the vertical direction by
some form of load limiting. The vertical dynamic response of
seat-occupant systems and, in particular, the effect of seat
behavior on the occupant deceleration excursions, has not been
sufficiently investigated to allow a full explanation of the
effects of this phenomenon. The factors affecting the response
of the seat and occupant and thus the final design of the load-
limiting system include:
"• Input pulse variables.
"* Orientation of the occupant and seat relative to the
resultant force vector.
"* Effective occupant weight.
"* Occupant spring rate and damping characteristics.
"* Weight of the movable part of the seat.
"* Spring rate and damping characteristics of the seat.
-e Spring rate and damping characteristics of the cush-
ion.
* Available stroke distance.
172
* Force-deflection characteristic of the energy-
absorption system.
* Any external influences such as those caused by loads
transmitted through dummy legs, or binding of the
seat mechanism.
Army-sponsored research in these areas is currently underway
and results will be incorporated into this document when avail-
able.
The effective weight in the vertical direction of a seated oc-
cupant is approximately 80 percent of the occupant's total
weight because the lower extremities are partially supported
by the floor. The effective occupant weight may be determined
by summing the following:
* Eighty percent of the occupant's body weight.
* Eighty percent of the weight of the occupant's cloth-
ing (less boots).
* One hundred percent of the weight of any equipment
carried on the body above knee level. Combat gear
is not included in the effective weight of the pilot
or copilot (see Section 8.2.1).
The dynamic limit load for the load-limiting system should be
established by use of a load factor fGL) of 11.5. The dynamic
limit load is determined by multiplying the summation of the
effective weight of the seat occupant, and of the movable or
stroking portion of the seat, by 11.5. The resulting dynamic
limit load includes the total force resisting the vertical
movement of the seat in a crash; the dynamic limit load of the
energy-absorption system, simple friction, friction due to
binding, etc. This requirement is difficult to satisfy with
a sliding guidance system because the frictional load varies
with contact load which, in turn, varies with the impact load
vector direction. A relatively friction-free rolling mechanism
or collapsible structure is therefore recommended.
The 11.5-G design cftternon, taken from Reference 24 and modi-
fied to provide a tolerable deceleration of the 5th-percentile
occupant, considers the dynamic response of the seat and occu-
pant. The factor of 11.5 was established to limit the decel-
erative loading on the seat/occupant system to less than 23 G
for durations in excess of 0.006 sec (the tolerable level for
humans as interpreted from the Eiband data) in crashes that do
not exhaust the stroke of the seat.
173
~i
Crewseats should be designed to stroke a minimum distance of
12 in. when the seat is in the lowest position of the adjust-
ment range. This distance is needed to absorb the resi-dual
energy associated with the vertical design pulse. Further, the
load-limiting system should be designed to stroke through the
full distance available including the vertical adjustment dis-
tance. Since a vertical adjustment of ±2-1/2 in. from neutral
is typically required by crewseat specifications, proper design
can provide up to 17 in. of stroke, depending on seat adjust-
ment position.
The minimum of 12 in. of stroke is recommended to provide the
minimum required level of protection. As illustrated later in
this section, even with 12 in. of stroke, heavier occupants in
more severe crashes will exhaust the available stroke distance
and bottom out. The following reasons point out the need for
obtaining the greatest possible energy-absorbing stroke from
the seat:
* It is easier to provide energy-absorbing stroke in
the seat than in the fuselage or landing gear. The
distance from the floor of the helicopter to the
ground is usually specified either directly or im-
plicitly by overall dimensional requirements.
* The energy-absorption capacity of the seat is much
easier to demonstrate than that of the airframe, as
the energy-absorption capacity of the airframe is
difficult to predict and hardware is usually not
available for testing in the early design phases of
a new aircraft.
Energy absorption assigned to landing gear can be
lost depending on the type of terrain upon which
the aircraft crashes; i.e., soft versus hard, as in
marshes or water as opposed to a landing strip. Air-
craft attitude at impact has a significant influence;
also a relatively high roll angle, for instance,
could render the landing gear energy-absorbing fea-
ture inoperative. *
Based on the two previous points, the seat is a low-

risk approach for providing energy-absorbing stroke.
* 12 in. of stroke has been shown to be practical in
existing aircraft.
If it is absolutely impossible to obtain a minimum of 12 in.
of stroke, a lesser amount is acceptable, but in no case should
it be less than 7 in. The reduced stroke is acceptable for a
174
retrofit application or for use in small aircraft in which it
is simply impossible to find the space for a 12-in. stroke.
In such cases a systems analysis is mandatory; the analysis
must show that occupant protection is equivalent to the system
in which the 12-in. stroke is available.
For retrofit applications, the maximum protection possible
should be obtained in any component being modified, i.e.,
* seats, gear, etc. Separate test criteria have been established
for seats not having the required 12 in. of stroke and are pre-
sented in Section 8.6.3.2 of this document.
Energy-absorbing systems should be designed for 11.5 plus 1 G
minus 0 G considering the effect of the dynamic loading rate.
To obtain the static test loads, dynamic limit loads should be
reduced by the amount due to rate sensitivity of the particular
device used. Further, in the design of the system the desired
total resistive load on the seat should be obtained by summing
the resistive load provided by the energy-absorbing system and
the resistive load resulting from friction and/or other mechan-
isms unique to the particular system. Thus, the resistive load
of the energy-absorbing subsystem must be reduced from the load
required to decelerate the seat by the amount of the other
stroke-resisting variables.
If the energy-absorbing system is to provide only one force
setting, the effective weight of the 50th-percentile occupant
from Tables 9 and 10 should be used for sizing it in order to
ensure a tolerable stroke for the majority of the occupants,
not exceeding the stroke limitations of the seat. These weights
are 142.3 and 160.7 lb for pilot/copilot and troop and gunner
seats, respectively.
The following is an example of the calculations made for a
seat designed to stroke under the decelerative load imposed by
a 50th-percentile crewmember. The average deceleration and
stroke of the 5th- and 95th-percentile seat occupants are ap- -
proximated. First, using weights from Table 9, the 50th-
percentile effective weight is calculated according to
Wteff = 0.80 (Wt5 0 + Wtc) + Wth (35)
where Wteff x effective weight of 50th-percentile occu-

pant, lb
Wt50 - nude weight of 5th-percentile occupant, lb
175
- 4-. ... I I I __I . . .I . .. . .I~ l f

Tc
Wt a weight of clothes, lb
Wth a weight of helmet, lb
Thus, Wteff - 0.80 (170.5 + 3.1) + 3.4
142.3 lb
which is shown in Table 9 as the effective weight of the 50th-

percentile crewmember. The effective weights for the 95th- and
5th-percentile aviators are 175.2 and 112.6, respectively.
Assuming a 60-lb movable seat weight, the total weights that
the load-limiting system must be designed for are:
5th percentile: 172.6 lb
The 50th-percentile static limit load (LL) is calculated as
follows:
LL = GL Wteff = (11.5) (202.3) z 2326 lb
The static load factors for the 95th- and 5th-percentile avia-
tors are then
GL95th 2326
25.2 =9.9
2326 -
GL5th .6 13.5
Seat deceleration spikes approaching 23 G could be expected for

the seat occupied by a 5th-percentile aviator. Decelerations
of this magnitude would not be expected to cause severe injury.
Also, in more severe crashes, the stroke would exceed 12 in.
for a seat occupied by the heavier percentiles. This would
mean that the 95th-percentile vertical survivable crash could
not be fully protected against in the seat adjusted-down posi-
tion. With the seat in the neutral or up position, however,
protection over the entire range might be provided (see Refer-
ence 24).
176
For comparison, the same type of calculations for a system
limit load sized for the 95th-percentile crewmember yields the
following:
LL = (11.5) (235.2) = 2705 lb
and
G 2705
GL50th = 270 13.4
2705
GL5th = = 15.7
The entire population in the 5th- to the 95th-percentile weight

range could be expected to receive deceleration spikes in ex-
cess of 23 G in seats in which the limit load was designed for
the 95th-percentile occupant. Also, the natural distribution
of occupant weights places the majority of aviator weights near
the 50th percentile. It is therefore expected that more over-
all protection can be provided by sizing limit loads for the
50th-percentile rather than for a heavier occupant.
In order to use the stroke distance available at maximum effi-
ciency, regardless of occupant weight, a variable-force load-
limiting mechanism is desirable. With an infinitely variable
force system, the deceleration levels can be maintained within
acceptable limits (if the stroke is not exhausted) for the full
range of occupant weights for either crew or troop seats while
using equal stroke lengths for identical pulses. A compromise
is possible for a seat design that uses a load-limiting device
rather than collapsing structure. The device can be designed
to produce two or more limit loads that can be selected by the
seat occupant. An example of one such system is discussed in
Reference 24. The selection would be made on the basis of
aviator weight. For example, for a dual-limit-l-oad device,
"the lowest load device might be-established by using the weight
of a 5th-percentile occupant. The second force might be de-
signed for the weight of a 50th-percentile aviator. In opera-
tion then, the aviator would be required to select a limit load
by movement of a lever or dial upon entering the seat.
It is recommended that at least a dual-level load limiter

(preferably three or more levels) be used to provide maximum
protection over the complete occupant weight range. As an
illustration, consider the limit-load factors calculated
on this page and on 181. With the limit load set for the
9 177
NI3
50th-percentile occupant weight, the calculated load factors
were 9.9 for the 95th- and 13.5 for the 5th-percentile occupant
I
weights. This produces a negative variation of 1.6 G for the
heavy occupant and a positive variation :f 2 G for the lighter
occupant from the design factor of 11.5 G. If two load set-
tings were possible, the variations could be halved producing
0.8-G and 1.0-G variations, respectively. An infinitely ad-
justable mechanism would reduce the variation to zero.
The interaction between the occupant and the movable seat
Therefore, the movable seat
masses increases with seat mass.
mass should be minimized. For integrally armored seats, this
mass is relatively high, varying between 60 and 120 lb. How-
i
ever, design in compliance with the criteria presented herein
should provide the desired degree of protection. "he criteria
were derived primarily from studies of an inter Iiy armored
seat prototype and projected to unarmored seats. The criteria
are expected to be somewhat conservative for seats of very
light (10 to 15 lb) movable sections.
Because troops do not have operational functions to perform and
troop seats are not armored, more flexibility exists in troop
seat design. Troop seats should be designed for the maximum
stroke feasible to maximize protection over the large weight
range represented by the fully equipped and lightly equipped
occupant. It is recommended that the full 17-in. seat pan
height normally considered desirable from the human engineering
standpoint be used for energy-absorbing stroke. It is further
recommended, as a minimum, that the limit load of the system be
sized using the 11.5-G load factor and the effective weight
of the 50th-percentile heavily equipped occupant (160.7 lb).
Variable-level load limiters sized as discussed previously are
also desirable for troop seats.
8.3.5 Upward Loads
A capabilitl to withstand a minimum upward load of 8 G is rec-
ommended for all aircraft seats. Occupant weight should be
that of the 95th-percentile crewmember or trooper as presented
in Section 8.2.
8.3.6 'ie.ral Loads and Deformation
The lateral load and deformation requirements for forward- and
aft-facing sents are presented in Figure 60. Two curves are
presented. One is for light fixed-wing aircraft and attack and
cargo helicopters, while the other is for other rotary-wing
aJ-craft. The deflections of the seat are to be measured by
recording the motions of the seat reference point. Occupant
weight should be as stated in Section 8.2 and should be that
of the 95th-percentile aircrew member or trooper.
178
25
Lateral
Acceptable failure area
M b•C (Example for forward-

"0 is Rejection .4 0 or aft-facing seat)
rea .Utility aed
area I j"e ,%
C observat ion
SIhelicopters
13. T t e (y)ttack, ando
Trotal controlled deformation (y), in.

measured at seat reference point
Figure 60. Lateral seat load and deformation requirements

for all types of Army aircraft.
Lateral loading in the forward direction (aircraft reference

system) on side-facing seats should be the same as for forward-
loading (Figure 59) except load limiting should be employed.
For crewseats, the lateral deflection should be minimized; how-
ever, it is doubtful if any great stiffness can be achieved in
lightweight hardware. As a matter of interest, many new ar-
mored buckets are made from Kevlar, a very tough and strong ma-
terial in tension. Its resin-starved condition (required for
good ballistic protection properties) leaves it with a rather
low flexural modulus, particularly after the seat has had other
loa=s imposed. The material is also rather rate sensitive
tE 'f under high loading rates, soft under low rates). For
th. . eason, it is believed adequate, as a design goal, to at-
tempt to limit the initial deflection to 1 in. with a 2-in. re-
quirement. Because of the possible loading rate sensitivity
of the seat materials, it is considered acceptable to demon-
strate compliance by analysis of test data. This analysis
might include adjustment of the static test data by use of mea-
sured or known deflection and load data from dynamic tests.
179
-I
Further, in cases where wells are provided under the seats to
increase the available stroke distance, the deformation should
be elastic. This may allow the seat to realign itself with the
well prior to entry after the lateral and longitudinal loads
are relieved, as explained in Chapter 4.
8.4 OTHER SEATS
The requirements presented for crewseats and troop and gun-
ner seats also apply to passenger seats and any other seat
installed in the aircraft for any purpose. Unique seats in-
stalled for special uses are not to be exempt.
8.5 PERSONNEL RESTRAINT HARNESS TESTING 1•
The restraint harnesses are to be statically and dynamically
tested along with the seat and/or structure to which they are
attached as noted in Chapter 7. However, the lap belt, shoul-
der straps, and tiedown straps, including all hardware in the
load path, should be statically tested separately to ensure
that all components possess adequate strength and to determine
elongation. The strength and elongation test requirements bf
restraint system subassemblies are specified in Table 4.
Specific component tests, including operational tests, are de-
tailed in a draft military specification (Reference 85). How-
ever, all components and subassemblies should be statically
load tested. Each subassembly should be tested to its full de-
sign load to demonstrate its adequacy. Elongation character-
istics should be measured to document these data for comparison
with requirements and use in systems analyses.
8.6 STRUCTURAL TEST REQUIREMENTS
Both stati. and dynamic tests are recommended. Dynamic tests
of aircraft seats have shown that individual components capable
of maintaining the design loads often fail when tested in com-
bination with other components. This could be a result of in-
accurate analyses. However, it is recommended that all seat
and litter systems be tested as complete units. This is not
to imply that component tests are not useful. Component tests
can be extremely useful and should be used wherever possible
to verify required strengths. This practice is particularly
valid where finite-element analyses have boen used to accu-
rately predict distribution of loads in redundant structures.
Upon acceptance of prototype systems tested under both static
and dynamic conditions, no further tests shotuld be required
except for quality assurance. Major structural design changes
in the basic seat system will require static retesting of the
180
new system to ensure that no loss in strength has been caused
. by the design changes. If the changes could affect the energy-
absorbing, or stroking, performance of the seat, additional dy-
namic tests should also be conducted. Major structural design
changes are those changes involving principal load-carrying
members such as floor, bulkhead, or ceiling tiedown fittings,
structural links or assemblies, seat legs, or energy-absorbing
systems. Minor changes, such as in ancillary fittings, can be
accepted without a structural test. A significant weight in-
crease, however, such as the addition of personnel or seat ar-
mor, would require additional testing. In summary, changes
that increase loading, decrease strength, produce significant
changes in load distribution, or affect the stroking mechanism
will require retesting.
All testing is to be conducted with the seat cushions in place
and, for seats with adjustments, the seats should be in the
full-up and full-aft positions unless another position is shown
to be more critical.
If desired, dynamic tests may be substituted for static tests;
however, loading in all principal directions are required. Al-
ternate dynamic tests are presented in Section 8.6.1.9.
8.6.1 Static Test Requirements
8.6.1.1 General: The purpose of the static tests is to demon-

strate that the seat has the strengths and other properties
required to provide the desired performance in all the princi-
pal loading directions. Static testing enables basic proper-
ties to be ascertained for known loads applied at a slow enough
rate so that seat response can be observed. Successful comple-
tion of the static tests does not guarantee passing the dynamic
tests, but it improves the chances. Weaknesses can be identi-
fied and corrected prior to conduct of the ultimate dynamic
tests. Also, due to the loading rate sensitivity of materials,
load distributions may be different in dynamic tests than they
are in static tests. Certain structures, statically soft, ma:
react as stiffer members under dynamic loading, and thus, pick
up more of the load than when the system was loaded statically.
Because of these reasons and because of dynamic avershoot, a
margin of safety has been added to the ultimate static load
factor on the design curves as compared to the peak accelera-
tions of the dynamic design pulses. It is recommended that
this margin not be sacrificed for reduced weight.
Table 11 presents the static test requirements for complete

seat units. The tests required include a series of unidirec-
tional tests to determine basic seat strengths along the major
181
~...................................- .
TABLE 11. SEAT DESIGN AND STATIC TEST REQUIREMENTS
Percentile
Test Loading direction occupant Load/deformation
L ref. with respect to Load used in load dei
no. fuselage floor required determination reguirementsai
I Upward 8-G minimum 95 No requirement
.+.0
-0 G G See
2 Downwardbn 8.3.4Section
bd 11.5 so
Aftward 12-G minimum 95 No requirement
4 Forward See Figure 59 95 See Figure 59
5 Combined
Forward e,f See Figure 59 95 See Figure 59
Downwardc 11.5 +2.0 G 50 Same as Test 2h
-1.0 G
Lateral f 9-G minimum 95 No requirements
6 Lateralg See Figure 60 95 See iigure 60
(
(a) ate aircraft floor or bulkhead should be deformed as detailed in
in Figures 61 and 62, simultaneously with, or prior to the con-
duct of all static tests and kept deformed throughout load appli-
cation.
(b) If more than one load-limiter setting is provided, a represen-
tative sample of settings spanning the range of loads should be
tested.
(c) If more than one load-limiter setting is provided, the highest
load should be used.
(d) Subsequent to the stroking of the vertical %,ergy-absorbing de-
vice, cockpit seats should carry a static load of 25 G, based on
the effective weight of the 95th-percentile clothed and equipped
occupant per Section 8.2 plus seat without loss of attachment
to the basic structure except when the seat pan has stroked to
and is supported by the floor.
(e) In the event that no load-limiting device is used in the forward
direction, a 20-G load for cabin seats and a 25-G load for cock-
pit seats may be used for this combined loading.
(f) For seats employing vertical guides which could distort under
combined loading and cause binding, the maximum forward and la-
teral loads should be reached prior to initiation of stroking.
This sequence demonstrates whether the seat will stroke downward
after transverse loads-are applied.
(g) The lateral loads should be applied in the most critical direc-
tion. In the case of symmetrical seats, the loading direction
is optional.
(h) Failure to meet the ll.5-G +2.0/-I.0-G static vertical load
limit should not be cause for seat rejection if the seat vertical
energy-absorbing system meets dynamic load requirements.
Wi) Plastic deformation is permissible: however, structural integrity
must be maintained.
182
axes. A combined loading test is also required to evaluate
the seat performance under static conditions simulating the
most severe, unsymmetrical loading condition anticipated. All
static tests should be conducted under simultaneous conditions
of floor buckling and warping as illustrated in Figure 61 or
bulkhead warping as illustrated in Figure 62. The warping con-
ditions must be introduced in the static test phase to evaluate
completely the performance of the seat under the most severe
requirements selected for design.
8.6.1.2 Unidirectional Tests: Where separate strength and
deformation requirements have been specified in Table 11 for
longitudinal, vertical, and lateral loading of seats, the loads
should be applied separately. Seats must demonstrate no loss
in structural integrity during these tests and should demon-
strate acceptable energy-absorbing capacity.
8.6.1.3 Combined Loads: Seats must demonstrate no loss of
structural integrity under conditions of combined loading as
shown in Table 11 and should demonstrate ability to stroke in
the vertical direction with the transverse loads applied.
8.6.1.4 Load Application Method: The static test loads are to

be applied at the expected center-of-gravity location of the
occupant or occupants of each seat. The loads should be ap-
plied through a body block (see Section 8.6.1.5) restrained in
the seat with the restraint system. Figure 63 shows the loca-
tion of the center of gravity that should be used as the ini-
tial static load application point for the seat occupant.
For the testing, the seat should be adjusted to its aftmost

and full-up position. The loads calculated by multiplying the
weight of the occupant and equipment plus the weight of the
seat by the required load factor should be applied continuously,
or in not more than 2-G increments while the load-deformation
performance of the seat is recorded. Maximum loads need not be
held for more than 1 sec. The maximum load reached, regardless
of duration, is to be used to assess compliance.
On integrally armored =rewseats, care should be taken to assure

that the loads are applied proportionally to the proper assem-
bly or test item to simulate the loads that would typically be
carried by the restraint harness and the seat support structure.
"* In other words, the portion of the load that could be expected
to be restrained by the restraint harness should be applied to
the body block as described above. The portion of the load
representing inertial loading of the movable assembly should
be applied separately at the center of gravity of the appro-
priate substructure through another provision. For example, a
183
. . .. .. ...
-_2. ... ... - . . .
F-,
Original *
position
VL20 position
0,
Figure 61. Suggested method of applying floor warping

for static testing of seats.
184
-- i
y :Hydraulic cylinder
~Test fixture structure
Test
structure
----De flected
' position
Seat .,inal
attachments ::•io
Bulkhead
Figure 62. Suggested method of applying bulkhead

warping for static testing of seats.
lever to proportion the load between the body block and movable
section of the seat, and a sling to apply the appropriate por-
tion of the load to the bucket, can be used. For seats with
relatively heavy frames, the inertial load of the frame can be
applied separately at its appropriate center of gravity. This
technique, although adding complexity to the test setup, as-
sures that all components in the seat and restraint system as-
sembly have been tested to their approximate static design
loads and that, as far as a static test simulation can be ex-
tended, performance and structural adequacy have been demon-
strated. For lightweight seats (less than approximately 45 lb
for total seat and restraint system), the total load can be im-
posed on the body block.
8.6.1.5 Static Load Body Block: The static test loads must
be applied through a body block contoured to approximate a
95th-percentile occupant seated in a normal flying attitude.
The body block must contain shoulders, neck, and upper legs,
and provide for passage of a belt tiedown strap between the
legs. The upper legs should be contoured to simulate the flat-
tened and spread configuration of seated thighs and to allow
the proper location of the buckle. Critical pelvis dimensions
9 185
}J
..... - A*4,. .. *
~.- . .n
- -mm----_---_--._--
It I
1..i
15.0 in.
Ur
Forward > Aftward 00 in.
5vIe0.0 t
0.5 in.
3.0-in, radius
o m Seat reference
".r 3: point
ajo
Figure 63. Static load application point and critical

body block pelvis geometry.
are shown in Figure 63. Buttock contours must be provided

to permit proper fit in a contoured seat pan. The leg stubs
should be configured to permit proper seat pan loading as the
body block rotates forward under longitudinal loading; i.e.,
the leg stubs should be only long enough to provide a surface
to react the lap belt load. The side view of the buttocks
should include an up-curved surface forward of the ischial tu-
berosities to allow the forward rotation of the body block
while maintaining the primary contact between the ischial tu-
berosities and the seat pan through the cushions.
8.6.1.6 Deflection Measurements: Deflection should be mea-
sured as lose to the seat reference point as possible to elim-
inate seat structure rotational deformation from influencing
the test results. To simplify thesc measurements, the seat
reference point can be projected to the outside of the seat pan
or bucket.
186
Normally the restraint system will be attached to the seat.
However, if a unique situation should develop in which the only

option for increasing crashwQrthiness is to attach the system
(lap belts and shoulder harness) to the basic aircraft struc-
ture rather than to the seat, certain factors should be con-
sidered. First, the forward and lateral deflection require-
ments of Figures 59 and 60 need not be considered because the
restraint harness limits torso and seat deflection. Second,
the vertical deflection of the seat pan still must be consid-
ered since the downward movement of the seat pan could cause
excessive slack in the restraint harness, or the harness could
limit the stroke ox the seat, depending on where the restraint
system is anchored. Neither of these conditions is acceptable
in the design.
8.6.1.7 Load Determination: The total load required for all

test directions, except vertical downward, is determined by
multiplying the required load factor from Table 11 by the
weight of the 95th-percentile clothed and equipped occupant
from Table 9 or 10 (Section 8.2) plus the weight of each
seat. The effective weight of the 50th-percentile occupant
should be used to calculate vertical components of loading
(Test Nos. 2 and 5 of Table 11) as discussed in Section 8.3.4;
the effective weight of the 95th-percentile clothed and equip-
ped occupant should be used for the bottomed test (Test No.
2(d) in Table 11). The weight of that portion of the seat that
strokes with the load-limited portion of the seat must be added
to occupant weight to determine the total required load in the
vertical direction.
8.6.1.8 Multiple Seats: Multiple-occupancy seats should be

fully occupied when tested. If it is determined that the most
adverse loading condition occurs in other than full-occupancy
situations, additional tests should be run for those conditions.
8.6.1.9 Substitution of Dynamic For Static Tests: It is rec-

ommended that static tests be conducted because of the advan-
tages previously described. In summary, static tests are more
economical to run than dynamic tests; because of their slow
rate of load application, closer real-time observation of seat
response to the loading is possibleaiand, static testing pro-
vides structural response information which is more comparable
to the static analyses typically used in the development of
present seat designs. In the future, when dynamic analysis
becomes more reliable, this latter point will no longer apply.
A significant consideration in static-versus-dynamic testing
is the cost of the hardware. Static testing can be conducted
with a minimum number of seats because the condition of the
seat can be monitored and judgments made as to its acceptabil-
ity for continued testing. If failures due to previous tests
occur, parts can be replaced and the test economically rerun.
187
If final acceptance decisions are based on dynamic tests alone,
a considerably more rigorous dynamic test matrix is required to
enable testing in all the principal loading directions. Dy-
namic tests are usually more expensive than static tests and
the increased number of tests will also require additional
hardware. If new hardware is not used for each test, the re-
sults may be inconclusive. If the seat passes the test, the
results are acceptable; but if the seat fails the test, another
test must be made since it will not be apparent whether the
failure was due to damage inflicted during a previous test or
due to a basic design or manufacturing flaw.
If for any reason, dynamic tests are substituted for the static
tests previously described, then loading in all principal di-
rections must be conducted. The dynamic test requirements are
presented in Figure 64. These three tests must be conducted
in addition to the two presented in Section 8.6.2, and all five
must be passed. These tests are to be conducted in accordance
with the same ground rules as those presented in Section 8.6.2
and are subject to the same testing parameters and evaluation
procedures. Further, the static upload of 8 G and the static
aftward loading of 12 G must be imposed and satisfactorily
passed.
8.6.2 Dynamic Test Requirements
8.6.2.1 Dynamic Test Requirements for Seats Having at Least
12-in, of Vertical Stroke: All U. S. Army prototype seats
should be dynamically tested to the two conditions specified
in Figure 65. These test conditions were determined from the
design velocity changes presented in Volume II. Test 1 is re-
quired to ensure that the vertical load-limiting provisions
will perform satisfactorily under simultaneous forward and la-
teral loading conditions. Test 2 is required to ensure that
the seat can resist the loads produced by the design pulse
when applied simultaneously in the forward and lateral direc-
tions. A 50th-percentile anthropomorphic dummy complying with
the Code of Federal Regulations, Title 49, Part 572 specifiea-
tion for dummies (Reference 90) should be used to simulate the
seat-system occupant for Test 1. A 95th-percenti]e anthropo-
morphic dummy simulating as closely as possible the features of
the 50th-percentile dummy described above should be used to
simulate the seat-system occupant for Test 2. Total weight in-
cluding instruments of these two test dummies should be:
90. U. S. Code of Federal Regulations, Title 49, Chapter 5,

Part 572: ANTHROPOMORPHIC TEST DUMMY, Government Printing
Office, Washington, D. C., (Rev.) 1978.
188
VAX.-
4*9 m-in
~''
_____ - ~ ... to
to4
load I
t sa 2I IS .5 04 .7
L~~ I
10 -
to in ft/s t 0303
2b croad&tc ac~e 0ockpi o.013 C.057 .03t3
Dummy
i ia
nert G min 0.03 1.03 105 10
load
r.Rx 19 19 19 19
AVin, ft/sec 42 42 425 25
j nylcotr
nbuvai
tn ia
lod
t see
2 sec
Gflmax
0.06Ž
0.100
21
0.036
0.100
33
.061
.104
21
.036
.124
27
AV min,ft/sec 30 s0 30 30
Ligurt 64fiqureens-fwinmitsti
cfrr susttu statd tests
heicptra~ 1 sc0.5, 0.33189 03

-I m0| m I I I I
aa
0
I I;
- IL2
tO t•l ti
Time, sec
Cockpit lest* Cabin seats
Test Confiquration Parameter QuaLification R&D Qualification R&D
Dummy inertial
load
t sac 0.043 0.024 0.059 0.034
t2 ooc 0.061 0.061 0.087 o.087
Av min, ft/sac SO s0 50 So
10'
2 t I sec 0.066 0.038 0.081 0.046
St2 sac 0.100 0.100 0.127 0.127
G min 28 28 22 22
inertial G max 33 33 27 27
load
Lv min, ft/sec 50 s0o 0 50
Figure 65. Dynamic test requirements for qualification

and for research/development testing.
91i)
50th percentile
Pilot/Copilot = 181.1 lb
Troop/Gunner = 196.6 lb
95th percentile
Pilot/Copilot = 222.3 lb
Troop/Gunner = 242.2 lb
Dynamic testing of multiple occupant seats should be performed
with the maximum number of occupants specified for the test
seat. Additional tests should be run if it is determined that
the most adverse loading condition occurs in other than full-
occupancy situations. For both tests of Figure 65, adjustable
seats should be adjusted to the full-aft and up position of
the adjustment range. Plastic deformation of the seat is per-
missible; however, structural integrity must be maintained in
all tests. For Test 1, the seat should limit the acceleration
as measured in the pelvis of the dummy to values which ensure
that the 50th-percentile clothed seat-system occupant (see Sec-
tion 8.2) will not experience vertical, +G , accelerations in
excess of human tolerance as defined in SJctions 4.3 and 4.8
of Volume II (see Figure 24 of this volume). The roll direc-
tion (10 degrees right or left) for Test I should be the more
critical loading for the specific seat design.
When determining compliance of the achieved test pulse with the

dynamic Xest requirements of Figure 65 (or Figure 64, as de-
scribed fn Section 8.6.1.9):
1. Determine the maximum acceleration and construct the

onset slope for the test pulse by the method ex-
plained in Section 8.6.3.
2. Compare the achieved onset and peak acceleration of

the test pulse with those allowed and presented in
Figure 65. The achieved onset slope should lie be-
tween the minimum and maximum onset slopes using the
values of t1 and t listed in Figure 65 for the spe-
cific test conditions. The maximum acceleration
should also fall between the upper and lower limits
allowed.
3. Integrate the actual acceleration/time curve of
the test pulse and establish the achieved velocity
change. The velocity charie achieved should be equal
to or greater than that tabulated for the specific
test conditions.
191
A-_ -f[&
8.6.2.2 Special Dynamic Test Requirements for Seats Having
Less Than 12 in. of Vertical Stroke: In the event that the
application of a systems approach permits the seat to have
less than 12-in. minimum vertical stroke, additional require-
ments are made of the dynamic testing. First, it would be de-
sirable to perform a full-scale crash test with the test speci-
men, including all assemblies involved in the energy-absorbing
process. This would include a section of the fuselage, land-
ing gear, and the seat or seats. This approach is totally ac-
ceptable for demonstrating the dynamic response and acceptabil-
ity of the system.
Since cost associated with the type of system testing described

above is usually prohibitive, a different approach is accept-
able. This approach includes dynamically testing the seat only,
as is done for systems with at least 12 in. of stroke, but mod-
ifying the input pulse to represent the energy-absorbing proc-
esses of the gear and fuselage. An example of such a modified
test pulse is presented in Figure 66. The initial plateau (t 1
to t') represents the acceleration-time history created by
stroJng of the landing gear. The sharp increase in accelera-
tion at t' relates to fuselage impact, and the pulse beyond t'
representR the crushing of the stiffer fuselage section. The
velocity change under the pulse should be the same as identi-
fied for the particular crash force direction for other estab-
lished tests (50 ft/sec for Test No. I or No. 2 of Figure 65).
G2
0
4.J
.. Fuselaq
Landing gear stroking crushing
t t0 ti
Time, sec
Figure 66. Example of input pulse for seats

having less than 12 in. of stroke.
2.92
[
. ___- _. - - . . . .
It will be difficult to determine accurate dynamic crush char-
acteristics of the various portions of the system to enable
establishment of a representative, and thus acceptable, test
pulse. The best analytical techniques, supported by test data,
should be used for determining the properties of the fuselage.
Since drop tests of landing gear are required, a much more ac-
curate approach exists for obtaining the landing gear influence
on the pulse. Seat testing should await completion of landing
gear tests so that the results can be used to establish the
initial plateau (or other shape) between t1 and t6 of the input
pulse.
Typically the landing gear will stroke at loads below those

required to stroke the seat; therefore, much of the kinetic en-
ergy of the occupant and seat will be absorbed prior to fuse-
lage impact. If the systems analysis is accurate, the energy-
absorbing capacity of the seat will be sufficient to absorb
the residual energy at limit loads tolerable to the occupant.
Since each system may display different characteristics, it is

not appropriate to present in this document specific quantita-
tive limits for use in evaluating the acceptability of the test
pulse. However, the same general approach and tolerances al-
ready presented for the standard pulse apply and should be
used. The technique described in Section 8.6.2.1 for estab-
lishing compliance with the required test pulse applies di-
rectly to the portion of the special test pulse following t'.
8.6.3 Data Acquisition and Reduction
Data acquisition and reduction should comply with the require-

ments of SAE J211 (Reference 91) for measurements of an anthro-
pomorphic dummy, body accelerations, and structures.
Data should be presented in both analog and tabular form in
compliance with the sign convention shown in Figure 4. Impact
velocity should be determined and recorded for the test plat-
form or vehicle. In the analysis of the data, velocity change
should be computed through either electronic means or graphi-
cally with a planimeter by integrating the area under the mea-
sured acceleration-time trace.
91. SAE Recommended Practice, SAE J211b, INSTRUMENTATION FOR

IMPACT TESTS, SAE Handbook 1979, Part 2, Society of Auto-
motive Engineers, Inc., Warrendale, Pennsylvania, 1979,
pp. 34.117-34.118.
193
The method recommended for use in
ity of the pulse (see Section 8.6.2)establishing the acceptabil-
parameters associated with the data and to determine other
is similar to that pre-
sented in MIL-S-9479(USAF); see Reference
as rise time, onset slope, and acceleration 92. Parameters such
may be obtained using the plateau duration
following graphic approximation
nique as shown in Figure 67. tech-
Gp
0.9 Gp 3
bseline
Calibration
j0.1 Gp
to tr
Figure 67. Graphic approximation example.

(From Reference 92)
9 Locate the calibration baseline.

* Determine the maximum (G ) acceleration
magnitude.
* Construct a reference line parallel
tion baseline at a magnitude equal to the calibra-
the peak acceleration (G ). The to 10 percent of
tersections of this line-with the first and last in-
acceeration-time
plot defines points I and 2.
92. Military Specification, MIL-S-9479,

SEAT SYSTEM, UPWARD
EJECTION, AIRCRAFT, GENERAL SPECIFICATION
FOR, Department
of Defense, Washington, D. C.
194
p- ~ - - -
' 1
• Construct a second reference line parallel to the *

calibration baseline at a magnitude equal to 90 per-
cent of the peak acceleration. The first and last
intersections of this line with the acceleration-time
plot define points 3 and 4.
* Some logic and practical judgment may be required for
selection of the first and last intersections depend-
ing on the noise, structural or electronic, apparent
in the data. Significant tendencies are important,
not noise.
* Construct the onset line defined by a straight line
through points I and 3.
9 If desired, construct the offset line defined by a
straight line through points 2 and 4.
* If desired, construct a line parallel to the calibra-
tion baseline, through the peak acceleration. The
time interval defined by the intersections of this
line with the constructed onset and offset lines
L (points 5 and 6) is the plateau duration (Ot).
1, * Locate the intersection of the constructed onset line

with the calibration baseline (point 7). The time
interval defined by points 7 and 5 is the rise time
(tR). Referring to Figure 65, the rise time should
beRgreater than t but less than t when determining
compliance with d4namic test requA~ements. Point 7
is the initial time to in Figure 65.
8.6.4 Seat Component Attachment
Since components that break free during a crash can become le-
thal missiles, it is recommended that attachment strengths be
consistent with those specified for ancillary equipment (see
Section 6.6.5.9, Volume III). Therefore, static-attachment
strengths for components, e.g., armored panels, should be as
follows:
Downward: 50 G
Upward: 10 G
Forward: 35 G
Aftward: 15 G
Lateral: 25 G
These criteria may be somewhat conservative for load-limited
seats; however, load limiting is mandatory in the vertical di-
rection only. In light of the potential hazard, the strength
requirements are considered justified.
195
fr~puwr"i
9. LITTER STRENGTH AND DEFORMATION REQUIREMENTS
9.1 INTRODUCTION
This chapter presents strength and deformation requirements
for litter systems. Aircraft systems are rather difficult to
design because of limitations including that of the strength
of existing litters and width of utility aircraft as compared
to the length of standard litters. The ultimate vertical
strength of existing litters with a 200-lb occupant and a total
system weight of 250 lb (see Section 9.2) is about 13 G. Since
the desired decelerative loads to be imposed on these litters
exceed 13 G, special techniques must be used to limit the de-
flection and to support some of the occupant load. A new
litter should be developed having the required strength to sup-
port loads in excess of 13 G, preferably 17 G, as presented as
a minimum in this chapter.
The other problem is associated with the length of the litter.
The width of the new Army utility helicopter does not allow
litters to be placed in the preferred lateral direction. The
lateral orientation is preferred because of the characteristics
of existing restraint systems used on litters which provide
more support when loaded laterally than when loaded longitu-
dinally. Since higher loads are more frequently seen in the
forward direction than in the lateral, it would be desirable
to orient the litters laterally in the aircraft. This is not
possible because the helicopter is not wide enough, so special
devices have been developed to permit loading the litters in a
lateral direction and then rotating the litters into a fore-
and-aft orientation inside the aircraft. Improved litter re-
straint systems are required to provide the desired support to
the supine occupant on litters oriented in the fore-and-aft di-
rection in these aircraft.
This chapter presents the design strength-deformation relation-
ships and testing requirements for aircraft litters and their
supports.
9.2 RECOMMENDED OCCUPANT WEIGHTS FOR LITTER DESIGN
The litter strength and deformation requirements defined below
are based on a 200-lb, 95th-percentile litter occupant with
20 lb of clothing and personal gear, a 10-lb splint or cast,
and 20 lb of litter and support bracket weight for a total
weight of 250 lb (the weight of a litter and patient as speci-
fied in MIL-A-8865 (ASG), Reference 93).
93. Military Specification, MIL-A-8865, AIRPLANE STRENGTH AND
RIGIDITY MISCELLANEOUS LOADS, Department of Defense, Wash-
ington, D. C.
196
9.3 VERTICAL LOADS
9.3.1 Downward Loads
In the case of litter systems, human to&erance is not the lim-

iting case in the vertical direction. The loads would be ap-
plied in a transverse direction to the body of a litter occu-
pant. However, design to the 45-G human tolerance level is
impractical due to the strength requirements for litters and
for the basic structure to support the litter systems.
Litters are either hung from the ceiling or supported at the
floor. In either case, the input deceleration pulses are the
same as for floor- or bulkhead-mounted seats (see Volume II).
The use of ceiling-supported litters is limited by the strength
of the overhead fuselage structure. The inefficiency of struc-
tural deformation of the ceiling of older aircraft requires
additional energy-absorbing stroke to provide the protection
desired. Litters should not be suspended from the overhead
structure unless it is capable of sustaining, with minimum
deformation, the downward loads from the tiers of litters.
Therefore, in the design co an efficient system, intentional
load limiting should be related to the floor pulse.
The vertical strength and deformation requirements for a litter

system are detailed in Figure 68. This curve is read in the
identical manner as the seat load-deflection curve shown in
Figure 59. The load factors in units of G are based on the
summation of the weights of the occupant plus clothing, per-
sonal gear, splint or cast, and the weight of the litter and
attachment brackets for a total of 250 lb as described in Sec-
tion 9.2. The curve of Figure 68 is based on the assumption
that 3 or 4 in. of vertical deflection will occur at the mid-
point of the litter. In the unlikely event that a rigid litter
is used, an additional 2 in. of deflection should be added to
the curve. The deflection curve is limited to 6 in., because
a larger deflection occurring on one corner of the litter due
to an asymmetric loading could cause ejection of the litter oc-
cupant. A larger- energy-absorbing stroke can bd used effec-
tively if a mechanism-is included in the system to control the °
amount of tilt allowed. For example, a system mechanism could
be designed that forced all four corners of the litter to
stroke the same distance (within elastic limits) thus achieving
this goal.
The additional problem associated with inadequate litter
strength must be dealt with in the design of litter systems.
The curve of Figure 68 assumes a litter capaile of at least
17 G with a maximum of 25 G. If the existing litter is used,
then a pan, net, or other device should be included under the
197
S..~. . . .
-v'" ~ ,- -
Aircraft ceiling
Downward fl j
Aircraft floor
ACCEPTABLE
m f : ::!: BASE AREA ;"l:::xi
10
0 1 2 3 4 5 6
Total controlled
deformation (z), in.
*G value based on 250-lb per litter
position.
Figure 68. Litter downward load and

deflection requirements.
litter to catch and support the litter occupant if the litter

fails. Actually the device should limit the deflection to a
value less than required to fail the litter and should stroke
with the litter. If all of these provisions are included, i.e.,
a rigid new litter or old litter with supporting pan under-
neath, together with the tilt-limiting mechanism, then the
stroke can be extended to 12 in. at a 17-G limit-load factor.
The load-deformation curve of Figure 68 would be extended at
17 G to 12 in. of stroke.
198
iA
Further background information on analysis and testing of heli-
copter litter systems can be found in Reference 22.
9.3.2 Upward Loads
All litter systems should be capable of withstanding a minimum
upward load of 8 G.
9.4 LATERAL AND LONGITUDINAL LOADS
Litter systems for all aircraft should be designed to withstand
the load and deformation requirements indicated in Figure 69
in all radials of the lateral/longitudinal plane. The litter
lateral loads are made equal to the longitudinal loads because
the litters may be oriented in either direction depending upon
the aircraft.
I ~ ~30. ..
Acceptable failure area

A ceotable
25
M
-4 20 Unacceptable
0115
0 1 2 3 4 5 6
Total controlled forward** deformation of litter bed, in.
*G value based on 250-lb per litter position.

**Forward is the direction towards the nose of the aircraft
regardless of litter orientation in the aircraft.
Figure 69. Litter forward or lateral load and deflection

requirements for all types of Army aircraft.
4 199
The 20-G acceptable load level indicated in Figure 69 is predi-
cated on the tolerance to acceleration of an individual re-
strained by straps on existing "table top" litters. If litters
and allied restraint harnesses are designed for improved crash-
worthiness, the 20-G load should be increased to 25 G.
Acceptable or nonacceptable load-deformation characteristics

are read from Figure 69 in the identical manner as the readings
from Figures 59 and 60 for seats. The deformation is measured
with respect to the aircraft floor along the longitudinal axis
toward the nose of the aircraft, regardless of litter orienta-
tion.
9.5 LITTER RESTRAINT HARNESS TESTING
The restraint used in existing military litters consists of two

straps wrapped around the litter. These straps should with-
stand a straight tensile minimum load of 2000 lb (4000-lb loop
strength). The maximum elongation should not be more than
3.0 in. under the straight pull (end-to-end) test on a minimum
strap length of 48 in. Elongation is restricted for litter
belts in order to minimize dynamic overshoot.
9.6 LITTER SYSTEM TEST REQUIREMENTS
9.6.1 Static Test Requirements
9.6.1.1 General: Table 12 presents the static test require-

ments for complete litter systems. Since previous studies have
shown that existing litters will not withstand the loads as
specified in this chapter, the assumption must be made that a
litter of sufficient strength will be developed prior to imple-
menting these recommendations. The tests required include a
series of unidirectional tests to determine basic litter and
attachment strengths in the major axes. Also, a combined load-
ing test is required to evaluate the litter system performance
under static conditions simulating a severe crash loading sit-
uation with loading components in multiple directions. Since
the litter orientation can bg either lateral or longitudinal,
a single requirement is made for-transverse loading in the hor-
izontal plane (Test 5).
9.6.1.2 Unidirectional Tests: The test loads for forward, la-

teral, and downward loading of litter systems as presented in
Table 12 should be applied separately.
9.6.1.3 Combined Loads: Litter systems must demonstrate no

loss of system integrity under conditions of combined loads as
specified in Table 12.
200
igu
TABLE 12. LITTER SYSTEM STATIC TEST REQUIREMENTS
Test Loading direction

ref. with respect to Deformation
no. fuselage floor Load required requirements
1 Forward See Figure 69 See Figure 69
2 Lateral See Figure 69 See Figure 69

3 Downward See Figure 68 See Figure 68
4 Upward 8 G No requirement
5 Combined loading
Downward plus See Figure 68 See Figure 68
transverse load
along any radial
in the x, y plane
of the aircraft See Figure 69 See Figure 69
' 9.6.1.4 Point of Load Application: The loads should be ap-

plied through a body block that simulates a supine occupant.
9.6.1.4.1 Forward (Longitudinal) - Lateral Tests: For systems

using the existing litter, a rigid simulated litter may be sub-
stituted for the actual litter. This will enable application
of equal loads at all attachment points between the litter and
the suspension system and allow testing of the suspension
system. The rigid litter substitution does not apply if the
litter has adequate strength to take the loads.
9.6.1.4.2 Downward and Upward Tests: Downward and upward
loads may be applied to each vertical suspension point sepa-
rately. If the suspension system has the tilt-limiting -ea-
tures, and the litter is adequate, then the load should be ap-
plied at the center of gravity of the body block.
9.6.1.5 Deflection Measurements: Downward, forward (longi-
tudinal), and lateral deflections should be measured at the
bracket attaching the litter to the suspension system.
9.6.1.6 Load Determination: The test load should be deter-
mined by multiplying the required load factor (G) as specified
in Table 12 by 250 lb.
201
I
II
II
9.6.2 Litter System Dynamic Test Requirements

A single test to evaluate the vertical load-limiting system is
required. Litter systems with 95th-percentile anthropomorphic
dummies (250-lb total)
and 30 belb subjected
litter should of additional
to a triangular weight in pulse
acceleration each
of 48-G peak and 0.054-sec duration (42-ft/sec velocity change).
The same test pulse tolerances, data, handling, and processing

requirements as presented for the seats in Section 8.5 apply.
At least three accelerometers should be placed in the dummy;
one in the head, one in the chest, and one in the pelvic re-
gion.
tions inThetheinstruments should be positioned
vertical directions (x axis of to
the sense
supineaccelera-
occupant,
z direction relative to the aircraft). The input acceleration- I
time pulse also should be measured. It is advisable to use re-
dundant accelerometers to sense the input pulse to assure ac-
quisition of the needed impact environment data.
I!
, I
202t
Li
kI
IIIII II m
10. DELETHALIZATION OF COCKPIT AND CABIN INTERIORS

10.1 INTRODUCTION
The kinematics of body action associated with aircraft crash

impacts are quite violent, even in accidents of moderate sever-
ity. The flailing of body parts is much more pronounced when
the aircraft occupant is restrained in a seat with only a lap
bilt. However, even with a lap belt and a shoulder harness
that are drawn up tightly, multidirectional flailing of the
)-ad, arms, and legs, and to a lesser extent, the lateral dis-
placement of the upper torso within its restraint harnessing,
is extensive. If it were possible to provide adequate space
within the occupant's immediate environment, this flailing ac-
tion of a fully restrained occupant would not be a particular
problem. Since space for occupants is usually at a premium in
aircraft, especially in cockpit areas, it is not feasible to
remove structural parts of the aircraft sufficiently to keep
the occupant from striking them. The only alternative is to
design the occupant's immediate environment so that, when the
body parts do flail and contact rigid and semi-rigid struc-
tures, injury potential is minimized.
An occupant who is even momentarily debilitated by having his

head strike a sharp, unyielding structural object or by a leg
injury can easily be prevented from rapidly evacuating the air-
craft and may not survive a postcrash fire or d water landing.
The importance of occupant environment designed for injury pre-
vention, therefore, sbould be emphasized if optimum crash pro-
tection is to be ensured.
Several approaches are available to alleviate potential secon-

dary impact problems. The most direct approach, which should
be taken if practical, is to relocate the hazardous structure
or object out of the occupant's reach. Such action is normally
subject to tradeoffs between safety and operational or human
engineering considerations. If relocation is not a viable al-
ternative, the hazard might be reduced by mounting the offend-
ing structure on frangible or energy-absorbing supports and ap-
plying a padding material to distribute the contact force over
a larger area.
10.2 OCCUPANT STRIKE ENVELOPES
10.2.1 Full Restraint
Body extremity strike envelopes are presented in Figures 70

through 72 for a 95th-percentile Army aviator wearing a re-
straint system that meets the requirements of MIL-S-58095(AV)
(Reference 14). The restraint system consists of a lap belt,
203
t
10 i.n.
IHeel ret line for cockpit.

Air-c-ra t floor line for
troop compartment.
Figure 70. Full-restraint extremity strike

envelope - side view.
lap belt tiedown strap, and two shoulder straps. The forward
motion shown in Figures 70 and 71 was obtained from a test
utilizing a 95th-percentile anthropomorphic dummy subjected to
a spineward (-G ) acceleration of 30 G. The lateral mot-ion is-_
based on an exthapolation of datar7frorn the same 30-G test. In
positions where an occupant is expect-ed to wear a helmet, the
helmet dimensions must be added to the envelope of head motion.
204
7. 4 . - _______________ - .-- --- ill" -.

/ ___i__. /
enelp -- tovew
Figureto71.upr reqxtremity strike

FUll-restraint- Ay i
10.2.2 Lap BeJ~t-Only Restraint

Although upper torso restraint is required in new Army air- 1•~
craft, strike envelopes for a 95th-percentile aviator wearing
lap belt-only restraint are presented in.Figures 73 through 75
for possible use. They are based on 4-G accelerations and
4 in. of torso movement away from the seat laterally and for-
ward. In positions where an occupant is expected to wear a hel-
met, the helmet dimensions must be added to the envelope of
head motion.
205
9••., - ..
0- -i-
t i in00
.. He.l rest line for cockpit.

Aircraft floor line for
troop compartment.
Figure 72. Full-restraint extremity strike

envelope - front view.
10.2.3 Seat Orientation

The strike envelopes of Figures 70 through 75 apply to all seat
orientations.
10.3 ENVIRONMENTAL HAZARDS
10.3.1 Primary Hazards
The primary environmental hazards are those rigid or semirigid
structural members within the extremity envelope of the head
and chest. It can be seen in Figures 70 through 75 that the
strike envelopes allow considerable upper torso movement for
various seating and restraint configurations. Since the upper
206
10 in.
* Heel rest line for cockpit.

oor line for
troop compartment.
Figure 73. Lap belt-only extremity strike

envelope - side view.
torso, and particularly the head, is the most vulnerable part

of the body, maximum protection must be provided within its-
strike envelope.-
10.3.2 Secondary-Hazards
Secondary environmental hazards are those that could result in
trapping or injuring the lower extremities to the extent that
one's ability to rapidly escape would be compromised. The
enelop.
strie
movement of unrestrained lower extremities in a crash impact is I
not significantly influenced by method of body restraint. Con-
10.32an optimized
sequently, even with Haard
Seondaybody restraint system, those
areas within the Secodar
lower extremity strike
envronentaarethoe
haard envelope
tht include
culd must esut i
ample protective design. ijurng
he lwerextemites
trapingor o te exenttha
ones biltyto apdl esap

207 wul be- copoisd.h
-.. -. -% F
r'I -
- -
-9
K' -1
_ _ _ _ /
/ S
.14
0
4)
S
0
S
______ ______ I.
0
0
4)
41
S
4.)
'4
S
p-I
0
41
'U
i-1
0
LI
.14
208
$4p
31c 0
.14 0
P-0 rq,
Lm0
I41
*14
209
10.3.3 Tertiary Hazards
Tertiary environmental hazards are those rigid and semirigid
structural members that could cause injury to flailing upper
limbs to an extent that could reduce an occupant's ability to
operate escape hatches or perform other essential tasks.
10.4 HEAD IMPACT HAZARDS
10.4.1 Geometry of Probable Head Impact Surfaces
Aircraft in the U. S. Army inventory in 1965 were examined to
determine the kinds of contact hazards most commonly found
(Reference 94). Typical hazards in the cockpit area included
window and door frames, consoles, control columns, seat backs,
electrical junction boxes, and instrument panels. Reference 94
presents further details of these impact hazards and a statis-
tical analysis of head injuries in both civilian and military
aircraft accidents. Contact hazards commonly found in aircraft
cabin areas include window and door frames, seats, and fuselage
structure. Use of suitable energy-absorbing padding materials,
frangible breakaway panels, smooth contoured surfaces, or duc-
tile materials in the typical hazard areas mentioned will re-
duce the injury potential of occupied areas.
10.4.2 Tolerance to Head Impacts
Protection of the head in the form of protective helmets and
energy-absorbing structure and padding in the occupant's imme-
diate environment is considered to be essential since, under
certain circumstances, even the forces incurred in minor crash
impacts could cause unacceptably high head impact velocities.
Tolerance levels for head impact are discussed in detail in
Volume II, and the reader should refer there for an understand-
ing of the problem. However, for the case of forehead impact
on a flat surface, which is pertinent to the discussion of this
section, the most widely accepted collection of tolerance data
is represented in the tolerance curve of Figure 76. These data,
resulting from impact tests conducted on animals and human ca-
davers at Wayne State University, demonstrate the contribution
of both acceleration and pulse duration to the tolerance cri-
terion.
94. Haley, J. L., Jr., et al., HELMET DESIGN CRITERIA FOR IM-
PROVED CRASH SURVIVAL, Aviation Safety Engineering and Re-
search (AvSER), Division of Flight Safety Foundation,
Inc.i USAAVLABS Technical Report 65-44, U. S. Army Avia-
tion Materiel Laboratories, Fort Eustis, Vircinia, Jan-
uary 1966, AD 628678.
210
i
600 5oo IT 7,
300 Dengero s_-__to life 4--

'In
Time duration of effective

acceleratio~n, meec
unyielding sufcs (Fo eeec 5
in
brai dan eheadags aintplane,
Figur;e 400an SaeTlrac uv for the humann

10..3 estProcedures
procedure for evaluating the effectiveness
Thesimpitest
structure and padding in preventing serious head
injury makes use of an instrumented headform. The headform,
equipped with an accelerometer, can be propelled by a ram,
dropped, or swung on a pendulum to impact the surface to be
evaluated. The recommended procedure is described in SAE J921
(Reference 96). The measured acceleration pulse can be aver-
aged for comparison with the Wayne State Tolerance Curve, or
integrated to compute a Severity Index, as discussed in Sec-
tion 4.4.1 of Volume LI.
Figure 77 shows typical head velocities relative to the seat
as measured on anthropomorphic duammies, cadavers, and live
hurin subjects in dynamic seat tests. Various combinations
95. Patrick, L. M.1 Lissner, H. R., and Gurdjian, E. S., SUR-
VIVAL BY DESIGN - HEAD PROTECTION, Proceedings, Seventh
Stapp Car Crash Conference, Society of Automotivengi-
neers, Inc., New York, 1963.
96. SAE Recommended Practice, SAE J921b, MOTOR VEHICLE INSTRU-
MENT PANEL LABORATORY IMPACT TEST PROCEDURE - HEAD AREA,
SAE Handbook, 1979, Part 2, Society of Automotive Engi-
neers, Inc., Warrendale, Pennsylvania, 1979, pp. 34.133-
34.134.
-L-1
- live volunteer
* " cadaver
S+- 95% dummy
lIbo - lap belt only
lbsh - lap belt and shoulder harness
60
4. 30 . .. .. ..
40
0 V__
o 20 40 60 80
i Seat velocity change in test pulse, ft/sec
Figure 77. Measured head velocities in sled tests with

anthropomorphic dummies and cadavers.
of occupant restraint were used and are so indicated on each

curve.
10.5 INSTRUMENT PANEL STRUCTURE PROXIMITY
Most aircraft cockpits are, of necessity, very compact. It is
necessary, for instance, for &_pilot to be able to reach var=
ious controls on the instrument panel by leaning forward no
more than 1 in. (the extent of unlocked inertia reel exten-
sion). Consequently, instrument panels must be close enough
to be reached and seen easily. Unfortunately, this usually
requires that the instrument panel and its supporting struc-
ture be placed directly above the pilot's lower legs as they
rest normally on the rudder pedals. When a seated pilot is
exposed to -G. (eyeballs-out) accelerations in a crash, the
212
lower limbs are abruptly extended longitudinally with some up-
ward velocity. In this process, the lower leg usua'ly impacts
on the lower edge of the instrument panel. iepending on the
particular aircraft configuration, this contact can take place
from the kneecap down to the ankle. In -view of the high ve-
locities associated with such flailing, disabling lower leg
injuries are common in accidents where high -G forces are
present. It is essential that desigyiers considgr using suit-
able energy-absorbing padding materials, frangible breakaway
panels, or ductile panel materials for structure within the
lower leg strike envelope.
10.6 RUDDER PEDAL CONFIGURATION
In certain types of aircraft accidents, the pilot's feet remain
on the rudder pedals instead of flailing upward and outward.
If the rudder pedal is a simple, bar type of arrangement, the
heel may be forced under the pedal. When the body is exposed
to a combination of vertical (Gz eyeballs-down) and longitu-
dinal (-G eyeballs-out) forces, pelvic rotation around the
lap belt *ill almost invariably occur unless a lap belt tie-
down strap is used. This pelvic rotation, which forces the
feet hard against the rudder pedals, can occur even though the
lap belt is drawn up tightly. A loose or slack lap belt aggra-
vates the tendency toward pelvic rotation. If the forces are
great enough, a badly injured or trapped foot can result.
Therefore, it is desirable to design the rudder pedals and sur-
rounding structure to prevent this from occurring. This is
usually done by providing a pedal capable of supporting both
the ball of the foot and the heel, and by providing a surround-
ing structure of sufficient strength to prevent crushing and
trapping of the lower limbs. The geometry required by MIL-STD-
1290(AV) (Reference 1) to prevent entrapment of feet is illus-
trated in Figure 78.
10.7 CONTROL COLUMNS
Control columns located in front of flight crew stations can
present a serious hazard to crewmembers if they fail- at a-ny
appreciable distance above the-aircraft floor. Such a failure
often leaves a torn, jagged stump that can inflict serious in-
jury to a crewman should he be thrown against it during impact,
move into it as an energy-absorbing seat strokes, or come in
contact with it during egress after impact. It is recommended
that control columns be designed so that fracture due to the
occupant's striking the column will occur at a point no more
than 4 in. above the pivot point. The failure should occur in
the form of a clean break, leaving no jagged or torn edges.
Control columns that pass longitudinally through the instrument
panel are not recommended since these tend to impale the crew-
members in severe longitudinal impacts.
"213
A K -CK
IIL,
Dimensions A, B, and C must be either

less than 2 in. or more than 6 in.
Figure 78. Antitorque, or rudder, pedal geometry

to prevent entrapment of feet.
10.8 SIGHTING AND VISIONIC SYSTEMS
Delethalization of the copilot/gunner (CPG) station of an at-

tack or scout helicopter equipped with a weapon sighting opti-
cal relay tube (ORT) can present a difficult design problem.
The copilot/g-unner crewstation activities demand that the CPG
will be either in contact with the ORT eyepiece during hazard-
ous nap-of-the-earth (NOE) flight or close to the eyepiece when
sitting in the full upright, erect position. Operational loca-
tion of the CPG head, when not looking in the ORT, may be as
little as 8.5 in. from the eyepiece. Therefore, it can be ex-
pected that the CPG, when restrained by a MIL-S-58095(AV) re-
straint system, will contact the ORT eyepiece under nearly all
impacts over 4 G (see Figure 79). Any deformation of the bulk-
head which would cause the ORT to move rearward will only fur-
ther ensure head contact. Forward motion of the upper torso
after head contact with the ORT could cause spinal injury.
214
ON1
-0
"a.14"
hi4
0
04
"A4
2154
Under NOE conditions with the CPG looking through the ORT, it
can be expected that no warning of impending impact will occur.
Regardless, any courses of action taken by the CPG to hold him-
self erect will probably not help in keeping his head from
striking the ORT due to head flailing and body stretch. An-
other factor that further decreases the distance between the
head and the ORT eyepiece is the travel of the seat as it
* strokes under crash loads.
Possible ORT hazards to the lower extremities and the torso
consist of the sharp unyielding lower structure of the ORT. In
addition, the rudder pedals may be located adjacent to the ORT.
During a crash, the potential displacement of the ORT may cause
the CPG's legs to become entrapped. A summary of typical ORT
crash hazards is presented in Table 13.
TABLE 13. POTENTIAL OPTICAL RELAY

TUBE CRASH HAZARDS
Location Type of
Hazard of injury Injury Cause
1 Head Laceration, Head strikes ORT due to
Fracture, flailing forward and down-
Concussion ward on impact
2 Head/Chest Crushing, Head/chest strikes ORT due

Avulsion, to ORT displacing rearward
Fracture
3 Head/Chest Laceration, CPG seat displaces downward
Crushing, and forward during energy-
Fracture absorbing stroke. Contact
of the head/chest with
sharp edges of franged ORT.
4 Arm Laceration, CPG arms flail forward on
Fracture longitudinal impact - _
5 Lower torso Avulsion, ORT di-splaces rearward on
Laceration, longitudinal impact
p ~Crushing,
S~Fracture
6 Leg Laceration, CPG leg flails forward on
Fracture longitudinal impact
7 Leg Crushing CPG leg trapped between air-
craft structure and displac-
ing ORT
216
LI/7
The cockpit should be designed to minimize the probability of
the CPG head/neck striking the ORT and minimize injury if the
CPG should strike the ORT, for both the "head-up" and "head-
down" CPG positions. Some of the options available to the de-
signer given this task are:
"* ORT Eyepiece Relocation - Consideration should be

given to reducing occupant strike hazards by moving
the ORT further away from the CPG.
"* Restraint System - The restraint system of Figure 44 4
would offer improved upper torso restraint, particu-
larly when combined with the power-haulback inertia
reel.
"* Inflatable Restraint - Consideration should be given
to an inflatable restraint system (see Section
7.2.4). This type of restraint harness can prevent
injury to the CPG in both the erect and head-down
position by reducing slack and increasing the sur-
face area of the body over which the harness reacts.
"* Frangible/Breakaway Features - ORT or ORT components
designed to be frangible should break away at a total
force not to exceed 500 lb. For the frangible ORT,
this force should be applied along any direction of
loading within the plane normal to the axis of the
ORT, as well as along the axis of the ORT. Break-
away point(s) of the ORT should be outside the head
strike envelope.
"* Collapsible Features - If the ORT is designed to

collapse in order to avoid injuring the CPG, the
collapse load along the axis of the ORT should not
exceed 500 lb. Figure 80 illustrates one crushable
sight eyepiece concept (from Reference 97). Two ad-
vantages of the crushable sight eyepiece are that it
is always available and, it should function regard-
less of head location. A helmet crash-absorber pad
would attenuate crash loads to the helmet when avail-
able crushing is expended. - -
e Power-Haulback Inertia Reel (PHBIR) - On the basis

of Air Force testing accomplished for the devel-
opment of PHBIRs, the retraction time is 0.3 to
E., DESIGNING CRASHWORTHI-

97. NESS R.,
Fox, INTO Kawa, M., and paper
THE YAH-63, Sharp, presented at the Aircraft
Crashworthiness Symposium, University of Cincinnati, Cin-
cinnati, Ohio, October 1975.
217
Normal operation
SPH-4 helmet
,•- reference
Helmet crash-
energy absorber / --Pressure support
area approximately
Crash energy- 6 in.2
absorbing material
Pre
Opiasliploaded
rings-- prInner fa-cecontours
ofl• mas
Optical relay Crush stroke Of face mask

tube Outer contour
face mask
20-MM
eye relief
After crash stroke
Helmet crash-
energy absorber
Yielded
material
Deformed tube
Figure 80. Crushable eyepiece concept. (From Reference 97)
0.4 sec, which is too slow for effectiveness in most

crashes. If this time were reduced, the retraction
velocity of the torso would have to be increased
considerably over the current limit.of 9 ft/sec. A
retraction velocity greater than this is not recom-
mended due to the lack of human tolerance data on
this type of loading. In a crash with a single pulse
218
• m " m ' n' " m m m° '• ' " "I'

of say 30-G peak and 50-ft/sec velocity change, the
retraction velocity should be approximately 25 ft/sec;
therefore, thle known tolerance limits would be ex-
ceeded at the higher velocity. In summary, the PHBIR,
as currently qualified under both Air Force and Navy
military specifications, requires excessive time to
position the tcrso by crash sensing. To be fully
effective, the system should move the torso into po-
sition in approximately 0.06 sec, but the resulting
acceleration would exceed known human tolerance
limits. The primary crashworthiness advantage of the
PHBIR would be as a manually activated tightening de-
vice for the head-up CPG position; the PHBIR offers
only limited advantage for the head-down CPG position.
10.9 ENERGY-ABSORBING REQUIREMENTS
FOR COCKPIT AND CABIN INTERIORS
10.9.1 General
To minimize occupant injury, the acceleration experienced dur-
ing secondary impacts of the occupant with surrounding struc-
tures must be reduced to a tolerable level. The areas of con-
tact to be considered for energy absorption include instrument
panels, glare shields, other interior surfaces within the occu-
pant's strike envelope, and seat cushions. A padding material
should not only reduce the decelerative force exerted on an im-
pacting body segment, but should distribute the load in order
to produce a more uniform pressure of safe magnitude.
As an example of the need for an energy-absorbing system to
possess both these characteristics, consider the case of head
impact. Head injuries sustained from impact may be grouped in
two general categories. The first is skull fracture with its
inherent brain damage and danger to life. The second is in-
jury to facial tissue and bone structure with a lesser pi:oba-
bility of brain damage.
A system that is to absorb the energy of an impacting head

should cushion the head to prevent skull fracture or penetra-
tion from protruding objects as a result of decelerative forces.
It should also distribute the forces-to minimize injury to tis-
sue and bone Etructure. The cushioning material used must ef-
fect low peak deceleration and low average stress. Figures 81
and 82, taken from Reference 98, indicate the impact behavior
98. Lee, W. M., and Williams, B. M., CUSHIONING AND LOAD DIS-
TRIBUTION PERFORMANCE OF PLASTIC FOAMS, Paper No. 700453,
Society of Automotive Engineers, Inc., New York, May 1970.
21
.1
*...r-s..
*.-.
.
175-
Semirigid urethane
15o foam (9 lb/ft 3 )
S125
0
3 Polystyrene
beadboard
a)
C.) (1 lb/ft 3 )
10
E 75-
V
4)
4 50- _____ i____ ___
\,Polyethylene
foam (2 lb/ft
25 (Te
(Testing at room temperature)
0 5 105 20 25 30
Impact speed, mi/h
Figure 81. Impact behavior (headform deceleration

versus speed) of three padding materials.
of three plastic foams. The foam sample specimens used to ob-

tain these data were 6 in. thick to minimize any bottoming-out
effect. Although the semirigid urethane appears to be a fair
cushioning material, it does not distribute the load as well
as the materials with which it is compared. A fair cushioning
material is not necessarily an effective load distributor.
Both criteria must be considered in the selection of a material
that is to provide impact protection for the head.
220
I-lrrl rl --. 1- ml-rr'11*l'

. .i 120. I I
_ _I_
(Testing at room temperature)
I
S100 Semirigid urethane or
foam (9 lb/ft 3 )"
- S80 .
.U 6 0 -P
40
60 - -3
0 foam (2 lb/ft
20
2
fir Polystyrene
1/' beadboard
Z (1 lb/ft)
0 5 10 15 20 25 30
Impact speed, mi/h
Figure 82. Impact behavior (average dynamic stress

versus speed) of three padding materials.
In addition to protecting bone structure and facial tissue, the

energy-absorbing system must also afford protection- against
intracranial lesions. Cerebral concussion, and the loss of
consciousness which often accompanies it, may occur if the head
is subjected to excessive decelerative forces. Mattingly, et
al. (Reference 99), in discussing possible intracranial lesions
and cerebral trauma including concussion, swelling, contusion,
laceration, and hematoma, conclude that in order to prevent
99W Mattingly, T. E., e al., INVESTIGATION OF VIBRATION AND

IMPACT PROTECTION OF THE HUMAN HEAD AND NECK, Northrop
Corporate Laboratoriesi AMRL Technical Report 69-112, Air
Force Systems Command, Wright-Patterson Air Force Base,
Ohio, December 1969, AD 702124.
221
I.......~..
head injury, materials must be carefully selected to absorb
and attenuate the energy of impact. The material must reduce
the level of acceleration, the rate of onset, and the amount
of energy transmitted to the head.
10.9.2 Types of Padding Materials and Properties
The most useful types of materials for energy-absorbing pad-
ding are plastic foams. A foamed plastic is usually totally
unlike the same plastic in the solid: unlike in properties,
in processing, and usually in applications. Three steps are
involved in producing a cellular structure in a polymer: 1)
preparation of polymeric material into a viscous liquid state,
2) introduction of fine bubbles of gas to produce expansion,
and 3) solidification of the foamed plastic to stabilize the
foamed structure. The particular process used in manufacturing
foam materials has a direct effect on their properties and can
result in products of the same chemical composition being very
different in performance.
10.9.2.1 Material Form: The form in which the foam material
is commercially available influences its adaptability to vehi-
cle applications. Slab and molded foams are often used in the
construction of instrument panels and seat systems. Differ-
ences in properties due to varying the form should be consid-
ered in the selection of a material. For example, Figure 83
shows the variation of minimum tensile strength versus product
density for polyethylene foam in sheet and plank forms.
10.9.2.2 Classification of Foams: Foams can be described as
flexible or rigid. A flexible foam recovers when deformed,
whereas a rigid foam cannot sustain multiple impacts. Flexi-
ble foams are most widely used in situations where energy ab-
sorption is important.
Another method of classifying foams is open-cell or closed-cell.
An open-cell foam contains individuAl cells that interconnect
with the others, while in a closed-cell foam individual cells
are completely enclosed by a wall of plastic.---
Plastic foam materials also can be classified according to
their chemical composition. Several energy-absorbing plastic
foams and some of their typical applications are listed in
Table 14.
10.9.2.3 Material Properties: The selection of a foam mater-
ial for vehicle energy-absorbing applications involves an eval-
uation of its processabilityl its mechanical, thermal, and
222
• • *14
~SheetPln
heetPlank
No100 L
op
4 80
w /
Prodc d
Figure 83. Minimum tensile strength versus product

density for polyethylene sheet and plank.
(From Reference 100)
chemical propertiesi as well as its cost. Along with the pri-

mary foam materials, the characteristics of adhesives and sur-
face coatings must be considered, particularly with respect to
emission of smoke and toxic vapors. The characteristics of
suitable materials for such use are listed
'II below:
e Adaptability
-. and
4a). -ease of processing
* High energy dissipation

-'U m -..
a Effective load distribution
* Low rebound
o Temperature insensitivity
e Low w0ter absorption
200. Fradosj. j., PLASTICS ENGINEERING HANDBOOK OF THE SOCIETY

OF THEt PLASTICS INDUSTRY, I*NC., Van Nostrand Reinhold Co.,
New York, 1976, pp. 499-567.
S~223.
e Resistance to chemicals, oil, ultraviolet radiation,
and sunlight
* Nontoxic, fume generation
* Favorable flammability rating
* Minimal smoke generation
* Durability and long life
* Cost competitive I
* Aesthetic
TABLE 14. ENERGY-ABSORBING PLASTIC FOAMS

AND SOME TYPICAL APPLICATIONS
r. Semirigid and flexible urethane foam

Aircraft, automobile, and furniture seat cushions,
safety padding, arm rests, sun visors, horn but-
tons, bedding, carpet underlay, packaging delicate
products.
2. Polyvinylchloride foam
Crash padding in automobile head liners and sun
visors, flooring, shoe soles and heels, automo-
bile door panels, seating upholstery sealants,
gaskets, bumperstock.
3. Polystyrene foam I
I
Insulation, packaging.
4. Expanded rubber
Bus and subway seat cushions, truck and ship
mattresses, gaskets, hose insulation.
5. Polyester foam
Short-run, custom-type seat cushioning.
6. Polyolefin foam
Packaging, gasketing, water sports equipment, rug
underlay, athletic padding, antivibration padding.
Foam materials are most often characterized by the mechanical

properties listed below, where it may be noted that several
of the properties apply only to rigid or flexible foams. For
224
example, compressive strength is not relevant in considering
flexible foams. SA compression-set test, on the other hand,
The
applies only to flexible materials.
• Density
* Tensile strength
* Tensile modulus
* Compressive strength
• Compressive modulus
* Flexural strength
* Flexural modulus
* Tear strength
* Compression set
* Compression deflection
e Elongation
* Rebound
e Hardness
* Impact
Properties of possible interest in selection of a material for
energy-absorbing applications are presented in Table 15 for
several applicable materials (data taken from References 100
through 103).
101. ENSOLITE, Publication ASP 9997, Expanded Products Depart--

ment, Uniroyal, Inc., Mishawaka, Indiana, 1977.
102. TEMPER FOAM, Form TF-20, Edmont-Wilson, Division of Bec-
ton, Dickenson and Co., Coshocton, Ohio, 1975.
103. Sundquist, D. J., POLYOLEFIN FOAMS, Monographs on Plas-
tics, Vol. 1, Part 1, 1972, pp. 193-295.
1 225
* ..- _ ." . . .-- - ;.. _..*, ."....- • .-
Sllim
ad to
o Ad.
40L
'4~'u * v-
I
.in
U'-403
j 3 4
0.0
.4 0~o I
0U~ in V
0in
0 N nP-
14 (hN
n
00
>44). 4
U.. 0jN
o 0 H 41 0 n0 . 0
4, -,4 .,4 4j. 0
V. r.
0.4 4
41
a04
cc 0f co5 0
to4 4m in
* 4
. j.22 4 ~ . 4 J ~ 61
10.9.3 Standard Test Methods
ASTM standard test procedures. are widely used by manufacturers

to specify various properties of a particular type of material.
Table 16 summarizes ASTM test methods and specifications for
flexible cellular plastics that provide a basis for comparison
of materials. Here it may be noted that most ASTM tests in-
volve simple tests, whereas the operational environment in-
volves dynamic loading and more complex conditions.
In particular, ASTM D 1564-71 describes "Standard Methods of

Testing Flexible Cellular Materials-Slab Urethane Foam" (Ref-
erence 104). Among other tests, there are compression-set and
load-deflection tests. In the compression-set test, the me-
thod consists of deflecting the foam specimen under specified
conditions of time and temperature and noting the reduction of
specimen thickness after removal of the load. The compression
device consists of two flat plates larger than the specimen.
In the load deflection test, one method consists of measuring
the Indentation Load Deflection (ILD) value, which is the load
necessary to produce a specified 25-perceft or 65-percent in-
dentation in the specimen under a 50-in. circular indenter
foot. Acceptable deflection rates range from 1.0 to 15.0
in./min. A second method, which uses the same indenter, ob-
tains the deflections under specified loads of 4.45, 111, and
222 N (1, 25, and 50 lb) during loading and 111 N during un-
loading. Theae deflections are reported as Indentation Resid-
ual Gage Load (IRGL) values. The latter method, which involves
indentation to specified loads, is intended for use with seat
cushion materials.
The above tests provide results that specify the material, but
do not necessarily portray its performance under actual impact
situations. A simple dynamic drop test, such as ASTM D1596-64
(1976), "Standard Test Method for Shock-Absorbing Characteris-
tics of Package Cushioning Materials" (Reference 104), more
closely simulates actual impact conditions. An acceleratior-
time curve is obtained by mounting a transducer on the dropping
head. The parameters evaluated are peak deceleration and the
dynamic set of the specimen. This method allows the test
parameters to vary and yet is simple enough to ensure repeat-
ability among different test facilities. In a drop test, the
test parameters are: the drop height that determines the im-
pact velocity, the weight and surface area of the impactor,
and foam thickness.
104. Lukens, R. P., et al., 1977 ANNUAL BOOK OF ASTM STANDARDS,
American Society for Testing and Materials, Easton, Mary-
land, 1977, Parts 20, 38, 48.
227
- 4 -... . - . . . . . .. • _
r
TABLE 16. SUMMARY OF ASTM TEST METHODS AND

SPECIFICATIONS FOR FLEXIBLE
CELLULAR PLASTICS (REFERENCES
104 AND 105)
D1564-71* Testing Flexible Cellular Materials-

Slab Urethane Foam
D1667-76* Specification for Flexible Cellular
Materials - Vinyl Chloride Polymers
and Copolymers (Closed-Cell Sponge)
Materials - Vinyl Chloride Polymers
and Copolymers (Open-Cell Foam)
(1975) Materials - Latex Foam
DI056-73* Specification for Flexible Cel.ular
Materials - Sponge or Expanded Rubber
D3575-77 Testing Flexible Cellular Materials
Made From Olefin Plastics
D1596-64* Test for Shock-Absorbing Characteristics
(1976) of Package Cushioning Materials
D2221-68* Test for Creep Properties of Package
(1973) Cushioning Materials
D1372-64* Testing Package Cushioning Materials
(1976)
D696-70* Test for Coefficient of Linear Thermal
Expansion of Plastics
E143-61* Test for Shear Modulus at Room Temp-
(1972) erature
D412-75* Tests for Rubber Properties in Tension
D1433-76" Test for Rate of Burning and/or Extent
and Time of Burning of Flexible Thin
Plastic Sheeting Supported on a 45-
degree Incline
D1692-76 Test for Rate of Burning and/or Extent
and Time of Burning of Cellular Plas-
tics Using a Speciman Supported by a
Horizontal Screen
*Indicates that the standard has been approved as

American National Standard by the American National
Standards Institute.
228
Other standard test procedures include SAE J815, "Load Deflec-
tion Testing of Urethane Foams for Automotive Seating," as de-
scribed in Reference 106. This test points out the factors of
interest in testing materials for vehicle seat cushions: the
thickness of the padding under the average passenger load, a
measurement that indicates the initial softness, and a measure-
ment that indicates resiliency. SAE J815 determines load ver-
sus deflection by measuring the thickness of the padding under
fixed loads-of 1 lb, 25 lb, and 50 lb with a circular indentor
foot (see Reference 105).
Also, SAE J388, "Dynamic Flex Fatigue Test for Slab Urethane
Foam" (Reference 107), describes procedures for evaluating the
loss of thickness and the amount of structural breakdown of
slab urethane foam seating materials. A test specimen is mea-
sured for thickness under a specified load and, subsequently,
subjected simultaneously to compressive and shear deformation
under a constant load for a specified number of cycles. In
the constant load height measuring test, a flat, circular
indenter foot of 50 in. wilh loads from 1.0 to 75.0 lb is de-
flected at rates from 2 to 8 in./min. The constant load dy-
namic fatigue apparatus uses rollers in a more complicated
setup.
SAE J921, "Motor Vehicle Instrument Panel Laboratory Impact
Test Procedure-Head Area," describes a test procedure for eval-
uating the head impact characteristics of such areas as in-
strument panels (Reference 96). An SAE J984 headform with
an effective weight of 15.1 lb is impacted at specified posi-
tions. The parameters evaluated are the impact velocity, the
acceleration-time history of Lhe htuoiorm, and the start of
impact, with optional measure,4Ient of the rebound velocity and
the headform dynamic displacement.
105. ENCYCLOPEDIA OF POLYMER SCIENCE AND TBCHNOLOGY - PLASTICS,

RESINS, RUBBER, FIBERS, John Wiley and Sons, Inc..,- New
-York, 1965, Vol. 3, pp. 98-126.
106. SAE Recommended Practice, SAE J815, LOAD DEFLECTION TEST-
ING OF URETHANE FOAMS FOR AUTOMOTIVE SEATING, SAE Handbook
1979, Part 2, Society of Automotive Engineers, Inc., War-
rendale, Pennsylvania, 1979, p. 34.31.
107. SAE Recommended Practice, SAE J388, DYNAMIC FLEX FATIGUE
TEST FOR SLAB POLYURETHANE FOAM, SAE Handbook, 1979,
Part 2, Society of Automotive Engineers, Inc., Warrendale,
Pennsylvania, 1979, pp. 34.28-34.30.
229
-- A, ---- -------.. -.. ~.. . . . . .

S10.9.4 Research on Materials for Energy-Absobing plications
Static tests that deviate from ASTM test procedures and simple
dynamic tests that are intended to grossly simulate crash con-
ditions have been performed by manufacturers and users with
different types of materials. Several of their approaches and
their energy-absorption criteria are discussed below.
10.9.4.1 Acceptable Stress-Strain Characteristics: Haley, et
al., have investigated design criteria for pacding materials,
as described in Reference 94. According to their conclusions,
energy-ablorbing materials with stress values between 40 end
80 lb/in. at 50 percent strain would offer reasonable survival
potential for head impacts on flat surfaces at velocities of up
to 20 ft/sec with a padding thickness of 1.5 in. More recent
unpublished data gathered by the Army's Aeromedical Research
Laboratory (USAARL), Fort Rucker, Alabama, indicates that the
above stress values are too high because the values were based
on the compressive strength alone while it is probable that
tensile stresses and shear stresses around the periphery of the
compressed areas play a large role in the total force resisting
compression. Regardless of the stress distribution in the pad-
ding material, the USAARL research has sholn that the stress
level should fall between 30 and 45 2 1b/in. for padding less
than I in. thick and 20 to 30 lb/in. for padding greater than
I in. thick in order to prevent peak G pulses from exceeding
the tolerance values stated in Volume 11. These crush strength
values, as illustrated in Figure 84, are recommended. These
levels are expected to prevent unconsciousness (within the
limits of the crush depth) for head impacts. The lower stress
level for the thicker padding is based on: (1) the average de-
sign decelerative level must be reduced as the depth of the
padding and concomitant time duration are incr'.eased to meet the
known tolerance limits stated in Volume II, and (2) a larger
area of foam is crushed as the head sinks into the thicker pad.
Use of a padding as proposed in Figure 84, is intended to limit

head peak G values tc 160 for the thin pads and 120 G for the
thicker padding.
The criteria of Figure 84 are to be satisfied by the padding

material over the entire anticipated operating temperature
range if the potential for survival is to be maintained. Prac-
tical considerations and risk analysis, however, may reduce
the temperature range requirements. Figure 85, taken from
Reference 108, indicates the temperature dependency of the
stress-strain properties of a particular foam material. It
illustrates the variation experienced by many padding materials
108. PACKAGING WITH ETHAFOAM, Publication No. 172-221-IOM-767,
Dow Chemical Company, Midland, Michigan, revised 1966.
230
. 40Ot~l • '
0
~ vb.:cept~hi.
CUAhGceptAbl. Acceptebim
20
D) NrO paddinq thicknes1 lies thn on. Lnch For padding thickmi grq ater thah one inch
_ _ _ _ _ _ _ _ I ,- I ,
Str,!n, petrent
Figure 84. Recommended stress-strain properties for padding

material for head contact, with cushion thickness
of at least 1.5 in. (From Reference 94)
and indicates that temperature sensitivity must be considered

as a padding material selection criterion.
Stress-strain curves for a polyurethane-foamed plastic are
shown 4n Figure 86. The curves are taken from Reference 109
and indicate that a 6-in. tnickness of the foam with a density
of 5 lb/ft will satisfy the criteria of Figure 84 (superim-
posed as a crosshatched area) over at least part of the oper-
ational temperature range. The lowest impact velocity used to
obtain the data of Figure 86 was 50 ft/sec. A weight of 295 lb
impacting at this velocity requires the absorption of over
11,000 ft-lb of energy by the padding material. This require-
ment is obviously considerably more demanding than that of 90
ft-lb of energy at an impact velocity of 20 ft/sec, as de-
scribed above. Further work with variations of this foam may
yield a material that will satisfy the design criteria with a
thickness of 2 in. or less.
I
10.9.4.2 Bioengineering Approach to Material Evaluation:
Daniel investigated the injury-reducing functions of crash pad-
ding, considering strength of skull segments, as described in
Reference 110. He concluded that because the cranial vault
109. ENGINEERING DESIGN HANDBOOK, DESIGN FOR AIR TRANSPORT
AND AIRDROP OF MATERIAL, AMC Pamphlet No. 706-130, U. S.
Army Materiel Command, Wahhington, D. C., December 1967,
AD 830262.
110. Daniel, R. P., A BIO-ENGINEERING APPROACH TO CRASH PADDING,

Paper No. 680001, Society of Automotive Engineers, Inc.,
New York, 1968.
231
________.._________ .
m m _______ - - ... ...

-A
20
18 ._
16.
14 - ,
S12 ____
1i0 __ -40OF _
+73 F
~' 8
0-
0 10 20 30 40 50
Strain, percent
Figure 85. Stress-strain curves for polyethylene

foam. (From Reference 108)
(above the eyebrows) is strong under localized impact, padding

used for protection of this region has the primary-function of
energy absorption to reduce the possibility of brain damage.
On the other hand, padding for facial protection should dis-
tribute the impact load over the weaker facial bones, and re-
quired energy absorption would be provided by the supporting
structure. His suggested evaluation criterion for energy-
absorbing materials, based on t program of 91 impact tests, is
illustrated in Figure 87. For any given material, plotting on
these curves the results of a test conducted according to the
given parameters would enable the determination of a material
"efficiency," where a 100-percent efficiency would correspond
to the deceleration achieved by an ideal square-wave energy
232
M0 I
so 0
70 - all P": 00
,//7
70 -
C
-.4
N, 50
k lI 41
Wn 40
20
no Claim*
(Ib/Itl)1
I
Tcqts
/0en~tytong
Ye.ar
Drolped
1.)I) '
Velocity I (rad) Size
tsec) i n.)
40 4 1S ;95 I 51, 74x2lxi
7
10 3- 0 1 I f ) N/A Qus . 1 x1,sxO.4,
2.0 0. 2 1979 N/A - ,,e- 2.0Na.sox.O
*Foem taken from foa.-itn-place helmot.

rFom apec imen produced from U.S. Air Force Helmet Foaemtiq Kit
witI. equal tartm of Prlyol an-I Iaocyanate.
0 I , I I 1
0 20 40 60 80 100
Compressive stra!.n, percent
Figure 86. Effect of density on stress-strain curves

for polyurethane-foamed plastic.
233
10 o
- , : . : ..],1• ;-4
4J 4
04
0I 'P ,4400
Q-r4 U 0 t$4n 0
~4 04)- iv ~ N ~
'04 41qb .0
to x~JUU
CH
4' )
U 00
* 0C
- "4
Ln I 02
en~ N
'UOTvJ9aO-
waope
234 '4.
absorber of the given thickness. According to Daniels, energy-
absorbing materials might be selected on the basis of maximum
efficiency.
Evaluation criteria for load-distributing applications, which
are illustrated in Figure 88, are based on the following as-
sumptions:
* A load-distributing pad should permit the face to
penetrate its surface relatively easily and then
maintain a cushioning layer of foam between the base
and the underlying structure during collapse of the
understructureI
* The understructure should deform at close to the 80-G
(1200 lb) face tolerance level expressed in both SAE
J885 and Federal Motor Vehicle Safety Standard 201
(References 111 and 112, respectively).
10.9.4.3 Energy-Absorbing Efficiency Calculations.- Th- energy-

absorbing characteristics of foamed polymers were mathemati-
cally calculated by Rusch from low-speed experimental data for
compressive strain and modulus (Reference 113). Materials were
characterized by three parameterst energy-absorbing efficiency,
impact energy per unit volume divided by foam modulus, and the
maximum decelerating force per unit area divided by foam mod-
ulus.
An ideal energy absorber would provide a constant deceleration

from an initial speed, vii for 100 percent of its thickness, h
The maximum deceleration for an ideal absorber is then given by
dm /2h
vi d~i (36)
111. SAE Information Report, SAE J885, HUMAN TOLERANCE TO IM-

PACT CONDITIONS AS RELATED TO MOTOR VEHICLE DESIGN, SAE
- Handbook 1979, Part 2, Society of Automotive Engineers,
Inc., Warrendale, Pennsylvania, 1979, pp. 34,114-34.117.
112. U. B. Code of Federal Regulations, Title 49, Part 571:
MOTOR VEHICLE SAFETY STANDARDS, 201, OCCUPANT PROTECTION
AND INTERIOR IMPACT, Government Printing Office, Washing-
ton, D. C., (Rev.) 1978.
113. Rusch, K. C., IMPACT ENERGY ABSORPTION BY FOAMED POLY-

MERS, Journal of Cellular Plastics, Vol. 7, No. 2, 1971,
pp. 78-83.
235
--
...... - --- --------.---.--.-.--... - -- -- , 4'j
•.•
1444)4
4) 0 w.$Qt.)>
V X
o~~~~
4 Q~ Q~
x A'
UI H. 044) zaH>1
I CD) N~'> Vr
0,r k~w
-10C C 0
4.1'a 2 1i , 0
.44
0 W4 11y
., 41
0~1.f 0 ) 41
> P4
u* o c
rq v
4j tjW
(d C 04H9 ..
'44.4
t
0 41
0
41
Ln 4
5-4
(ad
236
J- I I
R.-
The energy-absorbing efficiency, K, is defined as the inverse

ratio of the maximum deceleration exhibited by a real material,
dm, to that for an ideal material of equivalent thickness, dmi,
12
K vii /2hd m (
(37)
Generally, K is expressed as a function of the impact velocity.

At low v , the impact energy is small relative to the stiffness
-of the foam, the degree of penetration is small, and K is low.
r. At high vi, the impact energy is large relative to the stiff-
[ ness, the impacting body "bottoms" on the understructure, and
K is low. At some intermediate vi, K exhibits a maximum.
The optimum material is one for which: (1) the K versus v.
curve is as broad as possible, (2) Kmax is close to unity,
and (3) K occurs at the most probable vi for the particular
applicati1
On the basis of his calculations, Rusch stated the following

conclusions: (1) the energy-absorbing characteristics of a
brittle foam are superior to those of a ductile foam; (2) the
optimum energy-absorbing foam has a large cell size, a narrow
cell size distribution, and a minimum number of reinforcing
membranes between the cells; and (3) foam composites offer no
significant advantage over a single foam.
10.9.4.4 Composite Foam System: Brooks and Rey (Reference 114)
found that a composite could be formed combining the high en-
ergy dissipation of polystyrene beaded foam with the load-dis-
tributing effects of semi-rigid urethane. Simple dynamic tests
consisted of dropping a 6-1/2-in. diameter aluminum hemispher-
ical headform weighing 15 lb at impact velocities up to 30 mi/h
(44 ft/sec). As shown in Figure 89a, the urethane exhibits the
lowest level of headform acceleration during impact. On the
other hand, the polystyrene exhibits the lowest level of pene-
tration as shown in Figure 89b. The urethane can be said to
absorb the least amount of energy, as indicated by the highest
rebound value in Figure 89c. - -
In-small-scale static tests, 2-in. cubes were compressed to

70-percent deflection and then relaxed with an Instron testing
machine at 2.0-in./min crosshead speed. Figure 89d shows the
relative energy absorption of the three materials tested, in-
dicating the composite foam as a compromise between polystyrene
and semi-rigid urethane foam.
114. Brooks, J. D., and Rey, L. G., POLYSTYRENE-URETHANE COM-
POSITE FOAM FOR CRASH PADDING APPLICATION, Limited Publi-
cation, Low Chemical of Canada, Sarnia, Ontario.
237
.. .
___-_____, -- -.-
.. .. - .. . .. --- , ,.4 . .
130
Polystyrene 7 Z
110 Urethane
Compos.te
90
00
70
30 o
35 0- 2 3
i5 20 25 30
Velocity, mi/h i 20 25 30
(a) Head form acceleration Velocity mi/h
(b) Headform penetration
40 - - 60 -- Polystyrene
------- Composite
........ Urethane
35 4S-
Urethane .,/
030 2 30 -
""'00
M 0
S25 --
Composite /.
15 - 4
"20 // ::
- Polystyrene 0 10 30 50 70
Compression, percent
-W (d) Compression load-
15 20 25 30 deflection curves
Velocity mi/h
(c) Headform rebound
Figure 89. Response of composite foam, compared with urethane,

and r-1-styrene. (From Reference 114)
238
ii
SA 10.9.4.5 Specific Energy and Relative Energy-Absorption Ratio:
Reference 115 discusses performance parameters of Dow composite
foam in energy-absorbing applications. It was concluded that,
on the typical response curve for a compression test, where the
area contained within the hysteresis loop shown in Figure 90
is directly related to the energy absorbed, three performance
parameters can be defined: the specific energy absorbed at
maximum strain, the relative energy-absorption ratio, and the
maximum stress. The total energy absorbed at maximum strain is
the sum of areas A and B. When this total energy is expressed
in terms of a unit volume (or unit weight), the quantity be-
comes the specific energy absorption at maximum strain. The
ratio of area A to the sum is the relative energy-absorption
ratio, which is a measure of the amount of energy actually dis-
sipated during compression. In effect, it corrects the per-
formance parameter for the energy that is momentarily stored.
The maximum stress is usually the stress at maximum strain.
Exceptions to this occur wthen some rigid cellular materials are
compressed and a spike is observed during the initial stage of
compression. Maximum stress levels are directly related to the
deceleration that the impacting object sustains.
ra
ý4)
o ~ I
L~n B
Deflection
Figure 90. Typical response of plastic

foam to compression test.
115. PERFORMANCE PARAMETERS OF DOW COMPOSITE FOAM IN ENERGY

ABSORBING APPLICATIONS, Dow Chemical Company, Midland,
Michigan.
239
Melvin and Roberts (Reference 116) measured the specific energy
absorbed and the relative energy-absorption ratio for the ma-
terials listed in Table 17 using three speeds: 20, 2000, and
13,000 in./min. Their results are summarized in Table 18 and
Figure 91, from which they concluded that the majority of foams
do not exhibit marked increases in properties with increasing
test speed. The vinyl foams, which exhibit dramatic increases,
are the exceptions.
TABLE 17. MATERIAL SUMMARY (FROM REFERENCE 116)
Initial strain rate

Density Specimen (sec )
Material and 3 height
code number (lb/ft3) (in.) Speed 1 Speed 2 Speed 3
Polyethylene E-1 2.34 2 0.17 17 100
Polyethylene E-2 6.65 1.5 0.22 22 150
Polyethylene E-3 9.05 2 0.17 17 100
Polystyrene S-1 1.09 2 0.17 17 100
Polystyrene S-2 3.35 2 0.17 17 100
Polystyrene S-3 1.21 2 0.17 17 100

(pelletized)
Polyurethane U-1 1.53 2 0.17 17 100
(rigid)
Vinyl V-1 7.35 1 0.33 33 220
Vinyl V-2 7.25 1 0.33 33 220
Vinyl V-3 5.04 1 0.33 33 220
Cork C-l 11.5 1.5 0.22 22 150
116. Melvin, J. W., and Roberts, V. L., COMPRESSION OF CELLULAR

PLASTICS AT HIGH STRAIN RATES, Journal of Cellular Plas-
tics, March/April 1971, pp. 97-100.
240 "
5>, 0 w ~
I N 0 N I I O0
">~ ON ! 0 co 0i 0 . N N o
G
o0 x
4)
6
Ok c0
00
41 c
""4 "4
.0-- N (N 0 0E9q
04 V). 1 v In
as 00 0 m
0 4 *q
;* *; *
Z~~
>i.4.4 LWN
n 4 " - N V
EC
WCJ ~ 0ý r f.
"M
Nn 0
c(I'
N
v.4
4
N
N
"4"
m
1)r% ~ 0
4
(
I
LA
o >'~-44 -M )
(04
£ a N4 0
Ln - 0.
Lm N r4
( 0 Ln N c
On m N " n 4~(
Cl) $44 p
1> 41 ' N
c!N ' I (N 0. 0 i '. n
04 NN
'.40
J-4 *q
"C4 rq
1> >. EU"4 X
.4.- ~~241.. ..
200
S• zooC-1
100
0 1 50 s-2__
20 00
00
S 10 V 0
U)
X
2 10 0 5-10 0 0 0000 0000 00
CLid
Crsha vlct, n/i
strain versus crosshead velocity.
10.9.4.6 1namic Property Index: Fan (Reference 1171 devel-

oped techniques for simulating the force-penetration properties
of viscoelastic materials based on results of pendulum impact
tests on polyurethane foam. The dynamic force-penetration re-
lationship of polyurethane can be approximated by a function of
three variables: penetration-thickness ratio, sample thickness,
and impact velocity. Fan suggested a criterion for energy ab-
sorption expressed as the dynamic-property index:
I E dd/Gm (38)
117. Fan, W. R. S., A SIMULATION OF THE DYNAMIC PROPERTIES OF

ENERGY-ABSORBING MATERIALS, 1970 International Automobile
Safety Conference Compendium, Society of Automotive Engi-
neers, Inc., New York, 1970, pp. 1075-1083.
242
I . ..... ... ... ..... .

where I - dynamic-property index of the material.
Ed = the amount of energy dissipation by the foam
material during impact.
GmGma during
the maximum deceleration measured at the impactor
impact.
A high index value implies a high degree of effectiveness. The

dynamic-property index of a material varies with the test con-
ditions. The material rated as the most effective in a cer-
tain case is not necessarily the most effective material in
other cases.
10.9.4.7 Dynamic Crushing Pressure: Furio and Gilbert (Ref-
erence 118) c~nducted a series of drop tests with low den-
sity (2 lb/ft ) urethane foam using a flat impactor weighing
729 lb at a drop height of 45 ft.
The dynamic crushing pressure, Pcr, which is the product of
the weight of the impact mass and the acceleration divided by
the impact area, is shown in Figure 92 as a function of temper-
ature and velocity for two samples of identical dimensions.
The increase in pressure with velocity is attributed to the
fact that the entrapped gas must escape in order for the foam
to collapse. Under dynamic loading, the gas cannot escape fast
enough, and a higher pressure results.
10.9.5 Application of Padding Material
In the absence cf data for extremity impacts, it is assumed
that padding material that is suitable for head impact protec-
tion will be suitable also for protecting extremities. Extrem-
ity impacts are not likely to have the potentially severe ef-
fects of head impacts. It is suggested that areas within the
extremity strike envelope having radii of 2 in. or less be pad-
ded and that such padding have a minimum thickness of 0.75 in.
Caution must be exercised in padding sharp edges and corners.
Padding installed in a manner that allows it to-be broken away
from the corner or cut through by sharp edges offers no protec-
tion. It is recommended that edges and corners to be padded
have a minimum radius of 0.5 in. prior to padding. A definite
volume of the padding must be crushed to absorb the initial
kinetic energy of the head and protective helmet.
118. Furio, A. J., Jr., and Gilbert, W. E., IMPACT TESTS OF

URETHANE FOAM, Report No. NSRDC 4254, Naval Ship Research
and Development Center, Bethesda, Maryland, January 1974,
AD 775903.
243
r
S60
24 x 24 x 12 in.
Ambient
I
1404// 24 x 24 x 12 in.
Cold (-100F)
• • ~30...
S 20....-
>11
UI
0•
10
0
I
10 20 30 40
Impact velccity, ft/sec
Figure 92. Dynamic crushing pressure versus

impact velocity for tests at two
temperatures.
10.9.6 Ductile Materials

In cases where the use of padding material is impractical or
the thickness allowed is inadequate to provide the necessary
protection, ductile energy-absorbing materials or frangible
breakaway panels should be used where possible. Window and
door
shouldframes, controlwith
be designed columns,
large electrical
radii (I in.junction
or more)boxes,
ratheretc.,•
than
with sharp edges and corners. -
Swearingen concluded in Reference 119 that at impact velocities

of 30 ft/sec against rigid structure padded with materials even
6-in. thick, unconsciousness, concussion, and/or fatal head
injuries will be produced. Where possible, a combination of
119. Swearingen, J. J., EVALUATIONS OF VARIOUS PADDING MATER-
IALS FOR CRASH PROTECTION, FAA Technical Report AM. 66-40,
Federal Aviation Administration, Civil Aeromedical Insti-
tute, Washington, D. C., December 1966, AD 647048.
244
4-!
deformable structure and padding material should be considered
to absorb the impact energy and to adequately distribute the
forces over the face. Surfaces to which this combination
shuuld be applied are instrument panels, seat backs, bulkheads,
and any other structure the head may impact during the crash
sequence.
I
A
Ii
245 1.
.. . - . - .- . ... .
REFERENCES
1. Military Standard, MIL-STD-1290(AV), LIGHT FIXED- AND

ROTARY-WING AIRCRAFT CRASHWORTHINESS, Department of De-
fense, Washington, D. C., 25 January 1974.
2. ENGINEERING ANALYSIS OF CRASH INJURY IN ARMY Oh .8 AIR-

CRAFT, USASC Technical Report, U. S. Army Safety Center,
Fort Rucker, Alabama, to be published.
3. ENGINEERING ANALYSIS OF CRASH INJURY IN ARMY CH-47 AIR-

for Aviation Safety, Fort Rucker, Alabama, June 1978.
4. ENGINEERING ANALYSIS OF CRASH INJURY IN ARMY AH-l AIR-

for Aviation Safety, Fort Rucker, Alabama, March 1978.
5. Carnell, B. L., CRASHWORTHINESS DESIGN FEATURES FOR AD-

VANCED UTILITY HELICOPTERS, in Aircraft Crashworthiness,
K. Saczalski, et al., eds., University Press of Virginia,
Charlottesville, Virginia, 1975, pp. 51-64.
6. Bainbridge, M. E., Reilly, M. J., and Gonsalves, J. E.,

CRASHWORTHINESS OF THE BOEING VERTOL UTTAS, in Aircraft
Crashworthiness, K. Saczalski, et al., eds., University
Press of Virg-iia, Charlottesville, Virginia, 1975,
pp. 65-82.
7. Rich, M. J., INVESTIGATION OF ADVANCED HELICOPTER STRUC-

TURAL DESIGNS, Volume I, ADVANCED STRUCTURAL COMPONENT DE-
SIGN CONCEPT STUDY, Sikorsky Aircraft, Division of United
Technology Corporation; USAAMRDL Technical Report 75-59A,
Development Laboratory, Fort Eustis, Virginia, May 1976,
AD A026246.
8. Hoffstedt, D. J., and Swatton, S., ADVANCED HELICOPTER

STRUCTURAL DESIGN INVESTIGATION, The Boeing Vertol Com-
pany; USAAMRDL Technical-Report 75-56A,-Eustis DirectOrate,
tory, Fort Eustis, Virginia, March 1976, AD A024662.
9. Hicks, J. E., AN ANALYSIS OF LIFECYCLE ACCIDENT COSTS FOR

THE ADVANCED SCOUT HELICOPTER, U. S. Army Agency for Avia-
tion Safety, Fort Rucker, Alabama, January 1977.
10. McDermott, J. M., and Vega, E., THE EFFECTS OF LATEST

MILITARY CRITERIA ON THE STRUCTURAL WEIGHT OF THE HUGHES
ADVANCED ATTACK HELICOPTER YAH-64, Journal of the American
Helicopter Society, Vol. 23, No. 4, October 1978, pp. 2-9.
246
I.
REFERENCES (Continued)
11. Haley, J. L., Jr., CRASHWORTHINESS VERSUS COST: A STUDY
OF ARMY ROTARY WING AIRCRAFT ACCIDENTS IN PERIOD JANUARY
1970 THROUGH DECEMBER 1971, paper presented at the Air-
craft Crashworthiness Symposium, University of Cincinnati,
Cincinnati, Ohio, October 1975.
i 12. Hicks, J. E., ECONOMIC BENEFITS OF UTILITY AIRCRAFT CRASH-
WORTHINESS, USAAAVS Technical Report 75-2, U. S. Army
Agency for Aviation Safety, Fort Rucker, Alabama, July
1976.
13. THE ECONOMIC BENEFITS OF CRASHWORTHINESS AND FLIGHT SAFETY
DESIGN FEATURES IN ATTACK HELICOPTERS, USAAAVS Technical
Report 77-2, U. S. Army Agency for Aviation Safety, Fort
Rucker, Alabama, June 1977.
14. Military Specification, MIL-S-58095(AV), SEAT SYSTEM:

CRASHWORTHY, NON-EJECTION, AIRCREW, GENERAL SPECIFICATION
FOR, Department of Defense, Washington, D. C., 27 Aug-ust
1971.
15. Cook, R. L., and Goebel, D. E., EVALUATION OF THE UH-1D/H

HELICOPTER CRASHWORTHY FUEL SYSTEM IN A CRASH ENVIRONMENT,
Dynamic Science, Division of Marshall Industries; USAAMRDL
Technical Report 71-47, U. S. Army Air Mobility Research
and Development Laboratory, Fort Eustis, Virginia, Novem-
ber 1971.
16. Domzalski, L. P., et al., U. S. NAVY DEVELOPMENTS IN

CRASHWORTHY SEATING, Naval Air Development Center; Pro-
ceedings 1978 SAFE Symposium, Survival and Flight Equip-
ment Association, Canoga Park, California, October 1978.
1i. Gell, C. F., TABLE OF EQUIVALENTS FOR ACCELERATION TERMI-
NOLOGY, Aerospace Medicine, Vol. 32, No. 12, December
1961, pp. 1109-1111.
18. Military Standard, MIL-STD-1333, AIRCREW STATION GEOMETRY
SFOR MILITARYLAIRCRAFT, Department of Defense, Washington,
D. C. -
19. Military Standard, MIL-STD-850, AIRCREW STATION VISION

FORC. MILITARY AIRCRAFT, Department of Defense,
REQUIREMENTS D.
Washington,
20. Haley, J. L., ANALYSIS OF EXISTING HELICOPTER STRUCTURES

TO DETERMINE DIRECT IMPACT SURVIVAL PROBLEMS, U. S. Army
Board for Aviation Accident Research, Fort Rucker, Ala-
bama, 1971.
247
21. Singley, G. T., III, and Desjardins, S. P., CRASHWORTHY

HELICOPTER SEATS AND OCCUPANT RESTRAINT SYSTEMS, in Opera-
tional Helicopter Aviation Medicine, AGARD Conference Pro-
ceedings No. 255, North Atlantic Treaty Organization, Ad-
visory Group for Aerospace Research and Development,
Neuilly sur Seine, France, May 1978.
.2. Weinberg, L. W. T., AIRCRAFT LITTER RETENTION SYSTEM DE-

SIGN CRITERIA, Aviation Crash Injury Research (AvCIR), Di-
vision of rlight Safety Foundation, Inc.; USAAVLABS Tech-
nical Report 66-27, U. S. Army Aviation Materiel Labora-
tories, Fort Eustis, Virginia, April 1966, AD 632457.
23. Military Handbook, MIL-HDBK-5, METALLIC MATERIALS AND

ELEMENTS FOR AEROSPACE VEHICLE STRUCTURES, Department of
Defense, Washington, D. C.
24. Desjardins, S. P., and Harri-on, H., THE DESIGN, FABRICA-

TION, AND TESTING OF AN INTEGRALLY ARMORED CRASHWORTHY
CREWSEAT, Dynamic Science, Division of Marshall Indus-
tries; USAAMRDL Technical Report 71-54, Eustis Directorate,
tory, Fort Eustis, Virginia, January 1972, AD 742733.
STRAINT SYSTEMS STUDY - HC-lB VERTOL CHINOOK, AvCIR 62-26,
Aviation Crash Injury Research (AvCIR), Division of Flight
Safety Foundation, Inc., Phoenix, Arizona, November 1962.
26. Haley, J. L., Jr., and Avery J. P., Ph.D., PERSONAL RE-
STRAINT SYSTEMS STUDY - HU-1A AND HU-IB BELL IROQUOIS,
AvCIR 62-27, Aviation Crash Injury Research (AvCIR), Divi-
sion of Flight Safety Foundation, Inc., Phoenix, Arizona,
December 1962.
STRAINT SYSTEM STUDY - CV-2 DE HAVILLAND CARIBOU, AvCIR
62-16, Aviation Crash Injury Research (AvCIR), Division of
Flight Safety Foundation, Inc., Phoenix, Arizona, April
- 1964.
28. Beedle, L., PLASTIC DESIGN OF STEEL FRAMES, John Wiley

and Sons, New York, 1958.
29. Hodge,. P. G., Jr., PLASTIC ANALYSIS OF STRUCTURES, McGraw-
Hill Book Company, New York, 1959.
248
i -
IIIIII I 'III ' -•- -* r - " i--
30. Turnbow, J. W., et al., AIRCRAFT PASSENGER-SEAT-SYSTEM
RESPONSE TO IMPULSIVE LOADS, Aviation Safety Engineering
and Research (AvSER), Division of Flight Safety Founda-
tion, Inc.; USAAVLABS Technical Report 67-17, U- S. Army
August 1967, AD 661088.
31. Reilly, M. J., CRASHWORTHY, TROOP SEAT TESTING PROGRAM,
The Boeing Vertol Companyl USAAMRDL Technical Report 77-13,
Development Laboratory, Fort Eustis, Virginia, August 1977,
AD A048975.
32. Reilly, M. J., CRASHWORTHY, HELICOPTER GUNNER SEAT TESTING
PROGRAM, The Boeing Vertol Company; USARTL Technical Re-
port 78-7, U. S. Army Research and Technology Laboratories,
Fort Eustis, Virginia, February 1978, AD A054970.
33. Mazelsky, B., A CRASHWORTHY ARMORED PILOT SEAT FOR HELI-
COPTERS, ARA Inc.; USAAVSCOM Technical Report 73-34 and
Report No. NADC-74018-40, Joint Report issued by Naval Air
Systems Command, Washington, D. C., and U. S. Army Avia-
tion Systems Command, St. Louis, Missouri, January 1974.
AD A007551.
34. Domzalski, L. P., and Singley, G. T., III, JOINT ARMY/NAVY
TEST PROGRAM FOR UTTAS SEATING SYSTEMS, NADC-79229-60,
Naval Air Development Center, Warminster, Pennsylvania, to
be published.
35. Dummer, R. J., QUALIFICATION TEST REPORT 613-1787 COOL-
QUALIFICATION TESTING OF ARMORED CRASHWORTHY AIRCREW SEAT
RA-30525-1 (FOR SIKORSKY AIRCRAFT CONTRACT 576344), U. S.
Army Contract No. DAAJ01-77-C-0001, Norton Company, Indus-
trial Ceramics Division, Worcester, Massachusetts, revised
January 1979.
36. Carr, R. W., and Phillips, N. S., DEFINITION OF DESIGN
CRITERIA FOR ENERGY ABSORPTION SYSTEMS, Beta Industries
Incorporated; Report No. NADC-AC-7007, Naval Air Develop-
ment Center, Warminster, Pennsylvania, 11 June 1970,
-AD 871040.
37. Bacchetti, A. C., and Maltha, J., MADYMO - A GENERAL PUR-
POSE MATHEMATICAL DYNAMICAL MODEL FOR CRASH VICTIM SIMULA-
TION, Report No. 753012-C, Instituut voor Wegtransportmid-
delen, Netherlands 1978.
38. Bartz, J. L., DEVELOPMENT AND VALIDATION OF A COMPUTER
SIMULATION OF A CRASH VICTIM IN THREE DIMENSIONS, Pro-
ceedings, Sixteenth Stapp Car Crash Conference, Socet'
of Automotive Engineers, Inc., New York, 1972, pp. 105-127.
249
* I . ... . - -,- - - - -• .* .-
39. Danforth, J. P., and Randall, C. D., MODIFIED ROS OCCUPANT
DYNAMICS SIMULATION USER MANUAL, Publication No. GMR-1254,
General Motors Corporation Research Laboratory, Warren,
Michigan, 1972.
40. Fleck, J. T., Butler, F. E., and Vogel, S. L., AN IMPROVED
THREE DIMENSIONAL COMPUTER SIMULATION OF MOTOR VEHICLE
CRASH VICTIMS, Final Technical Report No. ZO-5180-L-1 (in
four volumes), Calspan Corporation, Buffalo, New York,
1974.
41. Furosho, H., Yokoya, K., and Fujiki, S., ANALYSIS OF OCCU-
PANT MOVEMENTS IN REAR-END COLLISION, Paper No. 13, in
Safety Research Tour in the U.S.A. from the Viewpoint of
Vehicle Dynamics, 1969.
42. Furosho, H., and Yokoya, K., ANALYSIS OF OCCUPANT'S MOVE-
MENT IN HEAD-ON COLLISION, Transactions of the Society of
Automotive Engineers of-Japan, No. 1, Tokyo, Japan, 1970,
pp. 1=45-55.
43. Glancy, J. J., and Larsen, S. E., USERS GUIDE FOR PROGRAM
SIMULA, Report TDR No. 72-23, Dynamic Science, Division of
Ultrasystems, Inc., Phoenix, Arizona, 1972.
44. Huston, R. L., Hessel, R., and Passerello, C., A THREE-
DIMENSIONAL VEHICLE-MAN MODEL FOR COLLISION AND HIGH
ACCELERATION STUDIES, Paper No. 740275, presented at Auto-
mobile Engineering Conference, Society of Automotive Er-
gineers, Inc., Detroit, Michigan, 25 February - 1 March
1974.
45. McHenry, R. R., ANALYSIS OF THE DYNAMICS OF AUTOMOBILE
PASSENGER-RESTRAINT SYSTEMS, Proceedings, Seventh Stapp
Car Conference, Society of Automotive Engineers, Inc.,
Now York, 1963, pp. 207-249.
46. Robbins, D. H., THREE-DIMEN9IONAL SIMULATION OF ADVANCED
AUTOMOTIVE RESTRAINT SYSTEMS, Paper No. 700421, In 1970
International Automotive Safety Conference Compendium--
P-30, Society of Automotive Engineers, Inc., New York,
1970.
47. Robbins, D. H., Bennett, R. 0., Jr., and Bowman, B. M.,

USER-ORIENTED MATHEMATICAL CRASH VICTIM SIMULATOR, Pro-
ceedings, Sixteenth Stapp Car Crash Conference, Society of
250
48. Robbins, D. H., Bennetti R. 0., and Roberts, R. L., HSRI

TWO-DIMENSIONAL CRASH VICTIM SIMULATOR: ANALYSIS, VERI-
FICATION, AND USER'S MANUAL, Final Report, Report No.
HSRI-Bio-M-70-8, Highway Safety Research Institute, Uni-
versity of Michigan, Ann Arbor, Michigan, 1970.
49. Robbins, D. H., Bowman, B. M., and Bennett, R. 0., THE

MVMA TWO-DIMENSIONAL CRASH VICTIM SIMULATIONS, Proceed-
ings, Eighteenth Stapp Car Crash Conference, Society oTf
50. Segal, D. J., REVISED COMPUTER SIMULATION OF THE AUTOMO-

BILE CRASH VICTIM, Report No. VJ-2759-V-2, Cornell Aero-
nautical Laboratory, Inc., Buffalo, New York, 1971.
51. Segal, D. J., and McHenry, R. R., COMPUTER SIMULATION OF

AUTOMOBILE C11 A VICTIM - REVISION, Report No. VJ-2492-1,
Cornell Aeronautical Laboratory, Inc., Buffalo, New York,
1967.
52. Twigg, D. W., and Karnes, R. N., PROMETHEUS, A USER-

ORIENTED PROGRAM FOR HUMAN CRASH DYNAMICS, Boeing Computer
Services, Report No. BCS 40038, Department of the Navy,
Office of Naval Research, Washington, D. C., 1974.
53. Young, R. D., THREE-DIMENSIONAL MATHEMATICAL MODEL OF AN

AUTOMOBILE PASSENGER, Research Report 140-2, Texas Trans-
portation Institute, College Station, Texas, 1970.
54. Young, R. D., Ross, H. E., and Lammert, W. F., SIMULATION

OF THE PEDESTRIAN DURING VEHICLE IMPACT, Paper No. 27,
Proceedings, Third International Congress on Automotive
Safety, Vol. II, ;ociety of Automotive Engineers, Inc.,
New York, 1974.
55. Kroeger, W. J., INTERNAL VIBRATIONS EXCITED IN THE OPERA-

TION OF PERSONNEL EMERGENCY ESCAPE CATAPULTS, Memorandum
Report 340, Frankfort Arsenal Laboratory Division, Phila-
delphia, Pennsylvania- 1946. -
56. Latham, W. F., A STUDY IN BODY BALLISTICS: SEAT EJEC-

TION, Proceedings of Royal Society, London, England, 1957,
B 147s 121-139.
57. Stech, E. L., and Payne, P. R., DYNAMIC MODELS OF THE HU-
MAN BODY, Frost Engineering Development Corp., AMRL Tech-
nical Report 66-157, Aerospace Medical Research Laboratory,
Wright-Patterson Airforce Base, Ohio, 1969.
) 251
*4
. -:
. .. . .. . . .. .. . . .. .
I
4
58. Laananen, D. H., DEVELOPMENT OF A SCIENTIFIC BASIS FOR
ANALYSIS OF AIRCRAFT SEATING SYSTEMS, Dynamic Science,
Division of Ultrasystems, Inc.; Report No. FAA-RD-74-130,
U. S. Department of Transportation, Federal Aviation Ad-
ministration, Washington, D. C., 1975, AD A004306.
I
59. Chandler, R. F., and Laananen, D. H., SEAT/OCCUANT CRASH
DYNAMIC ANALYSIS VALIDATION TEST PROGRAM, Pape. go. 790590,
presented at Business Aircraft Meeting, Society of Automo-
tive Engineers, Inc., Wichita, Kansas, April 1979. j
60. Belytschko, T., Schirver, L., and Schultz, A., A MODEL
FOR ANALYTIC INVESTIGATION OF THREE-DIMENSIONAL HEAD-
SPINE DYNAMICS - FINAL REPORT, University of Illinois at
Chicago Circle; AMRL Technical Report 76-10, Aerospace
Medical Research Laboratory, Wright-Patterson Air Force
Base, Ohio, April 1976, AD A025911.
61. Auyer, W., and Turnbow, J., A STUDY OF THE DYNAMIC RE-
SPONSE OF A DAMPED, MULTI-DEGREE OF FREEDOM, SPRING-MASS
SYSTEM WHICH SIMULATES A SEAT, SEAT CUSHION, AND SEAT OC-
CUPANT SUBJECTED TO A VERTICAL IMPACT ACCELERATION, Avia-
tion Safety Engineering and Resaarch (AvSER), Division of
Flight Safety Foundation, Inc. (unpublished report).
62. Ezra, A., and Pay, R. J., AN ASSESSMENT OF ENERGY ABSORB- I
ING DEVICES FOR PROSPECTIVE USE IN AIRCRAFT IMPACT SITUA-
TIONS, in Dynamic Response of Structures, G. Herrmann and
N. Perrone, eds., Pergammon Press, Elmsford, New York,
1972, pp. 225-246.
63. Reilly, M. J., CRASHWORTHY TROOP SEAT INVESTIGATION, The

Boeing Vertol Companyl USAAMRDL Technical Report 74-93,
Development Laboratory, Fort Eustis, Virginia, December
1974, AD/A-007090.
64. Kroell, C. K., A SIMPLE, EFFICIENT, ONE SHOT ENERGY AB-
SORBER, Reprint from Bulletin No. 30, Shock, Vibration,
and Associated Environments, Part III, General Motors Re-
search Laboratory, Warren, Michigan, February 1962.
65. Guist, L. R., and Marble, D. P., PREDICTION OF THE INVER-
SION LOAD OF A CIRCULAR TUBE, NASA Technical Note D-3622,
Ames Research Center, Moffett Field, California, June 16,
1966.
252
-- -- -- - - --.-
o.
66. Haley, J. L., Klemme, R. E., and Turnbow, J. W., TEST AND
EVALUATION OF 1000-4000 POUND LOAD-LIMITING DEVICES, Dy-
namic Science, AvSER Facility Report M69-2, for U.S. Army
February 1969.
67. Rich, M. J., VULNERABILITY AND CRASHWORTHINESS IN THE DE-
SIGN OF ROTARY-WING VEHICLE STRUCTURES, Paper No. 680673,
presented at Aeronautic and Space Engineering and Manufac-
turing Meeting at Los Angeles, Society of Automotive En-
gineers, Inc., New York, October 1968.
68. Bendix Products Aerospace Division, ENERGY ABSORBING CHAR-

ACTERISTICS OF CRUSHABLE ALUMINUM STRUCTURES IN A SPACE
ENVIRONMENT, Report No. SPP-65-107 (NASA-CR-65096), pre-
pared for NASA Manned Spacecraft Center, Houston, Texas,
July 1965.
69. McGehee, J. R., A PRELIMINARY EXPERIMENTAL INVESTIGATION
OF AN ENERGY-ABSORPTION PROCESS EMPLOYING FRANGIBLE METAL
TUBING, NASA Technical Note D-1477, National Aeronautics
and Space Administration, Washington, D. C., October 1962.
70. Schwartz, M., DYNAMIC TESTING OF ENERGY-ATTENUATING DE-

VICES, NADC Report No. AC-6905, Naval Air Development
Center, Warminster, Pennsylvania, October 1969.
71. Singley, G. T., III, FULL SCALE CRASH TESTING OF A CH-47C
HELICOPTER, paper presented at 32nd Annual National V/STOL
Forum, American Helicopter Society, Washington, D. C.,
May 1976.
72. Desjardins, S. P., et al., CRASHWORTHY ARMORED CREWSEAT
FOR THE UH-60A BLACK HAWK, paper presented at 35th Annual
National Forum, American Helicopter Society, Washington,
D. C., May 1979.
73. Mazelsky, B., INVESTIGATION OF AN ALUMINUM ROLLING HELIX

CRASH ENERGY ABSORBER, ARA, Inc.; USAAMRDL Technical Re-
port 77-8, Eustis Directorate, U. S. Army Air Mobility
ginia, May 1977, AD A042084.
74. Woodward, C. C., et al., INVESTIGATION, DESIGN AND DEVEL-

OPMENT OF AN P7U-3 EJECTION SEAT ENERGY ABSORPTION SYSTEM
FOR REDUCTION OF CRASH FORCE LOADS, NADC Report ACEL-335,
Naval Air Development Center, Warminster, Pennsylvania,
June 1957. 253
............................... ,V
*. ..
rI
75. Turnbow, J. W., Robertson, S. H., and Carroll, D. F., DY-

NAMIC TEST OF AN EXPERIMENTAL TROOP SEAT INSTALLATION
IN AN H-21 HELICOPTER, Aviation Crash Injury Research
(AvCIR), Division of Flight Safety Foundation, Inc.1
TRECOM Technical Report 63-62, U. S. Army Aviation Mater-
iel Laboratories, Fort Eustis, Virginia, November 1963.
76. DYNAMIC TEST OF CRUSHABLE SEAT CUSHIONS, AvSER Report

M67-6, Aviation Safety Engineering and Research (AvSER),
Division of Flight Safety Foundation, Inc., Phoenix, Ari-
zona, August 1967.
77. Carr, R. W., and Desjardins, S. P., AIRCREW RESTRAINT SYS-

TEM - DESIGN CRITERIA EVALUATION, Dynamic Science, Divi-
sion of Ultrasystems, Inc.; USAAMRDL Technical Report
75-2, Eustis Directorate, U. S. Army Air Mobility Research
and Development Laboratory, Fort Eustis, Virginia, Febru-
ary 1975, AD A009059.
78. Carr, R. W., HELICOPTER TROOP/PASSENGER RESTRAINT SYSTEMS

DESIGN CRITERIA EVALUATION, Dynamic Science, Division of
Ultrasystems, Inc.; USAAMRDL Technical Report 75-10,
Development Laboratory, Fort Eustis, Virginia, June 1975,
AD A012270.
79. Reilly, M. J., CRASHWORTHY HELICOPTER GUNNER'S SEAT IN-
VESTIGATION, The Boeing Vertol Company; USAAMRDL Techni-
cal Report 74-98, Eustis Directorate, U. S. Army Air Mo-
bility Research and Development Laboratory, Fort Eustis,
Virginia, January 1975, AD A005563.
80. Schulman, M., and McElhenney, J., INFLATABLE BODY AND HEAD
RESTRAINT, NADC-77176-40, Naval Air Systems Command, De-
partment of the Navy, Washington, D. C., September 1977,
AD A046477.
81. Singley, G. T., III, TEST AND 7 EVALUATION OF IMPROVED AIR-

CRAFT RESTRAINT SYSTEMS FOR COMBAT HELICOPTERS, Paper No.
A.18, presented at NATO/AGARD Aerospace Medical Panel,
Aerospace Specialist's Meeting on Aircrew and Survivabil-
ity, North Atlantic Treaty Organization, Bodo, Norway,
May 20-23, 1980.
82. Roberts, V. L., and Robbins, D. H., MULTIDIMENSIONAL

MATHEMATICAL MODELING OF OCCUPANT DYNAMICS UNDER CRASH
CONDITIONS, Paper No. 690248, Society of Automotive En-
gineers, Inc., New York, January 1969.
254
83. Kourouklis, G., Glancy, J. L., and Desjardins, S. P., THE

DESIGN, DEVELOPMENT, AND TESTING OF AN AIRCRAFT RESTRAINT
SYSTEM FOR ARMY AIRCRAFT, Dynamic Science, Division of Ul-
trasystems, Inc.; USAAMRDL Technical Report 72-26, Eustis
Directorate, U. S. Army Air Mobility Research and Develop-
ment Laboratory, Fort Eustis, Virginia, June 1972, AD746631.
84. Farris, L., HIGH STRENGTH STITCHING FOR AIRCRAFT PERSONNEL

RESTRAINT SYSTEMS, Pacific Scientific Co.; Proceedings,
1978 SAFE SymPosium, Survival and Flight Equipment Asso-
85. Proposed Draft Military Specification,

RESTRAINT SYSTEM, AIRCREW, September 1974.
86. Reilly, M. J., ENERGY ATTENUATING TROOP SEAT DEVELOPMENT,

The Boeing Vertol Company; Report NADC-AC-7105 with Adden-
dum NADC-73121-40, U. S. Naval Air Development Center,
Aerospace Crew Equipment Department, Warminster, Pennsyl-
vania, May 1971.
87. Military Specification, MIL-R-8236, REEL, SHOULDER HAR-

NESS, INERTIA LOCK, Department of Defense, Washington,
D. C.
88. Churrhill, E., et al., ANTHROPOMETRY OF U. S. ARMY AVIA-

TORS 1970, Anthropology Research Project; USANL Technical
Report 72-52-CE, U. S. Army Natick Laboratories, Natick,
Massachusetts, December 1971, AD 743528.
89. U. S. Army, THE BODY SIZE OF SOLDIERS - U. S. ARMY ANTHRO-

POMETRY - 1966, USANL Technical Report 72-51-CE, U. S.
Army Natick Laboratories, Natick, Massachusetts, December
1971, AD 743465.
90. U. S. Code of Federal Regulations, Title 49, Chapter 5,

Part 572: ANTHROPOMORPHIC TEST DUMMY, Government Printing
Office, Washington, D. C., (Rev.) 1978.
91. SAE Recommended Practice, SAE J211b, INSTRUMENTATION FOR

IMPACT TESTS, SAE Handbook 1979, Part 2, Society of Auto-
motive Engineers, Inc., Warrendale, Pennsylvania, 1979,
pp. 34.117-34.118.
92. Military Specification, MIL-S-9479, SEAT SYSTEM, UPWARD

EJECTION, AIRCRAFT, GENERAL SPECIFICATION FOR, Department
of Defense, Washington, D. C.
255
93. Military Specification, MIL-A-8865, AIRPLANE STRENGTH AND

RIGIDITY MISCELLANEOUS LOADS, Department of Defense, Wash-
ington, D. C.
94. Haley, J. L., Jr., et al., HELMET DESIGN CRITERIA FOR IM-
PROVED CRASH SURVIVAL, Aviation Safety Engineering and Re-
search (AvSER), Division of Flight Safety Foundation,
Inc.; USAAVLABS Technical Report 65-44, U. S. Army Avia-
tion Materiel Laboratories, Fort Eustis, Virginia, Jan-
uary 1966, AD 628678.
95. Patrick, L. M., Lissner, H. R., and Gurdjian, E. S., SUR-

VIVAL BY DESIGN - HEAD PROTECTION, Proceedings, Seventh
Stapp Car Crash Conference, Society of Automotive Engi-
neers, Inc., New York, 1963.
96. SAE Recommended Practice, SAE J921b, MOTOR VEHICLE INSTRU-

MENT PANEL LABORATORY IMPACT TEST PROCEDURE - HEAD AREA,
SAE Handbook, 1979, Part 2, Society of Automotive Engi-
neers, Inc., Warrendale, Pennsylvania, 1979, pp. 34.133-
34.134.
97. Fox, R., Kawa, M., and Sharp, E., DESIGNING CRASHWORTHI-
NESS INTO THE YAH-63, paper presented at the Aircraft
Crashworthiness Symposium, University of Cincinnati, Cin-
cinnati, Ohio, October 1975.
98. Lee, W. M., and Williams, B. M., CUSHIONING AND LOAD DIS-
TRIBUTION PERFORMANCE OF PLASTIC FOAMS, Paper No. 700453,
Society of Automotive Engineers, Inc., New York, May 1970.
99. Mattingly, T. E., et al., INVESTIGATION OF VIBRATION AND

IMPACT PROTECTION OF THE HUMAN HEAD AND NECK, Northrop
Corporate Laboratories; AMRL Technical Report 69-112, Air
Force Systems Command, Wright-Patterson Air Force Base,
Ohio, December 1969, AD 702124.
100. Frados, J., PLASTICS ENGINEERING HANDBOOK OF THE- SOCIETY

OF THE PLASTICS INDUSTRY, INC., Van Nostrand Reinhold Co.,
New York, 1976, pp. 499-567.
101. ENSOLITE, Publication ASP 9997, Expanded Products Depart-

ment, Uniroyal, Inc., Mishawaka, Indiana, 1977.
102. TEMPER FOAM, Form TF-20, Edmont-Wilson, Division of Bec-

ton, Dickenson and Co., Coshocton, Ohio, 1975.
256
I
103. Sundquist, D. J., POLYOLEFIN FOAMS, Monographs on Plas-

I tics, Vol. 1, Part 1, 1972, pp. 193-295.
104. Lukens, R. P., et al., 1977 ANNUAL BOOK OF ASTM STANDARDS,

American Society for Testing and Materials, Easton, Mary-
land, 1977, Parts 20, 38, 48.
105. ENCYCLOPEDIA OF POLYMER SCIENCE AND TECHNOLOGY - PLASTICS,

RESINS, RUBBER, FIBERS, John Wiley and Sons, Inc., New
York, 1965, Vol. 3, pp. 98-126.
106. SAE Recommended Practice, SAE J815, LOAD DEFLECTION TEST-

ING OF URETHANE FOAMS FOR AUTOMOTIVE SEATING, SAE Handbook
1979, Part 2, Society of Automotive Engineers, Inc., War-
rendale, Pennsylvania, 1979, p. 34.31.
107. SAE Recommended Practice, SAE J388, DYNAMIC FLEX FATIGUE

TEST FOR SLAB POLYURETHANE FOAM, SAE Handbook, 1979,
Part 2, Society of Automotive Engineers, Inc., Warrendale,
Pennsylvania, 1979, pp. 34.28-34.30.
108. PACKAGING WITH ETHAFOAM, Publication No. 172-221-IOM-767,

Dow Chemical Company, Midland, Michigan, revised 1966.
109. ENGINEERING DESIGN HANDBOOK, DESIGN FOR AIR TRANSPORT AND

AIRDROP OF MATERIAL, AMC Pamphlet No. 706-130, U. S. Army
Materiel Command, Washington, D. C., December 1967,
AD 830262.
110. Daniel, R. P., A BIO-ENGINEERING APPROACH TO CRASH PAD-

DING, Paper No. 680001, Society of Automotive Engineers,
Inc., New York, 1968.
111. SAE Information Report, SAE J885, HUMAN TOLERANCE TO IM-
PACT CONDITIONS AS RELATED TO MOTOR VEHICLE DESIGN, SAE
Handbook 1979, Part 2, Society of Automotive aEngineers,
Inc., Warrendale, Pennsylvania, '1979, pp. 34.114-34.117.
112. U. S. Code of Federal Regulations, Title 49, Part 571:
MOTOR VEHICLE SAFETY STANDARDS, 201, OCCUPANT PROTECTION
AND INTERIOR IMPACT, Government Printing Office, Washing-
ton, D. C., (Rev.) 1978.
113. Rusch, K. C., IMPACT ENERGY ABSORPTION BY FOAMED POLY-

MERS, Journal of Cellular Plastics, Vol. 7, No. 2, 1971,
pp. 78-83.
257
114. Brooks, J. D., and Rey, L. G., POLYSTYRENE-URETHANE COM-

POSITE FOAM FOR CRASH PADDING APPLICATION, Limited Publi-
cation, Dow Chemical of Canada, Sarnia, Ontario.
115. PERFORMANCE PARAMETERS OF DOW COMPOSITE FOAM IN ENERGY
ABSORBING APPLICATIONS, Dow Chemical Company, Midland,
Michigan.
116. Melvin, J. W., and Roberts, V. L., COMPRESSION OF CELLU-
LAR PLASTICS AT HIGH STRAIN RATES, Journal of Cellular
Plastics, March/April 1971, pp. 97-100.
117. Fan, W. R. S., A SIMULATION OF THE DYNAMIC PROPERTIES OF
ENERGY-ABSORBING MATERIALS, 1970 International Automobile
Safety Conference Compendium, Society of Automotive Engi-
neers, Inc., New York, 1970, pp. 1075-1083.
118. Furio, A. J., Jr., and Gilbert, W. E., IMPACT TESTS OF

URETHANE FOAM, Report No. NSRDC 4254, Naval Ship Research
and Development Center, Bethesda, Maryland, January 1974,
AD 775903. -
119. Swearingen, J. J., EVALUATIONS OF VARIOUS PADDING MATER-

IALS FOR CRASH PROTECTION, FAA Technical Report AM 66-40,
Federal Aviation Administration, Civil Aeromedical Insti-
tute, Washington, D.C., December 1966, AD 647048.
2.5
258
S . A .
-
BIBLIOGRAPHY
Adomeit, D., SEAT DESIGN - A SIGNIFICANT FACTOR FOR SAFETY BELT
EFFECTIVENESS, SAE Paper 791004, Proceedings, Twenty-Third
Stapp Car Crash Conference, Society of Automotive Engineers,
Inc., Warrendale, Pennsylvania, 1979.
Alexander, G. H., Conrad, R. E., and Neale, M. R., DETERMINA-

TION OF THE TRADEOFFS BETWEEN SAFETY, WEIGHT, AND COST (,F POS-
SIBLE IMPROVEMENTS TO VEHICLES STRUCTURE AND RESTRAINT03,
Battelle Columbus Laboratories; Report No. PB 238325, National
Highway Traffic Safety Administration, Department of Transror-
tation, Washington, D. C., December 1974.
Anderson, L. R., Grimes, G. R., and Rogers, 0. A., STUDY AND

DESIGN OF ARMORED AIRCREW CRASH SURVIVAL SEAT, Hayes Interna-
tional Corporation; USAAVLABS Technical Report 67-2, U. S. Army
Aviation Materiel Laboratories, Fort Eustis, Virginia, March
1967, AD 812994L.
Armstrong, R. W., and Waters, H. P., TESTING PROGRAMS AND RE-

SEARCH ON RESTRAINT SYSTEMS, 1969 SAE Transactions, Society of
Automotive Engineers, New York, January 1969.
Begeman, P. C., and King, A. I., THE EFFECT OF VARIABLE LOAD

ENERGY ABSORBERS ON THE BIODYNAMIC RESPONSE OF CADAVERS, Wayne
I
State University; Report No. N00014-75-C-1015, Department of
the Navy, Office of Naval Research, Arlington, Virginia, Decem-
ber 1975, AD A022417.
Brinkley, J. W., and Shaffer, J. T., DYNAMIC SIMULATION TECH-

NIQUES FOR THE DESIGN OF ESCAPE SYSTEMS: CURRENT APPLICATIONS
AND FUTURE AIR FORCE REQUIREMENTS, AMRL Technical Report 71-29,
Aerospace Medical Research Laboratory, Wright-Patterson Air
Force Base, Ohio, December 1971, AD 740439.
Burrows, L., Lane, R., and McElhenney, J., CH-47 CRASH TEST
(T-40) STRUCTURAL, CARGO RESTRAINT, AND AIRCREW INFLATABLE RE-
STRAINT EXPERIMENTS, UMARTL Technical Report 78-22, Applied
Laboratory, U.
Technology (AVRADCOM),
oratories Army Research
FortS_ Eustis, and Technology
Virginia, April 1978,Lab- i
AD A055804.
Calvin, M., and Gazenko, 0. G., eds., FOUNDATIONS OF SPACE

BIOLOGY AND MEDICINE, Volume II, Book 1, National Aeronautics
and Space Administration, Washington, D. C., 1975.
Carr, R. W., and Phillips, N. S., DEFINITION OF DESIGN CRITERIA

FOR ENERGY ABSORPTION SYSTEMS, Beta Industries, Inc.; Report
No. NADC-AC-7010, Naval Air Development Center, Warminster,
Pennsylvania, June 1970, AD 871040.
259
BIBLIOGRAPHY (Continued)
Castle, A. B.. Jr., A STUDY OF SEAT BELT BUCKLE RELEASE METH-

ODS, NBS 10387, National Bureau of Standards, Washington,
D. C., January 1971.
Castle, A. B., Jr., and Armstrong, R. W., SEAT BELT RETRACTOR

STUDIES CONDUCTED BY THE OFFICE OF VEHICLE SYSTEMS RESEARCH,
NBS 10537, National Bureau of Standards, Washington, D. C.,
February 1971.
Chandler, R. F., and Christian, R. A., COMPARATIVE EVALUATION

OF DUMMY PERFORMANCE UNDER -G IMPACT, Proceedings, Thirteenth
Stapp Car Crash Conference, Aciety of Automotive Engineers,
New York, 1969, pp. 61-75.
Eggert, W. S., AN IMPROVED ARMORED HELICOPTER SEAT, Abstract

of Invention, Serial Number 418,339, Navy Case 51,992, Depart-
ment of the Navy, Washington, D. C., November 1973, AD/D000446.
Fasa:.lla, E. L., and Alfaro-Bou, E., NASA GENERAL AVIATION

CRASHWORTHINESS SEAT DEVELOPMENT, Paper 790591, presented at
Business Airci ft Meeting, Society of Automotive Engineers,
Inc., Wichita, Kansas, 3-6 April 1979.
Foret-Bruno, J. Y., et al., CORRELATION BETWEEN THORACIC LE-

SIONS AND FORCE VALUES MFASURED AT THE SHOULDER OF 92 BELTED
OCCUPANTS INVOLVED IN REAL ACCIDENTS, Proceedings, Twenty-
second Stapp Car Crash Conference, Society of Automotive En-
gineers, Inc., Warrendale, Pennsylvania, 1978.
Gordon, S. L., Kondo, A., and Breedon, D., ANALYSIS OF COMFORT

AND CONVENIENCE FACTORS IN IMPROVED RESTRAINT SYSTEMS, Report
No. DOT-HS-802-113, National Highway Traffic Safety Administra-
tion, Department of Transportation, Washington, D. C., December
1976.
Guarracino, J., Coryell, S., and Delvecchio, R., CREWMANS RE-

TENSION SYSTEM FOR PROTECTION AGAINST HIGH SPEED EJECTION UP
TO 600 KNOTS, Grumman Aerospace Corporation; Report No. NADC-
76119-40, Naval Air Develcpment Center, Warminster, Pennsyl-
vania, October 1976, AD A036898.
Haines, J. A., Jr., and W;.oers, H. P., RESTRAINT REQUIREMENTS

FOR SIDE AND REAR FACING SEATS, NBS 10386, National Bureau of
Standards, Washington, D. C., January 1971.
260
, -•• _________-_________...__________..._i•
•,'• -
Hammer, E. W., Jr., and Peterson, R. L., DESIGN AND MOCKUP
EVALUATION OF A HIGH-STRENGTH ARMORED CREWSEAT FOR TRANSPORT/
CARGO AIRCRAFT, Budd Company; AFFDL Technical Report 73-47,
Air Force Flight Dynamics Laboratory, Wright-Patterson Air
Force Base, Ohio, June 1974, AD 785145.
Hendler, E., et al., EVALUATION OF AN ADVANCED AUTOMOTIVE RE-
STRAINT SYSTEM USING HUMAN SUBJECTS, Naval Air Development
Center; Report No. DOT-HS-063-1-0811A, National Highway Traffic
Safety Administration, Department of Transportation, Washing-
ton, D. C., June 1975, AD A012469.
Hinckley, W. M., and Yang, J. C. S., ANALYSIS OF RIGID POLY-
URETHANE FOAM AS A SHOCK MITIGATOR, NOLTR 73-162, Naval Ord-
nance Laboratory, White Oak, Maryland, August 1973, AD 772484.
Knapp, Colonel S. C., ed., OPERATIONAL HELICOPTER AVIATION

MEDICINE, AGARD CONFERENCE PROCEEDINGS 255, Advisory Group for
Aerospace Research and Development, North Atlantic Treaty Or-
ganization, Neuilly sur Seine, France, December 1978.
Kyde H., Ph.D., and Reichwein, C. T., REVIEW OF THE DYNAMIC
RE" INDEX (DRI), NADC-MR-6810, Naval Air Development Cen-
te drminster, Pennsylvania, August 1968, AD 843496.
Pavlick, M. J., PROTOTYPE FABRICATION AND TESTING OF IMPROVED
RESTRAINT SYSTEM, Budd Company; Report No. NADC-73002-40,
Naval Air Development Center, Warminster, Pennsylvania, Jan-
uary 1973, AD 916893L.
Pavlick, M. J., Schwartz, M., and O'Rourke, J., PROTOTYPE FAB-
RICATION AND TESTING OF A MODIFIED MA-2 HARNESS, Report No.
NADC-75034-40, Naval Air Development Center, Warminster, Penn-
sylvania, April 1975, AD A013640.
Phillips, N. S., Carr, R. W., and Scranton, R. S., A STATISTI-

CAL INVESTIGATION INTO THE DEVELOPMENT OF ENERGY ABSORBER DE-
SIGN CRITERIA, Beta Industries, Tnc.; Report No. NADC-CS-7122,
Naval -Air Development Center, Warminster, Pennsylvania, Decem-
ber 1971, AD 749333.
Pierce, B. F., Woodson, W. E., and Selby, P. H., SOURCES AND

REMEDIES FOR RESTRAINT SYSTEM DISCOMFORT AND INCONVENIENCES,
Man Factors, Inc.; Report No. DOT-HS-801-277, National Highway
Traffic Safety Administration, Department of Transportation,
Washington, D. C., November 1974.
261
t~.!
- -.- -.-. •
Robbins, D. H., et al., DEVELOPMENT AND TESTING OF INTEGRATED
SEAT RESTRAINT SYSTEMS, Highway Safety Research Institute, Uni-
versity of Michigan; Report No. DOT-HS-800-528, National High-
way Traffic Safety Administration, Department of Transporta-
tion, Washington, D. C., June 1971.
Rothe, V. E., et al., CREW SEAT DESIGN CRITERIA FOR ARMY AIR-
CRAFT, TRECOM Technical Report 63-4, U. S. Army Transportation
Research Command, Fort Eustis, Virginia, February 1963.
Saczalski, K., et al., eds., AIRCRAFT CRASHWORTHINESS, Univer-

sity Press of Virginia, Charlottesville, Virginia, 1975.
Sarrailhe, S. R., and Hearn, N. D., DYNAMIC TESTS OF A YIELD-

ING SEAT AND SEAT BELT SYSTEM FOR CRASH PROTECTION, Report ARL/
STRUC. 358, Department of Defence, Australian Defence Scien-
tific Service, Aeronautical Research Laboratories, Melbourne,
Australia, March 1975.
Sarrailhe, S. R., COMPARISON OF YIELDING AND ELASTIC RESTRAINT

SYSTEMS FOR CRASH PROTECTION, Report ARL/SM 382, Department of
Supply, Australian Defence Scientific Service, Aeronautical Re-
search Laboratories, Melbourne, Australia, October 1972.
Singley, G. T., III, and Haley, J. L., THE USE OF MATHEMATICAL
MODELING IN CRASHWORTHY HELICOPTER SEATING SYSTEMS, paper pre-
sented at NATO/AGARD Conference, Paris, France, 10 November
1978.
Snyder, R. G., HUMAN IMPACT TOLERANCE, Paper 700398, Interna-

tional Automobile Safety Compendium, Society of Automotive En-
gineers, New York, 1970, pp. 712-782.
Stech, E. L., DESIGN AND EVALUATION METHODS FOR OPTIMIZING

EJECTION SEAT CUSHIONS FOR COMFORT AND SAFETY, Frost Engineer-
ing Development Corporation? AMRL Technical Report 68-126,
Aerospace Medical Research Laboratory, Wright-Patterson Air
Force Base, Ohio, February 1977, AD A036035.
Svensson, L. G., MEANS FOR EFFECTIVE IMPROVEMENT OF THE THREE-

POINT SEAT BELT IN FRONTAL CRASHES, FFV Industrial Products
Division (Sweden); SAE Paper 780898, Proceedings, Twenty-second
Stapp Car Crash Conference, Society of Automotive Engineers,
Warrendale, Pennsylvania, 1978.
Swearingen, J. J., ACCEPTANCE TESTS OF VARIOUS UPPER TORSO RE-

STRAINTS, Report No. FAA-AM 71-12, Federal Aviation Administra-
tion Civil Aeromedical Institute, Oklahoma City, Oklahoma,
February 1971, AD 726253.
262
'S
Underhill, B., and McCullough, B., AN ENERGY-ABSOPBING SEAT
DESIGN FOR LIGHT AIRCRAFT, SAE Paper 720322, presented at 1972
National Business Aircraft Meeting, Society of Automotive En-
gineers, Wichita, Kansas, 15-17 March 1972.
Zenobi, T. J., DEVELOPMENT OF AN INFLATABLE HEAD/NECK RESTRAINT
SYSTEM FOR EJECTION SEATS, Report No. NADC-76357-40, Naval Air
Development Center, Warminster, Pennsylvania, February 1977,
AD A038762.
263
I UC
A .. - .i
INDEX
Acceleration, definition 26
Adjustment, seat 88
Adjustment hardware, location 146, 163
Aftward load, definition 29
Aftward strength 172
Air Force head/spine model 102
Aircraft coordinate system 25
Airframe, energy absorption by 40
Anchorage, restraint system 67, 141, 144
Anthropomorphic dummy
definition 31
seat testing 188
Attitude
definition 25
landing gear 43
Back tangent line
definition 53
limit for comfort 37
tolerance 172
Body block, static testing 185
Bolted connections
restraint systems 147
seats 50
Bottoming, definition 36
Buckle
aircrew 130, 163
single-point release 139
troop/passenger 134, 163
Bulkhead
definition 36
distortion 46, 58
warping for test 182
Buttock reference line 34, 141
Buttock reference point 35
Cable energy absorber 121
Castings 50,-54
Ceiling-mounted seat 45, 51
Clearance for-seat stroke 42
Collapse
definition 36
prevention of 64
Combined load, definition 29
Comfort 37
cushion 125
restraint system 137
seat adjustment 38
vibration 39
264
,•[ii
II~iii I il II!11 li-i1-1111- II lil -- -- l l
INDEX (Continued)
Computer simulation 93
Air Force head/spine model 102
CVS 99
one-dimensional 103
PROMETHEUS 100
SOM-LA 95
Controls 213
Crash environment 40, 43
Crew station geometry 33, 37, 39
Cushion
comfort 38, 125
desirable features 126
dynamic overshoot 127
energy-absorbing 127
filled 127
net-type 128
ventilation 38, 125
CVS computer program 99
Damping, comfort influenced by 37
Deceleration, characteristic waveform 85
Design conditions for seats 44
Design eye position
definition 33
seat adjustment 39
Downward load
definition 29
strength requirement 172
Ductility, seat materials 49
Dummy
definition 31
seat testing 188
Dynamic overshoot
cushion 127
definition 29
restraint system 69, 187
typical seat response 89
Dynamic response
empirical 85
theoretical 82
Dynamic testing
data acquisition 193
litters 201
seats 18
Effective weight
calculation 175
definition 32
limit load 82, 173
seat design 166, 173
265
- ~---
,*, -
INDEX (Continued)
Efficiency, energy absorber
Emergency release, 81, 235
Energy absorber
cable 121
definition 36
design loads, crewseats
design loads, troop seats173, 175
178
honeycomb 118
inversion tube 112
notched 110
optimization of waveform
rate sensitivity 175 91
rod Pull-through tube 121
rolling torus 114
tension pulley 123
tension
tube tube 119
flaring 121
variable-load 92, 177
wire bending 109
S~Energy-absorbing stroke
lateral or longitudinal 42
Energy-absorbing webbing 159
Energy absorption
airframe contribution 40
landing gear 40, 174
optical relay tube 217
plastic foams 219, 230
restraint webbing 159
seat contribution 41,
seat stroke example 7069, 174
stopping distance 67
Equipment weights, seat design 167
Facial injury, prevention
Failure, definition 36 232
Floor
Attachment 45, 51, 64
distortion 46, 52
warping for test 54, 183
Forward load
definition 28
strength requirements 168
Frangible tube energy absorber
Frequencies, critical 39 121
Friction, in seat stroke
Head impact hazards 210 173
Headrest cushion 128
Heel rest line, definition
35
266
LJ L> - _ _ _ _ _ _ _
_ _ _
INDEX (Continued)
Honeycomb energy absorber 118
Horizontal vision line, definition 33
Human body coordinates 30
Human tolerance
definition 31
head impact 210
limit load determined by 85
vertical acceleration 89, 172
Iliac crest 33, 137
Impact velocities 44
Inelastic analysis 65
Inertia reel 164
Inflatable restraint system 137
Instrument panel 212
Inversion tube energy absorber 112
Ischial tuberosity 35
Isolation, vibration 39
Joint deformation, seats 45, 52
Landing gear
energy absorption 40, 174
failure 43
Lap belt
anchorage 141
tiedown strap 33
Lateral load, definition 29
Lateral strength, seats 178
Limit analysis 65
Limit load, definition 36
Litter
occupant weights 196
restraint 49
testing 200
Load factor, definition 27
Load limiter, definition 36
also see energy absorber
Loads, seat design 168
Longitudinal strength, seats 168
Major impact 26
MIL-HDBK-5 51 -
MIL-S-58095(AV) 22, 160
MIL-STD-850 33
MIL-STD-1290 22
MIL-STD-1333 33, 38
Motion envelopes 203
Net-type cushions 128
267
I.
D Continued)
Occupant weights
litter design 196
seat design 166
Optical relay tube 214
Padding
application 219, 243
ductile support 244
energy absorption 230
functions 219
load-distributing 232
Plastic analysis 148
Plastic foams
materials 222
properties 219, 222
test methods 227
types 222
PROMETHEUS computer program 100
Rate sensitivity, seats 175,
Rebound, definition 29 179
Reflected shoulder straps 130
Releases, structural 54, 61
Restraint system
adjusters 146, 163
aircrew 130
anchorage 67
buckle 139, 163
design loads 149
dynamic overshoot 130
emergency release 139
fitting roughness 158
gunner 134
hardware materials 146
inertia reel 164
inflatable 137
lap belt anchorage 141
litter 49
reflected straps 130
requirements for seat types
48
shoulder harness anchorage 144
single-point release 129
stitching 151
tiedown strap anchorage 144
testing 180
troop 133
webbing characterietics 146,
148
weight 161
268
INDEX (Continued)
Retrofit 23
Riveted connections 51, 147
Rod pull-through tube energy absorber 121
Rolling torus energy absorber 114
Rollover 43
Rudder pedals 213
Seat
aft-facing 48
cabin 42
ceiling-mounted 45
clearance for stroke 41
component attachment 195
cushions 125
design conditions 44
dynamic response 82, 85
forward-facing 48
joint deformation
loads during crash 41
reference point 35
52
I
side-facing, restraint for 48
stroke
computation 69
efficiency 81
minimum, crewseats 174
minimum, troop seats 178
testing requirements
dynamic 188
static 181
troop 42, 178
vibration 39
width for comfort 38
Shoulder harness anchorage 1441
Sighting systems 214
Single-point release 129
Skull fracture, prevention 210, 219, 231
SOM-LA computer program 95
Specific energy
definition 36 -
energy absorbers 106
plastic foams 239
Static strength, definition 35
Static testing
adjustment position 181
body block 183, 185
deflection measurement 186
269
,I
INDEX (Continued)
floor warping 55
litters 200
seats 181
Stitching, restraint system 151
Stopping distance 67A
Strength requirements *
litter 197 k
restraint webbing 149
seats 166 H
Stress corrosion 50 1
Strike envelopes 203
Stroke
clearance for 42
computation example 70
minimum 82, 174, 178
Structural integrity, definition 35
i
Structural release 54, 61
Submarining
definition 32
lap belt location 141
restraint system preventing 67
Survivable accident, definition 30
Tension pulley energy absorber 123
Tension tube energy absorber 119
Testing
litter 200
seat 181
Thigh tangent angle
definition 34
limits for comfort 37
Thigh tangent line, definition 34
Tiedown strap
anchorage 144
comfort 130
Tolerable deceleration 89
Tolerance,- human, definition 31
Transmissibility, definit-ion 29 . a
Troop seat energy absorbers 110
Tube flaring energy absorber 121
Upward load, definition 29 i
Upward strength, seats 178
Variable-load energy absorber 92, 177
Velocities, impact 44
Vibration sources 39
I
Visionic systems 214
Warping of attachments 52, 54, 58
270
. 1
S ', • , •u,"- m••

-' • ,::W•'•i••
;:"• '"'•
' "- . ..... :• . .. • " -
.........
INDEX (Continued)
Webbing
characteristics 146
energy-absorbing 159
low-elongation 150
stitching 151
strength 149
thickness 150, 151
width 150, 151
wrap radius 158
Weights
equipment 167
occupant, seat 166
litter 196
Welded connections 51, 147
Wire-bending energy absorber 109
271
DEPARTMENT OF THE' ARMY
APPLIED TECHNOLOGY
US ARMY RESEARCH AND TECMNOLOGYLABORATORY
LABORAIORIES AVRADCOM
F¥)RT EUSTIS VIRGINIA 23d04
READER'S SERVICE LETTER

(For use in submitting comments, recommendations, correction;, and
revisions for Aircraft Crash Suri,ial Design Guide)
"FROM:
TO: Director, Applied Technology Laboratory, US Army

Laboratories (AVRADCOM), ATTN: DAVDL-ATL-ASV, Research and Technology
Fort Eustis, Virginia
23604
REMARKS:
SIGNATURE: DATE:
._
,.1
S .......
•'• " •""-#;,
.. ... ,-"......i';.•S•
UUS
DEPARTMENT OF THE ARMY
APPLIED TECHNOLOGY LABORATORY
ARMY RESEARCH AND TECHNOLOGY LABORATORIES ,AVRADCOM
1
I
FORT EUSTIS VIRGINIA 23604

(For use in submitting comments, recommendations, corrections, and
revisions for Aircraft Crash Survival Design Guide)
o F ROM:
I
TO: Director, Applied Technology Laboratory, US Army Research and Technology
Laboratories (AVRADCOM), ATTN: DAVDL-ATL-ASV, Fort Eustis, Virginia
23604
REMARKS:
S
I
I
4
I
A IrII!
DEPARTMENT OF THE ARMY

-APPI. IEO TEýHNOLOGV I.ABORA T014
US AHMV HILESA4CH AND0 TECHNOLOGY LASORAIOHIES AVRADCOM
FORT EUISTIS VIRGINIA 23604

(For use in submitting comments, recommendations, corrections, and
revisions for Aircraft Crash Surviria Design Guide)
FROM:
TO: Director, Applied Technology Laboratory, US Army Research and Technology

Laboratories (AVRADCOM). ATTN: DAVDL-ATL-ASV, Fort Eustis, Virginia
23604
REMARKS:
SIGN'ATURE: DATE:

Ada 088441

Uploaded by

Copyright:

Available Formats

Ada 088441

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ada 088441

Uploaded by

Copyright:

Available Formats

Y USARTL-TR-7- 22 / y

AIRCRAFT CRASH SURVIVAL DESIGN GUIDE I

F' f=.•:.- AIMULA INC. 48TH STREET

Approved for public release; distribution unlimited.

THIS DOCUMENT IS BEST QUALITY

This revised edition of the Crash Survival Design Guide was

This document has been coordinated with USAAVRADCOM, the U. S.

The technical monitors for this program were Messrs. G. T.

Comments or suggestions pertaining to this Design Guide will

NOTE: This is a revised edition of the Crash Survival Desil 'S

.. . ....-. ,-.- .n ,..L.,.,,s.•.•diI1•4 -

W STATEMENTIO(of thISPAopoff)at ECURITSYIO Rtee

Volume II - Aircraft Crash Environment and Human Tolerance-'

A major portion of the data contained herein was taken from

Additional credit is due the many authors, individual compa-

Section Title Pa.e

LIST OF ILLUSTRATIONS ................

CHAPTER 1. BACKGROUND DISCUSSION. .......... . 20

CHAPTER 2. DEFINITIONS. . .......... . . 25

2.1 AIRCRAFT COORDINATE SYSTEMS AND

3.1 INTRODUCTION . . . ....... ... 37

CHAPTER 4. DESIGN PRINCIPLES FOR SEATS AND

4.1 INTRODUCTION . . . . . ....... 48

Section Title Page

4.4.2 Riveted Connections . . . . . 51

CHAPTER 5. ENERGY-ABSORBING DEVICES . . . . . . . . 106

5.1 INTRODUCTION .............. ... 106

CHAPTER 6. SEAT CUSHIONS .... .............. ... 125

6.1 INTRODUCTION .... ......... ... 125

'9-- -,;• ... . --.- '•' • •" q- ,, • -_

CHAPTER 7. DESIGN PRINCIPLES FOR PERSONNEL

7.1 INTRODUCTION .... ........ . . ... .. 129

CHAPTER 8. SEAT STRENGTH AND DEFORMATION -

8.1 INTRODUCTION ............ 166

Section Title•e Page

8.3.5 Upward Loads . . . . . . .............. 178

CHAPTER 10. DELETHALIZATION OF, COCKPIT AND CABIN

10.1 INTRODUCTION ... .............. .... 203

REFERENCES .. ................. .. 246

BIBLIOGRAPHY. . . ......... ........... 259

1 Aircraft coordinates and attitude

2 Typical aircraft floor acceleration

4 Seating geometry ... .............. 34

S5 Sketch illustrating buckling or

"6 Static test floor warpage requirement

8 Aft seat leg casting attachment. . . . . 57

30 Torsional release of joints. . . . . . . 58

11 Fully released joint . . . . . . . . .. 59

12 Bulkhead in-plane warping. . . . . . . . 60

13 Universal release of a joint . . . . . . 61

14 Pin joint releases oriented to allow

17 Deceleration-time, velocity-time, and

25 Twelve-segment occupant model. . . . . . 97

26 Occupant model contact surfaces ..... 98

27 SOM-LA energy-absorbiing seat model. . . 98

29 PROMETHEUS occupant model. . . . . . .. 101

30 Three-dimensional head-spine model . . . 103

31 Lumped-parameter model of seat, seat

32 Crashworthy troop seat . . . . . . . . . 110

33 Troop seat tension energy absorber