•am

Approved for pyMc rej«ffje#;

Distribution Unlimited

ro

UNSCREENED SCREENED!!
ELECTRONICS ELECTRONICS

July 1993 TRI-SERVICETECHNICAL BRIEF 002-93-08

DIICQDALITyiNSHBCTED
For additional copies, unlimited reproduction is authorized, or contact:

USA: AMCRD-1C
ALEXANDRIA, VA 22333-0001

USN: ASN(RD&A)PI
WASHINGTON, DC 20352

USAF: SAF/AQK
1060 AIR FORCE PENTAGON
WASHINGTON, DC 20330-1060
 TRI-SERVICE TECHNICAL BRIEF 002-93-08

TRI-SERVICE
ENVIRONMENTAL STRESS SCREENING
GUIDELINES

JULY 1993
 PREFACE

Our urgent requirements to improve the operational performance of weapon systems

while reducing their operating and support costs can be met only by improving their quality
and reliability. Environmental stress screening (ESS) has been shown to be a significant aid
towards meeting these objectives.

This document is the culmination of work that began in the mid-1980s when industry,
with Government encouragement, initiated the revision and improvement of existing
Government ESS guidelines. The Departments of the Army, Navy, and Air Force have
collaborated in its preparation. It provides guidance for implementing the ESS requirements
in DoD Instruction 5000.2, encouraging consistency in interpretation among all three
services.

By providing a single source of ESS management methods, engineering guidance, best

practices and issues to be considered in preparing contracts, this document will help
program managers, project engineers, and contracting officers implement a successful
environmental stress screening program. We encourage you to use it.

{5p&3&^
LLOYD K. MOSEMANN, H
Deputy Assistant Secretary
(Communications, Computers, and Support Systems)
Office of the Assistant Secretary of the Air Force (Acquisition)

KJUL^

W. J. WILLOUGHB Y/JR.
Director, Product Integrity
Office of the Assistant Secretary of the Navy
(Research, Development and Acquisition)

<^2, 4
^TEPH£N R. BURDT
Director for Program Evaluation
Office of the Assistant Secretary of the Army
(Research, Development and Acquisition)
 ACKNOWLEDGMENTS

TRI-SERVICE COMMITTEE ON ENVIRONMENTAL STRESS SCREENING

Douglas O. Patterson U. S. Navy OASN(RD&A)PI Chairman

Louis C. Gills U. S. Navy OASN(RD&A)PI
MAJ Kevin Kirlin U.S. Air Force SAF/AQXM
Dr.PatLarter U. S. Air Force SM-ALC
Bobby G. Whiteley U. S. Army AMCRD-1C
Frank DiGiovanni U. S. Army CECOM

INSTITUTE OF ENVIRONMENTAL SCIENCES

ENVIRONMENTAL STRESS SCREENING OF ELECTRONIC HARDWARE
DOCUMENT REVD2W WORKING GROUP

Michael W. Gust Mitre Corporation Bedford, MA Chairman

Henry J. Caruso Westinghouse Electric Corporation Baltimore, MD
Jack Dunn Mitre Corporation Bedford, MA
Dr.PatLarter U.S. Air Force Colorado Springs, CO
C. E. Mandel, Jr. Consultant Laguna Hills, CA
Arthur A. McGill Lockheed Missiles and Space Company Sunnyvale, CA
George Miller Naval Air Warfare Center Indianapolis, IN
Michael A. Roy Westinghouse Electric Corporation Baltimore, MD
Edward A. Szymkowiak Westinghouse Electric Corporation Baltimore, MD
Doug Allison Litton Guidance and Control Systems Woodland Hills, CA
Rudy S. Cazanjian GTE Government Systems Mountain View, CA
James R. Marano Alliant Techsystems Hopkins, MN
CONTENTS

1 INTRODUCTION 1

1.1 PURPOSE & SCOPE 1

1.2 BACKGROUND 1
1.3 DOCUMENT PREPARATION 1

2 PROGRAM MANAGEMENT GUIDANCE 3

2.1 WHAT IS ENVIRONMENTAL STRESS SCREENING? 3

2.2 WHERE IS ESS APPLICABLE? 3

2.3 BENEFITS OF ENVIRONMENTAL STRESS SCREENING 4

2.4 PLANNING CONSIDERATIONS 4

2.5 MANAGEMENT ISSUES 4
2.6 PROGRAM MANAGEMENT CHECKLIST ,.6
DEMONSTRATION & VALIDATION 6
ENGINEERING & MANUFACTURING DEVELOPMENT 6
PRODUCTION & DEPLOYMENT 7
REPROCUREMENTAND DEPOT LEVEL OVERHAUL 7

2.7 GUIDANCE SUMMARY 8

3 ENGINEERING GUIDANCE 9

3.1 TECHNICAL ISSUES 9

TYPES OF FLAWS TO BE PRECIPITATED 9
LEVELS OF ASSEMBLY AT WHICH ESS SHOULD BE PERFORMED 9
TYPES AND SEVERITIES OF STRESSES 10
FAILURE DETECTION 12
Measurements During Thermal Cycling 12

Measurements During Random Vibration 12

BASELINE ESS PROFILES 12

CONTENTS (Continued)

OPTIMIZING/TAILORING OF ESS 13

RELATIONSHIPS OF ESS TO OTHER ACTIVITIES

IN PRODUCT DEVELOPMENT AND PRODUCTION 15

Parts Rescreening and Quality „ 15

Test, Analyze and Fix Programs 15

Reliability Demonstration and Production Reliability Acceptance Testing 15

Failure Reporting and Corrective Action System 15

SAMPLING VS100% SCREENING 16

EQUIPMENT CONFIGURATION , 16

MOUNTING SCHEMES 16

PERFORMANCE MONITORING AND

POWERED VERSUS UNPOWERED CONSIDERATIONS 16

During Thermal Screening 16

During Random Vibration 17

CHAMBER AIR FLOW CHARACTERISTICS 17

REPEATED SCREENING 18

3.2 GUIDANCE SUMMARY 18

4 VIBRATION SCREEN DEVELOPMENT 19

4.1 METHOD A-VIBRATION SURVEY 19

GENERAL TECHNIQUE 20

CONFIGURATION 20

Item 20
Level 20

Strategy 20

Excitation System 20

Fixturing 20

VI
CONTENTS (Continued)

MEASUREMENT PHILOSOPHY 20

Selection of Measurement Locations 20

Accelerometers • 21

Data Acquisition 22

Data Acquisition Equipment 22

Recorder Setup 22

Documentation 22

Calibration 23

Data Recording and Review 23

DATA PROCESSING 23

Data Analysis Equipment 23

Data Analysis Parameters 23

Documentation 24

PROCEDURE 24

SIMPLIFIED TECHNIQUE 25
4.2 METHOD B- STEP-STRESS TESTS 25
PROCEDURE 25
4.3 METHOD C - FAULT REPLICATION 25
4.4 METHOD D-HERITAGE SCREEN 26

5 THERMAL CYCLING SCREEN DEVELOPMENT 27

5.1 METHOD A -THERMAL SURVEY 27
THERMAL SURVEY PURPOSE 27
THERMAL SURVEY GUIDELINES 28
CYCLE CHARACTERISTICS 29
TEMPERATURE EXTREMES 29

RATE OF CHANGE OF TEMPERATURE 29

Vll
CONTENTS (Continued)

DWELL TIMES AT TEMPERATURE EXTREMES 30

Stabilization Time 31
Soak Time 31
EQUIPMENT CONDITION 31
NUMBER OF CYCLES 32
5.2 METHOD B- HERITAGESCREEN 33

A GLOSSARY & DEFINITIONS A-1

A.1 GLOSSARY A-1
A.2 DEFINITIONS A-1

B REFERENCES B-1

B.1 MILITARY B-1

B.2 INDUSTRY B-3

Vlll
FIGURES

4-1 EXAMPLE OF A SCREENABLE ITEM SHOWING POSSIBLE MEASUREMENT LOCATIONS 21

4-2 RESPONSE MEASUREMENT LOCATIONS ON RECTANGULAR PWA WITH BASE CONNECTOR .22
4-3 STEP-STRESS CONCEPT: RELATION OF ENVIRONMENTAL LEVELS 26

5-1 TEMPERATURE HISTORIES WITH HIGH CHAMBER AIR SPEED AND LONG DWELL TIMES 30
5-2 TEMPERATURE HISTORIES WITH LOW CHAMBER AIR SPEED AND SHORT DWELL TIMES ....30
5-3 TYPICAL-TEMPERATURE CYCLING PROFILES 31

TABLES

2-1 ESS BENEFITS TO MANAGEMENT 5

3-1 SCREENING ENVIRONMENTS VERSUS TYPICAL FAILURE MECHANICS 9

3-2 RISKS AND RESULTS OF ESS AT VARIOUS LEVELS 11
3-3 BASELINE VIBRATION PROFILE 13
3-4 BASELINE THERMAL CYCLE PROFILE 14

4-1 VIBRATION SCREEN DEVELOPMENT CONSIDERATIONS 19

4-2 DATA ANALYSIS LOG PARAMETERS 24

IX
SECTION 1
INTRODUCTION

experienced by equipment due to infant mortality of

1.1 PURPOSE & SCOPE parts, burn-in at higher levels of assembly was insti-
Environmental stress screening (ESS) is a cost- tuted in the 1960s. Noneof these approaches, however,
effective means ofimproving quality and reliability of stressed the assemblies adequately to precipitate manu-
electrical, electronic, electro-optical, electromechani- facturing defects. ESS was introduced in the 1970s to
cal and electrochemical assemblies and systems at a stimulate the identification of latent defects as early in
time when defect removal is relatively inexpensive. the production process as possible and to correct the
This document provides guidance for implementing process to preclude their recurrence.
the ESS requirements in Part 6, Section C, Paragraph Each Service independently developed regulations
3f(l) of DoD Instruction 5000.2 dated 23 February and guidance on implementing ESS in their acquisi-
1991. It will help program managers, project engi- tions. Among companies making products for
neers, and contracting officers implement a successful tri-service use, this led to conflicts and confusion,
ESS program. It explains to management the benefits increased acquisition costs through inefficient utiliza-
of ESS, and when and how to implement it, and tion of ESS screening equipment, increased
conveys ESS fundamentals, planning and execution to configuration control efforts, and more complex lo-
engineers. It focuses on ESS in development, produc- gistics systems.
tion and overhaul, at levels of assembly from the The Departments of the Army, Navy, and Air Force
printed wiring assembly to the system. ESS of parts is have collaborated in the preparation of this document
covered in other publications. to resolve this problem. By offering a single source of
Random vibration and temperature cycling have ESS management methods, engineering guidance,
proven to be the most successful forms ofESS in terms best practices and issues to be considered in preparing
of effective flaw precipitation. The focus of these Statements of Work and contracts, this document will
guidelines is on these forms, and an acceptable meth- help to assure consistency in interpretation and imple-
odology for each is detailed herein and recommended mentation of ESS programs across all three Services.
for use. Otherforms ofESS whichmay prove effective
for specific hardware configurations and characteris- 1.3 DOCUMENT PREPARATION
tics will require approval by the procuring agency.
The Tri-Service Environmental Stress Screening
Guidelines acknowledges the Institute of Environmen-
1.2 BACKGROUND tal Sciences (IES) and the Environmental Stress
Screening ofElectronic Hardware (ESSEH) Committee
The Department of Defense and its contractors have fortheir comments and foruseof published information
made great strides in recent years in learning how to presented in the ES-ESSEH Environmental Stress
design reliable weapon systems. Most failures in fielded Screening Guidelinesfor Assemblies dated March 1990.
systems today are traced to defective parts and im- Major portions of the ESSEH Guidelines have been
proper workmanship during manufacturing. reprinted with permission from the IES. These guide-
In the 1950s, in order to reduce rework due to lines have been developed by the ESSEH Technical
defective parts, functional testing of incoming parts Committee of the IES as a continuing series to advance
was introduced. Because of the initial high failure rate the state of technical and engineering sciences.
SECTION 2
PROGRAM MANAGEMENT GUIDANCE

The purpose of this section is to assist program

managers in understanding the issues and implement- A viableESSprogrammustbe dynamic—thescreen-
ing environmental stress screening. Contracting ing program must be actively managed, and tailored to
specialists will also find this section useful in integrat- the particular characteristics of the equipment being
ing and implementing ESS as a part of the acquisition screened. This includes conducting a survey to deter-
strategy. mine the mechanical and thermal characteristics of the
equipment and refining the screening profiles as more
information becomes available and/or designs, pro-
2.1 WHAT IS ENVIRONMENTAL cesses, and circumstances evolve.
STRESS SCREENING?
A clear understanding of environmental stress 2.2 WHERE IS ESS APPLICABLE?
screening requires a good definition as a baseline. The
following definition addresses the key aspects of ESS: Best design and manufacturing practice calls for the
application of environmental stress screening to:
Environmental stress screening of a product is a
process which involves the application of one or • All material acquisitions that include electrical,
more specific types of environmental stresses for electronic, electro-optical, electromechanical or
thepurpose ofprecipitating to hardfailure, latent, electrochemical components in demonstration &
intermittent, or incipient defects or flaws which validation, engineering & manufacturing devel-
would cause product failure in the use environ- opment and production phases
ment. The stress may be applied in combination or • Reprocurements and the procurement of spare
in sequence on an accelerated basis but within and repair parts where the cost of ESS implemen-
product design capabilities. tation can be amortized economically or where
ESS was required in original equipment
ESS isolates manufacturing problems caused by
poor workmanship, faulty and/or marginal parts. It • Depot overhaul programs where opportunities .
also identifies design problems if the design is inher- exist for substantial cost savings and overhaul/
ently fragile and if qualification and reliability growth repair effectiveness
tests are too benign. The most common stimuli used in • Nondevelopmental items, such as commercial
ESS are temperature cycling and random vibration. off-the-shelf (NDI-COTS) and domestic or for-
ESS is a process rather than a test in the normal eign military (NDI-Military) items only to the
accept/reject sense. Those participating in the effort, extent ESS was implemented and documented
including the contractor, should never be led to be- during either current or previous production. NDI
lieve that a "failure" is bad and would be held against items are not to be used unless they comply with
them. ESS is intended to stimulate defects, not to all specified requirements, including ESS.
simulate the operating environment, and therefore, • Equipment and spares that have been specifically
factory "failures" are encouraged. designated to receive ESS
The root causes ofESSfailures need to befoundand
corrected before there is a complete process. ESS may be applied at any manufacturing level,
Initially, ESS must be applied to 100% of the units from piece parts to end items. It is intended to screen
manufactured, including repaired units. By using a defects in a manner that is not harmful to properly
closed loop feedback system, one will be able to manufactured material. Some components, such as
eventually determine if the screening program should plasma displays, vacuum tubes, etc., by nature of their
be modified. design, are not amenable to either vibration or tem-
perature screening. Hardware proven to be too fragile
 may be excluded, but rationale for exclusion must be and defined profiles) are in place when production
approved by the government. begins. Therefore, it is desirable to reach this state
during engineering & manufacturing development, so
2.3 BENEFITS OF ENVIRONMENTAL that hardware for qualification and reliability growth
STRESS SCREENING testing is of higher quality and can be screened (to
prevent failures that are not design related). This
Proper application of environmental stress screen- implies that experimentation and planning should
ing offers several benefits: begin early.
• Reduced overall life cycle cost The cost of rework in manufacturing escalates by
orders of magnitude as the assembly process proceeds
• On-time deliveries from piece part level to printed wiring assembly/
• Improved reliability after delivery module, unit, system, and to the user. Finding defects
at the lowest possible level of assembly will tend to
• Improved user confidence and/or satisfaction minimize rework costs by reducing corrective action
• Reduced support costs time. However, some flaw types manifest themselves
only at the higher levels of assembly. Tailoring the
• Improved readiness screen by means of the vibration and thermal charac-
• Improved production process teristicsofthehardware coupled withdefect population
at each level of assembly is essential.
While these benefits far outweigh the costs of imple- This document presents many management and
mentation, they do not come without a penalty. ESS technical details to be considered and some of the
must be implemented early in the program and closely trade-off decisions that will vary with specific pro-
supervised throughout. It will take time and commit- grams. The guidance presented may be limited in
ment of the senior managers, because the benefits are some areas but there is no intent to make this a
long term but the requirements for people and funds textbook on the many facets of ESS. Because of the
occur early in the program. obvious cost, schedule and performance impacts rela-
ESS is normally conducted during the manufactur- tive to the ESS decisions, both government and
ing process to detect latent defects in parts and contractor program managers must not treat their
workmanship, but may also disclose design limita- decisions lightly. Where the required ESS expertise is
tions that were not detected during qualification and not available or will not be in time to address these
engineering tests. In addition, there are distinct ben- issues, ESS consultants may need to be considered.
efits to conducting ESS during development as well. A Each service has at least one organization which
considerable percentage of the failures encountered specializes in ESS engineering and the Institute of
during a reliability growth (test, analyze and fix) test Environmental Sciences has published many appro-
program may be caused by poor workmanship and priate articles in addition to those listed in Appendix B.
defective parts. These non-design-related failures can
mask design-related failures, can cause schedule slip-
page, and can adversely affect performance. By 2.5 MANAGEMENT ISSUES
screening the item prior to this testing, these adverse The following ESS management issues and guid-
effects can be minimized. It is virtually impossible to ance need to be considered to increase the probability
achieve design reliability without reducing to a mini- of implementing a viable ESS program:
mum the reliability degradation due to screenable
flaws. Additional benefits are presented in Table 2-1. • How critical are the items proposed for ESS and
whatlevel of quality is required? Criticality would
2.4 PLANNING CONSIDERATIONS be high if a failure of the item results in high
probability of loss of life or an inability to com-
It is imperative that ESS resources, training require- plete a mission, high life cycle cost, or high cost of
ments, and detailed plans (includinglevels of assembly failure.
TABLE 2-1. ESS BENEFITS TO MANAGEMENT

MANAGEMENT TARGET BENEFITS/RATIONALE

ENGINEERING & PROGRAM MANAGER • ENSURES HARDWARE PERFORMANCE ON DEMAND

MANUFACTURING AND ENGINEERING • CONTRIBUTES TO PARTS LIST DEVELOPMENT
DEVELOPMENT MANAGER
• IMPROVES RELIABILITY GROWTH TESTING
• ASSURES READINESS OF PRODUCTION SCREENS
• ESS WEEDS OUT PROBLEM VENDORS
• 60% FAILURES ARE DUE TO WORKMANSHIP DEFECTS
• 30% FAILURES ARE DUE TO PART FLAWS
• ESS DESIGN IS BY NATURE ITERATIVE

PRODUCTION MANUFACTURING • REDUCES REWORK COST

MANAGER • MINIMIZES SCHEDULE DELAYS
• FACIUTATES ACHIEVEMENT OF DESIGN RELIABILITY IN PRODUCTION
HARDWARE
• IMPROVES PRODUCTIVITY
• REDUCES OR ELIMINATES WORKMANSHIP AND PART DEFECTS
• ENTIRE PRODUCTION PROCESS MORE EFFICIENT
• DEFECTS SURFACED AT LOWEST LEVELS OF ASSEMBLY AND ROOT
CAUSE CORRECTIVE ACTION IMPLEMENTED
• ACCEPTANCE TEST PASSED ON FIRST PASS, LESS HIGH ASSEMBLY
LEVEL REWORK

GENERAL • LESS PROGRAM COST

• LESS SCHEDULE AND VARIANCE IMPACT
• HIGHER FIELD RELIABILITY
• FAILURES ARE FORCED TO EMERGE AT CONVENIENT PRODUCTION
STEPS
• LESS LATENT DEFECTS SHIPPED TO FIELD

USER • INITIALLY DELIVERED HARDWARE MEETS RELIABILITY AND QUALITY

REQUIREMENTS
• LOWERS SUPPORT COST
• LOWERS LIFE CYCLE COST
• SPARES MEET RELIABILITY REQUIREMENTS
• ON-TIME DELIVERIES
• INCREASES CONFIDENCE AND SATISFACTION
• REMOVES DEFECTS NORMALLY PRESENT IN DELIVERED HARDWARE
• FEWER FIELD FAILURES OR MAINTENANCE ACTIONS
• ESS INFLUENCES THROUGHOUT ACQUISITION CYCLE

The quantity to be procured should be considered. other ESS parameters that are less than those of
Where small quantities are involved and the item the baseline provided in Tables 3-3 and 3-4. In all
does not qualify as a high criticality item as given cases, appropriate rationale and data should be
above, then it may be cost effective to use only the presented to justify the ESS conditions to be
relatively low cost thermal cycling screens. applied.
The tailoring and optirnizationprocesses described The type of random vibration should be consid-
in this document may result in stress levels or ered. Should true random vibration excitation or a
 low cost alternative such as quasi-random (pneu- 2.6 PROGRAM MANAGEMENT
matic)vibrationexcitationbeusedforthedetection CHECKLIST
of flaws?
■ The contractor's proposed ESS program plan The following checklist should be used in the devel-
should emphasize the following: opment of a management plan for implementing ESS
in each phase of the acquisition process:
- Commitment to and understanding of ESS
- Failure reporting and corrective action system DEMONSTRATION & VALIDATION
(FRACAS). A FRACAS should be in place and
• Establish adequate ESS funding. To facilitate this,
operating.
a cost/benefit analysis should be conducted to
- Span of control for ESS. If ESS is being per- justify funding. The basis of this analysis could be
formed by multiple subcontractors, what is their the development of a historical data base on costs
degree of implementation? to implement various screens versus return on
investment (cost avoidance).
- Planned ESS profile optimization technique.
The Services recommend one of four random • Assess the training needs of ESS personnel and
vibration techniques, and one of two thermal develop a plan to correct any identified training
cycling techniques. Each has both advantages and/or qualification deficiencies.
and disadvantages.
• Determine equipment availability, adequacy, ca-
- Managerial and technical approaches to ESS. pacity, etc., to perform the intended screens.
The plan should include proposed methods for
determining initial screening environment, ap- • Identify speciallongleadequipmentrequirements
plicable assembly levels, data collection, (e.g., fixtures, racks, etc.).
failure analysis and corrective action, and pro- • Determine appropriate initial profile.
cedures or methods to be used in altering the
program. • Establish a FRACAS to report and analyze faults
that are precipitated out during screening.
- Nondevelopment items (NDI), such as com-
mercial off-the-shelf and domestic or foreign
ENGINEERING & MANUFACTURING
military items, if those items have been deter- DEVELOPMENT
mined to meet government requirements
The Government program manager should also • Continue to tailor, refine, and evaluate the ad-
address the following additional issues: equacy of the ESS profile, striving for an optimum
screen. The absence of fault precipitation during
- ESS profile requirements should not be speci- initial production or reprocurement may be an
fied in the RFP. In general, it is better to indication of a weak screen that needs further
allow contractors to propose an ESS profile optimization.
than to specify a particular profile, unless the
contract is a reprocurement and the original • Establish or continue a closed-loop FRACAS to
profile holding fixtures, vibration machine report and analyze faults that are precipitated out
and chamber capabilities are contained or ref- during screening.
erenced on the drawings and are found to be • If a test-analyze-fix (TAAF) program is being
satisfactory. (Note: the original profile may implemented, apply ESS just prior to the start of
have to be modified due to changes in the the TAAF program, while continuing to strive for
production process and component manufac- an optimum screen.
turing variability.)
• Document ESS requirements and appropriate de-
- The ESS and quality history of the contractor tails such as profiles, screening equipment and
fixtures as part of the product technical data pack- PRODUCTION AND DEPLOYMENT
age (TDP). The requirements shall be referenced
on the appropriate part/assembly drawings orparts • Establish or continue a closed-loop FRACAS to
list. Include in the TDP the statement: 'To the report and analyze faults that are precipitated out
extentthatthe profiles are equipment and/ormanu- during screening.
facturer unique, they may have to be modified due • Establish procedures to correct/monitor any work-
to changes in material or production processes." manship/parts problems identified during
Finalize the ESS profile before the system enters screening. Screens help to identify processes (both
into production. The following guidelines are in-house and vendor processes) that are "out of
provided to assist program managers in determin- control."
ing whether or not a reasonable screening profile • Provide parts failure information to parts manu-
has been developed. One or more of the following
facturers and require continuous improvements to
techniques may be required. reduce these deficiencies.
- Verifymatmemoreseveretemperaturescreening • Establish procedures to track fielded systems and
profiles are used at the lowerassembly levels (e.g., evaluate field failure information against the ef-
printed wiring assembly,module,subsystem, etc.). fectiveness of the current screens wherever
A good ESS program should drive outmost faults possible.
at the lower levels where faults are more easily
corrected and less costly to repair. • Establish criteria acceptable to the Government
on when and under what conditions 100% screen-
- Verifythattheproposedscreeningprofilesmeet ing shouldbe reduced to sampling. See continuous
or exceed the Tri-Service baseline presented in
sampling plans in MEL-STD-1235 (Reference
Tables 3-3 and 3-4. When the profiles do not
B.l-12).
meet or exceed the baseline, verify that ratio-
nale for this deviation is acceptable.
REPROCUREMENT AND DEPOT LEVEL
- Verify that the proposed screening profile is not OVERHAUL
so severe that it is damaging to the item being
stressed. By reviewing failure analyses a deter- • Derive the same benefits of ESS in reprocurement
mination can be made whether or not a failed items and depot overhauled items as initial pro-
component has been overstressed. If the results duction environments. Though the frequency of
of this review indicate that the item is being failure may be lower for depot overhaul items
overstressed, the screening profile should be (infant mortality/design updates are already in
adjusted until failure analyses indicate no fail- place through field use), poor workmanship and
ures are due to overstressing. In some cases a bad replacement parts are still a problem in the
minor design change, such as additional support depot overhaul environment.
for a resonant component, would be a more
Note: Numerous applications of ESS may be
logical and cost effective solution.
harmful to equipment. Depending on the particu-
- Verification may be made that a screening pro- lar equipment, the ESS program and the frequency
file is adequate by seeding known faults into an of overhaul, some useful life of the equipment
item and mendetemiiningifthe proposed screen- may be consumed.
ing profile is adequate to precipitate them to
• Requirethatallequipmentthatarereprocurements
hard failure.
be screened if ESS was required on the original
The profiles should not change unless the manu- procurement contract. System level equipment
facturing processes are changed, the system is should be screened at the originally developed
redesigned, parts are changed, or a different type screen or at a government approved equivalent
of screen is found to be more effective. screen. Where original screens werenotdeveloped
 for replacement modules, a determination based - Assembly level at which to apply ESS
on criticality and exist should be made to determine
whether or not to develop an appropriate screen. - Level of automation versus manual labor
• Establish a FRACAS whenever there is a screen- - Specific rates of thermal change versus capital
ing effort. investment
- Adequacy of available in-house random vibra-
2.7 GUIDANCE SUMMARY tion equipment versus cost of off-site screening
While individual program managers have great lee- or the purchase of new equipment
way on implementing ESS, the overall direction is - Cost considerations of active power-on versus
clear. Top management's commitment and attention passive power-on screening
is the key element in a successful ESS program. The
following summarizes the ESS guidance: • Consider sampling for the ESS screen based on
screening data collected, but only with customer
• Define contractual requirements. concurrence.
• Identify a general approach to satisfy these • Coordinate the ESS program with other activities
requirements. relating to quality and reliability.
• Perform a cost analysis considering the following: • Ensure that a FRACAS has been implemented.
SECTION 3
ENGINEERING GUIDANCE

This section discusses the technical issues that arise the sensitivity to these responses, product unique ESS
when implementing an ESS program. These issues are profiles can be developed. Table 3-1 gives examples
related, and decisions concerning them should be of typical defects that are sensitive to either thermal
made in an interactive manner. cycling, vibration or both. This table can be used as a
ESS is a manufacturing process and should not be guide in developing tailored ESS profiles. Care must
confused with a test. Ideally, a screening program be taken that in tailoring to one type of failure, other
should be designed for the system (equipment unique) classes of failures are detected.
to which it is to be applied. Such custom designing
may require the expenditure of resources to perform LEVELS OF ASSEMBLY AT WHICH ESS
additional analyses. Baseline screening parameters SHOULD BE PERFORMED
have been provided in this section and should serve as
a starting point only for custom designing a profile The term piece part as used herein is defined as a
unique to the particular item. The contractor should monolithic integrated circuit, resistor, switch, etc.,
not implement screening at or below baseline screen- which is the lowest level of assembly. The next level
ing parameters without prior Government approval. of assembly is a multi-part assembly that has a defined
identity — e.g., one that is given a drawing number
3.1 TECHNICAL ISSUES and, usually, a name. A typical item at this level is a
printedwiring assembly (PWA) or an equivalent shop
TYPES OF FLAWS TO BE PRECIPITATED replaceable unit (SRU). The top level is a system, but
one person's system is another's subsystem (engine,
B ased on the types of failures expected in the prod- propulsion system, air vehicle, weapon system). In
uct, product responses to environmental stimuli and reality, there is always some aggregate that is the
TABLE 3-1. SCREENING ENVIRONMENTS VERSUS TYPICAL FAILURE MECHANICS

SCREENING ENVIRONMENT
THERMAL CYCLING VIBRATION THERMAL OR VIBRATION

COMPONENT PARAMETER DRIFT PARTICLE CONTAMINATION DEFECTIVE SOLDER JOINTS

PWA OPENS/SHORTS CHAFED, PINCHED WIRES LOOSE HARDWARE
COMPONENT INCORRECTLY DEFECTIVE CRYSTALS DEFECTIVE COMPONENTS
UJ
cc INSTALLED MIXED ASSEMBLIES FASTENERS
_l .
WRONG COMPONENT ADJACENT BOARDS RUBBING BROKEN COMPONENT
U.
u- HERMETIC SEAL FAILURE TWO COMPONENTS SHORTING SURFACE MOUNT TECHNOLOGY
o
UJ CHEMICAL CONTAMINATION LOOSE WIRE FLAWS
>- DEFECTIVE HARNESS POORLY BONDED COMPONENT IMPROPERLY ETCHED PWAs
TERMINATION INADEQUATELY SECURED PARTS
IMPROPER CRIMP MECHANICAL FLAWS
POOR BONDING IMPROPERLY SEATED
HAIRLINE CRACKS IN PARTS CONNECTORS
OUT-OF-TOLERANCE PARTS
largest entity reasonably possible to subject to ESS. In TYPES AND SEVERITIES OF STRESSES
any event, there usually are several levels of assembly
at which ESS can be contemplated. A variety of environmental stresses have been can-
It is more cost effective to do ESS at the lowest level didates for use in ESS overthe years. Of these, random
possible and at more than one level. The choices of vibration and thermal cycling currently are considered
how many levels and which levels will involve an to be the most cost effective. Table 3-1 identifies some
engineering evaluation. common types of failures and reflects whether random
The costs associated with a failure usually appear in vibration or thermal cycling is the more likely stress to
connection with a single part or interconnection and precipitate that particular failure. A failure may also
will increase dramatically with the level of assembly. be surfaced under one stress, but detected under the
Consider the following brief list, which will vary other. The references in Appendix B present other
depending upon the manufacturer, complexity of the screening techniques which may be appropriate for
item and how much control the manufacturer has of some products.
his process: Traditional ESS, consisting of temperature cycling
• At higher levels and random vibration, may not be the most effective
environment to use for certain hardware. For example,
- More assembly work has to be undone and power cycling is effective in precipitating certain
redone when failures occur types of latent defects; pressure cycling may be desir-
- More material may need to be scrapped able for sealed equipment; and acoustic noise may
excite microelectronics structures better than struc-
- More impact on production flow and schedule ture-borne vibration. Ultimately, ESS environments
usually occurs must be chosen based on the types of flaws that are
• At lower levels known or expected to exist.
In the past, fixed-frequency or swept-sine vibration
- Corrective action is quicker were sometimes used. These practices were attribut-
- Repair cost is lower able in partto costs and physical limitations of available
test equipment at the time. However, the major reason
The above factors tend to lead management to is believed to be the lack of recognition of the short-
decide to perform ESS at lower levels of assembly. falls of fixed frequency and swept-sine vibration in
However, each step in assembly and integration pro- comparison with broadband random vibration.
vides additional opportunities for the introduction of Today, true random and quasi-random vibration are
flaws. Obviously, ESS at a particular level cannot used almost exclusively for ESS. True random vibra-
uncover flaws that are not introduced until the next tion, which is well known in the ESS community,
level. Generally, this dilemma is usually controlled by applies all frequencies in a certain bandwidth (usually
performing ESS at each major functioning level in the 20 to 2000 Hz) and is neither cyclic nor repetitive.
manufacturing process consistent with an assessment Quasi-random vibration, on the other hand, is a rela-
of the potential defect population at each level of tively new technology using pneumatically driven
assembly. vibrators which generate repetitive pulses. Forscreen-
Resolution of these conflicting considerations usu- ing applications, several (usually 4 to 6) of these
ally involves screening at multiple (usually 2 or 3) vibrators are attached to a specially designed shaker
levels of assembly. ESS at lower levels should focus table which is allowed to vibrate in multiple axes
on surfacing and correcting flaws in piece parts and simultaneously. This complex motion (6 degrees of
PWA processing. Thus, most ESS failures at higher freedom vibration — 3 linear axes and 3 rotational
levels will reflect flaws introduced later in the manu- axes) is very effective in finding all types of flaws.
facturing sequence that are usually correctable without It is not difficult to visualize that the complex
tear-down to the PWA level. Table 3-2 provides a interactions possible under random vibration can in-
summary of the risks and results of doing ESS at duce a wider variety ofrelative motions in an assembly.
various levels and functional conditions. As indicated by Table 3-1, vibration is the area of

10
TABLE 3-2. RISKS AND RESULTS OF ESS AT VARIOUS LEVELS
ESS CONDITIONS/TRADEOFFS RISKS/EFFECTS

POWER IAD 2 MONrTORED 3 TECHNICAL

LEVEL APPLIED' ESS COMMENTS
OF COST
ASSEMBLY
YES NO YES NO YES NO RISK RESULTS

TEMPERATURE CYCLING

X X X LOW LOW POOR

CONDUCT PRE & POST ESS FUNCTIONAL TEST

PWA X X X HIGH LOWER BETTER SCREEN PRIOR TO CONFORMAL COATING.

X X X HIGHEST LOWEST BEST

X X X HIGHEST LOWEST BEST

IF CIRCUMSTANCES PERMrT ESS AT ONLY ONE

LOWER HIGHER GOOD
UNrr/Box X -*• X X LEVEL OF ASSEMBLY, MPLEMENT AT UNIT LEVEL

X X X LOWEST HIGHEST POOR

MOST EFFECTIVE ESS AT SYSTEM LEVEL IS

SHORT DURATION RANDOM VIBRATION
SYSTEM X X X HIGHEST SEE COMMENT TO LOCATE INTERCONNECT DEFECTS
RESULTMG FROM SYSTEM INTEGRATION.

RANDOM VIBRATION

RANDOM VIBRATION IS MOST EFFECTIVE AT PWA

X X X HIGHEST LOW GOOD LEVEL IF:

1. SURFACE MOUNT TECHNOLOGY IS UTILIZED

PWA ^ PWA HAS LARGE COMPONENTS
X X X HIGH HIGH FAIR
3. PWA IS MULTILAYER
4. PWA CANNOT BE EFFECTIVELY SCREENED AT
X X X LOW HIGHEST POOR HIGHER ASSEMBLIES

X X X HIGHEST LOW BEST RANDOM VIBRATION MOST EFFECTIVE AT THIS

LEVEL OF ASSEMBLY. INTERMITTENT FLAWS MOST
UNIT/BOX X X X LOW HIGHER GOOD SUSCEPTIBLE TO POWER-ON WITH TO ESS.
POWER-ON WITHOUT TO REASONABLY EFFECTIVE.
DECISION REQURES COST BENEFIT TRADEOFF.
X X X LOWEST HIGHEST POOR

COST IS RELATIVELY LOW BECAUSE POWER

SYSTEM X X X LOW LOW GOOD AND VO NORMALLY PRESENT DUE TO NEED FOR
ACCEPTANCE TESTNG.

NOTES:

1. POWER APPLIED — AT PWA LEVEL OF ASSEMBLY, POWER ON DURING ESS IS NOT ALWAYS COST EFFECTIVE — SEE TEXT
2. I/O— EQUIPMENT FULLY FUNCTIONAL, WITH NORMAL INPUTS AND OUTPUTS
3. MONrTORED — MONITORING KEY FONTS DURING SCREEN TO ASSURE PROPER EOUPMENT OPERATION

stressing that normally precipitates latent assembly occurs over a long period. Bum-in is not a substitute
flaws caused by the undesired relative motion ofparts, for ESS.
wires, structural elements, etc., as well as mechanical Effective screening usually requires large, rapid
flaws that lead to propagating cracks. temperature changes and broadband random vibra-
Burn-in has been defined many ways by different tion. Such thermal cycling is used for the detection of
agencies and companies; however, for this document assembly flaws that involve installation errors or inad-
it is the exposure of powered equipment to either equate chemical or mechanical isolation or bonding.
ambient or steady elevated temperature. This tech- Under rapid thermal cycling (e.g., in solder joints),
nique has been used in the past with some success and differential thermal expansion takes place without
needs to be considered as a possible supplement to the sufficient time for stress relief, and this is a major
ESS requirement It is of particular value where com- mechanism for precipitating latent defects to detect-
ponents are of high power and where heat buildup able failures.

11
 As indicated in Table 3-1, some types of flaws may a marginal component may show up as an equipment
be precipitated to failures by either thermal cycling or failure only during the stress application. Otherwise,
random vibration. However, it is important to note that the incipient failure will escape detection, only to
thermal cycling and random vibration are synergistic. show up in an operational environment. Some of the
For example, thermal cycling following random vi- initial work in random vibration screening indicated a
bration sometimes leads to detection of vibration 2:1 difference in the efficiency of the screen if the
induced failures that were not apparent immediately. equipment were powered and monitored versus not
There have been reported cases where a very small powered. The technical risks and costs are summa-
flaw did not propagate to the point of detectability rized in Table 3-2 at each level of assembly for random
during random vibration, but advanced to the point of vibration screening.
detectability during subsequent thermal cycling.
The combined efforts (synergism) between vibra- BASELINE ESS PROFILES
tion and thermal cycling suggests that concurrent
application of the two stress types may be desirable. The baseline profiles (Tables 3-3 and 3-4) represent
This combined environment is in fact sometimes used the combined agreement of the three Services on
in ESS, but more often is avoided because it requires minimum levels to ensure effectiveness. They are
more elaborate facilities. Also, concurrent application derived from experimental and analytical stress screen-
of random vibration and thermal cycling can make it ing studies plus surveys of screens used in industry.
difficult to determine what caused a defect so that The random vibration baseline profile given in Table
corrective action can be taken 3-3 shows the values for response levels, frequencies,
If random vibration and thermal cycling are to be axes, duration and monitoring. The thermal cycling
conducted sequentially, random vibration would usu- baseline profile given in Table 3-4 shows a range of
ally be donefirst. A more effective sequence would be values for the temperature extremes, the temperature
five minutes of random vibration prior to thermal rate of change and the number of cycles.
cycling, and another five minutes of random vibration These baseline profiles for random vibration and
following. temperature cycling are not recommended stress lev-
els, and should be used only as starting points to
FAILURE DETECTION develop unique optimum profiles for a particular con-
figuration. If the response levels in Tables 3-3 and 3 -4
Measurements During Thermal Cycling exceed the design capability of the unit and/or system,
the contractor should submit appropriate rationale
Two approaches exist to monitoring equipment dur- with supporting data to the Government for a waiver
ingthermal cycling. The firstapproachutilizes periodic or deviation.
measurement In this approach, limited performance The most significant conclusion of ten years of
measurements are necessary prior to and at the end of random vibration screening is that the excitation must
ESS. These performance measurements may be made be tailored to the response experienced by the compo-
on the first and last cycle. Additional measurements nents of the unit under test The selection of stress
may be taken at other cycles, if desired. Each measure- levels must be based on available data and structural
ment should be made at the hot and cold operating design due to the potential for highly resonant mem-
extremes. bers, as well as the existence of vibration sensitive
The second approach calls for continuous monitor- electro-optical and electromechanical devices. To avoid
ing of equipment operation during the "cold-to-hot" potential fatigue or peak level damage due to reso-
transition and the "hot" dwell portion of each cycle. nances, some level reduction of the input spectrum
may be done at points of severe resonantfrequencies
Measurements During Random Vibration which result in amplification ofthe applied stress level
by a factor of 6 dB or more. These resonances would
The strong argument for monitoring equipment dur- be obtained from data accumulated during develop-
ing vibration screens is that the resulting movement of ment tests, or by conducting a low-level sine sweep.

12
 TABLE3-3. BASELINE VIBRATION PROFILE

CHARACTERISTIC LEVEL OF ASSEMBLY

PWA1 UNIT SYSTEM

OVERALL RESPONSE LEVEL 2 6gRMs 6gRMS 6gRMs

FREQUENCY 3 20 - 2000Hz 20 - 2000Hz 20 - 2000Hz

AXES 4
(SEQUENTIALLY OR SIMULTANEOUS) 3 3 3

DURATION

- AXES SEQUENTIALLY 10 MINUTES/AXIS 10 MINUTES/AXIS lOMINUTES'AXIS

- AXES SIMULTANEOUSLY 10 MINUTES 10 MINUTES 10 MINUTES

PRODUCT CONDITION UNPOWERED POWERED, MONITORED POWERED, MONITORED

(POWERED IF PURCHASED AS
AN END ITEM DELIVERABLE OR
ASA SPARE)

NOTES:

PURE RANDOM VIBRATION OR QUASI-RANDOM VIBRATION ARE CONSIDERED ACCEPTABLE FORMS OF VIBRATION FOR THE PURPOSE OF STRESS
SCREENING. THE OBJECTIVE IS TO ACHIEVE A BROAD-BAND EXCITATION.

1. WHEN RANDOM VIBRATION IS APPLIED AT THE UNIT LEVEL. IT MAY NOT BE COST EFFECTIVE ATTHE PWA LEVEL HOWEVER, PWAs MANUFACTURED
AS END ITEM DELIVERABLES OR SPARES MAY REQUIRE SCREENING USING RANDOM VIBRATION AS A STIMULUS. HOWEVER, ATTHE SYSTEM LEVEL,
WHEN A RESPONSE SURVEY INDICATESTHATTHE MOST SENSITIVE PWA IS DRIVING THE PROFILE IN A MANNER THAT CAUSES SOME PWAs TO
EXPERIENCE A RELATIVELY BENIGN SCREEN. THAT PWA SHOULD BE SCREENED INDIVIDUALLY. EACH PWA SCREENED SEPARATELY SHOULD HAVE
ITS OWN PRORLE DETERMINED FROM A VIBRATION RESPONSE SURVEY.

2. THE PREFERRED POWER SPECTRAL DENSTTY FOR Ggpus CONSISTS OF 0.04 g2/Hz FROM 80 TO 350 Hz WITH A 3 (»OCTAVE ROLLOFF
FROM 80TO20HZANDA3 dB/OCTAVE ROLLOFF FROM 350 TO 2000 Hz.

3. VIBRATION INPUT PRORLE FOR EACH SPECIRC APPLICATION SHOULD BE DETERMINED BY VIBRATION RESPONSE SURVEYS WHICH IDENTIFY THE
CORRELATION BETWEEN INPUT AND STRUCTURAL RESPONSES. HIGHER FREQUENCIES ARE USUALLY SIGNIRCANTLY ATTENUATED AT HIGHER
LEVELS OF ASSEMBLY.

4. SINGLE AXIS OR TWO AXIS VIBRATION MAY BE ACCEPTABLE IF DATA SHOWS MINIMAL FLAW DETECTION IN THE OTHER AXES.

Notching (but not notching out) may be permitted with OPTIMIZING/TAILORING OF ESS
government approval, but should be the exception, not
the general rule. Where warranted, temporary stiffen- The Environmental Stress Screening Plan should
ing of the unit should also be considered to prevent allowthemanufacturerto optimize a particular profile
overstressing during the stress screen. The design as needed, with government approval. The flexibility
agency may find that the most economic approach is to change the screens as new parts, vendors, assem-
a minor design change to provide permanent stiffen- blies and new or alternate materials arise is also
ing. Whether temporary or permanent, the stiffening essential for a good ESS program.
should be done in a manner which achieves the desired For any given part or production process, there
flat response throughout the unit being screened. exists a level of ESS stress that is optimal, i.e., maxi-
The temperature cycling screens also have to be mizes the likelihood of flaw detection without
tailored to each specific equipment and are equipment significant degradation of the unit undergoing ESS.
unique. Differences in components, materials and Determining this optimal level is normally referred to
heat dissipationlead to variations in the thermal stresses as the optimization of a profile for an individual piece
throughout the item. of equipment.

13
 TABLE 3-4. BASELINE THERMAL CYCLE PROFILE

1
CHARACTERISTIC LEVEL OF ASSEMBLY

PWA2 UNIT3 SYSTEM

TEMPERATURE RANGE OF -50°C TO +75°C -40<C TO +70°C -40"C TO +60°C

PRODUCT

TEMPERATURE RATE OF CHANGE 15°C/MINUTE 10°CMNUTE 10°Cdv1INUTE

OF PRODUCT4 TO TO TO
20tCMNUTE 20°CMINUTE 15°C/MINUTE

STABILIZATION CRITERION STABILIZATION HAS OCCURRED WHEN THE TEMPERATURE OF THE

SLOWEST-RESPONDING ELEMENT IN THE PRODUCT BEING SCREENED IS
WITHIN 15% OF THE SPECIFIED HIGH AND LOW TEMPERATURE EXTREMES.
LARGE MAGNETIC PARTS SHOULD BE AVOIDED WHEN DETERMINING THAT
STABILIZATION HAS OCCURRED.4

SOAK TIME OF PRODUCT AT TEMPERATURE EXTREMES AFTER STABILIZATION

- IF UNMONITORED 5 MINUTES 5 MINUTES 5 MINUTES

- IF MONITORED LONG ENOUGH TO PERFORM FUNCTIONAL TESTING

NUMBER OF CYCLES 20 TO 40 12 TO 20 12 TO 20

PRODUCT CONDITION 5 UNPOWERED/POWERED POWERED, MONITORED POWERED, MONITORED

NOTES

1. ALL TEMPERATURE PARAMETERS PERTAIN TO THE TEMPERATURE OF THE UNIT BEING SCREENED AND NOT THE CHAMBER AIR TEMPERATURE.
THE TEMPERATURE PARAMETERS OF THE UNIT BEING SCREENED ARE USUALLY DETERMINED BY THERMOCOUPLES PLACED AT VARIOUS POINTS
ON THE UNIT BEING SCREENED.

2. PWA GUIDELINES APPLY TO INDIVIDUAL PWAs AND TO MODULES, SUCH AS FLOW-THROUGH ELECTRONIC MODULES CONSISTING OF ONE OR TWO
PWAs BONDED TO A HEAT EXCHANGER.

3. UNIT GUIDELINES APPLY TO ELECTRONIC BOXES AND TO COMPLEX MODULES CONSISTING OF MORE THAN ONE SMALLER ELECTRONIC MODULE

4. rT IS UP TO THE DESIGNER OF THE SCREENING PROFILE TO DECIDE WHICH ELEMENTS OF THE HARDWARE (PARTS. SOLDER JOINTS. PWAs,
CONNECTORS, ETC.) MUST BE SUBJECTED TO THE EXTREME TEMPERATURES IN THE THERMAL CYCLE THE TEMPERATURE HISTORIES OF THE
VARIOUS ELEMENTS IN THE HARDWARE BEING SCREENED ARE DETERMINED BY MEANS OF A THERMAL SURVEY.

5. POWER IS APPLIED DURING THE LOW TO HIGH TEMPERATURE EXCURSION AND REMAINS ON UNTIL THE TEMPERATURE HAS STABILIZED ATTHE
HIGH TEMPERATURE POWER IS TURNED OFF ON THE HIGH TO LOW TEMPERATURE EXCURSION UNTIL STABILIZATION AT THE LOW TEMPERATURE
POWER IS ALSO TURNED ON AND OFF A MINIMUM OF THREE TIMES AT TEMPERATURE EXTREMES ON EACH CYCLE

ESS tailoring (the modification of ESS parameters to more of these techniques is usually predicated on such
fit specific hardware), if not planned and done properly, things as availability of screening equipment or cost
could be a major consumer of resources. Experience of procurement, architecture of equipment to be
with similar hardware can be helpful in setting initial screened and type of manufacturing defects expected,
tailoring levels leading to a rough approximation of and maturity of design and manufacturing processes.
optimal parameters. However, a true optimization is Trade-offs are needed because the payoff between
likely to require an extensive, carefully planned effort. "reasonably good" and "optimal" ESS parameters
Recommended tailoring techniques are given in may not be commensurate with the costs of finding the
Sections 4 and 5 for vibration screens and thermal optimal profile.
cycling screens, respectively. These are not the only Some specific engineering considerations in deter-
techniques available but are recognized throughout mining optimal ESS stress levels and making a sound
the industry as viable approaches for developing an engineering decision that tends to be on the conserva-
acceptable profile. The selection and use of one or tive side (i.e., no overstressing) are as follows:

14
 • Differences in physical characteristics such as Parts Rescreening and Quality
thermal inertia, thermal conductivity, mechanical
coupling, and mechanical resonant frequencies Poor quality piece parts play havoc with printed
assure that differently configured assemblies will wiring assembly (PWA) yields, with a resultant in-
respond differently to identical thermal and crease in assembly rework, cost and scrappage. Current
vibrational inputs. guidelines being implemented by some Services call
for 100% rescreening of microcircuits and semicon-
• Stress profiles should be defined in terms of
ductors by Original EquipmentManuf acturers (OEM)
responses rather than input A uniform level of at receiving inspection. This is normally continued
stress may not be achieved throughout the unit, until a quality level of less than 100 defective parts per
because all units are not generally internally
1,000,000 parts shipped can be demonstrated. The
homogeneous. The response can be specified and emphasis is on vendor process control to improve
measured at only a few points, so it will still differ quality of parts to an acceptable level rather than OEM
locally within differently configured assemblies.
rescreening. For information on parts rescreening and
quality, see References B.2-4 and B.l-19.
RELATIONSHIPS OF ESS TO OTHER
ACTIVITIES IN PRODUCT DEVELOPMENT Test, Analyze and Fix Programs
AND PRODUCTION
Since the primary purpose of ESS is to precipitate TAAF reliability growth testing programs are used
latent problems associated with the manufacturing extensively by the Services to identify and correct
processes, its effective use is predicated on good design deficiencies on new systems while still in the
design with quality parts. Historically, ESS results engineering & manufacturing development phase. As
show that failures due to workmanship are approxi- mentioned in Reference B.l-20, ESS should precede
mately two thirds of the total with the other third due formal TAAF testing. This helps to minimize the
to bad parts and poor design. occurrence of failures unrelated to design inadequa-
The ESS effort is expensive initially, particularly cies. Unrelated failures tend to retard the TAAFprocess,
considering the associated costs of the capital invest- lengthen its duration, and increase its total cost
ment Additional recurring cost factors that will add to
the overall cost include the utilities, failure analysis Reliability Demonstration and Production
and corrective actions that go along with the associ- Reliability Acceptance Testing
ated FRACAS program and all the labor necessary to
control the ESS program. All reliability predictions, demonstrations or tests
The ESS effort will be much more cost effective if are related to the system design and quality of parts
it is not loaded down with failures due to an immature used and do not consider workmanship or process
design and inferior parts. On the other hand, ESS is a deficiencies. Therefore, ESS is a necessary prerequi-
major cost avoidance factor in manufacturing because site for success in any reliability quantification based
the production process can be optimized, resulting in: on failures and operating time. The failures that occur
• Lessteardown during ESS are not counted in subsequent reliability
demonstrations but are input to aFRACAS program to
• Less troubleshooting time prevent reoccurrence. SeeReferencesB.1-9 andB.1-10.
• Less failure reporting and corrective action
• Less repair time Failure Reporting and Corrective Action System
• Less inspection time
• Less reassembly time One of the bestpractices in successful system devel-
opment efforts is the proper implementation of a
• Improved production personnel efficiency and FRACAS. AsdefinedinMIL-STD-1629,FRACASis
proficiency a"closed-loop system forinitiating reports, analyzing
• More efficient utilization of production facilities failures, and feeding back corrective actions into the

15
 design, manufacturing and test processes." Thus, ESS
MOUNTING SCHEMES
is an essential tie to the design and manufacturing
processes during development and to statistical pro- Even with relatively simple configurations and small
cess control (SPQ of the manufacturing processes module sizes, poorly designed mounting fixtures can
during production and depot repair. severely distort the applied vibration spectrum and
even cause unwanted failures due to structural reso-
SAMPLING VS 100% SCREENING nances. Each vibration screen setup should ensure that
the stress excitation is evenly applied to the product
When an item has been in production for some time, throughout the spectrum. Enough problems are en-
manufacturing processes and purchased parts may countered within the product without confounding the
have reached a steady state and be well controlled. issue with resonances in the fixture. For example,
Undertheseconditions.ESSwillnolongerbe precipi- fixture resonances and cost were countered in one
tating a significant number of failures. At this point, it program by suspending the product on "bungee" cords
can be argued that ESS is no longerproductive and that and using tri-axial excitation applied at the comer of
resources could be conserved by discontinuing ESS. If the product
it can be demonstrated that the decline in ESS failures Many temporary schemes can be used to damp
is indeed due to improvements, and notto manufactur- excessive resonances within the product. These
ing changes that make the ESS conditions ineffective, schemes include clamping, strapping or supporting
suspension of 100 percent ESS may be considered. the resonating area only for the duration of the vibra-
However, monitoring should be instituted to make tion screen. Usually the amount of damping can be
sure that the improvements remain effective. The best adjusted to obtain the desired responses.
way to accomplish this is to develop a sampling plan,
with reversion to 100 percent ESS on evidence of loss
of process control. One hundred percent ESS also PERFORMANCE MONITORING AND
should resume when processes, parts or sources are POWERED VERSUS UNP0WERED
changed and after production breaks or new product CONSIDERATIONS
introduction. In developing a screening program, an important
In most military contracts the production quantities consideration is whether the product should be pow-
are not sufficient to justify the effort necessary to go ered or unpowered, monitored orunmonitored. Unless
from 100% screening to a sampling procedure. See they are the end items, PWAs are usually unpowered
Reference B.l-12. because they aren't used as stand-alone items in the
operational environment. In addition, appropriate
EQUIPMENT CONFIGURATION screen equipment is usually not available to function-
ally monitor PWAs during the screening process. On
If there are many small and different modules in the the other hand, units and systems should be powered
equipment, the cost of vibration fixtures for these and monitored because they usually function as stand-
modules may be prohibitive, especially if each is alone items and appropriate test equipment is usually
powered and monitored. A compromise, in this case, available to functionally monitor them.
may be to do power-off thermal cycling only at the
module level and do both thermal cycling and random During Thermal Screening
vibration at the next higher level.
Conversely, if some equipments or cabinets are odd During thermal stress screening, whether perfor-
shaped or have heavy cantilevered components, for mance monitoring should be required and/or when
example, then it may be more cost effective to do only power should be applied are primarily determined by
thermal cycling at this level and do both stress screens two factors:
at a lower level of assembly. It is essential that these • Without performance monitoring, intermittent
analyses result in a cost effective program to precipi- failures may go undetected (this is an argument
tate manufacturing defects. for performance monitoring with power applied).

16
 • With power applied, the parts may not be able to defects won't be precipitated since the random
be cycled over a large temperature range without vibration screen is of limited duration.
overstressing some parts (this is an argument for
unpowered equipment). CHAMBER AIR FLOW
The availability of electrical test equipment is tradi- CHARACTERISTICS
tionally limited, and conflicts are generated between When any item is subjected to thermal cycling, the
screening and bench test operations. In addition, sched- temperature of the item lags that of the chamber air
ules may be affected by the need to move and set up because of thermal inertia and imperfect heat transfer.
test equipment at each different location. If all of the The thermal lag, i.e., the difference between the cham-
failures that occurred were "hard failures" (i.e., fail-
ber air and hardware temperatures, increases with
ures mat stay failed once they occur), performance increasing equipment thermal inertia and with de-
monitoring might not be necessary. Unfortunately, creasing air speed. The thermal lag is greatest for
many failures that occur in electronic hardware are heavy assemblies and for low speed air cycled at high
intermittent failures and only occur while thermal rates of temperature change.
stress is being applied. If the chamber air temperature rate of change is too
Performance monitoring should be done at the low- high, the dwell time too short, and/or the chamber air
est temperature limit and at the upper temperature too slow, the part temperatures will not attain the
limit of each thermal cycle. Monitoring at these tem- chamber air temperature extremes, resulting in a less
perature limits will detect intermittent defects that effective screen.
would not show up at room temperature. Power need In thermal stress screening, the rate of change of
not be applied during the entire thermal screen. Rather,
temperature is as important as the temperature ex-
it can be turned off during the cooling portion of the
tremes. The fasterthe rate ofchange, the more effective
thermal cycle until temperature has stabilized at the
the temperature stress screen. But it is the individual
low temperature. It is desirable to monitor perfor-
components that must experience a particular rate of
mance while power is applied during the cold to hot
change of temperature and temperature extremes. To
ramp. The degree of monitoring needs careful study
attain the appropriate temperature rate of change and
regarding cost effectiveness. Any attempt to monitor
temperature extremes of the item being screened,
intermittent shutdowns for as short a period as 2 to 3
there are several things that the screen designer may be
milliseconds may be very expensive.
able to do:
During Random Vibration • Allow the ESS chamber to "overshoot" the tem-
perature parameters. Overshooting is a method of
Industry has developed the following information achieving an increased temperature rate of change
about power on/poweroff random vibration screening: and higher/lower temperature extremes when the
chamber air temperature exceeds the upper and
• POWER OFFis of some value. When power is not
lower screening temperature limits for a con-
applied, approximately 50% of the defects are not
trolled period of time. Controlled overshooting is
precipitated to failure and all of the intermittent
permissible and encouraged as an excellentmethod
failures are not identified.
of achieving higher temperature rates of change,
. POWER ON, OUTPUTS INACTIVE is of greater thereby increasing screen effectiveness. To avoid
value. When power is on, but the hardware is not overstress at temperature extremes, the tempera-
operating, about 70% - 80% of the defects are ture of (or immediately adjacent to) the part with
stimulated to failure. the smallest thermal mass should be monitored
with thermocouples, if practical.
• POWER ON, OUTPUTS FULLY ONis of mostvalue
in that all latent and intermittent defects are stimu- • If practical, remove the protective covers of the
lated if there is an effectively designed random equipment, thus allowing the chamber air flow to
vibration screen. However, all random vibration more easily reach the individual components.

17
 • Install an air circulating system. In many units, introducing workmanship flaws. Such assessments
the electronic parts are densely packaged, thus are appropriately made through Corrective Action
increasing the thermal mass of the unit As the Board/Failure Review Board actions.
thermal mass increases, the air flow becomes
more restricted. To compensate for this, an air 3.2 GUIDANCE SUMMARY
circulating system (e.g., a fan) can be installed to
direct the air to the areas of the unit with the The following summarizes the ESS engineering
highest thermal mass, thus causing the compo- guidance provided in this Section:
nents to experience a much greater temperature • Identify the nature of anticipated defects for unit
rate of change. design and manufacturing processes.
• PWAs and subassemblies which are not confor- • Select appropriate levels of assembly, e.g., printed
mally coated may suffer damage or intermittent wiring assembly, assembly, system, etc., at which
operation due to condensation in the chamber. ESS should be performed.
Consideration should be given to using an air - Review product design.
drying system or some othermeans ofminimizing
- Evaluate repair cost at various levels against
this condensation.
fixture and ESS costs, includingteardown, repair,
checkout, reassembly and the potential of
REPEATED SCREENING introducing new defects.
Repeated application of screens after correction of • Develop and finalize the ESS profile:
ESS flaws can very easily begin to use up significant
- Review available in-house and industry-wide
useful life and to initiate rather than precipitate flaws. data relative to the design of screening profiles
To avoid such counter-productive screening, the fol-
for comparable equipment.
lowing guidelines are recommended:
- Review product design information to identify
• After repair of failure during first operating vibra- any thermal characteristics of mechanical
tion screen, complete remaining duration of screen, resonances/weakness which could impact detail
or five minutes, whichever is greater. of screening profiles.
• After repair of failure during first non-operating -Tailor and finalize the temperature cycling
vibration screen, repeat screen as a confidence screen, at each level of assembly selected, for
check at full level and 50% duration. temperature limits, product rate of temperature
• After subsequent repairs and/or modifications, change, number of temperature cycles, and
repeat original screen at -3 dB level (70% g,^) for whether monitored during screen.
50% duration. - Tailor and finalize the random vibration screen,
• Do not exceed five vibration exposures. at eachlevel of assembly selected, for spectrum,
eve
• If failure is detected and repaired during the initial SRMS l l> number of axes, true random or
thermal cycling screen, the balance of the cycles quasi-random, and whether the product is moni-
scheduled, or a minimum of three, should be run. tored during screen.
• After subsequent repairs and/or modifications, - Optimize or modify the ESS profiles based on
run one complete thermal cycling screen. results from the screens and operational use.
The guidelines above should be used in conjunction • Assess the timeliness and comprehensiveness of
with alerting Government and contractor program the FRACAS and assure that the corrective action
managers and an assessment of the appropriate amount process for any inadequate manufacturing pro-
of rescreening which takes into account the nature of cesses has been extended back to the OEM.
the repair/modification, the amount of teardown, • Fmalizetheproceduresgovemingthereapplication
rework and reassembly involved, and the chance for of ESS after correction of ESS related flaws.

18
SECTION 4
VIBRATION SCREEN DEVELOPMENT

There are several viable methods for developing a The baseline vibration profile in Table 3-3 recom-
starting profile forvibration stress screening. Starting mends 3 axes vibration. However, screens developed
emphasizes that developing a screen is a dynamic using either Method A, B or C should be done in the
process. The effectiveness of any screen should be critical axis (usually perpendicular to the plane of the
evaluated by engineering analysis of the equipment printed wiring assemblies) first, with similar screens
and the expected flaws, using factory and field failure developed for the second and third axes. This proce-
data, and the failure history of the equipment during dure may eliminate vibration in the second or third
and subsequent to the screen, adjusting the screen axis as being ineffective in screening out defects.
parameters as the'screen matures. Where strong coupling exists between axes, all but the
Four methods are described herein in order of de- critical axis may be eliminated with Government
scending analytical complexity. Selection depends on approval as not cost effective in screening out defects.
such factors as: 1) hardware availability, 2) number
and production rate of items to be screened, 3) avail- 4.1 METHOD A - VIBRATION SURVEY
ability ofvibration equipment (shakers, data acquisition
analysis, etc.), and 4) availability of experienced dy- This is the preferred method and has been used
namic test or screening personnel. Table 4-1 gives extensively. Two techniques are available: (1) a gen-
some general considerations for selection of an appro- eral technique based on recording and analyzing the
priate method. data obtained to develop the spectral responses through-

TABLE 4-1. VIBRATION SCREEN DEVELOPMENT CONSIDERATIONS

METHOD PRO CON

A. VIBRATION SURVEY TWO TECHNIQUES TO DETERMINE GENERAL SURVEY TECHNIQUE

SPECTRAL RESPONSES FOR TAILORING REQUIRES SPECTRAL ANALYSIS
EQUIPMENT

B. STEP-STRESS TESTS STRAIGHTFORWARD EMPIRICAL SOME RISK OF OVERSTRESS IF

METHOD IF PERFORMED BY DESIGN LIMITATION IS UNKNOWN
EXPERIENCED ENGINEERS
EQUIPMENT MAY BE STRESSED TO
MAY PROVIDE EQUIPMENT WITH DESTRUCTION
INCREASED DURABILITY
SUFFICIENT EXPENDABLE ASSETS
DEFINES ITEM DESIGN LIMITS MAY NOT BE AVAILABLE DURING
EARLY PRODUCT DEVELOPMENT
IDEAL FOR EXISTING AND DEVELOPING PHASE
TECHNOLOGY

C. FAULT REPLICATION TESTS GOOD SUPPLEMENT TO METHOD B LACK OF HARDWARE WITH

REPEATABLE FAILURE MODES

DIFFICULTY IN 'SEEDING" HARDWARE

REALISTICALLY

D. HERITAGE SCREEN MINIMUM DEVELOPMENT RESOURCES TRANSPARENT DISSIMILARITIES MAY

REQUIRED YIELD INADEQUATE OR DAMAGING
SCREEN

MINOR DESIGN CHANGES MAY

INVALIDATE PREVIOUS ESS

19
 out the unit being screened; and (2) a simplified The controller, control strategy, and the number and
technique wherein overall g^ level readings are location of control accelerometers should be the same
obtained at the different sites to determine if some as for the proposed screen.
components are either overstressed or understressed.
Excitation System
GENERAL TECHNIQUE
The excitation system used for the survey should be
The development of a random vibration stress screen the same as for the screen.
is predicated on tailoring the input to achieve an
acceptable response throughoutthe unit being screened. Fixturing
A vibration survey is the most logical and straightfor-
ward way to determine these responses. The spectral The fixture, slip-plate, and head expander used for
responses from selected accelerometer sites identify the survey should be the same as for the screen. It is
the frequencies where high responses or damping good practice to perform a vibration survey on the
occur. The input vibration level at appropriate fre- mounting fixture only prior to the item survey.
quencies can then be tailored to eliminate undesired
high or low responses. MEASUREMENT PHILOSOPHY
CONFIGURATION Selection of Measurement Locations

The vibration survey configuration should replicate In an exhaustive survey, vibration response would
the configuration for the proposed screen. be measured at each component, wire connection,
mounting screw, etc., within the item to be screened.
Item This clearly is neither feasible nor desirable. What is
desirable is to measure vibration responses at loca-
The item must be representative of the product to be tions throughout the item that are representative of
screened. It should be possible to mount accelerom- responses at a majority of the potential failure loca-
eters internally within the item. It should be permissible tions. Approximately 20 locations should suffice for
to accumulate vibration time on the item. mapping most items.
An example of an item is shown schematically in
Level Figure 4-1 to illustrate the selection of response loca-
tions. The item is an electronics card box with cable
The vibration survey should be conducted at an connectors and a time meter mounted on the front
input random vibration level of 2-Sgj^, which is 6 to panel, and transformers mounted on the rear panel.
10 dB below the baseline screening level of Table 3-3. There are 11 standard cards spread throughout the
A low level sine vibration sweep can also be used to box: four heavier, suffer cards are located in the center
obtain a very good picture of resonance responses and an encased, thick module is located at the rear. The
across the desired spectrum. cards and module have connectors on the bottom
which mate with the motherboard at the bottom of the
Strategy box.
The three measurement locations on the cards are
The survey should be performed for each input axis depicted in Figure 4-2. The locations indicated by "X"
or combination of input axes specified for the screen. are suggested for arectangularPWA with components
For instance, a screen performed by the sequential mounted uniformly over the surface, supported along
excitation of three orthogonal axes requires three sur- the short edges and a connector on the bottom. If the
veys. A screen performed as the combination of a dual top of the PWA is supported by compression of a
axes excitationandasingleaxisexcitationrequires two rubber gasket on the lid, the locations depicted by "O"
surveys. A triaxial input screen requires one survey. would perhaps be a better choice. A square PWA

20
FIGURE 4-1. EXAMPLE OF A SCREENABLE ITEM SHOWING POSSIBLE MEASUREMENT LOCATIONS

I I I I
c:
c TRANSFORMER
T

u cz
\ ENCASED MODULE

THREE LOCATIONS ON EACH OF THE FOUR CARDS INDICATED BY 12

ARROWS.
THREE LOCATIONS WITHIN THE ENCASED MODULE ON THE COMPONENT 3
MOUNTING SURFACES.
ONE LOCATION ON FRONT PANEL NEAR CONNECTORS AND TIME METER. 1
TWO LOCATIONS ON MOTHERBOARD 2
TWO LOCATIONS ON REAR PANEL AT DIAGONALLY OPPOSITE CORNERS 2
OF ONE OF THE TRANSFORMERS.

TOTAL 20

equally supported on each edge could be sufficiently eters could probably be used on the motherboard
mapped with two locations: one in the center and the and suffer PWAs. The standard PWAs normally
other at the middle of one edge. Obviously there are require the smallest, lightest accelerometers avail-
many location choices within this example, and within able so as to not alter the dynamic characteristics
other items that differ significantly from the example. and to fit available mounting space.
This illustrates mapping of the entire volume and
Triaxial Measurement
indicates that engineeringjudgmentmustbe exercised
in the selection of measurement locations. The acceleration in three orthogonal directions
Accelerometers must be known for each chosen measurement
location. This does not mandate a triaxial mea-
• Physical Characteristics
surement at each location. A measurement from
Accelerometers should be small enough that they another location may be substituted for one of the
can be mounted in the chosen location and light triaxial measurements if the response is judged to
enough that they do not alter the dynamic charac- be the same over the frequency range of interest
teristics of the item. In most surveys a mix of As an example, triaxial response at the three
accelerometer types can be used. In the example measurement locations depicted on the card in
shown in Figure 4-1, relatively large, heavy accel- Figure 4-2 can be acquired by using five acceler-
erometers couldbeusedtomeasuretheacceleration ometers. A single accelerometer is needed at each
input to the connectors/time meter on the front of the three measurement locations with the sen-
panel. Similar accelerometers could be used on sitive axis oriented perpendicular to the plane of
the rear panel at diagonally opposite comers of the the PWA. The in-plane response should be the
transformer. Medium size and weight accelerom- same for all locations on the PWA and can be

21
FIGURE 4-2. RESPONSE MEASUREMENT LOCATIONS ON RECTANGULAR PWA WITH BASE CONNECTOR

ii i i i i i i i i i i i i i i i i i i i i i i i n
O TOP SUPPORTED
X TOP NOT SUPPORTED

acquired by placing the two accelerometers wher- Data Acquisition Equipment

ever there is adequate space. Triaxial measure-
ments will be required only if the contractor has The data acquisition system, i.e., accelerometers,
equipment capable of triaxial excitation. signal conditioners, and recorder system, should have
sufficient dynamic range to observe and record the
• Installation
response accelerations. The system should be compat-
Accelerometers should measure the input to com- ible withinitself and with the data analysis equipment
ponents or parts, not the response of a particular
component or part In the example shown in Recorder Setup
Figure 4-1, this means placing accelerometers on
the PWAs, not the components, and on the front The recorder speed should be sufficient to obtain the
and rear panel, not the parts mounted to the panels. desired frequency response for the acquired data.
For the first data acquisition run in each survey, all
Data Acquisition control accelerometers should be recorded along with
the response accelerometers. For all remaining data
It is assumed that the control and response acquisition runs in each survey, one control acceler-
acceleration data will be recorded and played back to ometer should be recorded with the response
a spectrum analyzer for data analysis. Alternatively, if accelerometers. The control accelerometer should re-
the spectrum analyzer has enough data channels, the main the same for all remaining runs to validate
data analysis could be performed "on-line," obviating repeatability in case of questionable response data.
the need to record and later play back data for spectral
analysis. If a recorder is not readily available, or if the Documentation
number of available accelerometer channels islimited,
the survey can be accomplished in segments by Documentation for the data acquisition should in-
analyzing the response ofeach available accelerometer clude the following information:
and moving the accelerometer to another location or
• Screen identification
direction. In most cases this canbe done quite efficiently
with minimum impact to the overall survey. - program name

22
 - item name screening station response of each channel and the transfer function
between any two channels to be measured. Any dis-
- recorder crepancies that are found can be compensated for
- engineer during analysis.
-date Data Recording and Review
- excitation system
The minimum duration for recording of data should
• Channel information bethetimenecessary to calculate acceleration spectral
- accelerometer identification density (ASD) functions over the desired frequency
range, using 50 averages. This minimum time will
- accelerometer serial number vary, depending on the analysis block size and band-
- accelerometer sensitivity width, the number of channels processed
simultaneously, and the analyzer computational speed.
- charge amplifier gain The entire run should be recorded ifthe screen is a non-
- charge amplifier serial number stationary process. The data should be reviewed after
the run to confirm that the amplitudes are appropriate,
• Run information that the waveforms appear reasonable, and that the
- run identification data segmentis properly identified. The gain setting of
each channel should also be verified.
- frequency range and level of excitation

Calibration
DATA PROCESSING

The full scale g level of each channel should be The end result of the vibration survey should be a
estimated for each survey location prior to performing collection of ASD functions on a mass storage device
the data recording. This calculation or estimate will available for "massaging." ASD functions should be
significantly reduce the instrumentation error caused calculated for all control and response accelerometers.
by noise threshold or saturation.
A calibration signal, preferably a sine wave repre- Data Analysis Equipment
senting the full scale g level of the instrumentation,
should be placed on each tape data channel. The run It is recommended that the data processing be per-
identificationshouldnote the voltage level, equivalent formed by playing back the recorded data to a digital
g level, and frequency of the calibration signal. The Fourier spectrum analyzer. The analyzer should have
calibration should be recorded for at least two minutes the capability to calculate ASD functions, label the
after any changes in the patching of charge amplifiers functions, and store the functions and labels on a mass
to the recorder, and at any time that there is a question storage device such as disk or tape. Additionally, the
as to whether the input gains have been adjusted since analyzer should be able to retrieve a stored ASD,
the previous run. integrate the function over selected frequency ranges
It is also desirable for a broadband, approximately to obtain g^ values, and print the g^ values.
white noise, random signal to be recorded. The fre-
quency range of the noise signal should extend over Data Analysis Parameters
the frequency range of the excitation and its voltage
amplitude should be within the dynamic range of the ASD functions should be calculated with 50 aver-
recorder. This signal, coming from one source, should ages. An analysis bandwidth of approximately 5 Hz
be recorded simultaneously on all active data channels should be used for ASD calculation overthe frequency
at the beginning of each run for a period of one minute. range of 20 Hz to 2000 Hz. Alternatively, a constant
Record the true RMS voltage level of this signal percentage bandwidth analyzer may be used if the
during playback. This signal permits the frequency bandwidth does not exceed l/6th octave.

23
 TABLE4-2. DATA ANALYSIS LOG PARAMETERS

• PROGRAM NAME NUMBER OF AVERAGES

• UNIT NAME CHARGE AMPLIFIER GAIN
• DATE
RECORDER CHANNEL
• RUN IDENTIFICATION
MASS STORAGE DEVICE &
• BLOCK SIZE LOCATION NUMBER
• FREQUENCY RESOLUTION
MEASUREMENT
• FREQUENCY RANGE IDENTIFICATION

Documentation 7. Perform vibrationdry run(s) to fixture and patch

to the controller and data acquisition system.
Each ASD function should be stored with a unique
identifier. A data analysis log should record the run 8. Mountunitinfixture. Torque to specified values.
information and analysis parameters shown in 9. Attach remainder of response accelerometers
Table 4-2. and cables for this data run (attach accelerometers
and cables for all runs if available).
PROCEDURE
10. Patch response accelerometers for this run to
The following vibration survey procedure assumes data acquisitions system.
that data is recorded on analog tape and played back to 11. Tap check all accelerometers to verify that they
a spectrum analyzer for ASD calculation. The proce- are properly patched to the input of the tape
dure can be modified for use with an on-line spectral recorder and that all instrumentation functions
analysis system. properly.
The procedure also assumes that the excitation sys-
tem is an electrodynamic shaker. For other types of 12. Install all lids, covers, and unit cabling that will
excitation systems, not all steps will be relevant. be on during screening.

1. Record the calibration signal on all data channels 13. Perform vibration run, recording all data.
of the tape recorder. 14. Verify that the recorded data is valid before
2. Record the white noise on all data channels of proceeding to the next run.
the tape recorder. 15. Repeat steps 9 through 14 for remaining groups
3. Attachanyaccelerometersandcablesthatrequire of response accelerometers.
special treatment (disassembly of unit, 16. Repeatsteps4throughl5foradditionalsurveys,
cleanroom facilities, obstructions when installed if applicable.
in the fixture, etc.) to the unit
17. Analyze recorded data to obtain ASD functions.
4. Create or retrieve input specification on the Label and store functions on the mass storage
controller. device for later retrieval and "massaging."
5. Mountfixturetoshakertable. Torque to specified
values. Compare vibration survey response spectra against
allowable stress limit criteria applicable to the assembly
6. Mount control accelerometer(s) to fixture and under evaluation. Subsequent engineering analyses
patch to controller and data acquisition system. may result in appropriate hardware modifications to

24
remove vibration screening concerns. In addition, satisfactory screening results. The overall input level
tailoring of the input spectrum is a viable alternative is tailored to the product.
for reducing response maxima to within allowable As in Method A, the vibration survey test configu-
stress limits. However, because extensive tailoring ration should replicate the configuration for the
can adversely affect the ability to stimulate defects proposed screen. The test item must be representative
throughout the entire assembly, it should be viewed as of hardware to be screened. It should be permissible to
the exception, not the general rule. Where warranted, accumulate vibration time on the test hardware. The
temporary stiffening or damping of the assembly fixture, slip-plate, and head expander used for the
should be considered to eliminate the need fortailoring. survey should be the same as for the screen.

SIMPLIFIED TECHNIQUE PROCEDURE

The simplified vibration survey technique is a modi- Step-Stress tests proceed as follows:
fication ofthe general technique. The general technique 1. Use a broadband spectrum of 20 to 2000 Hz and
is based on recording and analyzing the data obtained an initial input level of 2 or Sgg^. Vibrate for 5
to develop the spectral responses throughout the unit minutes for this and each subsequent stress step.
being screened, but there are many situations where
neither the equipment nor the associated trained per- 2. Inspect and, if powered, operate the unit during
sonnel are available to do this. For these situations the and after each vibration.
general technique can be modified so that only overall 3. Analyze aUfaüures occurring during or after the
evel
SRMS * readings are obtained at the different accel- vibration tests to determine whether the prob-
erometer sites throughout the unit being screened. lem is design related or a latent flaw.
Comparing these overall gj^ values will determine if
some components are either overstressed or under- 4. Increase input by 2gRMS for successive vibration
stressed due to structural resonances or damping, tests until tolerance limit is reached. The toler-
respectively. There is some risk that responses pecu- ance limit occurs when the unitdoesnot function
liar to random vibration may be missed. properly or at all and flaws are being induced
If the mean RMS response derived from multiple into good hardware.
locations on an assembly are within +6, -3 dB of the 5. Establish the design capability or overstress
input excitation level, no tailoring may be required. region ofthe hardware (i.e., hardware whichhas
Since overall level is only a crude indication of spec- weaknesses removed) based on failure analysis.
tral response, if the responses for individual locations
6. Use one-half of the design capability (g^) as
differ appreciably from the mean RMS level, a vibra-
the initial screening level.
tion response survey should be conducted at an
excitation level of 6 to 10 dB below the baseline Ideally, the final vibration input screening level
screening level of Table 3-3 (2 to Sg^). should be greater than or equal to one-half the design
capability of the hardware and greater than the operat-
4.2 METHOD B - STEP-STRESS TESTS ing level or possibly even the design requirement.
Figure 4-3 depicts this concept,
Step-stress testing is an empirical procedure that can
be used when resources for elaborate surveys, record- 4.3 METHOD C - FAULT REPLICATION
ing, analysis and technical support are limited. Due to
the associated inaccuracies and risks, however, its use Fault replication increases the screening input level
must be approved by the government. until known faults in the unit being screened are
The step-stress approach determines the "tolerance precipitated. These known faults may be recognized
limit" or design capability of the hardware for the manufacturing problems or faults that have been de-
screen. By knowing this limit, a safe screening level liberately seeded. As in the case of Method B, use of
can be determined and changed as required to obtain this method must be approved by the government.

25
FIGURE 4-3. STEP-STRESS CONCEPT: RELATION OF ENVIRONMENTAL LEVELS

DESIGN CAPABILITY

SCREEN LEVEL
DESIGN REQUIREMENT

FIELD (OPERATING LEVEL)

FREQUENCY

Method C can be used independently to establish a • Intermittent switch

screening level or can be used in conjunction with
Method B. In either case, the steps involved are similar • Insufficient solder
to those listed for Method B except that the testing
stops at the input level necessary to precipitate the • Connector or plug partially unseated
known faults. • Nicked wires
A minimum of ten faults should be available for
replication to establish an effective screening level. • Fractured hardware
Ideally, the faults should be "hidden" or latent, mean-
ing that they would not be detected during functional 4.4 METHOD D - HERITAGE SCREEN
tests. Representative faults to be used for seeding are:
A heritage screen is a starting screen derived from
• Loose hardware recent successful screening experience on equipment
of comparable design and manufacture. Ideally, this
• Flawed components (transistors, relays, etc.) experience would have been based on Method A. A
heritage screen should only be considered if there are
• Cold solder joints data substantiating its effectiveness. Government ap-
• Nicked component mounting legs proval is required for use of Method D.
The baseline vibration profile in Table 3-3 is a
• Incorrect bonding heritage screen derived from general vibration screen-
ing experience. It can be used as a starting screen when
• Removed bonding it is not possible to perform a vibration survey.

26
SECTION 5
THERMAL CYCLING SCREEN DEVELOPMENT

The thermal screens in widest use today are (1) • Hardware temperature history
thermal cycling, (2) steady high temperature, and (3) - temperature range
thermal shock. The thermal cycling screen is recog-
nized by the ES as being the most cost effective - temperature extremes
although the other two are used in some special situ- - stabilization criterion
ations.
The thermal screens described by the two methods - soak time at temperature extremes
herein, thermal survey and heritage, should be consid-
ered only as starting profiles. The effectiveness of any • Elements of the hardware to be subjected to this
screen should be evaluated by engineering analysis of temperature history. It is generally not cost effec-
the equipment and die expected flaws, using factory tive to perform a long cycle that subjects the
and field failure data, and die failure history of the entire mass of the item being screened to the
equipment during and subsequent to the screen, ad- temperature extremes. This is especially true
justing the screen parameters as the screen matures. with items (such as units, systems, and heavy
modules) having high thermal inertia. Accord-
5.1 METHOD A - THERMAL SURVEY ingly, the designer of the thermal cycling
screening profile must decide what elements
A thermal screen is characterized by: (such as parts, solder joints, PWA connectors)
are important to be subjected to the specified
• Cycle characteristics hardware temperature history. This decision is
- low temperature based on where in the assembly the defects are
expected to be precipitated by the screen. This
- high temperature could be, for example, in the semiconductor parts
- rate of change of temperature or in the PWA connectors.
- dwell times at temperature extremes • The method of heat transfer to the item being
• Equipment condition screened, such as:
- powered or unpowered - coolant circulated through a coldplate thermally
connected to the item
- monitored or unmonitored
- chamber air blown over the exterior of the item
• Number of cycles
- conditioning fluid circulated through the item
• Level of assembly at which screen is performed
With the aid of a thermal survey, Method A tailors To achieve a desired hardware thermal cycle, a
the cycle characteristics, equipment condition and certain temperature history of the heat transfer me-
number of cycles to the hardware to be screened. dium producing the thermal cycling is required. A
thermal survey evaluates the thermal response of
THERMAL SURVEY PURPOSE various elements in the hardware to changes in the
temperature of the heat transfer medium. The tem-
Developing a temperature cycling screening profile perature history of the heat transfer medium required
in terms of the thermal environment to which the to produce a desired hardware thermal response may
hardware is to be subjected establishes: then be developed.

27
THERMAL SURVEY GUIDELINES ware temperatures (a function of the velocity
and temperature of the heat transfer medium).
Ideally, a thermal survey should include the follow-
ing four steps. However, developing a computer 2. Construct a replica of the actual screening facil-
simulation may not always be practical, affordable, or ity. The thermal survey must be performed with a
necessary. setup that replicates the thermal characteristics
of the actual ESS setup in thefollowing respects:
1. Perform a computer simulation.
• Facility
Develop a detailed transient thermal model of the
heat transfer occurring in the thermal cycling • Mounting of hardware in chamber
screening setup. (This is different from thermal • Powering (if powered during ESS)
analysis or thermal mapping, which is the mea-
surement of the operating temperatures of the • Cooling (if powered and actively cooled during
deployed equipment in actual use.) The model ESS)
should be capable of predicting, as functions of 3. Instrument the important locations. Monitor and
the temperature history of the heat transfer me- record the following quantities:
dium, the temperature histories of the electronic
parts, the PWAs, and other elements in the hard- • Temperatures
ware targeted for removal of screenable defects. - hardware (the thermal analysis performed in
This model should simulate: the first step will aid the selection of hardware
• The dissipations of active parts in the hardware locations at which to measure temperature)
being screened (in the case of powered equip- - heat transfer medium, such as chamber air
ment) - coolants (if actively cooled)
• Themermalresistancesbetweenlccationswithin • Flow rates (velocities)
the hardware
- heat transfer medium, such as chamber air
• The thermal resistances between locations in
the hardware and the heat transfer medium - coolants (if actively cooled)
• The thermal capacitances of the elements of the • Power dissipations (if powered)
hardware The hardware temperature histories typically are
Use the model to perform parametric analyses of measured with thermocouples, which are point
the thermal responses of the elements in the hard- instruments (as distinguished from infrared ther-
ware being screened to changes in the temperature mography, with which a temperature map of an
of the heat transfer medium. The results of these area is obtained). Data are obtained only at the
analyses will be hardware and heat transfer me- preselected instrumentation locations, so it is im-
dium temperature histories. These analyses will: portantto instrumentthe important locations, with
the aid of the computer simulation. Thermocouples
• Identify the elements having the slowest ther- must be electrically isolated from measurement
mal response to the heat transfer medium. surfaces that are electrically "hot."
• Evaluate the temperature rate of change of the 4. Perform the experimental thermal survey by com-
heat transfer medium required to achieve the pleting the following three distinct procedures:
specified hardware temperature rate of change
(a function of the velocity of the heat transfer • The unit is soaked cold with power off until all
medium). thermocouples have stabilized at the test tem-
perature, then power is turned on for the soak
• Evaluate the dwell time of the heat transfer period, and then the rise to temperature at the
medium required for stabilization of the hard- required rate for the chamber. The chamber

28
 temperature is held at the high temperature until (the difference between the high and low tempera-
all thermocouples have reached the test tem- tures) dictates the thermal stress/strain to which the
perature. The data is used to establish the high hardware is subjected in each cycle. The number of
temperature stabilization time. cycles to failure varies inversely with the temperature
range: the wider the range, the earlier the failure. By
• A similar cycle is run to establish the cold optimizing the temperature extremes, the screening
temperature stabilization time. profile designer can minimize the number of cycles
• Several complete cycles are run to fine tune the required to precipitate flaws. Thus, the temperature
parameters to adjust for the shortened stabiliza- extremes also affect the cost of the screen.
tion times. The key to selecting the temperature extremes is to
stress the hardware adequately to precipitate flaws
In the same way as was" done analytically in the without damaging good hardware. In practice, tem-
computer simulation, measure the temperature perature ranges from aminimum of 90°Cto amaximum
histories as functions of the screening setup pa- of 180°C have been used. Minimum values are: 125°C
rameters. Perform at least three thermal cycles to for modules (usually -50°C to 75°C), 110°C for units
establish a thermal steady state. (usually -40°C to 70°Q and 100°C for systems (usu-
The results will be experimental plots used to estab- ally -40°C to 60°C). The following key factors should
lish the screening parameters required to achieve the be considered for the extreme values:
specified hardware temperature histories. The analy- • Storage temperature (high and low) limits of hard-
sis in the computer simulation should minimize the ware such as the materials in printed wiring
amount of iteration required in the laboratory to estab- assemblies
lish the screening setup parameters.
• Maximum turn-on and operating temperatures of
CYCLE CHARACTERISTICS electronic parts

In characterizing the thermal cycle it is important to RATE OF CHANGE OF TEMPERATURE

distinguish between the temperature histories of the
hardware elements and that of the chamber air. The The temperature rate of change affects the screening
hardware temperature histories determine the effec- effectiveness in a complicated way. It also affects the
tiveness of the screen whereas the chamber air duration and thus the cost of the screen.
temperature history is the controlling element To The physical effect of the rate of change of tempera-
achieve a desired hardware temperature history, a ture is quite complex. If a slab of material were heated
certain temperature history of the chamber air produc- and cooled uniformly, the thermal stresses and strains
ing the thermal cycling is required. would be independentofthe temperature rateofchange.
A thermal survey evaluates the thermal response of In thermal stress screening, however the heating/
various elements in the hardware to changes in the cooling is nonuniform because of:
temperature of the chamber air. The results of the
• Nonuniform heat transfer to the surface of the
thermal survey will be experimental plots of the ther-
hardware
mal responses, measured at critical elements of the
hardware, to changes in the temperature of the cham- • Thermal lags between the surface and interior of
ber air. The necessary temperature range and rate of the hardware
change of the chamber air can then be identified for a
• Nonuniform thermal inertia of the various por-
desired response.
tions of the hardware
TEMPERATURE EXTREMES Consequently, instantaneous temperature gradients
can exist throughout the hardware. These temperature
The temperature extremes in a thermal cycle affect gradients, and the resultant thermal stress/strains, in-
the effectiveness of the screen. The temperature range crease with increasing temperature rate of change.

29
 Consistent with this phenomenon, industry has FIGURE 5-2. TEMPERATURE HISTORIES WITH LOW
found that increasing the temperature rate of change CHAMBER AIR SPEED AND SHORT DWELL TIMES
increases the screening strength up to a point. For
example, the situation is more complicated for solder, RAMP RATE = 15°C/MINUTE
which creeps at temperatures encountered in thermal DWELL TIME = 5 MINUTES
— CHAMBER AIR AIR VELOCITY = 5 FPS
stress screening. Creep, which has been identified as
HARDWARE
the major cause of solder joint failure, requires time
to occur. If the temperature rate of change is too high,
the thermal stress screening profile may actually be
excessively benign for the purpose of precipitating
defective solder joints to failure. (If properly con-
ducted, environmental stress screening to precipitate
defective solder joints in a specific set of equipment
should have to be performed at only one level of
assembly.)
If the chamber air temperature rate of change is too
high, and/orif the dwell time is too short, and/orif the 20 40 60 80 100
chamber air is too slow, then the part temperatures TIME (MINUTES)
.
will not attain the chamber air temperature extremes.
The result can be an unduly benign screen. This is tailor chamber conditions to achieve the desired tem-
illustrated by comparing Figures 5-1 and 5-2. In the peratures and rates of change in the test items.
first case, the air is fast enough and the dwell time long The choice of temperature rate of change depends
enough to enable the parts to stabilize and soak at the on the nature of the hardware and the flaws expected.
temperature extremes. In the second case, in which A high temperature rate of change is expected to be the
the air speed is six times slower and the dwell time most effective forprecipitating flaws in such elements
one-sixth as long, the part temperatures do not stabi- as plated-through holes, whereas a slow rate of change
lize, and instead cycle over a much smaller range than with long dwells at high temperature is expected to be
does the chamber air temperature. However, ad- the most effective for precipitating flaws in solder
equate experimentation and analysis can be used to joints. In practice the temperature rate of change
varies from 10°CAnin to 20°C/min with the nominal
FIGURE 5-1. TEMPERATURE HISTORIES WITH HIGH values as follows:
CHAMBER AIR SPEED AND LONG DWELL TIMES
PWA Screening 15°C/min to 20°Cymin
RAMP RATE = 15°C/MINUTE
DWELL TIME = 30 MINUTES Unit Screening 10°C/min to 20°CAnin
AIR VELOCITY = 30 FPS System Screening 10°C/min to 15°CAnin
— CHAMBER AIR
HARDWARE
The screening strength does not increase indefi-
o
CO
UJ T~\I rf~ nitely with increasing temperature rate of change.
HI
'/ n
£50
UJ
o / A
i
i/ DWELL TIMES AT TEMPERATURE EXTREMES
Ul
tr
0- if The dwell time of the chamber air temperature
consists of two elements, as shown in Figure 5-3:
\
• ThetimeG^^requiredfortheparttemperaturesto
- , ,N
-50-

20 40
z^Li
60 80
i
100 120 140
stabilize
• The additional time (D2) required to "soak" the
TIME (MINUTES)
hardware at the temperature extremes

30
Stabilization Time most important factors are the thermal inertia of the
assembly being screened and the chamber air speed.
The stabilization time (DJ required for internal
components to reach the ultimate chamber tempera- Soak Time
ture (chamber set point) has to be determined by the
thermal survey. The choice of stabilization criterion The soak period (D2) serves two purposes. First, this
affects the duration and thus the cost of the screen. period allows solder to creep. The time required for
The recommended stabilization criterion is: stabili- solderto relax is on the order of 5 minutes. Second, for
zation has occurred when the temperatures of the screens in which the equipment is powered and moni-
slowest-responding performance-related elements in tored, the soak periods at the temperature extremes
the hardware being screened are within 15% (AT in enable functional testing to be performed to detect
Figure 5-3) of the ultimate temperatures. During the failures which do not manifest themselves at ambient
screening of unpowered equipment, the ultimate tem- temperature. The recommended values of the soak
peratures are lhe chamber air high andlow extremes as time (D2) are as follows:
shown in Figure 5-2. With powered screening, the • Unmonitored equipment: 5 minutes
hardware temperatures may have other values, de-
pending on the specifics of the equipment and the • Monitored equipment long enough for functional
setup. The designer of the profile must decide which testing to be performed or 5 minutes, whichever is
elements of the hardware being screened (excluding longer.
magnetics) are to be monitored.
Defining stabilization as the time required for the EQUIPMENT CONDITION
rate of change of the part temperatures to decrease to
some small specified value is notrecommended. Ther- Detection of failures induced by the environmental
mal analyses indicate that this criterion can result in stresses generally requires that the equipment be pow-
excessively long-duration and thus expensive screens. ered and monitored. Testing the equipment to detect
The stabilization time (Dj) for a specific screen, failures should be done during application of environ-
using the criterion recommended depends on the hard- mental stress screening, otherwise intermittent failures
ware being screened and the screening facility. The will go undetected. Testing only before or after stress-
ing results in high risk of letting the intermittent flaws
FIGURE 5-3. TYPICAL TEMPERATURE CYCLING remain.
PROFILES Thermal cycling differs from vibration in this re-
spect: The period of a vibration cycle is a small
KEY PARAMETERSTACTORS
fraction of a second and the duration of a vibration
screenisontheorderoflOminutes.Duringavibration
• THERMAL INERTIA OF HARDWARE
• TEMPERATURE RATE OF CHANGE screen, there is not enough time to fully test a complex
•CHAMBER AIR SPEED
AT
system. In contrast, the period of a thermal cycle is on
1 the orderofhours and the duration of a thermal cycling
TH
ir screen is on the order of several hours. In a thermal
IE
3
v>Kl ' |\
/ S ' ' x
\
screen, therefore, one can test the system at either or
both temperature extremes as well as at ambient tem-
<er perature.
z PARTS \\ When powered equipment is subjected to thermal
IS cycling, the situation is complex because of the tem-
j J \\
! ! ' £^\ perature rise produced by the dissipation (heat) of the
T
L ~ electronic parts. The relation between the operating
1 1 I part temperatures and the chamber air temperature
Klw-t- so™ -i depends on the specific equipment and the screening
TIME
parameters. In addition to the instantaneous 4hermal

31
 gradients occurring in screens of unpowered equip- In evaluating the effect on failure of the number of
ment, additional thermal gradients occur because of cycles, it is important to distinguish between fallout at
the flow of heat from the dissipating parts to the the point of screening and subsequent failures at
surroundings. higher levels of assembly and in the field. ESS takes
Some factors involved in deciding whether or not to life out of good and bad equipment although the
have the equipment operating are as follows: decrease in the useful life of good equipment is small
• A powered screen is more effective in precipitat- with a well designed screening profile. The number of
ing flaws than an unpowered screen. Powering failures occurring per cycle usually begins low, rap-
produces temperature gradients in the hardware idly increases, then decreases exponentially until
not present in unpowered equipment. The thermal stabilization. When stabilization occurs, usually an
stresses/strains resulting from these thermal gra- optimum number of cycles has been reached.
dients may precipitate flaws that escape in Thermal cycling produces thermal stresses which
unpowered screens. induce alternate expansion and contraction. The
stresses and strains are highest at flaws because each
• A powered and monitored screen may detect flaw creates a stress riser that allows the stress to
failures that escape in an unpowered screen (inter- precipitate a flaw (i.e., latent defect) to hard (i.e.,
mittent failures). Failures that do not manifest detectable) failure. The cyclic loading causes the
themselves in testing at ambient conditions may flaws to grow. Eventually they become so large that
show up in testing at high or low temperature or they cause a complete structural failure and thus an
during vibration. An example is a broken connec- electrical failure. For example, a cracked plated through
tion in which the pieces are touching just enough hole eventually cracks completely around and causes
to provide continuity in the absence of thermal/ an open circuit
vibration stresses. The lifetime of the product is governed by the
• A powered and monitored screen is more expen- number of cycles, that is, the number of stress/strain
sive than an unpowered screen. reversals. The number of cycles to failure is a decreas-
ing function of the stress/strain range per cycle, which
• A power-offscreen at the PWA level of assembly in turn is a monotonically increasing function of the
is often used as an effective screen for latent part temperature range per cycle. However, a properly
defects. However, it should only be considered if designed thermal screen will precipitate failures in
the PWA will see a powered screen at the next flawed items, while not consuming a significant por-
higher level of assembly. tion of the useful life of good items.
Although details will differ for any specific item to For solder, the physics of failures induced by ther-
be screened, the consensus of industry experience on mal cycling is more complex than for materials such as
the basis of technical and costtrade-off considerations aluminum and copper. The reason is that, at the tem-
is as follows: peratures encountered in electronics equipment, solder
creeps. Creep has been identified as the major cause of
SSEMBLY EQUIPMENT solderjointfailures.Soldercreepsatarate that increases
LEVEL CONDITION with increasing temperature. Consequently, the num-
Board Unpowered berof cycles to failure of solder joints depends on other
parameters as well as temperature range. The most
Unit Powered monitored severe thermal cycles forsolder are those in which creep
System Powered monitored has sufficient time to occur. However, a screen should
avoid unnecessarily inducing creep in solder joints.
NUMBER OF CYCLES Although the selection of the number of thermal
cycles is critical relative to the effectiveness and cost
As do the cycle characteristics, the choice of the of the screen, the procedure to do so is controversial.
number of cycles impacts the effectiveness and the What is recommended here is a practical empirical
duration and thus the cost of the screea approach instead of estimating the residual fault con-

32
tent of an item and a corresponding screening strength 4. Continue monitoring screen results to justify
necessary for an acceptable product. any other adjustments of screening cycles, ei-
The number of cycles varies with product complex- ther up or down, that may be warranted.
ity, design and process maturity and whether the other
thermal screen characteristics have been carefully
chosen. The recommended procedure for selecting the 5.2 METHOD B - HERITAGE SCREEN
number of cycles is:
Similar to the Heritage Screen for Random Vibra-
1. Be sure that the thermal survey and analyses tion discussed in Section 4, the Heritage Thermal
have been completed to identify the most appro- Screen would be one derived from past, successful,
priate values of temperature range (high and screening experience on equipment of comparable
low value), product and chamber temperature design and manufacture. Again, this should be an
rate-of-change, dwell times, and whether pow- iterative process where the fallout, or flaw precipita-
ered and monitored. tion results, are carefully monitored so that screening
strength can be adjusted to the most cost effective
2. Based on the above, select the initial number of
value as discussed in Method A.
cycles for the thermal screen from the following
ranges: A comparable approach to the Heritage Screen that
PWA 20 to 40 cycles is based on general thermal cycling results in the
Unit/System 12 to 20 cycles Baseline Thermal Screening profile given in Sec-
tion 4. TheBaselineapproachispresentedasastarting
3. Perfonnthermalscreenasplanned.Recordwhen screen to be used when it is not possible to use Method
failures occur, types of failures, and corrective A orno data are available for the Heritage screen. It is
actions taken to prevent reoccurrence. Plot fail- more important with the Baseline approach that the
ures as a function of temperature cycle. When results be monitored and the screening strength be
stabilization occurs in above plot, reduce the adjusted as necessary. Government approval is re-
screennumberof cycles to value corresponding quired for use of either the Heritage or the Baseline
to stabilization. Thermal Screen.

33
APPENDIX A
GLOSSARY & DEFINITIONS

A.1 GLOSSARY
ASD Acceleration Spectral Density
ESS Environmental Stress Screening
FRACAS Failure Reporting, Analysis and Corrective Action System
IC Integrated Circuit
IES Institute of Environmental Sciences
NDI Nondevelopment Item
IRIG Inter-Range Instrumentation Group (a U. S. Government Agency)
OEM Original Equipment Manufacturer
OTS Off the Shelf
PWA Printed Wiring Assembly
R&M Reliability and Maintainability
RAM Reliability, Availability and Maintainability
RFP Request for Proposal
RMS Root Mean Square
SRU Shop Replaceable Unit
SOW Statement of Work
SPC Statistical Process Control
TAAF Test, Analyze & Fix
TOP Technical Data Package

A.2 DEFINITIONS
Assembly A combination of parts joined together to perform a specific function.
Burn-In Bum-in is usually applied during production at the end item level only and consists of an
operational period for a specified number of hours with a specified failure-free period.
The operational conditions and environmental stresses in most cases attempt to simulate
field conditions and therefore usually are the same as the test conditions used for
demonstrating reliability. Bum-in is normally performed on 100 percent of the items in
each production lot
Design Capability The level of stress (thermal or mechanical) which an item is able to achieve or endure
without damage or significant reduction of its overall usable life.
Environmental Environmental stress screening of a product is a process which involves the application of
Stress Screening one or more specific types of environmental stresses for the purpose of precipitating to
hard failure, latent, intermittent, or incipient defects or flaws which would cause product

A-l
 failure in the use environment. The stress may be applied in combination or in sequence
on an accelerated basis but within product design capabilities.
Failure Mode The fundamental physical or chemical process responsible for a failure; the causative
agents of a failure, including circumstances during design, manufacture or use that may
lead to a failure.
Hermeticity The ability of a sealed item to remain impervious to outside contaminants.
Indenture Level Level of assembly; the highest indenture level is a system, the lowest is apart.
Infant Mortality Failures that occur early in the life of the unit.
Isolation The reduction in severity of response, force, or motion to input stimulus.
Latent Defect An inherent or induced weakness, not detectable by ordinary means, which will either be
precipitated to early failure under ESS conditions or eventually fail in the intended-use
environment.
Module A self-contained collection of chassis-mounted components and/or printed wiring
assemblies within one package which performs a specific function or group of functions,
and which is removable as a single package from an operating system.
Part Any identifiable item within the product which can be removed or repaired (e.g., discrete
semiconductor, resistor, integrated circuit, connector); used interchangeably with piece
part, component part, and device.
Parts Rescreening Usually refers to all microcircuits and semiconductors at receiving inspection being tested
to specification and environmental requirements.
Patent Defect An inherent or induced weakness which can be detected by inspection, functional test, or
other defined means without the need for stress screens.
Precipitation of The process of transforming a latent (undetected) defect into a patent (detected) defect
Defects through the application of stress screening.
Printed Wiring An assembly containing a group of interconnected components mounted on a circuit card
Assembly Comparable terminology includes printed circuit board and printed circuit assembly.
Screening Generally, a measure of the capability of a screen to precipitate latent defects to failure-
Effectiveness sometimes used specifically to mean screening strength.
Screening The probability that a specific screen will precipitate a latent defect to failure, given that a
Strength latent defect susceptible to the screen is present.
System A group of units interconnected or assembled to perform an overall function.
Transmissibility The ratio of output response to input motion.
Unit A group of modules interconnected or assembled to perform a specific function with a
system.

A-2
APPENDIX B
REFERENCES

B.1 MILITARY
1. Air Force Pamphlet 800-7, "USAF R&M 2000 Process"
This is the Air Force document whichforms the basis for the Air Force R&M program. ESS is given visibility
in a detailed appendix, with specific parameters for temperature cycling and random vibration included in
a chart titled "R&M 2000 Baseline Regimen."
2. Army Materiel Command (AMC) Regulation 702-25, "AMC Environmental Stress Screening Program"
This Army Regulation is the basis for ESS requirements in the Army. This regulation contains a Baseline
Regimen and requires that a FRACAS be implemented. Appendix A contains a baseline Statement of Work
to be used in Invitations for Bids, Requests for Proposal, and awarded contracts.
3. DoD 4245.7-M, 'Transition from Development to Production"
This document provides an excellent overview, in the Manufacturing Screening template, of the proper way
to use ESS. The Manufacturing Screening template also places strong emphasis on keeping ESS dynamic
and flexible through intelligent tailoring.
4. DoD-HDBK-344 (USAF) "Environmental Stress Screening of Electronic Equipment"
This handbook covers a variety of importantissues, including contractual aspects, planning for development
and production phase ESS, and incorporating results of different program test phases. This handbook also
contains a mathematical methodology for developing a screen.
5. DoD Directive 5000.1, "Defense Acquisition"
This directive describes the policies which govern defense acquisition by DoD components, the major
characteristics of the three decision making support systems affecting acquisition, and the acquisition
responsibilities of key officials and groups. Although the directive doesn't mention ESS, it does emphasize
developing reliable systems, which is the goal of ESS.
6. DoD Instruction 5000.2, "Defense Acquisition Management Policies and Procedures"
This instruction describes the procedures to be used for translating broadly stated mission needs into stable,
affordable, DoD acquisition programs. It emphasizes effective acquisition planning, improved communi-
cations with users, and aggressive risk management by both Government and industry. The reliability and
maintainability section (Part 6, Section C) requires that an aggressive ESS program be developed for
electronic equipment and applied to engineering development and production assets.
7. MIL-HDBK-338-1 A, "Electronic Reliability Design Handbook," Volume I
This handbook, which covers all aspects of reliability program planning and execution, has a section on
assembly-level ESS. The assembly-level ESS section contains a realistic approach for determining
appropriate screens based on thermal and vibration surveys. The need for tailoring, continuous reevaluation
of screen cost-effectiveness, and understanding root causes of failures are continuously emphasized.

B-l
 8. MIL-HDBK-727, "Design Guidance for Producibility"
This handbook, which covers all aspects of producibility, has a section on part screening. Although this tri-
service ESS Guidebook does not discuss Part Screening, an ESS practitioner who has the need for a detailed
discussion of Part Screening may refer to MIL-HDBK-727.
9. MIL-HDBK-781, "Reliability Test Methods, Plans, and Environments for Engineering Development,
Qualification and Production"
Although this is a handbook which was developed forReliability Testing, there is a section which describes
three methods for monitoring ESS (the Computed ESS Time Interval Method, the Graphical Method and
the Standard ESS Method).
10. MIL-STD-781, "Reliability Testing for Engineering Development, Qualification, and Production"
This Military Standard, although developed for Reliability Qualification Testing, Reliability Growth
Testing, etc., contains a Task for implementing Environmental Stress Screening in contracts.
11. MIL-STD-785, "Reliability Program for Systems and Equipment, Development and Production"
This Military Standard provides general guidance and specific tasks for reliability programs during the
development, production, and initial deployment of systems and equipment Task 301 of MIL-STD-785
provides specifics for specifying an ESS program in contracts.
12. MIL-STD-1235, "Single-and Multi-Level Continuous Sampling Procedures and Tables for Inspection by
Attributes"
This standard provides tables and procedures for applying five different types of continuous sampling plans
for inspection by attributes.
13. NAVMAT P-9492, "Navy Manufacturing Screening Program"
This is the "grandfather" of all current ESS standards and documents. This document contains a Baseline
ESS Regimen which has been extensively implemented on U. S. Navy programs. This document also
defines temperature change rates in terms of equipment response instead of chamber air conditions.
14. NAVSO P-6071, "Best Practices"
NAVSO P-6071, a companion to DoD 4245.7-M, offers a very useful executive-style summary of the
important issues associated with successfully using ESS. This summary is accompanied by a unique chart
that contrasts the traps and consequences of some current approaches with the potential benefits of applying
the Best Practices strategies.
15. RADC-TR-86-149, "Environmental Stress Screening"
This technical report, developed by Rome Laboratories, contains quantitative techniques for planning,
monitoring and controlling the cost effectiveness of stress screening programs for electronic equipment. A
method of estimating the number of defects remaining in the delivered product is also provided.
16. RADC-TR-87-225, "Improved Operational Readiness through Environmental Stress Screening"
This technical report, developed by Rome Laboratories, contains guidelines for the application of ESS to
field inventory hardware. Methods are presented for the selection of equipment for ESS application which
offer significant potential for operational readiness improvement and life cycle cost reduction.

B-2
 17. RADC-TR-90-269, "Quantitative Reliability Growth Factors for ESS"
This technical report, developed by Rome Laboratories, examines the measured field reliability improve-
ments resulting from multiple cycle ESS on avionics systems. For several systems, additional cycles ofESS
were applied. Through the comparison of the serialized field reliability records for those systems with and
without the additional ESS cycles, an assessment of the improvement in field MTBF resulting from ESS
was made.
18. Sacramento Air Logistics Center (SM-ALC), "Environmental Stress Screening Handbook"
This document was developed to provide program managers and engineers information on "how to set up
an ESS program." This handbook considers often overlooked administrative as well as technical concerns
such as previous contractor experience, decision criteria for ESS applicability, cost effectiveness in the
production process, and contractor development of appropriate ESS methodologies.
19. TE000-ABT-GTP-020A, "Environmental Stress Screening Requirements and Application Manual for Navy
Electronic Equipment"
This document, developed by Naval Sea Systems Command (NAVSEA), contains the basis for the
NAVSEA ESS program. It is intended for use by Navy program managers as the baseline minimum ESS
requirements for Statements of Work, and by design and manufacturing engineers and depot repair facilities
for implementation. It contains specific information on determining the natural frequencies and displace-
ments of PWAs. It also contains guidance on understanding the equipment's vibration and thermal
responses.
20. Tri-Service 'Technical Brief for TAAF Implementation"
This document was developed by the three services in an effort to have aunified understanding of the TAAF
process. This document provides in a single, concise source document the methods most likely to result in
a successful TAAF program

B.2 INDUSTRY
1. Institute of Environmental Sciences, "Environmental Stress Screening Guidelines, 1981"
This was the first ESS Guidelines document prepared by the Institute of Environmental Sciences. This
document, which was developed by ESS practitioners from industry and Government, provides technical
information on Piece-Part Screening as well as Module, Unit, and System Level Screening.
2. Institute of Environmental Sciences, "Environmental Stress Screening Guidelines for Assemblies, 1984,
1988"
This document was developed by the institute of Environmental Sciences as an update to the 1981
Guidelines document and provides detailed information on module, unit and system level screening.
3. Institute of Environmental Sciences, "Environmental Stress Screening Guidelines for Assemblies, 1990"
This document differs markedly from the 1981,1984, and 1988 IES Guidelines documents in that it is
procedural and tutorial in nature. This document incorporates the results of research conducted during the
1980s on the physical processes involved in ESS. The Guidelines include program management guidance,
cost-effectiveness analysis techniques, descriptions of vibration and thermal survey methodologies, and
ESS tailoring principles.

B-3
4. Institute of Environmental Sciences, "Environmental Stress Screening Guidelines for Parts, 1985"
"Iliis document addresses part screening methods, development of a screening or rescreening program, case
histories and screening results data. Also provided is an extensive treatment of integrated circuit packaging
technology, a subject which is important to anyone involved in piece part ESS.

B-4