I

NASA TECHNICAL NOTE N-

A S A-TN D-4416
L. /

LOAN C&Y: RETURN TO

AFWL (WLIL-2)
KIRTLAND AFB, N MEX

EXPERIMENTAL INVESTIGATION OF
REACTION CONTROL, STORABLE
BIPROPELLANT THRUSTORS

by R. Jdmes RoZZbzthZer
Lewis Research Center
CZeveZand, Ohio

NATIONAL AERONAUTICS A N D SPACE A D M I N I S T R A T I O N WASHINGTON, D. C. APRIL 1968

 TECH LIBRARY KAFB, NM

Ill1llllllllIIIIIIlllIlIll

EXPERIMENTAL INVESTIGATION O F REACTION CONTROL, STORABLE

BIPROPELLANT THRUSTORS

By R . J a m e s Rollbuhler

Lewis Research Center

Cleveland, Ohio

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION

For sale by the Clearinghouse for Federal Scientific and Technical Information
Springfield, Virginia 22151- CFSTI price $3.00
 EXPERIMENTAL INVESTIGATION OF REACTION CONTROL, STORABLE
BIPROPELLANT THRU STORS
by R. James Rollbuhler
Lewis Research Center

SUMMARY

In o r d e r to evaluate the performance of low-thrust, chemical bipropellant rocket

t h r u s t o r s for possible applications in spacecraft reaction control s y s t e m s , an experi-
mental investigation was conducted with readily available thrustors. The t h r u s t o r s were
designed for a thrust of less than 5 pounds force (22.2 N), combustion-chamber p r e s s u r e s
of 100 pounds p e r s q u a r e inch absolute (68.9 N/cm 2) o r l e s s , and for u s e with nitrogen
tetroxide - hydrazine- type propellants. A highly instrumented altitude test facility was
used to test each of the t h r u s t o r s over a range of test durations and oxidizer to fuel
ratios.
The maximum specific impulse obtained was 262 pounds force per pound m a s s p e r
second (2570 (N)(sec)/kg), and the maximum characteristic velocity was 5200 feet per
second (1588 m/sec). Total impulses were measured for run t i m e s as s h o r t as 10 milli-
seconds. The average value of the specific impulse, at minimum r u n t i m e s , was about
25 percent of the maximum specific impulse for the s a m e thrustor. Also reported are
response times and combustor-wall temperatures.

INTRODUCTION

Recent space missions have indicated the increasing need for spacecraft with the
ability to control closely their attitude and orbit position o r trajectory. These require-
ments have been met in most c a s e s by m a s s expulsion thrustors, such as chemical
rockets and cold-gas thrustors. For both attitude and orbit control, t h e r e have been
increasing requirements for the delivery of s m a l l impulse bits to control the spacecraft
attitude m o r e closely o r to make minor orbit or trajectory corrections. Small impulse
bits imply either s h o r t firing t i m e s o r low values of thrust. The reduction of firing time
is practically limited by the reaction time of propellant valves and the deteriorating
performance of the thrustor because of poor reaction kinetics and high thermodynamic
losses. Therefore, low values of impulse bit have been obtained principally by the use
of s m a l l e r thrustors.

I
 One f o r m of s m a l l t h r u s t o r that has found extensive applications is one employing
cold gas such as nitrogen as a m a s s expellant. Although the advantages of this thrustor
system lie in its simplicity, the specific impulse of the propellant is very low compared
with the conventional propellant combinations. Thus, for s y s t e m s with long mission
t i m e s requiring a substantial total impulse, the propellant weight often becomes prohibi-
tively large. Therefore, interest has increased in s m a l l t h r u s t o r s that employ the con-
ventional propellant combinations to produce a high specific impulse.
Presented herein a r e the results of experiments conducted to study the performance
of s e v e r a l t h r u s t o r s that use nitrogen tetroxide and a mixture of 50 percent unsymmetri-
cal dimethylhydrazine (UDMH) and 50 percent hydrazine as propellants and range in thrust
from 1/2 to 5 pounds (2.22 to 2 2 . 2 N). The two a s p e c t s of engine performance of prime
interest were the efficiency of the thrustor during normal steady-state operation and the
pulsing performance of the t h r u s t o r s in t e r m s of impulse bit as a function of firing time.
Certain other engine performance characteristics that were noted during the study in-
cluded the relation of thrust to combustion-chamber p r e s s u r e , combustion time lags,
temperature histories, and thrust buildup patterns during pulsing operation.

s YM B3 L s
2
thrustor nozzle throat cross-sectional a r e a , in. 2; cm
*t
C* characteristic exhaust velocity, ft/sec; m / s e c

cf thrust coefficient
F thrust, lb force; N
2
gravitational acceleration, 32.2 ft/sec2; 9 . 8 1 m/sec
gC
specific impulse , (lb force)(sec)/lb m a s s ; (N)(sec)/kg
ISP
It total impulse, (lb force)(sec); (N)(sec)
L" characteristic thrustor length, in. ; cm

O D oxidizer-fuel mass-flow ratio

pC
combustion chamber p r e s s u r e , psia; N/cm 2
total propellant weight flow r a t e , lb m a s s / s e c ; kg/sec
Wt

2
 APPARATUS AND PROCEDURE

Test Hardware

Seven different thrustor assemblies, complete with propellant control valves, and
four alternate valve assemblies w e r e obtained for this program. All were essentially
off-the-shelf designs. Because the only c r i t e r i a specified were the maximum thrust and
the type of propellants, the assemblies came in a variety of sizes, weights, dimensions,
and materials. Some had altitude expansion nozzles while others had sea-level expan-
sion nozzles. The rated combustion-chamber p r e s s u r e varied f r o m 35 to 100 pounds p e r
square inch absolute (24.1 to 68. 9 N/cm 2 ). The construction materials w e r e another
variable to consider f o r each of the thrustors.
The testing of two of the four alternate valve assemblies was discontinued because
they proved to be incompatible with the propellants being used.
Thrustor assemblies. - Each assembly consisted of a combustion chamber and a n
expansion nozzle, a propellant injector, and valve assemblies f o r on-off control of the
propellant flows. The seven thrustor assemblies and two of the alternate valve a s s e m -
blies a r e pictured in figure 1, and their pertinent physical data a r e listed in table I.
The thrustors and the individual valves a r e identified numerically. Thrustor 7
initially had valves that proved unsuitable; therefore, the assembly was modified so that
either valve unit 8 o r 9 could be integrated into the assembly. Valve unit 8 is shown
mounted on thrustor 7 , and valve unit 9 is shown both assembled and disassembled in
figure 1.
Injection techniques. - Several different injection patterns were used to flow the
propellants into the thrustor combustion chambers. Thrustor 1 injected a n oxidant and
fuel jet, each 180' a p a r t , on a wedge so that the resulting s p r a y s impinged on each other.
Thrustor 2 used three pairs of impinging jets with each pair being a fuel impinging on a n
oxidizer jet. Thrustors 3 and 6 each used a single pair of impinging jets. Thrustors 4
and 5 had two fuel jets impinging on a n axial oxidizer jet. Thrustor 7 utilized a spray
jet f o r the fuel and a s i m i l a r s p r a y jet for the oxidizer. The s p r a y s , 180' a p a r t , entered
the chamber through the side wall and w e r e directed toward the head of the chamber
rather than toward the combustor throat.
Combustion chambers
~- -
and expansion nozzles. - Thrustors 1 to 6a were designed to
operate over extended continuous durations (i. e. , longer than 10 s e c ) . Thrustors 3 to 5
had refractory metal combustion chambers and nozzles that w e r e coated with various
surface coatings to prevent high-temperature-metal oxidation. Thrustor 4 had a stainless-
s t e e l chamber and nozzle; f o r extended run durations, the interior of thrustor 4 was flame
sprayed with aluminum oxide. F o r runs of only a few seconds, heat-sink, r a t h e r than
radiation, cooling was satisfactory.

3
 I
I

C-74756 C-71929
(a) Thrustor 1. (b) Thrustor 2.

C- 71462 C-74754
(c) Thrustor 3. (d) Thrustor 4.
 c-74750

(e) Thrustor 5.

.,-
-.A

am
C-65863
 TABLE I. - PHYSICAL CHARACTERISTICS OF LOW THRUST, LIQUID BIPROPELLANT ROCKET THRUSTOR

'hrustor Rated Rated Combustor Nozzle Thrustor Combustion- Combustion- Nozzle

ombustion-
chamber
pressure
thrust
1 throat
diameter ratioI
expansion characteristic
length ~
chamber
volume
chamber
wall
material
coating injector
parts
current
a t dc of

1 75 51.6 0.5 22.2 0.082 2.081 1.6 20 50.8 None Torque motor 0.3

2 75 51.6 5.0 22.2 ~ .287 1 7 . 2 9 ' 1.46 20 50.8 None Torque motor ,4

3a 96 56.2 5.0 22.2 . 193 '4.91 40 6 15.2 1.97 TZM molyb- TZM molyb- Molybdenum
denum ' denum disilicide
alloy alloy

3b 96 56.2 5.0 22.2 . 193 4.91 40 11 27.9 . 37 6. 07 TZM molyb- TZM molyb- Molybdenum Solenoid .6
denum denum disilicide
alloy alloy

4 50 34.4 1.0 4.45 .115 2.92 60 38 96.5 .303 4.97 Tantalum- Tantalum- Silicide Torque motor .3
tungsten tungsten diffusion
alloy alloy

5 60 41.4 5.0 22.2 .242 6.15 40 23 58. 5 1.08 17.7 Tantalum- Tantalum- Silicide Torque motor .5
tungsten tungsten diffusion
alloy alloy
!
6a 35 24.1 1 , 3 2 1 8.16 40 8 20.3 .647 10.6 TZM molyb- TZM molyb- Molybdenum Solenoid .3
denum denum disilicide
alloy alloy

35 24. 1 . 3 5 1 8.92 20 8 20. 3 .78 12.8 304 Stain- 304 Stain- Aluminum Solenoid .3
l e s s steel l e s s steel oxide

35 24. 1 .367 9.33 1.6 8 20.3 .85 13.9 304 Stain- 304 Stain- Aluminum Solenoid .3
l e s s steel l e s s steel oxide

100 68. 9 \I If . 176 4.47 62 60 152 1.58 25.9 304 Stain- Tantalum- Aluminide Torque motor 3.2 to 0.3
l e s s steel tungsten
I
 Propellant valves. - The propellant control valves were fast-acting, 28-volt, torque
motor o r solenoid-actuated types. The solenoid type employs a 28-volt direct-current
magnetic a r m a t u r e to r e t r a c t a spring-loaded poppet f r o m the propellant-sealing orifice.
The torque valve employs a single electromagnetic actuation mechanism operating a
mechanical linkage that controls the opening and closing of the different propellant ori-
fices. The actuation mechanism is e x t e r i o r to the flow s t r e a m , and the linkage e n t e r s
the s t r e a m through a flexure sleeve o r joint. In most c a s e s , the valves could be removed
f r o m their respective t h r u s t o r s f o r servicing.
The valve operating power was supplied f r o m a 28-volt direct-current source. The
c u r r e n t consumed by each valve is given in table I. The value given is for a nominal
current while the valve w a s in a steady-state operating condition; initial valve opening
conditions resulted in a much higher (50 to 100 percent) current value.

Test Facilities

The thrustors and their associated propellant control valves were tested in a test
facility that handles the propellants nitrogen tetroxide and a mixture of 50 percent
hydrazine and 50 percent unsymmetrical dimethylhydrazine (UDMH). The thrustors
2
were fired in a vacuum environment of 10 millimeters of m e r c u r y (1333 N/m ) o r less.
They were mounted on the thrust stand inside a 1500-cubic foot (42. 5-cu. m. ) vacuum
tank (see fig. 2). Mechanical vacuum pumping was capable of maintaining fully expanded
nozzle gas flow of all the thrustors.

Figure 2. - Vacuum test facility.

7
 I I I

The thrust stand was especially designed for testing t h r u s t o r s that have a thrust
level of l e s s than 10 pounds (44.5 N). The thrustor was mounted horizontally on top of
the stand (see fig. 3). The stand is a compound pendulum that u s e s a closed-loop elec-

Figure 3. - T h r u s t stand in 1500-cubic-foot (42.5-cu-ml vacuum tank.

tronic force generating network in a null-balance s y s t e m for indicating thrust and total
impulse. A more complete description of this stand is given in the appendix. Physical
calibration of the stand indicated that the thrust data e r r o r was k l . 5 percent up to a f r e -
quency of 88 cps (88 Hz).
The fuel and oxidizer liquid propellant s y s t e m s w e r e directly behind the thrust stand
(see fig. 3). Each propellant s y s t e m consisted of a supply tank connected to the thrustor
control valve by means of tubing that contained propellant flowmeters. The flow r a t e of
the propellant was controlled by the supply tank p r e s s u r e .

Instrumentation

In each propellant s y s t e m , as many as t h r e e different flowmeters were placed in

s e r i e s to verify flow rate measurements. These types of flowmeters were orifice differ-
ential p r e s s u r e , turbine, and remote-reading rotameter. The calibration of these flow-
m e t e r s with actual propellants is described in the appendix. The orifice flowmeter gave
the most dependable data over a range of operating conditions. Its steady-state accuracy
w a s &2 percent of the true flow r a t e , and its transient response w a s 0 to 90 percent of the
maximum flow within 500 milliseconds.
Besides thrust and flow-rate data, propellant p r e s s u r e s and temperatures,

8
combustion-chamber p r e s s u r e , and combustor-wall temperatures w e r e also measured.
The p r e s s u r e s w e r e measured with strain-gage p r e s s u r e transducers, which w e r e cali-
brated against a precision gage s o that the test data accuracy was r t l percent. The fre-
quency response of the combustion-chamber p r e s s u r e instrumentation was in the range
of 150 to 175 hertz. The method of determining this range is presented in the appendix.
The temperatures w e r e measured with thermocouples, and the reading precision of their
data signals was *lof o r the propellants and *5O for the combustor-wall temperatures.
All the data signals w e r e recorded on a high-speed (1 to 80 in. o r 2. 5 to 200 c m of
r e c o r d p e r sec) oscillograph r e c o r d e r located in the facility control r o o m (shown i n
fig. 4). The frequency response of the oscillograph galvanometers varied f r o m 30 to
3000 hertz, depending on the requirements of the variable being recorded. Various prob-
l e m s i n verifying test temperatures are mentioned in the appendix.

Test Ope rations

The thrustor to be tested w a s rigidly mounted horizontally on top of the thrust stand.
The propellant system was connected to the thrustor, and the s y s t e m s were then p r e s s u r e
checked. The data s e n s o r s were calibrated, and the control system w a s checked out.
After the propellant tanks had been filled, the vacuum tank w a s sealed and the tank p r e s s u r e
was reduced to l e s s than 10 millimeters of mercury (1333 N/m 2).
T e s t operations w e r e controlled and recorded in a control room remote f r o m the
vacuum tank (see fig. 4). For each t e s t , the sequence of events was controlled automati-
cally. When the run button was pushed, the instrumentation r e c o r d e r began, and then the
thrustor propellant control valves were energized. The propellant flow rate to the t h r u s t o r
was regulated by varying the p r e s s u r e in the propellant supply tanks. The thrustor firing
time could be varied between 1 0 milliseconds and 600 seconds as could the time between
firings. After a p r e s e t single o r multiple firing, the r e c o r d e r automatically shut down
while the propellant lines and t h r u s t o r w e r e purged.
For the testing of a given thrustor, the p r e s s u r e in the propellant run tanks w a s v a r -
ied f r o m test to t e s t to find the optimum thrust at different oxidizer-fuel ratios. Although
the individual fuel and oxidizer flow rates were varied, their combined flow r a t e w a s kept
near a constant value. Once the optimum oxidizer-fuel ratio w a s determined, it was re-
tained while the t h r u s t o r on and off t i m e s were varied.
Of all the t h r u s t o r s tested, only thrustor 1 failed to deliver repeatable performance.
This thrustor was tested at both sea-level and vacuum-environment conditions, but
the data were e r r a t i c and fluctuated over a wide amplitude. The problem w a s apparently
the propellant injection technique; in spite of the 10-micron (10-7-m) propellant inlet f i l -
t e r s , the injection p o r t s slowly plugged during each test. The propellant flows decreased
erratically as the injection p r e s s u r e increased to that of the driving p r e s s u r e i n the sup-
ply tanks. Each port, which was less than 0.01 inch (0.00254 m) in diameter, had to be

9
I I I II I I II I I 1111 II 111 I I l l 111111111.1111.111111 I 1 II11.1111111 1.1 I I, I I I I, I II I, 111.1 -..--I- .-I .,,...." .-,,,..- ..-... ,.... ..--_--_-.

Figure 4. - Control room of vacuum facility.

h \ h \ \ h
0.10-sec blips
laximum t h r u s t

-Oxidizer flow (rotameter)

-TGGiXi (rotameter)
-
Thrust sta-+
Start

I I
Fuel flow at orifice-, ....I_.

I / / chamber pressure

I
I

I
Oxfdizer flow at orifice;

I
I

I
su;e:r;1: y
10 Percent maximum
Y h a m b e r - p r e s s u r e start
chamber pressure
Figure 5. -Typical 500-millisecond test record.

10
I

' pressure ' I

pressure start
start
(a) Test time, 0. 175 second. (b) Test time, 0.050 second.

0.10-Sec blips

signal
Oxidizer flow
(rotameter);
T ;Fuel
e t flow
e r )

Start
signal- Thrust

Flow -
start7

flow

'Chamber-
pressure
start
Oxidizer flow
at venturi orificr

* -Chamber-
pressure
start

(c) Test time, 0.025 second. (d) Test time, 0. 015 second.

Figure 6. - Disappearance of equilibrium conditions as test durations decreased.

11
reamed clean t o r e s t o r e the original injection p r e s s u r e drop. Because of this operating
problem, equilibrium and repeatable performance data could not be obtained to include in
this report.

Data Analysis

All the test data w e r e recorded on a n oscillograph recorder. A typical 500-

millisecond-run r e c o r d is shown in figure 5. The data signals f r o m the transducers (and
thermocouples not shown i n fig. 5) are expressed as a voltage change representing a
known data value. The r u n duration is defined as the time f r o m the s t a r t of valve power
to the time the power is stopped.
Equilibrium data conditions for all the thrustors w e r e attained after 500 milliseconds
of steady-state thrustor operation. At this equilibrium point, the test data w e r e used to
calculate the steady-state performance. However, as test durations were made s h o r t e r
(fig. 6), equilibrium conditions w e r e no longer apparent. The propellant flow consisted
of a surge, the thrust tended toward a shock wave, and p r e s s u r e s had only begun to in-
c r e a s e when the test ended. Because of these conditions, the instantaneous data values
f o r the short-duration tests w e r e not reliable.
To obtain performance data from the short-duration r u n s (fig. 6 ) , the thrust and flow
rate signals were integrated over the duration of each test. The tests were repeated at
the s a m e operating conditions until repeatable performance data were obtained. F r o m
such integrated data, average instantaneous data values could be obtained. Thus, the
performance data from tests of 500 milliseconds o r longer are based on actual instantan-
eous conditions; data from s h o r t e r duration runs are based on average conditions over
the time span of the test.

DISCUSSION OF RESULTS

Steady- S t at e Perf or ma nce

The experimental r e s u l t s of the investigation of several typical low-thrust. bipropel-

lant, chemical thrustor assemblies are presented in table 11.
Thrustor combustion .efficiency. - The combustion efficiency of each thrustor was
.
*
determined as a function of the oxidizer-fuel ratio i n t e r m s of characteristic velocity C ,
which is defined as the product of the combustion chamber p r e s s u r e Pc the thrustor
throat cross-sectional area At, and the gravitational acceleration constant gc, all di-
vided by the total propellant weight flow rate Wt

12
 The combustion efficiency is the ratio of the experimentally determined C to the theo-
*
retical C*. The theoretical data are based on an assumption of frozen equilibrium ex-
pansion of the combustion gases; that is, after the propellants r e a c t in the combustor,
the products do not change i n flowing through the thrustor throat and expansion nozzle
(unpublished data obtained by S. Gordon and E. Norris of Lewis).

TABLE II. - RESULTS OF INVESTIGATION OF LOW-THRUST BIPROPELLANT

CHEMICAL THRUSTOR ASSEMBLIES

Quantity Thrustor

2 3 4 5 6 7

Maximum characteristic velocity, 4700; 143 5190; 158 5000; 153 4620; 141 4400; 134 4960; 151
ft/sec; m/sec

a t oxidant-fuel ratio of 1.5 1.4 0.8 1. 4 1.6 2.1

for theoretical frozen charac- 5500; 168 5580; 170 5380; 164 5550; 169 5500; 168 5430; 166
teristic velocity, ft/sec; m/sec

Maximum specific impulse,

(lb force)(sec)/lb mass; (N)(sec)/kg 196; 1920 261; 2560 262; 2570 238; 2340 205; 2010 260; 2550

at oxidant-fuel ratio of 1.3 1.4 1.1 1.4 1.6 2.0

for theoretical frozen specific 140; 2360 315; 3090 313; 3070 313; 3070 308; 3020 315; 3090
impulse, (lb force)(sec)/lb mass;
("sec)/kg

Specific impulse after 100 m sec of 138; 1352 180; 1768 181; 1777 195; 1913 145; 1422 126; 1238
run time, (lb force)(sec)/lb mass;
(N) ( sec 1k g

Thrustor valve response delay, m s e c 11 14 10 4 10 6

Thrustor ignition delay, m sec 14 4 43 27 6 299

Thrustor startup delay to 90 percent

of maximum thrust, m s e c 35 30 109 50 26 7 04

Maximum thrustor wall temperature 1544; 111 2371; 1572 2107; 142! 1792; 125C 756; 676 1031; 830
recorded, O F ; OK

13

I
 In figure 7 the C* performance data f o r each of the tested t h r u s t o r s are presented
*
as a function of the ratio of the oxidizer-fuel weight flow rate. The C data are shown
as a n average curve with the data s c a t t e r being rt4 percent o r less on either side of the
curve in t e r m s of performance, and &4percent i n t e r m s of the oxidant fuel ratio O/F.
For clarity, the average performance curves are shown separately f o r each thrustor
*
along with the theoretical frozen C curves for each thrustor (theoretical data of Gordon
and Norris).
*
The experimental C performance curves peaked at O/F r a t i o s that ranged from
0.8 to 2.0, the majority being around 1.5. The peak performance varied from 4200 to
5200 feet p e r second (1282 to 1585 m/sec), which is approximately 80 to 93 percent of
the calculated theoretical.
*
Because two different propellant-injection methods w e r e used, the C performance
of thrustor 4 is shown with three different average curves in figure 7. For one series of
tests, the fuel was injected into the combustion chamber through the port designed for the
oxidizer, and the oxidizer was injected through the two ports designed for the fuel. The
resulting data a r e represented by the solid curve between a n O/F of 0 . 4 and 0.8. In the
second series of tests, the method of fuel injection was reversed, and the resulting data
are represented by the solid curve between a n O / F of 0 . 8 and 1.4. F o r tests of the s a m e
time duration, the C* performance was much lower f o r the second test series than for
the first. When the second test s e r i e s was continued f o r durations of 1200 percent longer
than the first series, the performance tended to match in each case. The performance
curves do not overlap f r o m one injection technique to the other because the injection p r e s -
s u r e limited the extent to which the O/F variations could be carried. A possible explan-
ation for this variation in combustion performance is that, in the first test s e r i e s , with
two oxidizer jets impinging on one axial fuel jet, the combustor walls are covered with
oxidizer s p r a y and rapidly heat; this heating effect, in turn, improves the vaporization
and reaction of the fuel with the oxidizer. With two fuel jets impinging on a n axial oxidizer
jet, the resulting fuel s p r a y tends to keep the combustor walls cooled, and vaporization
and propellant reaction are l e s s efficient until the combustor walls reach the s a m e oper-
ating temperature as in the first test series.
Thrustor 3 has the s a m e type of propellant injection as do t h r u s t o r s 2 and 6; how-
e v e r , its performance is 92 percent of theoretical as compared with 84 and 80 percent
f o r the other two thrustors. The differences in the performance of thrustor 3 and thrus-
t o r s 2 and 6 may be explained by their respective combustion-chamber length-to-diameter
ratios; f o r thrustor 3, the ratio was 7. 5 as compared with 3.2 f o r thrustor 2 and 2.3 for
thrustor 6. The large ratio of thrustor 3 affects its performance because the propellant
injection ports are much c l o s e r to the hot inner combustor-wall surface, and the fuel
jets have a longer chamber length to travel than they do in t h r u s t o r s 2 and 6.
Another thrustor that gave a relatively high performance was number 7. I t s

14

II I
 65M
-__ Theoretical frozen
Experimental
- 550C
1800 600c

.I
l7O0
55M 500C
theoretical

1400
1500 500c 4500

1300
1400
450C theoretical 1200 4000
1300

400@
, 1 I I I 3500 I I
D 1.5 2.0 2.5 1.0 1.5 2.0 2.5
(a) Thrustor 2. (b) Thrustor3.

6ooor 1
-
U
CA
a,

E
"
-
CA
3=
0
2: t
._
c
" >;
U
theoretical
&- 1500
data)
---
84 Percent---,
t-
> 4500 - theoretical

-\
c
U

k 1200 4000 4000 -

II
V
two fuel
(5-sec data)
1100
3500 I 3000 0 1.5 2. 0 2.5 I
0 .5 1. 0 1.5 1.1
(c) Thrustor 4 ( d ) Thrustor 5.

-\
- \

-- -\
-
\
\
\

1500
theoretical
1400 -
1200

1300
1000
3000
1.0 1.5 .20 2. 5 3.0
4000
0
I I
1
I
2 3
I
4
Oxidizer-fuel weight ratio, O F
(e) Thrustor 6. (f) Thrustor 7.
Figure 7. - Thrustor experimental characteristic velocity performance as function of propellant mixture ratio.

15

I
 combustor-chamber length to diameter ratio is only 1.4, but it has a very l a r g e c h a r a c -
t e r i s t i c length L* of 60 inches (152 cm) and u s e s a novel propellant injection technique.
Propellant jet impingement is not depended on for jet breakup and vaporization; instead.
the propellants are injected through atomization nozzles. Also, the resulting propellant
m i s t is injected away f r o m , r a t h e r than toward, the t h r u s t o r gas exit. These two steps
improve the propellant reaction by permitting better utilization of the existing combustor
volume; thus the combustion gas t e m p e r a t u r e is higher than that which would occur i n a
combustor having the s a m e volume but using a m o r e conventional injection technique.
The overall C* performance of these t h r u s t o r s is much lower than that of l a r g e r
thrustors. For example, with 100-pound (444.8-N) thrust s y s t e m s , C is expected to
*
be higher than 5500 feet p e r second (1675 m/sec)(unpublished data obtained by Bell Aero-
s y s t e m s , Inc.). The combustion efficiency of the s m a l l t h r u s t o r s is strongly influenced
by the atomization, mixing, and reaction of the propellants within the combustor. With
only two o r t h r e e orifice jets to inject the propellants into the combustor, any slight
misdirection o r obstruction in a s t r e a m will drastically reduce the efficiency of the com-
bu stion p r o c e s s .
Specific impulse as__function
~ - of oxidizer-fuel ratio. - The specific impulse of each
thrustor firing in a vacuum environment was determined at steady-state operating con-
ditions by dividing the thrust data by the total propellant flow r a t e obtained at the s a m e
time :

During testing, the vacuum environment w a s maintained s o that the thrustor nozzles were
always flowing full or, possibly, were flowing underexpanded.
The specific impulse efficiency is defined as the ratio of the experimentally d e t e r -
mined I to the theoretical I The theoretical I w a s determined on the b a s i s of
SP SP' SP
frozen composition conditions; that is, the combustion products flowing through the
thrustor throat and nozzle did not change in composition. The experimental and theoret-
ical I for each of the t h r u s t o r s is plotted as a function of O/F in figure 8, where
SP
the experimental data a r e presented as an average curve. With r e g a r d to performance
and O/F, the deviation f r o m this curve was +.4 percent.
The experimental I performance c u r v e s of figure 8 a r e s i m i l a r to the C
* curves
SP
of figure 7: the maximum I w a s obtained at O/F ratios that ranged from 1 . 0 to 2.0.
SP
with most of the I values reaching a peak a t about 1.5. The peak values varied from
SP
195 to 262 pounds thrust p e r pound of propellant flow p e r s e c o n d ( l 9 1 2 to 2570 (N)(sec)/kg).
These values a r e approximately 67 to 84 percent of the theoretical values obtained at a
corresponding O/F ratio. The two performance c u r v e s in figure 8 for thrustor 4 a r e

16

.. . . . , , . I
 3400 ---Theoretical frozen
Experimental

300
----_-- \

2000
220t 78 Percent
theoretical
2200

200
theoretical

1800

1800
Y 5 In 1.0 1.5
(a) T h r u s t o r 2.
2.0
I 150
1.0 1.5 2.0
I
(b) T h r u s t o r 3.
2.5
I

350-

VI
m
u) /----

E
300.

84 Percent

t
Y

250 -
2 2400

‘ t
70 Percent
._E -
;
G
>
+-
200 --
One oxidizer; \ g 2000

150 ’
I .5
two fuel

1.0 1.5 2.0

3E 1600
3

>
v
m
>
1.5 2.0 2. 5
( c l T h r u s t o r 4. (d) T h r u s t o r 5.

350r Nozzle expan-

sion ratio r

---
20 ( T h r u s t o r 6a a n d 6b)
c-
\
\
2000 \

theoretical 2800
I 61 Percent \ I I
1600 theoretical \
1200
1001 I 2M)o I I I
1.0 1.5 2.0 2.5 0 1 2 3 4
Oxidizer-fuel weight ratio, O F
(e) T h r u s t o r 6. ( f ) T h r u s t o r 7.
Figure 8. - T h r u s t o r experimental vacuum specific impulse performance as f u n c t i o n of propellant m i x t u r e ratio
f o r tests of more t h a n 500 milliseconds duration.

17
presented s i m i l a r l y to those in figure 7 for the characteristic performance. Two p e r -
formance c u r v e s are also presented for t h r u s t o r 6 i n figure 8 , but in this c a s e t h r e e
nozzles were used; two of the nozzles (sa and 6b) had a n expansion ratio g r e a t e r than 20
and the other nozzle (6c) had a n expansion ratio of 1.6 (sea-level expansion). A l l the
nozzles were u s e d on the t h r u s t o r t e s t e d i n a vacuum environment. The l a r g e r expansion
ratio did i n c r e a s e the specific impulse efficiency.
With rocket engines having a l a r g e r thrust (25 lb force o r m o r e o r 110 N), the spe-
cific impulse efficiency is generally slightly l e s s than the characteristic velocity effici -
ency (4 to 6 percentage points, unpublished data obtained by Marquardt Corp. ). However,
in the c a s e of these s m a l l thrustors, the difference between the two performance p a r a m -
e t e r s is 6 to 13 percentage points. The interrelation between the characteristic velocity
and specific impulse performance is the thrust-coefficient: Cf = ISp gc/C*. Using
this equation with the available experimental data gives Cf data between 1.38 and 1.82.
These values of Cf are 90 to 98 percent of theoretical (ref. 1). The Cf factor is also
equal to the thrust divided by the thrustor throat a r e a and the combustion-chamber p r e s -

N
E
u
z
Q
t

c
.-0
c
m
3
D
E
0
0

Engine thrust, F, Ib force

I l l I6 L . I I -- I-. 1 1 I 1
0 2 4 8 10 12 14 16 18 20 22
Engine thrust, F, N
Figure 9. -Engine chamber pressure as function of thrust.

18

..-. . . .- ... .
sure, Cf = F/(At)(Pc). A plot of the experimental combustion-chamber p r e s s u r e against
the t h r u s t o r experimental t h r u s t data (fig. 9) indicates that the thrust coefficient f o r each
thrustor is very n e a r t o being a constant over the experimental testing range. Thus,
with the thrust coefficient remaining relatively high (90 percent o r more), a l o s s in over-
all performance I f o r these s m a l l thrustors is due primarily to poor propellant com-
bustion C .
* SP

Pulsing Performance

Total impulse. - T h e thrustor total impulse is defined as the thrust integrated as

a function of time, f: F d t = It. This integration w a s performed f o r each t e s t by the
thrust-stand electronic network. The indicates the total f o r c e that will be obtained
from a thrustor operating f o r a given time duration. By itself, It does not indicate how
efficiently the thrustor is producing the thrust.
Each thrustor was tested over a range of operational time durations. Enough tests
were made at any one time duration to establish repeatable impulse data. The impulse
data as a function of test time are shown for each thrustor i n figure 10. The data from
all the thrustors show a direct, almost linear, relation between the experimental i m -
pulse and the valve open-time. A s the pulse on-time becomes a g r e a t e r percentage of the
total operating time, the n e a r e r the performance comes to theoretical impulse. This r e -
lation is illustrated in figure 10 with data for thrustors 6 and 7 . The individual test dura-
tions w e r e held constant while the time between pulses was increased.
Specific impulse - .
as .function of operating time. - A s mentioned in the section,
APPARATUS AND PROCEDURE, f o r tests of less than 500 milliseconds, little credibility
is placed i n the data at a given instant during the test, and integrated and average data
values are used to calculate the performance. In this section, the I is determined a s
SP
a function of decreasing operating time (less than 500 m sec) with the O/F held rela-
tively constant from t e s t to test. T e s t s were repeated at identical conditions often enough
to get identical, repeatable data. Thus, the I in this section is a n average value over
SP
the time span of the test.
The average I data for each of the t h r u s t o r s over a range of valve on-times is
SP
shown as a n average curve i n figure 11. While it was possible to obtain impulse data
(fig. 10) down to 10 milliseconds, the slow response time of the propellant flowmeters
placed a lower limit of about 20 milliseconds on the flow data. Therefore, 20 millisec-
onds is a l s o the lower limit on the I data shown in figure 11. The Isp performance
SP
decreases rapidly f o r all the t h r u s t o r s from the peak values (as shown in fig. 8) to less
than 25 percent of the peak value i n some cases at the minimum valve on-time.
Two performance curves, each representing different propellant injection techniques

19
 1°iE ---Theoretical

E Experimental

[l

-
.1 -
1 I11111 I I lllllll

(bl Thrustor 3.

~~ 11111 I I I11111
10 000 10 000
(c) Thrustor 4. (d) Thrustor 5.

.01 I I1111111 I 1 1 I Illll I I I I11111 .01 I 111111 rl I 11111111 I I1111111

10 100 1000 10 000 10 100 lOOC 10 000
Thrustor on-time, msec
(el Thrustor 6. ( f ) Thrustor 7.

Figure 10. - Thrustor experimental total impulse as function of thrustor on-time.

20
 2OOr
- I

1600

1200

10 100 1000
(a) T h r u s t o r 2. Oxidant-fuel (b) T h r u s t o r 3. Oxidant-fuel
ratio, 1.5. ratio, 1.5.
250
VI
v)

cn
Y
2
U
a,
VI

5 2400

>
m
>
m +-
._ ai
ai
v) one oxidizer .._
;” 1800 VI

1
U

E
n
VI ._
._
U
5 1400 ._
c
v
._ 3
’3
U
a, ro
U
I I I I1111 I I I I Ill >
G-3 10 100 1000 5 -001
3
10 100 1000
u (c) T h r u s t o r 4. Oxidant-fuel 5 (d) T h r u s t o r 5. Oxidant-fuel
9 ratio, 1.0. ratio, 1.5.

200
r Th r u s t o r 6b
1600
200

1600

800

400
800

o1 I I I I 11111 I I IIIIIJ
10 100 1000 10 100 1000
T h r u s t o r on-time, msec

(e) T h r u s t o r 6. Oxidant-fuel I f ) T h r u s t o r 7.
ratio, 1.5.
F i g u r e 11. - T h r u s t o r experimental vacuum specific impulse performance drop-
off as f u n c t i o n of t h r u s t o r on-time.

21
II I 1111 111 I I II I I 11111111 11.111 I I1 I1 I I I 1 I111 I I, 11
.1 111.11, I .I .... 111-1.. -.-. . _.-._..

used, are shown in figure l l ( c ) f o r thrustor 4. For thrustor 6 (fig. l l ( e ) ) two perform-
ance curves, each for a different nozzle-expansion ratio, appear to be converging at
some minimum valve on-time.
The fact that decreasing valve off-time between tests or pulses resulted i n higher It
is also shown with I For example, i n figure ll(f) performance curves are presented
SP'
f o r thrustor 7 operating with a duty cycle of m o r e than 5 percent and less than 1 percent.
The c u r v e s show that, f o r a given on-time and a constant O/F, performance i n c r e a s e s as
the on time becomes a l a r g e r percentage of the total t e s t time.
Since most of the It performance is very close to theoretical values (fig. 10) and
the I for the s a m e t h r u s t o r s falls off rapidly with d e c r e a s e d operation time, the p r o -
SP
pellant utilization is apparently v e r y poor. When the propellant control valves open, the
propellant must s u r g e into the vacuum environment of the combustion chamber. This
surging is shown by spikes in the data t r a c e s (especially f o r the oxidizer) of figures 5
and 6. For s h o r t tests, neither flows nor thrust has time to r e a c h equilibrium conditions,
and repeated t e s t s become a s e r i e s of detonations. The actual O/F during this surging

TABLE III. - THRUSTOR AVERAGE RESPONSE TIMES FOR TORQUE- AND SOLENOID-TYPE VALVES

[Data are average values for thrustor test times of 1 sec o r longer. J
I _ -
__ .. .-. - - ..- .__

rhrustoi Average time, m sec, from s t a r t until- Average time, m sec, from top signal until-
. . -.

Propel- Chamber Thrust Chamber Thrust Propel- Chamber Thrust Chamber Thrust
lant pressure begins to p r es s u re reaches lant pressure begins to pressure reachec
flow begins to increase isapprox- maxi - flow begins to decrease isapprox- prerun
begins increase imately mum stops decrease imately value
3 0 per cenl test LO percen
of maxi- value of maxi-
mum mum
~ .. . - - - .. -.

1 10 17 23 134 ___ -_- -_ __- ___ ___

2 11 17 25 177 35 7 8 7 2 45 23

3 14 19 18 210 30 7 7 8 280 21

4 10 51 53 177 109 6 6 10 64 30

5 4 29 31 60 50 3 5 9 119 3 16

6 10 14 16 323 26 7 5 8 122 27

a7 6 276 305 644 7 04 3 5 9 119 316

___ --- _-- _-- 5 --_ 15 ___ 3 42

b7 __6 - -

aWith value 8.
bWith value 9.

22
of propellants is apparently much different from what it would be once flows and thrust
have stabilized. For the very s h o r t pulse durations, the propellant valve opens and
starts to close before the thrust and flows reach their maximum values (see fig. 6). It
is little wonder, then, that combustion efficiency is s o low for the short valve on-times.
The thrust (and It) produced is a r e s u l t of high initial flow rates, propellant vaporizing,
and burning to some degree, and then expanding out through the thrustor nozzle. F o r
these short tests, no m a t t e r how often they are repeated, the combustor wall tempera-
t u r e increases little.
Thrustor propellant valve response time. - A limitation of how often o r how fast a
thrustor can be fired is the opening and closing response time to a n electrical signal.
In this investigation, the valve response time is defined as the time f r o m the initiation
o r termination of the valve power t o the time that a change in propellant flow downstream
of the valve is indicated. F r o m data such as that shown in figure 6 , a n average response
time was computed f o r each of the valves tested. These data a r e presented in table 111
in the columns "Propellant flow begins", and "Propellant flow stops". These response
time averages ranged from 4 to 14 milliseconds in opening and 3 to 7 milliseconds in
closing. The averages were distributed between the torque- and solenoid- type valves s o
that neither had a n apparent response superiority over the other.
Thrustor ignition response time. - Another limitation on the pulse o r repeated firing
of a s m a l l thrustor is the time necessary f o r the propellants to react and to produce p r e s -
s u r e and thrust. This time is called ignition delay and is the period that begins when the
propellant is detected in the thrustor injector and ends when a sudden change is noted in
the thrust signal. Again, the average ignition delay was determined f o r each of the
thrustors, and these values a r e presented in table II. In figure 12, these data a r e pre-
sented as a function of the thrustor characteristic length L*, which is defined as the com-
bustor volume divided by the cross-sectional a r e a of the combustor nozzle throat. A s
the L* i n c r e a s e s , a longer time is required f o r the propellants to vaporize and mix and
to reach a n ignition temperature condition. The ignition delay data of figure 12 are f o r
t e s t s in which the combustor and propellants a r e both at ambient temperatures. F o r re-
peated firings with the s a m e thrustor , the combustor-wall temperature increased with
each firing, and the ignition delay decreased accordingly. The s h o r t e r the off-times be-
tween pulses, the s h o r t e r the ignition delay time. This relation between off-time and ig-
nition delay is shown in figure 13 f o r thrustors 3 and 7. The ignition delay f o r the initial
firings of either thrustor is that shown in figure 12. As firings were continued, this de-
lay decreased until it reached a constant value dependent on the off-time between firings
o r pulses. As the off-time was decreased, both thrustors reached a n almost identical
ignition delay that approached zero at zero off-time (steady-state firing conditions).
Thrustor performance response time. - The final limitation to obtaining a particular
impulse o r thrust is the time required to reach the desired impulse o r thrust value from

23
 400

m - ~
/
0 Combustion-chamber
ressure startup /
100
80 ,/'
60

40

20
/

10
8
6

4
0 10 20 30 50 60
Thrustor characteristic length, L", in.

I I I I I
0 20 40 60 80 100 120
Thrustor characteristic length, Lc, cm
Figure 12. - Thrustor ignition delay as function of thrustor charac-
teristic length.

400

200 - delay for initial

start of 60-in.
I I
" 100 Thrustor I
g 80 characteristic-
>; 60 length, I
-
Io
L",
-E 40 in. (cm) -
c
.-
0
.-
c
c
-
ol 20

delay for initial

10
1 stari of Il-in.

1
8 (28-cm) thrustc
6
4
.1
I I 1 I I//// 100
Pulse off-time, sec
Figure 13. -Average ignition delay after start of pulsing operation. (Delay, time from
start signal until chamber pressure starts to rise.)

24
 0 Performance response
A Thrust termination decay

10 20 30 40
Thrustor characteristic length, L*, i n

1 I I I
40 60 80 100 120 140 160
Thrustor characteristic length, L', cm
Figure 14. -Thrust buildup and fall-off delay.

the t i m e an actuation signal is given. This performance response time w a s calculated

and averaged f o r each of the t h r u s t o r s and is presented in table II. The performance
response time includes both the valve response and ignition delay times plus the time for
the thrust to build up to its maximum equilibrium value. Therefore, the thrust o r p e r -
formance response is a l s o a function of the L", as shown in figure 14.
Also important in the performance response of a t h r u s t o r is thrust decay on signal
termination. Because thrust decay does not involve a n ignition delay, it is a much
s h o r t e r t i m e than the startup response. The curve f o r this termination decay is a l s o
shown in figure 14. Both the curves in figure 14 a r e for initial firings, that is, when the
combustor and propellants are a t ambient temperatures. A s the t h r u s t o r temperature
i n c r e a s e s with successive operations, the ignition time d e c r e a s e s , which results in a
s h o r t e r performance response time.
In computing the performance response, maximum and minimum values of thrust
w e r e used r a t h e r than the combustion-chamber p r e s s u r e maximum and minimum re-

25

I
sponse times. The large difference in maximum and minimum response t i m e s between
the thrust and the chamber p r e s s u r e is shown in table III. Although chamber p r e s s u r e
and thrust data have a linear correlation f o r a given t h r u s t o r , as shown in figure 9 , each
signal arrives at this correlation at a different rate. The chamber-pressure change
rate is felt to be restricted by combustion-gas sonic flow-rate limits in the measuring
transducer (this limitation is discussed in the appendix). The thrust data are believed
to reflect m o r e accurately the actual thrustor response behavior.

T h r u s t o r Heat Transfer

Another criterion t o b e considered in the design and operation of s m a l l thrustors is

the material temperature limits. The theoretical combustion temperature of nitrogen
tetroxide and a mixture of 50 percent UDMH and 50 percent hydrazine is 5500' to 6000' F
(3300' to 3600' K) f o r the testing conditions of the present investigation. The amount of
heat reaching the combustor walls is a function of the amount of the propellant that
merely cools the walls rather than that which enters into the combustion process. As
mentioned in the section, APPARATUS AND PROCEDURE, all the thrustors had either
radiation-cooled walls o r heat-sink walls.
The importance of temperature w a s observed during the tests. If the individual
pulse-run firings w e r e kept s h o r t enough o r the time between firings increased enough,
all the thrustors tested could pulse fire for long periods of time without a n excessive heat
buildup i n the combustor walls. However, as the individual firings increased in duration
and/or the down time (or off-time) was decreased, the combustor-wall temperature in-
c r e a s e d rapidly. In table IV, the temperature data f o r thrustor 3 illustrate this phe-
nomenon. Table IV(a) presents 10 successive 1/2-second firings with the wall tem-
peratures at the end of each firing, and IV(b) presents the data for a continuous firing
that was terminated when the combustor wall burned through. Thrustor 3 had a rela-
tively high C" performance level, which indicated that most of the propellant was
going into the combustion reaction. This temperature rise as a function of run time
is plotted in figure 15, along with data f o r thrustor 3 and the wall temperature data f o r
s e v e r a l other thrustors. An example of a continuous operational radiation-cooled
thrustor is thrustor 4 , which reached a n equilibrium wall temperature of about 1900' F
(1300' K) after about 60 seconds of steady operating conditions. T h r u s t o r s with lower
performance tended to reach a lower wall temperature equilibrium.

26
 TABLE IV. - THRUSTOR 3 PULSING AND STEADY-STATE TESTS

(a) Pulsing test; on time, 0.470 second; off time, 5 . 9 0 seconds

Pulse Maximum Impulse Combustor wall temperature

thrust

i
N lb force N-sec l b f o r c e - s e c OK OF OK OF

18.9
~~ ~

1 60. 5 88 4.25 ---- --- 494 428 484 410

2 60.5 88 ~ 18.9 4.25 -_-_ ___ 567 557 558 544
3 61.3 89 ____ 8.90 2.0 619 651 625 664
4 61.3 89 8.45 1.9 669 739 681 765
5 60. 5 88 8.90 721 835 723 840

10
6
7
8
9
59.8
59.8
60.5
59.8
59.8
87
87
88
87
87 ~
____
18.8
____
4.22 ~
I
8.45

____
2i0
1.9
-__
765
798
828
847
900
912
976
1026
1061
1158
764
793
821
846
858
912
967
1018
1061
1082

(b) Steady-state test; total run time, 10 seconds

Time Combustion- I Thrust Oxidant - Combustor wall temperature

after s t a r t
of run,
~
fuel
ratio I
I
Throat
I
I
Chamber
I
sec
N lb force OK OF
~

34.4 0. OE 0 0 413 284

61.3 89 19.6 4. 4 1 5 08 454 445
61.3 89 19.5 4. 3 8 592 604 636
60. 5 88 19.4 4.37 696 79 1

59. 8
59.8
87
87
I 19.3
19.2
4.35
4.35
4.35
4.33
4.31
767
850
980
1082
1207
920
1069
1306
1489
1710

27
 1400 -

1800 c

1200 - i

1600 .'
1400
x 1000 -
W-
I
3
c 1200
e Y

a.
W
E-
E 800 - J
+
e 1000
P
W

5
800
0
5 Unpublishei
600 - 600 manufac-
turer's
exp. data
400 0 4
v 3
400 -

[ I dl
200

50 60
0 10 20 30 40
Time after start, sec
Figure 15. -Throat wall temperature as function of run duration.

SUMMARY OF RESULTS

A s a result of this investigation of reaction control, storable bipropellant thrustors,

the following conclusions were established:
1. The steady-state performance for firing t i m e s longer than 500 milliseconds is
about 80 to 90 percent of theoretical in t e r m s of characteristic velocity, and 67 to 84
percent of theoretical in t e r m s of specific impulse. The thrust coefficient remained
relatively high (90 percent o r m o r e of theoretical). The low specific impulse results a r e
due primarily to poor propellant combustion efficiency, that is, low characteristic ex-
haust velocity.

28
 2. The pulse-firing data indicated a direct, almost linear, relation between the ex-
perimental total impulse and the valve open-time down to approximately 10 milliseconds.
For the s a m e pulses, however, the average specific impulse decreased to approximately
one-fourth of the steady-state performance value.

Lewis Research Center ,

National Aeronautics and Space Administration,
Cleveland, Ohio, September 11, 1967,
128-3 1-20- 50- 2 2.

29
 APPENDIX - TEST EQUIPMENT DETAILS
Mu1tipou nd Th r u s t Stand

The thrust stand a s s e m b l y consists of a mass-balanced platform s y s t e m , pivoted on

knife edges and r e s t r a i n e d by a n electric-force motor that is controlled by a displacement
pickup and electronic network. The system is designed to m e a s u r e thrust from 0 . 0 1 to
10 pounds f o r c e (0.044 to 44.5 N) in four range settings with a threshold of 1 percent of
any range setting. Impulse can b e measured f r o m 0.01 to 10 pound-seconds (0.044 to
44.5 N-sec) with the s a m e accuracy as that for thrust. T h r u s t variations can also be
measured with the s a m e accuracy a t frequencies up to 100 cps (100 Hz).
The mass-balanced platform system is suspended on six s e t s of precision knife edges
and bearing blocks. The platform counterbalances a r e remotely adjustable by electric
m o t o r s , and they p e r m i t the platform to b e balanced f o r a payload of up to 25 pounds
(11.3 kg). The moving elements of the displacement t r a n s d u c e r and of the force motor
a r e connected directly to the platform. Propellants and other s e r v i c e fluids a r e brought
to the thrustor platform through flexible tubing parallel to the pivots. The electrical
wiring is s e t up in a s i m i l a r manner. The electronic equipment associated with the
force-generating loop and balancing is remotely located.
The combination of a c c u r a t e f o r c e readout and high frequency response is obtained
by a force-generating servosystem that operates in the following manner: The f o r c e
f r o m the t h r u s t o r under study causes a horizontal platform displacement and results in
a voltage output f r o m the displacement transducer. This signal, after suitable amplifi-
cation and compensation f o r loop phase shift, is applied to the f o r c e motor that produces
a force opposing that f r o m the thrustor and displaces the platform toward null position.
At null position, the f o r c e exerted by the force motor is in balance with the f o r c e pro-
duced by the thrustor. The c u r r e n t in the force motor is proportional to the force pro-
duced and is used as the thrust readout for the s y s t e m . This system is described in de-
tail in reference 2.

Propel la nt Flow meters

A major problem in testing millipound t h r u s t o r s is determining the propellant flow

rates. An average flow r a t e can b e obtained by determining the total amount of propel-
lant used for a fixed duration of time; however, in this investigation, measurement of
the instantaneous flow r a t e s f o r each propellant was attempted. The flow rates f o r the
mixture of 50 percent UDMH and 50 percent hydrazine ranged f r o m 0.0008 to 0.0049
pound p e r second (0.364 to 2.04 g/sec) and f o r nitrogen tetroxide f r o m 0.0020 to 0.0100

30
 Figure 16. - Flowmeter calibrator.

pound p e r second (0.91 to 4. 54 g/sec). T h r e e different types of flowmeters were

selected that could operate with these propellants and indicate the flow rates. These
flowmeters, a differential p r e s s u r e orifice, a turbine wheel, and a remote-reading
rotameter, are shown in figure 16. The m e t e r s were calibrated (in the r i g pictured in
fig. 16) by flowing actual propellants f r o m one glass tank to another through the meters.
The indicated flow r a t e s f r o m each of the flowmeters w e r e thus checked against each
other and against the visual amount of propellant that had flowed f r o m one glass tank to
the other within a given time interval. Different p r e s s u r e heads in the propellant tanks
were used to obtain different flow rates. The calibrated flow r a t e s had a n accuracy of a t
l e a s t 0.0002 pound p e r second (0.091 g/sec).

Thrustor Combustion-Chamber Pressure Measurements

The combustion chamber p r e s s u r e was measured on most of the thrustors by a p r e s -

s u r e transducer connected b y a finite length of metal tubing to the combustion chamber o r
to the propellant injector face. Most of the thrustors w e r e so s m a l l physically that this
connecting tube w a s capillary size with a l a r g e length-diameter ratio.
Although the p r e s s u r e transducers have a high natural frequency capability (2 to
6 kHz), when they are connected to the p r e s s u r e source with a high length-diameter tube,
their frequency response is drastically reduced. The p r e s s u r e measuring system then

31

I
 111 II I I

transducer

Figure 17. - Comparison of initial thrust, flush transducer chamber pressure, and standard-type transducer chamber pressure.

r e s e m b l e s a Helmholtz resonator. Resonator equations indicate a typical frequency r e -

sponse of about 166 hertz for a setup such as that used in this investigation.
In o r d e r to obtain experimental verification of this frequency response, dual chamber-
p r e s s u r e measuring t r a n s d u c e r s w e r e installed on s e v e r a l thrustors. On thrustor 7, one
p r e s s u r e transducer w a s connected by the metal tubing, and a flush-mounted transducer
w a s mounted directly i n the combustor wall (see sketch in fig. 17). Because the flush-
mounted transducer was uncooled, its data w e r e only a c c u r a t e f o r a few seconds of any
given test. A typical oscillograph r e c o r d of the two t r a n s d u c e r s and of the t h r u s t s y s t e m
is presented in figure 17. The agreement between the two p r e s s u r e measuring s y s t e m s
f o r time spans of a few milliseconds appears very good, with some evidence of agreement
f o r time spans as s h o r t as a millisecond, thus providing reasonable verificiation of the
calculated frequency response.

T h r u s t o r Wall Temperature Measurements

Temperature measurement of the thrustor combustor walls , especially those made

of r e f r a c t o r y m a t e r i a l s , proved to b e a problem. Mounting thermocouples solidly to the
refractory metal combustor walls involved breaking through the protective coating on the

32
material, thereby risking oxidation failure at that point during high-temperature condi-
tions Optical means of determining accurate wall temperatures were complicated be-
cause the thrustors w e r e operated in a vacuum tank that w a s filled with haze and fumes
during actual tests and because emissivity data f o r coated refractory materials at high
temperatures were lacking.
One solution to the problems in making temperature measurements w a s to wire ther-
mocouples to the refractory material thrustor walls. First the wired-on thermocouple
was calibrated in the following manner. 'Ityo identical thermocouples w e r e mounted as
close together as possible on a stainless-steel chamber, One thermocouple w a s welded
to the stainless surface, and the other thermocouple was wired onto the chamber. During
testing with the stainless-steel chamber, a ratio w a s established between the temperature
data f r o m the two thermocouples. Then the wired-on thermocouple w a s remounted on the
refractory metal chamber. The data obtained by the wired-on thermocouple during testing
w e r e then multiplied by the ratio determined f r o m the t e s t s with the stainless-steel cham-
ber. This corrected temperature is the one reported in this investigation for refractory
material thrustors. The actual temperatures a r e reported only for those thrustors having
stainless - stee1 combus tors.

33
 REFERENCES

1. Sutton, George P. : Rocket Propulsion Elements. Second ed. , John Wiley and Sons,
Inc., 1956, p. 70.
2. Smith, J. D. : Modification to Government-Owned T h r u s t Stand System. Hughes
Research Labs. (NASA CR-72021), Aug. 1966.

34
 -

National Aeronautics and Space Administntioa FIRST CLASS MAIL POSTAGE A N D FEES PAID
NATIONAL AERONAUTICS AND
WASHINGTON, D . C. SPACE ADhfINISTHATION

OFFICIAL BUSINESS

"The aeronautical and space activities of the United States shall be

.
conducted so as to contribute , . to the expansion of human knowl-
edge of phenomena in the atmosphere and space. The Administration
shall provide for the widest practicabje and appropriate dinemination
of information concerning its activities and the results fbereof."
-NATIONAL AERONAUTICS AND SPACE ACT OF 1958

NASA SCIENTIFIC A N D TECHNICAL PUBLICATIONS

TECHNICAL REPORTS: Scientific and technical information considered
important, complete, and a lasting contribution to existing h o w l d g e .

TECHNICAL NOTES: Information less broad in scope but nevertheless of

importance as a contribution to existing knowledge.

TECHNICAL MEMORANDUMS: Information receiving limited distribu-

tion because of preliminary data, security classification, or other reasons.
CONTRACTOR REPORTS: Scientific and technical information generated
under a NASA contract or grant and considered an important contribution to
existing knowledge.
TECHNICAL TRANSLATIONS: Information published in a foreign
language considered to merit NASA distribution in English.
SPECIAL PUBLICATIONS: Information derived from or of value to NASA
activities. Publications include conference proceedings, monographs, data
compilations, handbooks, sourcebooks, and special bibliographies.
TECHNOLOGY UTILIZATION PUBLICATION$^ Information on tech-
nology used by NASA that may be of particular interest in commercial and other
non-aerospace applications. Publications include Tech Briefs, Technology
Utilization Reports and Notes, and Technology Surveys.

Details on the availability of these publications may be obtained from:

SCIENTIFIC AND TECHNICAL INFORMATION DIVISION

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
Washington, D.C. PO546