NASA Technical Reports Server (NTRS) 20000093963: Low-Cost Approach to the Design and Fabrication of a LOX/RP-1 Injector

24 Pages·0.71 MB·English

by  NASA Technical Reports Server (NTRS)

AIAA-2000-3400
LOW-COST APPROACH TO THE DESIGN AND FABRICATION OF A LOX/RP-I INJECTOR
M.D. Shadoan* and D. Sparks
Space Transportation Directorate. Marshall Space Flight Center. MSFC, Alabama
Abstract an FBC attitude when devising and developing new
prograrn acquisitions. :\pplying this to the design of space
NASA Marshall Space Flight Center (MSFC! ha_ hardware means we must adapt new practices that result
designed, built, and is currently testing Fastrac, a liquid in inexpensi,.e and reliable components. To accomplish
oxygen _LOXI/RP-I fueled 60K-lb thrust class rocket these goal:,, the designer must incorporate fabrication
engine. One facet of Fastrac which makes it unique is experience, such as material and process selection, along
that it isthe first large-scale engine designed and developed with innovative design approaches.
inaccordance with the Agency's mandated "'faster, better,
cheaper" (FBC> program policy. The engine was The initial goal was to build a LOX/RP-I injector that
developed under the auspices of MSFC's Low Cost Boost exhibits good performance and wall compatibility when
Technology office. operated with an ablative thrust chamber and nozzle
a_,sembl,, at, a fraction ot" the co_,t of a con,.entional
Development work for the main injector actually began equivalent urlit. The development injector was de,,igned,
in 1993 in subscale form. In 1996, work began on the fabricated, and tested in 16months. The design ,,,,'asthen
full-scale unit -I yr prior to initiation of the engine transformed, with minor modifications, into the main
development program. In order to achieve the value goals injector for the Fastrac engine.
established by the FBC policy, a review of traditional
design practices was necessary. This internal reevaluation The injector design, fabrication processes, verification pro-
would ultimately challenge more conventional methods cedures, and test results are detailed inthis paper. Also, a
of material selection, desi,,n= process, and fabrication cost breakdown is given for manufacturing the LOX/
techniques. The effort was highly successful. This "'new RP-I injector.
way" of thinking has resulted in an innovative injector
design, one with reduced complexity and significantly 2. DESIGN DESCRIPTION
lower cost. Application of lessons learned during this eftort
to new or existing designs can have a similar effect on The Fastrac 60K-lb LOX/RP-I main injector is shown in
costs and future program successes, figure 1. Excluding the gimbal assembly, the entire
component is made up of only three parts: the core. LOX
1. INTRODUCTION dome cap. and faceplate. Additionally, the injector
contains several unique features, resulting in a low-cost
Development of space hardware has traditionally been design.
done with the philosophy that the designer must use all
_ InjectorDomeCap/
available technological resources to maximize per-
formance. This philosophy placed great emphasis on high
thrust to weight ratios that greatly increased the cost and
LowerGimbalBlock
complexity of space hardware. However, in recent years
of budget reductions and downsizing, the Government as
a whole has been tasked with reinventing itself, to adopt
Member
Cop>right c) 2(_)0 b', the American Institute of Aeronautics and
Astronautics. Inc. No copyright is asserted in the United States under
Iniector PIate
Title 17, U.S. Code. The U.S. Government has aroyalty-free license to
exercise all rights under the copyright claimed herein for Governmental
Purposes. All other rights are reserved by the copyright owner. Fig. 1, LOX/RP-I N]K injector.
1
American Institute of Aeronautic and Astronautics
Time Line SILCORO-75 braze alloy isapplied directly tothe injector
I ! I core lands manually with a syringe-type applicator. Any
Procure Material for ]
IniectorCore,Cap,andPlate I holes smaller than 0.030 in. are filled with a braze flow
6Weeks
! i ÷ inhibitor to prevent the alloy from wicking into the holes
,,I during the braze operation. The injector plate is properly
Fabricate Fabricate aligned using index marks on the plate and injector, then
Inlector Injector Injector
Core I FabCricaapte Plate 4Weeks the injector isplace on the injector core. A centering pin
I I prevents radial movement during brazing. Reticulated
I foam is placed on top of the injector and a stainless steel
plate is added on top of the foam to provide even weight
2Weeks
I WCeoldreItnojeCctaoDr distribution during brazing. At the same time. two
I complementing witness samples are also brazed for later
use to detem]ine the integrity of the braze joint.
FinalMachine
6Weeks
Injector
The final step is to install inserts and proceed with braze
I ! verification.
I
4. BRAZE VERIFICATION
IBtoraFzeaIcnejepclatoter 1+Week
The braze joint isverified if itmeets two criteria: A tensile
! I i ill I
test of the withes,, san]pies used during brazing and a
PulTlest
xacuum check to determine if there are leak paths in the
Witness ! Week
CLheeac,_< I Samples imerpropellant joints. The _imess samples must meet
I I tension pull test requirements set forth in MSFC-SPEC-
I
2761. During the vacuum test the LOX side of the injector
Install is isolated from ambient pressure by sealing it with
Inserts and
LOXClean 1Week closeout plates and plugs in the instrumentation ports. A
I vacuum is drawn on the part using adiffusion pump and
is then isolated from the pump. The vacuum test must
meet the criteria of MSFC-PROC-2953: i.e., itmust show
Build
StartEngineI
no appreciable leakage for a period of 15minutes or the
21 Weeks part is rejected.
Total
Fig. 5. Fabrication process. 5. STRUCTURAL ANALYSIS
To enhance the dimensional stability of the injector, a A detailed repo.rt on the stress analysis performed on the
stress relief cycle is performed to remove the residual main injector will be documented in an internal memo
stresses induced by prior fabrication processes. The final that will be released later. This paper only address two
machining of the injector assembly will remove any areas of concern as cited by Sutton: l Stresses on the
distortion introduced by welding. Also, all key interfaces injector plate due to the large combustion forces and
such as gimbal location, attachment points, and braze keeping a positive seal between the fuel and oxidizer to
surfaces are final machined. prevent internal fires.
The next process is tobraze the injector plate tothe injector The high stress inthe injector plate is due to the pressure
core assembly by a vacuum brazing operation. Material forces and thermN gradients derived from the combustion
selection was key in making this braze reliable and cost process. Due to thermal growth in the injector plate, the
effective. The 304L stainless steel injector core will braze internal lands and the injector plate undergo plastic
exceptionally well to an oxygen-free copper faceplate deformation. Yet, the ductility of the 304L stainless steel
without the requirement of aplating process. The specifics and oxygen-free copper are large enough to prevent any
of the brazing operation can be found in MSFC-SPEC- low-cycle fatigue issues.
2761. Before brazing, the injector plate is prepared by
grinding the surface to be brazed to a flamess of 0.0005 The LOX issealed from the RP-I bythe braze joint, which
in. in the restrained condition. This ensures that all high measures 0.1 in. The braze alloy is sufficiently ductile to
spots are removed and promotes good braze flow. prevent any damage to the braze joint.
3
American Institute of Aeronautic and Astronautics
control again prechills the injector by partially opening
the main oxidizer valve IMOV) for 2 seconds prior to
initiating ignition with TEAFFEB flow to the chamber.
The chamber pressure _Pc Irises as the hypergol reacts
with the LOX. Once a threshold Pc is reached, the main
fuel valve is fully opened while the fuel purge is cycled
off. At this point Pc builds rapidly, and once a second Pc
threshold level is reached, the MOV isfully opened. This
sequence takes =2.4 sec and is a reasonably good
representation of the actual engine start time trig. 9).
8.3 TCA Mainstage
Steady-state design operating conditions for the main
injector are presented in table 2. Figure 10shows Pcversus
time tor both engine and component level operation. Run
Fig. 8. 15:1 TCA installed in TSII6. durations are limited to 150 sec at TS 116 due to the LOX
tank capacity. A view of the TCA during mainstage
8.2 TCA Start operation is shown in figure 11.
Tile TCA is started oxygen rich and uses tile hypergol 8.4 TCA Shutdown
mixture trieth} lalumit_um and triethylboron (TEA/TEB _
asthe ignition source. Atthe component level, the injector A fuel-rich cutoff sequence, designed to minimize the
is first thennallv conditioned by prechilling the oxidizer possibility of high-temperature damage to the injector.
side of the injector v+ith a reduced flov+rate of LOX for terminates TCA operation+ GN, purges are utilized to
5-l I)sec. immediately prior to committing to automated prevent residual propellants from bacMlowmg through the
control. Once autosequence has begun, the programmed elements of the faceplate and into the supply manifolds.
TCAComponentandEngineStartComparison
70O
600 f 1_,-J °_ -
500
Oll 400
o_
P7100--ICATest
e,l 300 "- t
........... 6522-Eng Test
E
m ?
200
m
--..r
t00 I
mJ ,,,_
2 3 4 6 7
Time (sec)
Fig. 9. TCA versus engine start Pc profile.
5
American Institute of Aeronautic and Astronautics
strength at temperatures >300 °F. Therefore, temperatures the theoretical performance (determined from
within the injector acoustic cavity had to be considered. performance prediction codes at the same conditions)to
since it is in intimate contact with the mounting flange of determine r/c,. It is v,orth noting that the data shown in
the chamber. The thermal environments in this region figure q shows that the Pc continues to rise during the
clearly play a major roll in meeting the life requirements test. Since flow rate remains constant due to the cavitating
of the component. venturies, the gradual pressure rise is a direct result of
carbon buildup in the chamber throat during TCA
The engine power balance and the requirements of the operation. Therefore. efficiency, values are quoted based
X-34 `.'ehicle established the operating conditions for the on measured pretest throat diameters and on data gathered
main injector. Even though the injector was designed to earl,, inthe test, before significant accumulations of carbon
be lo,a cost, it obviously must still meet minimum
can Occur.
performance and stability requirements. Component test
results indicate all of these goals have been achieved: D_namic stability goals were based on similarities with
however, much engine testing remains to fully determine the MA-5 sustainer engine. The requirement is lot no
system performance margins. spontaneous instabilities in development or flight
hardware ground tests. Additionally, 50 consecutive stable
Results of tests performed thus far are as follows: tests to mainstage conditions are required to meet this
criteria, with no pressure oscillations >I0 percent peak to
All testing has indicated that the goals of uniform peak. The ultimate goal of the program is to exceed this
temperatures and wall compatibility have successfull_ minimum requirement bydemonstrating dynamic stability
been met. Chambers tested with the baseline film cooling according to industry guidelines li.e., with bomb tests_.
percentages have not shov, nany e`.'idence of high them_al Some bomb tests have already been per_brmed with mixed
stress, or wide swings in temperature profiles, E`.en after re,,ult_,, However, no spontaneous instabilities have
=350 sec of testing on a single chamber, the region near occurred, and many of the 50 required engine level tests
the injector remains streak free, and only minor indications have already been performed with no anomalies. Future
of temperature stress inthe form of material delaminations manpower and funding resources will ultimately
in the ablative wrap have occurred. determine if bomb stability testing can continue.
Extensive measurements have been made of gas 9. FUTURE TESTING
temperatures within the acoustic cavity and metal
temperatures near the wall of the cavity. Results indicate, TCA testing at MSFC is set to resume in late 2000 and
with the exception of brief spikes during ignition, that the will include tests of the fleet leader chamber, chambers
acoustic cavity region operates consistently at with known manufacturing anomalies, and tests of an
temperatures well below 10(X)°E These low temperatures improved performance injector design. Additionally.
are largely responsible for the lack of negative structural several tests are planned using a regeneratively cooled
margins {low-cycle fatigue) on the injector. chamber designed and built specifically for the 60K
injector,
Faceplate cooling appears more than adequate for the heat
loads generated during the combustion process. None of Engine test objectives at SSFL are the completion of
the hot fire testing to date has indicated any large thermal development and verification testing of the engine system.
variances due to hot gas recirculation or radial winds near These tests include full and extended duration runs, testing
the copper faceplate. There has been no pitting or with environmental conditioning, margin testing, and
discoloration of any kind and no deterioration of the comer calibration verification. Future testing at either SSFL or
of the faceplate, which forms one side of the acoustic SSC will conclude the test program with certification of
cavity aperture. the final design.
Performance tbr the TCA is based on the characteristic Tests of the gimbal bearing will also be conducted under
exhaust velocity efficiency (r/c,). Chamber pressure engine load conditions in an integrated system test at an
measurements are made ina plane even with the faceplate. alternate location.
and one-dimensional isentropic relationships are used to
derive the nozzle stagnation pressure IPns). Nozzle Prior to the first flight of the X-34. the engine will be
stagnation pressure, along with flow rates calculated from installed into the vehicle, and an integrated static test will
data obtained from inline cavitating venturies, are used to be performed at White Sands Missile Range in New
determine the actual C*. That value isthen ratioed against Mexico.
7
American Institute of Aeronautic and Astronautics

0
r__
0
¢J
°I,,,_
r._
0
_< 0
0
0
<_
O0 0ar.,P
_Z
0
°v"_
<
c_ c,.)

0 >_
0
z
0 0 O< <0
<
0
< ZZ
<_-
Z c_
c_ 0
c_< _0
0 0
_0 0 ¢::::1
<
< _0 _
_ c,.)
NzZ
om o
E-
Z
0
o× D
omm_
Z o< _ X
0 __ _ 0
ZO
<
Z
0
<m<
Z ¢_ _ Nm
N