1、Lessons Learned Entry: 1310Lesson Info:a71 Lesson Number: 1310a71 Lesson Date: 2002-08-25a71 Submitting Organization: MSFCa71 Submitted by: Huu TrinhSubject: NASA/MSFC Army Vortex Chamber Test Incident Description of Driving Event: On February 19, 2002, an incident occurred during the hot-fire testi
2、ng of the Army vortex thrust chamber assembly at MSFC Test Stand 115. This was the fourth hot-fire test of the hardware, but the first test flowing both LOX (liquid oxygen) and RP-1 fuel as the main propellants and GOX (gaseous oxygen) and GH2 (gaseous hydrogen) for the torch igniter. The first thre
3、e tests, which were successfully completed, were hot-fire tests of the GOX/GH2 torch igniter only. For these tests, gaseous nitrogen was flowed through the main injector. The fourth test was to characterize start up transient conditions. The objectives of the vortex chamber testing were to demonstra
4、te the feasibility of vortex chamber technology for a liquid hydrocarbon/liquid oxygen system, demonstrate several ignition techniques (torch, laser and combustion wave), demonstrate two rocket plume measurement methods (emission/absorption and Raman scattering), and characterize the chamber perform
5、ance by means of thrust measurements and species uniformity in the plume flow field. At approximately 5.3 seconds into the automated firing sequence, a catastrophic failure occurred to the test article. A redline cut initiated shutdown at that time. The facility proceeded to follow the normal shutdo
6、wn sequence, which initialized the safeguarding of the facility and test article. The area was immediately roped off with quality monitoring activities. Once the facility safeguarding was completed, it was determined that there were no injuries to personnel, and damage to the test facility was minor
7、. Most of the test article pieces were recovered and provided important information in the investigation and analysis of the incident. A timeline was constructed from the high-speed video film, control sequence data, and both the low and high-speed instrumentation. Due to the lack of a time stamp on
8、 the high-speed video, there were some inherent inaccuracies in correlating the instrumentation data timing with the video. Although all data systems were operational at the time of the incident, the over pressurization/detonation Provided by IHSNot for ResaleNo reproduction or networking permitted
9、without license from IHS-,-,-occurred very rapidly and as a result there was limited evidence of the over-pressurization in the collected data. The test article failed at the mounting bolts as well as the injector and spacer, and hardware was recovered over a large area of the test facility and near
10、by fields/woods. The scenarios developed by the incident investigation team pointed out that three significant events occurred during the start up transients: 1) surface burning of the chamber head end hardware, 2) surface burning of the injector module hardware, and 3) accumulation of propellants i
11、n the chamber. The primary cause of the incident was the propellant accumulation in the chamber during the ignition delay. The first two events are not believed to have caused the eventual over-pressurization. Lesson(s) Learned: Several lessons learned are identified from the subject incident: 1. Th
12、e excessive amount of propellants accumulated in the chamber was the main contributor to the test failure. This accumulation might have been caused by the unintended time delay in the LOX/RP-1 ignition. Consequently, the eventual ignition led to the chamber over-pressurization. a. Although GOX/GH2 t
13、orch igniters have been used in igniting LOX/RP-1 systems, the database and experience of such an igniter for the LOX/RP-1 system are limited. The use of this igniter type for LOX/RP-1 along with its location with respect to the chamber might have caused some delay time in the ignition and created t
14、he first two events as stated in the previous section. In this test sequence, LOX was injected to the chamber prior to the RP-1 injection. It has been speculated that RP-1 might have become frozen during this ignition delay. Past data indicates that frozen RP-1/LOX combustion releases an excessive e
15、nergy creating an over-pressurization in the combustion chamber. Furthermore, due to the nature of the of the vortex chamber flow field, the startup transient is not well understood. For the aforementioned reasons, a better-characterized igniter, such as TEA/TEB, should be applied.b. To detect the i
16、gnition/combustion occurrence, the vortex chamber should be monitored for an increase in pressure. The value of this increase and the detection duration depend on individual test conditions and the chamber configuration. Theoretically, it is possible to estimate these values; however, it is very dif
17、ficult and tedious to accurately predict them. They are normally determined through initial hot-fire tests, which was also one of the objectives of the fourth test. For this particular test, the test duration was set for 2.00 seconds and the incident occurred at 1.38 seconds within the duration. Giv
18、en the lack of data, this duration should be set conservatively at a lower value. If the duration does not allow enough time for ignition, a small incremental step can be added to the waiting time. This method, however, may require more tests in order to study the startup conditions.Provided by IHSN
19、ot for ResaleNo reproduction or networking permitted without license from IHS-,-,-2. Historically, most traditional LOX/RP-1 chambers have used an L* (chamber volume/throat ratio) between 35 to 96 inches. The L* for this vortex chamber was approximately 12 inches. It should be mentioned that one of
20、benefits of the vortex chamber concept is to have a small chamber. At the time of the design, this L* value was selected based on an existing Army chamber configuration for a gelled hypergolic propellant system. Later, Army engineers learned that a longer delay in ignition requires a larger chamber
21、volume in order to eliminate the chamber over-pressurization during the startup. From the aforementioned data, a larger L* value should be considered in the final chamber configuration.3. The Army Vortex test program did not require the IRIG time stamp on the high-speed film . Therefore, it was not
22、active during this series of tests. Normally, the film is only processed in the event of an anamoly. In this incident, the high-speed film provided the best data of what occurred and without the time stamp it required personnel reviewing the film and manually linking the event with the test sequence
23、 and data . This method is not as precise as the IRIG method. For the future, the IRIG time stamp should be the standard without the program having to request this option.4. Normally test articles and facilities are not instrumented for failures. In this case, the thrust measurement system constrain
24、ts did not allow for close coupling of the facility and test article instrumentation and would have been fine for successful steady state tests. In the future the instrumentation requirements should be more carefully evaluated and where possible instrument for a failure while maintaining the program
25、 requirements, schedule and budget.In conclusion, the vortex chamber design concept involves flow vortices and combustion behaviors which are not observed in conventional combustion chambers. In addition, existing numerical prediction methods may not be able to accurately predict or describe all sta
26、rtup transients and other phenomenal aspects, including two-phase flow, propellant mixing, flow field characteristics, etc. A conservative approach in the test article design and test planning should be taken to minimize test failure potentials. A gradual change in the hardware design and testing, l
27、eading to the eventual optimum configuration, should be considered after experience with the vortex chamber concept is gained. Recommendations: 1. The vortex chamber configuration, test conditions and test sequence should be improved to minimize propellant accumulation during the ignition process. T
28、his may include:a71 An increase in the chamber volumea71 A reduction of the initial propellant injection flow rate to the chambera71 No pre-chilling or using gaseous oxygen during the startup to ease the ignition processa71 Shortening the duration of the chamber ignition detection 2. Other igniter t
29、ypes, such as a TEA/TEB, provide shorter ignition delays and therefore might prove to be better ignition sources. Changes in the igniter location should also be considered to minimize the ignition delay time.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IH
30、S-,-,-3. Prior to attempting to derive the optimum combustion chamber design, a conservative approach should be taken in the hardware design and test planning. This should help to ensure that experience is gained with the vortex chamber.4. A time stamp on the high-speed film should become the standa
31、rd practice where applicable.5. The instrumentation requirements should be more carefully evaluated and where possible instrument for a failure while maintaining the program requirements, schedule and budget.6. The Safety Assessment included a Quantity Distance of 160 feet, based on the amount of pr
32、opellant on the test stand. Fragments from the detonation traveled 480 feet. The Safety Assessment for hot fire testing should include fragmentation predictions for a detonation in the test article. The MSFC Industrial safety Department is working to acquire the necessary expertise and develop proce
33、dures for inclusion of fragmentation predictions in the Safety Assessment. This will provide for improved personnel and equipment hazard mitigation techniques.Recommendation(s): Startup transients for vortex chamber designs is not well understood. Therefore, a better-characterized igniter, such as T
34、EA/TEB, should be used in future vortex designs. Vortex chambers require a larger chamber volume in order to eliminate the chamber over-pressurization during the startup. Vortex chamber configuration, test conditions and test sequence should be improved to minimize propellant accumulation during the
35、 ignition process. Possible alternatives include: a71 An increase in the chamber volumea71 A reduction of the initial propellant injection flow rate to the chambera71 No pre-chilling or use of gaseous oxygen during the startupa71 Shortening the duration of the chamber ignition detectionSafety Assess
36、ments for future hot fire testing should include fragmentation predictions for a detonation in the test article. Evidence of Recurrence Control Effectiveness: NADocuments Related to Lesson: N/AMission Directorate(s): Provided by IHSNot for ResaleNo reproduction or networking permitted without licens
37、e from IHS-,-,-a71 Exploration Systemsa71 Space Operationsa71 Aeronautics ResearchAdditional Key Phrase(s): a71 Ground Operationsa71 Pressure Vesselsa71 Range Operationsa71 Research & Developmenta71 Risk Management/Assessmenta71 Safety & Mission Assurancea71 Test & Verificationa71 Test Articlea71 Test FacilityAdditional Info: Approval Info: a71 Approval Date: 2002-10-10a71 Approval Name: Lisa Boninea71 Approval Organization: MSFCa71 Approval Phone Number: 256-544-2544Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-
