REG NASA-LLIS-3396-2010 Lessons Learned - Dawn Ion Propulsion System (IPS).pdf

1、Public Lessons Learned Entry: 3396Lesson Info: Lesson Number: 3396 Lesson Date: 2010-04-6 Submitting Organization: JPL Submitted by: David Oberhettinger Subject: Dawn Ion Propulsion System (IPS) Lessons Learned Abstract: Although the Dawn IPS was planned for the most part as a built-to-print inherit

2、ance of the DS1 design, the Dawn project encountered significant cost and schedule escalation and technical difficulties. The IPS contractor suffered from management instability and from capabilities lost over the 6-year lag between the two projects. For relatively novel technologies, it may be nece

3、ssary to re-qualify the capabilities of the contractor as well as re-qualifying the design for the new mission application. Description of Driving Event: After its September 2007 launch, the Dawn spacecraft employed a solar-powered ion propulsion system (IPS) to gain the additional velocity needed t

4、o reach Vesta and Ceres, and it will use the IPS to spiral to a low altitude orbit around these asteroids. Compared to chemical rockets, ion engines make very efficient use of onboard fuel because the propulsive energy is derived from the sun. Solar electric propulsion (SEP) technology for navigatio

5、n beyond Earth orbit was successfully demonstrated from 1998 to 2001 by the Deep Space 1 (DS1) mission, and Dawn planned to use the same IPS design (Figures 1 and 2).Figure 1. Deep Space 1 is lifted from its work platform at NASA Kennedy Space Center, providing a close view of the IPSFigure 2. Hot f

6、ire test of the Deep Space 1 IPSThe development of the Dawn IPS by the NASA/Caltech Jet Propulsion Laboratory (JPL) proved much more difficult and expensive than expected. The principal unanticipated problem was a degraded ability to manufacture the DS1-legacy components, principally the ion thruste

7、rs and the Power Processor Units (PPUs) (References (1) and (2). The inherited DS1 components had evolved through informal engineering processes more typical of experiments than development of configuration-controlled flight hardware. Although the Dawn IPS contractor for the ion thrusters and PPUs w

8、as the same company that had built the DS1 hardware, and the Dawn thrusters and PPUs were proposed as build-to-print copies of the DS1 designs, the contractor encountered significant management and process problems in delivering the Dawn flight hardware. The contractor problems, which likely were ex

9、acerbated by the 6-year lag between these two projects, included: 1. JPL efforts to negotiate a fixed-price or incentive-based contract with the Dawn IPS contractor failed. The resultant cost-plus-fixed-fee contract was overrun by almost 100 percent, and the flight hardware was delivered 8 months la

10、te. 2. The Dawn IPS contractor for the ion thrusters believed it could reproduce the assembly processes for the DS1 thrusters, but major cost escalation during fabrication, assembly, and testing of the Dawn IPS revealed that the time lag between the two projects had degraded their capabilities. Also

11、 the vendor that fabricated the DS1 ion thruster plasma screen was no longer in business, and the vendor that performed chem-etching of the DS1 thruster grids did not bid on the Dawn thruster grids. No shock testing of the DS1 thruster had been performed except at the less critical spacecraft level

12、 3. The DS1 PPU assembly procedures and other information needed to reproduce the hardware had not been adequately documented or retained. 4. Over the 2-1/2 years of the Dawn IPS build, the IPS contractor for the ion thrusters and PPUs cycled through four general managers and four Electric Propulsi

13、on department managers. Each upper-level manager in the contractors corporate unit was fully aware of a pending sale of the unit by the parent company. These managers were motivated to please their parent companys customers, and there were frequent delays due to reassignment of Dawn resources to the

14、se other projects. 5. A total of 40 IPS component and subsystem design reviews, plus the hardware reviews, produced a huge number of action items- 630. The number would have been more manageable had the required documents- technical requirements documents (TRDs), interface control drawings (ICDs), m

15、echanical interface control drawings (MICDs)- been completed on time. 6. The Dawn IPS interface with the spacecraft was complex. Assembly, Test, and Launch Operations (ATLO, aka spacecraft Integration & Test) was delayed pending receipt and installation of the thermal hardware because responsibility

16、 for thermal hardware (e.g., heaters, blankets, and platinum resistance thermometers (PRTs) attached to the IPS components was divided between JPL and the Dawn spacecraft system contractor. 7. Two test-as-you-fly exceptions went undetected prior to launch and were only identified because of problems

17、 encountered during the initial spacecraft Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-checkout activities. The latch valve masking should have been installed for the second phase of spacecraft thermal-vacuum testing, and a gimbal problem was not

18、 identified due to insufficient testing of the gimbals with the spacecraft prior to launch.These management problems affecting the Dawn IPS development were accompanied by many IPS design implementation, fabrication, and test technical problems with the ion thrusters, grids, cathode magnets, PPUs, d

19、igital control and interface units (DCIUs), xenon tank (Reference (3), and xenon feed system (XFS). References: 1. John Brophy, “Dawn IPS Lessons Learned“ presentation, December 4, 2007. 2. John Brophy, “DAWN Ion Propulsion System: Project Element Managers Report on the Development of the Dawn Ion P

20、ropulsion System,“ JPL Document No. D-41251, May 29, 2008. 3. “COPV Propellant Tank Failure on the Dawn Spacecraft,“ NASA Lesson Learned No. 1777, NASA Engineering Network, March 7, 2007. http:/www.nasa.gov/offices/oce/llis/imported_content/lesson_1777.html 4. “Interface Control and Verification,“ N

21、ASA Lesson Learned No. 0569, NASA Engineering Network, October 9, 1997. http:/www.nasa.gov/offices/oce/llis/imported_content/lesson_0569.htmlLesson(s) Learned: Although the Dawn IPS design concept was fundamentally sound and flight worthy (Reference (2), p. 141), JPL experienced significant cost esc

22、alation (Figures 3 and 4) and technical difficulties in implementing the design. Figure 3. Cost growth by major Dawn IPS activity. Figure 4. Cumulative Dawn IPS workforce.Recommendation(s): The need to re-qualify an inherited spacecraft subsystem design for application on a new mission is widely rec

23、ognized, but careful consideration should be given to the need to re-qualify the capabilities of the subsystem contractor: 1. For development of a subsystem that is dependent on novel technology, persevere in devising an incentive-based contract acceptable to both parties. 2. When inheriting a subsy

24、stem design that has not been built for many years, consider verifying manufacturability by fabricating a qualification model. 3. When inheriting a design for a build-to-print, ensure that all of the essential resources are actually available to be inherited- drawings, tooling, fabrication procedure

25、s, models, personnel, etc. 4. Assess whether the contractor has the management stability and dedication to the project needed to meet contract commitments. 5. Place a priority on ensuring that TRDs, MICDs, and ICDs are completed and released in time for the technical reviews that require the documen

26、ts. 6. The NASA Center should provide all of the supplementary hardware attached to the Center-provided components to be delivered for integration and test (I&T). To mitigate I&T problems with controlling a complex subsystem interface, assure that the spacecraft contractor provides a dedicated indiv

27、idual (see Reference (4) to oversee the spacecraft side of the interface. 7. Exercise great care in the identification and elimination, where possible, of test-as-you-fly exceptions.Evidence of Recurrence Control Effectiveness: JPL has referenced this lesson learned as additional rationale and guida

28、nce supporting Paragraph 6.6 (“Engineering Practices: Inheritance“) in the Jet Propulsion Laboratory standard “Flight Project Practices, Rev. 7,“ JPL DocID 58032, September 30, 2008. Documents Related to Lesson: N/A Mission Directorate(s): Science Additional Key Phrase(s): Program Management.Acquisi

29、tion / procurement strategy and planning Manufacturing and Assembly Engineering Design (Phase C/D).Spacecraft and Spacecraft Instruments Engineering Design (Phase C/D).Propulsion Additional Info: Project: Dawn Approval Info: Approval Date: 2010-09-07 Approval Name: mbell Approval Organization: HQ Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-