Full report

Original page 1

. Monthiy Progress Report =31 fw the ~ ~ HIGH RESOLUTION~~ CAMERA~ (HRC)R A Princrpaf investigator Stephen Murray Sept 2004 Prepared by: Smiihsonian k r o p hyricaI 0bservafory 60 Garden Street Cambridge, MA 02138 50944

Original page 2

HRC Monthly Progress Report #31 Sept 2004 1 Management and General Status The HRC expenditures are at or below the budget estimates for this month. The Mission Support and Operations and Science Activities are detailed below. 2 Mission Support and Operations During this reporting period, the HRC mission operations team has supported the following the routine support activities: Real Time monitoringof the operationalstate of the HRC instrument during each real time pass. 0 Processing and reviewing the HRC portion of the weekly command loads. 0 Acquisition, archiving, and generation of HRC science and engineering productsfrom flight telemetry. 0 Investigatinglong term trends and analysis of HRC science and engineering products. In addition to the routine support activities listed above, the operations and engineering team has supported the following tasks: 0 POC modifications: We have completed work on modifyingthe HRC POC electronicssystem in anticipation of integrating the ASVT with the HRC POC. We are now working with the FOT/ASVT team on integrating the HRC POC into the ASVT. 0 Failure Analysis of HRC Flight Relay We have completed the Failure Analysis of HRC Flight Relay. This has been incorporated into a flight note being prepared by Mike Juda. (See Attached) 3 Science Activities The HRC science team has continued to support the Chandra operations during this reporting period. 2

Original page 3

Orginization: Smithsonian Astrophysical Observatory 60 Garden St. Cambridge MA, 02310 Program/Project: Chandra High Resolution Camera Title: Technical Memorandum Failure Analysis of HRC Flight Relay Project HRC Start Date:Time: Document Type Technical Memorandum End Date:Time: File hrc.relay-failure-analysis.tex Test Conductor: Revision R e v : 57 Prepared By G. Austin Make Date October 6, 2004 Document Configuration Management Information: Project Directory: hrc/doc/TM/shutter R e v : 57 Id : hrc.relayfaikre,nalysis.trn.tex572004- 10- 0616 : 47 : 1 4 Z j h c U RL : sun + ssh : //jhc@netsi.c f a.harvard.edu/d0/hrc/svn/hrc/doc/TMlrelay~ailure,nalysis/hrc.relayfailureanalysis.tm.t U

Original page 4

Chandra/HRC Failure Analysis of HRC Flight Relay R e v :57 Contents 1 Introduction 2 Background 3 Preliminary Failure Analysis 1 1 1 3.1 Residual Gas Analysis (RGA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3.2 Decapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3.3 High Humidity (50% RH) and Temperature Extremes (75C, 40C) . . . . . . . . . . . . 2 3.4 Oxidizing Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 4 Preliminary Conclusions 2 4.1 FailureAnalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 4.1.1 Residual Gas Analysis (RGA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 4.1.2 Moisture Content and Relay Operation . . . . . . . . . . . . . . . . . . . . . . . 3 4.1.3 Particle Contamination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 4.1.4 Oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 5 Conclusions 6 Recommendations 7 References List of Figures List of Tables 4 4 5 1 Residual Gas Analysis results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2 Relay Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3 RGAAnalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 .. SA0 CONTENTS II

Original page 5

Chandra/HRC Failure Analysis of HRC Flight Relay Rev : 57 1 Introduction Following the determination that the -Y shutter motor select relay failed, a failure analysis was initiated t o determine the cause of the failure. The intent was to provide a basis for deciding if it was possible I t o recover the proper operation of the relay, as well the impact of the failure on the continued use of the other motor select relays. 2 Background The Genicom relays (S311P754/09-003; DC9536) used for the Flight HRC motor selection were qualified t o GSFC specification S311P754 including PlND test per MIL-R-83536 for detection of fine particles. These relays have an excellent reliability history in space programs with no failures. A review of the documentation received with Chandra relays showed that the lot for Chandra was selected from a lot of 24 relays that had two failures during processing (1-vibration, 1-seal). The relays have gold plating over nickel underplate on contacts; such contact coating has no history of flaking and causing particles. Particle contamination is one of the major causes for relay failure. The Proof of Concept (POC) HRC electronics utilize the same relays except for additional screening imposed by GSFC S311754 requirements. The POC system has been in operation since 1994. 3 Preliminary Failure Analysis Unfortunately there was only one flight relay in spare stock. So preliminary analysis was conducted on POC relays and MIL GRADE relays from Genicom and Tyco. 3.1 Residual Gas Analysis (RGA) An RGA was performed on a POC relay (3SBH1347A2, Genicom) to determine contamination levels in the relay package. The data is presented below (1). 11 Element I % I I L J Nitrogen (N) 99+ Carbon Dioxide (CO2) 0.113 1 Moisture (HzO) I 0.177 Table 1: Residual Gas Analysis results S A 0 1

Original page 6

ChandralHRC Failure Analysis of HRC Flight Relay Rev :57 3.2 Decapping The relay was decapped by grinding, taking care not t o introduce particles within. The decapped relay failed a switching test in a similar mode as observed in the flight HRC. The failure was found t o have been caused by distortion of the assembly during decapping. After repair the relay was found t o operate normally. A particle contamination was simulated by inserting a .002” thick piece of paper in the contact mechanism. It produced the failure mode observed in orbit. Particle contamination was thus established as one of the possible failure modes in orbit. 3.3 High Humidity (50% RH) and Temperature Extremes (75C. 40C) POC and Genicom MIL GRADE relay were subjected t o extremes of temperature (75C. 4OC) and high humidity (50% RH) for varying periods of time. No failures were observed. 3.4 Oxidizing Environment The RGA suggested the possibility of an oxidizing environment within the relay package, consisting of H, N, and 0 (constituentsfound in nitric acid fumes). A decapped POC relay wa5 subjectedt o 5% nitric acid fumes at 25C for 12 hours, resulting in a failure similar to that observed in orbit. Six consecutive switching attempts were required t o correct the relay operation. (Extended exposure t o fumes ( 300 hours) caused heavy oxidation of copper elements a worst-case scenario). Thus oxidation in gaseous environment within the relay package was established as a source of particles and a possible failure mechanism. 4 Preliminary Conclusions The analysis suggested the following factors as probable cause for the HRC failure in orbit: 1. Particle contamination 2. Oxidizing environment in the relay package as the primary source for particles. 4.1 Failure Analysis On the basis of the findings of preliminary analysis, failure analysis was conducted on following relays: 4.1.1 Residual Gas Analysis (RGA) The results of the RGA are shown in Table 3. SA0 3.2 Decapping 2

Original page 7

Chandra/HRC Failure Analysis of HRC Flight Relay Rev : 57 Relay Part No./Date Code Manufacturer Status 1 S311P754/09-003/9536 Genicom 2 3 SB145134K / 9452 3 3SBH1665A2 / 2002 Flight Genicom MIL GRADE Genicom MIL GRADE b 1 4 i 3 SB145134K1’/ 2003 Tyco I MIL GRADE Table 2: Relay Identification Elements Relay 1 Relay 2 Relay 3 Relay 4 Argon (Ar) .058 .063 .083 Carbon Dioxide (C02) Moisture (H20) .258 12.10 .75 II Hydrogen (H) I .32 I .174 I .161 I .167 1 Table 3: RGA Analysis The data confirmed the previous finding of an oxidizing environment. In addition, it showed significant moisture content within the flight relay. 4.1.2 Moisture Content and Relay Operation To study the effect content (up t o 12.1%), a measured amount of moisture (0.87 cc), determined from the RGA data and dimensions of the relay package was introduced in decapped relays (#l,3 and 4). These relays were selected because of their high moisture content. The relays were tested a t 25C and were found operational. The relays were then stored a t 45C in quiescent state for 22 hours. At the end of 22 hours, the relays operated normally a t 45C. Relays were then stored a t 60C in quiescent state for 2 hours. Tests a t 60C, 40C and 1OC showed normal operation. It was therefore concluded that moisture content up t o 12% had no effect on relay operation under normal humidity condition. 4.1.3 Particle Contamination Siltek, a diamond powder with particle size of 0.5 (m was used for contamination of two relays (#3 and 5). Three particles were deposited on top of the decapped relays, and the relays were then tapped lightly t o allow random, free fall of the particles into the contact assemblies. Both relays failed in a similar mode as observed in orbit. Relay #3 was then stored in quiescent state a t 40C for 150 hours in an oven that tended to vibrate whenever the blower fan was energized. At the end of 150 hours, the relay exhibited the same failure mode. So particle contamination as a source of relay failure was S A 0 4.1 Failure Analysis 3

Original page 8

Chandra/ HRC Failure Analysis of HRC Flight Relay Rev :57 established. 4.1.4 Oxidation The objective of the oxidation tests was t o simulate the long-term storage condition experienced by the failed YSMS relay in Chandra, and t o prove oxidation as the primary source for particles within the relay package. Two relays, #2 and 4, were exposed t o nitric acid (5%) fumes after initial switching tests in a quiescent condition for 24 hours. The switching test at 24 hours showed normal operation, with the appearance of thin oxide layers. The relays were then subjected t o nitric acid fumes for 150 hours. At the end of the 150 hours relays were found t o be heavily oxidized worst case situation in long-term quiescent storage. The oxide layers are shown in the attached photographs. This strongly suggested the growth of oxides on copper elements of the contact assembly over a long period of exposure t o oxidizing environment and the oxidation as the primary source of particles in the relay. 5 Conclusions The RGA of the Chandra flight relay showed significant moisture content (0.92%) in the relay package. Such an environment a t room temperature is considered ideal for oxidation of copper and copper alloys1 The possibility of particlesfrom external sources (such as gas supply, processing) was discounted because the relays were processed and tested per GSFC 5311574 specification including PlND test. Additionally, the documentation received with the relays showed no anomalies. The failure analysis concluded that the relay failure in orbit was likely caused by particles; the particles being generated as a result of oxidation of the copper elements of the contact assembly during long term storage (4 years) in oxidizing environment. This failure mechanism suggests that all the HRC motor select relays on- orbit have the potential to fail in a similar fashion. 6 Recommendations The HRC door mechanism should not be exercised. The door requires the its motor select relay t o operate properly t o both close and open the door. The door does not have any failsafe features in the mechanism in the event that the door was closed and the motor select relay failed a t that time. This would leave the HRC door permanently closed and render the HRC useless for the remainder of the . mission. The HRC shutters do have a failsafe mechanism t o remove the shutter blades from the field of view in the event that a motor relay fails. (Note that the failsafe mechanism is a one-shot, non-yesettable mechanism). At the current time, the -Y shutter relay has failed and is presumed t o still be failed. S A 0 4

Original page 9

Chandra/HRC Failure Analysis of HRC Flight Relay Rev : 57 Periodic attempts to select the -Yshutter motor could be done without actually deploying the shutter to see if the relay can be recovered. There is no known problem with the +Y shutter. The only drawback t o the use of the shutters and reliance on the failsafe features t o overcome a relay failure is the use of Starsys wax actuators in the failsafe mechanisms. Actuator of a similar type experienced failures due to burn-out of the heater prior to full actuation. Such a failure would result in the shutter blade still in the field of view, again rendering the HRC useless. This problem with the wax actuators was known prior t o launch, but a decision was made not to replace them, but t o overcome any potential problem operationally. The plan was to develop a procedure whereby the actuator would be turned on for brief intervals t o allow the wax to heat up slowly without having t o leave the heater on continuously. Heater elements of the type used in the HRC actuators were obtained from the manufacturer of the actuators for use in laboratory testing to determine the proper heater duty cycle. These tests were never conducted and no procedure was ever tested on the flight instrument. There is, therefore, some risk that the failsafe function of the shutter mechanisms could fail t o function properly in the event of a shutter motor relay failure. The HRC calibration mechanism is of no interest as the strength of the Fe55 source has diminished t o the point that it is no longer useful. 7 References 1. Handbook of Materials and Techniques for Vacuum Devices, Walter H. Kohl, Rheinhold Publishing. S A 0 5