United Space Alliance (USA) and Michoud Assembly Facility (MAF) engineers have installed the replacement external LH2 Feed-through connector inside Atlantis' ET-125, marking a key milestone ahead of Atlantis' NET February 7 launch.
The modified element of hardware is the "most likely" solution to the Engine Cut Off (ECO) system anomalies that have delayed the mission - as tests continue at the Marshall Space Flight Center (MSFC).
The replacement of the external element of the LH2 Feed-through connector was called after data - gained via Time-Domain Reflectometry (TDR) equipment - showed the sensor circuits issues common to the aft ET Feed-through connector system.
In total, LH2 ECOs 1 to 4 and a 5 percent level sensor showed anomalous readings during STS-122 / ET-125 tankings (two launch attempts and one tanking test), with the open circuit conditions occurring at times consistent with rapid cryogenic conditioning of connector hardware.
Engineers do not have a confirmed root cause, though a combination of physics-based analysis, reviews of ET-120 hardware data, and ET-125 investigation data were used to determine most probable candidate - namely the External Plug / Feed-through Receptacle Pin Contacts.
With only the external element of the feed-through hardware deemed at fault, engineers were given the green light to conduct work at the pad. Rollback would have been required, had the data shown the internal connector was the culprit.
Engineers worked off a complex and delicate plan for foam TPS (Thermal Protection System) to be removed incrementally from ET-125, in order to provide unobstructed access for in-situ inspections.
While the ultimate goal was to remove the external connector/plug, engineers carried out a detailed dissection of TPS within J-box surrounding the connector) so as to identify any potential contributors to the problems.
The connector and plug assembly was then removed intact to perform X-ray and visual inspections, while continuous circuit monitoring was performed during hardware manipulation. This presented early data points on the health of the connector, prior to it being shipped to MSFC (Marshall Space Flight Center) for extensive testing - which is still ongoing.
The in-situ inspections at the pad were also used to mitigate potential causes associated with internal plug and feed-through connector (contamination, bent pins, damaged sockets, etc.) due to the lack of an absolute confirmation of root cause.
'Feed Through Pins - Remote: Open circuit requires break in pin - Recovery of circuit performance not possible. Ambient and cryogenic leak testing performed on ET-120 feed through connector ? No issues identified,' noted the results from the in-situ inspections.
'Wire Harnesses / Socket Crimps - Remote: Unlikely to occur simultaneously. Design and manufacturing processing requires wire slack during assembly. Broken connections unlikely to 'heal' during subsequent tanking and recover during drain back. Nanofocus NDE of ET-120 hardware showed no evidence of crimp degradation. Acceptance testing performed to verify connection (21 lbs).
'Internal Plug / Feed Through Receptacle Pin Contacts - Remote: Minimal relative motion between socket and pin contacts. CTE-induced effects increase contact pressure between sockets and pins. CTE- induced effects have minimal effect on pin/socket engagement (around -0.007''). Sockets fully locked to retaining ring. Thermally-induced delta pressure loading within connector cavity not possible (GHe purge / LH2).
'Teflon grommet material thermally stable at cryogenic temperatures (no cracking). Less chance of contamination as compared to external plug. Contamination due to solid air / moisture not possible (GHe purge). Other contamination possible but unlikely to remain following 3 LH2 loadings.
'ET-120 internal plug flown with no issue. Inspection and testing of ET-125 internal plug (in-situ) and feed through replacement being used to mitigate potential causes associated with internal plug and feed through (contamination, bent pins, damaged sockets, etc.). No anomalous conditions identified.'
Interestingly, one flag on the internal connector was noted on one presentation, relating to trace amounts of Krytox, observed on internal connector through KSC lab analysis. However, it was noted that it was not located in location that would result in open circuit.
'Swab samples obtained from ET-125 feed through connector analyzed by MSFC M&P lab using an attenuated total reflectance (ATR) infrared technique,' noted the Marshall presentation. 'No apparent differences between the control swab and swabs used on the pins, the glass surfaces and the housing surfaces.'
The results - gained via previous analysis and their latest inspections - left engineers with just the one main culprit.
'External Plug / Feed Through Receptacle Pin Contacts - Most Likely: Significant relative motion possible between socket and pin contacts. CTE (Coefficient of Thermal Expansion)-induced effects increase contact pressure between sockets and pins. CTE-induced effects can cause around -0.019'' decrease in pin engagement. Sockets allowed to move towards receptacles.
'Review of ET-120 plug showed evidence of grommet material / socket retention degradation. Thermally-induced delta pressure loading test demonstrated at cryogenic temperatures (vacuum). Greater chance for contamination as compared to internal plug
'Contamination due to solid air / moisture probable (cryopumping through harness). Moisture proof material cracked on ET-125. Most likely a result of out-gassing. Other contamination possible. Bench testing demonstrated open circuit condition with ice contamination and relative motion between pin / sockets.
'Root cause of ET-125 anomalies most likely attributed to separation of plug socket / feed through pin contacts. Contact separation caused by thermally-induced effects (CTE, delta pressure, contamination, relative motion between parts) result in open circuit at external plug / feed through connection.'
With these findings collated early in the troubleshooting process, engineers devised a plan of action, one that would mitigate the aforementioned issues of the pin contacts separating from the plug socket. Ironically, they would receive a welcome break, via a ready-made solution from the Atlas/Centaur program.
Used to mitigate intermittent circuit anomalies, isolated to connection between external plug and feed-through receptacles on the Atlas vehicle, Lockheed Martin engineers had devised a 'pin soldering' solution, one that has proved to work without issue on their subsequent Atlas flights.
'ECO Circuit Anomaly Root Cause Mitigation Plan: Pursue design solution that addresses most likely root cause by implementing solid state connection between external harness and feed through connector. Pin soldering provides proven technology and can be screened prior to implementation (NDE, acceptance test loading).
'Technical Basis for Redesign: Eliminates significant contributors to most likely root cause of STS-122 ECO circuit anomalies. Contamination / relative motion effects between external plug socket and feed through pins eliminated.
'Pin soldering used on Atlas/Centaur program. Similar connector design as used on ET. Same pin / socket material and pin diameter. Testing performed on Atlas/Centaur program to verify design performance. Pin / socket tensile strength and qualification testing with cryogenic cycling (LN2) with vibration environments. Similar solder design (Atlas/Centaur) successfully flown with no issues (> 12 missions).'
The process of soldering the pins involved the requirement to solder the 'split tine socket' onto the feed-through pins, requiring the modification of the internal contact, by removing the outer sleeve (silver hood) and removal of gold plate from pins and sockets - which provided a better surface for the solder.
Tests were carried out on the soldered pins, including Cryo cycle testing to -407 degrees F, and then pull tested to the point of failure - seen at the crimp, as opposed to the solder joint - after 39lbs worth of pull.
'Atlas / Centaur Design and Usage: Atlas/Centaur used a Capacitance Level sensor system for their LH2 and LO2 tanks. Contact resistance critical to Atlas/Centaur operation (Operating voltage and current is well below ECO system). Cryo feed through located on forward dome of LO2 and LH2 tanks,' added related information on the Atlas approach.
'Cryo feed through similar to ET design with the exception is was a twelve contact version of the Deutsch connector. Internal plug was teflon insert. External was silicone insert. Feed through was glass/teflon insert. Plug sockets were originally 'leaf spring' design. Grease was used in external connector to prevent moisture affecting the circuits.'
So the plan was in place. Engineers at MSFC got to work on the replacement connectors for both ET-125 and ET-126, using the Atlas approach of soldering the pins, whilst an additional mitigation plan was worked on the replacing of the silicone grommet with a teflon insert - following the findings of cracks on the ET-120 silicone grommet.
Additional areas of mitigation were also put into play, with an increase to the wire gage and a replacement back shell design.
'Replace 22 gage wire with 20 gage. Add wire strain relief to each conductor after the crimp and before backshell strain relief. Replace RFI backshell with strain relief backshell. Contact of overall metallic braid retained through solid state connection to straight backshell,' added a presentation on additional tasks.
'Background / Reason for Design Selection: Larger wire provides better contact, wire support and crimp. Larger gage wire suggested in connector specification. Meets approved crimp configurations for contact and splices. Wire strain relief and straight backshell used to mitigate stresses on solder contact due to restrained wire movement.
'Current design strain relief only through silicone insert support of wires. RFI backshell does not support wire. Verified through vibration Qualification Test.'
A dual troubleshooting approach is continuing, as engineers attempt to bring Atlantis back into a launch configuration, while tests continues on the removed external feed-through connector at MSFC.
At the pad, the installation of the replacement connector (Saturday night/Sunday morning) marked a milestone en route to the February 7 launch attempt. Engineers were tasked with the removal of the vertical strut cable tray, J-box cover and SRB PAL TPS ahead of numerous tests on the replacement connector.
The process of installation took slightly longer than planned, due to issues with 'wire and splice issues.'
The connector's installation into ET-125 - involving the splicing of wiring - is being continually monitored by electrical equipment at the pad. Leak checks will follow, along with further electrical testing, prior to the reapplication of foam TPS to the work area.
'TPS Re-Application Materials and Processes: TPS shall be replaced using current materials and processes. Vertical strut cable tray and fairing. Ablator closeout (hand pack) of new trays brought from MAF. PDL pour under and in fairing around connector. Segmented manually spray BX-265 on cable tray cover bolt hole runs and J-box,' noted a MAF presentation.
'SRB PAL Ramp (BX spray foam). Manually sprayed BX-265 in three segments. Manually spray BX-265 closeouts configuration on cable tray covers will be slightly different due to R&R. Spay segment bondline locations. Manual sprays will be Vertical versus Horizontal. Vertical spray demonstration performed.'
Once the foam has been re-applicated to the work area, MAF information notes it will typically take two weeks for the foam to 'cure' - which allows for plenty of time ahead of the current launch target of February 7.
This is the final element of the pad work, with all other areas of implementation scheduled for completion by Jan 27.
Read More...
Summary only...
