NASA solving its ECO problems

by Chris Bergin

Last week’s PRCB meeting once again dismissed rationale for reducing the LCC (Launch Commit Criteria) to allow a launch to proceed with one failure out of the four primary ECO (Engine Cut Off) sensors in the Shuttle External Tank.

Believing they now have a greater understanding of the reasons for the anomalous readings the vital sensors have been showing since the lead up to STS-114, NASA will proceed with the LCC as 4/4, as they close in on the launch of STS-121. NASA also gave its reasoning for believing a manufacturing issue is to blame for its ECO problems.

The resulting information was collated from information presented at last week’s PRCB meeting, which has been available on L2 since the weekend. To join L2, click the advert —>

**Click here for selected images and the article’s related thread**

The ECO sensors are responsible for the indications on when the External Tank (ET) is running out of propellant, thus allowing the Orbiter to gauge when to power down its three main engines. Too early, and the Shuttle might not make it to its planned orbit, too late and the Orbiter could be ripped apart as the engine turbines overspin without their supply of fuel.

A presentation, written by Boeing, NASA and the United Space Alliance and acquired by, shows the discussion that has been taking place behind the scenes on what would be the best course of action, both ahead and during the first loading of the super cold propellants into ET-119 – the tank that is part of the STS-121 stack.

‘Given the experience with the recent LH2 ECO anomalies, review proposed LCC options for a failed WET LH2 ECO sensor during STS-121 cryogenic loading,’ the document opened with.

Interestingly enough, the document reveals that STS-114, which had ECO sensor issues throughout, actually launched with a 3 out of 4 LCC criteria – available to proceed with a launch should a particular failure occur.

Each tank has two main ECO sensors and two redundants in the LH2 tank, with the same set-up for the LOX tank, though the LO2 sensors, which work in the same manner as the LH2 sensors, are actually located in the liquid oxygen feedline on the orbiter side of the ET-Orbiter interface. With such a system, if one fails, another will still be there for redundancy.

‘MMT approved launching with 3 of 4 ECO sensors in combination with ECO wiring swap on an LCC waiver for STS-114 if certain conditions were met.

‘If LH2 ECO 1 or 3 fails – SCRUB. If LH2 ECO 2 fails WET – GO. IF LH2 ECO 4 fails WET – GO.’

Noted as rationale for the 3 out of 4 LCC, the document notes: ‘Testing to date provides confidence in health of redundant components. Fail to WET of a single sensor is benign failure mode. Flight software requires 2 DRY indications to command MECO (Main Engine Cut Off). Can handle two false fail to WET if no other failures. Flight Software rejects a single fail to DRY (different but more critical signature) prior to start of checks (arming mass)

‘Three failed WET sensors would still require an additional failure (SSME or ET problem causing need for LH2 cutoff). System is biased to a LOX cutoff – always load fuel bias to avoid oxygen rich shutdown.

‘System design is for LO2 low-level cutoff, likelihood of LH2 cutoff much greater than 3 sigma dispersions + launching at end of launch window. System redundancy was restored by recent modification so no single failure in supporting systems can take down multiple sensors – original reason for 4 of 4 was lack of isolation.’

Potential causes of the multiple issues seen prior to STS-114’s launch had been blamed by some on electrical issues with the Orbiter’s Point Sensor Box. However, the document lists seven of the 18 reported issues that show ‘low confidence’ in this being the actual reason for the failures. However, it appears NASA aren’t yet able to give definitive reasoning for the problems seen last summer. An understanding, however, is listed in the second half of this article.

‘Upon completion of troubleshooting, we will have narrowed the problem to specific areas: MDM and associated data path. Power and associated circuit. Point Sensor Box Functionality. Card, ECO Sensor and associated wiring.

‘Exact root cause of failure on the day of launch will not be known, just narrowed to a given area of the circuit. E.g., most probable cause of STS-114 launch attempt anomaly is a cryogenic tanking induced, high impedance/open in measurement pat. Unable to narrow down to PSB, Orbiter wiring, Orbiter connectors, ET wiring, ET connectors, or ET sensor.’

STS-121 Mission T-Shirt. Use discount code ‘nasaspaceflight’ when ordering from Countdown Creations & receive 5% off!

Should a repeat of the issues seen prior to STS-114’s launch occur again, NASA has a plan, although some of the options are undesirable to say the least given the delays that would be associated.

‘Option 1: 4 of 4 LH2 ECO LCC. De-tank and perform troubleshooting similar to STS-114. Option 2: 4 of 4 LH2 ECO LCC, with pre-planned contingency procedure to allow 3 of 4 LCC on the day of launch. Troubleshooting for a failed sensor will be performed per S00E00.227.

‘Option 3: 4 of 4 LH2 ECO LCC, with pre-planned contingency procedure to allow 3 of 4 LCC if failure repeats on the same sensor on 24 hour scrub and re-tanking. If a failure occurs on a different sensor during de-tank or during re-tank, perform troubleshooting similar to STS-11.

‘If either option 2 or 3 is chosen, details of the pre-planned contingency procedure would need to be developed by the community and approved through the appropriate channels.’

The theme that runs throughout is to have some allowance for a single ECO sensor failure, should it mirror expected data from such an event. However, the 4 out of 4 LCC will be the basis of their launch attempt.

‘Suspect reliability of LH2 ECO circuit requires full redundancy for launch.’

Order Now - Countdown CreationsHowever, if ‘pre-planned contingency procedure to allow 3 of 4 LCC if failure repeats on the same sensor on 24 hour scrub and re-tanking. Perform 24 hours scrub and re-tan. If failure only repeats with same signature – GO.’

As far as finding causes for the ECO issues, another document that was presented to last week’s PRCB did make an interesting note on a matter that was touched upon by Shuttle manager Wayne Hale – notably the ‘swage’ issue.

While conducting tests on ET-118’s ECO sensors (now being confirmed as to be swapped out following the exclusive article published on this site), the document noted: ‘Documented tooling issues in the time frame associated with 1996 sensor swaging process. Resistance shift noted on ET-119 – sensors built in 1996 time frame.’

Such issues with the ‘tooling’ of these sensors, all associated with the recent problems in the ETs with ECO sensor ‘phase/resistance shifts’ and failures points to the problem being with the sensors themselves, as opposed to any of the wiring or Point Sensor Box on the orbiter.

Historical records with the Michoud Assembly Facility (MAF) go some way to backing up that reasoning, with a list of ECO issues noted after a certain date and ET production manifest.

‘MAF trend data available for ET-64 and subs (expanded from 03/13/06 discussion on ET-119). ET 64-91 and ET 96-100 sensors were from pre-1996 sensor build period. No occurrences of circuit resistance shifts noted during MAF acceptance tests

‘ET 92-94 and ET 101-121 were from 1996 sensor build. 12 occurrences of circuit resistance shifts noted during MAF acceptance tests. 10 sensors removed / replaced in process at MAF. ET-116 shipped with known 2-ohm shift on LO2 100%(-) loading sensor (used for prelaunch, Crit 1R non-CIL) – No performance anomalies noted during STS-113 cryo loadings (3). ET-119, LH2 ECO #3 – Sensor removed / replaced during post RTF 2 processing.’

Delving deeper into the possible causes of these anomalies shows NASA has investigated in-depth, with some disturbing revelations at the fabrication level.

‘Excessive wear on swage tool – No criteria for tool wear on drawing. Modifications had been made to the element winding tool without approval by Lockheed Martin. The supplier had used NASA tools, such as the swage tool bed, on other products, in violation of contractual agreements.’

Thus, it was concluded, that known sensor swage process and tooling control issues noted on sensors fabricated during 1996 timeframe is a possible cause of ‘erratic’ sensor behaviour.’

All in all, the replacement of the sensors in both ET-119 and ET-118 make the possibility of an anomaly during the tanking stage of the STS-121 – and subsequent tankings – highly unlikely.

NASA, it would seem, has found its problem with the ECO sensors.


Related Articles