-
#340 by jimvela on 09 Dec, 2007 22:01
-
Quote
jimvela - 9/12/2007 3:38 PM
This document has a great deal of analytical detail, I'll go study it.
Mods feel free to kill this post if it's excessively lengthy or off-topic...
A quick look reveals the text below. Hopefully someone else will comment further.
My reaction is that transorbs are regularly misused by EEs (I've seen this myself), but that this issue feels to me like some kind of wiring/signal path problem.
The rant I originally quoted above looks to be just that, a rant.
Just because the Transorb is rated for a higher voltage than nominal operating range of the parts in the circuit doesn't make it wrong- many circuits with high input impedances get AFU if there is enough leakage current from the transorb into the circuit. I've personally seen many more problems where a transorb value was lowered by some 'expert' causing enough leakage current to completely hose a design...
This is in my limited knowledge (no direct knowledge other than what I've read here as a non L2 member)- I shouldn't be considered an expert at anything.
Here's the FORs from that document:
http://www.nasa.gov/pdf/160734main_STS-114%20Engine%20Cut-off%20Sensor%20Anomaly%20Technical%20Consultation%20Report.pdfQuote8.0 Findings, Observations and Recommendations
8.1 Findings
F-1. STS-114 troubleshooting did not isolate the root cause of the ECO sensor failures.
F-2. The SSP does not test ECO sensors prior to launch countdown in LH2 at -423 deg F, but tests them in liquid nitrogen (LN2) at -320 deg F.
F-3. Isolation and continuity checks of PSB connectors are not routinely performed.
F-4. PSB transient suppressor electronics are not subjected to routine functional tests.
F-5. An Electromagnetic Interference (EMI) induced failure observed during testing of modified PSB electronics was reproduced at the NASA Shuttle Logistics Depot (NSLD). The failure was traced to circuit differences between a flight unit and the modified PSB under test, which was removed from the Shuttle Avionics Integration Laboratory (SAIL). It was initiated by non-standard sensor test resistance values and has been explained by circuit analysis and testing.
F-6. PSB electronics were subjected to stability testing in the past for several hours with no heat-sinking and no temperature monitoring. While current practice is to test the boxes on a heat-sink, latent effects of the earlier testing are unknown.
F-7. PSB electronics have been the subject of a number of workmanship issues including poor solder joints, lack of solder on some connections, and discrepant flex ribbon assemblies (the flex ribbon assemblies have been especially problematic, according to NSLD personnel). Extent of remaining workmanship problems is unknown.
8.2 Observations
O-1. While the root cause of the STS-114 failure was not identified, two probable causes surfaced including an intermittent open somewhere in the wiring from the PSB to the sensors and return, an intermittent high resistance or open of the sensor itself, or a thermally-induced intermittent failure internal to the PSB. To minimize flight risk, the SSP implemented a set of real-time screens during launch countdown to help detect and isolate a failure. The SSP determined, and the NESC concurred, that launch with a single failed ECO sensor was acceptable providing the problem could be isolated to the one sensor/electronics channel which failed during the first STS-114 launch attempt.
Sensor 2 and 4 PSB inputs were switched to facilitate real-time troubleshooting and problem isolation during launch countdown, but the failure did not recur.
O-2. There are a number of open questions pertaining to the ECO sensor. The sensors are not screened at LH2 temperatures prior to installation as noted in Finding F-1. This may affect their reliability. The design incorporates swaged terminals and stainless steel surface passivation which may also contribute to intermittent cold opens.
O-3. Since multiple power grounds enter the PSBs, they must be shown to be isolated through routine continuity and isolation tests performed at the interface connectors. The chassis continuity connections need to be verified, as do the transient suppressors installed across the sensor lines and 28-volt bus in the power converters.
O-4. Transient suppressors used in this design may be incorrectly sized. Review of the PSB power converter schematic (1500067, revision T) shows the Transient Voltage Suppressors (TVS) to be back-to-back 1N5611s. According to the data sheet for this part, it begins to zener at 43.7 volts and clamps at 63.5 volts at peak surge current. The IC1 and IC2 op-amps are 741s and are directly across the unfiltered 28-volt supply in the drawings reviewed (1500062-005 revision D). According to the data sheet, the LM741 absolute maximum Vcc rating is +22 volts, or 44 volts total. It appears that input voltage surges, particularly strong ones, may be able to overvoltage stress the op-amps if these schematics are correct.
O-5. A problem with loose card guides noted during troubleshooting of the STS-114 ECO problems has not been fully resolved. The guides in at least one PSB were re-glued (S/N 110). However, when the unit was disassembled due to a card guide lug short circuit, the guides were found to have again de-bonded. The guides are necessary to help dissipate a relatively high heat load (over 100 watts of waste heat) and are a potential source of internal contamination when de-bonded due to the potential for liberation of copper beads incorporated in the epoxy adhesive to improve thermal conductivity.
O-6. There are a number of lingering issues with parts used in the PSBs. Government-Industry Data Exchange Program (GIDEP) Alert BZ-A-80-01D covered many Fairchild Semiconductor part numbers and Lot Date Codes (LDC). Parts considered suspect during the STS-114 troubleshooting were researched (LDC 7830 2N2222), but the PSBs contain other transitors including 2N2907 and 2N2219, which may also be covered by the Alert if they were manufactured by Fairchild and are in the LDC range of concern.
O-7. Specific issues with two of the PSB’s tested during the STS-114 troubleshooting have yet to be resolved. The PSB S/N 112 power converter has been noisy. This may be related to power converter instability or limit cycle regulation. Age may affect capacitor properties that can also possibly affect stability. This should be investigated further to ensure high reliability. In addition, the potential for copper path damage between the signal conditioner card circuitry and electrical connector in PSB S/N 111 should be assessed.
O-8. Test method CS116 has replaced CS06 as a method of testing EMI / EMC compliance. Although CS116 is a less-stringent test, unlike CS06 it applies to signal lines as well as power lines. The Point Sensor Box has been designed with bandwidth-limiting networks only on the power leads. Suspicion remains that the signal lines were never designed to be subjected to high input levels and CS116 testing could be over-stressing the circuitry, particularly the transient suppressors intended to provide ESD protection.
8.3 Recommendations
R-1. Conduct a series of cryogenic tests to determine whether reliability issues exist with the ECO sensors. These tests should address issues surrounding the use of staked connections and passivated stainless steel surfaces.
R-2. Modify PSB pre-flight testing and incorporate checkout of transient suppressor electronics and verification of multiple power grounds.
R-3. Verify CS116 EMI / EMC testing does not overstress PSB input circuitry.
R-4. Complete a review of the PSB circuit design and ensure transient suppressor electronics are properly sized and that all parts have been screened for any GIDEP-related issues.
R-5. Ensure all PSBs have been screened for potential workmanship and assembly issues, including loose card guides and damaged flex-ribbon assemblies, and that such problems have been eliminated to the extent possible.
R-6. Determine whether power-on testing of PSB electronics without a heat-sink may have resulted in component overstress and, if so, what components are likely to be affected and how failure of those components may manifest itself in flight.
R-7. Isolate the cause of PSB S/N 112 power converter noise and inspect PSB S/N 111 for copper path damage between the signal conditioner card circuitry and electrical connector. -
#341 by ChrisGebhardt on 09 Dec, 2007 22:06
-
A nice view....
-
#342 by shuttlefan on 09 Dec, 2007 22:54
-
Quote
psloss - 9/12/2007 6:25 AM
I've been following the program for20 years and I AM patient. Just was curious about that option. Thanks.Quoteshuttlefan - 9/12/2007 7:23 AM
Even if they tried filling again, it would be for troubleshooting.
Earlier it was mentioned that they may drain back down to 5% and try filling again. Could this possibly be done and still be successful today in time for a possible launch?
I know there's some anxiousness out there for a launch, but if you follow the shuttle program, you will learn to be patient.
-
#343 by Chris Bergin on 09 Dec, 2007 22:57
-
Watch the webcams. They are due to RSS mate very soon.
-
#344 by kimmern123 on 09 Dec, 2007 23:11
-
When is the crew scheduled to head back to Houston?
-
#345 by on 09 Dec, 2007 23:30
-
Quote
kimmern123 - 9/12/2007 6:11 PM When is the crew scheduled to head back to Houston?
The astronauts will depart for NASA's Johnson Space Center in Houston at 5:00 p.m. EST (2200 GMT) and resume training for their space station construction flight. from Space.com
-
#346 by C5C6 on 09 Dec, 2007 23:46
-
excuse me, i tried to find the answer but got no success, are we having a rollback to the VAB?
or the ECO issue can be fixed at pad and the next launch window starts on jan 2?
-
#347 by jimvela on 09 Dec, 2007 23:53
-
Quote
C5C6 - 9/12/2007 5:46 PM
excuse me, i tried to find the answer but got no success, are we having a rollback to the VAB? or the ECO issue can be fixed at pad and the next launch window starts on jan 2?
That depends on the outcome of the ongoing testing/troubleshooting, doesn't it?
You can't find a definitive answer because there isn't one yet. -
#348 by apollo13 on 09 Dec, 2007 23:59
-
Found this:
During the post-scrub news conference, Launch Director Doug Lyons said that a rollback to the Vehicle Assembly Building was not a situation managers were considering currently, and explained that the pad offers extensive access to the systems for troubleshooting and investigation. -
#349 by Ford Mustang on 10 Dec, 2007 00:08
-
RSS Rotation to MATE position has been underway for a while...
-
#350 by Ford Mustang on 10 Dec, 2007 00:13
-
-
#351 by Ford Mustang on 10 Dec, 2007 00:19
-
-
#352 by Avron on 10 Dec, 2007 00:26
-
Quote
C5C6 - 9/12/2007 7:46 PM
excuse me, i tried to find the answer but got no success, are we having a rollback to the VAB? or the ECO issue can be fixed at pad and the next launch window starts on jan 2?
lets see if the tests of today, narrow things down.. I hope the tests finally target the problem area, so that it could be understood .. based on that there may be an option to work the loading so as to get 100% repeatable /understandable results, without doing any work internally... who knows...maybe a R&R effort, maybe something that requires redesign..
I would be very mpressed if the NET Jan 2 date can be targeted.. -
#353 by Ronsmytheiii on 10 Dec, 2007 00:30
-
see you next month Atlantis
edit: Make that next year.
-
#354 by Jim on 10 Dec, 2007 00:55
-
Quote
Bruce - 9/12/2007 5:26 PM
1. There's lots of different rocket designs being used. OK, not many that are man-rated, but surely someone else has solved this problem of eco sensors. Is it something that can be bought elsewhere? Presumably the whole system would need to be changed, and that's not going to happen quickly.
2. Maybe there's more to this 2015 extension. Maybe NASA have said there's too much pressure to get ISS complete AND finish by 2010. Maybe someone has finally realised that NASA will look just a bit silly with a 5 year must-rely-on-soyuz plan.
3. Either design to fly safely without eco sensors.
1. not for this application
2. This was a congressman's proposal, not NASA's. There is plenty of time before 2010
3. The ECO system is for safety and can't fly without it -
#355 by Justin Space on 10 Dec, 2007 07:55
-
Quote
Ford Mustang - 9/12/2007 7:08 PM
RSS Rotation to MATE position has been underway for a while...
Nice images Ford, as always. -
#356 by Chris Bergin on 10 Dec, 2007 08:38
-
We'll go to a new thread later today, as part of the troubleshooting effort.
-
#357 by Polecat on 10 Dec, 2007 12:06
-
Quote
Chris Bergin - 8/12/2007 1:21 PM
In the meantime, I've written up some of the MMT presentations on L2:
http://www.nasaspaceflight.com/content/?cid=5304
That is long article!
-
#358 by Chris Bergin on 10 Dec, 2007 12:14
-
Quote
Polecat - 10/12/2007 1:06 PM
QuoteChris Bergin - 8/12/2007 1:21 PM
In the meantime, I've written up some of the MMT presentations on L2:
http://www.nasaspaceflight.com/content/?cid=5304
That is long article!
Not as big as the next, which will easily be the most fascinating I've written (not my writing, the quotes I'll be using). -
#359 by rgerhards on 10 Dec, 2007 14:52
-
I have written an article about the ECO sensors and how they work. I was motivated by a lot of false info (obviously not in this forum) and my own desire to understand. The question now is where have I myself screwed up?
I'd appreciate if those with a little time could have a look and pinpoint me to what is wrong. It is available at:
http://spacelaunch.gerhards.net/2007/12/space-shuttle-eco-sonsors-in-depth-view.html
Many thanks,
Rainer
or the ECO issue can be fixed at pad and the next launch window starts on jan 2?