infrequentflyer789

Interesting theory...

Problem: I would expect spar valve position would be recorded (although I don't know details on 777 FDR or QAR), and that a difference between recorded and commanded position would raise a warning (which would also be recorded). AAIB reported no recorded anomalies.

Counter: I think you've previously theorised that spar valve could be detected / recorded as only open/shut and therefore "partly shut" could go unrecorded ?

Problem: Spar valves were found to be open.

Counter: [Again, I think you've theorised this] could be a transient effect that reset/cleared before impact.

Problem: The issue didn't clear before impact, because we know the engines stayed at below commanded thrust until impact (AAIB).


Thoughts: impact cleared the problem - bounced the valves open again, or the electrical disruption ceased at impact. Hmm.

Or, there is a return flow of fuel around the HP pump, and that feedback makes me wonder if this system could possibly get kicked into an abnormal flow state which then self-sustains. ie. your transient spar valve incident causes reduced flow and cavitation at the HP pumps, affects the return flow, which then continues to cause reduced flow even after the spar valve is open again.

My gut feeling is it can't happen like that, on the other hand, it's a long time since I did any fluid dynamics, so maybe...

infrequentflyer789

But if scavenge was still in progress (and maybe sending a big slug of water through to the wings) then CT wouldn't be reading empty.

I've read that before also, and I understand it as no explicit indication / warning of scavenge system failure, not that you couldn't infer it from CT fuel level readings, and I am pretty sure that all 777s have fuel level sensors in every tank. Happy for any 777 pilot to tell us we're (maybe both) wrong though.

Checkers

I like that line of thinking, but fuel temperatures often stay below freezing until well after landing. It seems like there would be, at most, a "snowstorm" of small ice crystals in the liquid fuel.

Sunfish

We have been around all these theories in the previous 30 something pages. If,for example, there was a "snow storm of ice crystals" then why wasn't there a "fuel filter" EICAS warning (or whatever the correct terminology is)?

Let's wait for the report.

Volume

This is perfectly normal icing of the wing in the area where the cold fuel is stored. Fuel temperature during landing is very often well below 0°C, and most probably was during this flight. You can see this icing even on the upper surface of the wing, if there is still a larger amount of fuel present. I have several such photographs taken from the passenger windows.

This is why I seriously doubt that melting ice was somehow involved in the accident. All water in the tank was still in the solid state during impact, which of course could have been at any size of particle from fine "ice dust" to solid chunks.

Green-dot

Quoting infrequentflyer789:

That is indeed the question, spar valves in transit from open- but not reaching the closed position may perhaps not be recorded. Warning to the crew in case of unscheduled closing of the spar valves (with the fuel switches still in "Run" position) is unlikely. I refer to the AMM and/or WDM for further information as to why this may be the case. Also, if the fuel synoptics page was not displayed (or not observed) at the time, the crew would have no visual indication of actual spar valve position.

Correct, also see my comments below.

- I would not know if the disruption (if present) did or did not clear before impact. If, as i have theorized, spar valve control was disturbed by reverse logic: the valves take 15 seconds to fully close and another 15 seconds to fully open again. Theoretically the valves could have been in transit for over 25 seconds (from open - to not fully closed - and back to open) limiting fuel flow to the engines for this period of time and perhaps causing cavitation to the pumps when the engines demanded more fuel but did not receive it.

- The AAIB Special Bulletin 1/2008 from page 5 onward explains why the spar valves were found open. Aircraft wiring was in pre SB 777-28-0025 configuration. The SB was introduced to ensure the spar valve closes when the fire handle is pulled regardless of sequencing. I do, however, still have a question regarding the AAIB findings and subsequent explanation related to the spar valve control relays found in open position. Were they found in the open position because of incorrect sequencing as reported or perhaps because of possible reverse logic still in effect at the time of- or just before impact? With reverse logic this would imply:
- Fuel switch in "Run": control relay in Cutoff position;
- Fuel switch to "Cutoff": control relay in Run position.

In other words, could selecting fuel switches to "Cutoff" at the time of the fire drill have inadvertently re-opened the spar valves (already closing or closed) due to a reverse logic condition still in effect before the fire switches were activated, thus assuming in this case that the sequence was correctly performed, contrary to the report? In the pre SB configuration the second wire path goes directly to the spar valve actuator and de-energizes the control relay circuit, latching the relays in last selected position if i am correct, i could be wrong. Depending on when exactly the fire switches were activated, the impact and damage to the landing gear may have severed the wiring to the valve actuators, preventing the close signals from reaching the valves?


To my knowledge there is a return flow path of excessive fuel returning from the fuel control unit to the HP pump. As fuel flow reduces due to the suspected flow restriction, return flow to the HP pump reduces as more fuel is demanded by the now "fuel hungry" FCU.

See also my post #590 on page 30 of this thread which describes engine behaviour when fuel flow to the engine is deliberately restricted. The test with a 95 percent closed fuel shutoff valve was done because the aircraft type involved had encountered unscheduled closing of the valve due to reverse logic on several occasions, resulting in incidents and accidents.

Green-dot

airfoilmod

In reviewing this thread I noticed a seemingly pre-cognitive post: #251, Page 13. Read "shortfinals" post #662 on Page 34 and pull his Flightglobal.com's URL to read Captain Carbaugh's comments re: cavitation, Fuel Restriction, Temps and reference to ICE. The Captain's comments post-date my post by a month. I want attribution. Dopey Internet. I guess Boeing is lurking. LOL- Airfoil.

airfoilmod

What the ultimate findings will be, but I think that there is a good chance most of the possibilities have been explored here, without FDR CVR QAR, Pilot testimony, destructive testing, etc. I think the quality and tenor of the posting here has been first-rate, So much knowledge and experience in evidence from all.

Albert Square

I wonder if the fuel factor in this investigation may have triggered the current fuel supply problems at Manchester
http://news.bbc.co.uk/1/hi/england/m...er/7331296.stm

HotDog

Surely, you jest!

Spaz Modic

The ultimate update - it was outa gas!

jackharr

Is the AAIB required by regulations to publish its findings within a certain time scale? Or is it able to procrastinate indefinitely like a certain African leader?

Jack

lomapaseo

The AAIB responds to both the government as well as ICAO.

Send your letter of support directly to them and CC either of the above.

ribt4t

I thought one year was the typical timeframe for accident investigations.

Green-dot

Reading the AAIB Bulletin S/1 2008, I found the spar valve description a little bit puzzling regarding the examination of the spar valve circuitry and which valves are actually addressed. The spar valves installed on the rear spar of the wing or the engine shutoff valves:

On page 5 of the bulletin:

On page 6 of the bulletin:

1. First of all, I assume the AAIB refers to the engine fuel spar valves in the aircraft fuel system, not the engine fuel shutoff valves (SOVs) on the engines? As I read the text, it leaves room for misinterpretation as to which valves they actually refer to. No mention is made of the spar valve control relays, which in case of operating the fire switches are the relays directly isolated, down stream of the run/cut-off relays. The run/cut-off relays referred to by the AAIB are multi-function relays. Depending on which relays are actually addressed, other run/cut-off relays control the engine SOVs and are also wired to the fire switches.

2. Furthermore, has the AAIB positively established (proven by evidence/testimony) that the open spar valves were a result of incorrect sequencing or is that an assumption purely based on the pre SB 777-28-0025 configuration? Or is the AAIB also trying to find other possible causal scenarios that may have lead to a similar effect?

3. As I show in bold text above, and assuming the AAIB refers to the aircraft spar valves in the engine fuel feed system, there is perhaps a contradiction in their findings. On page 5 the AAIB found the run/cut-off relays "were still in the valve OPEN position." However, on page 6, in the case of the right engine, "the valve CLOSE wire from the run/cut-off relay" was still continuous?

I assume the AAIB means the run/cut-off relays (Reset/Fuel Spar L-(R-) Eng) located in the left (right) power management panel. From that relay onward the normal wire paths have yet to pass through the spar valve control relays before reaching the spar valves. If so, how could the right wire path still be continuous when the right run/cut-off relay was found in the OPEN position? Under these circumstances this would imply that the right spar valve control relay should also have been found in the OPEN position, hence no continuity. The bulletin, however, does not mention any examination regarding condition/position of those control relays. My assumption is that they must have electrically closed them in order to test continiuity and subsequently found the right valve wire to be continuous. Or were the control relays (latched) closed without having to move them to the closed position for the continuity test. . . .?

The final report may perhaps answer that question.


Green-dot

Sunfish

Now for a new fact, a rumour and an apology.

BA have NOT made a claim on it's insurance policy for the hull loss yet. The insurers have of course been advised, and the aircraft has been assessed as a hull loss. The reason for not yet submitting the claim is that they still don't know what caused the incident - so the prospect of cross claims between Boeing's, Roll's Royce's or other insured entities is as yet unquantifiable.

The theory on the part of the floor at Lloyd's (as of last week) is that it ain't the fuel that caused it, it's "RF interference". The theory goes that after George Brown was dropped off at Heathrow, his driver, or some other entity either deliberately or accidentally forgot to switch off their "jammer" and that the aforesaid vehicle on it's return journey may have gotten within "range" of the affected B777 and caused the mischief.

According to the Lloyd's rumour, the "jammer" in Brown's convoy apparently does very, very considerably more then just "jam cell phones", doing bad things to guidance and detonation systems over short ranges and I will leave it to your imagination to work out what that means.

IF (and it's a very large "if") this theory has any credibility, then the AAIB and HMG are in a pickle since this stuff is classified, and a suitable cover story will have to be constructed.

So my apology to Green Dot, Lloyd's currently agrees with you, although I'm buggered if I can think how such a signal would get into what is almost a perfect Faraday Cage.

wymonwold

The AAIB report does not state the actual fuel quantity that remained contained within the fuel tanks after the accident. A significant amount of fuel spilled from the aircraft at the crash site but what is a significant amount? If a hundred kgs of fuel spilled, it would still be significant due to the fire risk.

airfoilmod

Surprising no one considered that here. Insurors can have Subrogation clauses, that distribute responsibility and financial payments. I may as well put forward a thought I had as an offshoot of GreenDot's persistence and technical corroboratives. My intuition tells me if EMI, it is most likely Microwave. Though I didn't fly FAC, many friends did in Vietnam, I have heard much of interest about "Interrogating and Painting" while flying fast movers that were unarmed in forward sectors. I think that Fuel is still the immediate cause here, but perhaps not the procuring cause. I can't escape thinking about how ILS information is acquired and processed; this was a low level incident, so Glideslope and R/ALT. are obvious sources that could have been corrupted to the extent that the aircraft may have thought it was above/below glideslope or had actually landed, R/Alt. Or Even Tyre buttons telling lies. Again, thanks to Sunfish and Mr. GreenDot. Hats Off.

Airfoil

PBL

Sunfish tells us of the latest rumors at LLoyds.

First, a word of caution. Insurance people do not necessarily know any more than anyone else does, and they are not necessarily technically trained (indeed, most of them whom I have met aren't technical people at all).

Second, some sort of EMI effect has by no means been ruled out by any info available to us. Neither has the manifestation of a digital logic error of some sort (the EEC has of the order of hundreds of thousands of lines of code (LOC; a kLOC is a thousand LOC). The crude average error rate for safety-critical systems is of the order of 1 error per kLOC. The very very best measured rate of which I know in a largish system with hundreds of thousands of lines of code is 1 error per 25 kLOC. That doesn't mean the errors show up in any safety-critical way, but that cannot be ruled out either).

Concerning EMI with a source outside the aircraft, let's do a sanity check on this.

If somebody on the ground is EM-radiating in such a way as to cause well-shielded systems some 700 ft in the air to go haywire, the question is how that could happen without any other effects being noted. If that signal is broadcast, and it happens to be on a road, then lots of less well-shielded auto electronics are going to go crazy also, let alone stuff in buildings. Which means that if it is somebody travelling down the road, lots of people within 700 ft (oh, let's say 1000 ft) are going to experience problems with their cars and their company electronics. And our intrepid journalists would have had that all over the front pages within a day or two. Didn't happen. So not a broadcast.

Suppose it was directed. Then somebody would have been pointing a strong EM source at or near a landing aircraft for some reason. I think there is no chance at all that anybody working for the PM would have done such a thing.

External source does not cover the EMI hypothesis fully, of course. There is also the possibility of EMI from within AC systems themselves, which is the case which Green-dot has been pursuing, since he knows of instances.

PBL

airfoilmod

From within the A/C itself, unrelated to systems or Gordon Brown. Some extraneous pulse that affected FCU or FADEC, or the Spar Valves logic master. Or cables (conductors) phasing from energy fluctuations induced by exterior (transmitted) energy. Would be so nice for a crumb from the chaps.