GasPath
 

Then I see a problem. See Machaca's provided pictures of the Heat exchanger specifically the dozens of narrow tubes in the large cylindrical case. Any transfer of energy requires time, hence the narrow (large surface area) multiple passages. Adding time to passage transit requires restriction. I submit it is a logical place to consider a constriction. Though there are two systems, I refer to an earlier post of mine showing that when independent systems rely on common Fuel (Though stored separately), the possibility is enhanced for failure when certain parameters are shared, and possess an insurmountable extreme, (Temp. Viscosity, restriction, contamination, ambient energy, etc.). This possibility is well within consideration given the seven second delay.

#1159 Machaca, See also Arcniz#1160 for a good interpretation of my
point about heterogeneity in standard and duplicative systems.

Airfoil

Just read this in an aviation e-mag: wonder if some such thing could have happened in the case under discussion in this forum>>

Well Airfoil.

I think it is quite important to discover whether GA was called. Don't you think to retract the flaps at that moment was 'unusual'? There wasn't an opportunity to discuss the matter with the handling pilot presumably.

Fancy this being overlooked by all you amateur theorists?

That is the first thing I would do if struggling to get enough power to reach the runway is attempt a GA.

PoofBoot
 

I think you're confused; a look at the history of the thread would do you well. The Donks conked @ ~800 no power from then on. No one knows whether F/O Coward was attempting GA or simply trying to throttle up while short. The Flaps issue was aggressively dealt with by mods, who diverted all discussion of Flap reselect to JB. Ordinarily one would want a Positive ROC prior to flap retraction, very little is known except the A/P flew up to I think 17 degree AoA and 108 knots at 200 feet AGL. It's not difficult to look like an "amateur" with very little Data, as you now must know. One can Attempt a Go around with no power; One can attempt a HOVER with no power. (Down is Down).

Airfoil

If flaps were raised to less than landing position before GA was selected with unresponsive engines (and EAI was not ON), the engine EECs would have automatically reverted from approach idle to minimum idle, aggravating the situation in this particular case. However, there is no mention of raising flaps or GA selection in the AAIB reports.


Green-dot

The flaps were reduced from F30 to F25. Approach idle would have been minimum idle speed had the thrust levers been closed, which they weren't anyway.

Even if the flaps had been reduced to less than F25 the idle speed would be irrelevant unless the thrust levers were closed, therefore, I fail to understand how the situation would have been aggravated.

Pitot Heat in the B-2
 

Seekayess, I've never flown a B-2, but chances are that pitot heat is ALWAYS on unless selected otherwise. It may well be that they couldn't even select it off except by pulling a circuit breaker and I can't imagine their doing that.

To M.Mouse:

There is no mention made about selecting F25 in any AAIB reports i have read sofar. If there was an anticipation to GA, flap selection would have been less than F25, at least to F20, taking the flaps out of the landing position which would mean (as explained earlier) engine EECs would be set for minimum idle. If the thrust levers were even momentarily moved to idle, to minimum idle they would have gone.

Same as for the flap scheduling, there is also no mention if the thrust levers had been retarded to idle, even if only for a short period. Thrust levers in idle, same as your claim flaps were selected to F25, are not mentioned in the AAIB reports but it it does not necessarily mean it did not occur.


Green-dot

It seems obvious to me that the flap retraction to 25 was only to clear the fence and maybe make the runway. Nobody in that situation with two sick engines would attempt a go around. No pilot on the face of the earth would have attempted a go around. Why even bring this up? If they had enough power to go around they would have just landed.

Green Dot you are correct here is no mention in the AAIB report of the flaps being moved.

The flaps were retracted from F30 to F25 and the thrust levers were not selected to idle. Your hypothesis is flawed.

To M.Mouse:

Could you kindly provide a source to back up your statement?

I agree, from a procedural point of view, that it would not be logical to retard the thrust levers to idle in this situation but with unresponsive engines and no thrust increase with the thrust levers fire-walled, would it not be a natural reflex to retard the thrust levers to idle in an attempt to try a last ditch effort to spool up the engines manually, an attempt to initiate a thrust increase from idle conditions?

(Somewhat like gunning your car engine when the engine stalls and you try to recover by letting go of the accelerator and pushing it back in, in an attempt to recover from a stalling engine.)


Regards,
Green-dot

No.

If I know something for a fact I post it to avoid pages of tedious and inaccurate speculation developing.

It really makes no difference to me whether you believe what I post or not. The final report will show whether any statement I have made is right or wrong.

Not for anyone who was a jet pilot.

About the only time you'd think of retarding a thrust lever when you're undershooting, in an attempt to 'cure' an engine problem, would be if it was surging badly - and this is not the case with BA38.

When people talk about manual thrust on the 777, they mean without the autothrottle engaged. All the A/T does is try to move the thrust levers to satisfy demands from from the A/T computer; when you take it out it means you have to move them yourself. The thrust levers have sensors on them which provide position data to the EECs; that's it. The engines are FADEC = Full Authority Digital Engine Control: there is NO 'manual' control; the only thing you can influence is the mode they operate in - the RR Trent has a 'hard alternate' mode you can select from the flight deck that references N1 instead of EPR and removes some of the thrust (overboost) protection.

Says it all, really.

I disagree.
Green-dot put it right, just a human natural reflex for a mopped driver as well as a jet pilot.
Surprise effect + no time to spare !
That's exactly what did Asseline in Habsheim when he realized thrust was not coming.

And it helped, did it? :rolleyes:

It's possible to 'rich cut' a carb-fed piston engine by being aggressive on the throttle at a low power setting, on something with an accelerator pump... Maybe people get confused by this and expect jet engines to exhibit the same behaviour.

The only way I know to 'reset' the Trent engine is to momentarily put the fuel control switch into cutoff and hope that it will accelerate up again; even the manual says it could take a loooonnng time. Not a technique I'd apply at 700'. :ooh:

To M.Mouse:

Fair enough, patience is a virtue.


To FullWings:

Hindsight has 20/20 vision.

In the case of BA038, the crew were confronted with non responsive engines. They had no time to accurately analyse their engine problems. According to the AAIB reports there were no alerts or EICAS procedures to assist the crew. Could very well be that they perceived engine surging to be their problem at hand at the time.

I am aware of how the system works, including the RR configuration.

In order to select the EEC in the 'hard alternate' mode the pilot has to retard the thrust levers before manually selecting the guarded EEC switch to the 'hard alternate mode' (manual selection is referred to as the 'hard alternate' mode, automatic selection is referred to as the 'soft alternate' mode).

Retarding the thrust levers before selecting the 'hard alternate' mode is necessary because there is no thrust limit protection in the alternate modes (in case of the BA038 situation the thrust levers would have been fire-walled). If the crew had decided to select the 'hard alternate' mode (regardless of an EEC fault or not) they would have retarded the thrust levers before doing so.

To do this, the action is generally triggered by an EEC fault which initially results in a (automatic) 'soft alternate' mode and an EICAS advisory message/procedure. A 'hard alternate' mode selection is the next step in order to acquire a [boxed] N1 reference indication on the affected engine.

Since the AAIB reports explain there is no evidence of EEC faults, hence no alerts such as EEC ALTN light or EICAS messages, logically there would have been no reason for the crew to act and select the 'hard alternate' mode. The only remaining scenario would have been if the crew decided to select the 'hard alternate' mode at their own discretion in an attempt to "wake up" the engines as a last resort. In this scenario they could have retarded the thrust levers to idle although it would not have been necessary to retard the thrust levers that far back.

I stress this probably did not occur or the AAIB would have mentioned it in one of their released bulletins but it could have been a scenario as a last resort attempt to avert premature contact with terra firma in that final minute. If it did occur, it could have resulted in engines reverting to minimum idle if flaps were retracted to less than landing position and EAI not ON. But based on the information that M. Mouse provided (he must have first hand information to back his claim), minimum idle can apparently be illiminated as a factor.

It would have been risky to cycle the spar valves (by selecting fuel control switches from run-to cutoff-to run) which could result in jammed spar valve actuators which would certainly have resulted in fuel flow restrictions.

No but in the BA038 scenario, if this action took place, it didn't either.



Regards,
Green-dot

If you're out of thrust, retarding is not necessary.
What if they had firewalled with such a switching? - It would have saved their day, wouldn't it?
This is some silly behind the desk comment. If you are in a emergency situation without thrust, you really don't care about technicalities, formalities, bulletins, sops or similar: Anything that gives you a notch more is very welcome....

I believe Green-Dot
 

is saying closed throttles are necessary because Alt mode is locked out at advanced levers, to prevent over speed, No? Once throttles are retarded and Alt mode select is done, then truly manual Fuel control is obtained.(FAMEC?)

Airfoil

Mariner9. Thanks for your detailed response.
It is obvious that you know a lot about Jet fuel compliance.

I glean from your response that fuel compliance (or at least production) is as I suspected somewhat of a "Black Art".
Your statement re viscosity measurement only at -20C is interesting.
I still suspect the fuel.

However, I never INTENDED to suggest that the fuel wouldn't burn (although others have suggested this).
I INTENDED to suggest that some property of the fuel caused it to exhibit strange TRANSIENT viscosity characteristics under SUCTION, as in fact you have stated above.
Maybe I didn't express myself clearly enough.

I have NO experience in aviation but as a professional engineer I find it difficult to come to any other conclusion (ON THE INFORMATION AVAILABLE TO US) other than some aspect of the FUEL, conditioned by ENVIRONMENT, and FUEL SYSTEM DESIGN caused a condition that has not been previously considered.

Airfoilmod,
There are no 'lockouts' that I'm aware of. Selecting alternate mode makes N1 the controlling parameter and this is referenced to the thrust lever position. This means it is possible to exceed the rated thrust of the engine in certain conditions but not overspeed anything; also, the system is designed so that when you switch, you get at least the thrust you had before, possibly more. If you had firewalled before selecting alt. mode, it's very likely one of the the RPM limits would come into play; at much less than TO/GA to begin with you're unlikely to bust anything.

The manual also says: "If the EECs are in alternate mode, advancing the thrust levers full forward provides some overboost and should be considered only during emergency situations when all other available actions have been taken and terrain contact is imminent." Good one to have up your sleeve if you ever need it...


Green-dot,
I really can't see that: all engine parameters were stable and within limits, just not where they wanted them! An engine surge(ing) is difficult to overlook, even from a mile away... Two very different scenarios, IMHO.


snanceki,
I think your comment is pretty much the 'state of the art' at the moment. There are so many possibilities but, like you, I think the probabilities lie here...

Loss of fuel system suction feed
 

From:
Federal Register / Vol. 73, No. 110 / Friday, June 6, 2008 / Proposed Rules.

DEPARTMENT OF TRANSPORTATION
Federal Aviation Administration
14 CFR Part 39
[Docket No. FAA–2008–0618; Directorate Identifier 2007–NM–355–AD] RIN 2120–AA64
Airworthiness Directives; Boeing Model 777 Airplanes
AGENCY: Federal Aviation Administration (FAA), DOT.
ACTION: Notice of proposed rulemaking (NPRM).
SUMMARY: We propose to adopt a new airworthiness directive (AD) for all Boeing Model 777 airplanes. This proposed AD would require performing repetitive operational tests of the engine fuel suction feed of the fuel system, and other related testing if necessary. This proposed AD results from a report of inservice occurrences of loss of fuel system suction feed capability, followed by total loss of pressure of the fuel feed system. We are proposing this AD to detect and correct failure of the engine fuel suction feed of the fuel system, which could result in dual engine flameout, inability to restart the engines, and consequent forced landing of the airplane.
DATES: We must receive comments on this proposed AD by July 21, 2008.
---- END QUOTE -----
No mention of using any specific grade or type of fuel, or of testing at low temperatures, just testing.

P.S. Many thanks to precept for the link to the US Government AD Notice Search Location in his/her Post #1135 on 13th May, 2008. http://pickyperkins.home.infionline.net/pi.gif

Thanks!

The docket contains more detailed info, such as:
http://edocket.access.gpo.gov/2008/E8-12691.htm

There you go
 

If True, and the cause of BAE038 mishap, Renton/Rolls have some 'splainin to do. IMHO

Thanks for the pointer to the docket, it's an interesting read.

However, I don't see its relevance to BA 038.

According to the AAIB's reports ...

So, no suction feed.

If some sort of condition was present on flight BA038 that restricted fuel flow even with the boost pumps operating, why would the FAA propose rulemaking specifically for the case of suction feed?


Bernd

Similar NPRM issued for 747-400 too.

You might like to refer to my posts, some pages ago, concerning the possibility of trapped air in the fuel suction feed disrupting the fuel supply when released.

Page 28 post 555 para 4. If I knew how to cut and paste, it would be here:

Bernd
 

Just because Boost pumps are operating, does this mean suction feed was not compromised in other ways? The 777 was designed to operate sucking Fuel only, w/o pumps. The cavitation on the High side of the HP engine pumps points strongly to a lack of available Fuel, which means that demand was overpowering supply. To me, this means that the pumps/filters/flow somehow caused the problem, in other words, if the aircraft can run just fine (By Test) at all settings on suction alone, what in addition to suction only would cause the FAA to distribute a Rule that referenced "Dual Flameouts"? AoA? Fuel Viscosity? Fuel Temperature? Sensing Failures? #1 I feel this incident is not complicated, #2 AAIB/FAA have had the solution for some time, #3 Proper time is passing for reasons other than the complexity of the challenge.

Airfoil

Any detail on that test ... ?

There currently is a pressure test carried out at I believe 'A' check level.
So the penny has finally dropped to include the 744.... I wonder why that is;);):D
The 744 AMM will require a rewrite to include a vacuum check.:hmm:

To Pool:

Although i agree with you this was an emergency situation and lacked thrust, i was only explaining how the system functions, how the 'hard alternate' mode is designed and operated. Obviously it was not designed with the BA038 incident in mind. Fact remains that there is no thrust limit protection and that is why thrust levers are (partially) retarded if fire-walled and that is how operation in this mode is trained. N1, N2, and N3 overspeed protection remain available. Again, with hindsight, in the case of BA038 none of these limits would probably have been reached. And again with the crew not exactly knowing what was affecting the engines might or might not have followed trained alternate mode procedures, all hypothetical anyway because they had no indication or alert to select the alternate mode and most likely never did.

airfoilmod:

Alternate mode selection does not lock out the thrust levers which, in this case, do not have to be fully retarded to idle and as FullWings explains, in this mode N1 becomes the controlling parameter.

Green-dot

Some random thoughts on this AD

It seems to be a data gathering and not a conclusion.

7500 hours does not relate to a confirmed safety of flight situation.

But it is movement in a direction to at least gather pro or con information about a possible problem in a timely manner.

It reminds me of the initial response to the fuse pin issue with the B747 between CI and El Al accidents.

So I will await the inspection results (data vs postulations)

FAA Document 2008-0618 URL sources
 

A direct link to the FAA document regarding 777 suction feed follows. Interested parties may want to click on the imbedded links to both the Docket and to the Document. Comments by the Airlines and interested parties are posted on the Docket response. As can be seen Air New Zealand has already posted comments.

http://www.regulations.gov/fdmspubli...=FAA-2008-0618

My knowledge of the 777 fuel system comes from this thread, however, a suction test indicates that the engines are run and the boost pumps switched off, if the engines run down, it would appear that air is entering the system. Air could enter the system from, for instance, a NRV not seating properly. That both engines on BA038 ran down at the same time is still peculiar. This AD addresses the system not the fuel.

It still doesn't make sense, though. With the tank pumps running and the manifold(s) pressurised, fuel is going to leak out rather than air leak in... Unless there was a massive failure of NRVs leading to a kind of short circuit in the fuel supply... But even then, the fuel in the pipes is still at a higher pressure (atmospheric plus a bit of gravity feed) than that at the entry to the engine pumps. :confused: Still pointing towards some sort of restriction, I feel.

If a NRV or PRV is in the system to bleed off excess pressure when the boost pumps are operating, it would not be an external leak when it is working correctly as it is doing the job it is designed for, depending on the pressure it is set at. But if the valve is not seating as it should air can be drawn in from the return line.

Wing boost pumps ON. Start engines and run at idle-
Turn off APU (to disable APU boost pump)
Switch off all 4 boost pumps.
Engine shall run at idle for 5 minutes.
If not look for leaks in suction feed system.

When I used to do this on the L1011, we did it at take off power. Turned off all the boost pumps at take off and watched that nothing else happened. The engine always continued to run as normal.

The 7500 hours from publication allowance in the testing requirement suggests a complete lack of any sense of urgency in this inquiry. Cold prospect as far as they're concerned.

The NPRM's have been issued to all Boeings apart from the 717 as far as I can see

Suction Test
 

A similar test was called up on the 737-200, much as described by Swedish Steve, but the fuel level had to be quite low, actual figure escapes me, so as to uncover the most of the in tank fuel tubing. The method will work on any aircraft and has been used with success to find holes & bad joints on the in tank fuel tubes.