So much of what we see from the authority seems to depend on blind faith. A "tiny sample" can NOT be relied upon to exclude a growing and more pervasive problem. It is shortsighted to assume it is only Trent, based on information in the public domain. It is as you say, sooty.

Without a complete audit of every possibility, this AD rings hollow. A "Patch". An expensive patch, but a patch nonetheless.

Will

Agree:ok:

reminds me of the uncommanded thrust reverser problem. A Boeing specified, requirement but at first unique to only some PW models. An AD appropriately addressed the known. Then more data arrived (Airbus, Douglas, RR, GE) and additional ADs addressing the knowns in a specific fashion.

So I'll watch this space :) for additional knowns that can be addressed in a specific fashion.

Some may feel that this is guessing at the location of the bullet in the chamber, but then again throwing a design change at something that currently works just adds more bullets.

The FOHE (Trent) was not specifically designed as a fuel heater/ice melter.

I can't stress this enough. Modifying an existing component to mitigate a problem its overall design was never meant to address is unusual, to say the least. In the absence of further work on fuel/fueling issues, (is there any?), the actual cause of the icing induced rollbacks is being neglected?

The ICE is not the cause, it is what causes the ICE.

will & sooty
 

100+ on that...

wonder, whether that situation IS in fact causing some nailbiting, especially in regards to the ultra ETOPS qualification of the B777..??

not that a common denominator like water / ice contaminated fuel ( for whatever reasons, i.e bad supplier quality and/or bad draining routines, sometimes caused by that fact that during the short downtimes of long range flights you cannot possibly drain all condensation out etc etc) could not down a quad either, BUT the statistical probabilities for a twin to be affected on both motors would be considerably more severe..

so, anybody know?? were there any discussions to at least consider temp withdrawal of ETOPS qualifications????

dont get me wrong, still very impressed with the 777 in general, but that accident shook me up a bit...

and it may point to a far wider reaching problem, and that is fuel quality in general..and maybe insufficient means to get the water out of it..

and that again may not have much to do with the type of aircraft..

Will, As a fuel guy, this is the way I see it:
They cant use the fuel recovered from the aircraft (would be gone very quickly) but they have tried VERY HARD to bung as much water as possible into the test fuel and STILL they have not been able to replicate the fault although they have seen icing. So while I still believe it was fuel related I havent seen anything yet that leads me to believe it was simply water in fuel. I want to know why the fuel FP was elevated over typical RP-3 even though it met the spec for Jet A-1. Do I believe the Trent before the mod was more susceptible? Yes. Do I think that other powerplants would suffer the same problems as the environmental conditions became more severe? Of course.

Regarding your question on handling, it doesnt really matter - if the fuel was on spec and was badly handled and became unfit for purpose despite being on-spec it is no different to it being unfit for purpose and well handled.

Cast your mind back to the Australian avgas incident which caused a plane to fall out of the sky and grounded the Victoria GA fleet. Trace amounts (parts per billion) of Di Ethyl Amine - a chemical that is used widely in refining to remove Sulfur, and something you wouldnt expect to be an issue - caused a wierd reaction with the metal piping / braze in some aircraft fuel lines. The fuel met spec. In this sort of incident it is simply not enough to say "the fuel met Defstan 91-91" or whatever. You have to take it apart molecule by molecule and look for things that you "dont know that you don't know".

The fix is not melting the ice as much as it is mitigating ice on the move, en masse. According to what I've heard, the FOHE design has been modified for reasons of flow, not heat transfer.

Reliable fuel delivery is something we should have mastered by now. Perhaps after the introduction of new fuel formulations, certification was not thorough enough on this particular aircraft and engine combination.

vapilot2004

To now, the fix is an operational workaround, a cycling of thrust to heat the engine, melting the Ice.

Until AD compliance, that is the 'fix'.

After fleet wide compliance, a combination of operational and mechanical defense against Ice? There is an assumption here that Trent/777 is an isolated example of inability to resist ICE in FUEL. Absent a broader approach, it has all the appearance of a 'Patch'. The passage of time dulls some of the attention paid to this problem.

Given the nature of design, the workaround will have what effective measure of performance? Unknown, is my guess. Until the Fuel is understood better.

Will

Will,

I think the interim recommendation addressed two areas of concern. The power increases not only add heat to the FOHE but also cause the fuel flow to increase dramatically above idle rates, forcing accretions to break up and move on down the pipe.

My comment earlier was referencing the Trent's new FOHEs, so I may have mis-quoted your meaning. Apologies. :8

I agree that there is something odd about what has happened here and once again am left wondering out loud how this could have been missed by Boeing, Rolls and the regulatory bodies during certification. I think something changed and it smells like kerosene to me.

I haven't smelled Chinese Fuel from China. What drops an FP 20 degrees below 'spec'. If its expensive, why bother ? If it's cheaper, why ?

Fpts and intermolecular bonding
 

Will,

You can lower the FPt (i) by branching the hydrocarbon chain (use more of a catalytically cracked high Mr feed stock) or (ii) by shortening the chain length (use more of a more petrol/naptha like cut). Its a Van der Waals / London dispersion forces argument.

I would guess that either of these routes would push up costs but the fuel guys will answer that better than I can, I'm just a chemist....and yes yes I appreciate the possible use of (non hydrocarbon) additives.

CW

Would be interesting to take a sample of RP 3 / JF #3 NOW and compare with the recovered sample via GC MS as well as distillation range... and scouting for all those ex-PEK retention samples that will be hanging around from the last 5 years. There are only a few general possibilities for lowering the FP and they include, firstly, changes in the relative proportion of hydrocarbon components that SHOULD be there (eg wider than normal wide cut fuel) and secondly, the addition, either deliberately or accidentally, of hydrocarbon components that SHOULDNT be there (e.g. FAME). Then of course, there's the non - hydrocarbon components.

It would help if we knew which tests were done. If the material was on spec, what is the harm in releasing the data?

FAME can't improve FP in Jet, and is of course tested for nowadays in the standard spec.

Pinkman, I know you keep harping on about GCMS but do you really think it wasn't done? I agree however that there would have been no harm in releasing the data or even add a simple line to the report that nothing unusual was noted on the GCMS profile.

Personally, I would be interested in a comparison of profiles between the fuel samples from the aircraft and the retention samples. A "depressed" FP (if it was in fact depressed) for the aircraft fuel could have arisen from a reduction in the heavy HC components present (wax dropout? - though no wax deposition was apparently found doesn't mean it didn't happen)

M9 (who investigates Jet A-1 contaminations for a living)

Are we all not in danger of "examining the molecules" while forgetting the big picture.

What needs to be done to avoid the engine stopping again? The FOHE is not a fuel heater but an oil cooler and is designed as such. Presumably the engine is capable of taking a reasonable amount of water or ice in the fuel so does the revised FOHE allow slush ice to go through if not should there be a by-pass?

In common with many others, I've read every post made on this venerable and informative thread and it’s good to see their quality and thought-provoking nature being maintained. (Until this one…….)

The issue of bypassing the FOHE has been covered in detail already. If memory serves, one contributor, presumably with tongue firmly in cheek, even claimed a pat pending on his particular solution.

But I'm with you sky9, how can we actually deal with the rollback problem is the key. As we don't know what the problem actually is, that makes things really tough.

Presumably the question is how much ice in the injectors can be handled before blockage and presumably flameout never mind rollback, occurs?

I have the greatest of respect for the quality of the engineers working for RR and I'm quite sure that this will already have been looked at exhaustively by RR. I'm also naïve enough to take on trust that if it was just a matter of putting in a simple bypass system, RR would've done it long ago.

CW

chris

I have no naive trust, at least not relative to that which I had years ago.
I think a 'simple bypass system', may have been looked at prior to original design vis a vis Trent8, but frankly, I think it is being actively avoided at this point. Retro isn't simple, it isn't cheap, and it isn't logistically benign to any of the principals.

As read, the AD focuses solely on the FOHE. airfoil was the first to point out its design purpose did not include Fuel heating. If it did, Oil would not bypass its chamber as a designed mechanism (oil bypasses now only when it itself does not need cooling). This also raises questions about your faith in an a priori engineered look-see at the potential for Fuel Ice.

Since there is no provision for Fuel supply to the powerplants if for any reason the FOHE is clogged, one doubts the thoroughness of the investigation for potential Fuel starvation due to such a problem.

One could make the compelling claim that in the absence of an alternate path for fuel bypassing the FOHE, the sacrifice of FLOW (fuel) was not deemed to present a critical path, ipso facto; there was a determination made to exclude such a safety implementation. To think that it was excluded by accident, is beyond even my cynical approach and suggests our shared faith in the engineering, though from different perspectives.

Will

We Do Know
 

Chris:
I'm not being facetious here, but we really do know what the problem is/was. In it's simplest form, some of the "stuff" in the fuel tank froze and blocked path of the fuel to the engines. There may well be a number of scenarios in which this could happen so what we have to do is to prevent freezing during all of them. That means heating the fuel near the in-tank pumps. Is there a source of heat, now wasted, that could be put to good use?

Will, I share your concern about the cost of retro-fitting a change like I propose above, but how does that compare to the cost of losing a plane, not to mention crew and passengers? Fortunately, this time it was only the plane.

Smilin'Ed

You may have missed the irony in my post, I suggested that a retrofit of a full on bypass for fuel was being actively avoided.

If you think something more than 'converting' a cooler to a heater is necessary here, we are in complete agreement. I also agree with the conclusion that we know precisely what happened to 038 and Delta. The 'mystery' of 'unknown' characteristics of Fuel is a dodge, simply put.

Will

Will, Ed, thank you both.

I very much take the point that we know that it was solid phase water that triggered the roll back, my point was a little wider and not well made.

Without going over old ground too much and with due caveats ........

What we don't know is why, if the fuel was shall we say "well within spec" (neutral language) and the flight path and temperature profile not that statistically unusual, the problem has not occurred in something close to this form in many other flights too.

After all I've read, I'm still mostly in the "it's the fuel" camp. Was it made right, was it handled right, was the water drainage right etc ad nauseum.

CW

Question - has BA or Boeing changed anything on the B777 to avoid this happening again? I will be honest, I have not read through the 125+ pages to see if this has been answered so apologies, but hopefully someone will either be able to say yes or no. :ok:

Cloud1

The answer is yes the details are all in the thread including the past week.

Fuel Heater Needed
 

Will:
I believe we are in agreement and I think any heating needs to be upstream in the wing to help prevent icing in the pipes as was experienced in testing.

Chris, yes it is the fuel but I think it more practical (cheaper) to modify the planes so that they can better tolerate water than to change the specs. Trying to keep water, and other undesirable materials, out of the fuel would be much more expensive and more frustrating than a strategically located heater in the airplane.

A random thought
 

Liquid-fuel rocket engines use regenerative cooling, wherein the cryogenic fuel and/or oxidizer is used to cool the case and nozzle of the engine, and in so doing vaporize the liquid.
Kill two birds with one stone.

In a gas turbine engine, there is a strong thermodynamic case to be made for mid-stage cooling of the compressor airflow. It would not be a pretty picture, with additional air ducts or fuel lines running hither and thither, but it would be a useful way to heat fuel in the tank and simultaneously optimize the engine cycle.

I'm sure it's been considered and rejected because of cost and complexity, but who knows? If fuel approaches $10/gallon, should it be reconsidered?

I guess I missed something
 

Have read through the thread but can't quite see if there is positive evidence that the blockage/impingement is at the FOHE or if it's by logical analysis/plausability argument only. Could some patient soul enlighten me ? Thanks.

Don't know about the thread (I've only scanned it) but the last version of the report I read said that yes, ice* blockage in the heat exchanger can cause the symptoms seen. However, it went on to say that the timing of things observed at the engine would tend to put the blockage some 20 feet farther up the fuel line, and they are at a complete loss currently as to how that can happen - they can't find a point 20 feet back that can feasibly be blocked by ice.

I think the summary (of the report, not necessarily this thread) is: did ice* blockage cause the problem? Yes. Was it in the heat exchanger? Don't know. Was it farther back? Maybe, but don't see how that could happen. More research is required, and more research is happening.

* The report goes to some lengths to note that the "ice" is NOT 100% water. It is a frozen slush of water and fuel, and the water makes up a fairly small percentage of the slush.

Ice blockage of the FOHE inlet has been demonstrated in lab testing. It was not expected and operational procedures have been introduced to alleviate it. RR has designed a new FOHE which is now subject of an AD.
There is a picture in this thread, and also in Flight Global.

No Bleed Air, PLEASE
 

barit1:
Please don't even think about ducting bleed air out of the engine. I had to investigate an accident where three of my colleagues died because of a bleed air leak. Yes, it was decades ago, and yes materials have improved, but bleed air is highly dangerous stuff. Introducing bleed air into a fuel tank frightens me and I'm fearless.

barit1

In theory, if what you mean is transferring some of the heat of compression to the Fuel, it makes sense. Cooler air can be compressed further, and adds efficiency. The Thermodynamics are beyond me, I'm more a mechanical sort, and tend to agree that running Fuel through or even around the second hottest part of the engine, seems problematic.
Intercooler?

Smilin Ed

727, LAX ??

Douglas rules the waves. Boeing, not so much

No. Navy A-3 in about 1972.

Smaller is better
 

Back in post #2473, Pinkman says:

Using actual engines and fuel systems is probably neither the least expensive or most efficient way to go about investigating fuel system icing. Small-scale lab bench and workshop rigs, with a bit of mathematical modelling, will be a much more effective way of identifying the relevant parameters and their interaction.

On a lab scale, it should be relatively easy to identify the dependency of ice accreation rates on pipe external temperature, pipe size, internal surface characteristics, fuel temperature, fuel flow, fuel composition, including water content, effect of flow rate on the stability of the ice, and so on. Only when these elements are reasonably well understood will it be possible to predict what will happen in a full-size system, and reasonable to verify those predictions.

Perhaps such investigations are already under way in the industry. If not, it seems likely that they are a potential source of PhD theses in accademia.

On a slightly different tack, the AAIB report suggests that it was difficult or impossible in many cases to observe ice build-up inside their test rigs. I should have thought that techniques such as ultrasound would have been effective, and that it would not have been too difficult to send a miniature camera through the system.

how much evidence can you need?
 

Isn't it unsettling that even with an intact aircraft, all the systems ready to be examined in any detail you like, all the witnesses standing ready, fuel still available in the tank, all left in a convenient location and with time to think, there should still be any uncertainty ?

Mr Optimistic, Hi.
I am not sure what uncertainty you are referring to and why that would be unsettling, particularly after Swedish Steve's post just above. Can you say more?
rgds, pf

Pleasure
 

Thanks for the reply. Have read through thread and seen discussions on the meaningfulness of specifications, anxiety about FAME (and about the possibility that there is some unknown factor which isn't even tested for), questions over twin engine ops for the airframe/engine combo, but there seems to be no clear smoking gun.

This gave me the impression that the ('interim' ?) findings are based on plausibility: cold long flight, fuel starvation, there is water in fuel - so ice is a main suspect, where might it most likely accumulate ? Yes we can mimic that failure mode in the lab, so lets go with that one then as NFF will have implications.

If it's so plausible now mother nature has pointed it out to us why didn't the original design anticipate it and why is it so rare an occurrence ?

All very well to cover office walls with fault trees, cut-sets and 10 to the power of minus 9, but this surely should be cut and dried forensic engineering.

And I haven't got that impression from reading this thread (no headline claiming 'CRACKED IT').

Doesn't this one need putting to bed more firmly ? (..the issue, not me !).

As an after thought, has the fuel system on the 777's with alternative engine supplier demonstrated differences in design to explain why it is only RR ?

fuel pipe heating
 

In oil refineries it is common practice to pump large amounts of heavy lube and fuel oil for distances of up to a mile.In order to keep the oil from cooling and becoming too viscous to pump "trace heating" is used. This consists of a small diameter pipe wrapped round the pipeline which is then heavily insulated. Low pressure steam is passed through the tracepipe.By substituting steam for electrical resistive heating tape and insulating the fuel pipe is it not possible to warm the fuel before it gets near to the engine? This would be a more elegant solution than piping bleed air around the wing. Just a thought.

This is exactly what is used in the cargo bays to stop water from freezing inside the service pipes (resistive heater tapes that is). I'd imagine it'd take a load of research and money thrown at it before they'd ever certify electrical heater tapes on a fuel system pipe.

Already well-proven in industrial apps. However, the current needed to pump that much heat into such an enormous flow-volume would require some seriously big alternators and cabling. HUGE power-loss!

mid-stage heat-exchange intercooling would appear to be the simplest solution,a portion of the surplus,returned, warmed fuel could recirculate through the wing/pylon-piping, whilst the rest returned to the tank.

although power-losses are involved, the thermodynamic gain would offset some or all of this, thus making a virtue from necessity.

Good Idea Steve
 

Steve, your analysis of electrical heating problems is right on. Not very practical. Also, I don't have much problem using air bled off the engine as long as it stays in the nacelle and can't leak out to burn things like wings. I'm really paranoid about that. :ok:

I still don't believe an ETOPS airliner can have BOTH engines display the same behaviour at the SAME time due to fuel freezing or boost pump contamination.

It can't happen. Sure, it can happen in one tank and after a while maybe even in the other but AT THE SAME TIME? Forget it.

The chances of fuel freezing in both tanks at the same time is mathematically irrelevant.

Just my .02.

Not knowing the details of the Trent/777 installation -

The HP fuel pump is typically sized to pump fuel for the SL takeoff condition, which means it is oversized in cruise, and VERY oversized in descent. It being a positive-displacement pump, the excess fuel must be routed somewhere, and depending on where the excess is dumped, it can either create an unwanted thermal runaway, or else solve an overcooled condition.

Or did I read that the RR HP pump is a variable-displacement type, in which case the above is null and void? :confused:

No. the HP fuel pump is a gear type positive displacement pump. Remember it is turned by the gearbox, so the speed is proportional to engine speed, and so therefore is the output.
Inside the FMU is a Pressure drop and spill valve which controls the downstream pressure by spilling fuel back to the HP pump intake.

That's what I thought.

My point is that pumping fuel up to burner pressure takes a lot of energy, and when there's an excess of high pressure fuel available, that's heat that can be put to good use.

Or not. :uhoh: