phil gollin

I would again point out that the ice accretion photo, tests, etc.... have all come from UNREPRESENTATIVE tests using unrepresentative fuel.

Before trying to redesign everything, a proper understanding of what actually happened and the "new" fuel/ice behaviour is needed.


(I will now go back to sleep until the "accumulation" of posts ignring the facts starts to build-up to a critical level again.)

.

bsieker

No, the rate of heat exchange has been designed the way it is. Having more heat exchanged may make the oil too cold. The other name for the FOHE is "Fuel Cooled Oil Cooler", which clearly states is primary function: cooling the oil. Heating the fuel is only a secondary effect.



Countercurrent flow heat exchangers are much more efficient than parallel flow desings. The design goal of the FOHE was not to melt ice on its fuel inlet face, but to cool the oil, and weight being a critical parameter in aviation, this was obviously the right engineering decision. I'd be surprised to see a lot of parallel flow HEs anywhere.


Bernd

Dak Man

This is a normal arrangement, LP Pump - Oil Cooler - LP filter (with bypass) - HP Pump. Nothing unusual in that system and it's comon to all 3 big manufacturers.

Again, where T800 is different is the return circuit of spill flow (that flow delivered to the FMU in excess of what the burnt flow demand is) which is not an inconsiderable amount of fuel. On T800 it spills around the HP system in most other applications it spill around the LP system therefore introducing a large heat source into the FCOC.

Also, it is normal practice to arrange the flow path so that the hottest oil is cooled by the coldest fuel.

However, the effective "fix" is, in my opinion, treating the symptom not the cause, which is essentially fuel with high entrained water content.

Controlled testing of the engine fuel system whereby a "slug" of ice or "snowball" is introduced into the LP inlet is extremely difficult to control and replicate that which develops in the aircraft system.

As has been previously stated the engine fuel system has been optimised within the aircraft flight envelope, I do not know the intimate detail but it appears that there may have been some envelope excursion with BA0038 which the engine fuel system was not designed to accommodate, combined with the upload of fuel that may have had a high water content and the result could be considered as academic.

For information engine fuel system ice testing is carried out at a concentration level of ~260ppm.

Pinkman

That is why I continually ask what the results were from analysis of the 'retention' samples from the airport tankage. Practices vary; in the old days deliveries often had a representative sample from the bowser taken and it was sealed in something resembling a kilner jar with wire seals and a lead crimp.

Nowadays with hot hydrant systems the procedures aren't much changed: Ideally (i.e. CAP 748) there should still be a sample taken:


g) immediately prior to fuelling an aircraft.

Soooo.. again, I ask, where are the samples, were they subject to a proper Chain of Custody arrangement, and what are the analysis results?
It would be nice to see some records of the last delivery into tankage and the total rainfall increments between the delivery and the fuelling.

Pinkman

Jumbo Driver

Why is this?

Is this not the opposite of bsieker's assertion ?


JD

bsieker

Basic thermodynamics, really.

Ideally, a countercurrent flow HE can make the "hot" fluid (oil) at its exit almost as cold as the "cold" fluid (fuel) at its inlet (when it is coldest).

Ideally, a parallel flow HE can make the "hot" fluid (oil) at its outlet only almost as cold as the "cold" fluid (fuel) at its outlet (when it is warmest).


It is. And I think Dak Man might have been momentarily confused when he wrote it.

Because what he said also means that the hottest fuel would be used to cool the coldest oil, which would be inefficient.


Bernd

Oilandgasman

Countercurrent exchange - Wikipedia, the free encyclopedia
Here is a very basic diagram on the difference between the two. Showing that counter current exchange is the more efficient. This of course will only apply if the lube oil bypass on the FOHE is closed. If it is open then little heat is transferred to the fuel. As the lube oil flow through the shell side of the FOHE is a torturous path across the many tubes and around the baffle plates, it would be interesting to know how far the bypass needs to be open before there is no flow on the shell side of the exchanger and therefore no heat input to the fuel. Are we sure that this exchanger arrangement is counter current flow? If it is concurrent then less heat transferred to the fuel even when it is fully in service.

airfoilmod

The Fuel flow is 6,000 pph at Idle, and the Fuel is very cold.

The Oil Flow is what? Far far less?

The Oil side of the HE is exposed to almost static low temps, it affects the Fuel very little.

At the critical time being discussed, (FOHE plug, Face) the Hottest Oil would be most effective at melting the Ice, per AD. But remember, as I pointed out a while back, the FOHE is not intended to be an Ice melter, it is an Oil cooler. So the design of concurrent or counter is not relevant to the discussion; except insofar as an installation is now intended to accomplish that which it was not designed to do. Solving the Plug problem doesn't solve the Ice problem. Until Pipe Ice is eliminated, someone's tilting at windmills. The Trent may be taking the fall for a problem no engine was designed to function with, but happened to be more vulnerable to, (the Ice), by chance, than the other engines.

AF

Dak Man

Perhaps my post was slightly misleading, there is no hottest or coldest fuel "end", the fuel temp is what it is and is largely uniform as it enters the FCOC, again remember that it's an oil cooler - not a fuel heater but you can take "benefit" from the hottest oil in raising fuel temps at what is evidently a critical point in the system, i.e the heater matrix end plate.

The fix that I mentioned is, in my opinion, little more than a sticking plaster, if the focus is to be an engine fuel system orientated fix then it should be focussed on getting a bit more heat to fuel in the FCOC loop, and as I have said previously this is normally achieved via the FMU spill return to LP rather than HP as in the T800.

However, in saying that, does anyone know what Fuel Return to Tank (FRTT) philosophy is for B777 / T800, this may well be an insufficient Flow Number (FN) for these extremely long sector cruises and not prevent the tank fuel volumes dropping below say, -8 Deg C.

I suspect that if system testing is done with fuel in the sticky ice temperature window then lots of engine fuel systems will, shall we say, "fall down" which is a huge can of worms. It's perhaps somewhat fortunate that it was a T800 where a fix is easily "sold". If it had been a GE90 where I believe the FCOC loop is in the HP fuel circuit, therefore intrinsically hotter, then the proverbial would be hitting the fan. As an aside the problem with HP side FCOC loops is matrix tube failure, as significant oil dilution would occur leading to lubrication / bearing "issues".

Again, fundamentally the engine is being asked to deal with an aircraft fuel system phenomena, not exactly chicken and egg.

airfoilmod

If HP upstream is problematic for tube failure, why did GE choose the design? The Fuel as heating fluid may not be GE's strong suit, but it's HP design. Again, the Fuel does not gain much Heat, it wasn't the design issue. The idea would be to Heat the first place Ice forms, which is not the FOHE. As a final fix, if it is the FOHE that is to remediate the Icing problem Plug, it's the same as accepting the presence of Pipe Ice. As a final fix, that is ludicrous, for reasons stated above.

AF

sooty655

I can understand the suggestion that placing the FOHE after the HP pump may protect the GE system by "mashing" the ice to slurry in the pump, but doesn't the GE arrangement have an LP filter?

If so, the ice would collect there, and presumably melt eventually, and any blockage would be relieved by the bypass.

Rightbase

With nice clean fuel and nice clean oil, it doesn't matter which flows through the little tubes, and which flows round them.

But if the fuel is lumpy, perhaps because some water has been pumped into it via an exceptionally cold main tank, it might be better to put the nice engine oil through the little tubes, and let the lumpy fuel go round the outside.

airfoilmod

Thing. And Ice wasn't a design constraint. IMO.

Sooty - I can't find the GE schematic. The RR has an LP Filter downstream the FOHE, and appears to be embedded in an Oil flow system as is the Fuel. The LP seems to be an Impellor design, the HP a lobed, or geared design. So the Fuel goes unimpeded from Tank to LP to FOHE. (Roughly).

Only in Hindsight, and knowing Pipe Ice is a Real problem, are all these comments pertinent. But it would seem that putting the FOHE as the skinny kid, closer to the Nozzles, may have prevented Trent from taking the flak. But then we wouldn't be aware of the Problem at all. Perhaps never. So far, a totalled Bird, a broken leg and some brown Laundry offloaded Delta at Atlanta. So Far.

RightBase - To be fair, the FOHE wasn't designed strictly for high and cold. Down low, and warm, the Oil is in more need of cooling than up high, where most everything's cold to begin with. To get best exposure to the Fuel (which isn't so cold when low,) the design (IMO) slows the Fuel down, by necking down the cross section and adding surface area F/O.
Gives the transient Fuel more time to soak up some calories. Ice? What Ice. The increase in pressure and slower velocity becomes problematic up high, especially when Pipe Ice can occlude the passage. Which is what is frustrating, where the Oil isn't so hot, and the Fuel is very cold, the need for an FOHE is debatable, especially when Thrust is way short of TOGA or Cl. Hence, why not a (relatively) simple bypass for the Fuel? The Oil has one, it bypasses the FOHE when it cools down, Shouldn't the Fuel bypass at the same time? Or in an Icing regime?

AF

tanimbar

Warning: I'm non-professional; not crew, not engineer - just guest here, thanks.

The thread has now turned to an area that intrigued me some time ago (419, page 21) and caused me to write the following.

The question was not answered then but I'm hoping one of you will be kind enough this time.

Thanks in advance.

Regards, Tanimbar

Dak Man

ImageShack - Image Hosting :: ge90dacb777.tif

Indeed Sooty the GE90 does have an interstage LP filter which would go into bypass if clogged with ice. When in bypass the ice can migrate and block the bypass loop, cavitating the HP pump and ceasing all flow to the FMU.

Edit.

Having had a closer look at the GE90 schematic I was mistaken here, the LP filter (and hence bypass) is in the HP loop so the chances of ice ever getting there is slim as it would get mashed in the HP Gear Stage. However there is no bypass loop for the interstage strainer

sooty655

...to collect in a GE90 interstage strainer (or block a T800 FOHE) but not enough to fill the strainer and overspill into the bypass. I wonder if there are any records of GE90s with transient high strainer dp or bypass opening - would the pilot be aware of it?

It may be that variable conditions (and thus probably variable quantities of ice released) make the GE90 also an accident waiting to happen.

Sooty

Dak Man

Fuel systems are routinely Qualification tested with fuel temperatures down to -54°C in the tank and -45°C at the LP inlet. It's fuel in the region of -8°C that is causing the headaches.

tanimbar

You said,
.

I know. I should have made it clear that my question arises purely from curiosity. I'm not implying anything by asking, "What happens next if fuel at or below 3C enters an RR engine inlet?".

Regards, Tanimbar

Dak Man

Nothing untoward.

This table you mentioned seems a bit suspicious to me.

As I've said the engine fuel system is designed to accept fuel (with 260ppm of water present) at the LP inlet down to -45°C (fuel temp) without issue (and they do). What is now evident is that this is possibly not the worst case scenario due to the type and formation of ice present.

This kind of system icing testing has in the past prompted modifications to FMU Flow Washed Filters where sevo flows are traditionally tapped off. Obviously if a flow washed filter becomes clogged (it shouldn't because as it's name applies the filter is constantly "flow washed") then you can lose control of the engine as VSVA contol et al are servo flow controlled / powered.

tanimbar

Thanks for the reply. I think we might not be discussing the same thing.

If I understand the Airbus article properly ( http://www.airbus.com/store/mm_repos...6_low_fuel.pdf ) then the 'minimum engine inlet temperature' is the minimum temp. the fuel should be AFTER passing through the 'oil cooling system', i.e. the fuel is about to be burnt.

To quote from the article, paragraph entitled Minimum Inlet Temperature,

Of course, I might be completely misunderstanding the terms used and their meaning, in which case I'll find a dark corner .......

Regards, Tanimbar