snanceki

Hi,
Thanks for that report.

I assume these were all tests on supposedly the SAME type of fuel (Jet A1 or whatever it might be called in the USA).

The report doesn't indicate the specification limits nor the expected range limits for most samples.
So were these 19 samples in spec?

Differences look considerable. So why such differences?
This supports my fuel quality is a "Black Art" proposition.

I wonder why vapour pressure, calorific value etc wasn't (aren't) checked?

If I was buying aviation fuel in vast quantities I would want to be sure that the stuff wasn't being "diluted" or in other words providing less bang (mpg in car speak) for my buck.

Interesting.

chris weston

Snanceki sir, I very much agree that the evidence so far points to it being a fuel related event.

Without going over old ground too much, suffice to say that if we have an abnormally low fpt, and this fuel seems to have had just such, the lower intermolecular bonding levels that allow this fpt will also make for a correspondingly increased propensity for the fuel to go gas phase in the HP pumps - ie to cause cavitation when the FADECs commanded an increase in fuel flow rate at circa 700 feet.

A proper statistical picture of evidence of cavitation damage in other 777s would be very useful and I bet Boeing/RR have looked at this, hard. It could be more of a problem than is realised with all specs/SOPs et al on 038 being met with the unforeseen until now problem built in.

I have deliberately omitted further comments on additives and stratification.

Much respect from me for the chaps on the flight deck that day

CW

snanceki

Chris Weston

You appear to be in agreement with my position.

i.e. Looks like a "standards" / design parameter issue.

airfoilmod

Good Morning, Sir- Very difficult to read for understanding on the Net. I like to hear and see while discussing, old pilot. Your purpose in quoting my rhetorical question isn't clear to me. I must say though that the rest of your post describes my conclusions quite well, such as they are. My Bottom Line in this has to do with ETOPS. This isn't a generic incident for me. I get from your concerns that you may share my view. Without being specific, I believe that mitigating alterations to procedures have been created. I know of a few and believe that there are sufficient changes to protect the domain of Twin flight. Some are mentioned on this Thread, as well as other places, some are proprietary. To repeat something I wrote in February, just because one isn't directly aware of solutions doesn't mean they don't exist and haven't been engaged. To think in a fearful way that because AAIB haven't released a report sufficient to please all the interested public and somehow there is an unknown danger afoot is silly.
I have perfect faith in flying and in ETOPS.

BRGDS Airfoil

chris weston

Snanceki you are correct, that is my basic view too.

But as ever, the devil is in the detail and for that we must wait.

I have unequivocal respect for AAIB too.

CW

Mariner9

Someone above published a survey undertaken by Grumman. It confirmed what I (and some others) have been saying from the start - namely that the formulations of jet fuels vary hugely depending on crude source, production process (eg hydrotreated, severly hydrotreated etc) additives used etc etc etc. The above list does all meet spec by the way. But for commercial privilege, I could publish a list of a few thousand jet samples, several hundred or so of which are off spec for a variety of parameters (but the high proportion of off spec results is not surprising given that I investigate Jet contaminations for a living)

The reported freezepoint for the subject fuel falls within the limited sample range (19 samples?) in the Grumman list - it was ABSOLTUELY TYPICAL of the value you'd expect to see. If low freeze points were a problem, Defstan/ASTM specs would specify a range of acceptable freeze points not just a maximum. So please, do not continually harp on that at the FP was lower than it should be - it was not.

Few other pointers from preceding few posts:

Vapour pressure: This depends very much on fuel temperature. The lower the fuel temp, the lower the vapour pressure. If the fuel had an abnormally high vapour pressure the problem would have arisen during highest fuel flow rate and highest fuel temperatures - which would have been at takeoff. In any event, if the vapour pressure was too high, it would have been off spec on flashpoint (which it apparently wasn't)

Cal val: If cal val of the fuel was insufficient to provide sufficient thrust to arrest descent, how the heck could the a/c take off in the 1st place? In any event, Specific Energy forms part of the spec, and there is no indication that this parameter was off spec.

Additives: The allowed additive list for Jet A-1 is very restrictive, and additives used must be stated on QC's. If excess or unauthorised addives were used, it would have been picked up on GCMS.

Another point: Various people have talked about stratification. If the fuel had stratified, and the "good" components been consumed during the cruise leaving only "bad" stuff (technical terms )- this would have been plainly evident to the AAIB by simple comparison of the measured quality parameters of the remaining fuel with the reported quality parameters of the supplied fuel. There is no suggestion in the AAIB reports that fuel quality differed significantly than certified at loadport.

Edited to improve the English (originally typed upon return from the pub )

snanceki

Hi Mariner9,
So you are a Jet fuel "expert". Excellent.

If your job is Jet fuel compliance then this suggests that fuel quality is important although I have no knowledge of what parameters in the fuel may be of greatest concern and why. Maybe you could enlighten us.
What is the most common cause of non compliance?
What causes this non compliance?
Why is this deemed a concern?
Incidentally what happens to fuel that doesn't meet spec?

I suggest that you missed my point.
I did not intend to convey that the fuel in the subject aircraft was substandard in some way (the AAIB appear to have eliminated this by stating that the fuel was "in spec") but merely pointed out that ACCEPTABLE "quality" appears to have a wide range.

I certainly didn't suggest that the fuel was "so bad" that it wouldn't burn, for the reason you pointed out (would have been determined during take off/climb) but since Cal value = heat from fuel = Thrust, a lower cal value will give you less "bang for your buck".

HOWEVER, this wide range of acceptable quality MAY.... in some yet to be determined way..... have been implicated in the chain of events that occurred to BA038.

SOMETHING caused this incident.

As I stated I suspect we are talking about STANDARDS / SPECIFICATION /PROCEDURE boundaries in some way, otherwise the cause would have been determined by now.

snanceki

Good morning.
I quoted your rhetorical question since I believe the answer lies in something that is COMMON to the two separate, isolated and independently controlled systems.

ETOPS, from my extremely limited knowledge of the subject, appears to be working well, but we should not overlook that fact that certain aspects of the (duplicated) installation are COMMON.

It appears that we may not be looking at something having gone wrong, but rather a situation where the interface between
FUEL, ENVIRONMENT, DESIGN, STANDARDS and PROCEDURES has created an UNFORSEEN situation / scenario.

FullWings

From reading the AAIB reports, it would seem every component they've tested so far does/has work(ed) according to design parameters. This includes the integrity of the fuel supply system from both tanks.

We have symptoms of fuel starvation to the power units, cavitation damage to the engine pumps, tanks pumps still running (no pressure warnings) and plenty of fuel in the tanks: logically, this points to some restriction between the tank pumps and engine pumps. So far, nothing has been found, so the restriction appears to have been temporary in nature. I presume a lot of time is being spent trying to figure out what this might have been, especially as it affected both L & R systems almost simultaneously.

We're down to contaminant (nothing significant found yet), odd fuel behaviour (ditto), physical change to fuel lines to reduce cross-sectional area (no evidence for this at the moment) and ??

I'm hoping something might come from a detailed environmental test of the fuel supply chain - no news yet...

Smilin_Ed

Quoting snanecki, "If, as I suspect, the range is normally in the order of +/- 5 degC from whatever is the agreed "norm" is, then what made this fuel have such a spectacularly different freezing point?? OR is this very low freezing point understood, and simply a red herring?"

Right on snanecki! I tried to say this earlier but couldn't bring up the words that you have.

Mariner9

My job is not routine compliance testing (though it was a decade or so ago), we only get involved once problems are identified.


Perhaps not surprisingly, the most frequent problem by far in Jet fuel is failure of the simplest test available - visual appearance, where the spec is "Clear, bright, and visually free from solid matter and undissolved water at ambient temperature". Generally, this problem can be solved by simple filtration. In extreme cases, the fuel is downgraded to regular kero or blended into gasoil.

Another common problem is flashpoint, particularly if the fuel is transported at sea after a previous low flash cargo. Reduced flashpoints are very difficult to resolve, necessary blend ratios are impractically large. Typically, the most economical action can be to re-refine the material, but that will always pose production problems to a refinery that is probably already maximising Jet output - additional kero in the crude stream will reduce output of gasoil and gasoline.

An interesting one, and possibly of relevance to BA038, is particulate contamination. This is measured by filtering some fuel, and weighing the particulates collected. An associated test (which forms part of the military spec) is filtration time, which specifies the minimum time for 1 US gallon to be vacuum-filtered. We have had numerous problems where anti-icing additives have been incompletely mixed, which for reasons we dont fully understand, cause difficulties in filtration. Obviously, filtration difficulties would impact on the flow of fuel through aircraft filters, and I would imagine that the effect would worsen with decreasing temperture. However, I have no idea if the subject fuel contained any FSII - chances are it didn't. I also don't know if the AAIB have had particulate contamination/filtration time checked as strictly speaking, the limits only apply at the point of manufacture. It would seem unlikely that they haven't checked this basic point however.

The remaining common failure is the JFTOT (Jet fuel Thermal Oxidation tester"). This arises where the fuel storage or transportation includes incompatible metals such as copper. It can generally be treated by the addition of approved metal de-activator additives to the fuel.

I didn't, but concede that I didn't answer it properly!

Briefly dealing with the "bang for buck" point first, the answer I give below should hopefully answer the main query posed by you and others, which I interpret as "even though the fuel meets spec, could the wide range allow the presence of unsuitable components that wont burn properly"

Specific energy for Jet is specified at minimum 42.8 MJ/Kg. However, it does not have to be measured. Instead, it is acceptable to calculate the specific energy based on certain measured quality parameters (density, distillation recovery temperatures, aromatics content, and sulphur content). This shows that the authorities consider it reasonable and safe to conclude that a fuel which meets spec for those key parameters must produce sufficient energy, even if the material is not wholly "kerosene". Rest assured, this has been based on years of research and analysis. Clearly, they do not consider that other materials with similar parameters but adverse energy characteristics could be substituted. In any event, other burning characteristics of the fuel such as smoke point are checked.

However, I (and presumably the AAIB) accept that while this is a reasonable assumption, it would have to be checked out for the subject fuel. I would be amazed therefore if they have not yet measured the specific energy of the fuel. The point remains however, if the SE was low, the aircraft would never have got off the ground. Also, flamibility issues could not account for the apparent cavitation damage.

Certainly did, and its clear that the fuel has to be one of the major suspects. However, if it was, I would suspect it was more to do with its fluidity/pumpability in a combined low fuel temperature/lowish fuel quantities (minimising the net positive suction head available) environment than any other quality issues. (Cant recall whether I've mentioned this before, but the viscosity of Jet is specified only at -20C)

tanimbar

Some here are saying the -57C freezing point of the remaining fuel is spectacularly low.

Actually this does not seem to be true. If you examine what Shell have to say about fuels around the world, you will read that there are many different grades, different constituent standards and different methods of measuring temperatures.

http://www.shell.com/home/content/av..._10081004.html

There are Russian and Chinese sections on the above page. I mention Russia because there have been reports that the indigenous Chinese suppliers cannot produce enough fuel and have been importing Russian equivalents which have "a low freeze point (equivalent to about -57 degrees C by Western test methods)" - quote from the page above.

My view is that we here don't have enough information to make any conclusions about the provenance, name, constitution etc. etc. of the fuel on BA038. All we know is that the AAIB say the fuel, "complied with the Jet A-1 specification".

Regards, Tanimbar

Mariner9

They are, but rest assured, it isn't.

If it was TS1, it would not have met ASTM/Defstan Jet A-1 specs for flashpoint.

gas path

The entire fuel manifolding from boost pumps to the engine interface has been removed intact from the left wing and is at Farnborough. It is installed in a test rig for simulation purposes.

The filters have a bypass! The filter is monitored by the diff. pressure s/w which would send a discrete signal to the engine EEC before the bypass valve opened.
The result would be a status msg. ENG FUEL FILTER L (R)

airfoilmod

A friend @ Northrop/Grumman, a Thermal engineer has pointed out the discrepancy that follows. Freezing point of -78, -53, (Average)and -42 Maximum are expressed in degrees Fahrenheit. Quoted OAT (at times) along the noted Flightpath are -78 degrees with Fuel Low Temp of -38 degrees, Centigrade. Should the discussion revolve around values expressed in the same measuring system?

Gas Path, does the Fuel/Oil Heat exchanger also have a bypass, allowing for flow should the Heater Plug Up?

Airfoil

Poof in Boots

Gentlemen. Sorry to divert attention away from fuel for a moment

I have trawled through this massive thread and this fact does not appear to have been picked up, with regard to the credit the crew received for their "quick thinking", for reducing the amount of flap at the last moments.

I understand the command for a 'go around' was given, whereby the flap setting would be changed anyway from a landing configuration. If this is so, then the flap retraction and its subsequent benefit, was by default.

Was the command "go around" given?

FullWings

I doubt it, given the situation was a developing undershoot being caused by a lack of power, despite firewalling the thrust levers. What would you "go-around" with at this point?

Not all go-arounds require a change of flap setting, so a call could be "go-around" or "go-around flap X", depending...

airfoilmod

Poof, where did that information derive? It's reasonable, and I can't recall seeing it here. With a serious AoA problem and 108 knots @ 200 feet, I think we can say it was an "unstabilized" approach, demanding GA. The 108 knots happened under A/P, so the GA command is definitely something to entertain, whether it happened or not, I think is unknown to most.

Airfoil

Mariner9

Correct Airfoilmod, the Grumman report does show temps in F. By conversion, the lowest freeze point they identified was -61C. (Interestingly, the highest freeze point they measured would have been off spec for Jet A-1on that parameter!)

Regarding your point about mixed units, both the prescribed test method for freezing points and Defstan Jet specs both specify the measurements to be in degC.

gas path

No! But the FCOC and filter are 'part and parcel' one assembly. The fuel temperature is raised prior to reaching the FCOC as it has already passed through the (LP centrifugal pump) side of the combined engine driven fuel pumps.

apologies for the poor picture!