777 crash at Heathrow: what chance of another such accident?

From the start, the British Airways Boeing 777 crash at London Heathrow felt like a rare event – perhaps a one-off. But is it actually likely to happen again?





After all, like most big jets of its generation, the 777 has had a pretty trouble-free history since service entry in 1995, and aviation is now such a mature industry that, surely, there are few surprises left for the operator or air traveller? 


But the Heathrow accident has provided some surprises. The UK Air Accident Investigation Branch’s latest interim report says ice particles in the fuel system is the only likely cause of this event, and adds that the behaviour of water in fuel below certain temperatures that are regularly experienced is an unknown quantity. The Branch explains: “When the fuel temperature reduces to approximately ‑18°C (0°F), the ice crystals adhere to each other and become larger. Below this temperature little is known about the properties of ice crystals in fuel and further research may be required to enable the aviation industry to more fully understand this behaviour.” The fuel in G-YMMM’s tanks was -22degC on the fateful final approach.


So what is the risk that an event like this will recur, perhaps with a much worse outcome? If the fuel supply restriction that caused G-YMMM to crash-land just inside Heathrow’s boundary had occurred 30 seconds earlier, the aircraft might have hit a light industrial site and crossed two busy roads before coming to rest.


That said, there is a fair amount of evidence in the AAIB’s latest interim report to show that the risk is very low and could be lowered further by the use of ice-inhibiting fuel additives widely used in military aviation. The report narrative confirms that, since every component in the aircraft’s fuel system and in the engines is known beyond doubt to have performed as it should have done in terms of functionality, and that “waxing” of the fuel at that temperature would not have been an issue, ice particles or crystals clogging the fuel system is the only explanation for what happened.


Why it occurred on this and not on other flights is not yet completely clear, but it seems likely to have been the particular combination of a long fuel tank cold-soak at lower than usual temperatures, combined with an extended period of commanded low fuel flow until high power was demanded just before the unwanted power reduction on final approach.


Working with Boeing, the AAIB was able to mine data from 13,000 Rolls-Royce-powered 777 flights. Among these, they found 118 flights during which the fuel temperature profile for the whole flight was at or below that for G-YMMM’s final trip, and others in which the fuel flows demanded were close to the profile for this flight, yet they did not experience any problems with fuel supply to the engines.


Because there was no fuel supply system component malfunction in G-YMMM, one of the AAIB’s recommendations is: “The Federal Aviation Administration and the European Aviation Safety Agency should take immediate action to consider the implications of the findings of this investigation on other certificated airframe / engine combinations.” That means everything flying, not just 777s.


Meanwhile it has taken 17.5 million flying hours and 3.9 million flights by all 777s for this one event to occur. Specifically for Trent 800-powered Boeing 777 s, they have flown 6.5 million hours and 1.4 million flights.


So all but exceptionally nervous fliers should be able to feel safe in the 777.

9 Responses to 777 crash at Heathrow: what chance of another such accident?

  1. Blu Yonder 12 September, 2008 at 3:04 pm #

    “So all but exceptionally nervous fliers should be able to feel safe in the 777.”

    Exactly why should I feel safe on a B777 after this half backed report makes me wonder. Where in this report is a conclusion of any sort and a definite reply to the questions people are posing regarding this incident?

    The B777 has had a number of uncommanded rollbacks in it’s histroy, recently an AD on the GE90-115s on the same situation.

    You mention millions of hours flown atc. What about the inverse? Why is it that after these hours this has happened and NO other aircraft with the same or similar conditions has had such problems?

    The lack of actual correcting measures smells of smokescreen to me. If, as your article suggests, this can happen to any aircraft, why is it that no preventive action was disseminated to operators?

    In the meantime Boeing’s cashcow continues flying unrestricted everywhere.


    Blu Yonder

  2. Adam 12 September, 2008 at 11:17 pm #

    Air Canada had issues with early DC9-30s having fuel valves freezing and fuel quantity indication systems fail in winter. I have seen rafts of ice in tanks. Anti-ice additive injection during winter Canadian feulling went a long way to fix this. Also retrofit of teflon covered valve sealing platforms. Sumping during winter was accentuated at southern stations.

  3. Branimirzahariev 13 September, 2008 at 10:52 am #

    The cause of this accident is still unclear. I think that the interim report issued by AAIB (UK) does not give definitive conclusion and does not reveal the possible reason for both engines thrust to be retarded. If the problem (which is most probable) is a momentum software malfunction hiden somewhere in the AIMS and EECs, than it can not be confirmed now. The ice in the fuel can cause problem during Take off phase, but it is less probable to be during the phase of landing or flare. Let us see what UK AAIB and NTSB will find further.

  4. Andrew 18 September, 2008 at 3:26 am #

    What appears to have been overlooked here is that if as an example there was an emulsion of 40% water/ice and 60% fuel (I believe the fuel delivery pumps would create an emulsion) then 40% of the energy available in that fuel has been lost, which explains a loss of power without engine shutdown.
    The greater the proportion of ice and water, the lower the power produced.
    I cannot conceive that ice would signifantly block a 100mm dia fuel delivery line.

  5. peter field 18 September, 2008 at 9:41 am #

    The “elephant in the room” in the current AAIB interim report is the lack of detail of fuel conditions AFTER the preheater and the design operating parameters of the preheater.
    All combustion engineers will cringe at the kerosene layout wher it passes through a preheater(that could malfunction), then to the pump, then to a SPILLBACK type metering valve, that passes fuel back INTO THE INLET OF THE FUEL PUMP. Any problems (like ice restricting flow in preheater causing overheating) could start a minor gassing or foaming process that would inexorably get worse as gassing kerosene is passed back to the inlet of the fuel pump. The pump performance would slowly deteriorate. On panic advancement to full open metering valves, several seconds are usually required to flush out gassing kerosene. All this time the cavitating fuel pump could get cavitation marks on it.
    All kerosene combustion engineers prefer to spill back kerosene back into the fuel tank NOT NOT into the inlet of the pump.
    Please AAIB, give us more details of fuel conditions after the preheaters and operating parameters of the preheaters.

  6. Blu Yonder 22 September, 2008 at 10:24 am #

    Mr. Field, very interesting insight. From an operational point of view, makes you wonder flying one these planes for 16hrs+ maybe transpolar route! Still no one ever dares question any system on this particular aircraft.


    BLu Yonder

  7. Gerald Wilson 20 November, 2008 at 12:04 am #

    I still fail to understand why there is no recommendation to use fuel with FSII, which is the norm in military aviation.

  8. Gareth Davies 23 December, 2008 at 6:53 pm #

    This is another problem like the Challenger and Columbia orbiter losses where there is a known issue (in this case low fuel temps and an imperfect fuel system design that is potentially fail unsafe near the ground) and solutions, that are not being addressed, it seems, because there is a low probability of occurring. In this (BA EGLL) case it occurred. Since every B777 with RR trents has to land it has a probability (albeit small) of happening again, near the ground. The bottom line is when the pilot needs power it should be there. In safety terms it is ludicrous if not deliberately negligent not to address this fully.

  9. David Cement 24 September, 2009 at 1:21 am #

    I trust Boeing over Airbus with their computer knows best avionics and gay joysticks. Pilot trust and yokes please.