Engine Fire on Ground

[Captain Smiley]

What are your views on putting the burning engine downwind after a rejected take off because of engine fire.

This is what my company SOP is, but if you were taking off with a 90 degree xwind from the left and your left engine caught fire then that would mean making a 180 degree turn before stopping.

The last company I worked for wanted the crew to turn directly into the wind.

[BOAC]

Any turn towards a fire (headwind) improves survivability. Do what you can - and follow SOPs! Maybe that should be the other way round...............

[Carnage Matey!]

A large dose of common sense required here. If the wind is 90 degrees off then turning in either direction is just going to direct the flames towards a different part of the fuselage so why waste the time turning? Any other scenario the turn might be helpful but not at the expense of an expeditious evacuation. Turning the aircraft directly into wind seems somewhat impractical unless you operate an aircraft with a tight turning circle off wide runways. More chance of going off the hard stuff before you accomplish the turn and now you've got two problems to deal with!

[BOAC]

I reckon a 45 left would save a significant chunk of the fuselage, but each to his own.

[Tee Emm]

First of all you have to find a windsock which at many aerodromes is situated a long way from a runway. At night it will seldom be possible to see a lighted windsock and even more difficult to pick its indication in the glare of a fire wrecking night vision - if any. Of course you could always remember what the ATIS wind was and mentally subtract or add a desired aircraft heading from there. In any case if the fire is on the take off run, and an abort takes place, it is all too easy to get be-fuddled with the mental maths of remembering which side is the upwind or downwind side after stopping and the cabin crew are hammering to get the pax out. Perhaps therefore the clue is to include the final direction of stopping during the before take off safety brief.

[PAXboy]

Non pilot speaking. Let's be brutal: Fire/RTO/evac/many dead both inside and outside the machine. At the enquiry the chief pilot/cheir ops/cheif of cheifs says:

"Out commander followed their instructions to the letter and we commend him/her for that. Given the tragic loss of life, we must look again at our prodcedures."

"It is with surprise that I must report our commander did not follow company instructions. Given the tragic loss of life, this is particularly regrettable ..."

You know that they are going to shift the blame because that is how business works. Of course, you need to do your best efforts but, as I understand it, SOPS are there so that - in the confusion of an incident - you have instructions based on experience.

[BOAC]

Tee Emm - it is not THAT difficult! Most Tower controllers give you the wind as you are cleared for take-off? I feel we are making mountains out of molehills here.

As for

Perhaps therefore the clue is to include the final direction of stopping during the before take off safety brief.

- you mean you don't? I always ask my handling F/O after HIS/HER brief which way they EXPECT me to turn as well. At least then they think about it as they 'progress'.

In a former airline I flew with an excellent safety conscious Captain who even briefed the c/crew on the wind direction before take-off as well.

Paxboy has it in a nutshell - apart from what I hope is your natural desire to save lives, there is always 'the management' to think of. SOPs SOPs SOPs....................... If they are wrong, get them changed.

What is VITAL is to try to keep the fire away from the fuselage if you can. IIRC, BA's SOPs up to the tragic Manchester 737 accident were to vacate the runway if possible, and the unfortunate Captain followed those with the resultant loss of life, as it put the fire upwind. The SOPs changed. The Captain suffered badly psychologically.

[lomapaseo]

I am concerned about the discussion here.

When dealing with emegencies their should be top level steps to perform (peeling an onion) followed by lower level, time and conditions permitting.

The top level steps should be from wrote (simulator training, etc.) Fly-the-aircraft-first, the lower level steps may be checklists types to ensure that you haven't missed anything assuming that you have your aircraft under control.

For engine fire on the ground (below decision speed) , I assume that the first thing you do is to stop, immediately followed/or in conjunction with turning off the fuel to the affected engine.

Individual companies may offer additional procedures, but if you really do have an external (outside the engine) fire, the crew is unlikely to know this sufficient for confirmation and decision making, before they stop by that time I suspect the cabin crew have made their decisions.

[safetypee]

I agree with lomapaseo, crews must deal with the high priority issues before considering alleviations to less likely situations. Much of the discussion has been focused on a fire external to the engine zone, yet aircraft certification has stringent requirements for fire containment and extinguishing; some of these depend on correct crew action.
The Manchester accident was a very rare occurrence, and one single incident should not drive ’vital’ procedures unless the investigation identified major weaknesses. Stopping safely, shutting the fuel off, and evacuating the aircraft are vital actions; they should not be compromised by further restricting the mental processes, already under stress due to the surprising situation, with consideration of wind etc. Note that in the Manchester event, most if not all of the deaths were due to smoke and not fire; the uncontained fire was outside the cabin and only subsequently burnt the aircraft.
Turning into wind is a good safety procedure, but it is a nicety, and thus should not detract from the critical issues of stopping, fuel off and evacuating.

[BOAC]

Turning into wind is a good safety procedure, but it is a nicety, and thus should not detract from the critical issues of stopping, fuel off and evacuating.

- nor does it. Since it takes place (where possible) in the last second or two of the stop it interferes with nothing and would have prevented a large number of deaths at MAN had it been the SOP then. I would raise your 'nicety' to a 'desirable'.

[safetypee]

Originally Posted by BOAC

I would raise your 'nicety' to a 'desirable'.

Yes I would agree with that. Re-reading the accident report (Manchester 737 Aug 87), a point of concern was that the fire burnt through the fuselage allowing smoke from the pooled-fuel external fire to contribute to the toxic conditions in the cabin. The wind was a factor in this and hence the AAIB recommendation for crews to have procedures.

4.1 Procedures should be developed to enable the crew to position an aircraft, when a ground fire emergency exists, with the fire downwind of the fuselage. Visual indicators of local wind direction located within the manoeuvre areas would be valuable aids to the implementation of such a procedure. (letter to CAA 14 March 1986)
4.2 Research should be undertaken into methods of providing the flight deck crew with an external view of the aircraft, enabling them to assess the nature and extent of external damage and fires.
4.3 Operators should amend their Operations Manuals, if necessary, to direct crews on any rejected take-off or emergency landing to stop on the runway and review the situation before a decision on clearing the runway is made.
4.4 Consideration should be given to the requirement to fit an evacuation alarm permitting flight deck crew to instruct cabin crew to initiate an evacuation immediately, or if the aircraft is still moving to prime for an evacuation immediately the aircraft is brought to a halt.

Note that 4.2 recognised that crews may not know what the external situation is; therefore I suggest that without a clear external view time should not be wasted in further assessment before evacuating. I think that a similar comment applies to 4.3, if you stop on the runway and you are not sure of the situation then get out – without considering turning as this only delays the evacuation.
I would be interested in the results if the AAIB were to reconsider the impact of their findings, and hence recommendations in the light of new cabin layout requirements which speed the evacuation and the use of fire/smoke retarding furnishings.

BOAC, I would not agree with your assumption that the last second or two makes no difference, the Manchester report suggests otherwise.

To answer the original question; if an operator’s procedures require a turn then it should place the fire downwind of the cabin.

[BOAC]

I would not agree with your assumption that the last second or two makes no difference, the Manchester report suggests otherwise.

- no - you have misread it! 4.3 refers to 'clearing the runway', not turning AS YOU STOP. The stop takes a finite time. That cannot be reduced. In the last 'one or two seconds' the a/c can be turned. That is all there is to it. No delay to 'clear the runway'. No delay in stoppng. No turning AFTER the stop. No delay in any subsequent evacuation. Let us not make this simple concept any more difficult than it need be, please. Manchester cannot be dismissed as "a very rare occurrence" - it was a severe fire during take-off, for whatever reason - yes, thankfully rare - but what we train for.

I do not know how deeply you are involved in 'evacuation decisions', but it needs a little careful thought - it is not just 'chuck them out' regardless since injuries will occur if you do. There is often a period of evaluation required, even following a fire warning. Such does not require an 'automatic' immediate evacuation.

[safetypee]

BOAC it would be futile to quibble over1-2 sec while the aircraft is rolling to a stop, I am sure that we both appreciate that not delaying the evacuation is important and the decision will depend on the situation; – lets not debate the words as are there are more interesting issues.

We have different views of the uniqueness of the Manchester accident; it was a rare uncontained engine failure and a severe fuel leak which resulted in a pool of burning fuel; even individually these events are rarely seen in what are already rare events (RTO).

Re your “>>… it is not just 'chuck them out' regardless since injuries will occur if you do. There is often a period of evaluation required, even following a fire warning. Such does not require an 'automatic' immediate evacuation. <<
It would be interesting to understand what you would evaluate, why, and when these items are considered in deciding to opt for an evacuation.
If the aircraft is in a situation that requires an evacuation, then the decision to evacuate is self evident. A fire outside the engine or within the cabin could be such a situation – automatic – predetermined by SOP/company policy.
The thread discusses an RTO with an engine fire (a fire warning). The advice to turn may be specific to rare situations, but if it can be incorporated in generic SOPs (turn for all fire warning RTOs) without detriment, then it can be a positive safety aspect.
A concern would be having a drill specific to one situation so that alternative drills would be required for other similar situations; this could increase crew workload in situation assessment and decision making. Do operators do this unwittingly? Do pilots also do this unwittingly by attempting to ‘double think’ all situations – requiring an evaluation period. In my experience of surprising and potentially stressful situations, the crew workload should be kept as low as possible; this can be achieved with good SOP design and controlled thinking behaviour - strong mental discipline.

[BOAC]

In the case of the MAN accident, most people agree that a turn in the direction NOW taught would probably have improved survivability.

It would be interesting to understand what you would evaluate, why, and when these items are considered in deciding to opt for an evacuation

- hmm! Any pieces of string you need measuring while we are at it?

In three airlines now, a period of collection of information - using whatever sources are available - has been actively encouraged following an engine fire warning or engine failure RTO. That is the area of the 'loneliness of command'. I would hate to have broken pax ankles or worse, chucking them off an aircraft with a false fire warning, and equally would hate to keep them on one with a real fire. Sometimes it is easy. Sometimes not. Are you 'hinting', perhaps, that in your mind any RTO with a fire warning should be an automatic, instant evac?

I recall many 'false' fire warnings in aviation history - I believe at least one a/c had 3 aborted take-offs in succession with such (what fun!), and no evacuation took place.

Regarding a 'turn', either towards or away from an engine,

if it can be incorporated in generic SOPs (turn for all fire warning RTOs) without detriment, then it can be a positive safety aspect

- again, taught and emphasised in three airlines, and also for engine failures, BTW, and VERY SIMPLE to absorb - and it IS a positive safety aspect.

[Best foot forward]

Anyone who has seen the pictures of the MAn 737 won't need any convincing that if you ave a fire you put the fire down wind. Simple, like most SOPs they are there for a reason, thats so we can learn from past experiences and not have to live through them. I've worked for a few airlines and all have had the same philosophy, thats how you train thats how you fly.

Say that the Man 737 was a one off and shouldn't be used to formulate a policy is sheer folly.

[safetypee]

Originally Posted by Best foot forward

... shouldn't be used to formulate a policy is sheer folly.

Rare events such as Manchester may shape SOPs, but they should not ‘dictate’ policy and in doing so overlook more probable scenarios. One of the major sources of error is that crews/operators fail to correctly define the problem – failure to assess the relative risk, thus SOPs could be focussed on the wrong problem – they might provide inappropriate emphasis which could bias a crew in their decision making. This is typical of concentrating on the latest or the ‘brightest’ accident without understanding the relative risks. ( Errors in Aviation Decision Making.)
In addition we should not forget the ‘unforeseeable’ events which constitute the majority of accidents – you don’t have the same accident twice (situations), but there are ‘no new accidents’ (same contributions – including human error), thus it is essential to look for the warning signs; in this case ambiguities or conflictions in SOPs.

Ever since power plants have had fire detection there has been the dilemma about false warnings. The reliability of modern warning systems is much better than those of even a decade ago, thus where human judgement was required for confirmation, nowadays most warning systems can be taken as correct, particularly as the probability of a false fire warning is less than the probability of a human misjudging the situation i.e. we can cut one option out of the decision making process. We often fail to appreciate these improvements in our SOPs (SOP revisions).
This could be a hint for automatic evacuation for a fire, but I would balance this with improved fire containment, better extinguishers, and lower engine failure rate – thus the lowest risk might still be with the crew assessing the scene after the stop – the warning system is only one cue – take a look, get other advice, but for the obvious situations then there is no delay - evacuate.
These technological improvements have taken place in many systems and therefore might apply to many SOPs and decision scenarios; although this concept still does not provide absolute safety, the level of safety achieved (achievable) today is better than in those situations where a greater range of judgments were required.

The industry must not forget accidents such as Manchester, but we have to acknowledge the many improvements since then and look ahead for other, as yet unidentified threats.

Slightly off topic; who retains a ‘conditional’ statement in their EGPWS Pull Up SOP? i.e. “if visual with … etc”. Is this a hangover from the old GPWS; is the conditional evaluation (decision) still warranted with the modern highly reliable terrain displays – with significantly lower nuisance rates than pervious equipment? Mounting evidence suggest that crews encountering TAWS warnings due to error, perpetuate the error with their conditional evaluations – some failing to pull up in hazardous situations.
A subject for a new thread ? - Mod BOAC please consider.

P.S. Bff, in your freighter do you turn as you stop from a RTO – for interest, why/why not?

[Gary Lager]

See also http://www.pprune.org/forums/showthread.php?t=230934