Evaluating engine fire vital actions
 

A bit of an intellectual exercise here. There's an interesting thread running on Private flying concerning the various actions that might be taken in the event of an engine fire in the air. Various opinions have been offered and discussed as to what you should do in the case of an engine fire in the air.

But it begs an interesting question in my mind. Let's say we've got a particular reason for being very certain we're offering the best possible advice on how to deal with an engine fire. How would we go about developing and evaluating that advice? It presents some interesting questions about flight test risk and conduct which I'm glad nobody has ever really asked me to address, but it might be an interesting subject to bat around.


My initial thoughts were to divide this into sections, since generating a real engine fire is just asking for trouble. Avoiding smoke and fumes in the cockpit could be fairly easily addressed by fitting a non-noxious coloured smoke generator in the engine compartment and investigating various manoeuvres / vent settings to see what gives the minimum (and an acceptable!) degree of cockpit smoke ingress.

But as to evaluating means of extinguishing an engine fire for flight conditions - I'm mildly stumped. I can think of ways of evaluating this (probably using the Boscombe blower tunnel, a strapped down aeroplane, and a TP in a fireproof suit) that would work - but also cost a fortune and probably destroy half an airframe and several very expensive engines in the process. I've never seen any analytical tools for the purpose - has anybody else? So how could you do this within a reasonable test cost, if you really had to?

Thoughts anybody?

G

Aircraft Fires. A Good topic particularly for those who have already had one and for those who might be concerned.

Having had one engine fire resulting in a need for a wing change and several false fire warnings I believe the biggest problem is initially knowing whether you have a real fire and then its extent. There just aren't enough clues available to the crews to assess a fire event adequately leading to optimum actions for fire suppression/extinguishment.

Now that miniature high resolution video cameras are available at low cost we can all come up with numerous locations from which we could easily have presented a monitor view of the primary areas of interest. Throw in a bunch of temperature sensors in vital locations, other than those already in place, and we could then have the means of either dealing with an airborne fire or taking the most appropriate actions. Some of those action could even be automatic.

Tis a subject that has been neglected for too long.

A vital discussion. Although this strays somewhat from Ghengis's original post on how to simulate the real thing, I am concerned at the generally accepted airline pilot viewpoint that an engine fire is no great deal and there is no compelling hurry to take extinguishing action.

For instance one operator on the 737 states in the company Ops Manual that for an engine fire warning after take off, no action should be taken until a safe height of at least 600 ft is attained. With one engine failed (on fire) this could mean a low rate of climb and consequent delay in taking the first action. Especially as initiating the first action involves a fair amount of talking and recalling of the fire drill. Others prefer 400 ft agl before the first action is discussed. Either way, the inference is that there is no need to act swiftly.

In all cases, it seems that a fear of identifying the wrong engine and firing a bottle into it, instills a complacent attitude towards the stark reality of an engine on fire and how swiftly other components and their systems may be affected. One airline blithely states that no problem with an engine fire because it will probably fall off the wing eventually.

The best person to consult about this would be a fire engineer or even a fire investigator.

I believe that the way to approach the situation is to first of all considder the fire triangle (heat, fuel and oxygen) and put some figures on how much of each is available in various situations.

Since to put out the fire requires removal of one of the components of the triangle, it makes sense to have a procedure which seeks to remove that component that is easiest to remove.

Another point to be considdered is simply doing nothing with the fire other than landing the aircraft which brings up a whole different set of situations. The first question being, how long will it take for a fire to damage something that makes a safe outcome doubtful.

-----

As for doing nothing until at a safe height. This is often based on the principle that if the fire is the only problem, you don't want to shut down a working engine until you are at a safe height. If the engine no longer works then the priority is to avoid obstacles until at a safe height.

What defines a safe height when your ass is on fire is purely personal!!! :D

Regards,

DFC

I seem to remember from my time working as a TP for an engine manufacturer that engines are certified to contain a fire for at least 7 minutes. Thus, crews should be confident in taking a fire into the air after V1, and climbing to 400 (or 600) ft agl before taking action. The important thing is to carry out the correct action, rather than screwing it up by going too quickly.

Airbendane has a good point. For airline transport civil machines, the fire zone is very beefy, and can withstand 2000 degrees F for 15 minutes without affecting the basic airworthiness of the machine. Military engine fire zones are not nearly so capable. This gives some leeway for careful diagnosis, I think, and the ability to look for confirming indications. I also think the history of the type should give some guidance, since some fire detectors are sensitive at sunset, or on hot days, etc.

For a civil transport generally, to truly believe the fire warning enough to act on it, one should carefully look for some confirming secondary indications, I think. If performance is good, and the available landing area close, then a shutdown isn't so drastic a choice, so do it. But if it is a cold sea, or a long way across inhospitable territory, and all there is is the one light, a delay in shutting down that engine is reasonable. For a less protected military or Part 27 helicopter, there is less comfort in waiting, so the harsh choice might have to be taken.

Nick

While I take your point about not shutting down a servicable engine if you are a long way from land, surely a precautionary shutdown can always be reversed say two hours later should your good donk quit?

JF

The 400" rule on a 737 was explained to me with above mentioned plus the fact that burning engine still produces thust and main priority is the terrain clearance. And indeed, that is how the sim plays it on us. Which brings me to the original idea, is this correct? On the other hand, seized fan creates extreme drag and shutting down the engine comes as a relief. What would be the effects of a stalled/surging engine...?

FD

As an engineer who has had to design fire containment zones, the 2000 degree, 15 minute rule is a correct assumption - this is the FAA definition of fireproof.

The airflow computational modelling techniques around these days can give you lots of insight as to the ability for a defined compartment to sustain a fire (does the cooling air flow overpower the mix?). You still need to perform an FAA sanctioned concentration test to get your ticket but the tools available should get you to within one test instead of a series of expensive, iterative tests

I was always taught that the sensing elements had a certain delay, as do the crew, hence the time scale.

It was assumed the procedure would be to power back (not shut down) determine if the warning goes away (a broken HP line would be one conclusion) and if not, shut off the fuel. If this doesn't put the fire out, the first shot of the bottle should be pulled.

Does this go with or against operators SOP's ??

It will be interesting to hear the users views on the subject ;)

Having personally flown for several civil jet transport operators, I find that they can be divided into two basic types, where engine fires are concerned.

1. This operator mentions that engine fires are definately not good for ones health, and action to shut down the offending engine, and extinguish the fire at the earliest opportunity, should be taken.
Sometimes as low as 400agl.

2. This operator mentions that engine fires, altho potentially serious concerns, nevertheless should take the back seat to using the burning engine, (if indeed it is still producing useable thrust, which is many times the case) until a reasonably safe height is achieved which, in the FAA's thinking in years past, was at 1500agl.

In both cases, we are talking about pod mounted jet engines, not buried in the wing, Comet style.

MY personal preference is for number 2, as in my opinion, this is the most reasonable.

Individual circumstances clearly could dictate, however, shuting down the offending engine earlier...severe vibration, explosion, etc.

I recall an incident involving an HS748 that suffered an engine fire at lift off somewhere in UK and the captain made the instant decision to bunt over and land straight ahead on the remaining runway length. It was fortunate that he did because the fire was so severe I believe the aircraft was burnt out very soon after everyone escaped. Taking his time to climb out and casually identifying the problem when above 400 ft then setting about fighting the fire, would have been a deadly mistake.

HSWL, your HS 748 example is not a particularly good one. Without judging crew actions with hindsight – they were the only ones there at the time; there was no evidence to suggest that the aircraft structure would have failed if the engine shut down drills had been completed i.e fuel cocks selected off. With such a catastrophic engine failures, there may be attention getting symptoms that cause a crew to deviate from SOPs, yet if you decide to land straight ahead and hit the airfield boundary ditch with serious consequences, the view of such a decision could be open to severe criticism.

Some regulators / operators who decide to approve drills that delay shut down drills until a safe altitude may not consider or have all of the facts i.e a WAT limited, slow speed turboprop may take 5 mins to reach 1000 ft, let alone 1500 ft.
Most fire drill SOPs are written by the manufacturer and ‘certificating’ authority, however if an operator or ‘operational’ authority changes these without due consideration, all of the good design work if of little use.

I understand that the 400 ft shut down procedure originated from 400 ft being the minimum flap retraction altitude during an engine out procedure.
--------------------
Airspeed and Upwardness