Fire warnings - an intellectual debate on this contentious subject
 

The AW139 is unusual insofar as the Flight Manual provides an additional line in the proposed drill for Engine Fire In Flight that really needs further explanation but that explanation, it appears, is absent. That line says "Confirm engine fire". The lack of explanation has led to ambiguity in the way the drill should be interpreted. It appears that the consensus at the ATO is that the drill should always be completed as printed regardless how you answer the question posed by that line - is there a real engine fire..... or not?

Here is an interesting intellectual puzzle. Imagine two AW139's flying around IMC over the mountains, there's no radar service available and the cloud goes all the way down to the mountain tops and halfway to the valley bottoms.

Both aircraft have an engine fire warning and the crews begin the drill as per the QRH. One captain proceeds to adopt the 'as taught' procedure and arrives at the end of the drill with no more fire extinguishers but the light remains on. He knows that this could mean the fire is still burning or it could mean that it was a false warning and there never was a fire. Either way he is duty bound to follow the QRH drill or suffer the consequences of a subsequent legal case should his attempts at survival be only partially successful and passengers die as a result. He must now obey the strict instructions in the QRH to LAND IMMEDIATELY. This requires a descent 'at once' and implies that any risk is a lesser risk than staying airborne. Although he has doubts about the sense of his actions he is committed to obeying the book rather than his instincts. He begins a descent towards the mountains and trusts to luck.

The second captain has spent hours debating this very subject and his mind is prepared for a different analysis. Upon reaching that line in the drill the captain calls a halt and asks for help from crew and pax in confirming that there is a genuine fire. There are no signs on the instrumentation, no smells, no signs no smoke (actually ionic emissions) being detected in the baggage bay. Both pilots take turns to loosen their harness and open their doors to peer outside, no external smoke or flame visual from the engine bay area. He knows that if he takes the drill beyond that critical line he will be committed to the possibility of a forced descent as he too is a believer in the discipline of the checklist and if he gets to the line that says LAND IMMEDIATELY then he too will make a forced descent. He waits, still no signs of fire other than the lights. He's descends to MSA in preparation and LANDS AS SOON AS POSSIBLE. At any stage he is prepared to extend the drill beyond that critical line should the fire turn out to be real.

Who was right?

G

Can they fly single engine to the nearest airport with a suitable instrument approach?

G,
If I may, most types I have flown have also had the line - check for signs of fire- and certainly, if it hadn't got that particular line then it certainly has been taught to me or, if you prefer, implied. I believe it is with the belief that no system is infallible and therefore, it remains with the aircraft captain/crew to come to the conclusion that there is a threat to the aircraft. Therein lies the reason for the big bucks!
Nothing that I have experienced in aviation has happened in such a timeframe that there hasn't been time to take that "condor moment" in order to achieve the aim - that of getting the people you are responsible for on to terra firma and for you to walk through the door that evening and tell your tale.
Me personally, I'd make bloody sure before pulling levers/ pressing buttons that could quite easily put in a bad place. No external influence of a legal nature would have any bearing on my decision, of that I can assure you.
Better to be judged by twelve than carried by six......
Nice thought provoking post BTW :ok:
RTL

Sir K
 

That's effectively what the second guy did for one of the engines, (the one with the indication of fire) would have been at idle.

To secure the engine completely would take the second captain beyond the line in the drill that asked for confirmation of the engine fire. Once passed it he would have been committed to the possibility of a forced descent in cloud into the mountains because the drill is adamant in it's assertion that the pilot lands immediately, at once.

The second guy stuck at the line that called for confirmation of the engine fire and went no further, waiting for evidence, other than the lights, that there really was an engine fire. His wider experience led him to take a sceptical view of all complex electrical and avionic systems. He wanted some hard proof that if he was going to put everyone on board through the extremely risky descent then it would be justified. The first guy was a slave to the drill and was more concerned about his possible day in court.

RTL - just caught your post. I teach at a factory school and the policy is 'follow the checklist'. As a weary 'old hand' I think a lot more like you do but will have to go on teaching by the book. Yes I hope it does provoke some thoughts, thanks.

G.

G
That's what crew rooms are for!!! ;)
Sadly a diminishing environment......
RTL

Every Authority I have flown under has always had a nice Escape Clause written into the Regulations.

For the FAA it is as follows:


Now however one wishes to argue it.....when I am the PIC and I have to make a decision or set of decisions that directly affect the safety of the Aircraft, my Passengers, Me, and people on the Ground....I call the Shots based upon my best training, experience, and knowledge of the factors I have to draw upon.

The Checklist is a piece of Paper written by folks seeking to cover their own Ass and is massaged by Lawyers.

Their interest is not necessarily coincidental with yours as the PIC if for no other reason their Ass is sat in a comfy chair in some fancy office somewhere and not in the Cockpit with you.

Refer to Sasless Rule One: "Ass, Tin, Ticket!".

Do what is best for what is setting on your Wallet and generally you will do right.

If you have to use up that Airframe to protect what is setting on your Wallet...the Boss can replace that machine.....and if the Authority doesn't like you decision and you are there to see them take your License.....you can take up Transcontinental Truck Driving for the time you are out of the Cockpit.

Fire Warning Systems in Helicopters are not renown for their infallibility particularly with secondary indications of Fire.

In the situation described....over mountains obscured by Cloud...and a for real fire that will not go out even after all the drills are done....well sometimes Fate smacks you right between the Eyes.

how many of you have had a "real" engine fire and how many have been false engine fire alarms?

The role of a pilot is to manage at best ....

Diagnose - that doesn't mean blindly believe the red light.
Options - what are the variables? If I'm in the granite infested cloud I'm less inclined to shut down an engine that is providing power without smoke.
Decide - choose one of your options.
Act - get on with your chosen option.
REVIEW - is it going to plan? Probably the most important bit of the whole process.

PS. I've lost count of the number of binned/abridged sorties I've had due to sensor error. I've yet to have an engine fire.

Geoffers,
You are quoting (correctly) from the EASA RFM. The CHC ECL however, includes a further step prior to LAND/DITCH IMMEDIATELY as follows:

If warning remains illuminated and there are positive signs of fire other than the fire warning:
LAND/DITCH IMMEDIATELY

and,

If warning extinguishes, or there are NO positive signs of fire other than the fire warning:
Land as soon as possible

This is hardly a new issue and certainly not exclusive to the AW139. ECL's for many types, for many years have advocated land/ditch in the event of a persistent fire warning, and it's always been up to the aircrew to make the "best" decision commensurate with the circumstances and not just blindly follow the checklist regardless of the consequences.

I'd do whatever I thought was best at the time, taking into account the historical reliability of the fire warning system. For example in the Super Puma family the ratio of false fire warnings to real ones was probably a couple of hundred to one.

In the case you mention, if you get it wrong you die and so don't have to face any repercussions. If you get it right and everyone is OK it's unlikely you would be sued for failing to dollow the ECL.

It's a win-win situation!

Mk 6 et al
 

I am quoting correctly. What CHC have is a proper ECL, the kind of thing you would find in an FCOM.

I am unfortunately bound to teach the party line uninfluenced by a lifetime of SASLESS-like experiences. If this consensus continues then maybe I'll stick my hand up at the next Standards Meeting.

G.

We had the same problem at a Sim School I taught at....never mind reality....teach strictly according to the approved syllabus using exactly the approved phraseology.

I can see a need for Standardization but Industry Best Practices should be a part of that.

Compound that with the "Dipped Shoulder Syndrome" and sometimes the outcomes were not as optimum as we would like.


Another one of these Topics we might dredge up again is Engine Malfunction Analysis and the most useless Diagnostic Instrument known as the "Torquemeter".

Fire
 

Geoff,

The scenario you presented sounds like a flight from Douala to Yaounde.

Or outbound to Ngaoundere. How many times I reviewed emergency procedures out there where there were few if any alternate options.

Those situations still exist in some parts of the world so not a far-fetched "what if".

One day, many yeras ago, during a ferry flight of 140 nautical miles over the southPacific Ocean, alone in a single engine, I got to lighting the red "low-pressure engine oil". I drop the pitch down to going autorotation and light go off. . . . I check my gauges , everything was fine. I put the pitch up slowly up to 14 ° (14.5 ° cruise) the light stays off. I check again the oil pressure gauge and oil Temp : eveything is fine,
Oil pressure a little weak, but in the green arc.
I concluded that it is a sensor problem but without understanding the relationship with the pitch ....
I was 50 NM from the nearest land, so I continue my flight and climb a little be to keep some air under my ass . I continue and monitoring gauges.
10 minutes later, the light turns back on "LOW PRESSURE OIL" again I put pitch down slowly and at 13.5 ° light goes out .... I confirms the sensor failure but one more time i don't understand why ?
I ask for a short circuit and maintaining an angle of autorotation during most of the descent.
When i land, the mechanic came to me .I cut the turbine and she stop in few seconds ...
The mechanic ask me " Where is the engine oil ?"
After check of the turbine, it was An internal seal had gone in flight and I had burned all the oil through the turbine .....
In this case, i apply totally the manual about a LOW Pressure oil gauge failure ... And i am very lucky ... maybe few minutes or seconds more and the engine cut off in flight... Maybe no...

Of the 3 fire warnings in flight I have had, all were found to be false indications from detector and wiring faults. Still I would have to find good reason not to act on a fire warning and shut it down.

In the situation you describe, I would shut it down and fire both bottles and then continue flight OEI to the nearest landing site where I could safely get visual. Descending towards mountains in IMC with an aircraft still capable of flight is madness in my opinion, and I would only do this if I knew for certain that we were on fire and unable to contain it.

My logic would say this:

If it were a real fire, then hopefully two bottles have extinguished it and even if they haven't it should hopefully burn itself out before breaching the engine bay. (Remembering that those engine bays are required to be able to contain a fire for a period of time to meet certification standards.)

If it were a spurious warning, oh well all I have done is shut an engine down and some maintenance is required.

In my view, the chance of a real fire continuing to burn with nothing but a fire warning as the only indication is highly unlikely.

"Rules are for the obedience of fools and the guidance of wise men." -Harry Day.

I have spent considerable time thinking about this. There was an article written by Nick Lappos on engine fires hovering around the sim in WPB some years ago. It was a very interesting read mentioning additional indications and false warnings. I never got a copy but if anyone has I link please share.

In the case of confirming an engine fire on the AW139 the tail camera would be one of my first choices of gathering more info.

Coyote
 

I can see you are leaning my way but to do the drill means you end up reading the last line in the checklist which tells you to LAND IMMEDIATELY. If you do anything else but make that crazy descent then you are unquestionably going against the checklist. By stopping the drill at the line that requires CONFIRM ENGINE FIRE and going no further until you see positive signs of fire you remain in compliance with the checklist. That is a possible way out for the factory instructor as it combines the wisdom of experience with the technical correctness of the drill. Any takers?

Here at the factory school where we all try to teach the same thing and going against the checklist is forbidden we need a workable policy. You can see my dilemma.

Experience over obedience or vice versa?

Memorable quotes generated by this thread so far:

"Rules are for the obedience of fools and the guidance of wise men"

and

"Better to be judged by 12 than carried by 6."

Along with the reminder from SAS about the FAA regs on the role of the aircraft commander.

G. :confused:

I have had numerous false fire warnings and a couple of real ones. At one time I took off with both fire lights on. (That should stir it!)

Believe me: You KNOW when a gas turbine catches fire.

Unfortunately, in many cases it is a 'damned if you do, damned if you don't' scenario for the subsequent inquiry, and has led to much crew room discussion.

'Check for signs of fire': great in most circumstances, but remember the sensors are (generally) around the engine bay and it could be something else. A hot gas leak - should go out at low power so may be a good indicator, but is it doing any other damage?

Like Fareastdriver I too know what a real turbine fire looks (and sounds) like, but not the insiduous problems, although I have also seen the results. Spurious warnings are much more common, particularly in wet conditions, but I've only had two real fires on rotary - one was indicated but accompanied by a loud bang and associated indications; the other was a call from the back - 'hey capt, did you know the port engine's burning?'. Both times I was also well placed for a land immediately situation, but for a few of the spurious ones I was IMC and took the other line with 'no other signs'. It doesn't half concentrate the sphincter though.

Geoffersincornwall, I understand your dilemma. But I'll take going against the checklist over a good chance of flying into a mountain any day. Alarm bells ring for me if a pilot is letting the legal implications of non compliance with a CHECKLIST prevent them from acting on a fire warning IMC over mountains.

In your specific example, I think there is a good case to amend the checklist detailing a course of action should the fire not be confirmed.

However the checklist also says a 1 or 2 minute cool down run prior to shutdown. If you landed a helicopter immediately say, due to an fire in the cabin (I haven't seen a checklist for that), would you sit there complying with the cool down run?

Blind compliance is not the answer for all possible scenarios, and there will never be a checklist for every scenario possible. We are there for a reason, and one being perhaps the ability to reason and sometimes make choices in the best interests of safety when either there is no checklist or the checklist just isn't adequate or appropriate for the given situation.

In my view we should continue to view the checklist as an aide memoir, and not a legal contract. The regulator and operator can mandate the level of compliance with it, and the exemptions justified in the best interests of safety.

Otherwise are we not only designing the pilot out of the cockpit, but also legislating them out of it?