PDA

View Full Version : B737 Abort decision with engine overheat indication.


Hudson
13th Oct 2004, 14:25
It is known that Boeing recommend that you do not abort the take off above 80 knots for master caution light indications. Some operators have their own rules setting that limit airspeed at 100 knots. Master caution - engine overheat annunciation as example.

Engine fire warnings in the simulator are usually preceded by an engine overheat indication. At least that is how some 737 simulators are programmed. I have no statistics readily at hand to advise what percentage of real engine fire warnings are preceded by an overheat indication - and in fact how many seconds can be expected between first overheat indication and a fire warning (if a fire is detected). This suggests there is a grey area in the decision making process to therefore abort on an overheat if the event is sufficiently below V1 to warrant an abort over 80 knots.

Example: One simulator is set up to give overheat indication approximately 4 seconds after the instructor presses the fire warning button. Keep in mind that the aim of the exercise is to have the crew initiate an abort at the fire bell providing the speed is an appropriate safe figure below V1. 15 knots, maybe which allows for shock, horror and then reject action.

Another simulator for same aircraft type may be programmed to actuate the fire bell the instant the instructor hits the engine fire warning button on his panel - that is, no prior overheat indication. That makes the abort decision more clear cut.

The problem arises when in the first example the instructor is forced to tell the pilot that he should play the game and pretend not see the master caution and overheat light come on and only abort at the sound of the fire bell and lit fire switch. This is a dodgy way of achieving the aim of the exercise.

Notwithstanding the "official" Boeing view, and in the light of a paucity of reliable gen on the subject, I tend to the view that an engine overheat indication during the take off run warrants strong consideration of an abort well beyond the 80 knot limit and probably up to 15-20 knots below V1. The purists and pedantic will have other views no doubt.

My reasoning? Worst case is strong possibility of fire warning following the overheat indication. After all, isn't that why the simulator is certified to cover that probability? Not a good thing to take a fire into the air when a competent pilot should execute an abort safely and sort out the fire on the ground.

Also an overheat indication is either a false warning or something is very wrong with the engine. It is unwise to gamble on a false warning. If something is very wrong with an engine then at 15 knots below V1 an abort is probably a safe bet.

No situation will be exactly the same but in the simulator an instructor has the responsibility of giving the best advice he can offer commensurate with his own experience on type.

Strict adherence to Boeing recommendations will earn the pilot kudos in the simulator, but it is these grey areas of "should I - or shouldn't I" abort with a overheat indication that concerns some instructors in their efforts to give the most reliable advice.

Comments appreciated.

keithl
13th Oct 2004, 14:43
Two comments from non-Boeing Sim Instructor:
1. V1 allows for "shock horror then reject action", so you don't need a "safe limit below".
2. Sim instructors should not "ask pilots to play the game by ignoring...". If they do, everyone gets into a game of "What does he want me to do?", rather than "What would I do?"

Right Way Up
13th Oct 2004, 14:59
<<V1 allows for "shock horror then reject action", so you don't need a "safe limit below>>

I have probably misunderstood you, but V1 is the speed at which the reject actions must be started by, so in effect the failure must occur at a speed lower than V1 for a safe reject to be carried out.

GlueBall
13th Oct 2004, 15:13
Practical reality in the actual airplane is that neither pilot's (F/E's) eyeballs may be focused long enough to detect a slow, increasing EGT/TGT creep 15 or 10kts prior to V1.

After takeoff-power/thrust has been set prior to 80 kts, the pilots' scans primarily will be "outside," maintaining centerline with rapidly increasing ground speed/momentum with frequent airspeed scans "inside" and only a general overall "glance" at engine instruments.

If an EGT/TGT warning light came on that close to V1 I would throttle back on that engine rather than abort the take-off roll. In the real airplane my hand comes off the throttles usually 10 kts before V1, and I instruct my crew that this is a "go" decision no matter what happens.

With a heavy jet an abort at or near V1 can be a very violent maneuver, especially if the runway is short and wet, or at a high altitude airport. History has proven that a "go" decision is always better than a violent high speed abort.

Keeping in mind that practical reality differs from simulator reality. :ooh:

keithl
13th Oct 2004, 15:21
RWU - Sure, Hudson's words were:to have the crew initiate an abort at the fire bell providing the speed is an appropriate safe figure below V1. 15 knots, maybe
I don't think you need to be 15kts below V1 to initiate the abort. Otherwise what does V1 mean?

I have myself (real life) initiated an abort 2kts below V1 at max t/o weight, and the figures worked fine.

Spearing Britney
13th Oct 2004, 15:21
Right Way Up, I can't find the figures but in JAR/CAA there isn't a requirement for actions to be started by V1. There is, as keithl says, an allowance for the decision making and lag in taking action calculated for a failure that occurs immediately before V1 i.e. one for which the start of the abort actions would neccessarily be after V1.

As many briefings go 'for a failure at or after V1 we will continue the takeoff..."

Glueball, do you take your hands off at your arbitrary V1-10 on a sim check?:sad:

Right Way Up
13th Oct 2004, 18:55
Spearing Britney
In JAR 25, performance is based on engine failing at Vef speed with a lag time accounted for reject actions to be started by V1. Initiating reject actions at a speed greater than V1 will greatly enhance the chances of a runway excursion.

KeithL - sorry I did not read the bit about 15kts below V1 for a fire bell, I read the bit about 15 kts for an overheat. Just out of interest on your abort did the warning occur 2 kts below V1 or was that when you started your actions?

lomapaseo
13th Oct 2004, 18:57
In an attempt to clear away the subjectivity of what a sim instructor desires, here is some data based recommendations coordinated thru many responsible parties.

http://fromtheflightdeck.com/Stories/turbofan/

Spearing Britney
13th Oct 2004, 19:56
I stand to correct myself, as of 1998 I am wrong! Quoting the Flight Safety Foundations October 1998 Digest....

"Recent revisions of the US FAR and European JAR redefine V1 as the maximum airspeed at which a flight crew must take the first action to safely reject a takeof..."

Wish they would tell me when they change the fundamentals:p

oldebloke
13th Oct 2004, 20:08
As of 1978 Boeing and the industry started to include VEF in their performance info' and references in the V speeds.
VEF is when the engine is failed prior to V1 on the tests(constructor selects this speed).(sometimes as high as 3 sec's,to SEE fail'on TRIJETS)
V1 is the reaction Action limit(stop or go)after the fire/failure HAS been recognised after VEF...The first action.or no action,must be DONE by V! to stop in the Runway distance..
The Airbus FCOM stipulates the perf complies with FAA25..
FAA 25-107 stipulates that the pilot must have RECOGNIZED the failure and ACTED by V!
Some pilots 'feel' the 2 second time stipulated in 25-109 means that they can DECiDE at V!...The 2 second limit is to the DISTANCE for the ACC/STOP distance.....Decisions are BELOW V!.
Cheers..:ok:

VEF is genarally about 1 second(5knots)below V1.....:uhoh:

Right Way Up
14th Oct 2004, 08:46
Looking back at the original post, I cannot see why the instructor is having to ask you to play the game. On the 737 the manual says above 80 kts not to abort for anything other than engine fire, failure or adnormality that makes the aircraft non-flyable. The airbus is very similar and in fact an engine overheat will be inhibited above 100 kts. Therefore if you see an engine overheat in the sim above 80 kts, you ignore it as per SOP but you would be ready for a worsening condition. As far as I can see aborting at a speed close to V1 for an engine overheat will leave you wide open for blame should anything go wrong.

80/20
14th Oct 2004, 09:04
Hudson: Engine fire warnings in the simulator are usually preceded by an engine overheat indication.
All the 737 sims I have used, made by CAE and Flight Safety, did not work like this. The malfuction descriptions were:
Engine fire: Engine fire handle/bell and engine overheat light will come on at the same time.
Engine overheat: (simulates a bleed duct leak upstream of the bleed control valve) Engine overheat light will come on, retarding the thrust lever (below 75%) will extinguish the overheat light.
I thought this would be the same for all but obviously not. What sims are you referring to Hudson? (Not doubting your facts just curious.)

The background for Both Boeing and Airbus’ go-minded philosophy came after a RTO study by ATA and AIA in 1989-90. Part of their conclusion was that:
Only slightly more than one-fourth of RTO accidents and incidents actually involved any loss of engine thrust.
They studied 46 RTO overrun accidents for the western transport jets which caused more than 400 fatalities. Additional 28 serious incidents would likely have been accidents if the overrun areas had been less forgiving.
Approximately 80 percent of the overruns were avoidable.
The conclusion was that in most cases the airplane was capable of continuing the takeoff and either land at the departure airport or divert to an alternate. In other words, the decision to reject the takeoffs appears to have been “improper” in many accidents.

Most pilots know how difficult it is to make good decisions during a real life high-speed malfunction. The best go no-go decision are based on good underlying procedures, always ready for the unexpected and common sense. Common sense can not be given as set rules by manufacturers. Boeing gives the following statements in their FCOMs:
This manual is written under the assumption that the user has had previous multi–engine jet aircraft experience and is familiar with basic jet airplane systems and basic pilot techniques common to airplanes of this type. Therefore, the FCOM does not contain basic flight information that is considered prerequisite training. It is not possible to develop checklists for all conceivable situations…. the captain must assess the situation and use sound judgment to determine the safest course of action.

One example of this common sense and judgments is to consider your stopping margin during preflight. Typically we will use a balanced field V1 - that is in theory that engine-out-go distance = engine-out-stop distance

But the actual runway available is usually longer than the minimum balanced field length required. I like to discuss this with myself and my crew. E.g. "today we have more than 1000 mtr stopping margin, which is good"… or .."today we have only 100 mtr stopping margin and must be go-minded"


Given the chance to consider, would I RTO for engine overheat light & master caution at 100 kts?
With 1000 mtr stopping margin – yes
At exactly balanced field with no stopping margin – no.

The Puzzler
14th Oct 2004, 09:45
Puzzle me this....

Glueball, I dont know if you fly the 737, but if you get a high EGT warning on take off or any other engine exceedance, and the decision is to continue, Boeing suggests not to ease back the thrust on that engine as it will invalidate take off performance. At a minimum safe altitude and airspeed (usually 400AGL min) it is then up to the crew to take action as they deem appropriate e.g. reducing thrust on the affected engine.

Interestingly, an Engine Overheat warning is a recall drill, requiring closing of the affected thrust lever. This is not a procedure one would follow on the take off run. We either take off or reject. Consequently, as it is only a Master Caution alert, above 80 knots I'm continuing the take off.

And finally V1 is the speed that we must have taken the first action to reject the take off. It is also the minimum speed, following a failure of the critical engine, at which the pilot can continue and meet the minimum screen height within the take off distance. ;)

keithl
14th Oct 2004, 10:40
RWU To answer your question, (quite a lot of intervening posts) I don't know exactly when the warning occurred (I was looking outside, of course). "Abort!" was called (it was before "Reject" became fashionable), and when I immediately looked in, we were V1-2kts. I started the abort actions at that moment. Of course the speed continued to increase for a moment, but that's taken into account, too.

Right Way Up
14th Oct 2004, 10:46
KeithL,
Interesting it brings up the problem of the now fashionable procedure of only the Capt calling stop. In my present company and my last the f/o if PNF would call the failure and I would decide to Stop or Go. In your case the fact that the other pilot could make the decision probably saved a couple of seconds.

keithl
14th Oct 2004, 11:28
RWU - Yes, it certainly did. I approved of that system, and certainly trusted my colleagues to make good decisions - which is the necessary requirement for that system to work.

It might also be relevant to add that we would always(pre-start) discuss my intentions for the "grey area" take off decisions, such as various overheat indications. I don't want to get type-specific (because the original question was re: Boeing) but for one particular failure indication, I had a different policy from most other people, so I made sure that if it occurred before V1, I wanted an "Abort" call.