View Full Version : Min Flap Retraction Versus Acceleration Height

18th Aug 2004, 18:09
What procedure do you follow for airports where the MFRA is higher than your standard. Do you just use this value for the engine out acceleration or do you also use it as the all engine acceleration height?


Capt Chambo
18th Aug 2004, 19:48
At any airport that I have operated into that has this potential problem it is normally covered by an explanatory note, along the lines ".......min flap retraction is 1860' AAL.....".

We were also told that when we saw these non standard MFRA's the RTOW had been adjusted/decreased accordingly to make the second segment climb.

It was also useful to be wary of the MFRA in the event of a missed approach if the runway was also a landing runway. (Again assuming the normal climb gradient in the event of a GA)

19th Aug 2004, 05:57
Capt Chambo,

If the MFRA is increased, do you also use that value for the ALL ENGINE DEPARTURE, If so, WHY?



19th Aug 2004, 14:02
At least one operator disregards the published (runway analysis obstacle clearance figure) one engine inoperative acceleration altitude if, on arrival at the level off altitude, the crew are still carrying out any Recall Items of a non-normal check-list.

The directive is that all Recall Items will be completed and only then the pilot enters the third segment regardless of the third segment altitude. This immediately negates the obstacle clearance requirements. I would have thought it should be well within the capabilities of a competent crew to both fly the aircraft to obstacle clearance parameters and at the same time carry out vital actions such as Recall Items.

19th Aug 2004, 14:24

In our performance manual, if the FRA is >1000' AGL, then the MFRA is the FRA for both ENG OUT and standard ALL ENG cases.


My guess (and its only that) is that at a field where the standard ENG OUT FRA @ 800' doesn't have you clear of an obstacle, the performance gurus iterate the calculation until they get a profile that does.

Obviously flying this profile ENG OUT will be the more limiting case, so its simple for them to say the new non-standard FRA applies in both cases.

Presumably, if having calculated the obstacle clearance profile, based on the reduced gradients and you still can't clear the obstacles, you're looking at an OCP.

What I'm not clear on is that normally the T/O profile ends at 1500'. In the case where this doesn't have you clear of an obstacle, segment 2 of the climb is extended until you find a FRA which does have you clear of this obstacle, but how far out do you extend your protected trapezoid in this case?


19th Aug 2004, 16:20

If this is the case, where are they getting the protection for the 5/10 minute takeoff thrust limitation? If they cant be trusted to conducted a checklist and fly, how can they be trusted to watch a clock?


You are correct in your assumption that if the FRA of 800ft doesnt clear an obstacle the takeoff program will reiterate the calculation until it finds an altitude that will clear the obstacle or it gives you an error!

We had worked on the principle that the ALL ENGINE climb performance is significantly greater than the ENG OUT performance, therefore we had an ENG OUT FRA which was adjusted according to the runway requirements and a standard ALL ENGINE FRA of 1000 ft. Powers that be want to make both of them the same, I'm just investigating the PROS & CONS.

You will extend your projected trapezoid until limited by your engine thrust time limit or as limited by the AFM.


Empty Cruise
20th Aug 2004, 14:38
Hi Mutt (et al...)

Look at it this way: Min. FRH (2Eng) 600 ft. AGL, min. FRH (1Eng) 1400 ft. AGL. You arrive at 600 ft. AGL, start acceleating and are retracted to say Flap 2 at 1000 ft. AGL.

At this point, you suffer an engine failure. Now, the question is - is you prevous advantage of 2 engines operating (offset to some part by acceleration) from 600 ft. AGL up to 1000 ft. AGL enough to save your proverbial from hitting anything from 1000 ft. to 1400 ft. AGL?

The problem is that from you start accelerating, the engine might fail at any point. My guess would be that at first, you have not accelerated very much and are still in the proper config, so bleeding the speed back to V2+15 is not a big problem. However, the further you accelerate, the deeper into no-no country you get. however, at one point, the picture starts reversing, since you are now closing on your 1Eng FRH - receeding from no-no country again.

So in order to make sure we never go too far up :yuk: creek, the FRA(H) is the same for 2 and 1 engine (in many companies). It also leaves you with one less variable to think about, keeps your briefing down to bugging only one altitude, avoiding possible confusion and making the task simpler.

Regarding the horizontal extend of the obstacle clearing trapezoid - it can be longer than 5/10 min. worth, only from that point, it will be calculated with MCT. Once got a GWC handed out that took into consideration an obstacle some 44.800 m. away from DER. Only this would be in the 5th climb segment. Some GWC suppliers extend their terrain coverage to 25 NM from DER, no more. But stuff like that should be in the performance manual (hopefully).

Just me 2 eurocents worth - have a safe one!;)


20th Aug 2004, 16:05
at least one operator disregards the published (runway analysis obstacle clearance figure) one engine inoperative acceleration altitude if, on arrival at the level off altitude, the crew are still carrying out any Recall Items of a non-normal check-list. etc etc

sound quite sensible, fly the aircraft first.. the whole point of min FRA is the word min. as in not below. So if they dont start accelerating till 1300 feet instead of 1000 then its hardly the end of the world. It should not compromise their terrain clearance as they will have an improved overall gradient during that extra couple of hundred feet and its far better to do things in the taught sequence than faf around trying to level and accelerate while running checks.

Old Smokey
20th Aug 2004, 16:35
Engine out acceleration height and all engine acceleration height the same, both increased above the standard company 1000 ft to MFRA if higher.

The rationale is to ensure 3rd segment obstacle clearance should an engine fail between the standard 1000 ft and MFRA.

Empty Cruise
20th Aug 2004, 17:19
Btw, our GWC also states the maximum acceleration altitude. Gen. in the order of 1000-1500 ft. above the minimum altitude. Should leave plenty of space for flying the aircraft & doing recall items :cool:


20th Aug 2004, 23:57
Oddly enough, Mutt, this was discussed at least twice before...in 1984...and 1987, an no changes were made.

Third time luckey perhaps?

Captain Ahmed Sais, the TriStar fleet manager was a very smart guy, but was outvoted.

For crews...keep it simple.....and safe! :ok:

21st Aug 2004, 15:28
Daysleeper. If obstacle clearance requirements dictate a certain third segment level altitude then this should normally also take into account the 5 or 10 minute engine power limitation. See Mutt's reply to my post.

To climb to a higher third segment merely because someone in management decides that Recalls should be completed before third segment, defeats the certification purpose of the engine limitations. Thus obstacle clearance is compromised. The performance engineers responsible for designing the runway analysis chart with its various caveats on third segment level altitudes where dictated by obstacle clearance rules - are entitled to assume that the pilots are competent to handle Recalls as well as play by the rules.

If clawing your way to a higher third segment than laid down by the performance engineers simply because it seems a good idea at the time, then you may as well say throw the engine limitations out of the window and gain an unfair weight advantage on your competitors who play fair.

Empty Cruise. Bleeding back to V2+15 after an engine failure which has happened in excess of V2+15, does not comply with one Boeing type which requires that the speed at the time of engine failure should be held unless it happened above V2+20 - in which case speed should be reduced back to V2+20.
See FCT 737 (TM) Page 2.44 date October 31, 2002.

21st Aug 2004, 16:39
Its best not to rush these things, 20 years for a policy change seems about right :):)

Old Smokey,
How do you justify the weight loss associated with a FRA of 1000 ft? Rather than the minimum of 400 ft?

Empty Cruise,
Accounting for an obstacle 44kms away, wasnt that the purpose of designing Obstacle Clearance Procedures?


21st Aug 2004, 20:42
just to check we are not talking at cross purposes here.

My point is that it is safer to do one thing at a time. Which recall item dont you run in order to call for the level segment. From a human factors point of view interrupting a recall checklist is asking to forget an item.
I know we are not talking aircraft specifics but as an example of what I mean on the 757 at the weights I operate the performance is such that, by the time you have run the recall items including the 30 second wait between fire bottles if reqd, you could be well through min flap retract. The engine limit does not really come into it as it has taken the same time you just reach a higher altitude light compared to heavy.

Old Smokey
22nd Aug 2004, 01:10

Re FRA of 1000 ft - I can't justify it, company policy is rigidly at 1000 ft with Regulatory Authority insistance. The penalties not too bad considering that 98% of takeoffs are at reduced thrust and end result is usually only a requirement to flex to a slightly lower temperature.

22nd Aug 2004, 03:42
At the risk of repeating some of what has been listed previously, it might be useful to summarise the continued takeoff approaches adopted by the typical operations engineer .. and I have no doubt that Mutt and Old Smokey will take me to task for any errors and omissions I might make ....

(a) work out the maximum runway limited TOW.

(b) starting with this figure, we iterate (if, and as necessary) until we come up with the best weight (being not greater than the runway limited weight) for the obstacle requirements

(c) the takeoff flight path commences at end limiting TODR (keeping in mind that the declared runway data for obstacles will be referred to end TODA - which can create problems so far as do we redo the obstacle trapezoid or do we somehow ensure that we pass through the centreline/end TODA gate ?)

(d) normally we run the minimum third segment (400 ft referred to the runway head or such higher height as might be SOP for a particular jurisdiction or company) and see what the clearances are.

(e) if the clearances are OK, no problem

(f) if the clearances are inadequate, reduce the weight and start the exercise again - including the runway cases to account for a reduced TODR. If the critical obstacle is distant, it might be useful to look at running a higher third segment as that might produce a lesser necessary weight reduction.

(g) one of the problems to be addressed is where is the critical obstacle located (and noting that the critical obstacle may change as the weight is varied)

(h) if the critical obstacle is located in the first or early second segment, there is not much to be done other than pulling back the weight to reduce TODR and improve climb gradients

(i) if the critical obstacle is late second segment (especially), late third segment, or fourth segment, use of overspeed V2 or improved climb (depending on which manufacturer's camp you are in) may help out by running a higher V1 to reduce TODR and higher V2 to improve second segment climb gradient

(j) if the critical obstacle is third or fourth segment it may be useful to run the second segment higher so that the net flight path is forced above the critical obstacle. Main limitation here will be dictated by the need to get to the end of the fourth segment within the engine takeoff rating time limit. Sometimes there will be other limits to be observed .. eg as I recall the Dart had a feather pump time limit which constrained the third segment (at least on the F27 and HS748) to a maximum of 600 ft.

(k) provided that the AEO flight path is constrained always to be above the OEI gross flight path, then one can sensibly ignore the AEO case and post V1 failure as the aircraft will be in a conservative position compared to the V1 OEI case.

(l) one matter which has always concerned me is the case where (some) operators address the V1 failure case but then ignore post V1 failures in the case where the OEI escape route differs from the AEO departure. It is exceedingly facile, not to mention potentially dangerous, in this case to leave it to the pilot to sort out - to me, it is clearly the operations engineers' responsibility to address this case so that, for all failure cases, the calculated flight path to the en-route phase of flight is addressed with respect to obstacle clearance requirements.

(m) keep in mind, also, that the gross to net fudge factor provides an increasing vertical margin between gross and net flight paths. .. which is why we are far more actively interested in first and early second segment obstacles than, say, fourth segment.

(n) in respect of Mutt's original question, a similar reasoning to (l) and (m) ought to apply ... if the AEO flight path is not constrained to be above the OEI escape path, then the individual runway analysis needs to take care of the AEO differences .. ie, if the published procedure doesn't refer to using a lower AEO acceleration, then I would err on the side of being a bit suspicious .. in the same way I would any time that the AEO and OEI flight paths differ.

22nd Aug 2004, 15:15

A very interesting question which will always polarise opinions.

Most operators with whom I am familiar match the two altitudes for simplicity. There are all sorts of "after-market" performance reasons offered for using a single FRA, the vast majority of which are specious - the only sensible reason is simplicity.

Depending upon the aeroplane and the departure path, combining the two altitudes may not only be inefficient but also counter-productive in terms of mindsets and flightpath management.

The reason we have special departure/emergency escape/OEI procedures is to reconfigure the aeroplane as quickly as possible to the clean enroute climb configuration within the engine limitations without hitting the obstacles. We plan for the worst case, ie an engine failure at Vef at the maximum weight that satisfies all of the rules. That, by definition, will create the lowest safe flight path to be flown. Therefore, every second that elapses after Vef without an engine failure is money in the bank because, despite what Empty Cruise posited, the OEI profile is based on minimum obstacle clearance and a level acceleration and thus any AEO profile will normally be well above the OEI profile. The only way that the OEI profile could be compromised is if the pilot chose to accelerate at a lower gradient than the OEI second segment, which is not only stupid but illegal in some jurisdictions (eg Oz, which I thought JT would mention).

So now we need to look at simplicity versus efficiency.

The best flap configuration to be in at the time of failure is clean, followed closely by the lowest flap setting authorised for take-off. The AEO FRA is determined largely by legal minimums plus the time and obstacle clearance factors that allow for human error at a critical phase of flight. Having achieved AEO FRA without a failure, accelerating at a gradient not below the OEI gradient to then achieve a better flap configuration is not only safe but also offers the prospect of even better climb/acceleration performance in the event of a subsequent failure. That is potentially even more "money in the bank", easily verified by comparing climb gradient for each flap setting at the same weight and recognising that the potential climb performance will in most cases improve significantly as the speed increases beyond the relevant V2.

I accept the simplicity argument but I advocate the efficiency argument.

Now that I am old and grumpy, I really hate sitting through an AEO take-off at the worst possible flap setting that staggers up to a very high OEI FRA without configuration change "in the interest of simplicity" when I could have an even greater margin of safety by taking advantage of my AEO performance to optimise my configuration prior to any failure that may happen along.

For all of that Mutt, operational or organisational or national culture may end up being the safety driver for the policy decision, rather than efficiency. Best of luck....

You raised some later issues after I drafted my initial response above.

Raising the FRA above the 400' minimum does not always reduce payload - I presume you are always spoilt for runway length! Nonetheless, raising the minimum FRA is a safety-based commercial decision in that it buys time for the crews to get organised before accelerating either AEO or OEI. In many cases, there is a range of acceleration altitudes that will work within the engine limits, particularly for flex take-offs using thrusts settings below MCT. However, we don't tell the crews because there are too many variables and we want them to respect the published FRA as if there was little or no latitude for the acceleration.

Hudson's reference to at least one operator delaying the third segment level-off was, I suspect, highly context specific to very few destinations and related only to fire situations. The policy did not respect the FRA but did respect the OEM advice to carry out the initial fire drills in the second segment. My counter view was to delay the fire drill until starting the third segment - another great argument to be had over a few reds.

JT, in your (f) and (j) you mention elevating the FRA to solve the problem. I note that some Jepp models consider that the only option for dealing with distant obstacles. Because we plan all the way the MSA/LSALT to satisfy the Oz rules, we quite often look at a lower FRA to achieve a safe 4th segment climb before getting to the obstacle(s). For our overseas friends, the traditional Oz rule does not truncate the splay either laterally or longitudinally on the basis of a DR climb in IMC and turning can often not alleviate the problem. They have introduced an ICAO laterally truncated splay as an option but survey baselines are a major limitation for most jet aerodromes outside the major cities.

Stay Alive,

24th Aug 2004, 05:44
Plan to print out your instructions and pin them to the notice board, might actually remind people that there is more to this than just entering numbers into a program! :)

Glad to see that you agree with the method based on efficiency, our justification for using this was that you would always achieve a higher gradient with AEO. Unfortunately this was great until we realized that when our non-native speakers study procedures, they memorize without understanding! There is a Boeing chart that shows the different stages of takeoff, the acceleration portion is shown as a level acceleration, around here that has come to mean "Altitude Hold", so whatever advantage was achieved by the ALL ENG climb is immediately negated following an engine failure!
We have one airport in particular where the FRA is 1600ft and the standard ALL ENG acceleration height is 1000ft. We are growing increasingly concerned that someone is going to reach 1000ft, have an engine failure and immediately select Altitude Hold!
So as 411A has stated its time for the KISS system, make both heights the same. While at the same time reiterate in training that its not meant to be a level acceleration.


24th Aug 2004, 12:53
4 Dogs .. I am probably a little more conservative than some. Starting off with a performance background prior to flying heavies, I knew (at that time) that I knew a lot more than I came to realise I knew, if you know what I mean ...... (hopefully this is me talking and not the exquisite red I had for dinner just now).

Looking to your comments ..

(a) one always should structure SOPs so that the AEO case keeps the aircraft above the presumed OEI flightpath - if the two tracks are the same. It is where the OEI heads off down the valley and the AEO keeps going towards the hills that I get worried ...

(b) so far as pushing the second segment up goes, it is largely a matter of preference and standardisation - depending on how hungry the operator is for those last few kilos. I, like you I suspect, don't like the oft seen approach of looking at the V1 failure and leaving the rest to the pilot. In my conservative, and admittedly un-heroic approach to dying - for both me and my buddies, I would far rather see the bumpy bits looked at in detail until the aircraft net path is above the obstacles in the vicinity .. some of the approaches I have seen to the departure analysis have given me considerable food for reserved thought.

Long time now since I looked over the way Jepp does the work .. I was impressed with some aspects, not so much with others. But the attitude seemed fine talking with their specialists at that time.

Mutt ... big problem with pilots, as you would know, is that we all necessarily require a high self confidence in our abilities, knowledge, etc., etc. (otherwise it is forever in the RHS) ... sometimes this creates its own difficulties in getting enough (training) knowledge across in the time available in order to avoid having that very self confidence create its own set of problems ... this is not a criticism .. merely an observation of the animal.

Suggest you tidy my ramblings up a tad before posting to the noticeboard ...

Empty Cruise
24th Aug 2004, 19:34
Hi all,
Hudson - off course, V2+20. I stand corrected (too fast on keyboard, too slow in thought :} )

mutt - exactly. But for the aircraft concerned, this was well outside the 10 min. RTO thrust limitation. So your GWC should take into consideration also 5th segment climb, the first 5/10 min. with RTO, then at MCT. Maybe I misunderstood your question as to the 5/10 min. limitation being applied - apologies if waste of bandwidth :{


25th Aug 2004, 11:33

Hence my comment about the cultural element of the policy making.

However, I would have thought that having identified the potential problem, there were a number of solutions:

first, emphasising that acceleration must not be at a rate of climb less than 5-700fpm (or as determined by the required departure gradient) when all engines are operating (draw a matching diagram) or acceleration must not be at a rate of climb more than 0-200fpm if one engine has stopped and acceleration altitude has been achieved (Boeing diagram);

second, structure all of the sim rides to have split AEO/OEI FRAs and drive in the correct solution; and

third, ensure that all Line Checks examine that policy.


Let me remove any doubt - in my view, any manager who directs that only the V1 cut scenario should be planned and everything else is the sole responsibility of the pilot is abrogating his/her duty of care and should be a prime candidate for industrial manslaughter if a death occurs through such professional neglect.

I also believe that employers and employees share an equal burden to keep the grass out of the windscreen wipers - any decision to depart from the OEI track needs to include a minimum MOC to start the turn that takes into account the climb performance at the time of a failure at any point of the turn and the ability to navigate into safe airspace post-failure (allowing for radius of turn!).

However, I also respect the need to maximise payload because that is the business we are in. Hence my previously expressed frustration at those who would seek the path of least resistance and climb to 25 mile MSA at the take-off flap setting before commencing any turn. The true beauty is in achieving a safe balance.

Stay Alive,

25th Aug 2004, 11:54
Quick questions....

1) Is what people are referring to as fifth segment climb requirements analogous to the en-route climb requirements a la JAR-OPS 1.500/JAR 25.123 ?

2) In the event that an obstacle obliges that the requirements of JAR 25.121 (Climb OEI) be extended until >1500 AFE, is the obligation to meet the fourth segment requirements dispensed with? In this case, one resumes the calculation at the end of the third segment within the constraints of JAR 25.123 (Enroute Flight Paths)?