Asymmetric Commital Height.

[Keygrip]

Please - for the purposes of this discussion - no comments on "don't do asymmetric go-arounds".

Assume you are teaching student pilot to fly in a light piston twin - what do you deem to be a suitable asymmetric commital altitude, and does it change with time, experience and rating they hold or are training for??

I've heard figures of 500' agl for new pilots on the class (6hrs training).

I've heard 300' for good daylight, good visibility - with an extra 100' for night or IMC/low vis. conditions.

I've also heard 200' agl (for the new student) "because that will lead them into the instrument rating decision altitude".

What do you teach - and how do you justify the figure? Just curious.

(Wonder how many self confessed experts will try to steer away from the original question, or insist that their opinion is the only correct one).

[BEagle]

My only asymmetric flight instruction has been on heavy jets. On the 4-engined VC10 we used to teach both 3-engined and 2-engined go-arounds, of which only the 2-engined had any engine-out allowance. We would add not less than 200 ft to any 2-e precision approach DH/DA, but had an absolute 350 ft a.g.l. 'Visual Commital Height' below which no go-around would be flown. At VCH, the pilot was required to make a very positive decision; if 'LAND' was called it would be followed by flap selection direction, rudder trim orders and some precise flying. It was practised repeatedly in the simulator and we used to do it for real with new captains on their conversion course under strict weather conditions. Only training captains were otherwise permitted to practise unsupervised double asymmetric flying in the air.

Does the concept of engine-out allowance apply in the MEP world?

Personally I feel that asymmetric training is one of the key flying training areas where certainly the RAF has pretty firm and very sensible rules in place concerning mandatory go-around heights. And that's for folk who have all had extensive aptitude testing....

[A and C]

This should be a personal thing for the student , take them up to a safe altitude and from the normal Vref at a normal approach ROD with the gear down and the flaps set for the normal approach setting for the aircraft get the student to "go-around" and observe the loss of altitude before a positive rate of climb is established.

Add the loss of altitude to a safety factor of lets say 200ft and you have a basic number to start with. more demanding airfields may require an additional safety factor.

The altitude that you select for this demo should reflect the aircraft performance at MLW WAT limit.

This will assure that the commital height is within the ability of the student and as more practice is undertaken the exercise can be repeated to lower the limit as the students performance improves.

[johnnypick]

I've always taught 200'agl, as that was the height the examiner for the MEP told me me he was going ask for a go around SE. I've always found that a comfortable height, and the students have never complained.

[BlueLine]

"Assume you are teaching student pilot to fly in a light piston twin - what do you deem to be a suitable asymmetric commital altitude"

As A and C says it will depend on the student. ACH is designed to allow the aircraft to sink whilst going through the go-arround drills. Initially the student will be slow, may make mistakes and will inevitably loose some height.

As the student progresses the drill should be performed in a way that minimises the height loss. In any event the exercise is flown visually at the rating training stage.

"and does it change with time, experience and rating they hold or are training for??"

Yes

When the student reaches the IR skill test level he will be required to demonstrate an approach to minimums on instruments and demonstrate an asymmetric go arround from that height; so ACH becomes the same as the minima used.

Would you go arround from 300, 400, or 500 ft if you have an engine out and a 250ft cloudbase?

[Keygrip]

Would you go arround from 300, 400, or 500 ft if you have an engine out and a 250ft cloudbase?
By the same token - if the w/x was CAVOK, wind calm - would a 6hr multi engine student (of which only 3½ hrs are asymmetric) go all the way down to 200 agl?

Should they be taught - at THAT stage (i.e. during initial training ) to go all the way to 200agl before making the decision to go around for whatever reason?

Or, should they be taught, at initial rating stage, to go to 300, 400, 500 - maybe even their own (individual) figure - and then improve with time?

[Tee]

For initial MEP training, our Ops Manual specifies 250' agl. Seemed reasonable to me when I did the training, but I was a "commercial" student and had a CPL.

[BEagle]

The more I hear about asymmetric work in light twins, the stronger my resolve never to go near one of the things.

I find it quite astonishing that there is no defined Engine Out Allowance or Visual Committal Height published for each MEP aeroplane in the POH, being:

EOA: An allowance made to cater for the height loss incurred when carrying out a single engine go-around. This should then be added to the relevant DH/DA, so that a pilot executing a SE go-around does so from a height/altitude which will not involve him descending further below the normal minimum had the go-around been conducted with all engines operating at the normal DH/DA.

VCH: The height below which the aeroplane is committed to land following an asymmetric approach.

These values should be based on the average performance of an average pilot rated on type following the failure of the most adverse engine. Quite why there should be any 'personal' flexibility to invent one's own allowances, I cannot quite understand. Of course initial training should be conducted prudently, but the student should be capable of executing a go-around at a properly defined minimum altitude specific to a/c model.

Bitter experience with asymmetric training on things like Meteors led HMFC to be very, very careful about SE go-arounds. I was once told that the VCH for the RAF's last Meteor 7 at Brawdy in the mid-'70s was 650 ft a.g.l. - and that it was rigidly observed!

[ROB-x38]

Or, should they be taught, at initial rating stage, to go to 300, 400, 500 - maybe even their own (individual) figure - and then improve with time?

Just for interest's sake Keygrip, at the flight school I did my initial twin training with company procedure was you're committed at 400' AAL when OEI. This is a standard figure for anyone doing their initial multi training, however with time an experience will reduce.

[A and C]

I agree with you that an EOA should be published for each aircraft but it is not.

I think that this is because the makers of these aircraft feel that a lot of the "variable factors" are out of there control and the legal implications of publishing such a "hard data" would lead to them being taken to the cleaners by a sharp lawyer.

This being the case the best way to protect your student is to get them to find there own personal ACH.

You should remember that light twins should be treated as single engined aircraft that can continue flight after an engine faiure if flown properly and that in some phases of flight the best course of action is simply to close both throttles and find a field.

In terms of flying skill and airmanship to fly safely a light twin is much more demanding than the B737-800 that I fly for the day job.

[Hudson]

Please accept this as a purely personal viewpoint. The "decision" height depends on several inter-related factors such as airspeed and flap setting at the time of go-around and expected climb performance on single engine clean.

The instructor should demonstrate the most critical scenario which is landing flap down, and Vref speed. Obviously this should be demonstrated at a safe height such as 1500 feet agl.

When the student has demonstrated consistent competency at this manoeuvre, then various other configurations can be shown, such as no flap and say 10 knots above normal Vref. Clearly this allows more latitude in terms of decision height.

The results usually reveal that a competent well trained student can safely execute a single engine go-around at sea level from 2-300 feet with full flap and Vref at point of go-around, and if flapless and Vref plus 10 knots then 100 feet. These are realistic figures.

Whether the pilot has low hours or not, he must demonstrate a high level of competency before being signed out. There should be no question of adding altitude for Mum and the kids just because of a low level of experience. He is either totally competent to execute a single engine GA or he is not. If not, then more dual until he reaches a safe standard.

It is also important to brief the student what actions he can take to minimise the chances of a bad pile up if he is forced to place the aircraft on to the ground in event of being below safe GA height. This vital point is often inadequately covered.

[Keygrip]

Interesting comments - all quite correct in their own way.

x38 seems to be the nearest to what I've been taught through the years - but that may not be the correct (or modern) answer, hence my question.

BEagle was, not surprisingly, spot on with what should happen in the ideal World - but A and C is probably correct in terms of "product liability".

BlueLine was perfectly correct in the IFR go around scenario - if the DA for an approach is 200' agl then, by that time, the pilot needs to be able to do the go around at that level - but by the time the pilot is carrying out asymmetric missed approaches in IFR then I would hope they have had more training than four to six hours on a light piston twin.

What I was saying about "personal minimums" was because there isn't a published figure in the POH for the average pilot (indeed, who could work out an "average" - and what factors would the below average pilot apply to that figure [assuming that they would even class themselves as 'below average']).

It was the discovery of the fact that certain schools want the commital height to be mandated as 200' agl for a brand new student on the aircraft - with only four or five hours on the class and type of aircraft when this part of the course comes up.

Would we expect the 4 hour multi engine piston student to carry out a go around as neatly and safely as a multi engine instructor with, perhaps, a few thousand hours on light piston twins??

This is where I wonder about the idea of a higher ACH until the individual has gained some experience - both of themselves as a pilot and of the performance capabilities of the aircraft (neither of which I think they have the brain power for when first confronted with the multi engine conversion course).

Would 500' agl be too high for the six hour muppet? Whilst the 200' is the ultimate aim for every individual (for all the reasons above) - should we teach it as a mandatory figure from approach number one??

Curious at BEagles comment about never going near a light twin - if you are prepared to fly single engine aircraft then there is no difference - for the reasons stated by A and C (if it's not working nicely as a single engine twin - then make it into a glider).

[BEagle]

OK - I'll exclude the Wing Derringer and Diamond TwinStar from my 'avoid' list!

In something like a typical commercial FTO operated Seneca, flying an approach at the correct POH approach speed and configuration is there:

1. Any modification in IAS or configuration applied when asymmetric?

2. Any additional height loss in the event of an asymmetric go-around when compared to a go-around with both engines operating?

If the answer to 2. is "Yes", then whatever that value is should be added to the normal DA.

[DFC]

The ACH depends on aircraft type, WAT and pilot experience.

For the brand new MEP rated pilot in general we teach an ACH of 400ft. At the subsequent renewal 1 year later, again depending on type and experience there may be a reduction in this figure but the lowest used is 200ft.

However, we teach that an engine failure in a twin is an emergency and a mayday call must be made. This can help to reduce the chances of making a missed approach.

For instrument pilots (including IMC rated pilots who fly MEP) the same ACH applies.

Our and the CAA's policy is that if possible an engine out approach in IMC should be avoided. If that is not possible then the minima to be used are the higher of the procedure minima or the ACH. In other words we do not change the ACH just because the pilot is instrument rated it depends on their experience on class and type.

BEagle.....basically don't reduce below blue line or extend landing flap until a landing is assured (or comitted) is the only requirement......thus having an ACH of 400ft not only ensures that a missed approach can be made safely but it also gives time and height for the inexperienced pilot who is landing to extend the landing stage of flap and get things set up without being in a rush which they could end up in if 200ft was used.

Regards,

DFC

[hugh flung_dung]

The answer, as usual, is: "it depends".
After Asymm2 the stude should be convinced that it will take him/her 100-200ft to sort out an engine failure and get the beast climbing again - and this was when they were expecting a failure because it was the asymm part of a ME course! Ask them how they think they would cope in 11 months and 25 days on a hot day at MLW! I was taught to teach a Vach of 300 ft and this seems to work well for the average stude in a Seneca 1 (with an addition of 100ft or so if an asymm approach is unavoidable in IMC) but in a Duchess (e.g.) 300ft would probably be excessive (but safe). As always: know your horse and stay in practice.

On a slightly different tack...
I recently had a candidate for his first MEPL renewal who (he said) had not been given the full asymm long brief during his training or shown the advantage of 5 deg bank towards the live - I won't mention the airfield for fear of libel suits :( Please peeps, give your studes the long asymm brief and teach them about the bank attitude for zero side-slip.

And on an even different tack...
I tend to end an MEPL course by failing an engine below Vach - it's surprising how many people try to go-around despite having said LAND, lowered full flap and started slowing down as they go through 300ft. It leads to some interesting aircraft attitudes before self preservation leads to an "I have control".
RAGA before Vach, commit after Vach.

HFD

[chicken6]

BEagle

I used to teach on the proverbial FTO-operated Seneca I which we also used pretty regularly for ATOs VFR (and IFR if we had to but the door leaked, as did the windows and we used to get ice forming inside but that's a different story). The answers to your questions from our organisation were 1. Yes and 2. Yes.

1. The change in configuration when flying visual or IFR approaches 2-eng vs 1-eng was that instead of steadily reducing through Vyse at 300' and making the decision to land at 300' 2-eng, that decision would be made through 350' when 1-eng. The profile was also flown 50' higher all the way in to allow for a bit of height loss to help accelerate to Vyse (the one operating engine would very rarely produce a noticeable acceleration without height loss). The net result was that the speeds were the same at a given distance from touchdown but with 1-eng we flew 50' higher. Power settings were basically standard 18" for 2-eng approaches in the descending part (base and final) but with 1-eng it was whatever you need, count on another 4-5".

2. Extra height loss? Gimme a hell yeah! Standard first-go-around height loss for our students was about 50' 2-eng from their "going around" call, but on their first asymmetric go-around it was more like 150-200' before they got stabilised. And that with two of us and three hours fuel. Once they got a bit more proficient they could normally get it down to 100' but the poor old Seneca just couldn't accelerate without losing height while you get the gear up.

As I recall none of this was in the POH but we had operated them since the eighties so had a bit of experience with the type.

And yes, when we taught ILS approaches later on, we had the DH discussion and got them to tell us that 350' was a committal height whether you could see where you were going or not, so if it's an ILS to minima, you are physically committed below 350' so that is the effective decision height. Below that the pilot has to be sure of making a landing using normal techniques.

[BEagle]

Thanks for all your education, everyone.

It seesm to me then, that, for the Seneca:

1. A 150' EOA should be applied to the 'symmetric' DA/DH.

2. A VCH of 350' should be used; at 350' a.g.l the IAS and configuration should be progressively adjusted to achieve Vref and full flap at the threshold.

Thus, if landing criteria have not been achieved on an asymmetric ILS (with a 'normal' DH of 200 ft), an asymmetric DH of 350 ft must be used and a go-around initiated if landing criteria have not been achieved at 350 ft.

No attempt at an asymmetric go-around should be flown below 350 ft.

[whatunion]

moderator you asked

Assume you are teaching student pilot to fly in a light piston twin - what do you deem to be a suitable asymmetric commital altitude, and does it change with time, experience and rating they hold or are training for??

I assume you are talking about what we used to call perf group c twins which only guarantee climb with the engine feathered and the gear in the bay and generally seem to take pilots to the scene of the accident after a real engine failure!

interesting to see that no one mentioned go around flight path obstacles. a big difference in a ga at lydd then say at mull!

everyone will have a different height i would imagine. surely the most important thing is to have a limit and stick to it.

if i was writing a pilots order book these are some of the things i would consider

my first consideration would be that in an a/c of marginal SE climb performance a ga has to be fully warranted and a last resort. certainly better to be on the ground wishing you were in the air scenario etc.

I would also take the following into consideration.

Problem-- fly the a/c to a safe place, if your on fire a safe place is generally the runway not the GA! Instructors always talk engine failure but engine fire happens too(happened to me on a GA in france on an aztec for real and its very unpleasant!) (in fact thinking about it I have never had an engine failure as such but only engine fires, one exactly at V1 at Jersey, christ that dosnt alf concentrate your mind!)

Wx.
obvious one but also consider nav aids do you really want to to come back for another NDB let down on one engine!

Obstructions
consider what you are climbing into.
dosnt matter how perfect your GA is, a hill or mountain will serious alter the profile of both you and the a/c!

Experience/ability
Who rembers the citation that ended up in the garden on the right of 27 at jersey off a low level IMC GA and that was with both engines running!

who remembers the BMA viscount at manc were the first officer stuck in a bootful of wrong rudder and the a/c went right over on its back during an assymetric training exercise.

A/C

i seem to remember on ceratin aztecs you had to pump the gear up if a certain engine failed. I think I would like to have a few more feet under my backside in that situation for real.

PS One of the most dissapointing features i found in twin instruction is the belief by students and some instructors is that a perf C twin a/c would always climb away on one engine.

the company i worked for even used to close the cowl flaps for t.o. to improve single engine perf.


Personally if I had an engine fire i would be landing off the approach on a perf c a/c not considering a go around unless landing was physically impossible.

[Chuck Ellsworth]

I try to not comment to often on the instructor forum, but occaisonally I am facinated by the different thought patteren of people who teach flying.

So I have a couple of questions that maybe someone can answer for me as I seem to have difficulty with understanding the need for trying to fly low performance twin engine airplanes close to the ground on one engine.

Do any of you do this with the non operating engine actually feathered?

Chuck

[whatunion]

whatunion says, hey chuck this is a forum for bull****, anybody that got that near to the ground on one of these kites with an engine feathered can only be spoken to through a medium!

[hugh flung_dung]

Chuck, you asked Do any of you do this with the non operating engine actually feathered?
During initial training for a MEPL rating we feather an engine (usually only once) so that the stude gets to go through the full feather/unfeather drills but this is done at/above 3000ft and within easy reach of an airfield.
The course is 6 hrs min of which 3.5 hrs min is asymmetric, during this period all failures are simulated (except as above). My practise is to either pull mixture (at/above 3000ft), sneakily cut off fuel (in the cruise, at/above 3000ft) or by hiding the throttles and closing one (below 3000ft) - as soon as drills are correctly performed I set zero thrust (net effect equal to a feathered prop, typically around 10" MAP).

Asymmetric approaches are always flown with the inop engine set to zero thrust so that the engine is available if required. The only exception to this is simulating a failure below Ach when a throttle is closed once the stude has committed, it's more important to reach the runway than try to mess around with feathering drills so the engine/prop are windmilling but power is available if required.

whatunion: slightly melodramatic old chap.
As I'm sure you know there is no problem at all controlling "these kites" with an engine feathered providing the minimum control speed is maintained - whether or not the height you can maintain is sufficient, or whether you can gain height, is another matter and depends on a variety of factors.

HFD

[whatunion]

whatunion: slightly melodramatic old chap.
As I'm sure you know there is no problem at all controlling "these kites" with an engine feathered providing the minimum control speed is maintained - whether or not the height you can maintain is sufficient, or whether you can gain height, is another matter and depends on a variety of factors.

no problem at all controlling these kites with an engine feathered providing the minimum contol speed is maintained!!!!!!!!!
( you forgot to mention that the gear needs to be stowed +, 5 deg bank and below max weight also)

as im sure i know!!!!!!!!!
i am sure i know that a lot of pilots have been killed trying to do just that.
so you, old chap, would say to your student, never worry about having an engine failure on these a/c because there is NO PROBLEM AT ALL in controlling them providing the minimum control speed is maintained! interesting, i hope you have a very good solicitor who can argue, " ah but he flew into the hill in perfect control!"

its all very well to have a light a/c on a good day and know you are going to feather an engine. its a complely different matter when it happens for real at max weight on the worst day.

take the aztec accident at glenrothes were a very experienced air charter pilot took off with 5 engineers on board on a c of a air test, the front cargo hatch opened and he immeadaitely feathered that side engine. the a/c hit the ground and all 5 engineers were killed.(piper say a/c will fly safely with hatch open and it will not touch prop)

then there was the guy at jersey that had just done his base check took off on 27 had an engine failure lost control and was killed just off the the south of the threshold.


you are putting in the mind of the student that the a/c will always climb away and i know you said whether it will climb is another matter. these a/c should be treated as single engine a/c in some situations. its always better to fly to the scene of the accident rather then to arrive at the scene of the accident wing tip first!

[hugh flung_dung]

whatunion:
control depends on adequate speed, performance, however, requires gear stowed, 5 deg, etc. Min control speed itself depends on CG, weight, drag, power used, leg strength, etc.

I agree that there are times when they should be treated as SE acft but this does not justify your melodramatic comment.
The standard light twin departure brief contains something like this (I haven't got time to type the whole thing):
- eng problem on the ground: close both and stop
- eng problem airborne below Vtoss and ach: close both, lower nose, pick landing area, use power on live as required
- eng problem airborne above Vtoss and ach: control, attitude, power, drag, ident, verify, feather, 5 deg, etc
... this recognises that it isn't always possible to "go" and the stude is left in no doubt of this during their training. The concept of approach ach reinforces this - but this is down to performance and not control.

I prefer not to discuss accidents. We're all capable of getting it wrong and, unless you're in the hot seat at the time you don't know what actually happened.

HFD

[Tinstaafl]

Point of accuracy: 5 deg AoB is not about performance. It's a certification limitation related to establishing Vmca. It prevents the manufacturer specifying an unreasonably low Vmca by using an excessive amount of bank.

Best performance is typically at ~2-3 deg AoB.

This has been discussed before on PPRuNe so a search should find the thread.

[hugh flung_dung]

Thanks Tinstaafl, I keep saying that I'll stick the piece of wool on again but never quite get round to it!

HFD

[whatunion]

Vtoss and ach whats that? i like the connection with tossers though!

14 years since i instructed on twins so u need to explain that one to me.
i seemed to remember there was a blue line and a red line on the asi so shouldnt your brief relate to these, as thats whats in front of you. what you learnt in the classroom may not be but it could impress people at the coroners inquest.

whats the problem with discussing accidents. most advances in aviation have been the result of someone else killing themselves.

its always easier and cheaper to learn from someone elses mistakes.

whatunion says, man who never makes mistake never makes anything

[African Drunk]

We used to teach 300ft aca however the CAA examiner insisted on 200ft so it did not conflict with carrying out the IR test.

[FJJP]

Asymmetric training is potentially lethal. Over decades in the RAF it is a statistical fact that sim or practice asy accidents outstrip real asy accidents by a huge margin.

RAF asy rules are incredibly strict for just this reason. You never bust the asy speeds and heights. For example, the Canberra, perhaps the worst asy ac in recent years, the VCH was 600 feet. And that's genuine - I have been with a trapper who demonstrated just why it is so - and believe me, it was a hair-raising demo. The term 'parting grass' springs to mind.

I am astonished that the CAA and/or manufacturers haven't introduced rigid rules to protect pilots that have no experience or concept of the dangers involved - after all, it is exactly what test pilots are supposed to do, defining the limits to be flown by the average pilot. It should not be a question of the skill level of the pilot - some of the most experienced pilots in the world make mistakes and kill themselves because they break the rules. Rather the rules must be pitched at the average pilot; by all means add additional safety margins to the less experienced, but you have to have a definitive figure from which to start your calculations.

Having experienced years of sim, practice and real asy flying on a variety of 2 and 4 engined pistons and jets, I had developed a healthy respect for the published limits. I, and most of my friends are alive today because of them.

[bookworm]

So what are you proposing FJJP? Mandatory asymmetric commital heights? Let's say we make it 600 ft. That sounds like it has a nice additional safety margin included.

So I'm approaching my destination which is giving 2000m vis, BKN004 in my light twin. Not too challenging for a symmetric ILS approach. I'm solo, and at the end of my flight so I'm light. An engine starts to surge, to the extent that I feel the best strategy is to shut it down. Now what?

Options:

1) Fly the approach, knowing that by 600 ft I probably won't be visual so I'll have to do an asymmetric go around?

2) Fly to my alternate with a non-precision approach, reporting 3000m BKN007 but sufficiently far away that I need to fly for 30 mins on a single engine and I consume all my alternate fuel getting there.

3) Something else?

Your call.

[BEagle]

Well, I for one certainly think that there should be mandatory Engine Out Allowances and Visual Committal Heights applied to light piston singles......

The alternative? Press on with your approach, try to go-around......and die screaming?

[Oktas8]

Regarding mandatory visual committal heights - given that ten pilots will collectively have eleven opinions about anything technical, there would end up being so many "exceptions" that the basic rule would be swamped.

Compare the private owner of a well maintained twin in hilly country, with an experienced ME instructor operating at a flat airfield.

A suggested Engine Out Allowance in the POH / FM might work. But from this thread, it seems that most people have 200' - 300' in mind already, assuming no extraordinary circumstances.

O8

[whatunion]

interesting fatal at humberside for those who think these kites are easy near the ground!

[BEagle]

That's a bit tasteless, old chum. But I understand your drift.

[Oktas8]

Well, tell us then. What why how at Humberside?

[whatunion]

sorry not meant to be tasteless especially considering the examiner was killed but its a good lesson for all those instructors who expect their students to replicate training performance when it happens for real.

I cant think of one incident were a pilot has got away with a real engine failure near the ground in a perf c twin.

do you remember the clown who cleared customs at birmingham after a foreign trip and then positioned to wellesbourne. well that was his intention but he was low on fuel and instead of waiting to refuel at bhx he did a quick rolling take off from a taxyway turn and found out the meaning of surge and starvation just as he was getting airborne. the seneca ended up on its back and he was lucky to get away with his life.

this particular individual was an accident waiting to happen and one of the most arrogant and self opinionated amateur pilots i have ever come across. he was however well experienced on twins for a ppl with over 1000 hours and owned his own aircraft.

i believe not enough instructors drive home the point about the danger of light twins and single engine performance. there is a mighty big difference between a planned training simulated engine failure and one for real when you are not expecting it, as our friend above found out, the hard way!

ps the humberside accident detail is current in the uk flying mags

[bookworm]

I cant think of one incident were a pilot has got away with a real engine failure near the ground in a perf c twin

How would you expect to hear about them? There's no mandatory reporting of engine failures.

Don't get me wrong: your point is a good one. But it's almost impossible to build any sort of statistical picture.

We've also come a long way from discussing the risks of a planned asymmetric go around to the risks of an unplanned engine outage at low level. While the former is non-trivial, the latter is what seems to kill.

[whatunion]

how would i expect to know about them;

same way i get to know about anything in the industry i have worked in for 30 years.

your quite right about statistics but most twin pilots i have spoken to that are ppls have very misunderstood expectations of single engine performance.

i base some of my theory on some advice an ex air force jet pilot and very experienced exec prop jet and piston examiner gave me one day.

why have a higher miinima for a single engine app than an all engine app. the last thing in the world you want to do is do a se go around in a perf c twin, i totally agree.

the one difference i found between the pro flyng of twins in the charter world and the ppl flying of twins is that most pros had a very sceptical disbelief in se performance. whereas most ppl's really believed that a twin with an engine failure was just raring to go, not a problem, just the same as training.

ask yourselves how many ppl twin instructors actually show closing the live engine throttle and landing on the remaining runway as an option.

a suprise engine failure on take off or a low level go around on one engine is beyond the capabilities of most ppls in all but the ideal of conditions, thats my theory.

[Oktas8]

ask yourselves how many ppl twin instructors actually show closing the live engine throttle and landing on the remaining runway as an option
Problem is, that's a little difficult to do. As a twin instructor, I don't give engine failures at very low level after take-off (frankly, below 200 feet scares me) because, if the student is going to do something silly, it will be difficult to recover in time to avoid Terra Firma. Even if you do close the live throttle, you have to operate at an airfield with a long enough runway to land back on safely, after having allowed the student to try to climb for two or three seconds and demonstrate the futility / danger of the task.

EF at low level on a go-around? Yes, done it, student struggled to cope, I took over, student said "ah yes, I'll remember that next time." But human factors says that s/he probably won't in the heat of the moment. Plus, again, I would like to die of old age, and giving engine failures at 200' or less with someone else as PF is not a good way to achieve this... :}

It would be much easier if we had simulators that accurately reproduced this kind of problem - but how many flying schools have simulators of this calibre with good visuals?

Alternatively, many more things are possible with more dual experience on type - but then all the students will disappear down the road to the school offering ratings at the legal minimum hours.

This last point is perhaps the biggest drawback to improving standards in a highly deregulated industry.

cheers to all,
O8

[pilotbear]

Asymetric committal altitude is just normal decision ht 200' plus EOA depending on speed.
When initiating a go around single engine the power has to be increased before the pitch up action to allow an initial accelaration, if pitch is increased first the a/c will decelarate and if you are close to blue line trouble will ensue.
Therefore, the aircraft will continue to descend during this power increase. Typically it takes 80-100' of further descent for this accelaration this is the engine out allowance.

So, the asymetric go around ht is dependent on approach speed.
It should 200' unless 10kts above, at or below blue line in which case 100' should be added making 300' to commit or initiate.

[hugh flung_dung]

We seem to have a number of different threads going here for the different critical failure situations. This is what I propose (and teach):

At any time that control is not possible: shut both throttles, lower the nose ('tis better to hit the ground erect than inverted)
Take off
- Default: throttles shut, stop on the runway ('tis better to be on the ground than in the air)
- If below Vtoss (take off safety speed) and below some height (why not the same as Ach (asymm commital ht): throttles shut, land straight ahead (option to then use power to reach best area)
- If above Vtoss & Ach: control, lower nose, full power, gear/flaps up, identify/verify/feather, small bank to live, etc at Vyse
Base leg (no fire): control, lower nose, full power, gear/flaps up, identify/verify/feather, small bank to live, etc at Vyse.
Final approach
- Visual, above Ach (no fire): control, lower nose, full power, gear/flaps up, identify/verify/feather, small bank to live, etc at Vyse
- Visual, Below Ach: land straight ahead, use power to reach best area (but lead with rudder!), speed may decrease to normal Vat asymm
- IMC, add Ach to DA/MDA to generate new DA/MDA (because Ach is designed to allow the aircraft to be accelerated, cleaned-up and a climb initiated)
[/list=1]

Personally:
[list]
I insist on a full failure brief at the end of the pre-departure checks
I don't PRACTICALLY teach/examine engine failure on the runway because the margins are just too small and I believe it introduces an unacceptable risk for little benefit
I don't PRACTICALLY teach/examine EFATO below the commit height/speed for the same reason.
For those that advocate teaching EFATO and land ahead from around 250ft, have you done the sums? Very roughly (assume MAUW, +15C, paved, sea level, nil wind, 0 flap)...
TODR 2200
distance from 50ft to 200ft (9s @ 1000fpm & 85kts): 1300ft
distance from 200ft to 50ft (18s @ 500fpm & 85kts): 2500ft
LDR 1900
Total, without the fiddle factors for grass and dither: 7900ft or 2.4km - does anyone instruct on light singles from a runway this long? Even if you do it at 50ft (staggeringly unsafe in my view) the distance needed is around 4000ft without any faffing about.
I DO practically teach EFATO from a symmetric go-around just above Ach because this is fully representative of the take-off situation but with healthier margins and doesn't upset any NIMBYs at the end of the runway (well not as much, anyway)


Don't get me wrong, I agree wholeheartedly that if you are unable to control the aircraft at any time there is only one answer, become a glider and land the right way up in the best place available. My problem is that it is impractical to teach it at ground level, that's why the full EFATO drill is usually taught and examined at 2000-4000ft.

Health warnings:
1. I haven't got the Seneca POH to hand, these numbers came from some rough notes
2. the wine at lunchtime may have caused some imprecision in the calculatory department
... please don't shout too loud if I've boobed.

HFD

[whatunion]

how about going to a runway where it is possible to land ahead!
a twin rating is not just valid for the home airfield.

i did say showing the student, not letting them do it

[Oktas8]

hugh_flung_dung,

The way you & I teach is very similar!

You mention (point 2.) the below Vtoss, below ACH and the above Vtoss, above ACH cases - what about above the one and below the other? It gets complicated, which is not what one looks for in an emergency brief.

For this reason, I teach the same as you do, but without any discussion on heights. One is either at/above a safe speed or one is not. (This assumes a typical take-off profile where reaching a particular speed occurs at a similar height for all take-offs.)


Regarding EFATO, land back on:
Given h_f_d's numbers look broadly correct, I don't instruct near any 2400m runways where GA training is permitted. Could do it in the instrument ground trainer though...

O8

[BEagle]

Above and below Vtoss I can understand. But the concept of ACH applying on departure above Vtoss escapes me. The a/c is already effectively in the 'go-around' configuration and is hardly in the process of landing!

KISS would seem to apply here?

Why not just operate the a/c such that, once above Vtoss an engine failure at any stage of flight will not prove critical. Use such speeds and configurations as will enable level flight at least in the event of the failure of the most critical engine. Fly the approach (with both engines operating) in such a configuration that it may safely be continued following failure of the most critical engine (might require some demanding power, drag, trim and fancy footwork following a failure when close to Vmca in the landing configuration?) - and subsequently only configure fully for landing when at ACH on one?

[hugh flung_dung]

BEAGS: I didn't explain it too well.
In practise, from a failure at Vtoss in a Seneca 1, the average stude cannot get through the drills and start to climb without descending 100-150ft, even at training weights. Because of this I add a height to the go/nogo decision; historically I've used 200ft.
As I wrote my earlier mammoth offering it dawned on me that it would be easier for the stude to have one height gate to deal with - following the KISS principle let's call it Ach. Engine failure below Ach, whether taking off or landing, leads to a landing. What do people think?

Why not just operate the a/c such that.... The aircraft should always be operated at/above Vyse on the approach until Ach so that a go-around can be flown without needing to accelerate if there's a failure. Following a failure at 500ft (eg) on the approach there are 3 choices:[list=1]
Asymmetric go-around followed by pre-meditated asymm approach and landing
Continue the approach whilst trying to identify/verify/feather
Continue the approach without drills
[/list=1] My humble opinion is that the safest option is #1. I know from experience that many people coc& up a simulated failure below Ach and reach horrendous attitudes whilst trying to reach the runway from 250ft, I dread to think what they'd do from 500ft.

OKTAS8: Great minds think alike (or fools seldom differ!).
'Agree totally: what DO you do below Vtoss & above Ach or vice versa? I discuss this and usually we end up agreeing that logically you can dive to increase speed, or bleed the excess speed whilst going through the drills.
The bottom line is that below Vtoss the actions are: control, throttles shut, lower nose. Once this is done you can use available power to reach the best landing area - in practise this MAY mean that you find you can reach Vyse and climb. Having just re-read the previous sentence it sounds too positive - has anyone found a better way of expressing it?

(BTW, my only real engine failure was a slow affair at FL100.)
HFD

[DFC]

When comparing operation on one engine during the departure phase to the same during the approach phase one must remember the following very important facts;

1. At departure the weight is higher than during the approach at the end of (or later in) a flight;

2. During an approach, the attitude is low and a simple slight lowering of pitch can regain any lost speed without significant alteration of the profile;

3. During departure, the nose attitude is high, drag is relatively high and potential energy loss when an engine fails is significant

4. Many light multi engine aircraft have negative climb gradients at Max take-off weight

5. During the approach phase, the pilot is minded towards getting the aircraft on the ground and simply closing both throttles and gliding is not totally alien.

6. During the departure phase, the pilot is minded towards gaining height and closing both throttles can be very alien.

When departing, if no positive flight path can be constructed following engine failure then there are ceiling and visibility limitations that provide some posibility of a successful forced landing.

Regards,

DFC

-------
Pilot Bear,

Don't forget that blue line is best rate of climb speed single engine. Just like a single engine aircraft, it can be possible to climb at speeds above and below that speed but the rate is reduced. Of course if the gradient is negative at blue line then just like the best glide speed in a single, either side will vary the rate of descent.

[Oktas8]

With regard to height and Vtoss h_f_d:

I fly a Seminole, which is a tad lighter than the Seneca, and possibly has a better power-to-weight ratio. However, I deal with the possibility of height loss while doing EFATO drills by specifying Vyse as a decision speed (and gear retraction speed) rather than Vtoss - 88kts versus approximately 68kts.

This has some benefits for low-time pilots unfortunate enough to get a real EFATO.

-They are already at a speed offering best performance.

-The acceleration profile means they are further from the ground before they have to deal with the EF by doing drills.

-If drills are required, the gear is already at least in transit if not up.

-Lastly, the point marking the end of the "close the throttles" region and the beginning of the "deal with it and then assess performance" region is where they retract the undercarriage - an EF after this point does not require thought about airspeed and height, but merely "have I retracted the gear or not?"

I know that to use a height and a speed in the manner you've described h_f_d, is more flexible and possibly more operationally orientated than my method. However, from reading it appears that it takes about five seconds from EF to the pilot reacting logically, and longer if the reaction requires thought / memory. I do worry about pilots I've trained killing themselves while thinking about what has just happened and what they should do.

Hmm. Sorry about the wordiness - hope the sense comes through.

regards to all,
O8

PS never had an engine failure. Book smarts only, hope to stay that way! :ok:

[askmelater]

For what it's worth:

We distinguish between the decision height (ADH) and the committal height (ACH). ACH being the absolute minimum considered possible to achieve a safe asymmetric go-around and ADH being some factor added to ACH allowing for pilot ability etc etc. Making the decision at ADH allows some descent while aircraft configuration changes are made. In some cases ACH and ADH may be the same. We use 200' AGL for ACH and min 100' AGL added to this for training; (typically 200' with demonstrated student ability).

What instructors impress upon students is not necessarily what they take with them on their own operations, no matter how clear you think you were. There will always be students who consider themselves as good as or better than their instructor. Does this mean that you failed in your teaching? (This is not to say we shouldn't take any responsibility at all during teaching).

Lastly, in all the talk about IMC asymmetric go-arounds no one has yet mentioned the MAP climb gradient - 2.5%. Last time I checked most light piston twins would be struggling to achieve this at anywhere near MCTOW. This surely would necessitate a higher MDA/DA and an allowance on this height to ensure you were climbing by the new ADH/DA?

AML

[Tinstaafl]

Yes, it does. Oz rules specify that the MDA/DA must be adjusted if the a/c would be incapable of achieving 2.5%

[Oktas8]

Oz rules specify that the MDA/DA must be adjusted if the a/c would be incapable of achieving 2.5%

All very good. But to do this, you will need to know where the critical obstacles are, so you can work back from them at the actual climb gradient, and determine your new single-engine MDA.

This information is not in the ordinary approach plates. How many flying schools purchase the missed approach terrain survey maps from the certifying authority, just to do this calculation?

Two further complications:

a) The "ordinary" MDA may not be climb gradient restricted, but restricted for some other reason, such as not below 300' agl at the missed approach point. (This would be the case where a missed approach takes you over water or flat ground, with a turn at a non-specified distance to return to the airfield hold).

b) Your clapped out light twin may be incapable of climbing to, or maintaining, altitude in the missed approach hold when asymmetric. In that case, what's the point of doing the missed approach at all? If asymmetric in IMC, you are effectively committed to land from the initial approach point.

Food for thought, I'd welcome corrections.
O8 :)

[askmelater]

Tinstaafl - absolutely correct and Australia is not alone. Any country that constructs their procedures using ICAO PANSOPS will have similar statements in their AIP (or equivalent). What I find interesting, (as far as I can see anyway and happy for someone to direct me here) is there is no guidance provided in any official publication as to how to go about adjusting your MDA/DA. I have a couple of ways and I am sure others have their own.

Oktas8 - I have to confess to being a little confused. I presume the 300’ you talk about in your first point (a) is the normal ADH that one would select, regardless of conducting the flight under IFR or not? And I am unsure exactly what you mean with the comment in parenthesise. Could you explain a little more?

On your second point (b) I agree but with one addition (not really a correction); as you say the MAP is the least desirable option and one should consider themselves committed from the IAP. Therefore (and here is my addition) it would be unwise to even start the approach if visual reference is not assured at some margin above the normal MDA/DA. Put another way; you must establish an actual cloud base prior to approach commencement, this could be from ATC or a preceding aircraft having successfully completed the same approach.

You could take it a step further and say that descent below MSA (be it the holding MSA or route MSA) must not be considered unless a successful approach is assured, the exception being airspace with radar terrain services. I am not suggesting that you will be able to maintain every MSA as clearly any mountainous terrain MSA in a normally aspirated light twin with an engine out will be nigh impossible, but where practical.

Interested in more of this topic if anyone has any.

AML

[Oktas8]

Yes, sorry askmelater, my post was a little too summarised.

In point (a) I am suggesting that some MDA's are not set at a height to guarantee terrain clearance on a 2.5% climb gradient, but instead are set because it is deemed unsafe to go lower on a non-precision approach.

For example, a certifying authority might decide that it does not want an aircraft descending below 300' AGL until the aircraft is visual to the runway - this is not the pilot's ADH, but is the specified MDA for the approach. For example, let's say that the certifying authority wants 296 feet of terrain clearance at circling MDA, so that is the MDA set.

If the missed approach takes the aircraft over water or flat terrain, then terrain clearance on the missed approach is not an issue, and there is no obstacle clearance related reason for the asymmetric aircraft to use a higher MDA to compensate for reduced climb gradient. The pilot of a light twin that has suffered an engine failure could quite happily go down to MDA, and climb back at a 1% climb gradient if necessary - he or she isn't going to hit anything, so why set a higher MDA than necessary? (I'm ignoring airspace and traffic flow problems in this scenario.)

In my post I am attempting to show that adjusting MDA to account for reduced climb performance is difficult, sometimes expensive, and unable to be done accurately by the pilot in the cockpit - there are too many variables.

Perhaps I should sell my CAA's half-baked effort at an AIP & buy a Jeppesen suite of manuals with some decent PANS-OPS reference material in it!

O8

[alf5071h]

Much of the discussion in this thread has been on the technicalities of asymmetric GA However; I would be interested to hear views on the philosophy of instruction particularly for students who aim to fly larger JAR 25 aircraft that in general do not have need for a committal height.

One concern is of negative training, where the necessary awareness of committal height in a light twin is then transferred into commercial operations and leads to reluctance to GA from low altitude. First learnt, best remembered.

In addition, how do the instructors on the heavier JAR 25 aircraft, un-teach those different aspects between training aircraft and commercial operation. I exclude the problems of modern avionics etc; at least until a Cessna 172 has a HUD.