Air North Brasilia Crash in Darwin (Merged)
Acceleration will only occur if forces are not in balance, on take-off thrust is much greater than drag, therefore rapid speed increase is the result. The moment thrust is removed acceleration will stop. Inertia will only affect the present motion of the aircraft at that time, it can not cause an object to change speed or direction in any way. Acceleration will only continue as a result of the slow spool down of the engines, depending on type it may continue to produce positive thrust for a few more seconds after ground idle is selected.
Removed missleading rubbish regarding certification...
Removed missleading rubbish regarding certification...
Last edited by 43Inches; 29th Mar 2010 at 12:19.
Check Attitude
Join Date: Nov 2001
Location: Queensland, Australia
Posts: 476
Likes: 0
Received 0 Likes
on
0 Posts
This thread has evolved into one of the best technical discussions on assymetric operations.
Out of tragedy, much will be learned and incorporated.
My area of experience is GA twins, not transport category turbo prop twins.
When sitting in the right hand seat, about to fail an engine during training, checking or an endorsement,
I recognise that EFATO drills in a GA twin have a high risk factor.
Was the exercise adequately briefed, is it a long runway or a short runway, will the gear be up or in transit, flaps position etc.
The accident and the accompanying fatality rate in GA twin assymetrics makes sobering reading.
The training is meant to be eventually life saving, yet too often it becomes life threatening.
Vmca demos should and are demonstrated at a safe height, and to some degree initial EFATO drills can also be carried out at or above 3,000 ft.
When a satisfactory level of demonstrated competence is established, it may be beneficial to conduct actual simulated failures off the runway.
There will be a high degree of risk, and prudence may dictate waiting until gear retracts before failing the engine, depending on type.
Simulators are preferable for obvious reasons of safety, but PA31, C402, C310, BE58, BN2, P68 etc simulators
that can adequately be used for this exercise are either not available in Australia, or very rare.
By continuing this excellent technical discussion we are helping to ensure that some benefit will come out of this tragedy.
Out of tragedy, much will be learned and incorporated.
My area of experience is GA twins, not transport category turbo prop twins.
When sitting in the right hand seat, about to fail an engine during training, checking or an endorsement,
I recognise that EFATO drills in a GA twin have a high risk factor.
Was the exercise adequately briefed, is it a long runway or a short runway, will the gear be up or in transit, flaps position etc.
The accident and the accompanying fatality rate in GA twin assymetrics makes sobering reading.
The training is meant to be eventually life saving, yet too often it becomes life threatening.
Vmca demos should and are demonstrated at a safe height, and to some degree initial EFATO drills can also be carried out at or above 3,000 ft.
When a satisfactory level of demonstrated competence is established, it may be beneficial to conduct actual simulated failures off the runway.
There will be a high degree of risk, and prudence may dictate waiting until gear retracts before failing the engine, depending on type.
Simulators are preferable for obvious reasons of safety, but PA31, C402, C310, BE58, BN2, P68 etc simulators
that can adequately be used for this exercise are either not available in Australia, or very rare.
By continuing this excellent technical discussion we are helping to ensure that some benefit will come out of this tragedy.
Join Date: Jul 1999
Location: PLANET ZOG
Posts: 24
Likes: 0
Received 0 Likes
on
0 Posts
Mainframe
Yes I agree, it has turned into a worthy technical discussion. The thread has got off the original posted subject, but there is nothing further to be gained by speculation or ‘quarter backing’.
However there is much to be gained by sharing experiences and ideas that may help prevent a future tragedy occurring.
This thread has evolved into one of the best technical discussions on assymetric operations.
Yes I agree, it has turned into a worthy technical discussion. The thread has got off the original posted subject, but there is nothing further to be gained by speculation or ‘quarter backing’.
However there is much to be gained by sharing experiences and ideas that may help prevent a future tragedy occurring.
Join Date: Jul 1999
Location: PLANET ZOG
Posts: 24
Likes: 0
Received 0 Likes
on
0 Posts
remoak
Yes true, the same principles still apply, however I proffer that the handling technique may well be quite different.
The Brasilia does have a reduced thrust schedule and typically during training we’d set around 80% Tq (79% being min Tq from memory) on T/O to simulate a loaded a/c. From my experiences it was not necessary to increase thrust during a simulated engine out T/O, but reserve thrust was available up to 120% Tq (time limited) and beyond 120% if you still really needed it (emergency use only).
So power per say or the ability to climb away on one, was never the issue.
With the big 4 bladed carbon fibre blades on this bird, there was a big difference in managing a critical engine failure (#1) as compared to a #2 engine failure and required sound technique to maintain directional control. As I have commented previously, you could sometimes find yourself surprised at the amount of control inputs required on the critical failure case, so there was a considerable difference between the two ‘engine out’ exercises.
I haven’t flown the F27 so can’t comment on any differences in handling or performance between the two a/c.
The same principles apply in a turboprop. Those that don't have a reduced thrust schedule, normally have a thrust augmentation system (ie water meth in the F27 or some J32s).
The Brasilia does have a reduced thrust schedule and typically during training we’d set around 80% Tq (79% being min Tq from memory) on T/O to simulate a loaded a/c. From my experiences it was not necessary to increase thrust during a simulated engine out T/O, but reserve thrust was available up to 120% Tq (time limited) and beyond 120% if you still really needed it (emergency use only).
So power per say or the ability to climb away on one, was never the issue.
With the big 4 bladed carbon fibre blades on this bird, there was a big difference in managing a critical engine failure (#1) as compared to a #2 engine failure and required sound technique to maintain directional control. As I have commented previously, you could sometimes find yourself surprised at the amount of control inputs required on the critical failure case, so there was a considerable difference between the two ‘engine out’ exercises.
I haven’t flown the F27 so can’t comment on any differences in handling or performance between the two a/c.
Last edited by ace from space; 29th Mar 2010 at 05:06.
Join Date: Oct 2003
Location: In an Airplane
Posts: 125
Likes: 0
Received 0 Likes
on
0 Posts
As I have commented previously, you could sometimes find yourself surprised at the amount of control inputs required on the critical failure case,
Join Date: Sep 2006
Location: back of the crew bus
Posts: 1,312
Likes: 0
Received 0 Likes
on
0 Posts
I haven’t flown the F27 so can’t comment on any differences in handling or performance between the two a/c
Certainly in Europe, the F27 would never be certified if it was introduced as a new type - it simply doesn't have the performance margins.
The F27 does have a thrust reduction schedule - kinda - via the fuel trimmers. Nothing like a modern turboprop though.
Join Date: Oct 2003
Location: In an Airplane
Posts: 125
Likes: 0
Received 0 Likes
on
0 Posts
Might have been this one:
Title: Light Twin Engine Aircraft Accidents Following Engine Failures, 1972 - 1976.
NTSB Report Number: AAS-79-02, adopted on 12/13/1979
NTIS Report Number: PB80-177306
Link: http://libraryonline.erau.edu/online...s/AAS79-02.pdf
Might be on page 15 or there about.
Title: Light Twin Engine Aircraft Accidents Following Engine Failures, 1972 - 1976.
NTSB Report Number: AAS-79-02, adopted on 12/13/1979
NTIS Report Number: PB80-177306
Link: http://libraryonline.erau.edu/online...s/AAS79-02.pdf
Might be on page 15 or there about.
Last edited by privateer01; 29th Mar 2010 at 05:59. Reason: link
Join Date: Oct 2003
Location: In an Airplane
Posts: 125
Likes: 0
Received 0 Likes
on
0 Posts
Good discussion, however, I will have to put my 2 cents worth in.
Remoak, consider if you accelerate in your car to 100km/h, where you enter a steady state at 100km/h. The last part -say from 90 to 100 km/h - isn't deceleration, it is a reduction in acceleration until you are in a steady state, i.e. equilibrium. Consider also a propellor loss on an aircraft, it doesn't stay on the shaft until landing. Indeed it departs at a great rate as the rotating momemtum has to be brought also to a state of equilibrium before the dynamic pressure could hold it in place. Condolences to the families.
Remoak, consider if you accelerate in your car to 100km/h, where you enter a steady state at 100km/h. The last part -say from 90 to 100 km/h - isn't deceleration, it is a reduction in acceleration until you are in a steady state, i.e. equilibrium. Consider also a propellor loss on an aircraft, it doesn't stay on the shaft until landing. Indeed it departs at a great rate as the rotating momemtum has to be brought also to a state of equilibrium before the dynamic pressure could hold it in place. Condolences to the families.
Join Date: Sep 2006
Location: back of the crew bus
Posts: 1,312
Likes: 0
Received 0 Likes
on
0 Posts
Well the discussion on acceleration is probably a diversion that isn't really worth the time... however... just to finish it off..
That's fine - because in that case, thrust isn't removed, it is just reduced until thrust=drag (to put it crudely). Not the same thing at all as taking your foot off the throttle. You could argue that a car engine behaves in the same way as a high-bypass gas turbine (ie the effect of inertia), however as discussed inertia will not produce an acceleration (well it will, but in the opposite direction).
No... it leaves the shaft because a) it has inertia; and b) the aircraft suffers a negative acceleration due to the loss of the thrust of the prop (ie it slows down a bit). Nothing to do with the other case at all.
As I said before, if you can make a suddenly-free prop accelerate (positively) when there is absolutely no power being applied to it, you have invented perpetual motion. Can I buy shares in your company?
consider if you accelerate in your car to 100km/h, where you enter a steady state at 100km/h. The last part -say from 90 to 100 km/h - isn't deceleration, it is a reduction in acceleration until you are in a steady state, i.e. equilibrium.
Consider also a propellor loss on an aircraft, it doesn't stay on the shaft until landing. Indeed it departs at a great rate as the rotating momemtum has to be brought also to a state of equilibrium before the dynamic pressure could hold it in place.
As I said before, if you can make a suddenly-free prop accelerate (positively) when there is absolutely no power being applied to it, you have invented perpetual motion. Can I buy shares in your company?
Moderator
The speed might be increasing slightly, but overall there is a deceleration.
if we take the failure point acceleration as reference then, of course, there is a subsequent deceleration. However, if you take an earth based reference, there is a progressive reduction in acceleration (during which the aircraft is still increasing its speed relative to the reference frame) eventually ending in a deceleration and the speed reduces .. is this not really a vital consideration for the discussion ?
What you are SEEING on the instruments is not necessarily what is actually happening,
reality dictates that your observation is incorrect, I'm afraid. I would be a bit concerned if the ASI were such an unreliable gadget ..
for a variety of reasons
perhaps you might list the variety of reasons to which you refer ?
Maybe, but I really can't see why you wouldn't have the taps closed and the brakes/antiskid/spoilers doing their thing.
during a local exercise (sim/aircraft) the pilot is primed and hot to run with a reject .. knowing that the problem is coming.
Reality on the line is quite different and startle factor has to be contended with .. it is for this sort of consideration that A/L 42 was introduced
I'm having trouble with the idea that you could ever get into that situation in the first place.
if all goes well and the speed/thrust situation puts you above the real Vmcg/Vmca .. fine.
However, the book figures are just that .. reference data for specific circumstances.
Change the circumstances and the values change .. consider, for a min weight, min speed schedule takeoff and a reasonably aftish CG ..
(a) significant adverse crosswind and there goes book Vmcg out the window .. something in the order of 0.5 kt/kt (twin) to in excess of 1.0 kt/kt (quad) increase in Vmcg .. and away you go into the weeds off the side of the runway.
(b) higher than certification thrust .. likewise for both Vmcg and Vmca
(c) out of left field drag situation on the failed engine .. failure of the NTS/autofx/decouple/whatever system ... likewise for both etc ..
(d) bank miscontrol in the air .. Vmca is VERY bank dependent. For instance, a well known lots of engines strategic bomber, as I recall, has a Vmca delta of around 43 kt if the bank is 5 degrees into the failed engine .. maybe it was 34 kt ? ... long time ago now since I did that course ... and the memory is a bit scratchy on the specifics ... either way .. a lot of knots however you look at it ..
.. need I go on ?
Vmcg is never going to be above V1
see above ... realworld Vmcg can be quite different to book Vmcg on the day if things aren't going your way ... if that happens to be above V1 .. and that's not hard to have happen for a min weight min speed schedule takeoff .. then the pilot can find himself in a whole world of hurt.
if a failure has occurred below Vmcg, you are (unless you are Centaurus) going to stop.
thank heavens for that .. knowing Centaurus as well as I have over the years, I suggest that he will be quite aware of his closeness to Vmcg and tailor his management strategies accordingly...
I am still wondering what aircraft you are thinking of
still most of them ..
even the F27 will accelerate quickly through that particular danger area with an engine out.
now, I had three years of great fun on the Friendly .. but it, the same as most other aircraft, WON'T accelerate through the problem area if, for whatever reason, the aircraft has already departed in yaw ... keep thrust on and you just wind yourself into a tighter ball in the weeds ...
Why would you be trying to GO from below Vmcg?
I wouldn't.
The discussion relates to the situation when, for whatever reason, the pilot finds himself below the real world Vmcg on the day
I am still wondering what aircraft you are thinking of.
I'm still thinking of most aircraft ..
All the turboprops I have flown will quite happily accelerate (assuming that you rotated at the correct speed) and climb away, at training weights
refer to the above discussion re real world Vmcg/Vmca
But that is a gradual process.
granted .. but, for an equivalent sort of problem on takeoff there is nothing very gradual about the departure ...
Being fast enough to avoid a yaw departure means, basically, being above Vmca.
or Vmcg according to where you are on the takeoff ..
if you are flying the departure correctly, there isn't really a reason not to be well above that speed.
.. providing that the book speed hasn't been corrupted on the day due to unforeseen circumstances ...
once you are at V2 there is no reason that I can see for a yaw departure to occur.
as before, for a min weight, min speed schedule .. if the real world Vmca is significantly higher ... then you are in the middle of it all. Perhaps you are putting far too much of your trust in the "guaranteed" transference of the artificial world of certification into the real dirty world of day to day operations ?
if you mis-handle the departure, maybe, but that isn't what we are talking about.
well, actually, we are .. that's one of the scenarios I am discussing
inertia is more of a factor in heavies
I don't think so .. just changes the numbers ..
GA twins are far more marginal than most transport-category turboprops.
not at all .. make the numbers appropriate and you can have just as miserable day in a heavy as on a puddlejumper
It does take a little while for an engine to run down
I'm glad we seem to be getting to an agreed position ..
the core has inertia
the fan is the main problem ..
that doesn't mean that it is producing any meaningful thrust.
have to disagree there .. during the rundown the thrust starts high and progressively runs down to whatever drag pertains to a failed engine
Remove the fuel, and you remove most of the energy
to cite a somewhat silly analogy .. consider a motor car .. the driver turns off the engine (no fuel flow now). You are standing in the middle of the road 50 odd metres away in the path of the car ... outcome ? I think I'd opt for exiting stage left pronto ?
as soon as the rate of increase of velocity peaks, you have deceleration (by definition)
depending on your frame of reference .. we are more concerned about earth based frames of reference, I suggest.
You simply can't have an acceleration if thrust is reduced below that necessary to counter drag
that sounds about right to a simple minded engineer like me
most of what you are seeing as a speed increase is actually instrument error
don't think so
I would love to see some hard facts on that whole subject
the following link
is to a scan of a page from the Boeing Performance Engineers Training Manual. Cursory review infers the failed engine progressive thrust rundown following Vef.
Similar pictures abound in the literature and reflect physical reality .. unless you are postulating infinite forces coming into play at the time the engine fails ...
There will be a little residual thrust, but most of it is being absorbed by the rotating mass of the engine (Newton's Third Law again).
we probably might be interested in having your expanded discussion on this point ?
Whatever thrust there is, is highly unlikely to be enough to counter drag and inertia sufficiently to produce a significant speed increase.
the numbers will vary across different Types .. Centaurus' figures reflect my recollections of the 737 ..
Think about it - if you launched down the runway at say, 25% thrust, would you ever get to V1? Probably not
I think I'd agree with that .. but the analogy essentially is irrelevant to the discussion
This is how the aircraft is certified
I would recommend a read of AC 25-7A first
I recognise that EFATO drills in a GA twin have a high risk factor.
lots of risks for heavy iron as well .. however, that statement is the start of doing some sensible risk assessment and mitigation when it comes to in-aircraft local proficiency training sessions ..
Vmca demos should and are demonstrated at a safe height
where they cease to be valid (other than for generic training value) and expose the aircraft to the risks of stalling and spinning ... for the life of me I can see little reason for Vmca sandpit playing beyond the initial multiengine endorsement .. and, even then, why not just stay away from Vmca ? A sensible quasi demo (not of Vmca but something approaching the sort of problems relevant to a static Vmca departure) at a sensible height has some validity for reinforcing the briefing but, even then, is of questionable value
it was not necessary to increase thrust during a simulated engine out T/O
or even a real failure .. and, then, be very wary of simply shoving the throttles up .. lest you be caught out with a thrust overshoot and the potential for a yaw departure if at min weight and min speed schedule. I can recall at least one fatal where this was postulated at the time (by me, as it happened) as a contributory cause.
the F27 would never be certified if it was introduced as a new type
the same can be said for just about any dated Type ..
it leaves the shaft because
.. of aerodynamic forces associated with its motion and, subsequently, gyroscopic precession once it is launched into independent flight ... never a nice thing to contemplate ...
when there is absolutely no power being applied to it
thrust is the key, not power ... one uses power to end up producing thrust.
you have invented perpetual motion
I have a suspicion that you really might need to revisit a dissertation on perpetual motion ?
I'm not really trying to be overtly difficult and a PITA here .. but the reality is that we spend most of our training time in comparatively benign areas of the operating envelope ... any fool/monkey can handle an overspeed takeoff failure at V1 .. in my view, ALL pilots should be exposed to the edge of this particular area of the envelope if for no other reason than to engender a very healthy respect for how hard an aircraft can bite if it's not treated with the healthy respect it demands ...
if we take the failure point acceleration as reference then, of course, there is a subsequent deceleration. However, if you take an earth based reference, there is a progressive reduction in acceleration (during which the aircraft is still increasing its speed relative to the reference frame) eventually ending in a deceleration and the speed reduces .. is this not really a vital consideration for the discussion ?
What you are SEEING on the instruments is not necessarily what is actually happening,
reality dictates that your observation is incorrect, I'm afraid. I would be a bit concerned if the ASI were such an unreliable gadget ..
for a variety of reasons
perhaps you might list the variety of reasons to which you refer ?
Maybe, but I really can't see why you wouldn't have the taps closed and the brakes/antiskid/spoilers doing their thing.
during a local exercise (sim/aircraft) the pilot is primed and hot to run with a reject .. knowing that the problem is coming.
Reality on the line is quite different and startle factor has to be contended with .. it is for this sort of consideration that A/L 42 was introduced
I'm having trouble with the idea that you could ever get into that situation in the first place.
if all goes well and the speed/thrust situation puts you above the real Vmcg/Vmca .. fine.
However, the book figures are just that .. reference data for specific circumstances.
Change the circumstances and the values change .. consider, for a min weight, min speed schedule takeoff and a reasonably aftish CG ..
(a) significant adverse crosswind and there goes book Vmcg out the window .. something in the order of 0.5 kt/kt (twin) to in excess of 1.0 kt/kt (quad) increase in Vmcg .. and away you go into the weeds off the side of the runway.
(b) higher than certification thrust .. likewise for both Vmcg and Vmca
(c) out of left field drag situation on the failed engine .. failure of the NTS/autofx/decouple/whatever system ... likewise for both etc ..
(d) bank miscontrol in the air .. Vmca is VERY bank dependent. For instance, a well known lots of engines strategic bomber, as I recall, has a Vmca delta of around 43 kt if the bank is 5 degrees into the failed engine .. maybe it was 34 kt ? ... long time ago now since I did that course ... and the memory is a bit scratchy on the specifics ... either way .. a lot of knots however you look at it ..
.. need I go on ?
Vmcg is never going to be above V1
see above ... realworld Vmcg can be quite different to book Vmcg on the day if things aren't going your way ... if that happens to be above V1 .. and that's not hard to have happen for a min weight min speed schedule takeoff .. then the pilot can find himself in a whole world of hurt.
if a failure has occurred below Vmcg, you are (unless you are Centaurus) going to stop.
thank heavens for that .. knowing Centaurus as well as I have over the years, I suggest that he will be quite aware of his closeness to Vmcg and tailor his management strategies accordingly...
I am still wondering what aircraft you are thinking of
still most of them ..
even the F27 will accelerate quickly through that particular danger area with an engine out.
now, I had three years of great fun on the Friendly .. but it, the same as most other aircraft, WON'T accelerate through the problem area if, for whatever reason, the aircraft has already departed in yaw ... keep thrust on and you just wind yourself into a tighter ball in the weeds ...
Why would you be trying to GO from below Vmcg?
I wouldn't.
The discussion relates to the situation when, for whatever reason, the pilot finds himself below the real world Vmcg on the day
I am still wondering what aircraft you are thinking of.
I'm still thinking of most aircraft ..
All the turboprops I have flown will quite happily accelerate (assuming that you rotated at the correct speed) and climb away, at training weights
refer to the above discussion re real world Vmcg/Vmca
But that is a gradual process.
granted .. but, for an equivalent sort of problem on takeoff there is nothing very gradual about the departure ...
Being fast enough to avoid a yaw departure means, basically, being above Vmca.
or Vmcg according to where you are on the takeoff ..
if you are flying the departure correctly, there isn't really a reason not to be well above that speed.
.. providing that the book speed hasn't been corrupted on the day due to unforeseen circumstances ...
once you are at V2 there is no reason that I can see for a yaw departure to occur.
as before, for a min weight, min speed schedule .. if the real world Vmca is significantly higher ... then you are in the middle of it all. Perhaps you are putting far too much of your trust in the "guaranteed" transference of the artificial world of certification into the real dirty world of day to day operations ?
if you mis-handle the departure, maybe, but that isn't what we are talking about.
well, actually, we are .. that's one of the scenarios I am discussing
inertia is more of a factor in heavies
I don't think so .. just changes the numbers ..
GA twins are far more marginal than most transport-category turboprops.
not at all .. make the numbers appropriate and you can have just as miserable day in a heavy as on a puddlejumper
It does take a little while for an engine to run down
I'm glad we seem to be getting to an agreed position ..
the core has inertia
the fan is the main problem ..
that doesn't mean that it is producing any meaningful thrust.
have to disagree there .. during the rundown the thrust starts high and progressively runs down to whatever drag pertains to a failed engine
Remove the fuel, and you remove most of the energy
to cite a somewhat silly analogy .. consider a motor car .. the driver turns off the engine (no fuel flow now). You are standing in the middle of the road 50 odd metres away in the path of the car ... outcome ? I think I'd opt for exiting stage left pronto ?
as soon as the rate of increase of velocity peaks, you have deceleration (by definition)
depending on your frame of reference .. we are more concerned about earth based frames of reference, I suggest.
You simply can't have an acceleration if thrust is reduced below that necessary to counter drag
that sounds about right to a simple minded engineer like me
most of what you are seeing as a speed increase is actually instrument error
don't think so
I would love to see some hard facts on that whole subject
the following link
is to a scan of a page from the Boeing Performance Engineers Training Manual. Cursory review infers the failed engine progressive thrust rundown following Vef.
Similar pictures abound in the literature and reflect physical reality .. unless you are postulating infinite forces coming into play at the time the engine fails ...
There will be a little residual thrust, but most of it is being absorbed by the rotating mass of the engine (Newton's Third Law again).
we probably might be interested in having your expanded discussion on this point ?
Whatever thrust there is, is highly unlikely to be enough to counter drag and inertia sufficiently to produce a significant speed increase.
the numbers will vary across different Types .. Centaurus' figures reflect my recollections of the 737 ..
Think about it - if you launched down the runway at say, 25% thrust, would you ever get to V1? Probably not
I think I'd agree with that .. but the analogy essentially is irrelevant to the discussion
This is how the aircraft is certified
I would recommend a read of AC 25-7A first
I recognise that EFATO drills in a GA twin have a high risk factor.
lots of risks for heavy iron as well .. however, that statement is the start of doing some sensible risk assessment and mitigation when it comes to in-aircraft local proficiency training sessions ..
Vmca demos should and are demonstrated at a safe height
where they cease to be valid (other than for generic training value) and expose the aircraft to the risks of stalling and spinning ... for the life of me I can see little reason for Vmca sandpit playing beyond the initial multiengine endorsement .. and, even then, why not just stay away from Vmca ? A sensible quasi demo (not of Vmca but something approaching the sort of problems relevant to a static Vmca departure) at a sensible height has some validity for reinforcing the briefing but, even then, is of questionable value
it was not necessary to increase thrust during a simulated engine out T/O
or even a real failure .. and, then, be very wary of simply shoving the throttles up .. lest you be caught out with a thrust overshoot and the potential for a yaw departure if at min weight and min speed schedule. I can recall at least one fatal where this was postulated at the time (by me, as it happened) as a contributory cause.
the F27 would never be certified if it was introduced as a new type
the same can be said for just about any dated Type ..
it leaves the shaft because
.. of aerodynamic forces associated with its motion and, subsequently, gyroscopic precession once it is launched into independent flight ... never a nice thing to contemplate ...
when there is absolutely no power being applied to it
thrust is the key, not power ... one uses power to end up producing thrust.
you have invented perpetual motion
I have a suspicion that you really might need to revisit a dissertation on perpetual motion ?
I'm not really trying to be overtly difficult and a PITA here .. but the reality is that we spend most of our training time in comparatively benign areas of the operating envelope ... any fool/monkey can handle an overspeed takeoff failure at V1 .. in my view, ALL pilots should be exposed to the edge of this particular area of the envelope if for no other reason than to engender a very healthy respect for how hard an aircraft can bite if it's not treated with the healthy respect it demands ...
Grandpa Aerotart
However, if you take an earth based reference, there is a progressive reduction in acceleration (during which the aircraft is still increasing its speed relative to the reference frame) eventually ending in a deceleration and the speed reduces .. is this not really a vital consideration for the discussion ?
Moderator
.. a wee glass of Amarula, followed by one of Wyanga Port .. or the other way around if that's what it takes to float your boat ... and one's literary success is assured .... both fine drops, if I may so say myself... only problem is that the former was far cheaper in RSA than it is in Oz ... memories of fine days gone by ..
.. at least, in the written word, one can go slowly enough not to get excessively tongue-tied and tripped up.
.. at least, in the written word, one can go slowly enough not to get excessively tongue-tied and tripped up.
Join Date: Sep 2006
Location: back of the crew bus
Posts: 1,312
Likes: 0
Received 0 Likes
on
0 Posts
OK well it may be as well to leave the discussion on acceleration to one side, as it isn't entirely relevant and in any case, depends on the framework you are using. I was looking at it from a purely physics-based viewpoint (rate of change of velocity), you seem to be looking at it from a purely mechanical standpoint (ie change of velocity). So unless you want to explore it further, I suggest we put that one away for now.
[edit... on second thoughts...]
We may (possibly) be talking about two different things. There is obviously an increase in speed (and acceleration) from the failure point to the recognition point... however, as soon as you close/retard the thrust levers, you have removed the source of energy that is used to produce thrust. Now what I am waiting for you to explain to me, is how an object, being acted upon by a tiny amount of thrust and a whole heap of drag, can continue to increase it's speed, other than for a fraction of a second? Irrespective of how you define acceleration, there is simply no way that an object that is subject to an overwhelming retarding force (drag) and virtually no thrust, can do so. Please explain how that is possible. It is like saying that if you accelerate to 100km/h in your car and then take your foot off the throttle, the car continues to increase speed. it simply doesn't - inertia cannot produce a positive acceleration. By all means prove me wrong, I am hanging out for an explanation! Even your graph agrees with me.
Unless, of course, we are using different definitions of acceleration... that might explain the confusion. I am using the standard physics definition (rate of change of velocity).
Fixed and variable position error, hysteresis, software error in EFIS systems, sometimes ADC-related errors, for example incorrect temperature inputs where the temp probe has been sitting in direct sunlight.
Hmmm well we always trained our guys to assume that they would have a failure and be ready, and in all my years conducting line checks I never found anyone not taking that seriously... although it is obviously possible. Not familiar with A/L 42, is that an Aussie thing?
For the next 20 or so points, it seems to me that we are seeing a different philosophy in both training and operation between the JAA system (with which I am familiar) and the Australian system (which you are familiar with). I spent a lot of hours in the F27 sim trying to get it to depart from controlled flight on departure, because that was a training emphasis that we had a after the Prestwick J32 accident. In many hours of experimentation, we came to the conclusion that you would have to seriously mis-handle the departure to get into the sort of trouble you are describing. Part of that may be that we opted to round our speeds up a few knots, and make everything generally safer, even if it did cost us a few kilos on marginal runways. So, for example, our low-weight speeds would always be a few knots higher than strictly necessary to enhance safety. What is the point of performing a low-weight takeoff at minimum speeds? There isn't one, so why do it?
Sure, if you have already departed in yaw there is not much you can do to save the day, but our training concentrated on never getting into that position in the first place. We used speeds that simply didn't put you in the position where Vmca/Vmcg could be limiting.
All your examples seem to be predicated on gross mis-handling at low weights and minimum speeds... and while that is certainly possible, no prudent operator is going to operate the aircraft that way, nor train that way. Firstly, because it isn't sensible, and secondly, because it isn't necessary. You can kill yourself in an aircraft in a variety of ways, and you simply can't legislate for all of them... but if you operate it sensibly and with a little airmanship, there is no reason for that to happen.
Anyway, back to answering your points...
But surely the point is that, if you fly it correctly, you WON'T depart in yaw in the first place?
...or maybe certification in Australia is different to certification in Europe... dunno.
My understanding of Euro certification process is that it takes account of all adverse factors to a level that can be expected 90% of the time (to put it really, really simply... I don't particularly want to start quoting broad swathes of EASA manuals). Maybe it is different in Oz - how about you give us a real-world example of how you can get into the min speed schedule/Vmca/Vmcg problems you are suggesting?
From memory, inertia is proportional to mass... might be wrong though.
Maybe, but the certification requirements are different... I doubt the performance reserves of a Navajo would be considered adequate for a 737 (proportionally of course).
That has nothing to do with acceleration, and everything to do with momentum and inertia. The car isn't accelerating towards you, it is slowing down (ie accelerating in the other direction).
The frame of reference is a simple kinematic one... about as earth-based as you can get.
What that graph shows is a continued acceleration from the point of failure to the point of recognition of the failure, at which point acceleration peaks... and then reduces. If we assume the point of recognition is slightly before the point at which the thrust levers are retarded (which it always will be), it is clear that from that point onwards, acceleration is decreasing. In other words, there is no evidence at all from that graph that speed increases past the point of recognition.
Sure. It takes energy to rotate the engine... most of the ground idle fuel flow is dedicated to simply keeping the turbine rotating at around 50% to ensure a self-sustaining cycle. There is very little actual thrust being produced. maybe I worded it badly... how about "there is very little residual thrust, as most of the energy required to produce thrust is being absorbed by the rotating mass of the engine".
Yeah but the rest of that paragraph is very relevant:
"Think about it - if you launched down the runway at say, 25% thrust, would you ever get to V1? Probably not... and an idling engine is probably producing no more than 10% thrust. There is no way that an engine that is only receiving enough fuel to produce 10% thrust, is going to accelerate the aircraft it is bolted to at anywhere near the levels of drag it will be experiencing near V1."
And the the "aerodynamic forces associated with it's motion" come from where? That's right... inertia. Which is what I think I said.
Power + propeller = thrust...
Could have fooled me... 
Sure... but ONLY in the sim, or at a safe altitude. Nobody is against exploring the dark corners of an aircraft's performance envelope, in fact most of us enjoy it... well, I do, anyway.
But better yet... operate the aircraft in such a way that a pilot never needs to even come close to finding out the limits the hard way... prevention being always better than the cure...
[edit... on second thoughts...]
if we take the failure point acceleration as reference then, of course, there is a subsequent deceleration. However, if you take an earth based reference, there is a progressive reduction in acceleration (during which the aircraft is still increasing its speed relative to the reference frame) eventually ending in a deceleration and the speed reduces .. is this not really a vital consideration for the discussion ?
Unless, of course, we are using different definitions of acceleration... that might explain the confusion. I am using the standard physics definition (rate of change of velocity).
perhaps you might list the variety of reasons (for airspeed errors) to which you refer ?
Reality on the line is quite different and startle factor has to be contended with .. it is for this sort of consideration that A/L 42 was introduced
For the next 20 or so points, it seems to me that we are seeing a different philosophy in both training and operation between the JAA system (with which I am familiar) and the Australian system (which you are familiar with). I spent a lot of hours in the F27 sim trying to get it to depart from controlled flight on departure, because that was a training emphasis that we had a after the Prestwick J32 accident. In many hours of experimentation, we came to the conclusion that you would have to seriously mis-handle the departure to get into the sort of trouble you are describing. Part of that may be that we opted to round our speeds up a few knots, and make everything generally safer, even if it did cost us a few kilos on marginal runways. So, for example, our low-weight speeds would always be a few knots higher than strictly necessary to enhance safety. What is the point of performing a low-weight takeoff at minimum speeds? There isn't one, so why do it?
Sure, if you have already departed in yaw there is not much you can do to save the day, but our training concentrated on never getting into that position in the first place. We used speeds that simply didn't put you in the position where Vmca/Vmcg could be limiting.
All your examples seem to be predicated on gross mis-handling at low weights and minimum speeds... and while that is certainly possible, no prudent operator is going to operate the aircraft that way, nor train that way. Firstly, because it isn't sensible, and secondly, because it isn't necessary. You can kill yourself in an aircraft in a variety of ways, and you simply can't legislate for all of them... but if you operate it sensibly and with a little airmanship, there is no reason for that to happen.
Anyway, back to answering your points...
now, I had three years of great fun on the Friendly .. but it, the same as most other aircraft, WON'T accelerate through the problem area if, for whatever reason, the aircraft has already departed in yaw ... keep thrust on and you just wind yourself into a tighter ball in the weeds ...
as before, for a min weight, min speed schedule .. if the real world Vmca is significantly higher ... then you are in the middle of it all. Perhaps you are putting far too much of your trust in the "guaranteed" transference of the artificial world of certification into the real dirty world of day to day operations ?
My understanding of Euro certification process is that it takes account of all adverse factors to a level that can be expected 90% of the time (to put it really, really simply... I don't particularly want to start quoting broad swathes of EASA manuals). Maybe it is different in Oz - how about you give us a real-world example of how you can get into the min speed schedule/Vmca/Vmcg problems you are suggesting?
inertia is more of a factor in heavies
I don't think so .. just changes the numbers ..
I don't think so .. just changes the numbers ..
GA twins are far more marginal than most transport-category turboprops.
not at all .. make the numbers appropriate and you can have just as miserable day in a heavy as on a puddlejumper
not at all .. make the numbers appropriate and you can have just as miserable day in a heavy as on a puddlejumper
to cite a somewhat silly analogy .. consider a motor car .. the driver turns off the engine (no fuel flow now). You are standing in the middle of the road 50 odd metres away in the path of the car ... outcome ? I think I'd opt for exiting stage left pronto ?
as soon as the rate of increase of velocity peaks, you have deceleration (by definition)
depending on your frame of reference .. we are more concerned about earth based frames of reference, I suggest.
depending on your frame of reference .. we are more concerned about earth based frames of reference, I suggest.
scan of a page from the Boeing Performance Engineers Training Manual. Cursory review infers the failed engine progressive thrust rundown following Vef.
There will be a little residual thrust, but most of it is being absorbed by the rotating mass of the engine (Newton's Third Law again).
we probably might be interested in having your expanded discussion on this point ?
we probably might be interested in having your expanded discussion on this point ?
Think about it - if you launched down the runway at say, 25% thrust, would you ever get to V1? Probably not
I think I'd agree with that .. but the analogy essentially is irrelevant to the discussion
I think I'd agree with that .. but the analogy essentially is irrelevant to the discussion
"Think about it - if you launched down the runway at say, 25% thrust, would you ever get to V1? Probably not... and an idling engine is probably producing no more than 10% thrust. There is no way that an engine that is only receiving enough fuel to produce 10% thrust, is going to accelerate the aircraft it is bolted to at anywhere near the levels of drag it will be experiencing near V1."
it leaves the shaft because
.. of aerodynamic forces associated with its motion and, subsequently, gyroscopic precession once it is launched into independent flight ... never a nice thing to contemplate ...
.. of aerodynamic forces associated with its motion and, subsequently, gyroscopic precession once it is launched into independent flight ... never a nice thing to contemplate ...
thrust is the key, not power ... one uses power to end up producing thrust.
I'm not really trying to be overtly difficult and a PITA here ..
but the reality is that we spend most of our training time in comparatively benign areas of the operating envelope ... any fool/monkey can handle an overspeed takeoff failure at V1 .. in my view, ALL pilots should be exposed to the edge of this particular area of the envelope if for no other reason than to engender a very healthy respect for how hard an aircraft can bite if it's not treated with the healthy respect it demands ...
But better yet... operate the aircraft in such a way that a pilot never needs to even come close to finding out the limits the hard way... prevention being always better than the cure...
Last edited by remoak; 29th Mar 2010 at 14:08.
Moderator
Now what I am waiting for you to explain to me, is how an object, being acted upon by a tiny amount of thrust and a whole heap of drag, can continue to increase it's speed
I don't know what sort of aircraft your experience is on .. the reality is that, especially for high bypass fans, you cannot cause the thrust to disappear instantly. The engine runs down .. and it just can't stop producing thrust. The rundown is faster in the case of a failure compared to a throttle chop but nonetheless associated with a very observable and finite time delay.
If you don't accept that, that's fine .. but your reticence doesn't change the real world fact of engine rundown ...
During the rundown period, the thrust progressively reduces, resulting in a speed overrun with a hump such as shown generically in the Boeing sketch. Just the way the real world works, I'm afraid.
By the time that the rundown has finished, or nearly so, the residual thrust is either low/negative and your statement is quite correct. However, it is during the intervening period of rundown where the thrust is comparatively high and this results in the speed overrun.
It is like saying that if you accelerate to 100km/h in your car and then take your foot off the throttle
then let's follow your analogy .. while you are taking your foot off the throttle, the car WILL continue to accelerate with the characteristics relating to the vehicle's mass and the engine rundown characteristic. At some point during that foot movement (rundown), the engine power (think thrust line at the driving tyres) will reduce to a point where the net acceleration is negative (deceleration) and the speed trend reverses with the car starting to slow down .. is this not pretty well exactly what happens in the aircraft .. replacing foot movement by a throttle chop ? The only difference of note is that the rundown characteristics for throttle chop and failure vary with the latter being somewhat more rapid .. but you still get significant thrust in the intial stages of that rundown while the fan slows down. Keep in mind that the blades don't know whether they are being driven by the hot end or are just slowing down .. so long as the air is coming over the blades then we have the production of lift forces resulting in thrust for a measurable time interval.
inertia cannot produce a positive acceleration
.. but continued (slowing) motion consequent to inertial effects certainly can continue to produce forces which can produce such acceleration ... for a period.
Even your graph agrees with me
I don't think so .. perhaps you might expand on this suggestion ?
Unless, of course, we are using different definitions of acceleration
we are quite consistent .. I suggest that your confusion arises from different frames of reference. The end result is the same for the aircraft .. with a speed increase for a bit coming over the hump ...
Fixed and variable position error, hysteresis, software error in EFIS systems, sometimes ADC-related errors, for example incorrect temperature inputs where the temp probe has been sitting in direct sunlight.
.. and these only become significant when you suffer an engine failure during the takeoff ? I don't really think so.
Indeed, how do you show any causal relationship between an engine failure and such errors ?
In any case, the altimeter accuracy requirements, via the static interconnection, result in typical maximum ASI errors of, say, 4-5kt ... all rather independent, however, of engine failure.
Not familiar with A/L 42, is that an Aussie thing?
The FAR amendment which introduced the accel-stop 2 second delay. If you go back far enough in the local Industry, you will recall the equivalent Oz ANO 101.6 A/L 62 which introduced the parallel Oz requirement.
I spent a lot of hours in the F27 sim trying to get it to depart from controlled flight
that's because the sim does not replicate the dynamics outside the programmed (tested) envelope. If you have been a bit slower, and depending on the fidelity characteristics, you may/may not have seen something representative. As you would be aware, sims are computers, not aeroplanes, and the fidelity only is reasonable in the flight test validated envelope (for the sim). For example, I used to play in a 732 sim pre- and post- the FAA mandated mod to introduce the updated rudder model. Pre-mod the low speed characteristics were a bit of joke and not even of much use for generic training .. post-mod it was a whole new tiger. Similar considerations apply to whatever F27 sim you cite ie you were trying to get a confused computer to depart .. not the aircraft.
Part of that may be that we opted to round our speeds up a few knots
.. now, why ever would you have thought to do such a thing ?
What is the point of performing a low-weight takeoff at minimum speeds?
.. generally because the runway is so length critical .. and you want those last few kilos. The more important consideration is that the AFM data is based on idealised circumstances ... if you don't replicate that ideal .. then you can get bitten very badly in this area of the operating envelope .. this is main thrust of the discussion as most pilots have very little understanding of how things work in this (not often played in) sandpit .. which is why old pharts like me and Old Smokey (Mutt and MFS are not quite as decrepit as we are yet but they're working on it) continue to harp on such matters.
Sure, if you have already departed in yaw
thank heavens .. at last ... congruent thought.
there is not much you can do to save the day
Actually you can reduce thrust by bringing the throttles back a bit and, for Vmca, push the nose down to accelerate/crank in a bit more bank .. but you have the main idea ... once you depart, you lose all the other nice things and you either die or do something a bit non-SOP in a valiant attempt to avoid death.
but our training concentrated on never getting into that position in the first place
and more strength to you and your fine attitude, good sir.
We used speeds that simply didn't put you in the position where Vmca/Vmcg could be limiting
fine if you are well above book speeds. The concern in question is real world effects if you are somewhere near book speeds.
All your examples seem to be predicated on gross mis-handling at low weights and minimum speeds
that's only one precedent scenario .. the others are just as important - crosswind, systems failures, etc. Doesn't really matter what sets up the critical scenario. Captain Speaking still has to do sometime at the time to try to resurrect the situation ..
but if you operate it sensibly and with a little airmanship, there is no reason for that to happen.
in the low speed end of the envelope, there is a perceived lack of knowledge amongst the piloting fraternity. The value of this sort of discussion is that some of the problems (which are not generally well understood) are tossed around for the potential learning benefit.
Operating sensibly is fine ... if the aeroplane and the ambient conditions approximate the book story.
if you fly it correctly, you WON'T depart in yaw in the first place?
that is a good starting point .. providing that the other gremlins don't come out and bite you.
or maybe certification in Australia is different to certification in Europe... dunno
the basics of certification are reasonably well harmonised amongst the various Authorities. However, you appear, quite steadfastly, to be rejecting the proposition that the certification is an idealised version of the real world and, if the real world doesn't approximate the ideal .. then you can get bitten and bitten hard.
to a level that can be expected 90% of the time
I don't know that that's a valid proposition .. the basis is on a measure of statistical risk but the concept is similar. I suggest that the certification idealisation gets it reasonably applicable to a far more higher level than you are suggesting ... the concern in the discussion has been the low probability situation which overwhelms the pilot and arrives on Page 1 of the tabloids that evening ..
how about you give us a real-world example of how you can get into the min speed schedule/Vmca/Vmcg problems you are suggesting?
too easy ...
(a) significant adverse crosswind for Vmcg
(b) banking the wrong way for Vmca
(c) systems failures causing higher than certification yaw couples for both
etc...
From memory, inertia is proportional to mass
that's close enough ... the point is that inertia affects any machinery ... the numbers will depend on the specifics.
I doubt the performance reserves of a Navajo would be considered adequate for a 737
of course not .. perhaps my earlier comment was a bit too flippant ?
That has nothing to do with acceleration, and everything to do with momentum and inertia.
my point has been, and perhaps I have been unsuccessful in trying to get it across, we are not overly concerned about acceleration but more about the forces which might follow consequent to the effects of mass and motion ...
The frame of reference is a simple kinematic one... about as earth-based as you can get.
as I am not an expert in the discipline of mechanics I think we probably will just agree to disagree on minor semantics ? However, from an earth based frame aircraft speed continues to increase in the period immediately following the failure due to rundown charateristics ... only for a short period, agreed, but it is a real effect and affects distances etc.
there is no evidence at all from that graph that speed increases past the point of recognition.
the curve shapes to me tell a different story .. best we just agree to disagree. The discussion related to the rundown characteristics .. so the section of interest is post failure rather than post recognition ?
maybe I worded it badly
that's fine .. once the rundown has completed its sequence and you are back at idle (or failed, as the case may be)
the rest of that paragraph is very relevant
not really .. the discussion point centres around the thrust variation during rundown, not the steady state.
That's right... inertia. Which is what I think I said.
the inertia results in the motion .. which is the driver for the forces ... pedantic point with which we probably will go nowhere ...
Nobody is against exploring the dark corners of an aircraft's performance envelope, in fact most of us enjoy it... well, I do, anyway
we are in heated agreement when it comes to training .. far too much training risk has converted into prangs over the years ... the concern is the real world situation when, for reasons often outside the pilot's control, he/she is thrust into a situation which is outside the idealised certification world
But better yet... operate the aircraft in such a way that a pilot never needs to even come close to finding out the limits the hard way... prevention being always better than the cure...
I think this statement highlights our differing points of view .. you are taking the position that, by adherence to AFM procedures and SOP, you are ironclad. That applies MOST of the time but not ALL of the time. Historical reality shows that numerous accidents result from a disjoint between the ideal worlds of certification and SOP and the dirty, unfair, real world.
We should have a beer sometime to discuss philosophy ....
I don't know what sort of aircraft your experience is on .. the reality is that, especially for high bypass fans, you cannot cause the thrust to disappear instantly. The engine runs down .. and it just can't stop producing thrust. The rundown is faster in the case of a failure compared to a throttle chop but nonetheless associated with a very observable and finite time delay.
If you don't accept that, that's fine .. but your reticence doesn't change the real world fact of engine rundown ...
During the rundown period, the thrust progressively reduces, resulting in a speed overrun with a hump such as shown generically in the Boeing sketch. Just the way the real world works, I'm afraid.
By the time that the rundown has finished, or nearly so, the residual thrust is either low/negative and your statement is quite correct. However, it is during the intervening period of rundown where the thrust is comparatively high and this results in the speed overrun.
It is like saying that if you accelerate to 100km/h in your car and then take your foot off the throttle
then let's follow your analogy .. while you are taking your foot off the throttle, the car WILL continue to accelerate with the characteristics relating to the vehicle's mass and the engine rundown characteristic. At some point during that foot movement (rundown), the engine power (think thrust line at the driving tyres) will reduce to a point where the net acceleration is negative (deceleration) and the speed trend reverses with the car starting to slow down .. is this not pretty well exactly what happens in the aircraft .. replacing foot movement by a throttle chop ? The only difference of note is that the rundown characteristics for throttle chop and failure vary with the latter being somewhat more rapid .. but you still get significant thrust in the intial stages of that rundown while the fan slows down. Keep in mind that the blades don't know whether they are being driven by the hot end or are just slowing down .. so long as the air is coming over the blades then we have the production of lift forces resulting in thrust for a measurable time interval.
inertia cannot produce a positive acceleration
.. but continued (slowing) motion consequent to inertial effects certainly can continue to produce forces which can produce such acceleration ... for a period.
Even your graph agrees with me
I don't think so .. perhaps you might expand on this suggestion ?
Unless, of course, we are using different definitions of acceleration
we are quite consistent .. I suggest that your confusion arises from different frames of reference. The end result is the same for the aircraft .. with a speed increase for a bit coming over the hump ...
Fixed and variable position error, hysteresis, software error in EFIS systems, sometimes ADC-related errors, for example incorrect temperature inputs where the temp probe has been sitting in direct sunlight.
.. and these only become significant when you suffer an engine failure during the takeoff ? I don't really think so.
Indeed, how do you show any causal relationship between an engine failure and such errors ?
In any case, the altimeter accuracy requirements, via the static interconnection, result in typical maximum ASI errors of, say, 4-5kt ... all rather independent, however, of engine failure.
Not familiar with A/L 42, is that an Aussie thing?
The FAR amendment which introduced the accel-stop 2 second delay. If you go back far enough in the local Industry, you will recall the equivalent Oz ANO 101.6 A/L 62 which introduced the parallel Oz requirement.
I spent a lot of hours in the F27 sim trying to get it to depart from controlled flight
that's because the sim does not replicate the dynamics outside the programmed (tested) envelope. If you have been a bit slower, and depending on the fidelity characteristics, you may/may not have seen something representative. As you would be aware, sims are computers, not aeroplanes, and the fidelity only is reasonable in the flight test validated envelope (for the sim). For example, I used to play in a 732 sim pre- and post- the FAA mandated mod to introduce the updated rudder model. Pre-mod the low speed characteristics were a bit of joke and not even of much use for generic training .. post-mod it was a whole new tiger. Similar considerations apply to whatever F27 sim you cite ie you were trying to get a confused computer to depart .. not the aircraft.
Part of that may be that we opted to round our speeds up a few knots
.. now, why ever would you have thought to do such a thing ?
What is the point of performing a low-weight takeoff at minimum speeds?
.. generally because the runway is so length critical .. and you want those last few kilos. The more important consideration is that the AFM data is based on idealised circumstances ... if you don't replicate that ideal .. then you can get bitten very badly in this area of the operating envelope .. this is main thrust of the discussion as most pilots have very little understanding of how things work in this (not often played in) sandpit .. which is why old pharts like me and Old Smokey (Mutt and MFS are not quite as decrepit as we are yet but they're working on it) continue to harp on such matters.
Sure, if you have already departed in yaw
thank heavens .. at last ... congruent thought.
there is not much you can do to save the day
Actually you can reduce thrust by bringing the throttles back a bit and, for Vmca, push the nose down to accelerate/crank in a bit more bank .. but you have the main idea ... once you depart, you lose all the other nice things and you either die or do something a bit non-SOP in a valiant attempt to avoid death.
but our training concentrated on never getting into that position in the first place
and more strength to you and your fine attitude, good sir.
We used speeds that simply didn't put you in the position where Vmca/Vmcg could be limiting
fine if you are well above book speeds. The concern in question is real world effects if you are somewhere near book speeds.
All your examples seem to be predicated on gross mis-handling at low weights and minimum speeds
that's only one precedent scenario .. the others are just as important - crosswind, systems failures, etc. Doesn't really matter what sets up the critical scenario. Captain Speaking still has to do sometime at the time to try to resurrect the situation ..
but if you operate it sensibly and with a little airmanship, there is no reason for that to happen.
in the low speed end of the envelope, there is a perceived lack of knowledge amongst the piloting fraternity. The value of this sort of discussion is that some of the problems (which are not generally well understood) are tossed around for the potential learning benefit.
Operating sensibly is fine ... if the aeroplane and the ambient conditions approximate the book story.
if you fly it correctly, you WON'T depart in yaw in the first place?
that is a good starting point .. providing that the other gremlins don't come out and bite you.
or maybe certification in Australia is different to certification in Europe... dunno
the basics of certification are reasonably well harmonised amongst the various Authorities. However, you appear, quite steadfastly, to be rejecting the proposition that the certification is an idealised version of the real world and, if the real world doesn't approximate the ideal .. then you can get bitten and bitten hard.
to a level that can be expected 90% of the time
I don't know that that's a valid proposition .. the basis is on a measure of statistical risk but the concept is similar. I suggest that the certification idealisation gets it reasonably applicable to a far more higher level than you are suggesting ... the concern in the discussion has been the low probability situation which overwhelms the pilot and arrives on Page 1 of the tabloids that evening ..
how about you give us a real-world example of how you can get into the min speed schedule/Vmca/Vmcg problems you are suggesting?
too easy ...
(a) significant adverse crosswind for Vmcg
(b) banking the wrong way for Vmca
(c) systems failures causing higher than certification yaw couples for both
etc...
From memory, inertia is proportional to mass
that's close enough ... the point is that inertia affects any machinery ... the numbers will depend on the specifics.
I doubt the performance reserves of a Navajo would be considered adequate for a 737
of course not .. perhaps my earlier comment was a bit too flippant ?
That has nothing to do with acceleration, and everything to do with momentum and inertia.
my point has been, and perhaps I have been unsuccessful in trying to get it across, we are not overly concerned about acceleration but more about the forces which might follow consequent to the effects of mass and motion ...
The frame of reference is a simple kinematic one... about as earth-based as you can get.
as I am not an expert in the discipline of mechanics I think we probably will just agree to disagree on minor semantics ? However, from an earth based frame aircraft speed continues to increase in the period immediately following the failure due to rundown charateristics ... only for a short period, agreed, but it is a real effect and affects distances etc.
there is no evidence at all from that graph that speed increases past the point of recognition.
the curve shapes to me tell a different story .. best we just agree to disagree. The discussion related to the rundown characteristics .. so the section of interest is post failure rather than post recognition ?
maybe I worded it badly
that's fine .. once the rundown has completed its sequence and you are back at idle (or failed, as the case may be)
the rest of that paragraph is very relevant
not really .. the discussion point centres around the thrust variation during rundown, not the steady state.
That's right... inertia. Which is what I think I said.
the inertia results in the motion .. which is the driver for the forces ... pedantic point with which we probably will go nowhere ...
Nobody is against exploring the dark corners of an aircraft's performance envelope, in fact most of us enjoy it... well, I do, anyway
we are in heated agreement when it comes to training .. far too much training risk has converted into prangs over the years ... the concern is the real world situation when, for reasons often outside the pilot's control, he/she is thrust into a situation which is outside the idealised certification world
But better yet... operate the aircraft in such a way that a pilot never needs to even come close to finding out the limits the hard way... prevention being always better than the cure...
I think this statement highlights our differing points of view .. you are taking the position that, by adherence to AFM procedures and SOP, you are ironclad. That applies MOST of the time but not ALL of the time. Historical reality shows that numerous accidents result from a disjoint between the ideal worlds of certification and SOP and the dirty, unfair, real world.
We should have a beer sometime to discuss philosophy ....
Join Date: Sep 2006
Location: back of the crew bus
Posts: 1,312
Likes: 0
Received 0 Likes
on
0 Posts
you cannot cause the thrust to disappear instantly. The engine runs down .. and it just can't stop producing thrust. The rundown is faster in the case of a failure compared to a throttle chop but nonetheless associated with a very observable and finite time delay.
I am talking about acceleration as defined in kinematics and physics...
"Acceleration is the rate of change of velocity as a function of time. It is vector. In calculus terms, acceleration is the second derivative of position with respect to time or, alternately, the first derivative of the velocity with respect to time".
You are talking about acceleration as it is commonly used ie "an increase in speed".
What you don't seem to be able to see is that when you close the thrust levers, the speed may increase slightly (which I am not in disagreement with you about), but the acceleration (remember: the RATE of CHANGE of velocity) is negative.
We should probably be more careful with our use of the terms "speed" and "acceleration".
the car WILL continue to accelerate with the characteristics relating to the vehicle's mass and the engine rundown characteristic.
.. but continued (slowing) motion consequent to inertial effects certainly can continue to produce forces which can produce such acceleration ... for a period.
Even your graph agrees with me
I don't think so .. perhaps you might expand on this suggestion ?
I don't think so .. perhaps you might expand on this suggestion ?
Fixed and variable position error, hysteresis, software error in EFIS systems, sometimes ADC-related errors, for example incorrect temperature inputs where the temp probe has been sitting in direct sunlight.
.. and these only become significant when you suffer an engine failure during the takeoff ? I don't really think so.
.. and these only become significant when you suffer an engine failure during the takeoff ? I don't really think so.
Indeed, how do you show any causal relationship between an engine failure and such errors ?
The FAR amendment which introduced the accel-stop 2 second delay. If you go back far enough in the local Industry, you will recall the equivalent Oz ANO 101.6 A/L 62 which introduced the parallel Oz requirement.
Similar considerations apply to whatever F27 sim you cite ie you were trying to get a confused computer to depart .. not the aircraft.
However, you appear, quite steadfastly, to be rejecting the proposition that the certification is an idealised version of the real world and, if the real world doesn't approximate the ideal .. then you can get bitten and bitten hard.
The "real world" has to get a pretty long way from the real world on which certification is based to "bite you hard". That is why there are (statistically) so few accidents.
too easy ...
the point is that inertia affects any machinery ... the numbers will depend on the specifics.
perhaps my earlier comment was a bit too flippant ?
we are in heated agreement when it comes to training .. far too much training risk has converted into prangs over the years ... the concern is the real world situation when, for reasons often outside the pilot's control, he/she is thrust into a situation which is outside the idealised certification world
you are taking the position that, by adherence to AFM procedures and SOP, you are ironclad. That applies MOST of the time but not ALL of the time. Historical reality shows that numerous accidents result from a disjoint between the ideal worlds of certification and SOP and the dirty, unfair, real world.
If you are an experienced pilot, and choose to go to the places where dragons live, you shouldn't be surprised if you eventually get singed by their breath.
You seem to be saying that aircraft can quickly kill you and that it is the fault of certification for not adequately protecting you; I am saying that the process is as good as it can reasonably be, and that pilots are required to exercise judgement and airmanship.
Putting that into the context of this accident, I'm more than happy to stick my neck out and say that it will turn out to be an unnecessary and easily preventable accident, because the history of training accidents is full of similar examples.
We should have a beer sometime to discuss philosophy ....
mmmm physics... Remoak sorry mate I disagree..
Remember that the derivative of a graph at any point is the slope of the line, and that acceleration is the derivative of velocity (which in itself is the derivative of position). Thus if at any time the speed (velocity) is increasing, the aircraft is accelerating. When that speed (velocity) reaches its peak, the acceleration is instantaneously zero. As the speed (velocity) starts to deccelerate the aircraft is now deccelerating or has negative acceleration.
To put it even simpler, accelerating to a steady state there will never be negative acceleration, as you can see below...
What you don't seem to be able to see is that when you close the thrust levers, the speed may increase slightly (which I am not in disagreement with you about), but the acceleration (remember: the RATE of CHANGE of velocity) is negative.
To put it even simpler, accelerating to a steady state there will never be negative acceleration, as you can see below...
Join Date: Sep 2006
Location: back of the crew bus
Posts: 1,312
Likes: 0
Received 0 Likes
on
0 Posts
Thus if at any time the speed (velocity) is increasing, the aircraft is accelerating.
Acceleration being a measure the rate of change of velocity, not the change of velocity.
So if we graph speed against time, and have say the following values...
1 sec/10kts, 2 secs/20kts, 3 secs/30 kts, 4 secs 40kts -
...we can say that the change of velocity is 10kts/sec, and the rate of change of velocity is also 10kts/sec.
But if the figures change to, say -
1 sec/10kts, 2 secs/20kts, 3 secs/25 kts, 4 secs 28kts -
... the velocity is still increasing, but the rate of change of velocity is decreasing, which amounts to an acceleration in the opposite direction, or a deceleration relative to the velocity vector.
As the speed (velocity) starts to deccelerate the aircraft is now deccelerating or has negative acceleration.
1 sec/50kts, 2 sec/40kts, 3 sec/30kts, 4 secs/20kts -
... we have a change of velocity and rate of change of velocity of 10kts/sec. Now if those figures become -
1 sec/50kts, 2 sec/40kts, 3 sec/45kts, 4 secs/42kts -
... the velocity is still decreasing, but so is the rate of change of velocity - in other words, an acceleration in the original direction of travel, or, if you like, a negative deceleration.
To put it even simpler, accelerating to a steady state there will never be negative acceleration, as you can see below...
A cr*pload of nasty maths 'n graphs 'n stuff about what comprises acceleration
The analogy and thesis to me is thus: observer and airplane are increasing in speed almost together (travelling in the same direction) but the 'plane is creeping ahead of the observer (ie. it's accelerating away from the obsserver because it's rate of change of speed is greater than the observers, relative also to a third - fixed - observer say). If you cut the throttles that 'plane relative to observer one will begin to get closer (ie. the rate of change will become negative relative to that observer) and the observer will now catch up and rapidly pass the 'plane. But the increase in speed relative to observer two for the 'plane [and obs 1] may continue for a time - albeit the rate of change (read: acceleration) is immediately affected.
I'm not sure if that muddies the water or clears it - or if I've managed to make a total arse of myself
- but all this other stuff was beginning to hurt my head and I just had to try and simplify it for myself...Cheers, P.
Not in physics it isn't.
Acceleration being a measure the rate of change of velocity, not the change of velocity.
... the velocity is still increasing, but the rate of change of velocity is decreasing,
which amounts to an acceleration in the opposite direction, or a deceleration relative to the velocity vector.
alternatively can you please explain how it is possible to accelerate an object as you are proposing (i.e. its average rate of change of velocity with respect to time from t3 to t4 is 3/m/s^2) with an acceleration (which implies a net force) in the opposite direction...
Acceleration, velocity, and Position
Jerks - Third derivative of position
Anyway, to paraphase Forrest Gump, "im sorry for ruining your thread"...
Last edited by flighthappens; 30th Mar 2010 at 11:18. Reason: format & 2r's in Forrest