King Air down at Essendon?
Look for the most simple possibility. Say for some reason the Kingair rudder trim was offset before the pilot entered the aircraft. Accept that why it was offset is not determined. I haven't flown a King Air but would an offset rudder trim show up while taxiing? In other words is there a nose wheel steering connection like a 737?
A 737-400 crashed during a botched abort in USA and the rudder trim (electrically operated from the cockpit) was found to be full scale one side. The captain was taxiing but the F/O was doing the takeoff and soon discovered the aircraft was pulling to one side during the early part of the takeoff roll so he applied corrective rudder but to no effect. The captain took over and delayed aborting initially. When he finally aborted his procedure was incorrect and the aircraft went off the end into a river.
The point here was with the 737 the fully offset rudder trim causes the rudder pedal to move by about three inches. This because there is limited nosewheel steering availibility with the rudders. The pull to one side taxiing is immediately noticeable. To correct for this, the captain uses his nosewheel steering wheel to keep the aircraft straight while taxiing. In other words similar to crossed controls.
In the case of the 737 accident there was CVR evidence the captain did not mention to the F/O there was something strange happening that required the captain to use excessive nose wheel steering to keep the aircraft tracking straight while taxiing.
Unless the F/O just happened to be keeping his own feet on the rudder pedals while the captain was taxiing (very unlikely) he would not have seen the split rudder pedal position on his side since it would have been only a couple of inches difference.
Once lined up for takeoff and when the captain handed over control to the F/O for take off, the aircraft would have tended initially to pull to one side which the F/O would have corrected for by rudder.
As the aircraft picked up speed during the takeoff roll it gets more difficult to keep straight. When the captain in the 737 realised the F/O was having a problem with directional control and took over control the aircraft was well off the centreline.
So that was what happened to the 737 according to the NTSB accident report. We replicated that in the simulator. As mentioned earlier I have not flown a Kingair so know nothing about its taxiing characteristics with its rudder trim fully offset. But if its rudder pedal steering system is tied into the rudder trim system, then this would have been obvious when taxiing as it was in the 737. That being so it would have been unwise to takeoff.
A 737-400 crashed during a botched abort in USA and the rudder trim (electrically operated from the cockpit) was found to be full scale one side. The captain was taxiing but the F/O was doing the takeoff and soon discovered the aircraft was pulling to one side during the early part of the takeoff roll so he applied corrective rudder but to no effect. The captain took over and delayed aborting initially. When he finally aborted his procedure was incorrect and the aircraft went off the end into a river.
The point here was with the 737 the fully offset rudder trim causes the rudder pedal to move by about three inches. This because there is limited nosewheel steering availibility with the rudders. The pull to one side taxiing is immediately noticeable. To correct for this, the captain uses his nosewheel steering wheel to keep the aircraft straight while taxiing. In other words similar to crossed controls.
In the case of the 737 accident there was CVR evidence the captain did not mention to the F/O there was something strange happening that required the captain to use excessive nose wheel steering to keep the aircraft tracking straight while taxiing.
Unless the F/O just happened to be keeping his own feet on the rudder pedals while the captain was taxiing (very unlikely) he would not have seen the split rudder pedal position on his side since it would have been only a couple of inches difference.
Once lined up for takeoff and when the captain handed over control to the F/O for take off, the aircraft would have tended initially to pull to one side which the F/O would have corrected for by rudder.
As the aircraft picked up speed during the takeoff roll it gets more difficult to keep straight. When the captain in the 737 realised the F/O was having a problem with directional control and took over control the aircraft was well off the centreline.
So that was what happened to the 737 according to the NTSB accident report. We replicated that in the simulator. As mentioned earlier I have not flown a Kingair so know nothing about its taxiing characteristics with its rudder trim fully offset. But if its rudder pedal steering system is tied into the rudder trim system, then this would have been obvious when taxiing as it was in the 737. That being so it would have been unwise to takeoff.
Join Date: May 2000
Location: Australia
Posts: 344
Likes: 0
Received 0 Likes
on
0 Posts
Centaurus,
There is no interconnection between rudder trim setting and nose wheel on a King Air. There is a connection between rudder pedal and nose wheel which you would expect. It is my belief that you could input sufficient force to overcome fully deflected rudder trim during the take-off roll and the initial climb long enough to recognise and correct. That is if there was no other contributing factor like an engine failure. The point I have been trying to make all along is that regardless of the trim setting the aircraft would have climbed like a homesick angel, even with the gear still extended, while both engines operated normally. The accident aircraft didn't perform just as it does not when one engine rolls back with PLM. The other point I should make is when the power lever migrates rearward it doesn't go back to flight idle but to a point where its still producing about 600 ft lbs of torque however the prop will disc up and cause significant drag along with associated performance reduction normally resulting in about a 300-500 ft/min descent rate with gear extended.
Groggy
There is no interconnection between rudder trim setting and nose wheel on a King Air. There is a connection between rudder pedal and nose wheel which you would expect. It is my belief that you could input sufficient force to overcome fully deflected rudder trim during the take-off roll and the initial climb long enough to recognise and correct. That is if there was no other contributing factor like an engine failure. The point I have been trying to make all along is that regardless of the trim setting the aircraft would have climbed like a homesick angel, even with the gear still extended, while both engines operated normally. The accident aircraft didn't perform just as it does not when one engine rolls back with PLM. The other point I should make is when the power lever migrates rearward it doesn't go back to flight idle but to a point where its still producing about 600 ft lbs of torque however the prop will disc up and cause significant drag along with associated performance reduction normally resulting in about a 300-500 ft/min descent rate with gear extended.
Groggy
Thanks grogmonster.
Pending ‘approval’ of your attachment, I anticipate that it will show that the breakage of one cable in the rudder trim system with simultaneous tension put on another in that system will drive the rudder trim in one direction. I wish others luck in telling us what actually happens to those cables during a crash event.
Even so....
It remains odd that PLM would account for all of the ‘veer left then prang’ incidents, unless:
(1) it’s just the product of random chance (eventually there’ll be a bunch of ‘veer right then prang’ incidents due PLM (or the rudder trim being inadvertently being set, unnoticed, at full right); or
(2) there’s a design characteristic that means PLM usually only results in the left power lever migrating.
Any thoughts on #2?
Pending ‘approval’ of your attachment, I anticipate that it will show that the breakage of one cable in the rudder trim system with simultaneous tension put on another in that system will drive the rudder trim in one direction. I wish others luck in telling us what actually happens to those cables during a crash event.
Even so....
It remains odd that PLM would account for all of the ‘veer left then prang’ incidents, unless:
(1) it’s just the product of random chance (eventually there’ll be a bunch of ‘veer right then prang’ incidents due PLM (or the rudder trim being inadvertently being set, unnoticed, at full right); or
(2) there’s a design characteristic that means PLM usually only results in the left power lever migrating.
Any thoughts on #2?
Thanks grogmonster.
Pending ‘approval’ of your attachment, I anticipate that it will show that the breakage of one cable in the rudder trim system with simultaneous tension put on another in that system will drive the rudder trim in one direction. I wish others luck in telling us what actually happens to those cables during a crash event.
Even so....
It remains odd that PLM would account for all of the ‘veer left then prang’ incidents, unless:
(1) it’s just the product of random chance (eventually there’ll be a bunch of ‘veer right then prang’ incidents due PLM (or the rudder trim being inadvertently being set, unnoticed, at full right); or
(2) there’s a design characteristic that means PLM usually only results in the left power lever migrating.
Any thoughts on #2?
Pending ‘approval’ of your attachment, I anticipate that it will show that the breakage of one cable in the rudder trim system with simultaneous tension put on another in that system will drive the rudder trim in one direction. I wish others luck in telling us what actually happens to those cables during a crash event.
Even so....
It remains odd that PLM would account for all of the ‘veer left then prang’ incidents, unless:
(1) it’s just the product of random chance (eventually there’ll be a bunch of ‘veer right then prang’ incidents due PLM (or the rudder trim being inadvertently being set, unnoticed, at full right); or
(2) there’s a design characteristic that means PLM usually only results in the left power lever migrating.
Any thoughts on #2?
shorter cable run run to the left engine, less internal friction on the cable, than compared to right.
thats was a lot of people in the king air community believe.
While I would not totally discount the power lever migration theory, I find it hard to believe experienced pilots would get sucked in by it. Most light aircraft drivers keep one hand on the throttles during takeoff and initial climb, only momentarily removing the hand long enough to retract the gear. The hand goes back on the levers until flap retraction and other non essential activities such as switching off lights. Initiating gear retraction would take 3 seconds at the most - hardly enough time for the power to go right back to negative thrust, and even if it did it would be naturally corrected.
A pilot with an airline background could be in the habit of removing hand from levers at V1, but a pilot with that discipline would probably retract the gear then go back immediately to the power levers if something did not feel right. I don’t think this pilot had any airline training, so would expect he would have kept his hand on the throttles throughout takeoff and rotation.
Either way, I would think it instinctive to push both levers forward if performance was being degraded. Auto feather would take care of a major power loss, but only if the system was armed. If the system was not armed, control with total loss of power close to V1 is very difficult, in fact probably beyond the capability of pilots who have not practiced it in the simulator. We do know that this pilot was not a believer in regular simulator.
The pilot was in a state of panic - that’s really as much as we know. But whether brought on by a medical episode or not is unlikely to ever be discoverable, due to the fireball.
A pilot with an airline background could be in the habit of removing hand from levers at V1, but a pilot with that discipline would probably retract the gear then go back immediately to the power levers if something did not feel right. I don’t think this pilot had any airline training, so would expect he would have kept his hand on the throttles throughout takeoff and rotation.
Either way, I would think it instinctive to push both levers forward if performance was being degraded. Auto feather would take care of a major power loss, but only if the system was armed. If the system was not armed, control with total loss of power close to V1 is very difficult, in fact probably beyond the capability of pilots who have not practiced it in the simulator. We do know that this pilot was not a believer in regular simulator.
The pilot was in a state of panic - that’s really as much as we know. But whether brought on by a medical episode or not is unlikely to ever be discoverable, due to the fireball.
Last edited by Mach E Avelli; 31st Dec 2020 at 20:52.
As Centauras says, look for something simple.
Somehow before the flight the trim was moved to full deflection, during maintenance or preflight or kids in the cockpit or otherwise.
Pilot misses this due to interrupted pre-flight (as per video evidence) or lack of checklist (anecdotal evidence of this) or otherwise.
Power is applied, takeoff roll, control problems start, boost is off, pilot has never seen this before and the rest is history.
It’s a simple scenario that doesn’t require a complex sequence of events to transpire in order to explain, such as one specific trim cable breaking in the crash and then another being stretched or pulled straight after and the rudder trim being moved.
I can’t help but feel that a cable with a jack screw would need a firm and consistent pull to rotate the thread, possibly requiring both cables to activate, rather than a short sharp pull on a single cable, as the impact forces pulled on it abruptly. (Awaiting attachment approval above).
For a rollback scenario or friction issue, you’ve got too many things that add up to get to that end result, such as an engine failure / power loss AND the trim cable scenario has occurred? Plausible maybe, but unlikely... yes.
In option 1 above, only one thing has happened to cause the end result, not multiple things, working backwards from the end result.
Somehow before the flight the trim was moved to full deflection, during maintenance or preflight or kids in the cockpit or otherwise.
Pilot misses this due to interrupted pre-flight (as per video evidence) or lack of checklist (anecdotal evidence of this) or otherwise.
Power is applied, takeoff roll, control problems start, boost is off, pilot has never seen this before and the rest is history.
It’s a simple scenario that doesn’t require a complex sequence of events to transpire in order to explain, such as one specific trim cable breaking in the crash and then another being stretched or pulled straight after and the rudder trim being moved.
I can’t help but feel that a cable with a jack screw would need a firm and consistent pull to rotate the thread, possibly requiring both cables to activate, rather than a short sharp pull on a single cable, as the impact forces pulled on it abruptly. (Awaiting attachment approval above).
For a rollback scenario or friction issue, you’ve got too many things that add up to get to that end result, such as an engine failure / power loss AND the trim cable scenario has occurred? Plausible maybe, but unlikely... yes.
In option 1 above, only one thing has happened to cause the end result, not multiple things, working backwards from the end result.
roll down the runway, airborne, hand off lever to raise the gear... and this happens.
hardly enough time for the lever to go back?
now imagine that you grab the control wheel with that hand you took off the power lever to raise the gear, rather than putting it back on the power levers... because all of a sudden you have a large yaw that requires control input to help manage...
and some panic thinking “engine failure”...
not such an implausible theory as S7700 might imply.
once again, I’m not saying this was the cause of this accident.
its hard to investigate and prove as being the cause; look up the report for VP-BBK, that’s got a good write up on this scenario.
It’s plausible, but doesn’t explain why the trim was all the way to the left. Sometimes the most obvious answer is staring you in the face.
Once the diagram is available, you’ll likely see that both cables are required to turn the jack-screw and will be able to discount that happening afterwards. It must require two cables as Conned has suggested, as otherwise if a cable broke in flight and you were turning the trim, you wouldn’t be able to turn it back again.
Once the diagram is available, you’ll likely see that both cables are required to turn the jack-screw and will be able to discount that happening afterwards. It must require two cables as Conned has suggested, as otherwise if a cable broke in flight and you were turning the trim, you wouldn’t be able to turn it back again.
Out of those posting, who's actually flown the scenarios in the Sim? So far as I can tell, grogmonster is the only one.
I have personally experienced PLM in a King Air and I have seen it happen to other pilots. I have also simulated the Essendon scenario in the simulator and even though I knew what was coming ended up over the top of the same building. The ATSB have made a massive mistake which in their defence I can understand because they don't have experience on the type. I stick to my comments in my previous post.
The report states that the trim was to the left at initial impact due to witness marks on the trim tab. That sounds quite definitive and goes to the extent of discussing the abrasive surface that it came in contact with.
It feels like a long bow to draw to suggest that the cable pulled full left trim during the impact sequence as you’d have to think that it would be tearing out of the trim actuator so damn fast it would likely jam up or go past the stop or at least bend something in the process.
Remember that the report says that they started looking at the trim mechanism because a witness saw the aircraft yawing significantly to the left.
It feels like a long bow to draw to suggest that the cable pulled full left trim during the impact sequence as you’d have to think that it would be tearing out of the trim actuator so damn fast it would likely jam up or go past the stop or at least bend something in the process.
Remember that the report says that they started looking at the trim mechanism because a witness saw the aircraft yawing significantly to the left.
I'll take that as a 'no' from you, S7700.
Im only talking about the power lever moving, in generic terms, not in any specific relation to this event and the trim position.
however, if we want to get into combinations, another plausible scenario is, IF he did have the power lever drop back, did he incorrectly wind the rudder trim the wrong way in the heat of the moment?
We will never know. But it’s not an impossible scenario that the trim gets wound the wrong way.
i would have liked the power lever issue to be investigated/reported more, but without a working CVR from the accident flight a spectral analysis cannot be done to find out (unlike the VP-BBK crash).
LB, unfortunately I don’t get to have fun in the king air any more, so I’ve not been able to access a sim or aircraft to look at things.
however, I have personally experienced the left power come back whilst sitting in both seats.
As Mach has mentioned, shouldn’t be an issue (and it wasn’t) because if you put your hand straight back into the power levers, you’ll sort it out quick smart.
Clearly you know or are fairly sure I haven’t and I’m presuming you haven’t either, but that doesn’t mean I can’t engage in the conversation by analysing the information available, particularly around human behaviour.
Aircraft haven’t changed much in many years. They still have 3 axis, two or four wings generally, including a rudder, elevator, ailerons and an engine or two, three or four.
Human behaviour on the other hand does change, through training, procedures, SOP’s, competency, awareness, tiredness, health and other factors, which are never consistent.
Start looking at the end result here and regardless of what caused it, what would have been done to avoid it. If it was a rollback, what can be learnt from it? If it was engine related, it doesn’t change the outcome other than a witch-hunt to the manufacturer or maintenance organisation, but what does matter is how the event is handled.
Aircraft haven’t changed much in many years. They still have 3 axis, two or four wings generally, including a rudder, elevator, ailerons and an engine or two, three or four.
Human behaviour on the other hand does change, through training, procedures, SOP’s, competency, awareness, tiredness, health and other factors, which are never consistent.
Start looking at the end result here and regardless of what caused it, what would have been done to avoid it. If it was a rollback, what can be learnt from it? If it was engine related, it doesn’t change the outcome other than a witch-hunt to the manufacturer or maintenance organisation, but what does matter is how the event is handled.
The friction would have to be very loose for the lever to run back as quickly as it did in the video. Also the horn would be hard to ignore. Not saying impossible, merely improbable given the crash evidence of the mis set trim. In the very short time he was airborne it is doubtful that full rudder trim could have been wound on - it is fairly slow. Again, not impossible but I believe improbable given that he was yelling out "Mayday" repeatedly at the same time.
I have flown several scenarios in the King Air sim on more than one occasion, including power lever cut back at rotate, engine failure with no auto-feather, FCU over-torque, rudder boost switched off with engine failure, and spurious rudder boost. Knowing what was coming, all were flyable, though the engine failure with no auto-feather took some practice to make it climb. None of the events caused loss of directional control and none required both hands on the control wheel.
I don't recall trying the full rudder trim scenario (but may have; simply don't remember?), but those who have report it gets very ugly as speed increases toward 'blue line'.
However, the old King Air simulator in Melbourne is only a level B, so the fidelity may not be a true reflection of how the real aeroplane would handle in those various situations.
Those who have flown the level D simulator may be able to offer more comment.
I have flown several scenarios in the King Air sim on more than one occasion, including power lever cut back at rotate, engine failure with no auto-feather, FCU over-torque, rudder boost switched off with engine failure, and spurious rudder boost. Knowing what was coming, all were flyable, though the engine failure with no auto-feather took some practice to make it climb. None of the events caused loss of directional control and none required both hands on the control wheel.
I don't recall trying the full rudder trim scenario (but may have; simply don't remember?), but those who have report it gets very ugly as speed increases toward 'blue line'.
However, the old King Air simulator in Melbourne is only a level B, so the fidelity may not be a true reflection of how the real aeroplane would handle in those various situations.
Those who have flown the level D simulator may be able to offer more comment.
Last edited by Mach E Avelli; 1st Jan 2021 at 01:03.
The friction would have to be very loose for the lever to run back as quickly as it did in the video. Also the horn would be hard to ignore. Not saying impossible, merely improbable given the crash evidence of the mis set trim. In the very short time he was airborne it is doubtful that full rudder trim could have been wound on - it is fairly slow. Again, not impossible but I believe improbable given that he was yelling out "Mayday" repeatedly at the same time.
as for the gear horn, you won’t get it with one lever back and the gear still down, as in the accident case gear was down, the video I posted gear was up.
speed of trim? I could spin it to one side pretty quick, probably from neutral to full one side in 1-2 seconds without difficulty, but that’s me. Not a normal rate of using it though, mind you, but if I had to I could that quick.
I genuinely don't want to offend or get personal, S7700, but do really mean what you said?
How do you come up with sensible ideas to avoid an end result, if you don't know the cause/s of the end result? It seems to me that that which should be done to avoid the risks arising from e.g. an unnoticed rudder trimmed to FSD is not the same as that which should be done to avoid the risks arising from e.g. unnoticed power level migration. And if there was a failure which no competent pilot could adequately overcome to save the aircraft, what then?
I'd suggest something different than if it was unnoticed rudder trim at FSD on take off. The first question that would arise is: What causes PLM in this type of aircraft? Should it happen in the first place?
There's a difference between a system design flaw or defect and the proper avoidance response to that problem, on the one hand, and a pilot's failure to use the system properly to prevent or reverse the migration, and the proper avoidance response to that problem, on the other.
On my interpretation of what grogmonster has posted, unnoticed PLM on this aircraft type is a far more dangerous situation than unnoticed rudder trim at FSD, with the former rather than the latter more likely to have resulted in the aircraft ending up where it did.
Goodo, then. ATSB, CASA and us 'Monday morning quarterbacks' can confine our witch-hunts to pilots.
One wonders why they bothered grounding the Max 8s. Any competent pilot should easily be able to handle the consequences of MCAS activation on erroneous input from an AOA sensor.
Start looking at the end result here and regardless of what caused it, what would have been done to avoid it. ...
If it was a rollback, what can be learnt from it?
There's a difference between a system design flaw or defect and the proper avoidance response to that problem, on the one hand, and a pilot's failure to use the system properly to prevent or reverse the migration, and the proper avoidance response to that problem, on the other.
On my interpretation of what grogmonster has posted, unnoticed PLM on this aircraft type is a far more dangerous situation than unnoticed rudder trim at FSD, with the former rather than the latter more likely to have resulted in the aircraft ending up where it did.
If it was engine related, it doesn’t change the outcome other than a witch-hunt to the manufacturer or maintenance organisation, but what does matter is how the event is handled.
One wonders why they bothered grounding the Max 8s. Any competent pilot should easily be able to handle the consequences of MCAS activation on erroneous input from an AOA sensor.
One wonders why they bothered grounding the Max 8s. Any competent pilot should easily be able to handle the consequences of MCAS activation on erroneous input from an AOA sensor.
To answer your question above.. practice, but without significant investment, how does a one-man operation achieve task?
It’s not much different to someone that learns to fly in a C150, then flys a 172 for years, then his business does well and he buys a turbine Mirage to fly his staff around.
Good point. But if the gear was not selected up, it is even less likely that his hand would have been off the power levers, no? Of all the speculative causes, only one thing is certain - rudder trim was found fully displaced to the left and the aircraft went left as soon as it lifted off. So, it's also unlikely the pilot was busy winding trim on while barrelling down the runway. But possible...I suppose.
Hand comes off to grab the gear, migration occurs, yaw happens, possibly think “holy **** engine failure” and grab the control column with that hand. Gear hasn’t actually been selected up because you got distracted by the yaw and grabbed the column.
Now, the proper response is to get your hands back to the power levers and realise that one has migrated. but if you don’t, your now left in a potentially ugly situation.
some places still try to tell king air drivers at what is effectively V1 to take their hand off the levers.
there is NO need for that.