I still think Kamov got the AP wrong

[ARM505]

The AP doesn't fight back.

 

Yes, it quite clearly does. Or perhaps we are completely misunderstanding one another.

 

To test:

 

- Start on runway, no wind, lightly loaded helo (or whatever, but no gusts will help)

 

- Take off, get into a stable cruise at (say) 150 km/h

 

- All three AP channels default to on for a runway start, so let's test that condition to start with. With the helo trimmed, ball in the middle, apply back stick pressure. In my case, I could exert a CONSTANT back stick deflection that would cause a pitch up to 10 degress nose up, and then STAY there, despite the constant back stick deflection.

 

- Return to original conditions, or close enough (stable cruise, same speed etc)

 

- Turn off the AP pitch channel.

 

- Apply the same amount of back stick deflection. You will be rewarded with a constant pitch rate, not stopping at 10 degrees (or whatever value the initial test got you).

 

Logically then, the AP pitch channel was providing some other input that FOUGHT (ie went counter to) your input when it was engaged. It was attempting to return to a trimmed pitch attitude. That is what I (and AlphaOneSix I believe, if I understood his point correctly) am talking about when I mention 'fighting'. The Ka50 AP seems to be intended as an aid to stable flight, NOT as a 'pitch hold', 'roll hold' autopilot mode (and I know what these are, having used them daily) - HOWEVER DCS:BS presents them as such. Without the proof I have no leg to stand on - I fully concede that, so it remains a gut feel that it's not quite right in DCS:BS. But the AP does fight back - maybe it's like that in the real Ka50 too.

[EtherealN]

Well, yeah. What the AP is there to do is to stabilize, and when you are trimmed to a certain attitude it'll attempt to hold that attitude. When you want to change attitude you obviously need to change your trim so the AP is informed that you want a new attitude. Therefore you press-and-hold the trim when you initiate the attitude change (or turn) which causes the AP channel to be in standby for a new setting.

 

Whether this is all exactly as it should, well, I dunno, I haven't flown a Kamov IRL. But a proper trimming technique (that is, proper for the simulator whether or not it's proper for RL) means there is no fighting against you even with AP's on.

[Frederf]

There's an extraneous variable in that test, the fact that any AP channel has both the hold function and a stabilization filter. You cannot compare AP channel on vs. AP channel off and say conclusively that it is the AP hold that is fighting you.

 

Sure the AP filters you but it doesn't fight you. I wouldn't expect 1:1 control correlation between AP channel on and AP channel off. What I mean by fight you is that say perhaps you have the AP is set to keep the aircraft pitch 0° while it's trimmed at the center of its +/-20% authority. If you pitch down a small amount the AP doesn't add in its pitch-up authority to try keep its set aircraft attitude. That would be fighting the pilot.

 

The fact that you can stomp madly on the rudder pedals and get the aircraft to wiggle back and forth with the AP channels off but not with them on is not a case of fighting the pilot but merely filtering him. Now if you're saying the stabilization filters are unnecessarily harsh then fair enough. I don't have a good idea of how they work.

 

Well, yeah. What the AP is there to do is to stabilize, and when you are trimmed to a certain attitude it'll attempt to hold that attitude.

To be more exact: "when you are trimmer to a certain attitude it'll produce an input designed to achieve that attitude in the absence of pilot control input" which is not exactly the same thing as the AP using its full authority regardless of pilot control input.

 

I'm pretty sure the "official" real world Ka-50 trimming technique is not press-hold-maneuver-release but instead maneuver-press-briefly. I believe this is the case because the trimmer button, one pressed, releases all trim forces on the stick and is unnecessarily tiring or lacking in feedback for the pilot. Because most of us do not have FFB sticks (or perhaps the FFB sticks don't replicate this RL behavior) we don't appreciate this behavior.

Edited December 22, 2009 by Frederf

[ARM505]

There's an extraneous variable in that test, the fact that any AP channel has both the hold function and a stabilization filter. You cannot compare AP channel on vs. AP channel off and say conclusively that it is the AP hold that is fighting you.

 

Well...what else is doing it then? I'm not interested in which component is fighting me, merely that when the switch is on, I have to fight surely?

 

Sure the AP filters you but it doesn't fight you. I wouldn't expect 1:1 control correlation between AP channel on and AP channel off. What I mean by fight you is that say perhaps you have the AP is set to keep the aircraft pitch 0° while it's trimmed at the center of its +/-20% authority. If you pitch down a small amount the AP doesn't add in its pitch-up authority to try keep its set aircraft attitude. That would be fighting the pilot.

 

I believe it does. And yes,it's fighting me. It's exactly that AP applied force that keeps the aircraft from pitching at a constant pitch rate with the AP channel engaged in the test example I gave above. It's balancing my small control deflection in an attempt to regain it's trimmed attitude.

 

The fact that you can stomp madly on the rudder pedals and get the aircraft to wiggle back and forth with the AP channels off but not with them on is not a case of fighting the pilot but merely filtering him. Now if you're saying the stabilization filters are unnecessarily harsh then fair enough. I don't have a good idea of how they work.

 

Nope, I'm referring to the force applied when I apply a constant force, ie constant pedal deflection will result in a constant AP input in the opposite direction - it's fighting me!

 

 

 

To be more exact: "when you are trimmer to a certain attitude it'll produce an input designed to achieve that attitude in the absence of pilot control input" which is not exactly the same thing as the AP using its full authority regardless of pilot control input.

 

That's the thing - it produces that input in the PRESENCE of pilot input. THAT'S my problem with it. (And Alpha's too, again if we're on the same wavelength.)

[herr_LaSk]

Its been a while since I played Black Shark, but I think the people who complain about fighting the autopilot is somewhat missing the point. When you use the flight directors, or the autopilot is off, the flight controlls controll rate of roll/pitch, just like in a cessna 172. However when using standard autopilot settings the flight controlls changes to controll angle of roll/pitch, so the amount of stick deflection corelates to the bank and pitch angle of the chopper. Move the stick two degrees to the right, the helicopter banks a certain amount to the right and then holds that bank angle. Move the stick furder to the right and the helicopters bank angle increases the coresponding amount and holds that bank angle. Release the stick and the heli rolls back to the trimmed roll angle.

 

Its designed to work this way to make it less of an effort to fly for the pilot. With a standard configuration the pilot, in order to initiate a right turn, will first have to move the stick to the right in order to initiate a certain amount of roll rate, until the desired bank angle is reached at which point the pilot releases pressure on the stick in order to maintaing that bank ange. With the Ka 50 autopilot system, all the pilot has to do is move the stick the amount to the right which he knows from experience coresponds to the desired bank angle, hold the stick at that position (and trim if its a long turn) and the heli turns to the right. The first option requires the pilot to consentrate on several aspects of flying in order to initiate the desired turning manuver, while in the second option he only has to move the stick a certain amount and the heli turns just like he wants.

[Frederf]

That's the thing - it produces that input in the PRESENCE of pilot input.

 

Hmm, mine doesn't. Maybe Kamov can ship you a new Ka-50?

[astrospud]

Just a little off topic, I think the FD and Hover Mode should have been two ends of a 3 way toggle switch and not independantly operated.

Toggle up for FD, centre for off and down for HM.

I mean, who flys with both on?

[Frederf]

They are, you can watch the switch on the cyclic flick back and forth. ED only put in commands for the extremes D and R though. It would make those of us trying to fly with HOTAS sticks a lot happier if there was a key corresponding to the neutral position. The ridiculous hoops I have to go through to program a 3 position switch to replicate the real switch's behavior.

[ARM505]

Hmm, mine doesn't. Maybe Kamov can ship you a new Ka-50?

 

Ok, I'm not sure how to phrase this any clearer...

 

So, when you apply a constant (small) stick deflection, with the AP channels engaged, the helo keeps a constant pitch/roll/yaw rate? It doesn't stop at a certain point beyond the original attitude? ie You move your stick, say, five degrees forward, and the helo pitches nose down until it ends up trying to go vertically downwards? It doesn't stop at, say, ten degrees nose down? Because mine clearly, repeatedly, does. Therefore, the AP channel is resisting my attempts to create a constant pitch/roll rate when I move the stick. Turn off the AP channel, and this behaviour stops. Enter FD mode, and it stops. The way I see it, theres only one conclusion.

 

This post:

 

...snip... When you use the flight directors, or the autopilot is off, the flight controlls controll rate of roll/pitch, just like in a cessna 172. However when using standard autopilot settings the flight controlls changes to controll angle of roll/pitch, so the amount of stick deflection corelates to the bank and pitch angle of the chopper. Move the stick two degrees to the right, the helicopter banks a certain amount to the right and then holds that bank angle. Move the stick furder to the right and the helicopters bank angle increases the coresponding amount and holds that bank angle. Release the stick and the heli rolls back to the trimmed roll angle.

 

Its designed to work this way to make it less of an effort to fly for the pilot. With a standard configuration the pilot, in order to initiate a right turn, will first have to move the stick to the right in order to initiate a certain amount of roll rate, until the desired bank angle is reached at which point the pilot releases pressure on the stick in order to maintaing that bank ange. With the Ka 50 autopilot system, all the pilot has to do is move the stick the amount to the right which he knows from experience coresponds to the desired bank angle, hold the stick at that position (and trim if its a long turn) and the heli turns to the right. ..snip.

 

..is exactly what I'm talking about. I don't think it's really meant to be that way. It's counter intuitive. Even Airbus (knocked by Boeing pilots for years for it's FBW despite it being perfectly good) doesn't do it this way - a certain stick force or deflection produces a certain roll/pitch rate, not a certain attitude. Sure, the FCC moderates the amount of aileron input as airspeed varies to keep that roll/pitch rate the same across various conditions - but the point is, it doesn't try to change the fundamentals of how a pilot learned to control his aircraft. I can think of no aircraft that intentionally does this (not to say the Ka50 doesn't, but that's my point - I don't think it does.)

Edited December 23, 2009 by ARM505

[Bucic]

Ok, I'm not sure how to phrase this any clearer...

 

So, when you apply a constant (small) stick deflection, with the AP channels engaged, the helo keeps a constant pitch/roll/yaw rate? It doesn't stop at a certain point beyond the original attitude? ie You move your stick, say, five degrees forward, and the helo pitches nose down until it ends up trying to go vertically downwards? It doesn't stop at, say, ten degrees nose down? Because mine clearly, repeatedly, does. Therefore, the AP channel is resisting my attempts to create a constant pitch/roll rate when I move the stick. Turn off the AP channel, and this behaviour stops. Enter FD mode, and it stops. The way I see it, theres only one conclusion.

 

This post:

 

 

 

..is exactly what I'm talking about. I don't think it's really meant to be that way. It's counter intuitive. Even Airbus (knocked by Boeing pilots for years for it's FBW despite it being perfectly good) doesn't do it this way - a certain stick force or deflection produces a certain roll/pitch rate, not a certain attitude. Sure, the FCC moderates the amount of aileron input as airspeed varies to keep that roll/pitch rate the same across various conditions - but the point is, it doesn't try to change the fundamentals of how a pilot learned to control his aircraft. I can think of no aircraft that intentionally does this (not to say the Ka50 doesn't, but that's my point - I don't think it does.)

Very interesting. I'm looking forward to see ED response.

[therion_prime]

In missions without long straight lines between waypoints, I often find myself holding down the trim button during almost the whole mission (exception: hover, straight flight).

Try and fly through a narrow valley where you have to manouver constantly, you simply have to hold down the trim button most of the time.

 

I don't quite get this. Why would kamov build such a thing where the pilot has to press a button while manouvering? I simply could hold the trim button down with a piece of tape .....

[AlphaOneSix]

I don't quite get this. Why would kamov build such a thing where the pilot has to press a button while manouvering? I simply could hold the trim button down with a piece of tape .....

 

It's not just Kamov...

 

ALL modern helicopters are like this, from ALL manufacturers.

 

(EDIT: I am sure someone can find some modern helo without a force trim system, but the vast majority have a force trim system.)

[ericinexile]

Well A16, we've had this discussion for two years now and it looks like we have come to an agreement. The most basic precept of autoflight logic is a bit like the Hippocratic Oath: "Above all else, Do No Harm." That means the autopilot should never get in the pilot's way. It is easy to design a system to recognize cyclic input and accept that input as something the pilot desires as opposed to something that needs correction...which is what DCS does within its 20% control authority.

 

As a pilot, it is so counter-intuitive to me that normal procedure would be to fly with a system that is applying control inputs, no matter how light, counter to my own. Yes, I know of and have flown FBW and SAS, that's not what I am talking about here. The autopilot isn't stabilizing flight when it applies those inputs to maintain attitude or heading. Rather, it is saying one of two things, ONE, "I have no way of distinguishing pilot input from an outside disturbance", or TWO, "Pilot, I know you just told me you want to bank left but your last trim input locked me into this attitude or this heading and I am going to fight you to the limits of my authority!" Neither makes sense in a modern aircraft. What does make sense is an autoflight system that, once it senses cyclic movement, makes no input whatsoever until that input ceases (other than stability augmentation functions).

 

Fortunately, for those of use who find those uncalled for inputs irritating, there is the FD which we choose to use, albeit incorrectly. Or you can fly with the autopilot equally well by learning to work around its limited authority. Both techniques work.

 

Meantime, I'm with A16, my near-certain hunch is that the logic is not correctly inplemented. But until a Kamov engineer takes time out of his busy life to post on this forum, none of us will know for sure.

[therion_prime]

It's not just Kamov...

 

ALL modern helicopters are like this, from ALL manufacturers.

 

(EDIT: I am sure someone can find some modern helo without a force trim system, but the vast majority have a force trim system.)

 

While I can understand this system is adequate for most helicopters, it's tedious in the Ka-50 (being the only single seat attack helicopter).

Everything else is built to ease the workload of the pilot, but during every slight attitude change he has to hold down a button?

 

Come on! It has a complex autopilot that can follow waypoints and can auto hover while I can have the feet on the dashboard and sip on a Pina Colada, but requires my input for the simplest of things. I just don't get it.

 

Maybe this was the reason the Ka-50 project was stopped. :D

[ericinexile]

...Come on! It has a complex autopilot that can follow waypoints and can auto hover while I can have the feet on the dashboard and sip on a Pina Colada, but requires my input for the simplest of things. I just don't get it.

 

Maybe this was the reason the Ka-50 project was stopped. :D

 

 

I may have missed a post but I don't think you and A16 are in disagreement. He is correct in stating that the force trim button is a common and logical way to communicate to the AFS the you wish to hold current attitude, speed, heading, or hover point depending on the situation. The question is, and you both agree on the question, does the real Ka-50 actually require the application of the force trim button to temporarily stop AP inputs? You both intuitively feel that it should not. And I happen to agree.

[159th_Viper]

Try and fly through a narrow valley where you have to manouver constantly, you simply have to hold down the trim button most of the time.

 

I don't quite get this. Why would kamov build such a thing where the pilot has to press a button while manouvering? I simply could hold the trim button down with a piece of tape .....

 

Trim Often, Yes. Hold the Button down? Personally have never had to resort to this extreme......:joystick:

 

Post the Track/Mission where you are Obliged to - would like to have a Looksee :)

[liotczik]

I was used to (simulated) civilian helos without any autopilot, trim or stabilisation systems, with tailrotor (steering cross-coupling) and I've learned flying them the hard way. Learn to hover default FS9 or XP9 JetRanger on a gusty day or fly it manually for 5 straight hours in turbulence and only then try to complain how much Ka-50 is "unfriendly aircraft" :D At first I was shocked how Ka-50 was at the same time pefectly stable and easy to hold in one place, while a real pain to maneuver.

 

But after some time I've "adjusted" myself to its characteristics, exactly as was advised by many sources, only to find it is a very agile and pleasant aircraft to fly. It's easier to control than Cessna 172 :) The key to success is to understand how the machine works and to work along with her. It's all about flying the way the helicopter does want to, instead of making the helicopter fly as we want it to do.

 

To the point. I fly with all 4 AP buttons, for the most of the time without FD. When I want to change speed, heading or flightpath, I simply press the dreaded trim button, then manually stabilise helo on new attitude, and release trim button when I'm happy with the results. Simple and effective way of combining easy cruising with high combat/aerobatic agility. For the most of the time it works really good and (now) intuitive. Often I find cruising well-trimmed helo more useful, than enroute AP, because I'm not limited to only 6 INS waypoints. For that few occasions I can't constantly hold trim, I engage FD and throw the crate all over the sky with ease.

 

For everyone tired of pressing one button, I have the following thoughts.

 

First, when flying correctly, with a plan in your head, you barely need to perform dramatic maneuvers, so there is no real need to hold the trimmer all the time. When attacking, set up carefully and the rest is easy. When dodging that MANPAD, you really need only two buttons: trim and flares, and since you have two hands, you've got all you need in such situation (it can be performed even on a simple stick with at least two buttons, no need for HOTAS right away, just a little bit of imagination when configuring the buttons).

 

Second, my real-life aviation manual says about proper flight technique: stabilise attitude and speed and always trim - then re-trim and re-trim and re-trim to maintain that attitude. After major change in flight conditions, like shift from climb to cruise, stabilise in new attitude, trim and re-trim and re-trim and re-trim to maintain that new attitude. And so on. Trim even on traffic pattern and approach. Trim while holding. Always trim - it's a kind of mantra. The point is that you need only quick press to stabilise/correct aircraft attitude and then you release and let the helo fly itself. Again, no need to hold it for a long time. Want to perform aerobatics -> turn FD on.

 

Third. Come on, you've bought a serious simulation (not a weekend game!) with a manual of over 500 pages and now you complain about pressing one button? Maybe PacMan would be a better choice - no buttons to press at all :P Or try arcade mode. Correct me if I'm wrong, but as far as I remember, there is a similar "panic button" on F-16. It's a pinkie switch, which works as a temporary AP disable, to correct maintained attitude, as well as a mean to save the aircraft, if AP tries to do something weird while flying at low level in terrain following mode. I recall no one complaining about it - instead of keeping it pressed, you disengage AP and fly manually. It's that simple.

 

Fourth, like was said earlier, most of the helicopters with force trim work like that. Press to change attitude and release to maintain it or use a "chinese hat" - it depends on which system was installed on a particular helo. ED made a fantastic job of simulating power timming method on desktop josticks. Beat it - Ka-50 (and any other helo) is very different from any jetfighter. Old tricks doesn't work here, you need to learn completely new set of skills, but in the end that's exactly, what flying various sims is about. To learn and better understand.

 

Of course no one will get mad, if you fly it whatever way you like, instead of trying new things.:joystick: I've tried flying without even touching trim button, without any and all of 4 AP channels and so on and I have to say, that it's simply a no-go. Unrealistic, tiring and dangerous for your virtual-self ;)

[EtherealN]

Everything else is built to ease the workload of the pilot, but during every slight attitude change he has to hold down a button?

 

Not "everything else", it's "everything including this". If utilization of your pinky finger will cause irreparable fatigue then I don't want to know what will happen through the need to actually move your wrist when giving cyclic input. ;)

 

Come on! It has a complex autopilot that can follow waypoints and can auto hover while I can have the feet on the dashboard and sip on a Pina Colada, but requires my input for the simplest of things. I just don't get it.

 

Oh come on. You trim when changing attitude (or when outside parameters cause the previous trim to not be sufficient), which is something you do with your feet on the pedals and hands on cyclic and collective. At which point you utilize one of your fingers for trimming while performing those maneuvers.

 

It "requires [your] input" because we are talking about manual flight. I find the solution perfect - when I am looking to maneuver I'll have my hands on the stick and be able to use one of my fingers to adjust my trim. I'd much rather have manual trim than rely on a computer to read my mind. ;)

[ericinexile]

This notion that pilots of large helos trim every few seconds is nonsense. Pilots tap the force trim when they want the magnets to hold the stick in the new position because that new position is fairly close to where the plan on keeping the stick for awhile. Let's say we are in a stable hover (FD on, AP off) and the helo is trimmed. Now we want to translate left 10 feet. Nobody in that situation is going to apply a nudge of left stick and trim. Instead he is going to nudge left and let the helo coast to the new position, then nudge right to stop the movement. The helo is now in a stable and trimmed hover in the new location without ever touching the trim. The problem with FD Off (AP On) is that you MUST trim because otherwise the helicopter is going to apply cyclic counter to your desired sideward flight. THAT is what is so nonsensical (is that a word?) about the AP system as simulated in DCS. Where constant tapping of the trim occurs is during acceleration where the position of the stick is gradually moving farther forward and left (in the case of a Ka50 in forward flight). Without continuous trimming the pilot is slowly turning himself into a body-builder. But once the ship is stable, trimming stops and it shouldn't be needed if the pilot maneuvers with intent to return to the old, trimmed attitude and speed over a short time period. Constant blipping of the trim is the DCS pilot's way of telling the autopilot to "mind your own business". The other way is the Flight Director.

[Frederf]

So, when you apply a constant (small) stick deflection, with the AP channels engaged, the helo keeps a constant pitch/roll/yaw rate? It doesn't stop at a certain point beyond the original attitude?

 

I do not get this behavior. Pitch stabilizes at a certain attitude with the AP channel on or off. It's a different attitude but it always stabilizes. You might not have found the stabilization point with a large pitch change and a rapidly increasing airspeed but it is there. The idea of a "constant rate of change" input in pitch is completely alien to me, sure initially but not forever. Every Cessna I've ever flown has behaved quite similar to the Ka-50. You push the yoke, the pitch decreases, the airspeed increases, axial forces on the airplane rejigger, and the aircraft stabilizes at a new pitch. The same happens in bank unless you exceed the limitations of the zone of positive aerodynamic stability but I've never taken a C172 beyond 60-70° of bank.

 

I ran a test where I limited my X/Y joystick deadzones to produce at maximum deflection 20% input. I tried three states: Hold+Stabilization, Stabilization Only (FD), and Channel OFF.

 

Investigating pitch channel. At 1000m, 200kmph straight and level I pushed full forward on the cyclic (-20%):

H+S: -10° pitch stabilize

SO: -20° pitch stabilize

COFF: -20° pitch stabilize

 

Which makes it seem like somehow the stability program has more influence when the hold's authority is in effect. Comparing the states where the stabilization only was on and the channel was completely off showed nearly identical behavior except for the persistent oscillations when the channel was off.

 

At no point did the AP "fight me" by adding in control input to get back to a held value. The word fighting suggests that contrary and reactionary input contests my own. The word fighting does not suggest that not all of my input was delivered to the control surfaces. If that were to happen, "fighting" would not be the right word, it would be "partial compliance."

 

Again, the setup above (20% saturation on stick) shows to anyone easily how this fighting is a myth:

"Pilot, I know you just told me you want to bank left but your last trim input locked me into this attitude or this heading and I am going to fight you to the limits of my authority!"

Despite that it is obviously within the power of the AP's 20% authority to override very small inputs by the pilot, it did not, which I did indeed barely deflecting the joystick when even full deflection only adds up to a measly 20%.

 

If you're complaining that the damper-filter stabilization function dampens or filters your input... then I have to ask what did you expect? That's not fighting you, that's helping you. It's a design decision to increase stability (a.k.a. anti-maneuverability).

[ARM505]

I ran a test where I limited my X/Y joystick deadzones to produce at maximum deflection 20% input. I tried three states: Hold+Stabilization, Stabilization Only (FD), and Channel OFF.

 

Investigating pitch channel. At 1000m, 200kmph straight and level I pushed full forward on the cyclic (-20%):

H+S: -10° pitch stabilize

SO: -20° pitch stabilize

COFF: -20° pitch stabilize

 

Which makes it seem like somehow the stability program has more influence when the hold's authority is in effect. Comparing the states where the stabilization only was on and the channel was completely off showed nearly identical behavior except for the persistent oscillations when the channel was off.

 

At no point did the AP "fight me" by adding in control input to get back to a held value. The word fighting suggests that contrary and reactionary input contests my own. The word fighting does not suggest that not all of my input was delivered to the control surfaces. If that were to happen, "fighting" would not be the right word, it would be "partial compliance."

 

Again, the setup above (20% saturation on stick) shows to anyone easily how this fighting is a myth: Despite that it is obviously within the power of the AP's 20% authority to override very small inputs by the pilot, it did not, which I did indeed barely deflecting the joystick when even full deflection only adds up to a measly 20%.

 

If you're complaining that the damper-filter stabilization function dampens or filters your input... then I have to ask what did you expect? That's not fighting you, that's helping you. It's a design decision to increase stability (a.k.a. anti-maneuverability).

 

Well, I don't want to take issue with only your posts, but surely you've kind of proved the effect here? I mean, the aircraft moved further in COFF and SO than H+S? This is exactly what I'm getting at. And (I'm taking a flyer here), but maybe the effect would be even more pronounced at lower airspeeds, where the aerodynamic damping and straightening effects of the fuselage and it's horizontal and vertical stabilisers tended to resist change, ie reduce that 'contant pitch rate'? (I am aware that no control input results in a pitch/roll rate ad infinitum btw, obviously as conditions change, so does the rate). Perhaps when you refer to 'partial compliance' (sounds like a term the manufacturer would use when defending their product!), I would only ever call it 'fighting' (ie it resisted me). I know if it happened in an aircraft I was flying, I would snag it as 'fighting'.

 

I simply don't see how you can call it a myth. Anyway, I've seen enough verification and descriptions of exactly what I'm talking about in this thread.

[Frederf]

I mean the myth that if the AP is scheduled to hold an attitude, the pilot moves the cyclic, and the AP overcomes (partially or fully) to return the aircraft to the scheduled attitude. That is a myth. Clearly the AP pitch hold exercising 0% authority and trimmed for level flight would ramp up its authority (in this case to +11%) in the positive direction to counteract a -11% cyclic input (for example) and return the aircraft to the 000° pitch attitude. This is easily testable and is shown not to be what happens. This is the classic "AP is fighting me" argument but not the one you are proposing which has proven to be more confusing.

 

The idea that control authority is lessened and/or filtered by the AP is certainly valid. Partly it is expected that the stabilization routine will reduce control input in amplitude as well as reduce the rate of its onset. Pilot commands 100% RIGHT NOW and the stabilization gives 70% over the next little while. That's what it's supposed to do and how it achieves its purpose.

 

I tried some other tests with 100% rudder authority in a flight regime where weathervaning limited yaw to <90°. With the heading channel on or off I was able to achieve the same max yaw when starting from the same airspeed (250kmph or so) and applying rapid, full yaw in either direction. It would seem in that case that full input is not hampered by either the stabilization or stab/hold cases as I tried it in all three states to identical results in max yaw angle. I wouldn't be surprised if it took slightly longer for the stabilized maneuver to reach max yaw but it was difficult to detect.

 

Why I got differences in max negative pitch between Stab Only and Stab+Hold modes with my 20% stick setup I don't know. I would really like to see the scripts and/or coding that controls this behavior or at least a genuine design paper about the real-life system to get an idea of how the stabilization, holds, and inputs all combine in theory.

Edited December 24, 2009 by Frederf

[ericinexile]

Frederf,

 

You simply will not listen to evidence that counter your own convictions. In fact you don't even seem to fully trust your own experimentation. The evidence of AFS inputs counter to those desired by the pilot is there for you to witness in a 30 second flight: With flight director on, lift off to a stable hover, accelerate forward to 100 kps, turn 360 degrees, and decelerate back to a hover. Perform this little flight agressively--not as you would your 172 but as you would in a high powered killing machine. Now, do the same with the FD off. See the differece? If you do then to what do you attribute the difference? Stability? No, that's covered by the AFS dampening channels and is independent of the autopilot. Did you NEED to trim more with the FD off? Why? Finally (and ask yourself this honestly): Which mode was easier?

Edited December 25, 2009 by ericinexile

[ARM505]

Ok Frederf, I think I see why you resist the term 'fighting', since it seems that to you it would mean that the AP channel actually 'won' the fight, and produced a result totally opposite to pilot input. Did I understand you right? Because that's not what I mean by 'fighting'. To me, that means that not all of my control input was allowed to be exerted, ie the AP took something away from me, and I don't mean filtered the result (ie an AP moderated rate of onset), I mean genuinely exerted itself against my input and ultimately reaching a kind of 'stalemate' (regardless of the fact that I 'won' the fight, and the helo did ultimately move in the direction I told it to, abeit not at my commanded rate, and eventually stopping the commanded roll/pitch)

 

What I'm trying to ultimately say, is that FD mode feels more like the way it should be all the time - watching the RL Kamov videos doesn't show the FD HUD mode. I shouldn't have to press the trim button to be allowed to move the helo - the AP channels should take cyclic pressure/deflection commanded by the pilot into account when stabilising and not attempt to counter this. It's just a gut feel, ED may have it totally correct. In which case, Kamov, what were you thinking?! :)

 

Happy Christmas in any case!

[AlphaOneSix]

Until we get force trim feel springs in joysticks (real ones, not the wimpy ones we see today) then nobody is ever going to really understand why you need to press the trim button to move the controls.

 

NOTE: I'm being sarcastic. The force trim feel springs only exist in the actual aircraft, so until someone gets a set of them and installs them in a pit, it's going to be hard to describe it.

 

NOTE 2: I'll describe in more detail the issue as it relates to an Mi-17, but since that requires research, I must be away momentarily to do Christmas-like things. ;)

Edited December 25, 2009 by AlphaOneSix