I think the Trio guys are pretty helpful if the owner wants to make contact with them.
Responsive is one term I suppose, I would call it nervous…
There is an autopilot fitted and when we engaged it at cruise the aircraft oscillated about the pitch axis probably due to the sensitivity.
Would you say that this is normal?
Sure you can adjust to A/P gain though I think you are starting at the wrong end their. An aircraft with an A/P should be able to fly stable from itself. I would try the starter replacement first, or some other weight options to keep you away from the edge of the CG envelope. The basic operation should be fine first.
IMHO you should only adjust the A/P if required, when the aircraft is stable. Changing the A/P gain might also change operation at different speeds.
I wouldn’t spin it.
IMHO you should only adjust the A/P if required, when the aircraft is stable
Not sure I quite agree, as you put it, Jesse.
One “can” achieve “stability” with a normal proportional-only control loop, so long as the function is monotonic i.e. more movement on the control lever always produces more pitch G or more roll rate. Pilot-DAR is the expert here and he can put it much better than me…
And some uncertified planes do not meet the above requirement, which is pretty dangerous.
Some (more) designs don’t have monotonic control forces (more force = more G in pitch or more roll rate) which is not certifiable because it will confuse the hell out of the human brain trying to fly it, but a proportional autopilot should still work (perhaps with a crap transient response) because the is no torque sensing on the servos (except for pitch trim which is unrelated to this) and the servos are much more powerful than is needed.
So, so long as more control travel produces more G in pitch, or more degrees/sec roll rate in roll, and there is basic pitch stability in the airframe, an autopilot should be possible, even if the aircraft is very hard to hand fly.
What I am trying to say is that if an aircraft can be flown by a human, even with great difficulty, it is “stable” enough for an autopilot to be possible.
I would agree that dodgy handling should be fixed first, but how? It is in the airframe design and the W&B envelope.
The control loop can be made more sophisticated than just proportional, to improve transient response, but I don’t think you can solve porpoising in pitch if it is caused by having too much proportional gain. My recollection is that adding e.g. a derivative term only helps to improve transient response if this is poor due to the proportional gain being too low. Then you also add an integral term to get long term altitude stability, heading stability, etc. and every autopilot does that.
IMHO, if an autopilot is unstable, this needs expert input and a lot of flight testing.
There are special cases where e.g. an AI which is used to provide pitch/roll info is outputting too much signal. This can happen with misconfigured EA100 or GAD43 boxes (KI256 emulators) or if using a different AI whose pickoff coils have too much output (I know of one such case).
Thanks for all the input guys!
The local RV community seems to agree that the RVs are “pitch sensitive” and the solution is just to fly more… The comments basically say that at high speed you don’t move the stick, only apply pressure.. I’d also prefer to have the sensitivity sorted out before using the A/P even though it is possible to adjust it as Peter says, I mean, most modern fighters can’t fly without computer control, but this is not a Eurofighter…
In any case it has now been confirmed that RVs are extremely light in pitch at speed. I’ll see what he decides to do, there doesn’t seem to be a fix as such, it’s just the way it is. I guess what really surprised me is the way the aircraft felt and all the praise it gets.
I’ve flown some “aerobatic” and relatively maneuverable aircraft like the Giles, Saab Safir, Zlin 526, Bücker 131 etc and I think they all prove that an aircraft does not have to be super light on the controls to be fun to fly!
most modern fighters can’t fly without computer control, but this is not a Eurofighter…
I think that’s a different thing… on some types, you have control travel reversals. But even the F16 doesn’t actually reverse – it just gets too sensitive, according to this
There are GA homebuilts which have control force reversals (i.e. the force get heavier and then past a certain point in the travel gets lighter again – absolutely not certifiable, and arguably dangerous, but not unflyable) but a complete control travel reversal needs a clever FBW system which is way beyond any GA AP.
There are GA homebuilts which have control force reversals (i.e. the force get heavier and then past a certain point in the travel gets lighter again – absolutely not certifiable, and arguably dangerous, but not unflyable)
The Polish built version of the Rallye which I have flown exhibits an overbalancing of the ailerons whereby they got very much lighter at full deflection.
I think there is a misunderstanding there with regards to the original Falcon. Unlike the RV the problem wasn’t a problem with sensitivity of control surface versus speed, cg or deflection. The problem was that the prototype in its original version did not have any stick deflection so control surface deflection was driven by the force applied to the stick as opposed to angular deflection. This was subsequently updated to include half-inch travel as well as force sensing.
Why is the F16 fly by wire?
Because it is aerodynamlcally unstable, I think, and can only be flown safely with the help of a computer. Without it the risk of departing controlled flight would be too high.
“Relaxed Static Stabilty” they call it.
