What's new
Van's Air Force

The definitive Van's Aircraft support community! Buying, building or flying an RV? Join our exclusive family of mentors and enthusiasts!

Auto pilot in turbulence ?

P

Python

Active Member
Hi, I watch a youtube Video recently of a guy flying an RV 6 in IFR & for some reason every time he went into cloud ( light stratus) he would disconnect his Autopilot & hand fly ( I guess due to turbulence) ? I do have an instrument rating however I haven’t ever flown an RV in IFR ( I own a VFR RV 6). It looked like his autopilot was a 2 axis trutrak. Is there a reason for this that I don’t know ? To me this is where I would probably want it on ??? Regards 🐍.
 
Last edited:
I've done that also. But, in the RV and non-RV I've owned recently they both were equipped with a Trio Pro Pilot autopilot which allows independent activation of roll and pitch servos. Typically I would leave the roll servo on and hand fly the elevator. I did this because to pitch excursions in the clouds could overwhelm the autopilot at times.
 
I normally just shadow the controls in case it kicks off. If it’s egregious or the weather has proved itself to be rough in clouds I will hand fly it. I use Dynon AP, it is good at correcting but only so much. Even a big turbo prop like a PC-12 will kick off in rough turbulence.
 
I’ve never had any issues with the G3X AP flying through build-ups or light turbulence. But yes I do keep a close eye on things just in case.
 
I often find myself hitting the AP disconnect when flying in Wyoming turbulence. Maybe it's just my G3X setup, but in heavy updrafts the AP tries to maintain its set altitude to the point that it routinely puts my RV-14A's airspeed in the yellow. Same thing happened with my Carbon Cub (also G3X AP). I like to stay well within green arc in turbulence. Throttling back is also an option, but then I'm basically chasing the AP instead of letting it lighten my workload! :p
 
In the old days the theory was that if you hit a strong up or down draft the AP might overstress the plane trying to hold altitude. I once hit some clear air turbulence and my Trio Pro kicked off (g limit), so I’m inclined to think modern APs are pretty safe in this regard.
 
G3x on otto always in turbulence. G3x software flies the plane better than i ever could most of the time. Got 12 computers on board, i dam well going use them
 
I have a GRT autopilot and in most circumstances it flies much better than I can. When I fly in/over the mountains, I've had cases where it hands back control to me suddenly. I now have the habit of just hand flying when in these conditions. All VFR, so maintaining strict altitude and heading is not an issue. Reading about some of these big aircraft having their AP give up at exactly the worst moment is very surprising. What I have not seen is documentation on the exact scenarios where the AP will give up. I have not read every user manual for every AP, so perhaps many do document these limitations.
 
Many years ago when I was just getting my feet wet with complicated airplanes, I used to fly a 340 with the OEM autopilot. My boss/mentor advocated for hand flying in turbulence because he felt like it saved wear & tear on the servos. I seriously doubt that's a thing in this day and age, but I came to like hand flying in turbulence as the "fun" part of the flight.

I still tend to do that. Old habits die hard I guess. As far as just run of the mill stabilize air IMC, then I have no problem letting Otto fly unless I just want the practice. Fun comes into my decision, but so does the thought that hand flying is a perishable skill that needs to be practiced. Current and proficient aren't necessarily the same thing. I may be current, but am I really proficient if all I'm doing is poking buttons 90% of the time?
 
PilotjohnS said:
G3x on otto always in turbulence. G3x software flies the plane better than i ever could most of the time. Got 12 computers on board, i dam well going use them
Click to expand...

Desert Rat said:
Many years ago when I was just getting my feet wet with complicated airplanes, I used to fly a 340 with the OEM autopilot. My boss/mentor advocated for hand flying in turbulence because he felt like it saved wear & tear on the servos. I seriously doubt that's a thing in this day and age, but I came to like hand flying in turbulence as the "fun" part of the flight.

I still tend to do that. Old habits die hard I guess. As far as just run of the mill stabilize air IMC, then I have no problem letting Otto fly unless I just want the practice. Fun comes into my decision, but so does the thought that hand flying is a perishable skill that needs to be practiced. Current and proficient aren't necessarily the same thing. I may be current, but am I really proficient if all I'm doing is poking buttons 90% of the time?
Click to expand...
You can also look at it from the other side, too. If you are in turbulent IMC, the A/P is like having another pilot on board, freeing you to manage everything else, while it keeps the aircraft straight and level. Certainly not saying that you shouldn't hand fly but the autopilot is a tool...which does nothing if it is turned off.

Kind of like 4x4 pickups in the ditch after a snowstorm. "Were you using the four wheel drive?"..."Well, no..."
 
I wish I could leave the autopilot on all the time. The fact is I cannot because it wont come close to holding altitude anywhere near what is required under IFR. Yes, I've tweaked gain settings for years. I've tried different brand autopilots. Just can't get adequate performance to let it do it's thing in any turbulence. So, sadly, the pitch servo goes off in turbulence.
 
FORANE said:
I wish I could leave the autopilot on all the time. The fact is I cannot because it wont come close to holding altitude anywhere near what is required under IFR. Yes, I've tweaked gain settings for years. I've tried different brand autopilots. Just can't get adequate performance to let it do it's thing in any turbulence. So, sadly, the pitch servo goes off in turbulence.
Click to expand...
I think this is not normal. Maybe have someone else evaluate this with you. Might be worth a ride w a friend.JMHO
 
Yeah, that’s strange. My G3X has no issues remaining well within IFR limits. In fact, it’s not even close to the limits.
 
FORANE said:
I wish I could leave the autopilot on all the time. The fact is I cannot because it wont come close to holding altitude anywhere near what is required under IFR. Yes, I've tweaked gain settings for years. I've tried different brand autopilots. Just can't get adequate performance to let it do it's thing in any turbulence. So, sadly, the pitch servo goes off in turbulence.
Click to expand...
If this is in the Lancair not surprising.
 
IFR and turbulence are when the autopilot pays for itself! Yeah in heavy bumps you gotta watch it, but it's still a big help. Oftentimes after a big bump I'll leave the autopilot on but grab the stick and damp out some of it's response to keep it from overspeeding. Another pilot mentioned that if you have a vertical speed mode on your AP set it to zero in turbulence. It AP just nulls out the climb, and doesn't dive the plane back to the held altitude. I haven't tried it but makes sense.
 
Walt said:
If this is in the Lancair not surprising.
Click to expand...
It is, but even in the RV-9a I had some difficulties with the Trio Pro Pilot. In the RV-9a it would hold altitude fine except in rough air. I never was able to get it to climb or descend on autopilot; it would oscillate above then below vertical course in ever increasing amplitude.
 
I use it, but you must be be vigilent in watching what it is doing. In mountain wave induced up drafts, I have had it try to pull me into very low airspeeds.
 
lr172 said:
I use it, but you must be be vigilent in watching what it is doing. In mountain wave induced up drafts, I have had it try to pull me into very low airspeeds.
Click to expand...
I am surprised by the number of people that are having a problem with their AP holding altitude in turbulence. As PilotJohnS mentioned I do not think this is normal. I have a GRT Sport EX EFIS coupled with the GRT AP servos and it seems to be able to do a better job than I can do. Turbulence doesn't seem to be a problem but when in a strong updraft or down draft my IAS will vary as the AP lowers or raises the nose in order to maintain altitude. That is totally normal and short of me messing with the throttle, kind of the way it has to be. I can set the desired "climb to" altitude and it will climb up and level off smoothly and the same with descents. I have the climb parameter set to a certain airspeed (rather than a certain rate of climb) and the decent parameter set to a certain descent rate. That way the airspeed won't get too slow in the climb at higher altitudes. I don't doubt that others are having problems but just surprised. The AP is what makes IFR flight MUCH easier. Obviously I monitor things but so far no problems.

One other thought is that everyone's definition of turbulence is different. I am not sure what mine will do in very severe turbulence that borders on not being able to be controllable but for normal to moderate turbulence mine does a better job than I do. If I am VFR and am encountering large updrafts or downdrafts, I might turn it off and just "accept" the altitude variations and not worry so much about it as things seem to "even themselves out".

Just one more datapoint.
Keith
 
Python said:
Hi, I watch a youtube Video recently of a guy flying an RV 6 in IFR & for some reason every time he went into cloud ( light stratus) he would disconnect his Autopilot & hand fly ( I guess due to turbulence) ? I do have an instrument rating however I haven’t ever flown an RV in IFR ( I own a VFR RV 6). It looked like his autopilot was a 2 axis trutrak. Is there a reason for this that I don’t know ? To me this is where I would probably want it on ??? Regards 🐍.
Click to expand...
It’s fun to hand fly in the clouds.
 
Pilot8 said:
Very good point! 😀. Hand flying in turbulence keeps your mind in the game- Over reliance on the auto pilot isn’t a forgiving mode.
Click to expand...
Under reliance on the autopilot can also be a problem as well…the auto pilot won’t get vertigo or spatial disorientation.

Also, when flying IFR in the conditions being discussed, it is very easy to get task saturated, and that can lead to catastrophic results. The autopilot reduces the likelihood of task saturation. Not using it when necessary is just silly.

Also, if the autopilot won’t stay connected in turbulence less than severe, I’d say there is a problem with it. For severe and greater turbulence, all bets are off, and hopefully the outcome is favorable.
 
FORANE said:
It is, but even in the RV-9a I had some difficulties with the Trio Pro Pilot. In the RV-9a it would hold altitude fine except in rough air. I never was able to get it to climb or descend on autopilot; it would oscillate above then below vertical course in ever increasing amplitude.
Click to expand...
Sounds like a static leak or the config isn't optimized. In any case call the guys at Trio, they will walk you through trouble shooting your Pro Pilot. I have been flying the Pro since doing beta testing for it many years ago and what you are describing is definitely something that can be fixed.
 
BobTurner said:
I do know that they don’t work well with Piper style one piece blade pitot and static probes.
Click to expand...
Bingo
I do have one of those in the lancair. We didn't though in the RV-9a and while it was better it still misbehaved in climb or descent.
 
rocketman1988 said:
Under reliance on the autopilot can also be a problem as well…the auto pilot won’t get vertigo or spatial disorientation.

Also, when flying IFR in the conditions being discussed, it is very easy to get task saturated, and that can lead to catastrophic results. The autopilot reduces the likelihood of task saturation. Not using it when necessary is just silly.

Also, if the autopilot won’t stay connected in turbulence less than severe, I’d say there is a problem with it. For severe and greater turbulence, all bets are off, and hopefully the outcome is favorable.
Click to expand...

I toally agree! I remember a pilot flying practice approaches one weekend where Bay approach finally said in exasperation, "This is IFR you need to maintain the altitudes I assign you", to which he replied "We are trying to figure out the auto pilot". It's important to be proficient. :) Mike Seager says you need an autopilot to safely fly a RV in IFR.
 
Last edited:
rv8ch said:
I have a GRT autopilot and in most circumstances it flies much better than I can. When I fly in/over the mountains, I've had cases where it hands back control to me suddenly. I now have the habit of just hand flying when in these conditions. All VFR, so maintaining strict altitude and heading is not an issue. Reading about some of these big aircraft having their AP give up at exactly the worst moment is very surprising. What I have not seen is documentation on the exact scenarios where the AP will give up. I have not read every user manual for every AP, so perhaps many do document these limitations.
Click to expand...
rv8ch,

The reason working Autopilots give up the ghost is almost always associated with the aircraft being grossly out of trim. If the aircraft is properly trimmed a single autopilot can fly a multiengine aircraft even with an engine out.

Another point for the whole board is the autopilot has NO way of knowing it just went into IMC. This disconnecting a normally reliable autopilot because of entering IMC conditions is unnecessary. In fact, if a VFR pilot inadvertantly enters IMC engaging an autopilot will save his/her bacon in aiding to maintain proper flight attitudes.


If turbulence becomes a concern slow your aircraft to Va and maintain your level flight attitude rather than chasing airspeed or altitude. Just let the airspeed and altitude fluctuate as you maintain that attitude.Notify ATC that due to severe turbulence you are unable to maintain altitude. Usually, they will assign you a large block altitude to clear you while you ride out that turbulence. That is what the Navy taught us about handling inadvertant thunderstorm penetration..

Fly safely out there.

Bob
 
Last edited:
lr172 said:
I use it, but you must be be vigilent in watching what it is doing. In mountain wave induced up drafts, I have had it try to pull me into very low airspeeds.
Click to expand...
In the big iron they knock off the altitude hold from what I have been told in mountain waves.
 
Norman CYYJ said:
In the big iron they knock off the altitude hold from what I have been told in mountain waves.
Click to expand...
I don’t remember turning off altitude hold for moderate or less turbukence in large aircraft. But the autopilot altitude hold and capture on the heavies had altitude capture and hold limits that monitored and limited G loading. The Lockheed Tristar would soften the ride for passenger comfort by routinely accepting small altitude deviations of less than 100 feet during cruise. The MD10/11 during altitude capture would limit g loads and overshoot altitudes slightly to smooth out high v/s altitude captures. We didn’t want to spill the passenger drinks. Boeings had firmer rides and tended to lock on and hold altitudes more firmly but they too limited g during altitude captures. Perhaps some captains turned off altitude hold in moderate or less turbulence but I did not encounter any that I can recall. It was not a policy in any companies I worked for.
 
A bunch of people have mentioned pitch oscillations. When I was first setting up my autopilot I had some issues getting it dialed in. It seemed like if I set teh gain high enough to prevent it from overshooting, any little bump would set off it off into oscillations.

Eventually I came across a post by somebody (Walt maybe?) that mentioned if the trailing edge radius on the elevators isn't right it will cause problems just like it does for the ailerons. Vans doesn't mention this anywhere, but based on how much emphasis they put on getting the aileron trailing edges right it makes perfect sense that if you don't do the same with the elevators it could cause sensitivity issues.

I got out the straight edge and there were several areas where the needed to be massaged into a tighter radius. Once I did that,I was able to go back and play with the gain again and now it's rock solid.
 
rsultzbach said:
If turbulence becomes a concern slow your aircraft to Va and maintain your level flight attitude rather than chasing airspeed or altitude. Just let the airspeed and altitude fluctuate as you maintain that attitude.Notify ATC that due to severe turbulence you are unable to maintain altitude. Usually, they will assign you a large block altitude to clear you while you ride out that turbulence. That is what the Navy taught us about handling inadvertant thunderstorm penetration..

Fly safely out there.

Bob
Click to expand...
Bob raises a good suggestion here and we should all strive to be sure we understand how our APs work. I suspect some would struggle in a tense situation to remember how to get their APs in to pitch and roll hold mode.
 
Last edited:
Desert Rat said:
A bunch of people have mentioned pitch oscillations. When I was first setting up my autopilot I had some issues getting it dialed in. It seemed like if I set teh gain high enough to prevent it from overshooting, any little bump would set off it off into oscillations.

Eventually I came across a post by somebody (Walt maybe?) that mentioned if the trailing edge radius on the elevators isn't right it will cause problems just like it does for the ailerons. Vans doesn't mention this anywhere, but based on how much emphasis they put on getting the aileron trailing edges right it makes perfect sense that if you don't do the same with the elevators it could cause sensitivity issues.

I got out the straight edge and there were several areas where the needed to be massaged into a tighter radius. Once I did that,I was able to go back and play with the gain again and now it's rock solid.
Click to expand...
Actually Vans does mention it in Sect 5, IIRC. A whole area addressing how to ensure that all rolled edge flight surfaces are correctly set. They have other docs discussing heavy wings and repeat it, but it is still in the build manual. Same approach applies to the rudder as well.
 
lr172 said:
Actually Vans does mention it in Sect 5, IIRC. A whole area addressing how to ensure that all rolled edge flight surfaces are correctly set. They have other docs discussing heavy wings and repeat it, but it is still in the build manual. Same approach applies to the rudder as well.
Click to expand...
fair enough. I wasn't attentive enough to notice that section applies to all folded trailing edges, not just ailerons. Anyway, point being if the radius isn't right on your elevators it can cause autopilot pitch sensitivity.
 
Desert Rat said:
fair enough. I wasn't attentive enough to notice that section applies to all folded trailing edges, not just ailerons. Anyway, point being if the radius isn't right on your elevators it can cause autopilot pitch sensitivity.
Click to expand...
Agree 100% I got my rating in the 6. After a couple of flights, I couldn't understand why I couldn't hold an altitude. It was just too responsive. Like you, I had addressed the aileron edges, but not the elevator. I went out and tweaked them to the correct shape and low and behold, the plane was much easier to handle precisely in pitch. OTOH, my AP had been doing fine holding pitch.
 
lr172 said:
... It was just too responsive. Like you, I had addressed the aileron edges, but not the elevator. I went out and tweaked them to the correct shape and low and behold, the plane was much easier to handle precisely in pitch. ...
Click to expand...
Are you trading responsiveness for stability? Could that stability be implemented in the AP servo settings? My GRT AP has some values here to make things either more or less responsive - I assume that other APs have the same. Probably will take some time to tweak them just right.
 
rv8ch said:
Are you trading responsiveness for stability? Could that stability be implemented in the AP servo settings? My GRT AP has some values here to make things either more or less responsive - I assume that other APs have the same. Probably will take some time to tweak them just right.
Click to expand...
Absolutely not. The plane was still very responsive, but in a more "well behaved" manner. I tweaked the edges untill they were the way Vans intended them to be. I was very happy with the feel of the plane after adjustment. I had never flown an RV before mine, so had no reference for what they should feel like. When you have convex or concave shapes on a flat control surface, you get turbulunce at the trailing edge. This turbulence creates undesired reaction in the sensitivity of that control surface. Some folks actually use that turbulence in place of a trim tab to address a heavy wing. I feel bad that those folks never get to experience how the plane is supposed to handle.
 
FORANE said:
It is, but even in the RV-9a I had some difficulties with the Trio Pro Pilot. In the RV-9a it would hold altitude fine except in rough air. I never was able to get it to climb or descend on autopilot; it would oscillate above then below vertical course in ever increasing amplitude.
Click to expand...
The Dynon AP in my 9A works wonderfully in turbulence, but it does struggle a bit when trying to climb at a specified airspeed - it tends to hunt up and down on the climb rate. I usually just climb on vertical speed and adjust as needed during the climb.
 
FORANE said:
It is, but even in the RV-9a I had some difficulties with the Trio Pro Pilot. In the RV-9a it would hold altitude fine except in rough air. I never was able to get it to climb or descend on autopilot; it would oscillate above then below vertical course in ever increasing amplitude.
Click to expand...
The Trio servo installation manual has some conflicting instructions regarding the location of the rod end bolt hole on the elevator bell crank. I moved the bell crank bolt hole as far away from the pivot point as possible while complying with all the over centre/control stop requirements. This made the bell crank movement less sensitive to the servo arm movement and completely solved all my previous oscillation issues.

Fin 9A
 
You must log in or register to reply here.
Back
Top