What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Is Skyview Autopilot a "reasonably modern Autopilot"?

WingedFrog

WingedFrog

Well Known Member
My question is inspired by a quite interesting story from the March 2016 Issue of Sport Aviation by J Mac Mcclellan (Title "Pilot in Command"). It is all about cockpit automation. In the article section discussing all what can go wrong in the autopilot control, McClellan says:
"Every reasonably modern autopilot that I know of requires the human pilot to test the clutches as part of the preflight checks"
Of course he is talking about commercial airplanes way bigger and more complex than our RV-12 and I am being a little facetious in the title of this thread. Still, I have this question: Everyone of us using the Skyview autopilot has experienced instances of slippage of the pitch server. These slips are subject to warning and solved by adjusting the trim (or doing nothing if produced by a sudden strong turbulence). In the case of failure of the clutch, i.e. undetected stucked clutch, what would be the consequence? I assume that we would not get the warning but we also could get unusual attitudes. Should we (or better, Vans) not add this test of the clutches in the preflight check list?
And by the way, I have no idea how we could do that :confused:
 
It seems that if you really wanted to do it, you could test roll by moving the heading bug away from the current heading and putting the AP in Trk/Hdg mode. If you can keep the stick from moving (and feel the "clicking" of the clutch), it's working.

Pitch would be similar.
 
There are no clutches in a Dynon servo. They are inherently safe in that the internal drive motors are torque limited. Every time you "slip" the servos you are experiencing the maximum torque it can exert in a failure.
 
You also get a nice "ROLL SLIP" or "PITCH SLIP" annunciation on the EFIS when it happens. That's usually my cue to disconnect and hand fly when the skies are too bumpy.
 
Modern Autopilot

MGL's autopilot, if you use their servos, is accurate, is digital and the digital servos give you a force feedback on your screen for manual or auto trim function. I have an Iefis touch screen and it is wonderful.
 
WingedFrog said:
McClellan says:
"Every reasonably modern autopilot that I know of requires the human pilot to test the clutches as part of the preflight checks"
Click to expand...


I haven't read the article, but I don't have a clue what he is talking about. I have never flown an airplane with an autopilot, (inc. g/a, commuter, Boeing, Airbus, etc.) that required me to preflight or otherwise test the autopilot as part of the preflight checks.

It either works or doesn't, is MEL'd or inop.

Most PTS state clearly that that you may be required to demonstrate proficiency using the automation installed in the aircraft used for the practical test, which means you may be required to intervene if it is not working correctly by responding to inputs.

I tuned mine per the guide and it works pretty good. Set an altitude and waypoints, hit GPS, climb at 75-85 IAS, and she will level off and trim and maintain alt within a few feet.
 
It's been a while but I remember when calibrating and setting up the Dynon auto pilot, the instructions ask you to use break out force on each servo once. I assume to make sure you can over ride the servos if needed. My auto pilot has worked great for 10 months! I think it was on my second flight, I hit the AP button and SHAZAM! It worked perfect! I don't know why I was so surprised but i was!
 
I have installed many Dynon servos both driven by the D series and Skyview systems. I just installed a 2-axis system (as well as a full Skyview system) in a -9A and on the first flight, I turned on the A/P at 500' and it worked absolutely flawlessly. Nothing to adjust. Nothing to tweak. It just worked.

I have a Dynon A/P in my IFR RV-10 and my IFR RV-6A and love them. They are as modern and feature rich as it gets.
 
For full disclosure, I am a dealer of Dynon equipment, but that doesn't change my opinion of them. I sell their stuff because I like it, I don't like it because I sell it.

I also have installed a lot of TruTrak auto pilots. They work somewhat differently than the Dynon, but overall are very stable and dependable as well.

I have also installed some Garmin servos, and they work well also.

I have never had the need or request to install Trio or GRT equipment, but I hear good things about both.
 
cactuspilot said:
I haven't read the article, but I don't have a clue what he is talking about. I have never flown an airplane with an autopilot, (inc. g/a, commuter, Boeing, Airbus, etc.) that required me to preflight or otherwise test the autopilot as part of the preflight checks.
Click to expand...
The King autopilot in my Mooney had to be ground tested or it would not engage in flight, so it's not unheard of.

"POWER ON AND SELF-TEST

When power is applied to the KAP 150 the TRIM light will be illuminated indicating that the system self-test should be initiated. The test must be completed before the A/P can be engaged. To perform the test momentarily push the TEST button."
 
dynonsupport said:
There are no clutches in a Dynon servo. They are inherently safe in that the internal drive motors are torque limited. Every time you "slip" the servos you are experiencing the maximum torque it can exert in a failure.
Click to expand...

The servos also have a shear pins which will break, if the AP jams.

Mac is a good writer but he doesn't seem to understand the experimental products we enjoy so much.
 
The preflight checks in Even a basic late model 172 have you engage the AP during preflight and move the controls, I believe to verify that you can over power the controls, then you disconnect the AP and hear the annunciation.

The Garmin AP instructions tell you to do the same. During any preflight when I knew I would be using the AP on my G3X I would do this... AP master switch on, Engage AP (GMC 305), move controls, disengage using remote disconnect/CWS button. Sort of a loop to verify all is good, takes two seconds.;)
 
49clipper

In the King Air models, it is part of the checklist to check break out force and to see if they move in the right direction on each startup. I never do it on my TruTrak though. Maybe I should at least check the direction.
 
Dbro172 said:
The preflight checks in Even a basic late model 172 have you engage the AP during preflight and move the controls, I believe to verify that you can over power the controls, then you disconnect the AP and hear the annunciation.

The Garmin AP instructions tell you to do the same. During any preflight when I knew I would be using the AP on my G3X I would do this... AP master switch on, Engage AP (GMC 305), move controls, disengage using remote disconnect/CWS button. Sort of a loop to verify all is good, takes two seconds.;)
Click to expand...

This is exactly the answer I was looking for. Can I do that with my Skyview and how? I understand that the slip warning is a virtual slip as well explained by the Dynon representative (not knowing better I had taken the slip warning as the sign of a clutch existence:eek:) but still, is a preflight test of the Skyview AP really useless?
 
WingedFrog said:
This is exactly the answer I was looking for. Can I do that with my Skyview and how? I understand that the slip warning is a virtual slip as well explained by the Dynon representative (not knowing better I had taken the slip warning as the sign of a clutch existence:eek:) but still, is a preflight test of the Skyview AP really useless?
Click to expand...

Yes you can. The Dynon AP has a self test function.

Check your PM's.
 
Well I read the article, and it was excellent.

I will say that this thread probably misses the intent of the article all together.

Fly the plane...use automation to help you fly the plane...know the limits of that automation...intervene to correct when necessary.

First off, I am no expert on autopilot servos, but it strikes me that generally testing servos the way some suggest here is putting wear and tear on devices that don't necessarily need to be tested that way.

Second, a refresher:

9b6b7fa1e35cbab3ec4351d046e92fa4.jpg
 
automation and staying on track

Mac is pretty old school... well, so am I so look who's talking.
But, some of the writers don't get around much before they start typing these days. I do have" not so fond" memories of rental Cessnas that came with an autopilot. No one I ever knew would engage one of those without a firm grip on the wheel. More likely than not, the plane would be headed for an oil starvation situation in moments.
One item that always sticks in my mind though, was asking the FAA flight test engineers how much force was allowed to be in the pilots hands if elevator trim ever went full travel (un-commanded) at cruise speeds. It was at a time when an Italian professor was introducing his autopilot via STC here in America. It had no servo on the elevator. Only a brushless motor servo on the trim tab. It came to mind again this weekend when I was hefting a sixty pound sack of concrete pre-mix for a backyard project. The FAA, the designer and of course the marketing people could never give me a straight answer. But I do have a suspicion that the forces in pitch would certainly be a challenge until one got slowed down and or pulled a breaker.
My Tru Trak and my Trio are very well mannered BTW.
 
This thread is getting more and more interesting and funny... Eventually I would like someone to come with an idea on how to do this preflight test in a way that's not as cumbersome as the Dynon test which IMHO is more adapted to test the installation than for a preflight test. I personally (I may be wrong) do not see great harm in not having this test but as an avid user of the autopilot, it would bother me having to turn around after finding out that 10 minutes into my flight the autopilot is not working. Yes, I admit, I hate flying my RV-12 on a straight and level flight of one or two hours!:D
 
I have the Gamin system in my RV-9 and do as Derek mentions.
I once had moved the plane before the AHRS had calibrated. At this point the A/P does not function. Since I do not want to fly a two hour trip without A/P I have added the A/P check to my checklist.
Simply engage and make sure the stick has servo resistance after AHRS has calibrated. (there is a indicator on the PDF that lets you know)
 
Skyview autopilot

I don't know if my Skyview autopilot is a "modern autopilot" or not. What I do know is that it is an awesome autopilot that is feature rich. Punch the AP button and see if it arms the servos, punch the disconnect buttom to disengage before takeoff easy peasy. Set the climb and flight parameters you want at 500' punch the AP button and away you go. Hdg, Nav, VS, Alt, Vnav, whatever you want is just a button push away. Decents, approaches to minimums its all there.

After you've flown behind Cessna, STec, King and a bunch of wing levelers you cant believe how nice it is to have such an awesome piece of equipment. You can set the torque limits to allow over ride and if the servo jams the sheer screw setup works. When engaging and on an approach I certainly have my finger right next to that disconnect button and hand on the stick because you just never know!

So is the Skyview autopilot a "modern autopilot"? Well I dont fly a 777 or anything like that which I assume would have a modern autopilot but my Skyview autopilot is plenty modern for me.
 
Jrskygod said:
I don't know if my Skyview autopilot is a "modern autopilot" or not. What I do know is that it is an awesome autopilot that is feature rich. Punch the AP button and see if it arms the servos, punch the disconnect buttom to disengage before takeoff easy peasy. Set the climb and flight parameters you want at 500' punch the AP button and away you go. Hdg, Nav, VS, Alt, Vnav, whatever you want is just a button push away. Decents, approaches to minimums its all there.

After you've flown behind Cessna, STec, King and a bunch of wing levelers you cant believe how nice it is to have such an awesome piece of equipment. You can set the torque limits to allow over ride and if the servo jams the sheer screw setup works. When engaging and on an approach I certainly have my finger right next to that disconnect button and hand on the stick because you just never know!

So is the Skyview autopilot a "modern autopilot"? Well I dont fly a 777 or anything like that which I assume would have a modern autopilot but my Skyview autopilot is plenty modern for me.
Click to expand...

Good pitch, Ted, I will try that!
 
WingedFrog said:
I would like someone to come with an idea on how to do this preflight test in a way that's not as cumbersome as the Dynon test which IMHO is more adapted to test the installation than for a preflight test.
Click to expand...

Engage the AP on the ground and verify that it's restricting you in pitch and roll. There's really not more to do to test it in a preflight.

However, it's basically unheard of for the AP to be broken and not have an ERR indication in the AP top bar. The AP has enough sensors to be pretty sure when it's healthy or not.
 
dynonsupport said:
Engage the AP on the ground and verify that it's restricting you in pitch and roll. There's really not more to do to test it in a preflight.

However, it's basically unheard of for the AP to be broken and not have an ERR indication in the AP top bar. The AP has enough sensors to be pretty sure when it's healthy or not.
Click to expand...

Thank you Dynon, I like this and will try it. If it works I will add it to the preflight checklist.
What you say about the ERR indicator is probably correct but I like to follow this time tested rule: Trust but verify...
Would it not be nice if the ERR indicator came up while doing the test on the ground rather than in the air?
 
I did the test today, it worked as predicted:
- With avionics and AP powered, I set the AP on level
- Overpowering the pitch server (moving the stick forward and backward) results in a slip and trim message as well as a clicking noise
- Same thing for the roll server without the trim message
If the overpowering is not harmful for the servers I believe it would be useful to introduce this test in the preflight checklist.
Then we would be able to claim that the Skyview has a "reasonably modern autopilot" :D
 
WingedFrog said:
I did the test today, it worked as predicted:
- With avionics and AP powered, I set the AP on level
- Overpowering the pitch server (moving the stick forward and backward) results in a slip and trim message as well as a clicking noise
- Same thing for the roll server without the trim message
If the overpowering is not harmful for the servers I believe it would be useful to introduce this test in the preflight checklist.
Then we would be able to claim that the Skyview has a "reasonably modern autopilot" :D
Click to expand...

With the current design of the system, I don't think it is possible for the servos to produce more torque than you would be able to over come with the stick.
And if they were, there is a shear screw that would fail and disconnect the servo from the control system when you resisted with the stick.
 
WingedFrog said:
Would it not be nice if the ERR indicator came up while doing the test on the ground rather than in the air?
Click to expand...

The ERR will be there before you engage. The AP servos are constantly checking themselves even when not flying the plane.

About the only thing they can't check is for a broken shear screw.
 
You must log in or register to reply here.
Back
Top