Autopilot disconnects after 60 seconds - new install, AFS panel

[RV6-KPTW]

Hi,

I have just started test flying a brand spanking RV14 with a Skyview AFS quick panel. I did due diligence with the wiring, connections, and ground calibrations for the servos. All seems well. However, starting on the in flight fine tuning guide for the roll axis, I set the system to simplified and engage TRK / ALT mode and it will reliable perform for about 60 seconds and disengage. There are no warnings other than the aural disengage to point to an issue. The alert log file just reflects the disengage without elaboration.

The User log has some interesting data for the AP - the only really weird value to me is AP_Pitch_Position - I see -90's. But I don't know how to evaluate these numbers. Is there a reference for the log data?

Are there any obvious things I should look at?

Thanks

Dennis

 

[airguy]

The AP disconnect switch operates by grounding a wire to airframe ground momentarily via your pushbutton disconnect. If that wire is rubbing somewhere and has a bad spot in the insulation, and occasionally makes contact with the aircraft frame, the Skyview cannot tell the difference between that and you pushing the AP disconnect button and will dutifully disconnect the autopilot. Check your wiring on the disconnect switch for that.

 

[DBRV10]

The AP servo positions are "just a number" and they are relative positions in steps from where the servo arm was when it was powered up compared to its current spot. Or something like that..... It is not important to you with your dilemma.

 

[RV6-KPTW]

Today, I turned off the disconnect config switches in the software. It had no effect. The autopilot still disengages in 10 to 60 seconds. Should the system still consider a disconnect request when SETUP MENU > AUTOPILOT SETUP > DISENGAGE BUTTON > options are all set to no?

Yes, there may be a wiring issue but one would think not for a disabled function.

 

[RV6-KPTW]

I just realized that this config controls engaging, not disengaging. So, never mind....

 

[Dynon]

If either of the servos' disconnect wires are grounded (button pressed), the AP will disconnect. There is no way to disable that, for safety reasons (exception/caveat - on a servo used as a yaw damper, the disconnect switch is used instead as an optional toggle switch with each momentary press).

So while this is probably a wiring thing (or the disconnect button), one way to possibly gain some insight is to go into AUTOPILOT SETUP > STATUS. From there you can see the disconnect button states. If something is happening very quickly, you might not catch it in there, but it's worth looking at.

 

[RV6-KPTW]

That status page does show a very quick uncommanded change to the disconnect button state. I also notice that when the stick circuit is connected, a press of the disconnect button only changes the state of the pitch servo. So maybe I have an issue with the roll servo wiring. I took the sticks with the disconnect out of the circuit and still get uncommanded disconnects. They connect directly into the ACM. I have trim in the roll servo circuit and see revisions to that wiring diagram. I will buzz everything out on that side of the ACM and see if there is an “Aha”.

 

[airguy]

Yep - you've got a wire shorting to ground somewhere. You've identified the culprit, that's the easy part - now you have to FIND and fix it. Good luck...

 

[jakej]

Dennis - just a thought here. What software version do you have in the EFIS ?

 

[RV6-KPTW]

It is one of the latest versions. I am away from the airplane for the next week or so and will check for the specific version.

 

[Janekom]

I agree with the previuos posters and have the T-shirt. It is most likely a wire chaving and making contact with the airframe.