Everytime I reboot, Servo 3 and 4 change to TiltMotorFront (right/left)

I’ve tried a Pixhawk and a Kakute F7 - both on v4.6.3 (though oddly, Mission Planner keeps telling me a new version is available, and THEN says “no upgrade available”, which makes sense, cause I’m already on 4.6.3).

Anyway, I change servo 3 and 4 to disabled - but when I reboot, they are set to TiltMotorFront (right/left). Does it mattr they are like that?

I’m building a 2-motor tailsitter, non vectored.

For what it’s worth, @wsalopek I remember a similar issue earlier in 2025 and the end of 2024 when I was working on a similar build, tailsitter twin, non-vectored. I just attributed it to something strange with the little Speedybee F405 Mini I was using on the plane and eventually I had to move on from that project. But nearly every time I’d boot up, I’d have to go and disable a TiltMotor servo output, because in my case I think it was on an output I wasn’t using at all, and if I didn’t disable it, Mission Planner would yell at me about the error. Once I disabled it, everything was okay, until the next reboot. Unfortunately I never found a real “solution”.

You picture shows firmware 4.6.1 not 4.6.3.
Show your full parameter list.
Did you tried to change and save the setting by entering your full parameter list under CONFIG Tab?

That I actually have 4.6.1…

Yes very weird, even though I have 4.6.1 loaded (thanks for pointing that out), and even though Mission Planner prompts me with “New firmware available” - when I try to install, Mission Planner says:

”No need to upload, firmware already installed”. So THAT is weird.

And yes, I’ve changed the servo output both ways many times: by using the drop-down list on the servo GUI page, as well as by changing the values in the full parameter list.

Attached is my param file.

FX61VTOL.param (22.7 KB)

So you are saying that your 2-motor, non-vectored VTOL flew just fine even though Mission Planner (or whatever it was) kept changing one or more of your non-used servo outputs to “tiltmotor…”?

Thanks…

OK - I downloaded v4.6.3 from:

And then did a “load custom firmware”. which did allow me to upgrade from 4.6.1 to 4.6.3.

And I still have the same issue - when I reboot, servo 3 and 4 (two outputs that I am not using, and repeatedly set to DISABLED), are re-set to:

TiltMotorFrontRight
TiltMotorFrontLeft

Two questions…

Why does this happen? And - can I ignore this error and fly successfully?

And…

I’ll just add this extra info to this thread (from another thread), as it may matter…

When I try to do a motor test, and when I press A, B, C, or D, the left motor (when viewed from behind), and which is connected to Servo 6, spins for ALL FOUR letters - while the right motor, using servo 5, does not spin at all.

Does motor test work the same way for tailsitter VTOLS as it does for quadcopters, hexacopters, etc?

Yes. It flew okay overall, I was just trying to add in that my vehicle had exhibited similar behavior, and I had to adjust that servo output to be what it was supposed to be each time I rebooted the aircraft. I didn’t dig into it very much at the time when I was actively working with the vehicle, it was just a test platform, but I just thought I would add my similar experience so that you knew you weren’t the only one this had happened to in recent times.

Please do another test with AP 4.7 beta.

If there is still the same problem, report an issue on ardupilot github. Reference this thread in the description.

These “default” settings are are set by the frame type/class AFAIK but this should only happen once.

Tried 4.7 beta - same thing. So this error occurs now with 2 different flight controllers and several firmware version (including BD-shot and NON-BD-shot) - I went all the way back to 4.2.X(?) - same error (but I did not try the firmwares in between 4.2.x and 4.6.3.

Thanks…

1 Like

Posted to Github…

IMHO, the reply is not satisfactory (God bless the devs…but) and the firmware is in fact dangerous. My apologies in advance if I’m missing something, but under the current state of the firmware, I believe an attempted take off would have resulted in a very quick CRASH, as I found out I need to change ANOTHER parameter (Q_TAILSIT_VHGAIN) to actually tell the flight controller that I do NOT have vectored motors.