Those health errors deserve a closer look.
Do you use always armed for one flight mode and a normal arming for the other mode?
The only reasons I can think of that you would get this motor glitch (and even negative collective) in one mode and not the other are:
- The FC reboots in the bad mode.
- The channel positions themselves were causing errors. This might be possible if one end of the mode switch was close to an invalid value and the problem mode was on the invalid value.
- The channel positions add up to an invalid value. This used to happen with old PPM radios if the sum of all the pulses in the pulse train was too long. I have an old RC brand 1 Tx and brand 2 Rx that have this problem.
- The glitchless mode ignores the error. When you are running it in "manual" are you also "always armed"? Really a stretch here.
You could look at Firmware -> FlightTime. If it goes to zero right after the glitch, the FC rebooted.
You could set up some GCS scopes to watch things like:
- all ManualControlCommand.Channel
(may jump to invalid values).
- SystemStats.EventSystemWarningID
- StacksMonitor, see if any non-zero stacks go to zero right in that time frame
It may be a hardware problem. I don't like to say that. Is this an OP CC3D or a clone, and if clone, where did you get it? I trust OP CC3D's. I have heard of a very few bad CC3D clones.