Uncontrollable Flight in Stabilize Mode

I hate that my first post is seeking help…but help.

I’ve got a scratch built tricopter controlled my an APM 2.6 running 3.2.1. Until yesterday it was flying splendidly.

Yesterday started behavior that I can’t sort out. The tricopter will lift off as expected in stabilize mode and begin to fly normally. Then, after several seconds, it’ll pitch or roll wildly and shoot off in any direction while spinning at max rate.

During this time I have no control over pitch, roll, or even throttle. Oddly, however, yaw works fine.

The only way to get the thing back to ground again is to flip to RTL mode then back into stabilize. This gives me about enough control to crash before any damage can be done to people or property.

I’ve reflashed the firmware, tested motors and speed controls and can’t sort this issue out. If the problem was compass related I wouldn’t think it would present itself in stabilize mode. The compass/nav receiver are mounted up and away from anything and I have no problem with 3D lock or compass heading.

It looks like the Hdop value is way too high after takeoff, but I’ve been in a wide-open field with clear skies and everything was working great until yesterday.

I’m at a loss here. Any help is appreciated

Below are links to my log files:

.bin
drive.google.com/file/d/0B3tXxT … sp=sharing

.kmz
drive.google.com/file/d/0B3tXxT … sp=sharing

.log
drive.google.com/file/d/0B3tXxT … sp=sharing

.log.gpx
drive.google.com/file/d/0B3tXxT … sp=sharing

From the logs that provide very little information as logging is set to low it looks like you have a problem with your GPS. There are a number of BARO errors because the GPS is saying you are not moving up or down. The Barometer is showing that you went to a height of 50 meters and your fence is set at only 25 meters so it goes into land mode because of the breach. Switching to another mode resets this condition only to have it retrigger later.

Mike

Thanks for the reply, Mike. During each of these events I wasn’t more than two meters off the ground and attempting level flight. Would this still indicate a GPS error or could it be the barometer?

I’ve got the APM board in the plastic case, but haven’t covered the barometer with cotton or anything. I’m wondering if that would make a difference.

Either way I’ll have to adjust the logging settings to collect more data. New motors and batteries arrive today, so I should be able to get in the air tonight or tomorrow morning.