Hi there, I just did a new build with a new pixhawk and i am sometimes getting the compass variance error.
Did the compass dance several times but to no avail. Strange thing is the fact that sometimes it is just alright and just lifts off without the problem, but sometimes it refuses to arm.There is no metal in the neighborhood. I am running ver 3.3. Because of this i have not tested loiter yet.
Did plenty of builds before with version 3.2 without any problems. I have an external gps+mag on a stick. Offsets look ok. Anyone any ideas?
You need to be a member of diydrones to add comments!
Replies
I had this same error repeatedly. Compass variances, failure to calibrate compass, etc. I finally figured out that the problem was that my external compass with GPS was too close to my camera gimbal (mounted on the top side of drone). I relocated the GPS mast to the other end of the drone and the compass problems went away. It is worth a try, use double sided tape to try it out and if it fixes the problem, you can permanently mount the GPS there.
Hi, DId you change the orientation of the compass? Which compass are you using?
Micha said:
Tnx for the input people. I did a test flight today and everything looks fine. Loiter is very stable and no toilet bowling. Next weekend i will do a little mission and see how that works out.
Good point. I took a look at the gps. Looks ok, the antenna is on top. Because of the design of the plastic case and the place where the cable is soldered there is only one way it can fit. I have rotated the setting in GS to roll 180° and will test it tomorrow. Nasty weather today.
Some things that can help when pulling your hair out over Compass problems: twist up all your accessory wires, especially the mag wire. Avoid running the mag wire near the power distribution board or battery leads. Do a Compass Motor calibration. Reinstall firmware if all else fails.
It is not uncommon for GPS to have the unit rotated and the arrow wrong. So since you are stillhaving problems I would take the cover off the gps and check to make sure the compass chip is pointing forward and that it is indeed on the bottom or underside of the circuit board, since you have it set to rotation_none.
Thx for dropping in Mike. The gps on a stick has an arrow on the top. It is pointing forward. So is the Pixhawk. Rotation is GCS is set to none. All my electronics are mounted at least 3 cm away from the pixhawk. The cf top plate is big with plenty of room. Underneath are the opto esc's. But there is no current flowing in them yet as the error occurs before liftoff.
Remember the compass variance is caused by a difference in reported compass data between the two compass. The fc is not happy that it is getting very different data from the two compass. The most common problem is that one or both of the compass are not aligned to front correctly but it can be caused as you note by electromagnetic interference to one or both of the compass.