Page 2 of 2
Re: roll pitch yaw calibration
Posted: Sat Mar 07, 2015 11:37 am
by Greensprings
try setting the trim to 0 in the client
every time you boot it loads some trim which may not be what you need
just put 0 or any value in and hit enter
Re: roll pitch yaw calibration
Posted: Sun Mar 08, 2015 12:09 am
by poizone
Propeller guards do what they say and no more, guard the props. This seems to be a solder issue with the IMU, and crashes and bumps seem to be the common factor. I'm going to let one of the BitCraze guys take it from here, because this seems to come down to how well the chips are soldered on. The durability isn't holding up to the same standard the CF1 has had it seems.
EDIT: Crashes and bumps, even lightly with walls and ceilings, can do quite a bit of damage. These quads are pretty delicate.
Re: roll pitch yaw calibration
Posted: Mon Mar 09, 2015 1:38 pm
by tobias
I fly it indoor only. Moreover, I'm using the propeller guard. If at all it drops while flying, it falls on the carpet. Yes, it shows big drift in roll and pitch. Something like -59.
I received it 4 days back, I though I would play during the weekend and bought a PS3 WL controller but I'm hardly able to fly it. just after a few seconds of flight after resetting, the gyros get crazy.
It might be a soldering issue of the sensor as poizone suggests. Can you please check the console output in the cfclient after this happens. If it outputs something like "I2CDEV: Error callback nr: 1024" that is the problem. PM me if this is the case.
Else it might just be
unbalanced propellers.
Re: roll pitch yaw calibration
Posted: Mon Mar 09, 2015 1:52 pm
by tobias
The durability isn't holding up to the same standard the CF1 has had it seems.
We have actually done a lot of durability testing and we never ran into this problem before production. We have actually never damaged any of our many testing boards during the last 8 months of testing and usage. We have tried to analyse the problem and our current conclusion is that not all sensors has been properly centered during reflow and the soldering has therefore become week. It is still very few units that has this problem and we hope it will stay this way. This is normally Seeedstudios responsibility but we will also do what we can to help out so please PM me if any of you run into this problem.