Just downloaded and installed the 2.0.11 release* from the APM Planner. I'm still getting the same rudder effect.
Jason, I (think) I have successfully made a log of the test flight which should show that I applied left rudder many times but the copter kept 'unwinding' and pointing in its original direction. With the number of times I used left rudder I should really have done several complete turns. Do you want me to send you the log? What is the best way to send you the log?
Nigel
* That train keeps on rollin' down the tracks :-)
Replies
I uncomented #define MAG_ORIENTATION in AMP_config and it seems to work. So ACM2 does not automaticly determine mag orientation!
Will test some more!
This way everytime you land you are sure you have saved your offsets.
You could also simulate the older set offests, by arming motors without giving throttle and rotating your copter for first flight.
I used new YAW function (defined in APM_config.h) #define YAW_OPTION 1
I have offests set correctly (checked in CLI).
YAW response is good, but goes left and right with no input.
I am trying to download logs, but apparently have some problems.
I opened an issue, concerning the Log, maybe I'm doing something wrong, but I can't download them even usin the Mission Planner.
Jason !
Did not get you. Do I have to fly at least 7 minutes straight (without Yaw applied) to determine and save offsets
p.s. Is mag orientation determined every time on start up?
Hey, Haven't been able to keep the downloads up to date for the last few days. Too many changes.
I'd highly suggest using Tortoise SVN or equivalent. It has all the commit comments that you can look at. Also you can browse the commits via google code.
The latest version is an entirely different method for Yaw hold. The Yaw stick is a relative definition of direction for hold. It's a really simple method so that I could narrow down on the problem. It's also just as effective as the prior Yaw method so I may stick with it. But for now it's a compile time option. See APM_Config.h for options.
I think that the compass offsets are not being save for some folks, or being cleared by the firmware revisions. If you don't fly for 7 minutes straight the compass will revert to 0,0,0 the next flight. Once it's saved it will last until the firmware is revved.
I am also going to try and setup the I term by default. It really needs to be used to keep drift to a minimum. You can try now if you want. Make it .0001 for starters. increase it as needed. You can also increase YAW_IMAX to 2° or 3°.
Be careful not to go too high on imax or you will get some throttle jumps.
Jason
Perhaps we face the issue that compass "feels" magnetic field from power wires. In my case they are not close, but anyway I will try locate it somewhere further. Still I do not understand if ACM2 automatically determines compass position or do I have to determine it somewhere? I have components down, pins forward.
Similar problem!
I have Hexa. Code 2.11. Without compass everything is SOLID and perfect.
When compass is enabled:
There must be in the code correction: Magnetometer must be used NOT for yaw gyro compensation, but only to determine navigation direction.
Hi Nigel
Where do you see the version number ???
Where did you find out about 2.0.11. The downloads page is srill showing 2.0.8?
Thanks