MR60

How to explain the small YAW drift while flying and not touching the yaw stick

Posted by Hugues on July 21, 2013 at 1:11am

Hello,

I noticed in my last video filming some low frequency right/left (yaw) movements. I looked at my Tlog and noticed indeed that I have a small yaw drift although I do not touch the yaw stick.

How can this be explained ? What could be the causes ? Objective being to have a perfectly stable yaw in condition where I do not touch the yaw stick (and no wind of course).

This is my mavlink graph where i underlined in yellow the yaw drift. I also join in annex the full tlog file.

Thanks for helping me troubleshoot this.

3691041884?profile=original

 

2013-07-20 19-24-30.tlog

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –
Follow

Replies

  • MR60
    Forrest Frantz July 22, 2013 at 11:50am

    Too many reasons.  To guess, you'd need more data.  In alt-hold (hands off the throttle) simple mode:

    • Run a straight course out for 30 seconds and then back (using mostly pitch).
    • Run a straight course to the side for 30 seconds and back (using mostly roll).
    • Run a straight course up for 15 seconds and then slowly back down (using mostly throttle).

    [note:  between each, do something identifiable so you know when each run starts and ends]

    Also, 

    • Report what determines direction (magnometer alone or with GPS).
    • Report what determines declination (auto or input).

    From the log file we should be able to determine if it's

    • cross-talk between roll-yaw or pitch-yaw or throttle-yaw because your copter's mathematical center isn't the actual CG.
    • auto functions refining location and direction.
    • small differences between motors-prop combinations that result in small rpm differences required to generate the same lift. 
    • others?
    • or a combination of all possibles.

    Or, just get the actual CG as close to the mathematical CG as possible and don't worry about it.

This reply was deleted.