I’m hoping Pauls new control scheme will reduce this “pumping”. Paul gave me a cleaned up branch yesterday, and I’d like to try to do a beta release with his new control scheme as an option. You’d still be able to use the current control scheme, but if you set Q_FRAME_TYPE=13 then it will use Pauls new system.
I did my first flight with Pauls new controller on Sunday. It went OK until a servo slipped (again!) and the two motors ended up pointing off by 45 degrees. It didn’t fly so well after that 
Great video !
In your video the yaw control is much better with motors than with ailerons .
I hope that it will be so in Arduplane in a next future.
I agree, the non vectored design with wind would allow (difficult) landing only with the wing parallel to it .
I plan to make my plane work the best I can and then pass to vectored .
Did you try to set Q_A_ANGLE_BOOST - to 0 ?
Verry interresting to hear. I will try it when ready.
currently I’m tuning the wing in a rig.
I will do it systematically eg for Roll
Needs time to document:grinning:
Will report soon.
Otto
We noticed today that we had to set Q_A_RAT_PIT/RLL/YAW_FF to zero, and the fidget had ended.
After that we had the first successful transition flight:
Rolf
looks great, congratulations!
interesting, I’d previously found in SITL that some FF helped with stability, which is why for tailsitters it defaults to 0.2 for FF on pitch and yaw. If others find the same result I’ll change that default to zero.
Maybe you need some FF for conventional tailsitters.
There was hardly any wind yesterday. We’ll see what happens at higher wind speeds.
Second Flight (no wind) . Excellent software, thanks to @tridge and all firmware developers.
.
1 Like
That looks really nice!
Hi, Tridge, yes, an old beat up one very soft… thats why the oscillating motors in hover and it needs bigger servos, at high speed I loose some elevon control it has the original 9g servos… but flys great… will build another with some carbon in the wing it also needs more yaw but fogot to take my keyboard to the field
Thanks for the software!
[quote=“tridge, post:601, topic:15302”]
I assume this is the FT versa kit?
[/quote]Yes , from plans with the blunt nose
1 Like
@majelan looks great! Even managing to land without tipping over, which is tricky on tailsitters.
Can you share your DF log?
It would be interesting to know for each tailsitter plane we are testing where is the COG and verify the relevance of the distance between the COG and the propellers line in the design of the plane.
As reported, I made about 30 Testflights to find useful parameters.
And was not satisfied with the result.
Therefore I started Tuning in a Rig.
Nearly finished, Rolf reported his “discovery” to set the FF to Zero.
And this brought the success with the simple Params:
Roll P=0.2 ,I=0, D=0.01— Pitch P=0.14, I=0, D=0, Yaw P=0.2, I=0,D=0.
VHGAIN 0.5, and “FF for all = 0.0”
If somebody is interested to see why I’m shure, here the results.
Tuning Pitch https://drive.google.com/file/d/0Bxldq38bfbqHYlJEbHBab2EzaW8/view?usp=sharing
Tuning Roll https://drive.google.com/file/d/0Bxldq38bfbqHaDhYallzeURlbU0/view?usp=sharing
Otto