@Libor_Prilesky Interesting design… how does it perform in forward flight? And which exact parameter do you mean by “mixing_offset”? It looks like you have elevons and a canard, not sure what you mean by “elevons and ailerons”.
Tried rebooting after setting the parameter? I had a lot of troubles with setting up Flaperons with another plane till I figured out that its ONLY working as it should after I reboot.
I presume you mean MIXING_GAIN? That would allow you to change the ratio of ailerons and elevons.
where is the propeller, or is this a glider?
If the prop is meant to go at the back then how to you get enough airflow over the control surfaces in hover?
has anyone flown 3.8.3beta1 or beta2 on a tailsitter yet? I’d like to know if the new transition code works as well on real aircraft as it does is the simulator.
Short video of a PID tuning test of a Caipirinha tailsitter using a tether:
This is a reasonably safe way to do initial testing without breaking props. As you can see the tether has little effect on the vehicle if it is in a stable hover, and you can do an effective job of tuning the RPY pid controllers this way.
(I’ve cleaned the garage up a bit since then 
)
1 Like
re: Libor’s airframe: The props are a little hard to see, but the planform looks similar to my Stryker tailsitter, with slots in the wings and with the vertical fins on the thrustlines.
@tridge the (grey) props are right in front of the control surfaces. Trace the wires from the ESCs in the first picture.
ahh, thanks Peter! I must be blind to miss those.
yes, that setup should work as long as the CoG works out.
For tuning, using a “hang it from the ceiling” method like what @kd0aij showed in his video today is the way to go
Gents, thank you for your answers, I see that i Created a lot of confusion.
Here is explanatory picture, that will make it crystal clear.
as @peterbarker pointed out, props are directly in front of control surfaces. That makes them very effective, but the plane is quite noisy.
There are ailerons as well at the end of the each wing. Ailerons are not needed for VTOL flight but there are usefull in forward flight, because in FW the motors in front of elevons should be off and there is supposed to be a 3rd motor in the middle.
@kd0aij In forwars flight it performs quite well. You have to take care to set CoG properly, but it flies very nicely. It is guite noisy though, because propellers are comming through the space between trailing edge of the wing and elevon.
So, as I explained above, I have elevons and ailerons and I need to decrease elevon movement for aileron input, because the elevons are so effective. I wanted to use mixing offset - http://ardupilot.org/plane/docs/parameters.html#mixing-offset-mixing-offset to decrease elevon movement for aileron input. I hope that you can understand now beter, what I meant.
@palm369 Thanks for the advice, but I´ve tried and it does not seem to have any effect.
@tridge Kudos to you! We have a quite succesful hover. What is strange though, that our plane still wants to go slightly right. We had to fixed that by trimming the aileron input on the transmitter, but I don(t like this solution much. I have been trying to fix this by setting AHRS_TRIM_X and AHRS_TRIM_Z (I didn´t know, which one to use, due to tailsitter vs. fixed-wing orientation) and it does not seem to work and plane still tends to move right 
@kd0aij idea is not bad, but I´ve already tried it this way and in my opinion it is suitable only for fine tune testing and I would be quite worried, that the leash will get in the propellers. If the only purpose is to protect propellers, you can do something like X-Vert has (the propeller guards).
Not yet.
While waiting the release I tried to optimize the CG for FBWA with the end like this:
Log: https://drive.google.com/file/d/0Bxldq38bfbqHSnVWU1k0dWVXZXM/view?usp=sharing
I have no idea what happend at the end of the “old” Backtransition.
May be a servo failure.
Or can you find another reason in the Log?
So unfortunately the test of the long awaited new Transitions will last until the refurbish of the wing.
Regard, Otto
MIXING_OFFSET is for the ratio of pitch to roll movement in elevons. I don’t think that is what you want if I understand correctly.
I think what you’re trying to do is have relatively less movement of the elevons as compared to the ailerons. Is that right?
If so, you can do this by lowering the MIXING_GAIN, which affects both pitch and roll output of the elevons.
Also, MIXING_OFFSET doesn’t currently work, I’ve pushed created a PR to fix it
don’t do that. Put that trim back to zero then post a flight log, and maybe a link to a video?
I can’t see anything that stands out, sorry.
Did all the servos and the motors behave themselves after the flight in ground testing?
darn!
Good idea, verry simple.
But what happend if the line touches the Props?
To complete this idea you could lead the line via one pulley on top away from the wing and via a second one down.
And at the end with a small weight pull the line up as the wing climbes
By the way, I’m surprised about the stability as non vectored with the origin small elevons.
Can you post the Params?
I would like to compare with mines.
A shop has then a good order if you finde everything 
Regards, Otto
That is exactly what I want 
I want elevons to go full range of motion in pitch, but much smaller range for roll (because I have ailerons as well). The reason is, that in fixed-wing flight, if elevons would go full range to roll input it would be very sensitive.
That is exctly what I thought, because I am the only one using ailerons. Thank you very much. I’ll wait untill it’s ready.
The PR for MIXING_OFFSET is here in case you want to build it yourself:
I plan on adding it to plane 3.8.3 if it passes review
Wow…that was fast 
I don’t know, how to do that, right now, but I 'll do my research and try to build it myself.
Thank you very much.
ok, if you get stuck let me know what sort of board you have and I’ll do a build for it as long as you promise to test it
Thanks for analysing.
Will test Material after reasembled. To much disassembled after the " vertical landing". 
Regards, Otto