Hi, Tridge
thinking of putting my board in gray to test if it’s a hardware issue, also have a avau x2 which software
flavor would I use?
thanks
@lorbass Were you using Leonard’s attitude compensation patch when you had problems with the elevons oscillating? My testing indicates that the net effect of that patch is to significantly increase PID gains at low throttle, which would probably be the cause of the problem. See my comment here https://github.com/ArduPilot/ardupilot/pull/5971#pullrequestreview-38283254
and my flight log here: https://github.com/ArduPilot/ardupilot/pull/5971#issuecomment-301331357
No idea, I just flashed latest master.
But the shaking elevons where already without running motors.
And they shaked also with PID reduced fro o.35 to 0.1.
I think tailsitter with elevons only is a compromise between shaking elevons and good steering effect.
At least with small, ligth wings.
As a first reaction, I wanted to stop the chalenge but after sleeping some nights
I recovered the motivation and decided now to continue with the vectored tailsitter.
Glad to hear you’re going to keep at it. I think the vectored approach has great potential.
Leonard’s patch is not in master, so that wasn’t the cause of the shaking. Must have been due to excessive PID gains, as you say. Did you try moving your CG back to allow using lower gains in Q modes?
The CG was at that point as discribed in the manual of the Caipirinha. (Plane mode)
But may be, to high on the wing considered in plane mode. So the balance was difficult in hover.
The battery, the power modul, the Telemetrie TX and the BEC where sunk in the wing, the rest (RX, FC, GPS) on top.
Another fault was perhaps the usage of analog Servos. Yesterday Rolf told me thats not correct.
I will change it and see if the shaking is gone. If not, go back to the original elevons which are smaller.
I think with vectored they don’t need so wide. This helps compensate the weigth of the additional Servos for Tilt.
You see, hope dies last.
Shoot me an email through the contact form if you want any more details on the aircraft 
Looking to do transition tests in the next month or two.
Hey, started work on a TVBS stryker ,what’s the c/g range, just have the $ 20. wing no info[quote=“kd0aij, post:425, topic:15302”]
I’m having better luck with my Stryker
[/quote]
I have the F27C and the CG is about 15mm behind the recommended location:
http://www.parkzone.com/ProdInfo/Files/PKZ4275_F27C_PNP_manual.pdf
(step 5)
I’d expect you can use standard size elevons and the recommended CG with a vectored thrust arrangement though. My flying weight is 870g with a 3S/2200 LiPo
very helpful, thanks… I will post results soon
Hi all,
You remember my issues with shaking elevons?
Just tu help other greenhorns like me (6 years copter experience only) this information.
You need digital Servos for APM Plane (at least master) else it ends with shaking elevons.
Thank you Rolf !
No idea if this is everybody clear, I never read this somewhere.
Now, no shaking elevons anymore.
it kinda flies, sometimes
@tilt
Sometimes it dances.
Your simple wing was really stable for takeoff.
May be to light, the same for my planned Caipirinha?
Which FC are you using?
great! Welcome back
really nice! Needs a lot of tuning obviously, but great to see such a simple TVBS.
Why such large elevons? I would have thought with the vectoring that you wouldn’t need much more than an inch or so
what a great concept!
Now of course I’m trying to think about how to fly that with ArduPilot.
First step would be some EKF work so the state estimation doesn’t get dizzy in hover. I’m sure Paul could manage that. Knowing the exact sensor lag will be critical.
The hover should be doable using a rotating AHRS_View.
The control frequencies will be interesting - really low frequency for stability in hover, similar to a helicopter.
It would be good to create a FlightAxis model of this.
This type of config should have a high degree of natural stability. A friend of mine in CanberraUAV (Ryan) built a similar twist wing aircraft for his thesis and managed to fly it without an autopilot at all. It was naturally stable in hover with two nitro motors on the wings.
@tridge
I’m shure you will manage this.
But another chalenge will be the self finding landing pod.
If this fails, the whole flyer will be damaged by the hight dynamic energy of the rotating wings.
I started to setup the vectored reading back 250 posts to finde how to connect and how to set the params.
The only information I could find, are the logs from kd0aij. (20.bin,…)
But not shure what typ of aircraft (tailsitter or vectored) they concerne.
And there e.g.Log 20.bin Servo1_Function =4, Servo2_Function=19, Servo3_Function=4 and Servo4_Function=0
Servo3 the same as Servo 1, Servo 4 no Function(0)
Did I miss some infos?