VTOL Plane Project - Follow our journey!

Way too high. I ran your video in slow motion and your servo literally snap to 90 degrees. No good…

Are you all located in Germany?

We will for sure try a lower setting. No but close, we’re from Stockholm, Sweden. :slight_smile:

What do you think about the sudden yaw when entering FBWA? (Around the 8s mark on the crash video). To me that almost says esc issues.

Do you guys check log files after each test flight?

That would pretty much will tell you and you can analyze every aspect of the flight. Give it a try and also upload the tlog and bin files.

Also, the rpm of your motors in transition is way too much. Your First goal is to reach Cruising speed and then go full throttle.

The formula for the lift coefficient, Cl, is:
Cl = 2L ÷ (r × V2 × A), where L is the lift, r is the density, V is the velocity and A is the wing area.

Coefficient Of lift is published ratios based on airfoil type at versions speeds. You can lookup the Clark-y table on line.

Density of air as per google.
You will have to use newton 2nd law to estimate velocity based on motor/ prop thirst and than u can calculate lift.
:wink:

1 Like

Do you have counter rotating props on the front motors?

If you look at how the motors are tilted when you fly in multicopter mode, you will notice one motor is tilted towards the front, the other to the back. When you initiate the transition, the motor tilted forward will reach the Q_tilt_max angle faster than the one tilted back, hence it will create a yawing moment. There is nothing you can do about this, as far as I know, except having counter rotating props. It will fly with same props, but this effect is accentuated.

Also if you are not pointed straight into the wind, again you have more tilt difference in the front motors!

Hello,
Nice project!
You have to trust your plane’s behaviour in order to try it over the water :muscle:

We also used a flying wing (already available in our company) and tried to turn it into a VTOL.
We first tried to use only two front tilt motors but it would be too sensitive to the wind.
Recently we modified it to a tri-copter configuration (using the same front tilting motors) and all the pixhawk parametrization was based on the information available in the Aticof’s project:

We ran into several difficulties during this modification but now succeeded (at least with no crash) in a controllable vertical flight (no adjustments to PIDs were still performed):

We now have to try to correct the vibrations caused by the back motor as you can see/hear in the video (we still don’t know if it can be corrected by reducing the pitch PIDs or if we have to increase the rigidity of the rear part of the foam structure).

We also want to use AUTOTUNE to adjust the PIDs but we will not have enough battery to fly for a long period of time (unfortunately it has a 5800mAh 4S 60C battery and only flies for about 3min with wind :confused:).

Does anyone knows if we can “pause” the autotune, land the aircraft, change the battery, take-off again and continue the autotune?

1 Like

Hi! Cool project!

No as far as I am aware you have to complete the autotune for it to save its values. We ditched autotuning and went for manual tune as we found more success with it. We still have a oscillation in yaw but I’m confident it will be solved as soon as the plane is rebuilt and I get a chance to fly it a bit!

/Teo

1 Like

Yessir, counter rotating. Interesting note about the tilt servo speed!

Hello David, I am glad that my project has been useful to you.

To autotune you can do it for each axis if you do not have a battery for all 3 axes.

2 Likes

Well, it is not about the speed, because you set the speed for both servos not individually. It’s about their respective position. When I tried same props on the front motors I got about 90 degrees of yaw at the transition start.

Do you have your LOG_BITMASK enabled (i.e. =65535 for PX4/PixHawk Full Logs default)?

Yes, and the problem with logging seemed to be a full SD card.

1 Like

I’ve started looking into .lua scripting. I’m trying to make a servo move but I cant seem to figure out how. Either the pixhawk doesn’t run my script or there is some syntax error… I am currently only running example scripts for testing :confused:

“Unable to allocate scripting memory” is the error I get

EDIT: Switched to a CubeOrange from 2.4.8 and it seems to work run the script - albeit with errors

We’re trying a belly lander bicopter with both motors on the same axis, i.e., not able to move individually. Is there any way to do this? I am not aware of any settings for this and I’m currently playing with scripts to make the motors move :sweat_smile: Here are some photos

Got it to work! Heap size was too big! Getting familiar with .lua syntax helped too :wink:

Hey, we went for a waypoint mission yesterday but crashed the plane. It went up to the desired height and then just fell down to the ground. We couldn’t read the bin logs (don’t know - corrupted) but the tlogs are available. It seems the plane thought it was doing 66m/s when hovering on the spot. We think this is cause we had previously enabled an analog aspd sensor but forgot to disable it in the parameters.

If someone can try to look at the bin logs that would be awesome, we can’t seem to do it!

(Edit: uploaded bin)

I had a look and it doesn’t look like there is any data to recover.
Can you link to the tlog?

Here it is. We fly at the very end of the log. :slight_smile: