Well issue is as follows:
When I’m in AltHold or PosHold mode and tilt the quad forward, the FC is supossed to increase throttle in order to keep the quad at the same altitude- In other words compensate for the amount of thrust “lost” in forward movement.
However in my case the FC does compensate but just not enought to totaly keep the quad at the same altitude, so I always end up loosing 1-2 or even up to 5 meters depending on how mutch tilt it. (That doesn’t happen on my smaller quad)
So question is: Is there any parameter I can set to change and increase the amount of throttle compensation? Or is there something else I can do to fix that?
This issue comes up a lot (I have the same issue) but it seem’s to be more of a build problem than a setting issue. Some suggestions of a pressure “bubble” building up throwing off the barometer are the most plausible.
Yes after some testing I can confirm it is an aerodynamical related issue. There’s something that tricks the baro into thinking it’s gaining altitude, for example a low pressure bubble forming below the top plate which is very plausible in a frame with a large top plate like my tbs discovery. I tested medium-slow forward flight while looking at the live MAVlink baro data. It showed an altitude increase of up to 2 meters just before it started correcting (by loosing the equivelant altitude it thought it gained) and eventually showing the correct one- althought being -2m below the actual starting point.
I’ll try to close some of the openings between the two plates and test again. If that doesn’t work i’ll probably try filling the spaces inside the FC box with low density foam and see if that makes a difference.
We have the same issue then as I have my Pixhawk between the plates of my Discovery too.
I think Randy said the best design is to have the bottom, top and sides enclosed while leaving an opening at the top edges of the side. If that description makes sense lol.
I see, so that has to be pretty a common issue of that frame.
Ehmm and I didn’t really get what you mean to be honest lol. You mean sealing the top and sides of the disco but leaving a hole at the top of each side?
I see, ok I tried pretty mutch everything in terms of aerodynamics on the tbs, I covered the front, sealed the back, sides, little to no improvement, some changes also made it worse.
For now I have increased inav_tc_z parameter to 7, that didn’t fix the issue but it made the phenomenon less severe- thought I still loose a ton of alt on really slow forward movement!
Check this out if you haven’t already: diydrones.com/profiles/blogs/los … ized-modes
It’s a little to risky though
