ArduPlane: RTL Mode wind drift problem

Hello together,

maybe there exists an easy answer to my problem however I could not find it yet.

I use APM 2.5 with the 2.74b firmware in an Easy star from MPX. Manual and stabilize mode seem to work perfect.

I encounter problems when trying to fly RTL. The model directly starts to turn and drifts away with the wind. This looks like the toilet bowl effect as far as I understand but I am not sure if this is the problem.

Please see here the effect (that was a windy day). My position was the thin white line in the right part of the image (were the plane never was except for takeoff ;-)).

3691067512?profile=original

What I tried to solve this problem so far (besides reading) I connected an external compass and checked the values of the vibrations (-3~3 for x and y, -15 ~ -5 for z according to manual) and these seemed to be perfectly in range. I turned off AUTO_DECLINATION of the compass and used a manual value for my hometown. With my first tests the airspeed sensor (which I don't have) was mistakenly enabled and I thought this was it but unfortunately no. I also uploaded the copter firmware to the board to check the magnetic variation with compassmot (not existent for planes as well(?)) and it gave me values of 5% and less.

Now I am a bit lost since all I can find in the net are tips for wind drift that are related to copters, i.e. a variable LOITER_P for example doesn't even exist for planes.

Can somebody point me into the right direction? I someone is interested I also have config files and flashlogs available, no telemetry unfortunately.

Thanks in advance

cheers

Kai

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –

Replies

  • Hi again,

    I tried to follow the hints but still no effect yet. I added the telemetry stuff to my apm which made life much easier (i.e. being able to change parameters while not connected via USB, great !), however setting up a mission and playing around with NAVL1_PERIOD (I am now at a value of 15) did not give any results. The behavior is still the same, it's now a bit more obvious when I replay the tlog and see the plane first following the desired heading to the first WP and then starting a turn into the opposite direction.

    Can it be that this is the CIRCLE mode what I see here? Does any kind of fallback for APM exist in case the APM has some kind of problem with GPS? Not that I know of any (tlog seems to look good) but I am wondering why no GPS supported mode seems to work, while for example STABILIZE seemed to work well.

    It was just a short flight since it began to rain but maybe someone could have a look at my tlog and give me a hint?

    Would be great,

    Thanks in advance

    cheers

    Kai

    2013-11-03 16-53-42.tlog

    https://storage.ning.com/topology/rest/1.0/file/get/3692872796?profile=original
  • Uploaded config file for my easy star with rudder-only

    easystar_v13_20131001.param

    https://storage.ning.com/topology/rest/1.0/file/get/3692850006?profile=original
  • 3D Robotics

    The "toilet bowl" effect only applies to copters, not planes. What you have looks simply like the aircraft hasn't been tuned with enough navigation authority for that airframe (are you using the rudder-only Easystar or the one with ailerons?). The tuning guide is here

    -c

This reply was deleted.