Discussions

Moderator

Strange airspeed sensor outputs

Posted by Jost Laznik on August 7, 2011 at 1:59pm

Hy all

 

Today i have flown my drone for the first time in AUTO mode. Basic 4 point navigation route. The throttle sounded so strange (i am talking about fixed wing airplane)... every few seconds it turned full throttle for a fraction of a second and stopped... then i examined my logs and discovered that my airspeed sensor output suddenly dropped from some value to 0m/s !!!! ... many times.. every few seconds. I assume when airspeed dropped to 0m/s the engine went in full throttle?

 

I have attached image of airspeed log... Is there any solution it can be one?

3690882924?profile=original

thanks to all

 

jl

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

Join diydrones

Email me when people reply –
Follow

Replies

  • Moderator
    Jost Laznik September 25, 2011 at 2:24am

    Just after another test. I changed speed controller and things are much worse.... now there is 0 airspeed for longer periods...

    Does any one know how to eliminate this... as Helldesk said it is probably RFI from controller... But there is another strange thing..  if you check the image i have attached you will notice that at the beginning for first few seconds when flying in manual mode airspeed values are normal... after changing to auto RODEO begins.  I have aslo done tests on the ground with motor running in manual mode... the outputs were OK... i will try to repeat ground test in auto mode....grrr

     

    any idea what to do?

    airspeed4.png

  • Moderator
    Jost Laznik September 20, 2011 at 12:03pm

    Just bought a new sensor and test it on ground.. all readings were stable. When in the air, log is the same as previous ... airspeed fluctuation is still present... APM throttles full when AS falls to 0... and i must say it looks and hears pretty bad when observing my drone....  Still waiting for solutions.. Anybody experienced the same problem?

  • Developer
    Doug Weibel August 8, 2011 at 9:33am

    Sorry Chris,

     

    Your answer is not correct.  If the airspeed sensor is enabled then pitch is always controlled by airspeed and if your PID values are good airspeed will be maintained regardless of headwind/tailwind.  There is no concern of a stall due to any wind condition other than a sudden strong tailwind gust that would stall the plane before APM could react.  In that condition you would probably stall under manual control as well.  We do not compare to gps speed.

     

    There is a "nudging" feature in the code that allows you to increase the speed set-point in case you are making insufficient headway into a headwind.  If this feature is enables then the throttle stick should be left in the bottom 1/2 of the range while in Auto mode(s).  Moving it into the top 1/2 nudges the speed set-point up towards the max FBW speed.

  • Developer
    Doug Weibel August 7, 2011 at 11:05pm

    Another issue could be if the tube connecting the probe and sensor is intermittently getting kinked.

     

    The calculation is very sensitive around zero, so the reading jumping between 0 and 2 m/s on the bench is no problem.

  • 3D Robotics
    Chris Anderson August 7, 2011 at 2:13pm
    Sounds like you've got a loose connection on the cable. When you're testing in CLI mode and you blow into the active tube (upper one), do the numbers go up as you would expect?

    While you're testing, you can just disable the airspeed sensor in MAVLink and the plane will fly fine. it's mostly to help with fighting the wind.
  • Moderator
    Jost Laznik August 7, 2011 at 2:01pm

    I have to say i have cable extension of about 40 cm between APM and airspeed sensor.... Could that be the cause.

     

    Also when in test mode i dont get 0.00m/s values but it jumpes from 0 - 2 m/s

     

    thanks

     

    jl

This reply was deleted.

Activity