No output to servos, except in Manual mode

Posted by Mapper905 on May 13, 2014 at 9:17pm

I found this topic in "troubleshooting", but the suggestions didn't help.

The plane has been working fine in FBWA, auto, etc., but would only operate in manual mode today.  I don't see any errors in HUD or from telemetry logs. 

I DID make a change today -- I replaced 3dr radios with RFD900s.  Everything telemetry-related seems fine though -- HUD info is all correct (flying in manual).  I can read/write waypoints, and at least on the ground, the HUD correctly displays as  I switch modes between stabilize/manual/fbwa.  There's just no activity to the control surfaces, or throttle, except in manual mode.  Pixhawk main LED is solid green. No errors in HUD, nor helpful messages in telemetry logs that I can find.

Should I switch back to 3dr radios?  Anyone else had problem with RFD900s & pixhawk?  This doesn't seem related to radios, but it's all I'm aware I've changed...

I'm flying Skywalker 2014/Pixhawk w/ 3dr gps/compass, spektrum satellite receiver, 2.78b firmware.

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

Join diydrones

Email me when people reply –
Follow

Replies

  • Mapper905 May 15, 2014 at 8:19am

    Some interesting posts from some developers about problems with powering RFD900 externally when used with Pixhawk:

    https://groups.google.com/forum/#!msg/drones-discuss/-W1zbHc2gOM/MT...

    • Developer
      Philip > Mapper905 May 15, 2014 at 11:58am
      Will chase that up, not sure where that's at..
  • Mapper905 May 15, 2014 at 6:45am

    OK-"troubleshooting guide" was right when it suggested redoing the radio calibration; my bad. I visited the radio cal dialog in MP, but thought it was working fine since I saw movement when I moved sticks/switches  -- that wasn't enough; should have actually re-done the calibration.  I have now; everything seems to be working again.

    Anybody know WHY the radio calibration would get reset?  I can't imagine what I could have done to have caused that.  I had not visited the dialog in MP, nor had I manually changed or reloaded any parameters settings.

    Thanks, Phillip & Tom, for the comments on the RFD900s.

  • Developer
    Philip May 13, 2014 at 9:22pm
    How are you powering your RFD900? It must not be powered off the board.
    • Tom McKinnon > Philip May 14, 2014 at 7:06am

      @Philip.  Is it correct that the RFD900 can't be powered directly from the Pixhawk board?  The Pixhawk has a much more sophisticated power bus than the APM 2.x based around the LT4417 PowerPath controller.  As long as a BEC with enough current capacity is connected to the power rail of the servo controllers, the power bus SHOULD roll over to the BEC when the APM power block is maxed out.  That being said, I've had some weird problems with noise on the power bus when I run my RFD900s at 1W.  The problem diminishes when I turn down the power.  I need to get it on my scope to try to diagnose the problem better.

    • Developer
      Philip > Tom McKinnon May 14, 2014 at 8:02pm
      To roll over to the servo rail makes quite a few assumptions. None that are worth risking your airframe over. The main one being, noise from the servos can cause an inflight reboot if you don't set it up right.
    • Tom McKinnon > Philip May 15, 2014 at 8:00am

      @Philip.  Can you elaborate on your comment "if you don't set it up right"?  I'm not aware of any setup options for the rollover power supply controller.  It seems that since the Pixhawk has a single power bus, any noise generated by the servos will be getting to the microcontroller no matter what we do.  Is this correct? 

    • Developer
      Philip > Tom McKinnon May 15, 2014 at 11:54am
      There is an issue with some servos, they generate high voltage spikes (up to 11v)
      This is well above the upper limit of 5.7v that will be accepted as a valid power source.
      As such, if you were to roll over to servo rail, and this happened, the system has no further options....
      The fix for this.... Run a 5.6v Zener diode across the servo rail (stipe end to v+) and make sure your servo Bec is at 5.1v

      This is not a Pix hawk bug, it is a design flaw in the servos... But the effect is that it reduces the reliability of the backup power.
    • Tom McKinnon > Philip May 15, 2014 at 6:48pm

      Putting a zener clamp across the servo rail is a brilliant solution.  I'll make up a bunch of those -- very cheap insurance.  Thanks!

    • Mapper905 > Tom McKinnon May 14, 2014 at 10:44am

      So Tom, are you running RFD900s with APM 2.x or Pixhawk?

This reply was deleted.