ArduCopter-3.2 beta testing

Replies

- Max LEe > Richard Kennedy January 16, 2015 at 1:54pm
  
  Hi Richard, the prop is the bolt on type, with two screws and a retaining plate. The prop didn't move at all.
  
  I visibly saw the #1 motor stop turning entirely... :/
  
  We have a spare pixhawk to throw on there, but now we are back to square one.... We are considering scrapping the motor/esc set that we have and go with a different brand such as T-motor. But thats at a considerable cost.
  
  I will do what terry suggested, and do a rapid mode changes and throttle changes with the new esc in place to make sure the sync issue doesn't surface its ugly head again....
  
  any other ideas on what could've happened? could I call this a sync related failure?
  - Phill Simpson > Max LEe January 16, 2015 at 4:04pm
    
    If you do your test the way Randy suggests doing the compass mot test (flip the props and rotate all of them one position clockwise) you can do it on the ground without any danger of a crash.
    
    It may be a compatibility issue between the motor and ESC. ESCs with the SimonK firmware and low KV motors used to be a problem until the software parameters were updated. you might have something similar.
    - Max LEe > Phill Simpson January 16, 2015 at 8:39pm
      
      We will do this test, with the quad tied down to the table, with props on upside down, one position over, to reproduce the results with the current esc/motor. to see if we can get it to lose sync while switching through all the modes, and etc.
      
      Thanks for the inputs everyone.
      
      Another question, is there a way to get all the parameter changes I have made, so I can directly apply it to the new pixhawk without going through all the auto-tune and setups for the custom power modules and stuff? whats the easiest way to find the parameter lines of values within the log?
      - Artem > Max LEe January 16, 2015 at 11:44pm
        
        You do sync loss test without FC at all, best results with Rx ESC direct connection. Setup a throttle cut switch on your radio, than test each motor/ESc separatelly by first quicly changing throtle stick postition from 0 to max, 1/3 to max, 1/2 to max. Than engage throttle cut switch, put your thr stick in 1/3 than togle thr cut switch multiple times, repet with 1/2 thr stick and max thr stick. If still no sync issues you have good motor/esc combination. I generally do this on all my builds and each motor/esc as quality of both can vary enough to cause issues on even identical combinations.
      - Alex O. > Max LEe January 16, 2015 at 9:00pm
        
        All the parameters are listed at the beginning of each log. You can find everything there and input manually to your spare unit. But I think a better way to do is using "Save parameters" on your crushed Pixhawk unit and then doing "Load parameters" on the spare unit using the saved parameters file.
Michael Burmeister January 16, 2015 at 1:00pm

I can't believe the Auto Pilot is the cause of this crash. If there was an issue with the sync it should have effected all of the motors and not just motor 1.

The logs show that motor one was starting to fail before it went completely. Could be an ESC failure or got to hot or maybe a loose wire connection.

Switching mode may have just shutdown the ESC long enough for it start it back up before it just failed again. Would not put to much into the mode switching.

I have flown many missions with this version and have not seen this issue. I have although seen my prop fly away and have my copter tumble to the ground.

Have you checked for other causes?

These are just my thoughts.

Mike
- Terry > Michael Burmeister January 16, 2015 at 1:10pm
  
  Hi Mike, I don't know how often all the motors are effected but very often it is only one. Where the sync is marginal or if the FC legitimately puts more acceleration on one motor which was the case in my crash, only that motor will loose sync. All that said it could also be the FC but only testing sync will eliminate that issue.
- Max LEe > Michael Burmeister January 16, 2015 at 1:06pm
  
  Hi Michael, Thanks for your insight.
  
  Could you explain to me where you see the #1 motor failing? I fail to see where the motor one was starting to fail, because when I look on the log, on RCOUT on CH1, the 100% demand was right when I switched to Loiter from RTL. This is exactly the moment, the copter started to tumble.
  
  I am not blaming the autopilot, I will replace the affected #1 ESC and the damaged #4 motor. I am just curious if anyone had a random sync issue.
  
  BTW, the aircraft has had dozens of good stablize-althold-loiter-rtl flights, but I cannot test further as pixhawk was crushed on the crash.
Max LEe January 16, 2015 at 11:50am

Hello Randy and developers.
We experienced a catastrophic failure on our geo-surveying Turbo Ace Matrix quadcopter.

It was our first auto-pilot mission with this aircraft.

I will describe the all the steps we took.
1. We powered up the aircraft, and made sure we had good gps lock, idle for ~5-10 minutes. while we wrote the waypoints.

2. We took off in Stablized mode, once at ~20ft, we switched to altitude hold. Once that was satisfactory, we switched to AUTO.

3. Auto began to climb, and we realized our altitude was set far too high, 200 meters, so we decided to abort the mission by hitting the RTL mode switch. At around ~80 meters up.

4. Copter began RTL, and started to descend.

5. When copter was ~50 ft above the ground, we switched to Loiter, to bring it down manually, and all the sudden #1 motor failed, and it tumbled out of the air.

6. I switched to Stablized, and tried to straight out the aircraft, but still had a very hard belly landing, followed by a bounce, and another failure of the #1 motor, which resulted in a mangled mess.

I visibly saw the #1 motor fail, and stop spinning, but it only happened when I switched mode from Auto to Loiter. This is confirmed by the log, which pixhawk clearly demands 100% throttle from the #1 motor, but no luck from there.

I've attached a log below, Please advise on what steps we should take next to prevent this type of mishap from happening. We are planning to replace #1 esc already.

Why would a motor lose "sync" after a mode change? Everything was going perfectly fine from that. Is it bad to switch from Auto directly to Loiter? Is there a bug in the code, where upon mode change, it may send confusing signals to the ESC making it lose sync?

Thank you!

25.BIN

https://storage.ning.com/topology/rest/1.0/file/get/3701914743?profile=original
- Gustav Kuhn > Max LEe January 18, 2015 at 9:51am
  
  Hi Max, others,
  
  I've just had a visit from a friend that flies IRIS, and have similar problems.
  
  After a long discussion, it turned out his problem was that the camera was unbalancing his IRIS nose down.
  
  The two front motors were working almost at the limit, the two back motors not so hard.
  
  Sudden mode changes somehow caused the pixhawk to go belly up, crash, and not record logs???
  
  Without the camera, same mission, never a crash.
  
  So I am thinking, more attention should be given to balance the aircraft.
  
  And I might add, on my 3DR quad and APM 2.6, I regularly change modes in the air, never had a problem.

of 373

Discussions

ArduCopter-3.2 beta testing

Replies

Activity