Replies to This Discussion

3D RoboticsPermalink Reply by Chris Anderson on October 5, 2011 at 9:57pm

Did you do the setup/config process again? You always have to do that after loading new firmware.

• ▶ Reply

Permalink Reply by Jaques Viljoen on October 5, 2011 at 11:26pm

I agree with Chris - found that out last night while having difficulty in getting the AC2 HILS running. When in doubt start from scratch (I don't even use old setup files - just in case)

By the way, did you guys notice something strange about the Altitude Hold fields in APM Planner 1.0.77 not being updated when connecting?

 

 

• ▶ Reply

Permalink Reply by Johann Van Niekerk on October 6, 2011 at 10:22am

Hi Jaques how you keeping?

 

I redid the claibration and setup and the motor test works fine now but I am still face with the old problem of my left motor is underperforming badly, but then I did some more testing or call it by chance but I had enough of the copter and just gave it full power while holding it in my hand ....... the left motor fixed it self  and was reacting when tilting the copter, so i did some more testing and it seems as if the left motor only jumps into action once throttle is set to 50% or higher below that it under performs and copter tilts to the weakside and as result crashes. but when doing hand launch at 50% it floats like a boat and everthing is perfect only problem is at 50% my copter is already(or wants to be ) 20 feet up in the air. why would this be ? I re did calibration its not that they all start the same time. is it possiable that we need a 12v regulator between the PDB and the ESCs ? so that they all get same voltage at same time ?  

• ▶ Reply

Permalink Reply by Jaques Viljoen on October 6, 2011 at 12:53pm

Hi Johann

I'm keeping well, finally getting back to AC2 after crashing with it and switching to the kkPlus board (www.kkmulticopter.com) and having LOADS of fun. Was like flying with the AC2 in Acro mode, except the other board cost less than a tenth of the AC2.

On to the problem you face: Johann, I'd honestly

1: firstly look at both the motor and the ESC before anything else. If your using bullet connectors between the ESC and motor - remove the bullets and SOLDER THE WIRES. No jokes - crashed my 450 heli after burning out the motor - and that was from only one bad connector. You see, since the ESC drives the motor on any pair of the three wires, the other is used for sensing the motor's movement through the magnetic field and the processor then deciding when to switch the wiring pair based on that detection. If you have a poor connection you might get false information and thus poor firing timing that just causes too many other bad things to mention i.e. coil saturation leading to over current conditions etc.

2: To be really sure, after soldering and recalibrating your ESCs exactly the same (You did make sure all 4 have the same calibration, timing settings, voltage cutoff, soft starts, bla bla bla, didn't you?) that you move the offending motor and ESC combo to another position on the quad.

3: If the problem moved with debug #2 try using another motor with the suspect ESC i.e. just switch out 2 motors. If it goes along with the motor, that is the best bet to be the culprit, else one might have to take a good look at the ESC.

 

Your question about the same voltages to the ESCs -> All the ESCs are already getting the same voltages from the battery through the Power Distribution Board. You shouldn't be concerned about this. I think you might find a simpler answer ;-) I have read of another person having loads of problems with the 880kV motors due to poor manufacturing in his opinion, and all I can say is that I also think one might get a dud motor now and then - the same goes for ESCs. A friend's quad had problems spinning up one of the motors, eventually we couldn't even start it by hand anymore. We swopped the ESC out with a spare and all was fine. Upon inspection I saw a lot of poor soldering with poor quality solder so I resoldered most of the ESC - especially the MosFETs, and voila, the ESC worked perfectly again.

 

Many possible problems Johann - many. Don't look to solve them by trying the most difficult or unique solutions :-) You may find the solution to be relatively simple..

 

Best regards and GOOD LUCK

Jaques

• ▶ Reply

Permalink Reply by Jaques Viljoen on October 6, 2011 at 11:32pm

Hi Colby, Michael let me know that its fixed - some parameters were renamed. :-) Hopefully some testing this weekend!

• ▶ Reply

JDronesPermalink Reply by Jani Hirvinen on October 6, 2011 at 6:54pm

Johann,

 

at least on 2.0.44 it works just fine and there has no been any discussions to remove it yet so it should be there.

• ▶ Reply

Permalink Reply by Johann Van Niekerk on October 6, 2011 at 9:15pm

Hi Jani

 

 I got the motor test to work again , but if you read my reply ontop you will see the other issue im having and its really strange

 

Thanx Jaques

 

But I have replaced all the ESCs and two motors on the effect side , but problem still there.

I never had bullet connector its been soldered straight on from the first day.

• ▶ Reply

Permalink Reply by Jaques Viljoen on October 6, 2011 at 11:48pm

Agge nee... Sorry Johann, really thought it was the hardware on that leg.

Are you able to load previous versions of ArduCopter code, like 2.0.44? If you could, maybe you can see if that solves the problem.

Another thing you could try is to run the ArduCopter in your hand while looking at the values the AC2 sends to the ESCs, don't you think? On the Flight Data page you click the Raw Sensor View btn and I think the second tab in the popup window has visual information for radio signals in and radio signals out. Maybe you could see if the funny throttle behaviour above and below 50% throttle is visible. Then you'll know its the code.

Hope you nail it ;-)  -- Remember to update Mission Planner - Mine is now 1.0.79

• ▶ Reply

Permalink Reply by Johann Van Niekerk on October 7, 2011 at 4:32am

Hi Jaques

 

Thanx for the hope , but it nailed it self ..... straight into the tar road. started working after redoing the calibration for the 1000 time and flue nice for about 2 min then did a yaw wabble and toke off like a rocket about 100 meters into the air truned left and headed full spead down to earth.... 4 blocks later one ArduCopter is no more, 3 arms bent , 4 motors bent, all props gone , dome in two pieces, Ublox GPS smashed and two ports on APM bent, not sure if its still working brought everthing back in a black bag to the house.

 

starting all over again

• ▶ Reply

Permalink Reply by Jaques Viljoen on October 7, 2011 at 4:45am

AAAAG F.K

 

Johann, what bad luck - really sorry to hear that.

Could you please elaborate on the mode you were flying, which code base?
Better yet - if you could download the logs and post them, maybe Jason Short could give us a hand with figuring out what caused the accident.

 

Condolences once more matey - its really kark rebuilding and having no assurance of success next time out, by the way, thats why I have been playing with the kkmulticopter board for a while -  much less stress ;-)

 

• ▶ Reply

DeveloperPermalink Reply by Olivier ADLER on October 7, 2011 at 9:31am

 

But luck...

 

Wich radio are you using ? Did you check the radio inputs before to fly ? (available inside Mission planner raw sensor view)

 

You could be targetted by the PPM encoder "difficult receiver" compatibility problem.

You could have it working for quite a long period, but it could come back to decoding errors after some minutes.

 

In my experience, you need to test radio inputs during at least 1 hour to be sure. Use the Mission planner radio view (use raw sensor button to go here) to check, it is the best tool to check for radio reliability.

 

• ▶ Reply