Strange Motor Behaviour...PID?

Info and discussion about the original AQ v6 flight controller

Strange Motor Behaviour...PID?

Postby stud » Wed Jan 14, 2015 12:32 pm

Gentleman,

I hope I'm in the right section of this forum. Sorry if not!
In the video below, you see that my X8 coaxial octo runs up. There is no oscillation or instability, but just before lift off, one (random!) or more engines suddenly get very fast (diverge), hit the current limit of the ESC and stop. This happens if the octo is hanging from the ceiling (like in the video), or stationary on the ground.
https://www.youtube.com/watch?v=Lwr7lre ... e=youtu.be
[youtube]Lwr7lrelwLw&feature=youtu.be[/youtube]

What I've done:
-All ESC32 are calibrated with 13Amps and the Global limit is set to 15Amps. I tried 20 Amps, but that just causes a stronger tilt. I think the problem is not the CL calibration, the problem is why do they run up so fast at all?
-Trial and Error with the PID settings of the AQ6. I isolated every parameter and set to zero indepentently and it still happened

What I think it might be:
-ESC32 PID? So far I used default
-Interferece between AQ6 and ESC32 PWM signal caused by three-phase cable between ESC32 and motors
-I dont know, please help :)

Setup
Rosewhite X8 coax
AQ6 post Oct14
ESC32
4S 4400mAh
MN3510 630kV
13x4.4 CarbonProps (T-Motor)
Immersion RC LongRange UHF with Futaba 14Chnl

Thank you for your help!

Dominic
stud
 
Posts: 12
Joined: Wed Jan 07, 2015 2:41 pm

Re: Strange Motor Behaviour...PID?

Postby aBUGSworstnightmare » Wed Jan 14, 2015 12:56 pm

stud wrote:MN3510 630kV
13x4.4 CarbonProps (T-Motor)


Hi stud,

you should re-calibrate your ESC32 since your motors are loosing sync!
I'm using the MN3510-15 630KV with Graupner e-Prop 12x6 and I have calibrated this Motor+prop combi with a global current Limit of 23A and a calibration current of 22A. This motor has some huge spikes in his current demand when accelerating!
Find my calibration Parameters enclosed (just for info).

I've tried to use it with 13x5.5 CFK props but was not able to calibrate it with good results.

Then you should start thinking of PIDs. Start with defaults; messing around with PIDs is not a good idea? Have you calibrated the board (sorry, no info if AQ6 with analogue IMU or upgraded to DIMU (= TARE + onboard MAG calibration at least)).

Please post some Pictures of the setup (i.e. location of the ESC32). If you have mounted them on the arms near the Motor you should either use shielded wires for PWM of upgrade to DIMU + CAN.
Is your baro shielded with foam? Don't hit me, just asking ...

Joerg
Attachments
MN3510-15-630KV_eProp12x6_23A-22A-18kHz.zip
(2.39 KiB) Downloaded 225 times
aBUGSworstnightmare
 
Posts: 1460
Joined: Fri Jun 22, 2012 5:24 pm

Re: Strange Motor Behaviour...PID?

Postby stud » Wed Jan 14, 2015 1:43 pm

Thank you Joerg, some very good, experience based suggestions!
-So I will recalibrate the CL with the values you stated. What do you mean by "not able to calibrate 13x5.5 with good results"?
-I have the DIMU, tare and MAG calibration is done and stored, no calibso or temp compensation
-The DIMU baro is shielded with foam, the AQ6 baro is not (is that a problem?)
Picture of the setup
https://www.dropbox.com/s/l9sqrq73xdsio ... 133408.jpg
Could it be the interferece? Because the PWM signal cables are pretty close to the Motor Cables

Thank you
stud
 
Posts: 12
Joined: Wed Jan 07, 2015 2:41 pm

Re: Strange Motor Behaviour...PID?

Postby aBUGSworstnightmare » Wed Jan 14, 2015 3:31 pm

stud wrote:So I will recalibrate the CL with the values you stated


I don't think that you can use them 1:1; I only wanted to give you an idea which Parameter to Change and what to look for (when judging the results; I made 3 calibrations with different parameter sets were the 18KHz set turns out to be the most efficient one - in case of power consumption/estimated flight time).

stud wrote:-The DIMU baro is shielded with foam, the AQ6 baro is not (is that a problem?)

No Need to shield the board, but many users forget to shield the baro.

stud wrote:What do you mean by "not able to calibrate 13x5.5 with good results"?

It was not possible to accelerate from - lets say idle (low revolution) to (nearly) maximum without the motor stalling (loosing sync) although the calibration could be finished. If you do a calibration, always use QGC to test the motor!!!
aBUGSworstnightmare
 
Posts: 1460
Joined: Fri Jun 22, 2012 5:24 pm

Re: Strange Motor Behaviour...PID?

Postby chschmid » Wed Jan 14, 2015 4:08 pm

Hi Dominic

I agree with Joerg about the calibration of the ESC's.
Nevertheless I would double check the motor mixing table. Looking at you vid, I am not sure what comes first. And the X8 should stay level with one motor off.
What I had in build with the ESC'S out at the motor mount was a screwed PWM signal as soon a the amps went over 10A. A shielded PWM line did the trick.

Cheers

Christof
chschmid
 
Posts: 1800
Joined: Wed Jul 11, 2012 7:41 pm
Location: Herrliberg, Switzerland

Re: Strange Motor Behaviour...PID?

Postby bluuu » Wed Jan 14, 2015 4:24 pm

chschmid wrote:A shielded PWM line did the trick.


Or CAN wiring
bluuu
 
Posts: 1339
Joined: Mon Oct 29, 2012 8:02 pm
Location: Poland, Silesia

Re: Strange Motor Behaviour...PID?

Postby stud » Wed Jan 14, 2015 4:42 pm

Thank you for your inputs!
I will do the CL calibration (by using your values, Joerg, i mean the 22/23Amps, not your calibration table) tomorrow.
I am sure that the motor mixing table is correct. About what comes first:
1. stable run up of all 8 engines
2. one engine "goes crazy", gets very fast and
3. it hits the current limiter
4. octo tilts
5. i shut it down
Where the core of my problem is nr.2. Nr. 3 is just a consequence
And then I'll see if the shielded PWM do the trick. Thanks again, I'll update tomorrow
stud
 
Posts: 12
Joined: Wed Jan 07, 2015 2:41 pm

Re: Strange Motor Behaviour...PID?

Postby bluuu » Wed Jan 14, 2015 4:46 pm

Do some logs. It may help a lot.
bluuu
 
Posts: 1339
Joined: Mon Oct 29, 2012 8:02 pm
Location: Poland, Silesia

Re: Strange Motor Behaviour...PID?

Postby aBUGSworstnightmare » Wed Jan 14, 2015 5:20 pm

stud wrote:I will do the CL calibration (by using your values, Joerg, i mean the 22/23Amps, not your calibration table) tomorrow.


Please post your results when you're done!
As I mentioned before: The spikes (in current demand) of this motor are the challange here.

Some examples:
Duty 15 to 50, MAX current = 27.402510
--> the motor draws 27A to accelerate from 15% to 50% duty

Duty 35 to 65, MAX current = 29.743782
--> 29A (!)

Once you find a good calibration it runs like a charm (drawing ~12.6A at 4S for 7700rpm). O.K. don't expect to see that high current draws with the 4.4in blade, but you should be warned. Maybe you also need to Change the switching frequency (highest one I was able to do with the 12x6 is 19KHz for this motor; compared to 21kHz on the Tarot 4006 using the same prop).
Last edited by aBUGSworstnightmare on Wed Jan 14, 2015 5:22 pm, edited 1 time in total.
aBUGSworstnightmare
 
Posts: 1460
Joined: Fri Jun 22, 2012 5:24 pm

Re: Strange Motor Behaviour...PID?

Postby chschmid » Wed Jan 14, 2015 5:21 pm

Hi Dominic

The current limiter is not shutting off the ESC. It is limiting the current. Why the motor goes out of sync and shuts down has another reason. The "going wild" ect. sound like what I had with the PWM lines.
Why not using CAN?

Cheers
Christof
chschmid
 
Posts: 1800
Joined: Wed Jul 11, 2012 7:41 pm
Location: Herrliberg, Switzerland

Next

Return to AutoQuad 6 Flight Controller

Who is online

Users browsing this forum: No registered users and 2 guests