Notifications
Clear all

Startup from 0rpm issue  

Page 2 / 2
  RSS
(@meric)
Eminent Member

@milad Those we are talking about : Number of poles, pwm frequency.

=> I check again : number of pole is M, but I confirm : PWM frequency is Volatile.

 ****

I try to launch from I=0 A with Torque = medium value => Exactly the same behaviour.

I try both in torque and Speed => Same behaviour, modulo the servitude (speed increase continually after start of run when we are in Torque, and motor stop quickly when we put potentiometer at zero in speed mode). PID works properly.

Best result is when I reduce Imax to a medium value, and then I slightly raise the Torque. But, even, I must manually kick the wheel to start it.

This post was modified 4 weeks ago by Meric
ReplyQuote
Posted : 26/09/2020 10:45 pm
(@milad)
Chief Member Moderator

Hi @meric
Ok, I just now checked the PWM frequency changing and I confirm that it stores into Memory as I checked it with oscilloscope after power recycle also with 8kHz, so I want to know why in your case you see it's volatile and you should set it every time? Remember, when you read the PWM freq. from SOLO it sends you back the real value (it sends you back 8000 in Dec [0x00001F40], not 8), only in the setting we send the PWM freq. as a multiplicand of 1000. 

After this, I can imagine the main issue you are facing is the issue with the good tuning of the Current controller probably during the ID. process, so to see if that's the case I ask you to the following things:

***Pls follow the steps below one by one and report me what changes you see:

1- Turn Off SOLO and put all the piano switches up, make sure you are putting the current Limit POT into maximum value ( No current Limit ) and the Speed/Torque Pot is set at zero

2- Turn ON SOLO, and after some seconds, just push down the piano switch number 5 and wait for ID. to finish

3- Set the PWM freq at 8kHz if needed

4- Leave the piano switch pins for motor selection unchanged (Now you are selecting brushless motor NO# 3 by leaving PINs 1 and 2 UP)

5- Test the behavior of the Motor both in Torque and Speed mode and pls report it.

 

*** Beside the test above:

- I like you to send me the values that you read for the current controller Kp and Ki after ID.  which is done in both the conditions (Motor selected No#1 and Motor selected No#3), so what you need to do is do the ID. one time when the Motor selected is from type 1 (PIN1 Down, Pin2 UP and the other time when the motor selected it type3, Pin1 and 2 both UP), to read the current controller Kp and Ki you should send the commands with [0x91] and [0x92] code similar to reading the PWM you did, just share here the whole packet you receive and we will take care of the conversion here.

Regards

Milad

 

 

If you found the answers helpful and you could solve your problems, please kindly verify here to help others in the future.

ReplyQuote
Posted : 27/09/2020 8:55 am
(@meric)
Eminent Member

Hi Milad, 

OK, the write is in Hex[KHz] (ff ff 00 09 00 00 00 08 00 fe)  the write Acknowledgment is in Hex[KHz] (fe ff ff 00 09 00 00 00 08 00 fe ) but the read feedback is in Hex[Hz] (ff ff 00 8b 00 00 1f 40 00 ). 

=> I just identify that the read does not indicate the same number that the write ack, so for me the command was not registered. My bad.

After this, I can imagine the main issue you are facing is the issue with the good tuning of the Current controller probably during the ID. process

You're right in your diagnosis.

The Motor#3 has the same behaviour than before, but, performing the ID in Motor#1 config with the correct settings : 8KHz and 46poles = It works ! 🤩 

Motor#1 (functional setup)

ff ff 00 91 00 02 82 06 00 fe

ff ff 00 92 08 20 ec 4e 00 fe

Motor#3 (crappy setup)

ff ff 00 91 00 00 7b f4 00 fe

ff ff 00 92 03 25 89 d8 00 fe

ReplyQuote
Posted : 27/09/2020 8:47 pm
(@milad)
Chief Member Moderator

@meric

Perfetto! yeah, the reason is Motor No#3 is for Ultrafast brushless motors and they go up to 30000RPM which is strikingly fast and they need special treatment as they normally have extremely low phase inductance, but Motor No#2 is for Normal brushless motors, so the tuning is different, so how is the performance now? are you satisfied with it or we need to go deeper? are both Torque and Speed control functioning good?

Regards

Milad

 

If you found the answers helpful and you could solve your problems, please kindly verify here to help others in the future.

ReplyQuote
Posted : 27/09/2020 8:54 pm
(@meric)
Eminent Member

Behaviour is not perfect, but motor is running, and perhaps we deal with the (chinese) motor limits itself. Moreover, I run this motor undervoltaged, cause a lack of a pair of appropriate battery. => I'll come back to you when I receive the good one, to be sure of the motor well being.

Meanwhile, I'll probably try to run it with an arduino, so my next point of interest will probably turn around the connection topics.

Thanks for the (night shift) support 😉

ReplyQuote
Posted : 27/09/2020 9:15 pm
(@meric)
Eminent Member

Yes, Torque and Speed works properly. 
Speed mode is the Torque mode itself, behind a Proportional Integer servitude, right ?

ReplyQuote
Posted : 27/09/2020 9:17 pm
(@milad)
Chief Member Moderator

@meric
Ok great, also remember that now since you see it works, you can try to increase the switching frequency, 8kHz is fairly low and you might get better results with 12kHz or sth in this range, but it's very important to apply the nominal voltage on the motor to see it's really true behavior.
SOLO works in Feild oriented Control Mode and yes, the control loops are Nested on the back of each other with the order of  Position==>Speed==>Torque as you can see below an example for DC motors (BLDC, PMSM and ACIM motors are also similar to this)
How to control the speed of DC motor using ARDUINO and SOLO in closed loop sensorless mode 3

 

I will remain available, test it, and let me know if it went as expected or other issues. 

Regards

Milad

 

If you found the answers helpful and you could solve your problems, please kindly verify here to help others in the future.

ReplyQuote
Posted : 27/09/2020 9:23 pm
Page 2 / 2
Share: