DRV10983EVM: Problems tuning for my Maxon EC32 motor 226006

John,

It looks like the Rm is programmed quite right but that should be a big issue. I think 0x6B would be a better setting.

Looking at the inductance and resistance value I'm not too concerned about the L/R constant.

I see you have abnormal Kt lock disabled. Were you hitting this lock regularly and thus disabled?

The one thing I can think of is maybe the advance angle/time isn't set correctly. What I do when trying to optimize the advance angle is I look at the display page of the GUI and have it do autorefresh of the Display area. This will give you an indication of what the IC is estimating the BEMF to be. I then adjust the advance angle until the IC estimate and the programmed value are roughly the same.

A current measurement would definitely be helpful to debug. Could you at least provide a scope shot of FG and phase U voltage?

Thanks,
Brian

Hi Brian,

Thanks for responding.

I have updated the configuration so Rm is 0x6B 

Also I re enable the kt and other locks which haven't had any effect.

I am running open loop with IPD off and as a first step I am expecting to see the motor speed vary as I change the speed demand from 100 to 500 but what I find is that the motor starts, reaches a fixed speed which remains constant regardless of what I do to the speed demand.  I wonder I the algorithm ever leaves the startup setting state.

DRV10983 0x20 0x6B
DRV10983 0x21 0x2A
DRV10983 0x22 0x3B
DRV10983 0x23 0x0
DRV10983 0x24 0x40
DRV10983 0x25 0xFD
DRV10983 0x26 0x88
DRV10983 0x27 0xF0
DRV10983 0x28 0xA9
DRV10983 0x29 0xC9
DRV10983 0x2A 0xD
DRV10983 0x2B 0xD

Please find attached a scope plot with Channel 1 showing phase U and Channel 2 showing FG. at a steady state and the display page from the gui.

John,

I apologize for not looking closer at your settings. I missed that you have closed loop operation disabled. This will prevent you from going faster than the programmed open to closed loop threshold (which you have set at 25.6 Hz). Hence the reason the motor won't go faster.

Enable closed loop operation and you should be able to drive the motor faster as you change the input command.

Thanks,
Brian

Brian,

I enabled closed loop operation and then the motor will only run consistently when I un-check the Open Loop Stuck tickbox in the Advanced settings page. However I am still unable to vary the motor speed.

I then adjusted the T Control Advance(s) all the way up to 2.88m and saw the BEMF (Motor Velocity Constant) drop from 75 to 58, but again it did not result in me being able to control the motor speed.
The motor configuration settings I ended up with are attached below.

Thanks,

John

DRV10983 0x20 0x6B
DRV10983 0x21 0x2E
DRV10983 0x22 0x79
DRV10983 0x23 0x10
DRV10983 0x24 0x40
DRV10983 0x25 0xFD
DRV10983 0x26 0x78
DRV10983 0x27 0xF0
DRV10983 0x28 0xA9
DRV10983 0x29 0xB9
DRV10983 0x2A 0x5
DRV10983 0x2B 0xC

John,

It still sounds like the BEMF is not being estimated correctly.

Most motors I've worked with need and advance time of 100-800 us.

Can you try the following:

Increase the hand off frequency.

Tell me what the speed command and speed buffer command are (these should be the same unless the system is limiting the command because of software current limit).

Thanks,
Brian

Brian,

I increased the Open to Closed Loop Threshold from 12Hz to 25.6Hz.

Set the T Control Advance to 600us

All other settings as before.

With the speed set to 100 I get

Motor speed 1638Hz

speed command 19%

Speed cmd buf 23%

with the speed set to 400 I get

Motor speed 1638Hz

speed command 78%

Speed cmd buf 23%

Thanks

John

John,

Just to make sure I understand - with the open loop stuck lock disabled the motor continues to spin, but will not increase in speed with higher speed command?

Can you try programming Kt to be about half the value you have programmed?

Thanks,
Brian

Brian,

>Just to make sure I understand - with the open loop stuck lock disabled the motor continues to spin, but will not increase in speed with higher speed command?   Yes

I have tried Kt values of 51.3, 25.6 and 12.8, 18.3 each time changing the parameter, stopping the motor and then restarting.  The open loop stuck remains all the time and changing the speed control value has no effect on the actual speed.  

I am using a 24V power supply and the current draw is 140mA, If I increase the Open Loop/Align Current from 0.4 to 0.8A the current increases to 500mA but the motor speed does not change.

The only setting that affects the speed is the Open to Closed Loop Threshold,

With Kt at 18.3, Rm 6.8.

@ 38.4Hz FG frequency is 19.9Hz and the motor speed from a 2048 count encoder is 18KHz so 530rpm

@ 25.6Hz, FG frequency is 11.8Hz and the motor speed is 350rpm

@ 12Hz, FG frequency is 5.8Hz and the motor speed is 172rpm.

See motor config values used below along with Display page with different values of Kt.

John

DRV10983 0x20 0x6B
DRV10983 0x21 0x1A
DRV10983 0x22 0x4F
DRV10983 0x23 0x10
DRV10983 0x24 0x40
DRV10983 0x25 0xFD
DRV10983 0x26 0x78
DRV10983 0x27 0xF0
DRV10983 0x28 0xA9
DRV10983 0x29 0xA9
DRV10983 0x2A 0x5
DRV10983 0x2B 0x2C

Kt = 25.6

Kt=12.8

Brian,

I came across a note saying that the ground line should not be connected to the motor. Once this was disconnected, so the only motor connection is the 3 phases, then the motor speed became controllable!
The open loop stuck lock has cleared but I have a BEMF abnormal alert. When the BEMF lock is disabled I have my estimated Kt of 36.6mV/Hz and in the Display page is 80mV/Hz. I increased the T Advance all the way up to 2mS but that only reduced the BEMF level to 60mV/Hz.

With the BEMF lock disabled when I set the speed demand to 75 I get my desired motor speed of 600rpm,

With the T Advance at 1.76m and the Kt set to 51.3 then BEMF on the Display page is 67 and I have no lock flags.

Can you suggest how I could improve the performance so that I am not operating so close to a lockout condition?

Thanks very much for your help

John

John,

Thanks for the update! I didn't think to ask what you were doing with the common node of the motor.

It looks like you already have the largest lock window programmed so we can change that.

You can try changing the control coefficient to 0.75 to see if that helps. It's kind of a fudge factor in the algorithm, but may get the calculated and the programmed value to be closer to the same.

The other knob you can turn is changing the Rm up or down a little to get the calculated Kt closer to the programmed value.

Thanks,
Brian

Brian,

Thanks again for your update, the control is significantly better and working reliably, the only area that looks a little ragged is startup, when the motor pauses then rattles a little before accelerating up to speed.  Sometimes the motor heads off in the wrong direction for a 10 or 20 degrees before correcting itself.

Based on your suggestions I found the following:

Reducing the control coefficient to 0.75  reduces BEMF by a few mV/Hz and appears to have no negative effect, so I have left it at that setting.

Increasing Rm above 10 does reduce the BEMF a little but then affects the startup performance, i.e. the motor starts and then stalls with BEMF lock.

My settings are as follows, If you have any other recommendations that would be appreciated.

DRV10983 0x20 0x6B
DRV10983 0x21 0x2E
DRV10983 0x22 0x5C
DRV10983 0x23 0x4C
DRV10983 0x24 0x44
DRV10983 0x25 0xE3
DRV10983 0x26 0x89
DRV10983 0x27 0xFC
DRV10983 0x28 0x73
DRV10983 0x29 0xA9
DRV10983 0x2A 0x1D
DRV10983 0x2B 0x2E

Best regards,

John