This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DRV8301-69M-KIT: IPD-HFI Tuning / Motor Saliency Determination

Part Number: DRV8301-69M-KIT

Hello All,

I've gone through most of the labs and have the motor identified well, and the speed controller tuned. I would like to use HFI, since I would like to operate the motor at sub-100 RPM, perhaps 10RPM. The FAST estimator starts to get a bit into the noise floor and can't get enough back EMF and the motor vibrates a lot when turning at slow speeds.

I have a fairly large motor with a very significant amount of cogging torque. I would think it would be a fairly salient motor, but I am certainly not an expert. Is there any way to measure saliency? 

I've entered in as many parameters as possible into lab21, and I haven't been able to spin the shaft at all. At low HFI gain, I just get the motor singing at the HFI frequency (~625Hz). If I increase the measurement amplitudes, it just gets louder. I tried even 0.1 coarse and 0.05 fine, it's still quite audible. Is this normal? I've increased the gain to ~30-50 range, and the shaft oscillates rather violently. I am getting access to a current probe tomorrow, so I do not have much confidence the magnitudes are not heavily saturating the stator. 

When tuning kspd, there is mention of gThrottle, but it isn't in the user.h file. Do I need to adjust this in the expression watch window?

Do you know why the motor isn't attempting to move? It just usually sits there and hums.. I've added a watch to the angle returned from the HFI and it seems to be, at least, stable. Shouldn't it try to move?

It seems that this lab is setup for torque control only. How difficult would it be to port to a SpeedTAC control?

lots of questions, I know, but any attempt at helping me would be appreciated!

Stephen

  • In general, the motor manufacturer provide the Ls_d and Ls_q in the motor specification which shows the salient ratio. The inject current frequency and amplitude is important, so excFreq_Hz, lpFilterCutOffFreq_Hz, hpFilterCutOffFreq_Hz, iqFullScaleFreq_Hz, excMag_coarse_pu, and excMag_fine_pu need to be set a correct value based on motor.
    You can port it to SpinTAC speed control refer to lab6/lab13.
  • Thanks for your input, I appreciate it. I'll contact the motor manufacturer, but I suspect that they do not have this information. I wonder if there's a way to measure this, even if I have an LCR meter.

    I feel like TI should have more tips and steps on how to bring up the IPD/HFI solution, but I have confirmation that they aren't interested in supporting it unless you're a key customer.

    I do have hall sensors available, but I find the low-speed velocity ripple to be fairly poor. I think I will just have to live with it, unless more people express their interest to TI enough where they may reconsider their position.
  • I do not think it will be feasible to measure Ld and Lq with LCR, it is very difficult to do it this way. We don't currently support decoupled Ld/Lq identification, and we don't suggest a way to do it either at this time.

    Sean
  • If anyone else wants some HFI support, please +1 this posting, perhaps TI will attempt to continue where it left off.

    Proposed ideas:

    1) How to identify a motor that has saliency. Maybe an independent lq / ld measurement using the supported development boards?

    2) Tuning tips, hints, screenshots, etc. would be excellent; given (1) proves 'sufficient' saliency.

    Stephen