• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 82 subscribers
  • Views 159 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TMS320F2800132: Power calculation

Manu36
Expert 1420 points

Part Number: TMS320F2800132

Dear all,

I'm using FAST estimator on a board using TMS320F2800132.

HW is very close to HV kit, most of the software comes from the universal_motorcontrol_lab.

We used overmodulation (with current regen), field weakening, and also a ripple compensation.

We use a 4 poles motors on the boards and all is OK. Especially we have a power calculation using P = 1.5 * (Vd*Id+Vq*Iq). We use some filters on Vd, Id, etc... and at the end we get a really small difference from the power measured with wattmeters, also when we have the maximum field weakening the calculation is OK.

 

We want to do the same with a 10 poles motor (about same mechanical speed so electrical frequency is higher) but in that case, the power calculation give poor results, especially when field weakening is acting (10 to 15% error on the power). Do you have an idea on what can be the root cause? On the HW we change the filtering on phase voltage because of the higher electrical frequency. The setting in SW was changed accoringly.

For information for the power calculation, we use Vd,Vq OUT and Id,Iq IN this give the best results with the 4 poles motor. We try to calculate Vd and Vq IN but the result looks not accurate and we do not understand why. 

Beyond the power calculation, the 10poles motor runs OK, no noise, current looks OK etc...

Do you have any idea?

What we want is the electrical power send to the motor, the power at the input of the inverter will be also OK.

Regards

 

  • 0
    TI__Guru** 117270 points

    The power calculation formula is correct. The issue could stem from the sensing resolution, or the rated power of the motor may be too low for the board. You could try applying an averaging filter to calculate the power when the speed and load are stable.

  • 0 in reply to Yanming Luo
    Expert 1420 points

    Hello, rated power/resolution should not be the issue as there is no difference in the power betwwen the 4poles and 10 pole. The main difference is the electrical frquency.

    I think I have found part of the issue:

    -I use Vd and Vq out to calculte the power but as there are a lot of non linearity (dead time / overmodulation), the real applied voltages are different.

    -Using Vd and Vq IN give more error but I finally found why : the filtering on motor phases looks compensated by FAST but this is not compensated on the parke and clarke transform I use to calculate Vd and Vq IN. I finally change the filtering to reduce the phase shift and now Vd and Vq IN looks ok and allows a power calculation which sounds now OK. I need to perform more test in different configuration to confirm, I will let you know.

    Regards. 

  • 0 in reply to Manu36
    Expert 1420 points

    I'd like to take this opportunity to ask another question:

    In the main ISR, the phases voltages are measured and the measured values are phase voltage to Gnd. Then a few lines later, an offset is substracted (typical 50% of the Dcbus) to get positive an negative value. Can you please explain why this is done? If I'm right, the clarke transform will at the end remove any offset. Is it only to calibrate the input voltage during the startup or there is another reason?

    Regards.

  • 0 in reply to Manu36
    TI__Guru** 117270 points

    To calibrate the measurement error from the sensing circuit across the three phases. Correct. The offset, which has the same value at half of V_DC, will be removed in the Clarke transform.