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.

F28035 based PMSM for BLDC motor control. Motor as a generator

Other Parts Discussed in Thread: DRV8301

Hi,


I could successfully use the F28035 PMSM code for driving the BLDC motor with drv8301 experimental kit.  I would desire to run the BLDC motor as a generator with the same drive.  Is there any way.

thx

KVRao

  • FOC is capable of 4 quadrant operation, meaning both motoring and generating mode, clock wise and counter clockwise.

    For sensorless operation, InstaSPIN-FOC makes it much easier to handle 4 quadrant operation as we can go through zero easier and have more stable operation than the SMO based technique.

    There are many threads on this topic:
    e2e.ti.com/.../506294
    e2e.ti.com/.../500268
    e2e.ti.com/.../272539


    the one "gotcha" with this is you need hardware to handle the voltage/power generation. Most EVMs aren't built for this. You can build HW that has bleed / brake capacitors or that's connected to a grid. When using EVMs a good way to handle this issue is to run the motor and generator from the same bus, so there is a net zero energy transfer. See this article I wrote:
    www.edn.com/.../Testing-your-motor-control-algorithms
  • My understanding is:  when the BLDC runs as a generator, the output voltage will be rectified by the freewheel diodes embedded in the inverter, without any control applied. With a control of the BLDC motor as generator, are you still trying to do rectifier job or some other functions?

  • Hi Chris,
    Thanks for the threads. They are indeed interesting.
    Query:
    You can build HW that has bleed / brake capacitors - Could you please provide any links which could help me to build the hardware with F28069 motor control experimental kit.
    thx
  • Unfortunately our kit does not support this feature. If I'm interpreting Chris' comment correctly, building your own hardware would be a new design, not an extension of our EVM. There should be textbook publications of bleed/shunt circuits for braking, as well as circuits designed to present additional bus voltage to a source such as a battery. Sadly, we do not cover these designs and are not equipped to make a suggestion for the hardware.

    Sean
  • I have some experience with intra-spin. And I think it may work with a standard kit. When using a power source that can source and sink.

    Make a torque controller and set a torque value in the opposite direction. I think the controller will sink a constant current into the powersupply.
    I=KT * Toque
    (Kt is a motor parameter)

    An other option is add a parallel load with the power supply. So your bench supply in powering the load and the controller. When your motor is feeding current back to the load. current of you bench supply drops.

    I really like to know the results and how well the current can be controlled.
  • Hi evs,  Thanks for your inputs.  Actually I intend to use a battery for source and sinking should be possible too.  Making a toque controller is sensible.  Could you please suggest how to do this.  Maybe if I get the opportunity I could try it once.

    Thx

    KVRao

  • A battery is not really fail save.

    See lab4.c for a torque controller. Make it work before trying to make a generator. A small toque value shoot be enough to get the motor spinning.
  • Hi evs, Thanks for your inputs. I would try with lab4 and revert back.
    You have mentioned battery is not really safe. Could you suggest in additional to battery what we need to incorporate to make it safe. Also I believe Chris experimented with insta SPIN. I am not sure what additional precautions he has taken. May be you could throw some light.
    thx
  • I think a saver way is to use a bench power supply with a resistor load in parallel. To get thing working.

    After that. use a battery with a fuse and a serial resistor. So that the generator current controller has some resistance to control against and not directly into a voltage source. If that works correctly lower the serial resistor till it is 0 mohm.  I think that is the hard part to get that working safely because of the speed and accuracy of the current measurement.