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.

TMS320F28069F: [Instaspin]: Encoder Synchronization & Diagnosis functions

Part Number: TMS320F28069F


Hello,

I am working with a custom SW based on Instaspin control lab 12 and custom HW. The control works fine and now I would like to include some diagnosis functions. I would like to detect if the encoder alignment phase during de "Rs Recalibration" state is correctly done or not.

There are some situations in which the encoder synchronization could be not correctly done:

  • Encoder Cable is broken
  • Motor not connected

If the encoder cable is broken It is supposed that no movement should be observed, so HAL_getQepPosnCounts() should return 0. Is that true? If the encoder is currently aligned the result would be also 0. Is that true? If so, is there any way to ditinguish both situations? For the second situation, is there any way to make an small movement of the motor in order to desalign it (and then repeat the alignment process)?

For the motor not connected diagnosis. I was thinking on measuring the torque given using the following function

USER_computeTorque_Nm(ctrlHandle, gTorque_Flux_Iq_pu_to_Nm_sf, gTorque_Ls_Id_Iq_pu_to_Nm_sf)

but the result it is quite similar with no load (or small load) and when the motor is diconnected. Do you have any recommendation to detect it?

Thank you in advance

Borja

  • Hi,

    One of our experts will get back to you by Tuesday, since Monday is a US holiday.

  • Sorry, we don't have such a reference code to detect the connection error from the encoder or motor. You might add such function to do this by checking the feedback signal from the encoder, and the phase current and torque of the motor.

    The torque will be abnormal if the motor is disconnected, so it can be used for detecting the disconnection fault of the motor.

  • Hello,

    Ok I will tried to make it with inferred torque. Just another doubt: Is it possible to generate a small movement like the one used during the encoder aligment phase. It just to be used in which no movement is detected during the Rs recalculation phase. During this phase a current I is injected throught the U phase and -I/2 for V and W phases. If no movement is detected during this phases there two options:

    The motor is already aligned

    The encoder conection is broken

    To disiguish these two option a "disalignement" movement could be done. Is it possible? For instance, what happens it you inject a I current throught phase V and -I/2 throught phase U and V.

    Thank you in advance

    Borja.

  • That depends on the value and duration time of the injection current, and the inertia and friction of the motor. You might find some papers about the high-frequency cuurent injection method.