• State Resolved
  • Locked Locked
  • Replies 19 replies
  • Subscribers 29 subscribers
  • Views 2641 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

DRV8881P About current wavefrom at 1-2 phase

masahiko shimizu
Guru 11235 points
Other Parts Discussed in Thread: DRV8881PEVM

Hi,

My customer is using DRV8881PEVM and customer motor.
Then, he has a issue.

Please see the below picture.
This is current waveform.

Motor operates at 1-2 phase.
Do you know the reason of red circle of this picture?

Best regards,
Shimizu

  • 0
    TI__Guru** 114905 points
    Hi Shimizu-san,

    It appears the stepper code uses brake mode instead of coast mode in the areas highlighted.

    We will confirm this and reply soon.
  • 0 in reply to Rick Duncan
    TI__Guru** 114905 points
    Hi Shimizu-san,

    Can you confirm the 4 digital channels on the scope capture? What signals are they from top to bottom?

    Also, please provide the firmware version of the EVM (listed on the GUI).
  • 0 in reply to Rick Duncan
    Guru 11235 points

    Hi Rick,

    I confirmed your question.

    Please see below.

    and I got customer motor.

    Now I am waiting for EVM to arrive. I purchased EVM.

    1. About Input signal

    Customer does not use coast at FET OFF.

    2. Firmware

     My customer is using EVM. But not using GUI.

     DRV8881P on EVM is operated from external MCU. not MSP430 on EVM.

    Best regards,

    Shimizu

  • 0 in reply to masahiko shimizu
    TI__Guru** 114905 points
    Hi Shimizu-san,

    Thank you for the information. Using brake instead of coast may be the cause of this current.

    The MSP430 firmware for the EVM uses coast mode, and allows the current to decay to zero through the body diodes.
  • 0 in reply to Rick Duncan
    Guru 11235 points

    Hi Rick,

    Thank you for your reply.

    Please see the previous post waveform of lower left.

    Customer operated IN1/IN2 of DRV8881P to High/High for decay(slow decay).

    There is following table in datasheet.

    Is this correct?

    Best regards,  

    Shimizu

  • 0 in reply to masahiko shimizu
    TI__Guru** 114905 points
    Hi Shimizu-san,

    Yes, the table is correct. xIN1=xIN2=1 is Brake; slow decay
  • 0 in reply to Rick Duncan
    Guru 11235 points

    Hi Rick,

     Today, I took about this issue with my customer.

     This issue occurs only at low speed. For example 250 pps.

     When motor speed is 2000 pps, This issue does not see.    

     (VM=24V, Itrip=500mA, default EVM setting, ) 

     At previous post, I confirmed that Input signal of customer is correct.

     I think this issue is related to speed.

     Could you give me comment about this, please?     

    Best regards,

    Shimizu

  • 0 in reply to masahiko shimizu
    TI__Guru** 114905 points
    Hi Shimizu-san,

    Sorry for the delay. I also think this could be related to speed. At higher speeds, the next step appears before the current has a chance to increase.

    In the DRV8881PEVM, coast mode is used for the zero current step.
  • 0 in reply to Rick Duncan
    Guru 11235 points

    Hi Rick,

     Thank you for reply.

     Please tell me about below.

    >In the DRV8881PEVM, coast mode is used for the zero current step.

    I understood this is for mistaking of MSP430 firmware on EVM.

    Is it correct?

    I tried with DRV8881PEVM and customer motor.

    But at EVM(GUI), I can only operate 2-phase operation.

    For 0 current operation, It needs 1-2 phase operation.   

    EVM with GUI can not operate 1-2 phase operation.

    We need to input control signal from external MCU.

    Best regards,

    Shimizu

  • 0 in reply to masahiko shimizu
    TI__Guru** 114905 points
    Hi Shimizu-san,

    My apologies, you are correct that EVM operates in 2 phase (full step). To run 1-2 phase requires modification of the mcu code or control from external mcu.
  • 0 in reply to Rick Duncan
    Guru 11235 points
    Hi Rick,

    Thank you for reply.
    I understand your opinion.

    Then, Do you have something to comment about this issue?

    Best regards,
    Shimizu
  • 0 in reply to masahiko shimizu
    TI__Guru** 114905 points
    Hi Shimizu-san,

    The original comment "I also think this could be related to speed. At higher speeds, the next step appears before the current has a chance to increase." still applies.

    Has the customer tried using xIN1=xIN2=0 (coast) for the zero current step?
  • 0 in reply to Rick Duncan
    Guru 11235 points
    Hi Rick,

    >Has the customer tried using xIN1=xIN2=0 (coast) for the zero current step?
    (Ans)
    No, customer has not.
    For the zero current step, Customer input xIN1=xIN2=1.
    Please see waveform of previous my post.
    Bottom of the picture is input signal.

    e2e.ti.com/.../2311.DRV8881P.png

    Best regards,
    Shimizu
  • 0 in reply to masahiko shimizu
    Guru 11235 points

    Hi Rick,

    I confirmed customer motor and DRV8881P with external MCU at 1-2 phase operation.
    I could occur this issue.
    And also, I could other stteper motor, too.

    Please see the below picture.
    This is A phase output voltage of DRV8881P and motor current at customer motor.

    At 0A current operation, DRV881P output(AOUT1/2) is Low.
    In the other word, Brake(Slow decay).
    But current is not operated to 0A.

    I guess that the cause of this issue is that decay is not enough.

    Q1 : Is this guess correct? Do you think about this guess?

          -> I need to explain this issue mechanism to customer.

    Q2: If this guess is correct, is there countermeasure against this issue at DRV8881P?

          -> Coast step? this is your opinion at privious your post. 

    Q3: If we use DRV8881E with AutoTune, Is this issue solved?

    Best regards,
    Shimizu

  • 0 in reply to masahiko shimizu
    TI__Guru** 114905 points

    Hi Shimizu-san,

    masahiko shimizu said:

    I guess that the cause of this issue is that decay is not enough.

    Q1 : Is this guess correct? Do you think about this guess?

          -> I need to explain this issue mechanism to customer.

    Yes, brake mode (slow decay) causes the current to decay slowly. As the motor speed increases, there is less time for the current to decay to zero.

    masahiko shimizu said:

    Q2: If this guess is correct, is there countermeasure against this issue at DRV8881P?

          -> Coast step? this is your opinion at privious your post. 

    Yes, the countermeasure is coast step (asynchronous fast decay). This will remove the current quickly.

    masahiko shimizu said:
    Q3: If we use DRV8881E with AutoTune, Is this issue solved?

    AutoTune will have no effect.

  • 0 in reply to Rick Duncan
    Guru 11235 points
    Hi Rick,

    Thank you for your support.

    I could solve this issue by using coast step for 0A operation.

    Could you let me know disadvantage by using coast step for 0A operation, please?

    Best regards,
    Shimizu
  • 0 in reply to masahiko shimizu
    TI__Guru** 114905 points

    Hi Shimizu-san,

    I am glad to hear coast step worked.

    The disadvantage is the current is flowing through the body diodes as it decays to zero. This creates more heat during the decay time.

  • 0 in reply to Rick Duncan
    Guru 11235 points
    Hi Rick,

    Does "more heat during decay time " mean the loss of resistor of body diodes?

    And I think that Motor is more heat at fast decay than at slow decay.
    Is it correct?
    If it is correct, Is coast step same?

    Best regards,
    Shimizu
  • 0 in reply to masahiko shimizu
    TI__Guru** 114905 points
    Hi Shimizu-san,

    Does "more heat during decay time " mean the loss of resistor of body diodes?
    Yes, the current is flowing through the body diodes.


    And I think that Motor is more heat at fast decay than at slow decay.
    Is it correct?
    If it is correct, Is coast step same?

    Motor is generally less heat at fast decay than slow decay. The RMS current (torque) at each step is less.