• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 30 subscribers
  • Views 689 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.

DRV8412 sawtooth output?

Manuel Malagon
Prodigy 250 points

Other Parts Discussed in Thread: DRV8412, DRV8840

Hi all

   I'd like to know why instead of having a constant voltage at the DRV8412 output am I getting something like sawtooth voltage? and because of this the motor starts to spin but immediately the voltage starts to decrease and so the motor speed until it reaches what it seems to be a low-level-trigger and then jumps immediately to the supply voltage and the cycle starts again. 

The motor power supply is 12vdc, I´m using mode 4 (as explained in datasheet p.14) because I have a simple DC motor in outputs AB, so, when I put a 0 in the line "RESET AB" then the output jumps to 12 volts and the motor start to spin, immediately the voltage starts to decrease aprox 1.2v/sec (no charge at the motor) until it reaches 2.6V aprox and then it jumps to 12 V again but immediately the voltage starts to decrease and the cycle repeats.

If I apply some charge the voltage decreases faster to the same 2.6v aprox. but it does the same pulsing response. Any Ideas?

Thank you for the answers!! Regards

  • 0
    TI__Guru 55715 points

    Manuel,

    The device could be entering cycle-by-cycle current limiting mode where one of the outputs is going to a HIZ state.  Couple of questions for you:

    1)  What is the DC resistance of your motor?

    2)  You mention a motor supply of 12V.  Are you also connecting GVDD and VDD to the same 12V?  This is what you should do, but just checking.

    3)  Can you post a schematic for me or are you using an EVM from TI.  What are the markings on your EVM if you are using one?

    4)  What resistor are you using on the OC_ADJ pin?

    5)  What does the input look like to PWM_A?  Are you just holding it high or PWM'ing the input?  

  • 0 in reply to Ryan Kehr
    Prodigy 250 points

    Ryan Kehr said:

    Manuel,

    The device could be entering cycle-by-cycle current limiting mode where one of the outputs is going to a HIZ state.  Couple of questions for you:

    1)  What is the DC resistance of your motor?

    2)  You mention a motor supply of 12V.  Are you also connecting GVDD and VDD to the same 12V?  This is what you should do, but just checking.

    3)  Can you post a schematic for me or are you using an EVM from TI.  What are the markings on your EVM if you are using one?

    4)  What resistor are you using on the OC_ADJ pin?

    5)  What does the input look like to PWM_A?  Are you just holding it high or PWM'ing the input?  

    Hi Ryan and Thanks for answering.

    1)   The DC resistance of my motor is 14 ohm.

    2)   Yes, I´m connecting these two to the same 12V supply.

    3)   I don´t know how to attach a file here.

    4)   I´ve tried with two values based on the information found in the data sheet p.12. First with a 24K and then with a 200K and does the same thing.

    5)   I´m just holding it High and I´m doing this because is what it says at p.14 in the datasheet isn´t it?

    I just don´t know what´s happening...

  • 0 in reply to Manuel Malagon
    TI__Guru 55715 points

    Manuel,

    Please post the schematic by clicking on the "paper-clip" looking button on the bottom row of the header nearly all the way to the right.  

    If you hold the input high, the bootstrap under-voltage protection will kick in (top right of page 11) as the bootstrap capacitor will deplete with no minimum low pulse to recharge it. Is that maybe what you are seeing?  

    We have some more integrated devices with selectable current chopping levels and charge-pump based gate drive that might work better for you.  Not sure how far down the design path you are, but take a look at the DRV8840 as an example of another option.

  • 0 in reply to Ryan Kehr
    Prodigy 250 points

    Ryan Kehr said:

    Manuel,

    Please post the schematic by clicking on the "paper-clip" looking button on the bottom row of the header nearly all the way to the right.  

    If you hold the input high, the bootstrap under-voltage protection will kick in (top right of page 11) as the bootstrap capacitor will deplete with no minimum low pulse to recharge it. Is that maybe what you are seeing?  

    We have some more integrated devices with selectable current chopping levels and charge-pump based gate drive that might work better for you.  Not sure how far down the design path you are, but take a look at the DRV8840 as an example of another option.

    Here is the schematic and I´m out of time, I need to make it work as it is. In the schematic I´m connecting directly to an MCU the lines OTW, FAULT, M2, M1, PWM_A, PWM_B, PWM_C, PWM_D, RESET_AB and RESET_CD. And as I mentioned above, I can´t change this because the PCB is already done.

    Thanks and I´ll wait your comments

    8640.Motor Control Schematics.pdf

  • 0 in reply to Manuel Malagon
    TI__Guru 55715 points

    Manuel,

    I should have asked before, but are you getting any low pulses on the /FAULT or /OTW pin during this behavior?  Make sure to remove R10 and R11 before trying this as I don't believe our internal drivers on the /FAULT and /OTW pins have enough drive strength to overcome the pull-ups on your board.  We buffer the diode indicators on our EVM.

    If I understand it correctly (some scope captures would help if you can post them), the behavior is un-latched as the condition seems to continue to repeat as long as you let the driver run.  

    Have you tried more devices to rule out some damage to the device under test?  

    Finally, have you followed the layout guidelines on page 14 in terms of decoupling on the PVDD pins, thermally connecting the PowerPAD to GND, etc?  This is all important for proper device operation.  Specifically, I would add a larger, bulk capacitor (330uF in reference schematic) to the MVDD node in your schematic.  

  • 0 in reply to Ryan Kehr
    Prodigy 250 points

    Ryan Kehr said:

    Manuel,

    I should have asked before, but are you getting any low pulses on the /FAULT or /OTW pin during this behavior?  Make sure to remove R10 and R11 before trying this as I don't believe our internal drivers on the /FAULT and /OTW pins have enough drive strength to overcome the pull-ups on your board.  We buffer the diode indicators on our EVM.

    If I understand it correctly (some scope captures would help if you can post them), the behavior is un-latched as the condition seems to continue to repeat as long as you let the driver run.  

    Have you tried more devices to rule out some damage to the device under test?  

    Finally, have you followed the layout guidelines on page 14 in terms of decoupling on the PVDD pins, thermally connecting the PowerPAD to GND, etc?  This is all important for proper device operation.  Specifically, I would add a larger, bulk capacitor (330uF in reference schematic) to the MVDD node in your schematic.  

    Hi and thanks for the support, after several tests I found out that mode 4 is useless, I just changed the mode to mode 2 and everything works perfect now. What I´m doing is:

    1)   To turn clockwise I put a 192Khz, 90% duty cycle signal to PWM A and 0 to PWM B.

    2)   To turn counterclockwise I put a 192Khz, 90% duty cycle signal to PWM B and 0 to PWM A.

    And that´s it, my motor moves perfectly fine to both directions!!