• State
  • Locked Locked
  • Replies 8 replies
  • Subscribers 29 subscribers
  • Views 2818 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.

DRV8825 Reset Issue

Saumitra Deshpande
Prodigy 210 points
Other Parts Discussed in Thread: DRV8825

Hi,

We have designed a board with 3 DRV8825 Drivers. The main controller is STM32 uC with configuration interface(Microstepping, Reset, Enable, Direction, Decay etc) through 74HC595 & 3 separate PWMs for each. Power supply for Motor & Controller is Dual Output SMPS with ratings of 24VDC/4A & 12VDC/1.5A.

Note that 595 shift registers & DRV8825 are on same PCB while STM32 is on separate PCB with single FRC cable used for inter-connect.

I am able to configure DRV8825 correctly through SPI of STM32 The issue we're facing is that sometimes while starting the motors or sometimes while motors are running it resets all shift registers making all of its output 0, leading to reset of all DRV8825. Note that there is no activity on any of shift-registers lines when this happens.

1) I assumed that it may be issue with shift registers of noise interfering & tried many things, but issue isn't solved.I am unable to catch any glitches on SPI lines with my scope.

2) Also using 2 separate power supplies sometimes masks the issue, but in the long run shift-registers get reset.

3) Interestingly, we've another configuration using exact same motors, same power supply. Only Shift registers are on STM32 PCB while we've 3 separate DRV8825 boards inter-connected through 3 FRC cables. It's been running without any issues for last 12 months 24x7.

So is it possible to have multiple DRV8825s on same PCB, sharing common Power Supply tracks(+VM & GND)? Or issue is still with shift-registers?

I am attaching images of TOP & BOTTOM layers of my DRV8825 PCB.
Any help will be greatly appreciated.

Thanks in advance.

  • 0
    Prodigy 210 points
    Hi,

    One more question, can we leave the nHOME pin on all DRV8825s floating(not even a pullup) or is it mandatory to connect it to a pullup resistor?

    Thanks agains,
    Saumitra A Deshpande
  • 0 in reply to Saumitra Deshpande
    TI__Mastermind 26065 points
    Hi Saumitra,

    nHOME can be floalting if you don't use the signal. Have you checked all the power line and regulator line of the logic? Also make sure the GND plane connection that the control signal will not long cross the high current return path.

    Try to find the extact condition to triggle the issue. For example is it connect to any sepcial motor/load action. It will be help find the root cause. Also could show the schematic as well?

    Best regards,
  • 0 in reply to Wilson Zuo
    Prodigy 210 points

    Hi Willson,

    Thanks for reply.

    1) Power GND (for 24VDC supply for Motors) & 3.3V GND(used for ARef on DRV8825) from uC Board are connected to each other at single point and never cross over each other, either TOP or BOTTOM layer.

    2)There is no regulators on this board.  3.3V input from uC Board has 3x100uF capacitors & 2x100nF capacitors before its used by any of circuits.

    3) Each x595 has 100nF decoupling capacitors right next to their power pins.

    4) I also tried to terminating the SPI lines with various resistors or adding pullup/downs or adding small RC filters on SPI lines.

    5) Am unable to capture any ringing or ripples on SPI lines or power lines using scope. Though I might add, scope bandwidth is not great, so I might be missing things.

    Shift Registers get RESET,

    1) 90% of the times when I start the Motors, any of the three.

    2) Some 10% of the times, when motors are up & running.

    3) This happens even when Motors have no load on their shafts.

    I am attaching the schematic.

    Thanks,

    Saumitra A Deshpande

    motor-Motor.pdf

  • 0 in reply to Saumitra Deshpande
    TI__Mastermind 26065 points

    Hi Saumitra,

    Thanks for you info and sch. For the layout, one thing I find is the GND was splited but all the inputs control signals are cross the GND gap without dedicate return path. The signal return has to merge into the PGND motor driving current path. So I think the GND split is not necessary and it is possible to cause logic issue when the peak current happens.

    Could you try to bridge the splited GND and do some new tests?

    Best regards,

  • 0 in reply to Wilson Zuo
    TI__Mastermind 26065 points
    Also please check if there is any nFAULT signal came out in the abnormal time.
  • 0 in reply to Wilson Zuo
    Prodigy 210 points

    Hi Wilson,

    1) All x595 are supplied by 3.3V/DGND supply lines. This 3.3v is generated off-board & supplied to this board

    2) All DRV8825 connected to +24V/PGND lines.

    3) There is single point of connection between DGND & PGND as following image shows... R9 is 0E resistor simply to specify single point connection between 2 grounds.

    Is this okay?

    Thanks,

    Saumitra A Deshpande

  • 0 in reply to Wilson Zuo
    Prodigy 210 points
    Hi,

    Forgot to mention... nFAULT line doesn't get asserted. But since all x595 outputs are set 0, nRESET pin of DRV8825 gets asserted.

    Saumitra A Deshpande
  • 0 in reply to Saumitra Deshpande
    Prodigy 210 points

    Hello Wilson,

    Any advice for my problem?

    Thanks,

    Saumitra A Deshpande