• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 29 subscribers
  • Views 2476 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.

DRV8711: Pre-driver fault occurs everytime

user4873309
Prodigy 80 points
Part Number: DRV8711
Other Parts Discussed in Thread: , CSD88539ND

Hi, 

I'm using DRV8711 to drive multi types of stepper with 48V power supply, and the rms current range is 1.5A~5A, these motors need to work in variable speed environment with high frequency changing, I test them with increasing speed from 10 rpm to 600 rpm and decreasing back to 10rpm in loop, but I got PDF occurs on uncertain time after starting, maybe one or more minutes, even sencods.

I placed 10ohm~100ohm resistor on the low side gate, and changed the DTIME sametime, tried lots of lots of combination of IDRIVE and TDRIVE, it havn't solved the problem until now.

Another problem is I couldn't run the motor at All Mixed Mode with variable speed, but the Slow decay/Mixed Mode(default) works. I would prefer the smooth behaivor with All Mixed, but I failed due to change speed.

Here are to types motor I'm testing with(working on 256 microstepping):

  1. Irms=1.5A, R=4.3Ohm, L=15.2mH
  2. Irms=5A, R=0.48Ohm, L=3.6mH

Here is a bunch of parameters with motor 1(speed changing loop):

GAIN=5

TORQUE=10

TOFF=60

TBLANK=20~50

TDECAY=5

ABT=0

DECMOD=1

It could run under many configurations of IDRIVE and TDRIVE, but ends with PDF.

These problems torture me about half of month, did I miss someting important? I tested the parameters with 30V voltage power, and it works fine, but I need high power driving for higher torque. I designed board follow with BOOST-DRV8711. I tried to place 0.1uf and 0.01uf capacitor on the VM pin of MOSFET(CSD88539ND), nothing better.

Hoping get your advice, thanks.

  • 0
    TI__Guru** 114905 points
    Hi user4873309,

    What PDF do you get (APDF, BPDF, or both)?

    Please try the following:
    Place 120 Ohm resistors in series with the low side gates.
    Set the dead time to 850ns per section 8.1.2
    Lower the gate drives to IDRIVEN = IDRIVEP = 100mA
    Increase the TDRIVE to 2us

    If you still see PDF faults, please use the nFAULT to capture the gate voltage of the low side FETs.
  • 0 in reply to Rick Duncan
    TI__Guru** 114905 points
    Hi user4873309,

    Sorry, I neglected your second question.

    Do you need 1/256 microsteps? If not please try decreasing the number of microsteps with all mixed or auto mixed decay.
  • 0 in reply to Rick Duncan
    Prodigy 80 points

    Hi，Rick

    Thanks for your reply.

    I need to use 256 microsteps, yes. and I tried your parameters, tested with 47 Ohm and 120 Ohm resistors on low side gate. I still get the error and even run less than one second in All Mixed Mode,  and the PDF occurs ramdonly on two channels, here are waveforms what I catched when nFAULT is setted(two low side gate voltage in one channel):

    With 47 Ohm resistors:

    With 120 Ohm resistors:

    I could run these parameters with 30V nicely, but PDF occurs less than half second after power-on when using 48V. Is this the problem caused by the spike? How could I reduce them if it is.

    Hope get your reply, thanks.

  • 0 in reply to user4873309
    TI__Guru** 114905 points
    Hi user4873309,

    This does not appear to caused by the spike.

    Do you have a scope with more channels or a second scope? If so, please provide all 4 GLS signals with respect to the nFAULT asserting.

    Assuming the scope is triggering on nFAULT, please move the position of the trigger two divisions to the right. This will allow observing more information prior to the fault.
  • 0 in reply to Rick Duncan
    Prodigy 80 points

    Hi Rick,

    Sorry I only have one 2 channels scope, although I measured the voltage from one channel, but I would check the nFAULT register to ensure the fault occurs on this channel. 

    Here are latest measurement with constant speed, and they were in two cases(error channel or normal channel):

    Occurs with fault:

    Indicate anthoer channel occurs with fault:

    I tested times, and the scope shows simillar result. If the problem is not casued by spike, so what would it be, do the waveforms could show the problem?

    Thanks.
    Zheng Liang