• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 30 subscribers
  • Views 800 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.

DRV8301 Resetting

Zac Bogart
Prodigy 30 points
Other Parts Discussed in Thread: DRV8301, BOOSTXL-DRV8301

We are using the DRV8301 in a motor driver.   At high currents (13-16A) the 8301 is resetting as follows:

The contents of the Control Register reset to their default values

The Fault pin does not drop (stays high),  The Fault register does not indicate ANY faults (in any bits, nor in the "fault" bit via SPI)

 The OC mode is set to 01 so that the gate drives will completely shut down if we hit the current limit.  It's not happening.  (We have also tested OC mode at 3 so that we're sure it's not participating in this problem.)

The OC_Adjust_set is set very high (20) Rds-on =  13.5 mOhm  = 48A.  so it won't trigger.

We have read the AP note SLVA552 regarding noise on the DVDD line.  We have a cap (10 uf) directly on the Dvdd pin (thorugh a via) with the other side of the cap directly on the ground pad on the chip (thorugh a via).  DVDD is very clean: no spikes, solid 3.3V (ish).   (Per the datasheet, the only fault that will do what we're seeing, without reporting a fault,  is the DVDD undervoltage.)

EN_GATE is a bit noisy with narrow spikes coming down to about 2.5V  When the chip resets we do  NOT get a scope trigger on this line below 2V.

The ground pad on the chip is very well soldered to the ground plane.

PVDD is nominally 26V pulling down to 19V at 13A of motor current (5A coming from the PSU)  (The PSU is a linear, unregulated, bulk supply.)

Checking between the ground tab and other parts of ground plane proximate to the mosfets, we see about .8V of ground ringing coincident with the mosfet switching edges.  Some of it is false (airborne interference) but most of it is real.

We are using Infineon  IPU138N08N3  mosfets.  They appear to perform nicely.  Heating appears appropriate to the power being transmitted.  (They gett a bit too hot at 1.5W consumption)

Our belief now is that our board layout needs to be improved.  We will start that soon.  However, we prefer to find the "smoking gun" before we proceed.  We're looking for advice about where to look, what to probe, poke or bolster.   What can cause the chip to reset this way?  

Thanks for any advice.  


 

  • 0
    TI__Guru 55615 points

    Zac,

    The SLVA552 app note was written specifically to address this issue as it is quite common.  Our first EVM actually had this problem which prompted the application note. 

    The GNDs for DVDD/AVDD/GVDD/PVDD capacitors need to be closely coupled together.  We were able to fix a GND seperation issue on our EVM by simply taking a 24AWG wire and connecting GND together on all the capacitors and then to the PowerPAD connection to the main PCB GND.  This is shown in figure 9 of the application note. 

    Do you have any series resistors between the output gate drive pins and the FETs?  A 5-10ohm series resistor will help quiet the GND bouncing down as it will slow down the edges. 

     

  • 0 in reply to Ryan Kehr
    Intellectual 500 points
    We are experiencing the same problem with the BoostXL-DRV8301 board.
    I will assume that the layout considerations were implemented on this board. But the board has since been modified to drive our inverter. So the MOSFETs have been removed and short twisted pairs go from the BoostXL board to our MOSFETs. The ground of the BoostXL is connected to the ground of our inverter and is a twisted pair with the PVDD.
    After we try to jog the motor, the OCP value is reset from Report Only to Current Limit.
    Alan