DRV8306: DRV8306HRSMR芯片的nFault故障信号引脚在开机的时候，有一个异常的低脉冲，约770us

Hi Yuqiang,

Thanks for sharing this waveform with us - it really helps.

From looking at the waveform of all 3 signals, we see that your nFAULT low-pulse occurs at around the same time as VM power-up. 

We think that the observed nFAULT low pulse is not a problem, and here is the reason why:
1) At initial device power-up, the nFAULT pin will transition from high-to-low since VM and the charge-pump are not yet powered on completely. This is the device's UVLO protection. 
2) Once fully powered on, the nFAULT pin recovers from low to high automatically, indicating that the device is ready-to use. (UVLO fault condition is cleared).

If your motor driver system is able to drive the motor successfully after power-up, then this temporary nFAULT low pulse is expected and normal.

Image of datasheet reference to charge-pump UVLO: 

Please let me know if this helps answer your question, and if there are any further concerns.

Thanks and Best Regards,

Andrew 

Hi Yuqiang,

Thanks for your patience - please see latest responses below:

1. nFAULT signal at VM startup sequence
   1. The best and likely explanation for the nFAULT waveform in your scope capture is the device's normal power-on sequence and UVLO feature.
   2. Are there any other unresolved problems that occur after the device is finished powering on, that would be a cause for concern?
2. Short-circuit test data
   1. For a phase-to-phase short circuit condition (SHA short to SHB), the DRV8306 device is designed to trigger an nFAULT flag for either Vds OCP or Vsense OCP.
   2. In your waveform, the OCP is appearing to work properly, since the phase current starts lowering after some ~5uSec time (close to Datasheet typical value)
   3. One clarification about the deglitch time measurement is the start time cursor should be placed
      1. Vsense OCP and Vds OCP features will start their deglitch time period when a large-enough voltage difference is seen on Rsense or Vds to trigger the DRV8306 comparators.
      2. There is some delay between the very start of the current spike vs when excessive current is large enough to be detected.
      3. This would bring the OCP deglitch time measurement closer to datasheet 4.5uSec value, which is just a typical value (not a min/max limit).
3. Hall sensor HPC-pin voltage
   1. This is a question that has been resolved before on another e2e thread for DRV8306 - please see the link below
      1. Link: https://e2e.ti.com/support/motor-drivers-group/motor-drivers/f/motor-drivers-forum/985855/drv8306-hpc-pin-voltage-hall-voltage-is-not-aligned
   2. In summary, there is a parasitic path on the HPC pin that causes a slight voltage drop that impacts HPC pin voltage due to clamping.
   3. The design team has confirmed that this will not affect the hall logic, as the DRV8306 device will correctly interpret the hall signals even at the HPC voltage level described

Thanks and Best Regards,

Andrew

Hi Yuqiang,

Please see the updated response below to your questions:

1. Damage to external MOSFETs and motor
   1. for this question, I think it would be best to refer to the MOSFET and motor datasheet ratings.
   2. identifying damage, if any, would have to be done by checking individual components on the board-level near the affected signal path.
   3. it my also be a good idea to test the MOSFET's drain-source resistance values when it is on/off.
2. Overcurrent protection setting
   1. Tying Vds pin to AGND directly would produce the lowest threshold so that a Vds_OCP type fault will trigger more quickly in theory. (refer to datasheet table below).
   2. for Vsen_OCP type fault protection, only one setting is available and that is determined by the Rsense resistor value. 

Thanks and Best Regards,

Andrew