• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 30 subscribers
  • Views 2205 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.

DRV8871: Occasionally Burns pin out using esp32

Adrian Logan
Prodigy 20 points
Part Number: DRV8871

I am encountering an issue with about 10% of my PCBs where after the first cycle of turning a 24v DC motor on then off I seem to kill the pin attached to the IN pin(s). After this point, we no longer see the voltage go high on the in pin and can no longer turn on the motor. There are no additional components between our microcontroller (ESP32) and the DRV8871. We also don't have any additional capacitors for EMF from the motor. The ESP pins are max 3.6v 20mA. Any advice on how to improve our design to mitigate this issue would be much appreciated.

Schematic:

Layout:

  • 0
    TI__Mastermind 37683 points

    Hi Adrian,

    One reason why IN1 and IN2 would be damaged is if its absolute maximum rating of 7V is exceeded. Electrostatic discharge would be another cause for the pin damages. Are there any possibility of ESD happening in your system?

    Another potential issue I see in your layout is having the IN1 and IN2 signals very close to the high power 24V plane. It is recommended to not place any sensitive low voltage signals near high power traces which are often very noisy. Noise from the high power trace can potentially couple onto the low voltage signals. If the noise on the IN1 and IN2 signals are greater than 7V, the pins could get damaged.

    I recommended reading this app-note and watching this training video to learn more about best practices for PCB layout of motor driver systems.

  • 0 in reply to Pablo Armet
    TI__Mastermind 37683 points

    Hi Adrian,

    Any updates from your side?

  • 0 in reply to Pablo Armet
    Prodigy 20 points

    Hi Pablo,

    Thank you for your advice. Here are some of my thoughts after reviewing your recommended reference material.

    I am thinking that increasing the clearance between the signal traces and the 24v plane and adding bypass capacitors (0.1uF or another value that you may recommend) to the signal traces may be the simplest revisions that require the least changes to our current design. Please let me know if you think that these changes may be sufficient to reduce the inductive noise. If we were to be more drastic in changing our design, I see that we should eliminate the 24v plane on the top and directly route the power and instead use a ground plane on the top and bottom. As a part of this process, we could change to a partition ground scheme. This would also allow me to significantly improve our thermal performance (though we have not seen thermal issues at this point).

  • 0 in reply to Adrian Logan
    TI__Mastermind 37683 points

    Hi Adrian,

    My suggestion is to have as much separation between the noisy high power traces and the sensitive low voltage signals to prevent noise coupling. The picture below shows a layout example.

    If you have the time, I suggest replacing the 24V plane with a GND plane and directly route the VM trace to the IC. However, you don't have to partition the analog and digital GNDs since you are not dealing with very high power. You can keep one common GND.

    You may take a look at the DRV8871EVM layout files (https://www.ti.com/lit/zip/slvc633) as a reference.