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.

DRV5032: Output status

Part Number: DRV5032

Hi Bu:

My customer is evaluating the safety of the system which Is implementing the DRV5032FA , in one of the evaluation items, customer need to know the status of the DRV5032's output when cutting the ground connection to the device , like the below.

The output of 5032 is connected to the GPIO of the MCU , as my thinking , the output should be floating since it is the push-pull output structure .So the status should be reply on the external circuit of the output signal. It will be pull down to ground , am I right ?

  • Jason,

    Thank you for your question.

    You are correct.  With the push-pull output on the FA device, we should have a floating output if GND is not present.  Without an adequate supply path through the device the outputs should not be enabled.

    Thanks,

    Scott

  • Hi  Scott:

    Thanks for your reply.

    It's weird that the output is a stable 1.8V DC voltage actually when we power on 3.3V for the device and disconnect the ground , no matter we put the magnet close and far away from the device.

    Could you please check it in your lab ? Thanks!

  • Jason,

    On my lab bench, I have tried to replicate your experiment.  I configured my PCB with no pull-up resistor or output RC filter.  I set my multi-meter to monitor the voltage from OUT to GND and supply the PCB with 3.3 V.  With no magnetic input, I see the output drive to 3.3V as expected.  If I then remove the GND wire from the supply I observe the output will drop to 0V.

    I again repeated the experiment by powering on the supply and measuring the voltage with respect to the supply GND instead of PCB GND.  In this case when I disconnected the GND to the EVM, I saw the output float to about 2.5V for my unit since I now have different GND references.

    Thanks,

    Scott

  • Hi  Scott:

    Thanks for your test.

    Actually, the customer's test is a little different from yours, could you please replicate again?

    The below circuit is on the main board, 

    the rest in below is on another board which is integrated in to a module (the module is also including the magnet) , there are three wire out of the board ,

    1) VCC ( VCC PIN OF DRV5032)   2) out ( the left side of R3)  3) GND (The Ground of the DRV5032) 

    The DRV5032 is powered by the 3.3V from a LDO ( LDO's output connected to the VCC , and LDO ground is connect to the system ground , it is also on the main board).

    So if we disconnect the module ground ( the GND of the DRV5032) to the system's ground, and measure the signal between the right side of the R2 ( left side of R3) , it is stable 1.8Vdc. In this scenario, the magnet is also close to the hall in the module. 

    So could you please do one more test to simulate the customer's environment, you just need to set up the configuration as  customers, 

    and cut off the ground pin of drv5032, and measure the voltage between system ground and the R2.

    Thanks!  

  • Jason,

    I am inquiring about what might cause the result you have described and expect to be able to respond with more detail on Monday.  

    There are a few aspects of your setup that I do not think I could easily replicate in the lab (LDO and Microcontroller).  

    What I observed in my testing was that the Device GND and OUT both were pulled up to 2.5 V which is roughly a diode away from VCC.  The method of disconnecting GND was fairly similar to what you have described, however.  After observing that OUT and GND had the same potential, I then measured to system ground when the return wire was removed from the socket. 

    Is it possible for you to verify the voltage on OUT under the same conditions when the connection to the main board (between R2 and R3) is broken?

  • Jason,

    Thank you for reaching out with customer details directly through email. As we have transitioned this discussion from this E2E to allow more detail sharing, I will close this thread.

    thanks,

    Scott