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: Hot Swap / No Load Failure

Part Number: DRV8871

Hello,

 

We are using the DRV8871 to drive a brushed DC motor using motor with the following settings:

* Load = Brushed DC motor

* VM = 40V

* Frequency = 5kHz

* In1 duty = varies between 10% and 100%

* In2 duty = 0%

* Imot = Between 1.3A and 2.8A

* ILIM = 20k

* Test Duration = ~10 minute intervals with fixed In1 duty cycles. 

 

We were able to execute several tests where we ran the motor at a fixed In1 duty cycle for approximately 10 minutes. 

After each 10 minute test we adjusted the duty and ran another test.  This was performed for several hours over the course of several days.

During one test the DC motor was prematurely disconnected from DRV8871 prior to disabling the drive and turning off power (In1 was at logic high, In2 was at logic low, and 40V was supplied to the DRV8871).

Within 5 minutes, the DR8871 failed resulting in low resistances between the following pins: IN2, OUT2, VM, and GND.

 

We understand that this may be the result of user error by removing the DC motor prior to turning off the system, however, we do not want to dismiss other sources leading to the failure.

 

It is peculiar that the device would fail when operating with no load as there is no current path to cause heating in the high and low side transistors.

If it were due to back EMF when the motor was disconnected we would have expected immediate failure instead of the ~5 min it took for it to fail. 

Is there something about driving this IC with no load that overstresses the part?

Any thoughts on how/why this failed in an unloaded operating state would be appreciated.

Thank you.

  • Hey Timothy, 

    Was this using the EVM board or your own schematic?  If it's your own schematic posting a picture of the motor driver circuit could be useful for our debugging.  

    Do you know what duty cycle it was at when it failed?  The datasheet says you can set IN1/IN2 to a static voltage for 100% duty cycle (3.1 Bridge Control), but the note under 6.3 Recommended Operating Conditions  discusses the 800ns pulse width requirement.  Granted, for a 5kHz signal I think it would have usable duty cycle range up to 99.6%. Either way, I wouldn't think that would cause this failure.

     I would also be curious to see how much current the chip draws with no load in your setup.  I don't know of any reason that this chip would fail in an unloaded state like you say, but I will ask my team if they have heard of anything like this as well.  

    Regards, 

    Jacob T. 

  • Hello Jacob,

    We are using an EVM board from Adafruit (Adafruit PN 3190, Digikey PN 1528-1719-ND).

    Leading up to the failure, we were testing with a static 100% duty cycle (In1 full on, In2 full off).  Also of note, the Arduino we were using to control the InX lines was also damaged from this failure. 

    We do have one more spare EVM and Arduino, however, we are airing on the side of caution due to parts availability.  I will talk with the team to see if we can get a current measurement.  Please let us know if you get a response from your team.

    Thank you,

    Kind Regards,

    -Tim

  • Hey Tim, 

    We are leaning towards suspecting inductive kickback when the load was disconnected.  Agreed on the 5 minutes being weird, as that should have damaged the chip when it happened not several minutes afterwards.  

    Here's a video about driving inductive loads:  https://training.ti.com/driving-inductive-loads-power-switches

    TI does have several chips with Open-Load Detection that might be of interest to you, though they are all only rated up to 40V - The DRV8873 or DRV8243 might be of specific interest to you.  

    Cheers,

    Jacob T.