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.

DRV8844: driving with load disconnected

Part Number: DRV8844


I'm using a DRV8844 in a configuration where I parallel the outputs to drive a single BDC motor bidirectionally. It had been working fine so several weeks, until we wanted to do some testing on another part of the system that this motor/driver is a part of, and the easiest way to move forward with our testing was to disconnect the motor (to prevent it from moving unintentionally). After a few days of testing, we observed a board failure that seems to have stemmed from the DRV8844. I'm trying to figure out how the failure happened, and was wondering if there's any risk involved in commanding the DRV8844 to drive a motor that's disconnected? I can't find any info in the datasheet regarding protection against open load.



  • Hi Ryan,

    There is no risk commanding the DRV8844 to drive with no load. There is no protection against open load because there is no current flow.

    Assuming you are using a single supply, please look for any potential absolute maximum violations in the system when performing the other testing.

  • It's definitely possible that the main power rail (Vm on the chip) saw some transient positive spikes above its max allowable voltage (65V). The system is running at 54V, so when the other motors turn off there may be some power supply regeneration (this hasn't been investigated yet but I imagine there are at least some spikes, but I don't know at what level). I have a TVS on my board (the one with the DRV8844 on it), right at the input rail, with a min breakdown voltage of 64V and a max of 71V. So depending on where the actual component was in that range, it could have been clamping above the absolute max for the DRV8844. Do you know if that 65V listed in the absolute max rating for the chip applies to transients as well, or if it's only for continuous supply voltage?

    Another piece of information, in case this is useful for diagnosing the failure mode:  apparently when the failure initially happened, a wisp of smoke was seen to rise from the board (I wasn't there at the time so I just got a second-hand description). I saw that the board's fuse (on the positive rail) had blown, so I replaced it and repowered (after checking to make sure there were no obvious shorts), and upon repowering, the DRV8844 chip sparked and smoked, around its pin 11 VM connection. Can you say whether a failure localized at that pin would seem to point towards an positive supply overstress failure?


  • Hi Ryan,

    The absolute maximum is for both transients and continuous supply voltage. From the datasheet absolute maximum footnote:

    "Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability."

    The wisp of smoke may have been the initial damage to the DRV8844 (and the fuse protected the device from further damage). Replacing the fuse allowed the device to spark and smoke.

    When a device experiences this type of damage, it is usually difficult to pinpoint the cause. One suggestion is to operate the device at a lower voltage if possible and look for spikes around the device. Doing this can lead to possible reasons for the damage.

    During the unloaded testing, there should have been few voltage spikes.

    Is it possible that something was accidentally shorted, such as CP1 to CP2, VCP to GND, or V3P3OUT to something?

  • As far as I can tell, I didn't see any other shorts.

    Unfortunately we can't lower the main supply rail, since it would have an impact on the other devices. I'm going to swap out the TVS on my board for a model that clamps at a slightly lower voltage, to hopefully give a little more buffer, and I'll also try to get a scope on the supply to see what kind of spikes it might be seeing.

  • Hi Ryan,

    Thanks for the update. I hoped you could lower the main rail, but understand it is not always possible.

    Please keep us updated on what you find.