• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 622 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.

DRV8705-Q1: Application with motor drivers in parallel

Henrik Leegaard1
Prodigy 140 points
Part Number: DRV8705-Q1
Other Parts Discussed in Thread: DRV8874

Hello MDBU team,

I have a question regarding the DRV8705S-Q1

In my application I plan to use two H-Bridge and gate drivers in parallel (DRV8705 + MOSFETs and DRV8874).
They will drive the same motor.

The DRV8874 is only used in an emergency case and the DRV8705 is the main driver used for normal use.

In the application the supply for the DRV8705 + MOSFETs can be switched off by a MOSFET and the supply for DRV8874 can also be switched off by a MOSFET.

My concern is if there will be any problems or leakage current if f.ex DRV8874 is powerless (supply switched off) and DRV8705 is driving the motor in any direction.
I assume there will be a leakage from OUT1/OUT2 of DRV8874 to the supply pin through the body diode of the high-side MOSFETs in DRV8874

But will there be any unforeseen problems or will the DRV8874 be in Hi-Z?

Same question regarding the emergency case where DRV8705 is powerless (supply switched off) and DRV8874 is driving the motor is any direction.
Will this give any unforeseen issues with DRV8705?

In the application HW is implemented so the DRV8705 and DRV8874 can never be active at the same time.

Hope you can if you can comment if the approach is a no go and the concept is not functional or if the concept is ok or ok with minor tweaks

Looking forward to hearing from you

 

Regards

Henrik Leegaard

  • 0
    TI__Guru 56650 points

    Henrik,

    I don't see any issue with this approach if the DRV8705 and DRV8874 can never be active at the same time.

  • 0 in reply to Wang5577
    Prodigy 140 points

    I didn't assumed either, but I am concerned about the following.

    1. When (with reference to attached drawing) DRV8705 is active (SW2 enabled and SW1 disabled) voltage will run backwards in DRV8874 through the high-side MOSFETs from the output to SUPPLY-SEC.
      1. Will this be an issue for DRV8874?
    2. And when DRV8874 is active (SW1 enabled and SW2 is disabled), voltage will run backwards in the H-Bridge through the high-side MOSFETs from the output to SUPPLY-PRI
      1. Will this be an issue for DRV8705?

    I hope that you can comment my concern. H-Bridge.pdf

  • 0 in reply to Henrik Leegaard1
    TI__Guru 56650 points

    Henrik,

    If SW1 or SW2 is off, it could possibly violate the abs max of VCP with respect to VM. So, we can short out SW1 and SW2 to remove this concern and reduce the power loss of SW1 and SW2.  We can use nSLEEP pin to control DRV8874 and DRV8705 to set the motor driver in HiZ mode.

    Only other concern now is how the current will re-circulate (through the body diodes of the external FETs or the DRV8874). This could create problems if the current through the external FETs is higher than rated current of the DRV8874 and flows through the body diodes of the DRV8874. Adding external schottky diodes should help eliminate that as a concern.

  • 0 in reply to Wang5577
    Prodigy 140 points

    Wang li

    The two switches can't be removed as they are implemented for safety reasons.

    The application is in a medical device so the switches are implemented in other not to have any safety issue at single faults.

    I wonder if I added four diodes in the circuit. One from Motor+ to SUPPLY-SEC, one from Motor- to SUPPLY-SEC, one from Motor+ to SUPPLY-PRI and one from Motor- to SUPPLY-PRI wil solve the issue/concern
    The inactive driver will then be in sleep/Hi-z but not powerless.

    Will this approach be doable without violating the DRV8874 and DRV8705 circuits

    The current through the diodes will only be the sleep current of the inactive driver so they don't need to be that high performance

    Looking forward to hearing your comments.

  • 0 in reply to Henrik Leegaard1
    TI__Guru 56650 points

    Henrik,

    It looks good to me. But, please verify it before run mass production.

  • 0 in reply to Wang5577
    Prodigy 140 points

    I will veify before mass production.
    I also need to verify if the switches can handle voltage on the output when they are in off-mode