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.

DRV8323R: Rotating motor on power down

Part Number: DRV8323R
Other Parts Discussed in Thread: DRV8323,

Hi,

we have a motor where the user can rotate the motor when system is powered off.

Have used three relays on each motor winding but would like to use the coast feature to reduce BOM.

Now when the motor is rotated the power generated turn on the led, can we expect the entire board to turn on?

What can we do to prevent this or at least start the motor in coasting mode.

And another question, when placing an Emergency stop button, should this disconnect power to entire board or just VM/motor?

  • Hi Martin, 

    When the motor is in generator mode, there are two things to look out for:

    - SHx voltage conducting through HS MOSFET's body diode and backpowering the DRV8323 into the VM pin
    - SHx voltage not exceeding 15V so the VCP-VM voltage is <15V or else internal conduction paths in the DRV8323 could result in damage to the VM-VCP ESD diode rated for 15V

    The best way to miitgate any back powering and voltage pumping when placing the motor in coast is to ensure there is ample bulk capacitance at the supply and the drain of the HS MOSFET's to manage any energy dissipated from the motor. Having parallel bulk caps from 1uF up to hundreds of uF is sufficient, as well as minimizing any extra parasitic inductance at the powerstage to avoid voltage spiking and overshoots. 

    An emergency stop button should act as an active cutoff switch. We have an app note that describe how to implement different architectures of cutoff switches with motor drive systems: https://www.ti.com/lit/pdf/slva991

    Thanks,
    Aaron

  • Hi Aaron, thanks for your answer.

     

    I think I missed a crucial part in my first post.

    What “Powered off” in our case is batteries disconnected.

     

    Our emergency system is directly connected to a contactor that disconnects the battery from motor and controller.

     

    This means that the user can rotate the motor while batteries are disconnected.

     

    Today this is solved with relays that is normally open, so motor windings are disconnected from motor controller when e-stop is active

     

    The user needs to be able to rotate the motor when e-stop is active.

     

    Hope this is better explained.

     

     

     

     

  • Hello Martin,

    Thanks for the added details. We will review this tomorrow an provide some feedback.

    Best,

    Isaac

  • Hey Martin,

    Aaron is currently out of the office due to a holiday here in the US, so I will address your question for the time being.

    Typically from a coast we do not see enough power to turn on the entire system since it is only temporary, but if the user has access to keep rotating the motor manually then depending on your motor parameters this could potentially provide enough energy to power up the entire system.

    You mentioned that rotating the motor is enough to power an LED, was this power generated from a single phase or from all the phases combined? If your goal is to reduce BOM you might be able to reduce the amount of relays used to 2 or 1 dependent on the amount of energy being generated by each phase, quantifying how much power is produced by each motor phase will be key here. This would allow you to reduce the amount of relays used while still ensuring that the power generated by the motor does not accidentally turn on your whole system if a user keeps continuously rotating the motor.

    To address another question, if your emergency shutoff needs to shut power off to the entire board or just the motor driver/motor. Typically power is just disconnected from the motor driver. In a system, the microcontroller is usually left active since it may be running other operations and not just trying to operate the motor itself but this tends to be more application specific.

    Best,

    Isaac

  • Hi Isaac,

     

    Power comes from all three phases, tested today and it generates around 40V at max rpm(200).
    The user will be able to reach this speed but in very short bursts.
    This is enough to power up the MCU. The first thing MCU does on power up is activating coasting.
    I’m curious to know what happens when running it at max rpm?

    Not a big fan of a system that can power up while e-stop is active.

    Goal is/was to get rid of relay as they have finite switching and we would need to have them on a separate board as space is limited and in the case of failure this can be swapped.
    Since we plan to pot motor and pcb this will make it more difficult and expensive.
    But might probably be the only solution.

    E-bikes can be used without batteries, but these looks to have a clutch system so motor is not rotating while powered down?

    As the MCU only controls a single motor and nothing else just disconnecting the batteries on E-stop have worked fine for testing. DRV8323R is powering the entire board so disconnecting only the motor driver will also remove power from MCU.

    Can we expect damaging mosfet/drv8323R when suddenly disconnecting power while motor is rotating since commutation is stopped?

    Aaron sent a pdf for cutoff switch, “The NMOS Cut-Off Switch With Downstream Motor Driver” setup, this will probably give the same result as mention above unless rest of the system is powered by a separate SMPS and not the internal buck in DRV8323R.

    We are changing from DRV8323RH to DRV8353RS, this will give us more headroom regarding voltages.
    Is there a date for restocking of DRV8353RS, just looking for prototype amount(5-10pcs)?
  • Hello Martin, 

    The team is out of the office due to Thanksgiving holiday here in the US. We will be back in office on Nov 29th, thank you for your patience and understanding.

    Best,

    Isaac

  • Hello Martin,

    Thank you for your patience.

    You should not expect damage the DRV8323R when disconnecting power while the motor is rotating as long as your system is able to handle power generated by the motor as explained by Aaron above. Moving to the DRV8353RS will definitely give you any added headroom that is needed especially if your motor is generating 40V of BEMF.

    There is currently no info on the lead time for DRV8353RS devices, but there are EVMs available if you would like to start evaluating this device for your application: https://www.ti.com/tool/DRV8353RS-EVM

    Best,

    Isaac