• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 28 subscribers
  • Views 410 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Original question:

DRV8871-Q1: Some motors work some don't . Current regulation setting and inrush limit

DRV8871-Q1: Avoiding inrush problem

Helmut Aretz
Prodigy 10 points
Part Number: DRV8871-Q1
Other Parts Discussed in Thread: DRV8871, DRV8872

I have developed a board that is intended to drive different types of motors. One application is a "smart" servo motor which has an integrated controller and an RC (PWM) interface. It draws approx 200mA during normal operation but stall current might be 1.5 amps. R_lim is set for 2 amps, but inrush is unknown.

My false assumption was that the outputs of the DRV8871 can be switched ON/OFF statically e.g. to supply power to pretty much anything, like a load switch with two mutually exclusive outputs.

I am now finding that the servo does not switch on, and instead I am getting approx. 2-3 volts between OUT2 and GND with a pattern that seems to repeat every 3ms or so. My assumption is that the inrush current of the servo (which probably has a capacitor across the supply pins) trips the over-current protection. Note that the resistor R_lim is 33K to limit current to approx 2 amps during normal operation.

The only method I found to start up the servo is by inserting a current limiting component in the path, e.g. a 2 ohm power resistor, or a coil (10uH works sometimes, 470uH works always). But then I am getting the expected voltage drop across the component when the servo moves, and I am worried the servo might reset or the component will burn out during stall (e.g. 1.5 amps stall will require a 2 ohm resistor to momentarily withstand 4.5W). Since the boards are very small (1x1.5"), if they have to be modded, space is an issue. I am currently experimenting with "external" components, but still need to find a solution for the space constraints when reworking the boards. An off-board solution is not an option - only to find a working component.

Do you have any other suggestions, ideally without using any larger components?

I have the following ideas, not sure if they make sense:

- Use fast PWM to overcome the inrush phase, then, switch to static on when stable (after a few milliseconds)

- Start with both inputs HIGH, maybe the DRV8871 doesn't like the immediate H/L combination on startup.

  • 0
    TI__Guru 54645 points

    Helmut,

    Certainly does seem like OCP protection is kicking in especially with 3ms matching the recovery time.  Do you know how large the capacitor is across the outputs?  If too large, every time the output switches, there is the potential for OCP to happen in which case the only thing to avoid this would be to isolate the capacitance with an inductor.

    Regards,

    Ryan

  • 0 in reply to Ryan Kehr
    Prodigy 10 points

    Thank you for confirming my suspicion. I do not know the capacitor across the supply (it is assumed to be inside the servo), I do not have an additional cap in parallel on my board. I don't know but I might be able to get this information from the supplier.

    Can you explain in more detail what you mean by "isolate" the capacitance with an inductor? Is this basically what I have been trying by inserting an inductance in the supply path? What's a good value and wattage per your experience?

    Do you have any comments on using PWM during startup strategy? Is it worth trying?

  • 0 in reply to Ryan Kehr
    Prodigy 10 points

    From supplier, the inrush may be for the internal motor only:

    1. Actuator’s Capacitance (servo inside) across the power supply pins : About 37~38[uF]
    2. Actuator’s inrush current : About 18~20[mA]
  • 0 in reply to Helmut Aretz
    TI__Guru 54645 points

    Helmut,

    I was referring to placing the inductor between the outputs of the motor driver and the load.  37uF is very, very large.  The outputs switch with a high slew rate.  I = C dv/dt.  This most likely explains the OCP.

    Regards,

    Ryan

  • 0 in reply to Ryan Kehr
    Prodigy 10 points

    I have tried to supply just a 47uF capacitor and the voltage reaches the target level after a few OCP attempts. However, if the servo is connected, it won't work. The servo supplier claims they cannot replicate the issue with a DRV8872. I would like to confirm this, but I cannot find these parts available right now. Can TI provide a sample chip, so I can test it and order larger quantity later?

  • 0 in reply to Helmut Aretz
    TI__Guru 54645 points

    Helmut,

    It appears, as you mentioned, that we are out of stock for the samples.  

    The only solution would be to order an EVM.

    https://www.ti.com/tool/DRV8872EVM

    Regards,

    Ryan