• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 30 subscribers
  • Views 431 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:

Stepper motor controllers/drivers in parallel

DRV8899-Q1: Redundancy of the stepper motor driver

SlawomirM
Prodigy 50 points
Part Number: DRV8899-Q1

I am working on a Cubesat subsystem which main task is to operate with stepper motor. We are not able to use two motors (redundant motors) or dual winding motor. We have selected single stepper motor for this task (high-reliability, space grade). 

I am going to use DRV8899-Q1 to drive this step motor, but I need to have a backup driver for radundancy. My question is, can I add redundant DRV8899-Q1 and connect the outputs to the same windings? Redundant motor driver will be turned-off by default. In case of first driver failure, power supply to it will be cut off, and redundant one will be activated.

Do I need to separate outputs from each others? Or do I need to cut off also driver outputs? How to realise this? It's bipolar operation of the stepper motor. Should I use solid state relay configurated for AC operation, or load switch?

I will be grateful for any tips!

Best regards,
Slawek

  • +1
    Genius 3590 points

    Hi Slawek,

    I have never done such dual circuit though I have some thoughts about it.

    Issue 1. If you connect two drivers to the single stepper motor and operate one of them the second driver probably will get supply voltage (VM) through its Mosfets body diodes and it will be powered even if its power supply is cut off. This may cause some problems.

    Issue 2. In case one of the Mosfets in one driver is damaged and the damage is short circuit type the second driver will not be able to operate.

    Maybe switching motor between drivers with two relays would be an option. There are bistable relays that need power just for switching them like the one below

    https://www.tme.eu/pl/en/details/hfd2_012-s-l2/miniature-electromagnetic-relays/hongfa-relay/

    If you go for a relay there will be a few things to decide/check:

    - should relay be sealed or not ?

    - should contacts be gold plated or not ?

    - operating under load (it might happen) in working conditions (vacuum or some pressure, temperature)

    - will drivers survive should relays switch under load ?

    Regards,

    Grzesiek

  • 0 in reply to Grzegorz Pelikan
    Prodigy 50 points

    I can confirm that. I did some experiments, and second driver is conducting through the diodes.

    I have also found potential solution in this article:
    https://www.ti.com/lit/an/slva948/slva948.pdf?ts=1682391744613

    load switches/e-fuses can be connected back to back. When one is conducting, second one is conducting through the diode. That could work, next to proper isolation of the whole driver (I will cut off all the signals to be sure if the driver is damaged, it will be cut out from the circuit). Still I will just test this approach. 

    I was also thinking about solid state relays, but looks like e-fuses will be way more compact (it's small motor, <1A per phase). 

    thank You!
    Sławek


  • +1 in reply to SlawomirM
    TI__Mastermind 21095 points

    Hi Slawek,

    Thanks to Grzegorz who hit the key points to consider while connecting two drivers for redundancy. Glad you tried it. When an unpowered driver is connected (also in parallel to another active driver)  to a stepper motor and you manually spin the motor the back EMF (BEMF) will be conducted via the body diodes and appear on VM and if above the minimum operating voltage will power up the device. You can observe this by manually spinning the motor when connected to one of our EVMs or your circuit board. So some sort of bi-directional switches on motor terminals would be required.

    In addition to this all the signal pins have ESD protection diodes which would conduct when forward biased and provide biasing to digital circuit. While this would not damage the device if within the specifications the inactive device would not be truly unpowered. .

    That said, e-fuses are designed for DC hot swap applications. I'm not sure they'd be ideal for the motor output terminals especially to behave as a true switch all the way up to the current regulation chopping frequencies. You may want to check on this with the e-fuse experts at TI. It will certainly work for the VM power supply swap with suitable specs. For motor connections a 4-pole double throw latching relay (space grade) could be a solution. When considering SSRs please check with their frequency response while on.  

    Regards, Murugavel