• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 235 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.

LMR38020: Parallel operation

David Reed1
Intellectual 365 points
Part Number: LMR38020

As the name suggests, I am looking to parallel two of these to get 3A. I was planning to diode-OR the outputs and use a single voltage divider to generate the reference feedback. 

  • Would you recommend Diode-OR'ing the outputs?
  • Could the outputs be tied directly together if the same feedback is utilized?
  • Would you recommend placing small value resistors between the output nodes instead?

Thanks... 

  • 0
    TI__Mastermind 23845 points

    Hi,

    It may be possible but the answer would not be trivial with just a diode - OR since nothing control the balancing of the current. 
    You may want to refer to this app note on another device that using external components to balance and control the current
    https://www.ti.com/lit/an/snvaa21/snvaa21.pdf?ts=1644451019571&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FLM62460-Q1

    I hope that helps

    Thanks

    -Arief

  • 0 in reply to Arief Hernadi
    Intellectual 365 points

    Hi Arief,

    Thanks for the app note. The downside I see to that potential solution is the need to power the op-amp before the switchers are powered up.  If that is really required for stable operation that adds an additional supply just for the op-amp and  also requires power sequencing.

    Diode OR ing somewhat self regulates with the voltage drop across the diode increasing as more current is drawn from that branch. The question would be is that sufficient to get it "close enough" for my application.  I have ordered the eval boards to do some testing and am also looking at potential ways to divide up the loads and have multiple rails. 

    It is not clear from your response if  your concern for the imbalance is to support full load operation. If that is the case I would never expect to see 100% load on both devices and therefore was not too concerned if one side supplied 60% of the load and the other side 40%. 

    How about simply using low value resistors to sum the two outputs? Would you recommend that approach instead of the diodes?

    Thanks,

    Dave

  • +1 in reply to David Reed1
    TI__Mastermind 23845 points

    Hi David, 

    I think the switcher can be powered up first before the op amp because the op amp circuitry is needed to be active when you want to share the current. 
    My concern is that one part may hit current limit first before the other one. For example IC1 is at 2.5A and IC2 is at 0.5A. I know that these may be an extreme case but that was my concern since one IC is "blind" to what current the other IC is delivering. 

    You may have some kind of current balancing with a resistor or a diode OR ring but again that may require some testing and how much you can tolerate in terms of current differences between the two ICs. But again the method may need to be tested in prototype in a condition such as short circuit, overload, startup, etc.

    Thanks

    -Arief