• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 342 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.

LM5170-Q1: LM5170-Q1 - Current Sharing

Chris Salvo
Prodigy 20 points
Part Number: LM5170-Q1
Other Parts Discussed in Thread: LM5170

Hello.

I am using the LM5170 to control a 4-phase buck converter.  The buck is 48V/36V 1 kW.  The controller is regulating the output voltage and it is able to handle transients, but  I am having issues currently with current sharing among the four phases.  My phase 1 and 2 share one controller and my phase 3 and 4 share another controller.  Current is balance between phases 1 and 2 and current is balanced between phases 3 and 4, but the current balance is not balanced between the controllers.  Phases 1 and 2 have way more current in them than phases 3 and 4.

I am using 1 voltage loop to set the ISETA pin for all four phases.  I was wondering if I should be using 2 separate voltage loops, one for each controller?  I do not know if the references are slightly off in the controller and if it is causing a large discrepancy between both controllers. 

  • 0
    TI__Guru* 78400 points
    Hi Chris,

    Thank you for using the LM5170. You should NOT use two voltage loops. All LM5170s in a system should be controlled by a single voltage loop. Therefore, each will see the same ISETA and the each phase will be regulated against the same ISETA. Consequently the current will be evenly balanced.

    The problem that you have seems to be the ground issue. Your two LM5170s may not sit on the same potential. Use a wire to directly connect the AGND nodes of the LM5170, and this should overcome the error and bring back the current balancing among all four phases.

    Thanks,
    Youhao Xi, Applications Engineering
  • 0 in reply to Youhao Xi
    Prodigy 20 points
    Thanks for the reply Youhao.

    I did try this and it did not work. However, looking at my layout for my control circuit, most of the ground seems to all be shared, I did not split up the analog ground and power ground. If pretty much all of the grounds (power and analog) are connected in the controller area, could that lead to the same issue? Looking at my layout it probably needs to be redesigned after looking at the layout suggestions in the datasheet.

    Also, another separate question. I am trying to run the controller at 500 kHz (max Fsw), when I set the Fsw resistor to 8 kohms, it never reaches 500 kHz, it is always below. Right now it sits at ~482 kHz, which isn't a big deal, I am just wondering why the Fsw is a little lower then the set point.
  • 0 in reply to Chris Salvo
    TI__Guru* 78400 points
    Hi Chris,

    Yes, I believe that caused the problem. The LM5170 has high accuracy to regulate phase dc current against ISETA signal. If there is ground potential differences among the ICs, the same ISETA signal will become different for each IC, so the phase current will be different. When you redo your PCB, please follow the instructions in datasheet, and also refer to our EVM layout.

    Regarding the frequency, 482 kHz should still be inside the performance range. The oscillator programming has about +/-10 % tolerance. In this case, you can my reduce the resistor slightly to fine tune the frequency.

    Thanks,
    Youhao