• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 213 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.

LM5118 Load sharing

Alan Jacks
Prodigy 20 points
Other Parts Discussed in Thread: LM5118

Hi,

Is it possible to parallel connect two identical LM5118 circuits for load sharing? Both designs are currently frequency synchronised to each other in such that the SYNC pins are linked but I was wondering if there are other factors involved such as:

1. Does RT pin need a resistor on both devices 

2. Should both regulators use common feedback resistor and loop filter components (these are currently unique for each regulator circuit)

3. Should the current sense be common in some way?

Each circuit operates independently very well for >4A output loads and starts up well under full-load conditions.I have a requirement for a 7A (pulsed) radio system but my latest testing (based loosely on heat dissipation) suggests that the regulators are not sharing equally (if at all!) when the output is bridged.

Appreciate there may be better solutions but for morbid curiosity more than anything, would be good to see if this can be made to work or hear if there are any fundamental reasons why it shouldn't.

Thanks,

Alan

  • 0
    TI__Guru 58391 points

    Hi

    LM5118 is not designed for multi-phase load-sharing.

    Since there are two error amplifiers working at the same time, If you just connect the two ouputs into one, one converter will operate dominently unless the two internal error amplifier references are identical each other.

    If you connect two errror amplifier outputs into one, there is a case two error amplifiers are fight each other

    Regards,

     

  • 0 in reply to EricLee
    Prodigy 20 points

    Hi Eric,

    Thanks for your reply. I agree totally with your thoughts as I suspected this was the case.

    Just for reference though, what I can tell you is that my testing to date (on a reasonable production run) shows that the parallel configuration, when implemented, works well. Whether this is luck on my part remains to be seen or whether the layout and tight tolerancing of parts is playing their part is unknown. Further production batches may provide an insight.

    The interesting thing of note here is that a single LM5118 stage operates well into a 4.8A load but does not go much beyond this. This is satisfactory for my current design requirements though.

    I can draw 7.1A from a parallel configuration quite comfortably AND the measured efficiency of the regulator design improves from 90.6% (single stage) to 94.4% (bridged parallel stages).

    I agree totally with your comments. In the nature of R&D, I will continue to assess the performance as production continues. What buck/boost part would you suggest for higher current capability?

    Alan

  • 0 in reply to Alan Jacks
    TI__Guru 58391 points

    Hi Alan

    Thank you for sharing the information. For the higher current requirement, please refer 4-switch LM5118 design at http://www.ti.com/tool/lm5118-4-swxevm

    Also, you can contact TI's local FAE to see what's in TI's roadmap for the higher current buck-boost application.

    Regards,

    Eric