• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 84 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.

LM51772: Buck-boost selection for new product

Anthony Sotillet
Prodigy 20 points
Part Number: LM51772

Tool/software:

Hi folks

I’m working on a power stage that must accept 12 V to 50 V DC and provide four identical outputs that the user can program to 12 V, 24 V, or 48 V. Each rail is limited to 2.5 A by a TPS4811 circuit breaker. The loads are sensitive VHF/UHF RF modules. I select the LM51772 but would like to hear practical advice from the community.

My question is architectural.Would you recommend a single LM51772 configured for 48 V at 10 A feeding four eFuses and local LC filters, four individual LM51772 converters each set for 2.5 A, or a hybrid scheme with a front-end boost followed by separate buck stages? Efficiency and EMI at the low-input extreme of 12 V (where the system draws roughly 40 A) are my main concerns.

Best regards!!

  • 0
    TI__Guru 58621 points

    Hi Anthony,

    thank you for using the E2E forum.

    The LM51772 would fit very well to your configuration as far as I understood your requirements.

    On the other hand I need to say that I am also not sure if I full understood that.

    So if you need 4 voltage outputs on 4 you also need for power stages. So how to you think this could be done in one.

    When doing it in one power stage this might give you issues with handling the thermal losses and would need addtinal cooling.

    Doing a 2.5A output should not be an issue if the design is done properly

    Best regards,

     Stefan

  • 0 in reply to Stefan Schauer
    Prodigy 20 points

    Hi Stefan,

    thanks for your fast reply

    Let me clarify the two architectures we are evaluating:

    1) ONE buck-boost stage
    – One LM51772 delivers the three posible outputs (12, 24 and 48 V), the 48V output with the 12V input is the worst case scenario. It will put the highest stress on the converter.
    – The single rail is then split into four branches, each protected by a TPS4811 eFuse and its own LC filter.
    – All four loads always share the same voltage; the MCU updates the setpoint through I²C.

    2) FOUR buck-boost stages
    – One LM51772 per branch, each limited to 2.5 A.
    – All devices receive the same I²C command so the four outputs track (12 V, 24 V or 48 V).
    – Advantage: lower current per stage and true channel redundancy.

    I was asking whether option 1 is still considered “good practice” with the LM51772 if we size the inductor and manage the thermal path properly, or if it is recommended one controller per output.

    Follow-up questions:

    • For on-the-fly voltage changes, is the sequence “disable → change FB via I²C → enable with soft-start” the safest way to avoid overshoot?

    Hope this clarifies the scenario. Your guidance on which option is better from the EMI and thermal standpoint would be highly appreciated.

    Best regards,
    Anthony

  • 0 in reply to Anthony Sotillet
    TI__Guru 58621 points

    Hi Anthony,

    so you are targeting an output power of 48V * 2.5A  * 4 = 480W for the first power stage.

    As mentioned if done in one power stage this needs special thermal management if done in a single phase (heat sink, cooling).

    When targeting passive cooling without heat sink you should consider multiple phases.

    So i think the 2nd approach would be the better option from thermal perspective.

    The quickstart calculator can help you do get more insight into the expected losses. ( LM51772 Design Calculator)

    Best regards,

     Stefan