• State
  • Replies 3 replies
  • Subscribers 38 subscribers
  • Views 98 views
  • Users 0 members are here
Support feedback
Options
Options
Related

PMP41115: Efficiency optimisations

Rain Haase
Prodigy 141 points
Part Number: PMP41115


Tool/software:

Hi,

I am looking to design a board similar to PMP41115 that will operate with 30 - 40V battery voltage instead of 42V and 50V as shown in the reference design. The 240W output should be able to be sustained without hitting thermal limits for about 22 minutes, within a plastic casing that may or may not include additional heatsinking measures other than those related to the PCB (via stitching, 4 layers, 2oz copper) based on feedback here.

With these parameters, I've assumed that the main power losses are within the inductor and MOSFETs, so I have chosen alternative components that could yield greater efficiency.

Inductor CMLB136T-100MS --> B82559A7103A020 (lower DC and AC losses)

MOSFET SIR880BDP-T1-RE3 --> ISC0602NLS (higher Rds(on) but lower gate charge)

Could you please give me a rough estimate of whether these changes might be sufficient for the device to operate within these thermal conditions? Are there any other large sources of losses that I could reduce with bigger or more expensive components?

Thank you.

  • 0
    Prodigy 141 points

    I would like to further add onto this that I'm switching the capacitors from EEHZC1J560P --> EEH-ZV1J560V and PCR1J101MCL1GS --> EEH-ZU1J101V, with the same voltage and capacitance ratings but half the ESR and significantly increased ripple current (+50% for the former, +150% for the latter). With that, the inductor change, and the FET change, could I expect to make a good reduction in losses?

  • 0 in reply to Rain Haase
    TI__Intellectual 2160 points

    Hi, Rain

    Depending on your switching frequency, you can make a balance of switching loss and conduction loss for power MOSFET. making a good comparison of figure of Merit (FOM) of switching device to get better efficiency result.

    also reduce ESR of capacitor also helps the efficiency, but has to compromise the size and cost.

    BR

    Max

  • 0 in reply to Max Wang1
    Prodigy 141 points

    Hi Max,

    CMLB136T-100MS: 17.7mΩ changed to B82559A7103A020: 3.7mΩ

    I calculated the total losses within the new inductor B82559A7103A020 to be only 0.9W in absolute worse case boosting 30V to 48V at 5A using this calculator which should take into account all loss types.

    product.tdk.com/.../core_loss_simulation

    In the reference design, the worse case losses are 6.4W when delivering 140W in OTG mode and charging at 240W.

    I would imagine the original inductor CMLB136T-100MS contributes significantly more losses than 0.9W, but I would like to know from you whether this is the case, so that I can ensure my new inductor choice is justified, as the significant weight increase can be avoided if it is not needed in my weight-sensitive application.

    A breakdown of losses, or at least an estimate, out of this 6.4W in the reference design converting from 240W USB-C input to 50V battery charge would be very much appreciated just so that I can see.

    Thank you.

    Best regards,

    Rain