• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 371 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.

TPS23753A: Circuit / Magnetics Optimization for Low Line / APD Operation

Matt Chang
Prodigy 40 points
Part Number: TPS23753A

Hello--

I'm looking at using the TPS23753A in a PoE PD that must also be able to accept external adapter power down to ~12V. The application is an IIoT product that must be able to operate on SLA battery power (12V nominal) or PoE in environmental ambient temperatures >70C, enclosure ambient >85C. Overall power requirement is >5W at ~3.3V and I'd feel much more comfortable with >8W available.

While testing out TPS23753AEVM-004 (3.3V output, spec'd to ~6W at 12V input) I noticed significant heating of the transformer- no way the core will stay within operating temperature limits at more than room temperature. It looks like the EVM transformer (Coilcraft HA3801-AL) is optimized for PoE ~48V input at 1:0.15 and 150 uH. Per the design method in SLVA305C, at low line the transformer ought to look more like 1:0.25 and 12 uH...

A few questions...

  1. Does anyone know if an off-the-shelf transformer or eval design for the TPS2375x operating ~8W at ~12V exists?
  2. If developing a custom transformer, is there any rule of thumb, documentation or guidance on appropriate core size selection?
    1. It would be great to continue to use EFD15 if overheating isn't an issue
    2. I know EFD20 will work due to testing on a different vendor's solution- but that solution is excessively expensive.

  3. If developing a custom transformer, does any documentation or guidance exist on selecting an appropriate turns ratio for a drive winding for the synchronous rectifier FET, or is this just a matter of making sure Vgs stays within appropriate limits across input range? Not having a DRV winding isn't a show-stopper as I could use an ideal diode controller.

Thanks for any input anyone has!

--Matt

  • 0
    TI__Mastermind 41770 points
    Hi Matt,

    1. TPS23753AEVM-004 would be the best solution currently in our portfolio for this operating condition. However, PMP9282 has a DCDC that meets this operating condition. You can maybe copy the same power stage (has a transformer available in digikey EP13+ core) with the TPS23753A if the thermal performance (shown in the test report) is okay for your application.
    http://www.ti.com/tool/PMP9282

    2. There are some transformer design app notes like in the link below; however, it goes into detail more on transformer design rather than choosing the right core off the shelf. Usually, for off the shelf, you can search by power and input voltage, then by size.
    www.ti.com/.../slup126.pdf

    3. This is correct, the VGS must be high enough to fully drive the FET and not exceed the abs max VGS within the input range and max expected duty cycle condition. It would be good minimize leakage inductance as well.
  • 0 in reply to Darwin Fernandez
    Prodigy 40 points
    Thanks! I'll take a look at this reference design.

    --Matt