• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 2079 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TPS5420 board layout question

John M
Expert 2135 points
Other Parts Discussed in Thread: TPS5420, TPS2541

I noticed that the GND pin for the TPS5420 EVM board is not connected directly to the C4+C1 input cap ground plane but first connects to the catch-diode (D1) plane and then out to the main C4+C1 plane.

However, the board layout recommendations in the TPS5420's datasheet (Figure 9, Design Layout) show the GND pin being connected directly to the C4+C1 plane and the D1 plane.

Which method would you recommend for a new board layout? I'm guessing that the EVM's method is preferred as the EVM is shipping now and the datasheet is from 2007. But, it also seems that not also connecting GND directly out to C4+C1 keeps D1's noisy, high current path localized and away from the rest of the circuit. I wanted to be sure though.

Thanks!

  • 0
    TI__Guru**** 191445 points

    That family of parts doe not connect well in actual pcb layouts.  It is pin for pin compatible with certain competitor parts and that makes it not so easy to route.  I recommend that you use the layout of the board layout recommendations.  The EVM was supposed to be connected to the input capacitors at the GND pin, but that fill was pinched off by the VIN trace.  We did not catch that until the pcbs were already made.  Since the performance did not suffer we went ahead with them.  The difference in ground routing distance between connecting from the external side to connecting from under the device is minimal, so either way will work.

  • 0 in reply to JohnTucker
    Expert 2135 points

    Thank you for the layout tip!

    Might there be another SWIFT module you'd recommend as a better choice for 24V-to-5V step-down at 1.5A-1.8A continuous? It's for a Dedicated Charging Port for USB devices using the TPS2541. We selected the TPS5420 as a quick, inexpensive, way to get the prototype boards up and running but if the part isn't optimized as much as it could be (because of the requirement to match comp. pinouts), perhaps there's another chip you'd recommend?

  • 0 in reply to John M
    TI__Guru**** 191445 points

    There are certainly alternatives.  Can you let me know a little about your care abouts?  Then perhaps I can make a recommendation,

     

     

  • 0 in reply to JohnTucker
    Expert 2135 points

    I'm looking for a low-cost solution I can get up and going quickly to power a TPC2541 acting as a controller for a general-purpose 5.0V USB dedicated charging port. Here are some of the features/specs:

    - 100mA-1800mA loads consisting of anything from small LED accessory lamps to smartphones being charged.

    - The input is regulated 23.5V to 24.5V from a AC-DC power supply that's also providing power for other circuits.

    - The output must be limited to 4.75V to 5.25 volt for all loads and load changes over a -20C to 60C temperature range. We can sacrifice the low end, i.e., go to 0C to 60C, if necessary to keep the costs lower.

    - BOM cost of about $5-$6 (1000 pcs.) is good.

    - Safety is critical so short-circuit, over-voltage, over-current protection is a must (the TPS2541 provides current-limiting so we're covered a bit there). Low EMI emission is good too.

    - After the initial power-up of any loads, I expect the current draw to be fairly steady (or slowly lowering if acting as a charger). Fast load-change transient response is not critical.

    - Connect and disconnect of the load can be expected to occur at any time though.

    - I have about 2" sq. (1300 mm-sq.) available for this circuit on the PCB (4-layer, 2oz. copper), out of a much larger board. Fan cooling is available.

    Thanks!