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.

TPS543820: TPS543320, TPS543320EVM

Part Number: TPS543820
Other Parts Discussed in Thread: TPS543320

Hello colleagues,

My customer is evaluating TPS543820/TPS543320 as AOZ device replacement:

I started to work on the EVM and switched ON/OFF a 180uF Polymer ELKO. So far, I used the PCB “as is” with 3.3V default setting. Result is that I need to slow down to a rise time of about 200us, which means ~3A current to not trigger the short circuit detection. I also tried to add a 180uF bulk cap to the EVM PCB, which did not really improve things (I guess due to the reduced control BW by that).


I need to go on working a bit more detailed on the situation, but I think maybe I can share the updated power rails that I want to replace: “Alter Wandler”=Current Solution / “Neuer…”=Redesign

Cout: Total value all over the PCB, in brackets it is the capacitance right at the DCDC outputs in the AOZ designs. The caps right at the DCDC is what I can optimize.

Eingang=Input voltage is 12V on all DCDCs.

For the #2/+5V rail: Current is 3...4A, but to re-use parts, I’ll use TPS543820. There is a load switch from +5VALW to +5V. I will try to change that and use the enable. On this rail, there are load switches on our self-build board and to the USB ports. So, I would see some amount of bulk cap added to the DCDC converter output. But I need to figure out, how to do that best. On the EVM there is C27 introducing zero at ~250kHz and pole at >1MHz, maybe shift that to a lower frequency and change MODE pin?

Any comments or recommendations are welcome. I need to make that robust regarding load switching in the rails 5V/3.3V.

Thank you,


  • Hi Daria,

    Regarding a Soft start of 200us, the device has only 4 selectable SS option, the closest been 0.5ms which isn't on any of the MODE pinstrap selections on the EVM, unfortunately this device can't go any lower as SS settings are internally selected. To use 0.5ms you would need to swap Rmode for one of the selections on DS table 7-4. Also keep in mind that you have to recalculate the Cramp(through Rmode) when changing Cout per figure 8-3 as using a MODE resistor with incorrect Cramp may result in instability of the device and irregular switching and Vout. 

    You can refer to our TPS543x20 design calculator for COUT, CFF(C27) zero and MODE resistor calculations for each individual rail:

    You can lower C27 frequency by increasing the capacitance, use the calculator to get exact desired value or use our SIMPLIS model for this device to look at bode (