TPS54283: Designing wide input range dual buck converter (24 to 5V to 5V) using TPx8x with wide output-current-range

Julian Breiter

Part Number: TPS54283
Other Parts Discussed in Thread: TPS54286

Hello together,

I'm currently working on Dual-Buck-Converter using TPS54x8x to convert 24V input-voltage to dual 5V outputvoltages. In the datasheet of the TPS54286, on page 42 is a design example (example 2) for a similar application.

(Vin = 22 to 26V) ouput 2 --> 5V @ 2A. In this application, the specified input voltage range is 22 to 26V. Why is the input voltage range so small? What are the design hints to allow input-voltage range of about 10V to 28V (of course the Vout1 has to be smaller than Vin, in my application also 5V = Vout2)

Further, it is important to allow also a wide output-current-range of about 50mA to 1A. Is this possible? At the moment I see a huge voltage ripple at Vout. I just use MLCCs as ouput capacitors. Is it important to use an Elko (100µF) instead of MLCC?

Kind regards

over 1 year ago

0 Miranda Gu over 1 year ago

TI__Genius 15337 points

Hi Julian,

Thanks for reaching to us!

Please share your schematic and issue waveforms (Vout/SW/IL). I will review then feedback you.

0 Julian Breiter over 1 year ago in reply to Miranda Gu

Prodigy 10 points

Hi Miranda,

here is the schematic of one of the two similar output-stages. The measurements I'll submit asap.

0 Zixu Wang over 1 year ago in reply to Julian Breiter

TI__Expert 8100 points

Hi Julian,

My colleague will give you feedback next Monday!

BRs

Zixu

0 Miranda Gu over 1 year ago in reply to Zixu Wang

TI__Genius 15337 points

Hi Julian,

1. The schematic shows the Vin range is 10V~30V, but the device cannot support 30Vin, the max recommended Vin is 28Vin.

2. The FB resistance divider sets 0.8V*(27k+6.8k+390)/(390+6.8K)=3.8V, not 5V you mentioned in first post, please check.

0 Julian Breiter over 1 year ago in reply to Miranda Gu

Prodigy 10 points

Hi Miranda, yes, you are right. I forgot to tell, that I have some limiting elements on the supply, so that the maximum voltage on the regulator is below 28V.

Further, it's not clear yet, if the output voltage hast to be 5V or 3.8V, that's why both FB-resistance divider will be analysed. The other version (for 5V output) of FB-divider would be 27k and 5k14. Now it should be clear.

0 Miranda Gu over 1 year ago in reply to Julian Breiter

TI__Genius 15337 points

Hi Julian,

Understood.

I guess your question is that device works at DCM mode which caused large Vout ripple (picture in right).

Do you accept to add a dummy resistor loading? To increase the Iout and make sure device works in CCM mode which has smaller Vout ripple (picture in left).

0 Julian Breiter over 1 year ago in reply to Miranda Gu

Prodigy 10 points

Hi Miranda,

in case of an additional dummy resistor, how much minimum current would be useful? Is there another solution useful, such as higher capacitance or inductance?

0 Miranda Gu over 1 year ago in reply to Julian Breiter

TI__Genius 15337 points

Hi Julian,

It depends on your actual Iout. You can measure SW/IL waveform to check how much dummy resistance will enter CCM mode in your application.

Using larger Cout and L should decrease the Vout ripple as well. You can try.

Power management

Power management forum

TPS54283: Designing wide input range dual buck converter (24 to 5V to 5V) using TPx8x with wide output-current-range