TPS543C20A: TPS543C20ARVFR can not work correct

Rock Shao

Part Number: TPS543C20A

Hi team

My customer used the TPS543C20ARVFR for 12V to 5V & 3.3V, 5V & 3.3V the max output both are 25A (combine max output power 125W ), and OCP each needs 30A.

They found the following issue:

Ch3 is 5V switching
Ch4 is 3V3 switching

1. Large and small waves appearing in the no-load state

2. When large and small waves appear during loading, the PWM height will drop

to 1A

to 2A

to 3A the switch voltage is about 8V5

3. When a single group loads more than 7A, the 5V and 3V3 output are both out of control

Thanks and regards

over 1 year ago

0 Britton Jones over 1 year ago

TI__Expert 5706 points

Hello!

Please share your schematic and layout so that we can help you get to the bottom of this.

Best regards,

Britton

0 Rock Shao over 1 year ago in reply to Britton Jones

Expert 2895 points

Hi Britton,

Please check the ZIP file, thanks.

6170.layout.zip

0 Britton Jones over 1 year ago in reply to Rock Shao

TI__Expert 5706 points

Hi Rock,

Thank you for the files. We are taking a look into this, please allow a few days to get back to you

Best regards,

Britton

0 Britton Jones over 1 year ago in reply to Britton Jones

TI__Expert 5706 points

Can you please provide the component values for all inductors and capacitors? I will need this information to complete the review.

Thanks,

Britton

0 Rock Shao over 1 year ago in reply to Britton Jones

Expert 2895 points

Hi Britton,

Please refer to the BOM list, thanks.

DD bom.xlsx

L601, L603: EMPI1007B-2R2M-C7

L600, L602: SMB322521V5-350-X1

EMPI1007B-2R2M-C7-220105.pdf SMB322521V5-350-X1(V1.2)-180612.pdf

0 Britton Jones over 1 year ago in reply to Rock Shao

TI__Expert 5706 points

Thank you. I will review and get back to you shortly

Best regards,
Britton

0 Britton Jones over 1 year ago in reply to Britton Jones

TI__Expert 5706 points

Hi Rock,

The problem is likely related to the chosen L & COUT values. The inductor value, 2.2uH, is higher than the maximum allowable inductance for proper sensing. With a 25A solution, the maximum inductance is 1.22uH. The output capacitance is also too high for a stable design. In both the 5V and 3.3V cases, I recommend replacing the inductor with something closer to 0.8uH and reducing the output capacitance by removing one of the 560uF capacitors. Please take caution when selecting components for the design. For example, the original inductor that was chosen appears to be rated up to only 9.7A, when 25A is required. I have attached design spreadsheets to this message for both rails with acceptable L and COUT values.

On a separate note, we recommend placing a small high frequency decoupling capacitor (1nF-0.1uF) between VIN and GND, as close as possible to the VIN pin. This helps reduce SW node ringing.

3p3V_TPS543C20A_TPS543C20_Calculator.xlsx 5V_TPS543C20A_TPS543C20_Calculator.xlsx

Best regards,

Britton

0 Rock Shao over 1 year ago in reply to Britton Jones

Expert 2895 points

Hi Britton,

After changing the 12v input capacitor and output capacitor to Panasonic POSCAP, the GND of the two groups is wholly separated and the RSP & RSN feedback is temporarily connected to the PCB board. 5v & 3.3v can be output to 20A separately, and the switching waveform is also normal. But as long as there are the following conditions, it will not be able to boot

1. 2 GND connections
2. 2 GND connections for 12v input
3. 2 GND connections for 2 sets of outputs
4. Two GND connections at the far end

Do you have any suggestions here?

0 Britton Jones over 1 year ago in reply to Rock Shao

TI__Expert 5706 points

Did you adjust the inductor and output capacitor values mentioned above? This difficult to picture without seeing the schematic/layout. Input and output capacitors should both be connected to PGND.

0 Rock Shao over 1 year ago in reply to Britton Jones

Expert 2895 points

The customer reduce the inductor and output capacitor and the waveforms normal in full load, but now we have the GND issue, you can see the GND is in contact together in same GND pad. And I share the layout with you, thanks.

2311.layout.zip

0 Britton Jones over 1 year ago in reply to Rock Shao

TI__Expert 5706 points

Hi Rock,

Do you have vias connecting the GND of the top layer to the layer where PGND connects? It is difficult to see if any exist. I need a bit more clarification before I can help you resolve this. You are able to power up the devices normally. What is different with the layout in the case where the device does not power up?

Thanks,

Britton

Power management

Power management forum

TPS543C20A: TPS543C20ARVFR can not work correct