TPS552882: 12V 0.5A Webbench design implemented but only outputting 0.7V

Ray

Part Number: TPS552882

Tool/software:

I have implemented the buck/boost design as defined by the webbench tool (report attached) for a supply input of 6-36V and output of 12V at a maximum of 0.5A. Powering the design with various inputs in that range with no load and 0.1A load, the output voltage is always 0.7V instead of 12V.

I don't see anything obviously wrong. Could the lower load be causing instability? And are there any suggestions on alternate part values to allow the current PCB to function?

Thank you in advance.

My schematic is:

Layout top:

Layout Bottom:

and have internal solid ground plane.

Testing the design with 17V input:

EN/UVLO is 12.2V

and probed the following signals

SW1 (yellow) and SW2 (purple):

COMP:

ILM:

TPS552882_12V_0p5A_WBDesign15.pdf

3 months ago

0 Shipeng Cheng 3 months ago

TI__Genius 17430 points

HI,

I will check your design next week. Thanks.

+1 Shipeng Cheng 3 months ago

TI__Genius 17430 points

Hi,

The schematic looks correct. But your waveform shows a very slow frequency(60-70ms), this is not device behavior. Could you check if some component is soldered incorrect?

Or could you solder another board?

Thanks.

0 Ray 3 months ago in reply to Shipeng Cheng

Prodigy 10 points

Sorry I accidently clicked "resolved my issue" Issue is not resolved.

I have two boards built up and they are both behaving similarly. There are no obvious parts solder incorrectly. Is there a specific pin set I should look at more closely?

0 Shipeng Cheng 2 months ago in reply to Ray

TI__Genius 17430 points

Hi Ray,

Give you some guidelines to check.

1. From the waveform I find the period is close to hiccup period(76ms), so please check whether the output is short to GND or resistance is low.

2. You use large R75 so current limit is low, try to change to 20k see if any improvement.

3. SLVRBH3 Calculation tool | TI.com use this tool to check your compensation value, need to consider output capacitor DC derating.

Thanks.

0 Ray 2 months ago in reply to Shipeng Cheng

Prodigy 10 points

Responses:

1. Probing between output and GND on several boards, all are >100kohm.

2. I changed the R75 to 20k, and still saw the same .072V output voltage and scopes unchanged from above.

3. I used the Webbench tool's exact values for the compensation component values. (see above attached PDF). Does the tool not use the correct formulas? Using the spreadsheet provided and a center 20V input i get the following:

What are the key values I am looking for here? I don't see anywhere in the spreadsheet associated with DC derating.

0 Shipeng Cheng 2 months ago in reply to Ray

TI__Genius 17430 points

Hi Ray,

Could you measure VCC voltage waveform? Thanks.

0 Ray 2 months ago in reply to Shipeng Cheng

Prodigy 10 points

VCC waveform at same 20V input but 0.72V output:

0 Shipeng Cheng 2 months ago in reply to Ray

TI__Genius 17430 points

Hi Ray,

Do you have TPS552882EVM? You can do A-B-A test. First, exchange the IC, to see if IC has damaged. Then, exchange each other components to find the cause. Thanks.

0 Ray 2 months ago in reply to Shipeng Cheng

Prodigy 10 points

I do not have the EVM, but I have multiple boards populated (qty 10) and all that have been tested display this same behavior. So it is highly unlikely the IC or components are damaged on all boards in the same manor.

Could the compensation circuit be the cause? Are those values not adjusted correctly in the webbench tool?

0 Shipeng Cheng 2 months ago in reply to Ray

TI__Genius 17430 points

Hi Ray,

Is it possible that the actual component on the board is different with your schematic? For example the R61 is actually 1k instead of 1Ohm.

And the compensation might be the cause. You can try change R76 to 1k or lower. Thanks.

Power management

Power management forum

TPS552882: 12V 0.5A Webbench design implemented but only outputting 0.7V