TPS552882: My battery charger based on a workbench schematic stalls after some seconds ans die with a short.

Hi Olaf,

Thank you for reaching out. Can you provide schematic so that I can check if it is well designed.

Can you help provide SW1 SW2 Vin Vout inductor current waveforms? It will be really helpful to catch the failure moment waveforms.

Also please provide inductor and output capacitor part number or datasheet.

Regards,

Mulin

TPS552882 Schematic.pdf

Hallo Mulin, thank you for your response. 

OK, sorry, I must find out, how I can support an image first. Hire it is. 

Ir is impossible, to support the SW1,SW2, because the TPS552882 does not start, or dies.

The coil is a Würth  78433391000   with 10µH 14,4A / 5,6A 33.2mR  https://www.we-online.com/components/products/datasheet/78433391000.pdf

For both Transistors I use a CSD18537NQ5AT  CSD18537NQ5A 60-V N-Channel NexFET Power MOSFETs datasheet (Rev. B) (ti.com)

For the input capacitor I use also Würth  865080743010 https://www.we-online.com/components/products/datasheet/865080743010.pdf  

parallel to    885012214005   https://www.we-online.com/components/products/datasheet/885012214005.pdf

Also the output capacitor   885012214005   parallel  to  Würth  865080443008   https://www.we-online.com/components/products/datasheet/865080443008.pdf

You can see, that the voltage divider for FB use a e-Poti, MCP4022T-503E/CHSOT-23-6_MC_MCH-M    but for all my tests, the output A and W are shorted, so that we have a divider of 82k to 10k

TPS552882 PCB.pdf Hire is a part of the two layer PCB. Don't be afraid, the GND signal is faded for a better view. 

and hire with GND: TPS552882 PCB with GND.pdf 

Please for any information you need, I hope, you can find out, what happens. 

Kind regards

Olaf

Hi Mulin,

R5 and R6 are 102k . I try to get a better image...

Hire comes a complete PDF with a much better resolution. 

Mainboard-25.pdf

Hi Olaf,

In your schematic, ISN and ISP are wrongly connected. Please refer to EVM board.

Regards,

Mulin

Thank you Mulin, but I can not see, what's really wrong. 
Now in the schematick, I have shift the ISP to an other place, but the electrical connection is still the same. 

Only the 100nF (C13) between ISP ans ISN is missing. 
Do you mean that missing capacitor? 

Hello Olaf,

The expert to this device has a holiday today. Please expect an answer tomorrow.

Best regards,
Brigitte

Hi Olaf,

Thank you for your waiting. Yes, there should be a capacitor here. Also recommend to add a 0402 0.1uF capacitor in Vout pin and put it as close as to the chip.

I think the main possible problem is the poor layout. Main power pin, SW2, Vout, GND trace are too thin. This will cause large parasitic parameter. And since there is no decoupling capacitor in Vout pin, there is risk that rings cause chip die. To verify the assumption, please provide Vin, SW1, SW2 inductor current, Vout waveforms. Even the chip will die, you can trigger Vin and see what will happen when chip is enabled. And then we will see if the ring causes damage and if  we need to change layout.

By the way, what is default setting of MCP4022?

Regards,

Mulin

Hallo Mulin,

thank you for your suggestions. I'm out of the office for the next week, and I will try it after my return. 

The default setting of the 4022 is 20k Ohm. 

I will report what happens after my return.

Kind regards

Olaf

Hi Olaf,

I will wait for your reply.

Regards,

Mulin