TPS40210: TPS40210 driving 1000uF output cap

Reza

Part Number: TPS40210
Other Parts Discussed in Thread: LM5155

Hi,

I did some simulations with TPS40210 in Webench (Vin=24V, Vout=80V, Cout=1000uF, Cin=150uF, Iout=1A) the results were promising, boost was stable with about PM=56' but when I updated the TPS40210 EVAL-BD with the Webench values and parts it didn't work that way, working unstable with vout=32V or so.

Typically I don't use Webench but I saw couple of videos mentioning it as very close to real world results but didn't find it like what advertised.

Any thoughts about why Webench is NOT inline with bench testing and how can I get similar results on my desk and in Webench?

Thanks,

Reza

over 6 years ago

0 Helen chen over 6 years ago

TI__Mastermind 40080 points

Hi Reza ,

Could you please attach your webench schematic and report in pdf version?

0 Reza over 6 years ago in reply to Helen chen

Intellectual 765 points

Hi Helen,

Here is the Webench SCH and design report PDF.

Thanks,

Reza

7271.00_build_this_1.pdf

0 Helen chen over 6 years ago in reply to Reza

TI__Mastermind 40080 points

Hi Reza,

Thanks for the update!

The expert of this part is on holiday. They will address your issue after the new year. Please wait in patience.

0 Reza over 6 years ago in reply to Helen chen

Intellectual 765 points

Thanks Helen!

Reza

0 Garrett Roecker over 6 years ago

TI__Mastermind 21720 points

Hi Reza,

When you are starting up does the TPS40210 enter current limit operation? This could be a possibility since the output capacitance is 1.2mF. To test this please increase CSS by a factor of 10.

Thanks,

Garrett

0 Reza over 6 years ago in reply to Garrett Roecker

Intellectual 765 points

Hi Garrett,

Yes, that was the soft start cap and by changing it to about 2uF now voltage goes up to Vo=80V.

I use a power NMOS as a switch to dump the output cap to an LED array (load) for a Very short time of about 2ms with a repetition of 10Hz, during this 2ms current draw could be as high as 6A.

This will cause the ISNS pin to detect over current condition and it will take a long time to restart and charge the output cap again. I can lower the sense resistor to get rid of over-current detection but I need to take care of SOA of the boost NMOS as well as boost inductor. Any comments or recommendations regarding the application I described?

Regards,

Reza

0 Garrett Roecker over 6 years ago in reply to Reza

TI__Mastermind 21720 points

Hi Reza,

Yes you can attempt to lower the value of the current sense resistor bu be aware the inductor will need to be rated to handle this higher current.

The other option is to switch to a device that doesn't have hiccup mode peak current limit. The LM5155 would be a good choice to replace the TPS40210.

Please let me know if you have any questions.

Thanks,

Garrett

Power management

Power management forum

TPS40210: TPS40210 driving 1000uF output cap