TPS92512HV: Output Capacitor and Shunt Capabilty

Or simply to say that, could I implement the TPS92515HV capability such as shunt fet dimming on TPS92512HV and omit the capacitor? Because TPS92515HV does not provide me enough current. Any answer will be appreciated.
Thanks,

Hello,

The TPS92512 requires an output capacitor for stability since it is current mode control so shunt FET dimming is not an option. The TPS92515 does not require an output capacitor and is very capable of shunt FET dimming. However, if you need more current both the LM3409 and the TPS92641 are capable of shunt FET dimming with no output capacitor. They are controllers so you can get significantly more current and still shunt FET dim very fast. Feel free to let me know if you have any questions about either.

Regards,

Clint

Thanks for your response. Honestly, I have been thinking about these two chips.

LM3409 seems provide more efficiency thus lower power dissipation, but higher W/C compares to TPS92641. Also, I realize that TPS92641 has already built in an external pwm driver so that I do not need to implement a two off timers in order to do external shunt dimming.

However, my concern is why TPS92641 runs at very low Vin - Vout conversion? In the evaluation board, it runs from 48V input voltage and 32V output voltage. Is there any specific reason to this? Because in my opinion, it will dissipate a lot of power.

I was wandering if TI also will provide the design tools for TPS92641.

Thanks

Hello,

I don't believe a tool has been developed for the TPS92641 other than webench. But it is not too difficult to design with, it's very forgiving, and it is inherently stable. I am not sure exactly why the ranges are limited to what they are on the EVM, I guess maybe it was targeted for particular applications. I was not involved in that development. But the EVM will run easily at other operating points. I have changed the UVLO resistors and run it as low as 8V input and a single LED output. So if you have a range you would like to test it at you can just changed the UVLO set point.

Regards,

Clint

Correct me if I am wrong. I believe there is no WEBENCH tool for TPS92641. And also, I think what you meant by changing the UVLO resistors, is involving LM3409. As far as I know, TPS92641, only has UDIM resistor. If you can provide me with the webench for TPS92641, I would be appreciate it.

Thanks,

Hello,

You are correct, my mistake, there isn't webench available for that device. Yes, I did mean the UDIM resistor values that provide a UVLO function as well. You can change those on the EVM and it will have a much wider range. The step by step design guide in the datasheet is pretty easy to follow, but if you need assistance on anything feel free to let me know and I will help if I can.

Thanks,

Clint

I tried to make a spreadsheet to calculate the required component. In this case, I am considering to buy the Evaluation Board for TPS92641. Since I want to test it at Vin = 60v, Vout = 42v, and Iout = 2A, I made a modification to the parts on the EVM. I would like to verify to you that my scheme will work.

Thanks,

TPS92641-EvaluationBoard.xlsxTPS92641-EvaluationBoardReplacement.xlsx

That looks like it should be fine. You may just need to worry about the power dissipation in the FETs with the increased current. You also need to consider that 2A is really pushing the EVM inductor. I would choose something with a higher saturation current and DC current rating. If you are planning on shunt FET dimming you really also want to rely on the inductor for LED current ripple and not use an output capacitor. The fast transients caused by shunt FET dimmin will cause extremely high RMS and peak currents in the output capacitor. Not only that it could cause you to exceed the SOA of the shunt FET.

Regards,

Clint