This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM5156: Selectable Output Design Process

Part Number: LM5156
Other Parts Discussed in Thread: LMR36503

Tool/software:

I would like to design a boost circuit using LM5156.

VIN=10 - 42V

IOUT = 1A

VOUT = 48V, 36V, 24V

The thinking here is that I can use some FETs to modify the feedback resistors to select output voltage during runtime. Obviously if the circuit was at the highest input voltage range the lower output voltage setpoints would not be regulating and would just be in passthrough mode of the controller.

What are the implications of compensation?

What would the correct design approach be? For example, design per datasheet procedure at minimum VIN (10V) and maximum VOUT (48V), and should the other combinations just work?

I would appreciate any feedback on this topic.

Thanks!

  • Hi Evan,

    Thanks for using E2E.

    Yes, it is possible to change the output voltage during runtime by manipulating the FB voltage. This can be done by paralleling resistors to the lower FB resistor using MOSFETs, or by just manipulating the FB voltage using a microcontoller for example.

    Regarding the compensation, it has to be adjusted to ensure a stable operation over all operating points. For checking this, you can use our quickstart calculator for this device: https://www.ti.com/tool/download/SNVC224, or the power stage designer: https://www.ti.com/tool/download/POWERSTAGE-DESIGNER

    If you have further questions, please feel free to ask.

    Best regards

    Moritz

  • Moritz,

    Thanks, this is helpful. One other question. What will behavior of LM5156 converter be if VIN=42V and VOUT is set to the 24V? Will converter essentially just be in pass-thru mode (i.e. not switching)? I do see the note in datasheet section 9.3.8 about no current limit protection if VOUT <= VIN.

  • Hi Evan,

    You are right, LM5156 will not switch if Vin is higher than Vout.

    Also there will be no current limit protection in this case.

    Best regards,

    Moritz

  • Also can you please provide any comments on attached schematic snippet? Using LMR36503 to provide +10V rail for other purposes, but also using to supply BIAS pin of LM5156 for better efficiency.

    Boost inductor: SRP1265A-220M

    Boost FET: BSC146N10LS5ATMA1

    Boost diode: SR5H15_R1_00001

  • Hi Evan,

    The circuit itself seems to be ok. Only the compensation network might have to be adjusted afterwards. It seems like with the current values it is not stable for all operation areas.

    Regarding the +10V rail:

    It is possible to supply the IC with these 10 V.

    In terms of efficiency it would be better to supply the VCC pin directly with these 10V. With that the internal LDO between Bias and VCC would be deactivated.

    In this case the gate-drive voltage would be 10V but this should be no problem for your MOSFET.

    Best regards,

    Moritz

  • Moritz,

    Yes, I realized after I posted that I didn't have the correct compensation components in there. Here's updated values:

    Feeding VCC directly from +10V, that's fine - should I still connect BIAS pin to VIN or also to +10V rail ? Like such:

    Thanks!

  • Hi Evan,

    Yes, that looks better.

    You can have Bias connected to Vin in this case.

    Best regards,

    Moritz