I am using LM5116 for a 90V~100VDC in, 36V~80VDC out with 8 Ohm resistance load (but the load will be behind a PWM driver with 100Hz ~ 100KHz switch frequency) solution, and I am facing the following issues:
1. The SPICE model downloaded through Webench being used in Altium mixed simulation is behaving quite differently than in TINA-TI (RT/SYNC pin level is ~200mV instead of 1V).
2. In Webench, after I change MOSFETs of the design by duplicating 1 -> 3, the heat simulation etc are all disabled. Also it doesn't allow me to generate a solution with more than 3A under a 80V output, or any solution using LM5116 with 36V output.
3. In TINA-TI with the model schematic downloaded from Webench, I'm unable to get a converging result with the parameters suggested by LM5116's quick start Excel sheet. Note that the quick start Excel sheet is dated back to 2008 and actually the starting parameter suggested by the quick start Excel sheet is much different in optimization angle with Webench. Webench and quick start Excel seem to provide very different starting parameters even under same clock freq.
Actually I ready to set different clock frequency for different output voltages or loads, but it's difficult for me to tune the RAMP signal along with the output. Would you please suggest from TI's perspective whether it is feasible to achieve this design target with LM5116, or is it that a better solution is in place?
Thx you and best regards,
kpk