We've designed +15VDC and -15VDC power supplies using the TPS54060 device.
Now, we'd like to make the designs programmable to be able to adjust the outputs from 7VDC to 15VDC by just varying one resistor in the sense circuit.
However, I'm concerned that the loop filter won't be sufficient as a "one size fits all" and we need to know how to interpret the gain and phase results tohopefully reach a compromise response for various output voltages.
Can someone explain what all of the plots represent and if some parameters are more critical, etc.?
I've been out of the analog design area for quite a few years and I need a quick refresher one these plots.
Thanks,
Greg Lese
Hi Greg,
I have a bit of experience with SwitcherPro and what you are describing here can be done in the software with a little work.
However, this particular device doesn't really support the change one resistor approach as many of the other parts do. The reason is because in most designs the top divider resistor is "fixed" to a value like 10K. This resistor is part of the compensation circuit and will have an impact on crossover and loop response. By fixing that top resistor to some value it is possible to change the design output voltage just by varying the bottom diver resistor. This device is not supported that way because the apps team has chosen to have the bottom resistor fixed to a value and then the top resistor is varied to set the output voltage. This will change the loop response a bit.
What I'd suggest is If you design the supply first for 15V then select and lock the Inductor, output cap and compensation components then you can use the "Change inputs" button to set the output voltage to 7V.
when you do that the locked components will stay fixed and only the R1/R2 divider will change. this works with no issues. I tried it with a 40 - 60V input , 15V & 7V output at 300mA. My reistor divider changed from 10K 178K over 10K to 78.7K over 10K. In both case I checked the loop response to verify that I had adequate phase margin. Both designs had 61 degrees of phase margin and a crossove frequency of around12kHz. This is plenty stable. Then main things you want to look for are at least 30 degrees of phase margin at the crossover frequency of the "Total Gain" So for this case you can really just change the R2. Again it will change the loop response but it doens't look unstable to me, it just changes the crossover frequency a little bit.
I hope this helps and let me know if you need more help on the gain and phase plots.
Thanks
heath