TPS54560 power supply transfer function

steven heitbrink

Other Parts Discussed in Thread: TPS54560

I have a few of these out and they seem to be working like a champ. I'm just trying to make sense of the transfer function that was generated in web wrench. The web wrench design is given as Design : 1185265/43.

The poles and zeros are given as:

Pole1 : 1.32kHz

Complex Pole: 4.09kHz zeta=0.2393

Zero 1 : 4.09KHz

Zero 2 : 26.53kHz

The complex pole, and zero 2 can easily be explained by the RLC circuit placed on the output of the switch, but I can't figure out where pole 1 and zero 1 come from. Are they something that is internal to the tps54560 converter? Do they come from some sort of specialize analysis peculiar to switching converters?

I'm not having any trouble with these now, but I think it would be useful to understand this so I can avoid troubles in the future.

Thanks

Steve H

over 9 years ago

0 JohnTucker over 9 years ago

TI__Guru**** 191445 points

Can you post the inputs you provided to webench? I will try to reproduce it.

0 steven heitbrink over 9 years ago in reply to JohnTucker

Prodigy 230 points

I think the easiest thing would be to give you the whole design:

webench.ti.com/.../SDP.cgi

I was using a :

15uH inductor with a 40mOhm resistance
100 uF of output capacitance with 60mOhm of esr.

R comp: 30.9k
in series with :
C comp : 560pF

C comp2: 220pF

Rt=182.0 kOhm

0 JohnTucker over 9 years ago in reply to steven heitbrink

TI__Guru**** 191445 points

I'm not seeing the same results. The power stage output filter pole is at 1.32 kHz. The ESR zero is at 26.5 kHz.

0 steven heitbrink over 9 years ago in reply to JohnTucker

Prodigy 230 points

Yeah those results are something I pulled up in October. The weird thing is that the bode plots still look really similar. The gain and phase margin look about the same. Maybe someone is flip flopping on what type of analysis to do?

I guess the other question is should I be trusting web wrench for loop analysis? The results I get from web wrench give me better performance than the values calculated from the data sheet, and it is a real time saver.

0 JohnTucker over 9 years ago in reply to steven heitbrink

TI__Guru**** 191445 points

I prefer pspice modeling if it is available. Otherwise any webench designs are ok as long as they have the "Re-comp" feature enabled. The datasheet equations are kind of ball park approximations. The actual power stage transfer function gets rather complex once you add in slope compensation. I tend not to use datasheet equations except as last resort or unless they are based on the pspice model, in which case they work fine.

Power management

Power management forum

TPS54560 power supply transfer function