State Space Averaging vs Switch Model Averaging

[AMSA84]

Hello guys,

I am trying to design a dc-dc step-down converter. Right now I am studying the control part. But to do this, I first have to learn how to derive the transfer function of the converter.

That said, I'd like to ask first a couple of question that left me confused after reading parts of Basso's book as well as Erickson's Book.

When we are talking about the DC Steady State, it has nothing to do with large signal analysis or small-signall analysis, right?

In order to design the controller, we need to get first the transfer function of the converter. Right? (After we get the TG, we can see the behaviour of the circuit and then based on his response design the control)

Here this is referred as the small-signal analysis or large signal?

But to get the transfer function, we need a we to model the converter in order to work and manipulate the expression to plot the bode plot. Now here is one of the big questions I have:

I read, for example, in Erickson's book that we can model the converter using the State Space Averaging or the Switch Model Averaging.

Now I ask, what are the advantage and disadvantages of each one of them? Biggest differences, etc.

The idea here is to try to model the non linear switching device so that we can derive the transfer function of the converter?

Regards.

 

[mtwieg]

Neither of those methods deal with the nonlinear behavior of the system. They are both small signal AC approximations. If applied properly they should yield the same answer, it's just a matter of preference.

However before the small signal AC analysis can be done, the DC biasing (steady state) conditions must be determined. For most converters this will be a nonlinear process, but a simple one since it's just steady state.