The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> High-Power Design >> SpectreRF?
https://designers-guide.org/forum/YaBB.pl?num=1174094487

Message started by Ken Kundert on Mar 16^th, 2007, 6:21pm

Title: SpectreRF?
Post by Ken Kundert on Mar 16^th, 2007, 6:21pm

Anybody using SpectreRF for simulating power electronics? If so, what do you use it for?

I can imagine using the PStb (Periodic-Stability) analysis for exploring stability of the feedback system. I can also imaging using the envelope analysis for transient simulation. Is anybody using it for that? How well does it work? Is anybody using it in another way?

Thanks,
-Ken

Title: Re: SpectreRF?
Post by Matthieu on Apr 2^nd, 2007, 7:40am

Hi all,

I use SpectreRF for simulating a buck DCDC converter.

However this is one of the first time I use spectreRF so I am not sure I do it in the right way. I use it to determine the open loop response of the DCDC converter:

1-> Determine DC feedback level ensuring the right good output level
2-> run PSS simulation
3-> Determine PAC response in openloop.

This method gives quite good results when we compare transient response with phase margin computed previously. Problem is that PSS simulation is quite long as we need to wait for the DCDC output level to stabilize.

I have read one of your mail which tells that opening the loop is not a good thing, however I haven't found any equivalent for continous time stab analysis (IC5141 USR3) in discrete time analysis. Is it possible to calculate phase margin in close loop for a discrete time system with SpectreRF?

Regards
Matthieu

Title: Re: SpectreRF?
Post by Ken Kundert on Apr 2^nd, 2007, 8:38am

The latest version of SpectreRF provides pstb analysis, which sounds like what you would need.

-Ken

Title: Re: SpectreRF?
Post by Scottie on Jan 3^rd, 2008, 7:34pm

I have tried PSS + PAC to acquire the bode plot of only the power stage of a buck converter (e.g. dVout/dD where Vout is output of that buck converter and D is the duty cycle of a fixed frequency signal driving the PMOS of that buck converter).

So, the simulated result should follow the well-known and -accepted result of voltage-mode transfer function of a power stage of buck converter, such as with a complex pole located at the resonance frequency of LC of that buck converter. However, the simulated bode plot shows the peak off from the theoretical location.

Anyone has faced similar problem as I did?

Thanks
Scottie