The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> strobing noise in QPSS
https://designers-guide.org/forum/YaBB.pl?num=1255442703

Message started by gattoblu on Oct 13^th, 2009, 7:05am

Title: strobing noise in QPSS
Post by gattoblu on Oct 13^th, 2009, 7:05am

I'm simulating a DSM (Delta-Sigma Modulator) with an ideal linearized quantizer. The quantization noise is discrete-time and perfectly represented with PNOISE simulations with the strobing option "noisetype=timedomain". in such a case the only large-signal is the sampling clock.

Now I would like to put a large-signal source at the input of the DSM and therefore to use QPSS. But the strobing option is missing. Does anybody understand why or how to circumvent the issue?

Title: Re: strobing noise in QPSS
Post by Andrew Beckett on Jan 3^rd, 2010, 3:08am

Rather late response, sorry. Been a bit tied up the last few months... (for a change!)

The primary reason is that QPSS is a "mixed frequency-time" analysis, and so is partly in the frequency domain (the "moderate" signals) and partly in the time domain (the "large" signals). Since tdnoise relies on the time domain approach, that's why it isn't available.

Conceptually I would have thought it could be technically possible to strobe the noise output during the shooting interval used in QPSS (i.e. if the clock were the large signal), but this has not been implemented. I don't know if it really is possible (Ken would have a better idea than me), but it's worth contacting Cadence directly (via customer support) and suggesting it.

As an alternative though, if you can simulate over the common frequency of your large signal input and the clock, then it's possible with PSS. It just boils down to whether your common frequency is reasonable - and by adjusting the frequencies you may be able to make sure it runs in a reasonable time with a reasonable amount of memory.

Regards,

Andrew.