Vuk,

Thanks for the information. I have not seen this white paper. I am curious: where did the ripple come from in a linear regulator? And why did the SpectreRF phase noise differ so much?

The salesman claimed SpectreRF "failed to average out the ripple effects". I am still not convinced SpectreRF gave incorrect results. I still wonder if the HB simulator falsely introduced the ripple. Without any lab data, it is hard to say which simulator was right. Again, I did not see the white paper but the entire example seemed contrived and the conclusions struck me as a salesman's wishful thinking. Another problem I had with the results is that it appeared that the HB analysis was performed by an HB expert while the SpectreRF version of the analysis was performed by a SpectreRF novice. The difference in levels of expertise hardly make the comparison fair.

Vuk,

I believe the saleman's party had a copy of SpectreRF. That's why I don't think the comparision was fair. They are HB experts. Based on their misunderstanding of other SpectreRF features, I can only assume they are also not experts at simulating phase noise with SpectreRF.

As for averaging the ripple, I think he was trying to say that HB assessed phase noise of the VCO over a time long enough to average over the low frequency ripple, while SpectreRF only averaged over one cycle of the VCO. Of course, I may be mis-stating his claim. He was pretty vague at this depth.