The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Charge Pump Noise
https://designers-guide.org/forum/YaBB.pl?num=1155695214

Message started by William Ngai on Aug 15^th, 2006, 7:26pm

Title: Charge Pump Noise
Post by William Ngai on Aug 15^th, 2006, 7:26pm

Hi,

I'm using PNoise to simulate charge pump output noise current. As expected, there are 1/f noise and white noise coming out from devices so when charge pump is in its on (for very short time) and off (for most of the time) state, we're effectively sampling the output noise (neglecting off noise). So in PSS, there is "# of harmonics" and in PNoise, there is "Max sideband". If I want the simulation to be accurate, I'd need to set the # of harmonics and Max sideband to very large (since the die-off of sinc function for my short sampling pulse is really slow over freqency and up to 1GHz) and if my sampling frequency is only 1MHz that would be up to 1000 sidebands and harmonics. Am I correct on this?

Here is what I think: since white noise is flat over frequency, the overfold of such noise would be significant even from far away harmonics. Does PNoise smart enough to handle white noise and 1/f noise independently so that I don't need to simulate up to so many harmonics? Would the high harmonics/sidebands slow down PSS and PNoise calculation?

Thanks,
William

Title: Re: Charge Pump Noise
Post by Ken Kundert on Aug 16^th, 2006, 11:18pm

The accuracy improves asymptotically as you increase the number of sidebands. If you are willing to accept results within a reasonable tolerance of say 5-10%, you can often get away with a reasonable number of sidebands. Check out http://www.designers-guide.org/Analysis/sc-filters.pdf to see how the accuracy of a switched-capacitor filter noise simulation varied with the number of sidebands.

Still, for charge pumps a large number of sidebands will still be needed. I recommend artificially inflating test frequency to reduce the need for so many sidebands.

-Ken

Title: Re: Charge Pump Noise
Post by William Ngai on Aug 17^th, 2006, 10:08am

Thanks Ken, that's a really good paper on sampled noise and the using of PNoise.

William :D