Designer’s Guide Community

aaron_do Senior Fellow Offline Posts: 1398	Noise analysis May 29^th, 2007, 1:42am Hi all, just a quick question. If i run a noise analysis, does the simlated output noise have any dependence on input frequency. i.e. i'm not concerned with input referred noise, but the input frequency is not equal to the output frequency. Also it's not a mixer so i'm not sure if i can use pnoise analysis. thanks in advance, Aaron
Back to top	there is no energy in matter other than that received from the environment - Nikola Tesla IP Logged

ACWWong

Community Fellow

Offline

Posts: 539
Oxford, UK

Re: Noise analysis
Reply #1 - May 29^th, 2007, 4:29am

noise analysis is a frequency domain simulation on a linearised (time-invariant) operating point. At any particular small signal frequency it'll compute the output noise at that frequency, with all the components equations using the same value of frequency in their equations and using the computed linear transferfunctions/gains to the the output.
pnoise is useful if you have frequency translation effects in your circuit, such as mixing or sampling etc.
try ken's books and papers on the subject...
perhaps if you elborate a bit more on what exactly you are trying to simulate ?

Back to top

IP Logged

aaron_do

Senior Fellow

Offline

Posts: 1398

Re: Noise analysis
Reply #2 - May 29^th, 2007, 5:57pm

i'm trying to simulate a peak detector. The peak detector is followed by a RC low pass filter. My concern is, suppose the bandwidth of my signal before peak detection is 1 MHz: does all the noise in the 1 MHz band get peak detected and converted to DC? Is it possible to simulate this? i.e. if the noise gets clipped it technically will result in a DC component.

thanks,
Aaron

Back to top

there is no energy in matter other than that received from the environment - Nikola Tesla

IP Logged

ACWWong

Community Fellow

Offline

Posts: 539
Oxford, UK

Re: Noise analysis
Reply #3 - May 30^th, 2007, 2:28am

ok, i think no need for pnoise or anything here.... you can use noise analysis to give you the output noise spectrum.... from this you can get an idea of the amplitude of noise at any one frequency which will be additive to your signal... i think you'll find the noise levels should not be a significant percentage of the signal amplitude you are expecting....

If noise levels are going to problematic (and you can improve your design so that noise becomes insignficant), then a simulation solution would be to simulate using transient noise analysis.

cheers
aw

Back to top

IP Logged

aaron_do Senior Fellow Offline Posts: 1398	Re: Noise analysis Reply #4 - Jun 2^nd, 2007, 5:54pm Hi, I ran the noise analysis and you are right, there wasn't much noise. I was worried that with peak detection, the noise in the signal bandwidth would be converted to a DC component...do you know why that didn't happen? thanks Aaron
Back to top	there is no energy in matter other than that received from the environment - Nikola Tesla IP Logged