Stats

Locked Locked
Replies 8
Subscribers 126
Views 19414
Members are here 0

This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

kTC noise in Spectre transient noise analysis

Is it possible to use Spectre transient noise analysis to calculate kTC noise? I tried a test circuit, i.e., cmos switch clocked by 400kHz clock with 40% duty cycle. Cap is 10pF. Therefore, at Vout node, the expected kTC noise at 293K is 20.1uV. In this time domain noise analysis, stop_time=1mS, noisefmax=131.072MHz, noisefmin=1kHz v("/Vout") in time domain seems correct (I calculated 18uV rms just for these limited data from 0uS to 30uS for voltages at Vout node during CLK=0V). However, DFT output is just very confusing. Is it possible to calculate the DFT output data and come up with number close to the ideal 20.1uV? (I verified that SpectreRF can come up with 20.1uV but how about Spectre transient noise analysis?)

Tawna over 10 years ago

Hi Marklin, can you log into Cadence Online Support http://support.cadence.com ? There is a document available to Cadence customers there. File a Case with support and request it. The article number is 20026660 - an AE would need to send it to you.
Cancel
Vote Up 0 Vote Down

Cancel
marklin over 10 years ago

Thank you, Tawna. I have already filed a case to Cadence support. I hope this 20026660 is not "direct time-domain noise analysis using virtuoso spectre" which I have already had thorough reading.
Cancel
Vote Up 0 Vote Down

Cancel
Tawna over 10 years ago

There is also Article 11771942 How to reconcile the linear noise, pnoise, and transient noise results on a switched capacitor ckt? on Cadence Online Support. http://support.cadence.com/wps/mypoc/cos?uri=deeplinkmin:ViewSolution;solutionNumber=11771942
Cancel
Vote Up 0 Vote Down

Cancel
Tawna over 10 years ago

What is your Case number. I'll look it up. best regards, Tawna
Cancel
Vote Up 0 Vote Down

Cancel
Tawna over 10 years ago

Hi Mark, I found your Case. I've put the necessary information in it so you can get the article.
Cancel
Vote Up 0 Vote Down

Cancel
marklin over 10 years ago

I still don't get light-bulb moment when reading article 11771942. I believe time-domain output is pretty accurate since rms(v("/Vout")-average(v("/Vout")) came up 20.75uV, very close to ideal 20.11uV. I just don't get any intuitive info from DFT outputs.
Cancel
Vote Up 0 Vote Down

Cancel
marklin over 10 years ago

After following the instructions, yes, I was able to come up with the rms noise value from integrated PSD curve derived from transient noise analysis in Spectre exactly the same as the theoretical kTC noise, ie., 20.10uV cf. 20.11uV. Thank you. Interesting to note that "This 20.10uV is only true for the S/H portion. The continuous time portion is still 20-25% away from theoretical value." Probably it is because continuous time needs much more data points than the S/H data points.
Cancel
Vote Up 0 Vote Down

Cancel
Tawna over 9 years ago

No, it's not that appNote. There is an attachment to the Article 20026660: Switched Capacitor Simulation Guidelines.
Cancel
Vote Up 0 Vote Down

Cancel

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.