An issue with the noise summary tool of cadence

Hi,

I am designing a VCO. While debugging the phase noise issues, I am using the noise summary tool of virtuoso to find the contribution of individual transistors. There is one thing that is really confusing and I am not able to understand. My VCO has a capacitor bank, varactor, tail current source, cross coupled transistors (for negative resistance) etc. So I was initially placing everything as ideal elements and then replaced them one by one by the real elements to find out the addition of phase noise due to these elements along with their individual contributions on the noise summary window.
Case 1) with tail current source only
In this case, I had the tail current source and replaced the varactor with an analog lib cap (just for checking at a particular frequency). The cap bank was also present. I found on the noise summary window that the tail current source was contributing most of the percentage while cross-coupled pair was contributing less noise. Cap bank was also contributing some noise.
Case 2) with ideal cap bank
I used an ideal cap bank and everything else was same as before. Now when I looked up the noise summary, the cross coupled transistor contribution got increased and the tail current source contribution got reduced! I couldn't understand how this could happen.

Is "noise summary" tool reliable? I have attached schematic of the VCO (in fact it's a QVCO) that i used



Thanks! Please let me know if you have any questions.

Noise summary showing ordered noise contribution of elements, calculated on noise simulation, which is small-signal AC simulation. The results of this simulation are accurate for number of linear circuits like amplifiers, current sources or current conveyors.
In VCO, transistors (especially cross-coupled pair) works with periodic large signals, changing their operating points and thus a noise behavior. So I'm not sure about two things - if your test bench is proper for phase noise simulations, or if noise simulation is appropriate for VCO circuits.

Do You tried to calculate the noise by hand and compare both results - simulations and analysis?

iVenky said:

Is "noise summary" tool reliable?

Click to expand...

It is reliable in most cases.
Show me analysis statement.

I don't know what simulator you use.
However I recommend you to set "augmented=off", if you use Autonomous-Shooting-PSS/Pnoise in Cadence Spectre.

Try the followings, if you use Cadence Spectre.

(1) Autonomous-HB-PSS/Pnoise

(2) Autonomous-Shooting-Fullspectrum-PSS/Pnoise with augmented=on and augmented=off in Cadence APS
https://www.designers-guide.org/Forum/YaBB.pl?num=1438976388/2#2

(3) Set "noiseon_inst[...]" or "noiseoff_inst=[...]" in options statements.

Noise parameters
49 noiseon_inst=[...]
The list of instances to be considered as noisy throughout noise analysis, which include, noise, sp noise, pnoise and tran noise.

50 noiseoff_inst=[...]
The list of instances not to be considered as noisy throughout noise analyses, which include noise, sp noise, pnoise and tran noise.