The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Measurements >> Phase Noise and Jitter Measurements >> How to guide for noise_table usage in VCO
https://designers-guide.org/forum/YaBB.pl?num=1408572169

Message started by kg_maclean on Aug 20^th, 2014, 3:02pm

Title: How to guide for noise_table usage in VCO
Post by kg_maclean on Aug 20^th, 2014, 3:02pm

I would like to generate a noise table file for an all transistor VCO and then use that noise_table file in a closed loop PLL simulation using
veriloga models.

Is there a how-to guide or an example veriloga file showing the usage of the noise_table function and the exact required format for a
noise_table file?

Title: Re: How to guide for noise_table usage in VCO
Post by raja.cedt on Aug 21^st, 2014, 3:32am

Hi..
Why don't you save noise date into CSV file and import this table into VDC. Which works pretty fine for all PLL sims...

Title: Re: How to guide for noise_table usage in VCO
Post by kg_maclean on Aug 21^st, 2014, 11:41am

Thank you for the suggestion which helped very much.

I was not able to get spectre to accept a noise file which had been created in CSV format, but I used trial and error on the other file formats. I found that if I selected SPECTRE as the output file type in
my VCO pnoise simulation and then used the file created with the .in file name extension that spectre would run and generated a similar output ac noise spectrum to what I saw in the pnoise simulation.
The spectrum was not identical in that there is an offset of +0.535dB at all frequencies between the small signal ac noise simulation and the original pnoise simulation.

Title: Re: How to guide for noise_table usage in VCO
Post by raja.cedt on Aug 22^nd, 2014, 1:54am

Hello,
Could be some thing wrong in sim, because here ac sim is not even calculating any thing other than simply displaying .csv file(I don't know about .in file)
Check how do you plot spectrum...

Thanks,
Raj.

Title: Re: How to guide for noise_table usage in VCO
Post by kg_maclean on Aug 22^nd, 2014, 6:31am

Hi Raj

I figured out the cause of the offset. The pnoise result for the vco is displayed in dBc/Hz referred to the rms amplitude of the signal (at the vco frequency). When I save the pnoise data to a .in.spectre file the raw amplitude (in V/sqrt(Hz)) is saved, which must be referred to the actual rms amplitude to get the correct magnitude in dBc/Hz. When I did this additional step in my small signal ac simulation the magnitude was correct.

Thank you very much for your help on this.
Ken