The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> 'Noise Type' setting Pnoise in ADE
https://designers-guide.org/forum/YaBB.pl?num=1158924743

Message started by Paul Geraedts on Sep 22^nd, 2006, 4:32am

Title: 'Noise Type' setting Pnoise in ADE
Post by Paul Geraedts on Sep 22^nd, 2006, 4:32am

Hi all,

Seem to me unintentional: both the 'modulated' and the 'jitter' setting of 'Noise Type' within ADE, result in the same simulation command in the netlist and the same post-processing options. So it does not seem to matter which one you choose.

Both result in the two Pnoise analyses, of which the second has the negative frequency range of the first and has the noisetype=correlations added. By the way, why is this 'correlations' option removed from ADE (I belief it was PCR658211)?

Most likely, the ADE 'jitter' setting should translate to a Pnoise analysis with noisetype=pmjitter added. A bug?

Paul

Title: Re: 'Noise Type' setting Pnoise in ADE
Post by Andrew Beckett on Jan 15^th, 2007, 3:10am

Paul,

Sorry for the late reply to this.

pmjitter noisetype is for driven circuits only. For autonomous circuits, it should use the two-sweep approach (one correlated) for both jitter and modulated (you're quite right, the analysis is the same for both).

noisetype=correlations was removed from the UI because it was not really of practical use to be used directly by the user (I wasn't aware of anyone using it).

Regards,

Andrew.

Title: Re: 'Noise Type' setting Pnoise in ADE
Post by Frank Wiedmann on Jan 15^th, 2007, 6:40am

Andrew, I'm still not convinced that this is the correct way to simulate the jitter of an autonomous circuit, especially if you have additional circuits like buffers or dividers connected to the oscillator. Have you taken a look at http://www.designers-guide.org/Forum/YaBB.pl?num=1158788102;start=all ?

Title: Re: 'Noise Type' setting Pnoise in ADE
Post by Andrew Beckett on Jan 15^th, 2007, 7:47am

Hi Frank,

I'd missed that post in my catchup (there were a lot of posts for me to catch up on, since I've been rather busy the last few months). Having read through it, I tend to agree with you. I think the reason it is done this way (for autonomous jitter) comes from that fact that with tdnoise there was no way with an oscillator to reliably get the noise measurement at the right place in the cycle. With pmjitter, you can specify a threshold, and so the crossing point can be reliably defined.

So, when using the methods described in Ken's phase noise paper, tdnoise was not really a good way to do it - but spectre has moved on since then.

I'll also raise this with R&D.

Regards,

Andrew.

Title: Re: 'Noise Type' setting Pnoise in ADE
Post by Frank Wiedmann on Mar 9^th, 2010, 8:11am

According to Andrew Beckett, strobed jitter ("pmjitter") is now also available for autonomous circuits. Quoting from an email I received from him:

Quote:

This is (at last) available in 5.10.41.500.6.141 (and also 6.1.3.500.17 and 6.1.4.500.1) - i.e. the latest hotfix of all supported streams. Now when you select Noise type "jitter" with an oscillator you can choose whether it uses the "FM" approach (the default) or alternatively the "PM" approach - which you specify a threshold crossing.