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 >> Requesting LO buffer phase noise simulation information
https://designers-guide.org/forum/YaBB.pl?num=1422378806

Message started by rlm584 on Jan 27^th, 2015, 9:13am

Title: Requesting LO buffer phase noise simulation information
Post by rlm584 on Jan 27^th, 2015, 9:13am

I have been using Spectre for a few years, but I'm new to Spectre RF. Please be gentle.

I am designing an LO buffer intended to drive a mixer. One of the requirements that must be checked is that the buffer does not significantly degrade defined system phase noise performance. Could someone please take a moment to explain how to do this properly? I am trying to help myself by reading the Spectre RF manual, but I am not at all confident in my use of the tool yet. The only related posting I have found on this topic is http://www.designers-guide.org/Forum/YaBB.pl?num=1386865117, where Mr. Kundert states that mixers (LO input implied) are thresholding devices and suggests an appropriate analysis type. Unfortunately, no procedural details are included. Other forum topics cover jitter for driven circuits, but not phase noise. Please provide enough detail so that someone new can follow; some of the PNOISE and Direct Plot form entries are still a bit mysterious to me.

What would change if a divider is included?

Best regards...

Title: Re: Requesting LO buffer phase noise simulation information
Post by Ken Kundert on Jan 27^th, 2015, 9:33pm

I am always skeptical of transient noise results. It is difficult to get accurate results because the noise signals are quite small and often overwhelmed by simulator errors. In addition, requesting information about low frequency noise, such as flicker noise, causes the simulations to run extremely slowly. Finally, their primary claimed benefit, that they can handle nonlinear noise effects, is almost never needed. Ironically, to see nonlinear noise effects in an oscillator, you have to simulate very close-in noise, which is extremely slow and largely impractical.

Pnoise analysis, on the other hand, separates the noise signals from the nonlinear transient behavior, making it much more accurate and robust if you do not need nonlinear noise effects, which as I already pointed out, you almost never do.

-Ken

Title: Re: Requesting LO buffer phase noise simulation information
Post by rlm584 on Jan 28^th, 2015, 2:59pm

Hi Ken,

What I am requesting is a procedure showing how to setup a PSS/PNOISE phase noise simulation for a buffer to drive a mixer LO port. You mentioned that mixers (LO input) are thresholding devices in the post I referenced. There are several descriptions of using PNOISE jitter to get jitter results for a thresholding circuit on this forum, but I can't find anything on phase noise for this type of driven circuit. Please be as detailed as possible; I'm new to Spectre RF.

Thanks.

Title: Re: Requesting LO buffer phase noise simulation information
Post by Ken Kundert on Jan 28^th, 2015, 7:39pm

Sorry. The above response was actually intended for a different thread. I accidentally posted it on this thread by mistake. Please ignore it.

If you are designing both the buffer and the mixer, I would recommend you simply combine them for simulation and treat the whole thing as a mixer. If the mixer is not available, then you should characterize the buffer in terms of its jitter behavior. After all, it is only the timing of the edges that are significant.

I have described how to extract the jitter of a driven block in my paper on PLL jitter, but that paper is kind of old now, and Cadence has simplified the procedure considerably. Just run the PNoise analysis and ask for the output jitter.

Once you know the jitter you can easily convert it to phase noise by scaling it to convert seconds to radians.

-Ken

Title: Re: Requesting LO buffer phase noise simulation information
Post by rlm584 on Feb 2^nd, 2015, 4:05pm

I didn't design the mixer, but it is available. Combining them seems to be the safest approach to this task. Thanks, Ken.