The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> Simulation of DJ using SpectreRF
https://designers-guide.org/forum/YaBB.pl?num=1187679312

Message started by Visjnoe on Aug 20^th, 2007, 11:55pm

Title: Simulation of DJ using SpectreRF
Post by Visjnoe on Aug 20^th, 2007, 11:55pm

Dear all,

can SpectreRF be used to simulate deterministic jitter (DJ), e.g. due to power supply ripple/noise?
Will these show up as spurs in the phase noise profile of a clock signal?

The only alternative (that works) that I can come up with to simulate DJ is to run a transient simulation and to look at the period + deviations of the signal under investigation.

Has anybody used SpectreRF for simulating DJ?

Regards

Peter

Title: Re: Simulation of DJ using SpectreRF
Post by Frank Wiedmann on Aug 21^st, 2007, 2:56am

Sampled pxf analysis is the method of choice for this. You get the jitter by dividing the result by the slope (derivative) of the signal at the crossing point. I have used this method extensively in the past (using tdnoise and special noise sources at a time when sampled pxf did not yet exist) and the results have always been very close to those of transient simulations. Sampled pac analysis is still rather buggy at the moment but sampled pxf seems to work ok, probably because it is in some way part of the tdnoise analysis.

Title: Re: Simulation of DJ using SpectreRF
Post by Frank Wiedmann on Aug 23^rd, 2007, 12:05am

An additional hint: in transient analysis, the jitter is probably measured most easily by plotting the result with the eyeDiagram calculator function and then measuring the jitter with the crosshair markers.

Title: Re: Simulation of DJ using SpectreRF
Post by wwm101 on Oct 8^th, 2008, 7:25pm

Frank Wiedmann wrote on Aug 23^rd, 2007, 12:05am:

can we extract DJ and RJ separately from the total jitter?

Title: Re: Simulation of DJ using SpectreRF
Post by Frank Wiedmann on Nov 1^st, 2008, 2:30pm

The method described above simulates the deterministic jitter due to disturbances on the power supply. For the deterministic jitter due to intersymbol interference, do a transient simulation with a representative data signal at the input and plot the output signal with the eyeDiagram calculator function. For the simulation of random jitter due to device noise, see http://www.designers-guide.org/Forum/YaBB.pl?num=1224609785.

Title: Re: Simulation of DJ using SpectreRF
Post by Jacki on Sep 27^th, 2017, 9:56am

Hi Frank,

Thank you for your explanation of DJ simulation. How about an oscillator with a modified power supply, how to simulate the jitter?

Title: Re: Simulation of DJ using SpectreRF
Post by Frank Wiedmann on Sep 28^th, 2017, 12:46am

It's not quite clear for me what you mean by "modified power supply". However, I think that you should be able to use the same method for oscillators as well. Just select the Oscillator mode in the PSS setup form.

Title: Re: Simulation of DJ using SpectreRF
Post by Jacki on Sep 28^th, 2017, 5:10am

Hi Frank,

Thank you for your reply. Maybe I don't describe the problem clearly. For example, I have a ring oscillator, but the power supply is not clean (i.e. switched-capacitor DC converter or the ripple due to large current drop). I read some documents, your threads, and Ken's paper about the bang-bang noise. I think the jitter due to power supply should belong to deterministic jitter. I am not sure if we can simulate the DJ of an oscillator due to power supply ripple directly by PSS and PNoise because I am afraid PSS will not be converged since it is like a two-tone system. We can turn off the device noise and just look at the DJ due to the power supply ripple.
Regarding the PXF simulation you mentioned in this thread, do you know if there is any application notes or more detailed descriptions. It looks oscillator is an autonomous system, but when the jitter is generated by power supply, it becomes a kind of driven circuit?
Thank you.
Jacki

Title: Re: Simulation of DJ using SpectreRF
Post by Jacki on Sep 28^th, 2017, 5:39am

By the way, Frank, you talk a lot about the eye diagram plot. Would it be possible to evaluate an oscillator by eye diagram as well, just like a driven circuit?
Thank you.

Title: Re: Simulation of DJ using SpectreRF
Post by Jacki on Sep 28^th, 2017, 9:58am

Hi Frank,

I see you talk about the sampled PXF, what is the unit of the output from the sampled PXF? Does the transfer function has the output unit as sec/V? Because you say we can get the jitter by dividing the transfer function results by the slope of the output signal (I guess PSS can do it, do we need to define the time event as in PNoise simulation?). I think the slope is dv/dt, if so, the output should have the unit square(sec)/V? Please correct me, I am new to use the PXF transfer function. Thank you.

Title: Re: Simulation of DJ using SpectreRF
Post by Frank Wiedmann on Sep 29^th, 2017, 2:18am

The pxf transfer function has the unit V/V. When you divide by the signal slope (V/s), you get the transfer function from power supply ripple to output jitter in s/V. If you use the pxf analysis, you should not add an additional ripple to the power supply voltage. As a consequence, pss convergence should not be a problem because you only have a single tone (the oscillator signal).

You can of course also use the eye diagram. You will have to be careful to specify the correct time period for the eye diagram plot, however, because this may strongly influence your result.

Title: Re: Simulation of DJ using SpectreRF
Post by Jacki on Sep 29^th, 2017, 4:08am

Hello Frank,

Thank you. It seems the sampled PXF gives the sensitivity of the output jitter to the supply ripple, and it is a small-signal condition. Since the result from PXF divided by the slope dv/dt has the unit sec/V, how to calculate the jitter at the output due to the power supply ripple? Also is the PXF giving the transfer function in Frequency domain? If so, it should be a frequency dependent transfer function (at least plot from 0 Hz to f0, f0 is the beating frequency)? how to choose the value from the PXF result?
Assume we already know the sensitivity for the output jitter to supply ripple, but the ripple on the power supply is an unknown value, how to calculate the jitter? I am guessing if I need to average the ripple value, and use the averaged ripple voltage multiply the sensitivity to get the "RMS jitter".
Please give me a hint, and I hope I will not bother you on this problem again. Thank you. Have a nice weekend.

Title: Re: Simulation of DJ using SpectreRF
Post by Frank Wiedmann on Sep 29^th, 2017, 4:51pm

Take a look at http://www.designers-guide.org/Forum/YaBB.pl?num=1224609785/18#18.

Title: Re: Simulation of DJ using SpectreRF
Post by Jacki on Sep 30^th, 2017, 11:17pm

Thank you very much Frank. It seems PXF can be used to estimate a known power supply ripple with absolute frequency. I think the transfer function divided by the slope at the transition time gives the transfer function from voltage to jitter, and if we want to know the jitter, we need to know the amplitude of the ripple first. As you mentioned, for an unknown ripple, it is better to run a transient simulation. Have a good day.
Regards, Jacki

Title: Re: Simulation of DJ using SpectreRF
Post by Mos on Jan 11^th, 2020, 6:33am

Hi Frank,
I have been just going through your post on different threads, they have been really helpful.
For sampled PXF which sampling instance I should pick?
Say if I have 5 stage ring oscillator there are 5 different sampling events happening at different points. So do I just select one of those nodes when I specify my signal??

Thanks

Title: Re: Simulation of DJ using SpectreRF
Post by Frank Wiedmann on Jan 13^th, 2020, 12:48am

Select the node of which you want to analyze the performance. Usually, that's probably the output node.

Title: Re: Simulation of DJ using SpectreRF
Post by Mos on Feb 13^th, 2020, 1:59pm

Hi Frank,
I am trying to setup a ring oscillator followed by a driver for supply senstivity analysis.
Both oscillator and driver share the same supply.
If I give 10mVp(20mVpp) 5MHz ripple on the supply then in transients sim I see a variation of 150fs peak(300fspeak-peak around 370ps nominal)
I tried replicating the same with pss + pxf analysis based on your previous posts
So in pss I choose the following
Enginer: Shooting
Beat Freq : fosc(1/370p)
output harmonics(No of harmonics) : 20
Oscillator: ticked
case1:
Osc Node+ : Driver out
Osc Node- : gnd

case2:
Osc Node+ : Terminal inside ring oscillator
Osc Node- : gnd

Pxf settings
sweeptype: default
Relative harmonic: 0
Freq sweep : 10Hz - 200MHz
Sidebands: Max sidebands: 1
case1:
Pos output node : Driver out
Neg output node : gnd

case2:
Pos output node : Terminal inside ring oscillator
Neg output node : gnd

Threshold = Vdd/2

Max Samples (Have tried blank and 2000 getting the same results with both)

I am trying to interpret the results I get out of these sims. As mentioned above I have two sims with pxf
case 1: one with a output of final driver selected for pss and pxf
case 2: one with a node inside the ring oscillator selected for pss and pxf

I am measuring the voltage gain @ 5MHz offset from the oscillation frequency
I get different gains for both cases
case 1: Gain of 6.5V/V
case 2: Gain of 10V/V

Even if one of them is right with 30GV/s dv/dt I get peak dt of 3.3ps with 10mVp which is almost 20 times more than what I am measuring with transient.
Could you please help me in figuring out where I am getting this wrong.

Thanks & Regards

Title: Re: Simulation of DJ using SpectreRF
Post by Frank Wiedmann on Feb 14^th, 2020, 4:24am

The PXF frequency is the frequency of the power supply ripple, so you should look at the result at 5MHz (which isn't even inside your sweep range right now).

Title: Re: Simulation of DJ using SpectreRF
Post by Mos on Feb 14^th, 2020, 6:48am

Hi Frank,
Sorry for the confusion i am sweeping from 10Hz - 200MHz

I looked at 5MHz the gain is even higher 50
Which also does not look right.

Thanks

Title: Re: Simulation of DJ using SpectreRF
Post by Frank Wiedmann on Feb 14^th, 2020, 8:52am

I don't see any obvious problem with your setup. This has worked for me pretty well in the past, but I have not run this kind of analysis for quite some time and I also did not run it on oscillators back then. I suggest that you contact Cadence Support to let them sort this out, other people are currently also having problems with this kind of analysis (see https://community.cadence.com/cadence_technology_forums/f/custom-ic-design/43477/noise-jitter-transfer-function-along-clock-driven-inverter-chain).

Title: Re: Simulation of DJ using SpectreRF
Post by Mos on Feb 14^th, 2020, 10:23am

Thanks, Frank I will try following up with them
Also, I was reading PSRR_OSC_AN.pdf from the spectreRF Workshop folder.
There they suggest using modulated instead of sampled for power supply sensitivity analysis. Just trying to understand the difference.
Thinking about this I think any perturbation on the power supply is kind of causing phase modulation around oscillation freq.

Thanks