The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> Maximizing PSS simulation accuracy in presence ...
https://designers-guide.org/forum/YaBB.pl?num=1170976620

Message started by schehrazi on Feb 8^th, 2007, 3:17pm

Title: Maximizing PSS simulation accuracy in presence ...
Post by schehrazi on Feb 8^th, 2007, 3:17pm

How can we maximize the accuracy of a PSS simulation when signals with vastly different amplitudes are present? For example, think of a very linear circuit for which you want to simulate an intermodulation tone. Now if the difference between the main tones and the intermodulatin tone at the output is larger than 100dB will we have a simulation accuracy issue?

Title: Re: Maximizing PSS simulation accuracy in presence
Post by ACWWong on Feb 9^th, 2007, 2:47am

Using pss option "high order" should drop numerical noise floor enough to do your 100dBc measurement.
It invokes Multi-Inverval-Chebyshev analysis after pss has nominally converged, to improve noise floor. According spectre -h pss it perports to improve it to noise floor to -150 dB using "conservative", but i was told before by a Cadence guy that 200dBc distortion should be achieveable... unfortunately my circuits are never that linear so i haven;t needed to test it ;)

Title: Re: Maximizing PSS simulation accuracy in presence
Post by Ken Kundert on Feb 9^th, 2007, 3:28pm

Personally, I'd try the harmonic balance option.

-Ken

Title: Re: Maximizing PSS simulation accuracy in presence
Post by schehrazi on Mar 13^th, 2007, 10:29am

In fact, I did try flexible balance for measuring IIP2 of a current commutating active mixer. The resulting graph does not make sense! I have attached the graphs I got. The upper left corner graph is using shooting method (which is good) and the rest are harmonic balance but with different number of harmonics.
For some other type of mixer, both methods fail and IM1/IM2 vs input power graph looks like harmonic balance ones in the attached figure.

I have had this problem with IIP2 simulation for a long time and nobody has been able to help. Shooting method works sometimes, I have not been able to make flexible balance work. I can't figure out what the trick is since everything makes sense to me the way I run the simulation. Interestingly, with the same simulation, IM3 vs input is linear and looks very good!

I would be grateful if you can advise.

Title: Re: Maximizing PSS simulation accuracy in presence
Post by Ken Kundert on Mar 13^th, 2007, 3:01pm

Generally when you get this behavior of it is because of one of two things:
1. The Fourier analysis is not very accurate and has a noise floor that is corrupting the results.
The strategy in this case is to try to make the Fourier results more accurate by tightening tolerances, shrinking the timestep, etc. This is not needed with harmonic balance. There should be no noise floor with harmonic balance. This should only be an issue with normal transient analysis or shooting methods.
2. The circuit is discontinuous.
The expectation that doubling the input power should result in a factor of four increase in the level of the second-order distortion product is based on the assumption that the second-order distortion product is caused by smooth second-order nonlinearity. If instead the distortion is cause by a discontinuity, its level could be constant with input power level or it could go up slower than expected.
A discontinuity presumably would come from the models being used. I've seen this more with MOS passive mixers because the FETs are biased in an unusual way that exposes some problems with the model. I would not expect to see this with a traditional active mixer.

Either way, you might want to work with Cadence to resolve this issue.

-Ken

Title: Re: Maximizing PSS simulation accuracy in presence
Post by schehrazi on Mar 15^th, 2007, 2:15pm

Thank you I will contact Cadence. However, this is another observation of mine. Imagine the same current commutating active mixer with three large signal tones, LO=1GHz, RF1=1.1GHz and RF2=1.16GHz. After downconversion, I expect to see IF1=100MHz, IF2=160MHz, IM2=60MHz, IM3_1=40MHz and IM3_2=220MHz and probably a lot of spurious tones which are very small compared to these main tones (except maybe for IM2). This is exactly what shooting method gives (Right hand side of the firgure below).

When I change the PSS method to flexible balance, I get some resutls which don't make sense. I see a lot of spurious tones at integer multiples of the beat frequency (20MHz) which are relatively large. The next largest tone after the fundamentals is only 32dB lower than the IF1 and it is at 80MHz! (Left hand side of the firgure below). Besides that, magnitude of the desired downconverted tones and the upper IM3 tone match in both shooting and flexible balance methods however, that of the lower IM3 tone and IM2 tones don't match!

I should say that I am using 200 harmonics in both simulations.

What is wrong here? I expected to get similar results with both methods.

Thank you in advance

Title: Re: Maximizing PSS simulation accuracy in presence
Post by byang on Mar 15^th, 2007, 3:15pm

Hi, Schehrazi,

Just curious: did you set highorder=yes in the shooting method?

Thanks,

byang

Title: Re: Maximizing PSS simulation accuracy in presence
Post by schehrazi on Mar 15^th, 2007, 3:24pm

No, I did not select highorder in shooting method. The results look to be good and valid and the noise floor is low enough even without highorder option.

Title: Re: Maximizing PSS simulation accuracy in presence
Post by byang on Mar 15^th, 2007, 9:07pm

If tightening tolerance or setting highorder=yes all lead to the same shooting result, that would give you more confidence that the shooting result is correct.

What could have caused the problem in the flexible balance result? One possibility is that your time-domain waveform has sharp transitions. Harmonic Balance could have accuracy problem with sharp transitions.

byang

Title: Re: Maximizing PSS simulation accuracy in presence
Post by schehrazi on Mar 15^th, 2007, 9:40pm

My stimuli are sinusoid.

Title: Re: Maximizing PSS simulation accuracy in presence
Post by Andrew Beckett on Mar 16^th, 2007, 4:15am

Can you try using QPSS with flexible balance? Really if you have a multitone circuit and are using HB, you should use QPSS and specify the harmonics you need for each tone, rather than lumping them all together and doing a PSS. That makes sense with shooting, but less so with HB.

Andrew.

Title: Re: Maximizing PSS simulation accuracy in presence
Post by schehrazi on Mar 16^th, 2007, 11:02am

Results of shooting and flexible balance with QPSS match well.

Does HB simulation have a fundamental limit when there are multiple inputs? or the accuracy can be improved some way?

Title: Re: Maximizing PSS simulation accuracy in presence
Post by schehrazi on Mar 22^nd, 2007, 1:46pm

I contacted Cadence. Apparently this is a bug in my release of Cadence (sub-version 6.1.0.270
) which is fixed in the latest ISR of MMSIM 6.1 release. I am going to install the latest ISR now !

Title: Re: Maximizing PSS simulation accuracy in presence
Post by Bill Toole on Mar 28^th, 2007, 5:48am

schehrazi wrote on Mar 16^th, 2007, 11:02am:

Results of shooting and flexible balance with QPSS match well.

Does HB simulation have a fundamental limit when there are multiple inputs? or the accuracy can be improved some way?

I don't believe Cadence HB has a limit to the number of input but the more inputs you have, the more memory you will require to handle all the mixing products etc. If you have a large circuit you could quickly run out of memory with a large number of input tones. I generally have not had to use more than three inputs

The accuracy of the HB simulation depends on a number of factors
1. The order of harmonics of your input tones - the more the better but the slower the sim
- if you have squarish looking signal (say on an LO port of a mixer) you will need more harmonics to get good accuarcy
2. The order of the intermod products - again the more the better but the slower the sim
- cadence default is to calculate all intermods but you can select diamond rule and limit the number of intermods
3. Oversampling ratio can affect accuracy some as well - usually an OSR of 2 or 4 will suffice but will slow down the sim compared to 1

Unfortunately there is no golden rule about what order to use for your harmonics and intermod. It highly depends on your circuit, signals, and which tone you are interested in. (ie if you are only interested in the fundamental output you probably don't need that much accuracy but if you are interested in the intermods (IP2or3) then you need higher accuarcy.)

Title: Re: Maximizing PSS simulation accuracy in presence
Post by schehrazi on Mar 28^th, 2007, 11:49am

Just to give you an update on this, the problem exists with the newest ISR, in fact this is what I got from Cadence today!

"
I used the previous ISR of MMSIM 6.1 release (i.e., sub-version
6.1.1.332) to get the correct
results.

You are right in that the issue occurs in the latest ISR of MMSIM 6.1
(sub-version 6.1.1.340).
"

This should be some old wrong code which was fixed in some ISR but was reused by mistake in the latest one!

Title: Re: Maximizing PSS simulation accuracy in presence
Post by schehrazi on Mar 30^th, 2007, 10:21am

Another update for those interested in this thread:

I talked to somebody from Cadence a few minutes ago, this issue is not related to different versions of Cadence. It was his mistake last time, he read the result of the shooting method simulation instead of harmonic balance in his simulation. That's why he thought it was working with the previous ISR.

The key to make this work is to use a HUGE number of harmonics. I thought 200 was already a lot but he suggested to increase it to 1024 and now, the results well match! I am amazed why I should use that huge number of harmonics though!

Title: Re: Maximizing PSS simulation accuracy in presence
Post by Ken Kundert on Mar 30^th, 2007, 1:39pm

This is a risk with all harmonic balance simulators. In many circumstances they require a large number of harmonics and to date none of them automatically set the number of harmonics to control error.

Shame on them. Spice has had automatic error control for 35 years, and there is no reason why it cannot be provided with harmonic balance. Personally, as a user, I think you should be outraged and let Cadence know that you are disappointed in them.

-Ken

Title: Re: Maximizing PSS simulation accuracy in presence
Post by BERBEREPHONE on Apr 5^th, 2007, 2:18pm

Hi ,

There is someone who knows a way to measure IIP3 and compression point 1db under ELDO for an MIXER in a AC analysis or SST analysis??

Thanks a lot,

ecrivez moi les commandes a utiliser pour calculer 1db compression point

ibrdmmache