[SI-LIST] Re: Self and Mutual Inductance from S-Parameter
- From: "Alfred P. Neves" <al@xxxxxxxxxxxxxxxxx>
- To: ron@xxxxxxxxxxx, si-list <si-list@xxxxxxxxxxxxx>
- Date: Wed, 15 May 2013 09:04:44 -0700
Ron,
Agilent ADS is a very good tool. However...
We routinely use Simbeor to obtain stellar correlation between measurement and
simulation using causal/passive, well taken S-parameters. This includes eye
diagrams to 15G, step and TDR responses down to 10.7psec measured using Tek
TDR. We are in the process of a complete correspondence package to 32G eye
using Anritsu BERT products and our Channel Modeling Platform CMP-32.
We have numerous papers from 2009 on our website supporting this. Jim Bell
just finished an entire group of correspondence on the more pathologic networks
(stepped impedance, VIA fields, ground voids, resonators) and nailed them for
both TDR and time domain gated to 40GHz using Simbeor. The data is available
in a presentation we are presenting to SI folks interested in this topic.
Although we use Simbeor to design our platforms, our customers have had
positive correspondence with Hyperlynx, Ansoft tool correlation has worked well
for us, Matlab tools have worked well for us as well using RF toolbox, and yes
we did quite a bit of early work with Agilent ADS with our Channel Modeling
platforms and obtained good correspondence as well.
Some of the things we learned:
1. Take good S-parameters that includes 5X Nyquist and good DC point. DO NOT
correct for causality with any of the tools. They destroy the S-parameters
(from our DesignCon2013 tutorial with Anritsu). Dont fix junk measurements,
get it right with good metrology practice.
2. Our experience is the low frequency extrapolation to DC can be a problem
crux. Simbeor RCM works well and gets you to DC.
3. The stimulus for simulation needs to exactly replicate the source
measurement. This is not easy.
4. Windowing can create problems for IFFT. The mag errors hurt eye
correspondence.
5. If the launches and general signal integrity yield mediocre return loss all
bets are off in establishing good correspondence. Return loss needs to be
good for launches out to 40-50GHz at least for 28G.
6. VNA calibration validation using stepped impedance standards works very
well.
ping me, Ill send you a copy: "Methods of Improving 3D EM Model Development
and Associated Time/Frequency Domain Measurements"⨠Jim Bell and Al Neves,
Wild River Technologyâ¨Bob Buxton and Jon Martens, Anritsu Companyâ¨Josiah
Bartlett, Tektronix, DesignCon2013
Products for the Signal Integrity Practitioner
Alfred P. Neves
Chief Technologist
Office: 503-679-2429
www.wildrivertech.com
On May 15, 2013, at 8:31 AM, Ronald Miller <ron@xxxxxxxxxxx> wrote:
> Hi Vivek
> You are barking up the wrong tree.
>
> Forget lumped parameter like inductance. The basic parameter is the T-Line
> impedance and coupling approached from an RF point of view(frequency
> domain).
>
> To get back to the time domain where you are trying to work(eye pattern
> & BER) only
> a signal convolution of the S-Parameters will do the job accurately from
> 1 Ghz on up.
>
> Lastly, only Agilent ADS is designed to accomplish this and anything
> less is a waste of time
> because you will never get convergence between measurement and simulation.
>
> Ron Miller
>
>
> On 4/29/2013 11:48 PM, Vivek wrote:
>> Hi Experts,
>>
>> I have a 4 port S-Parameter of a differential trace structure, how can i
>> get the
>> self and mutual Inductance of the two nets of the diff pair using the 4-
>> port S-Parameter.
>>
>> Please guide me.
>>
>> Thanks,
>> Vivek.V
>>
>> ------------------------------------------------------------------
>> To unsubscribe from si-list:
>> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>>
>> or to administer your membership from a web page, go to:
>> //www.freelists.org/webpage/si-list
>>
>> For help:
>> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>>
>>
>> List forum is accessible at:
>> http://tech.groups.yahoo.com/group/si-list
>>
>> List archives are viewable at:
>> //www.freelists.org/archives/si-list
>>
>> Old (prior to June 6, 2001) list archives are viewable at:
>> http://www.qsl.net/wb6tpu
>>
>>
>>
>>
>
>
> --
> Ronald B. Miller, Microwave/SI Engineer
> Gigahertz Data Cell 510-377-9380
> 7721 Sunset Avenue www.ghzdata.com
> Newark, CA 94560
>
>
>
>
>
>
>
>
>
>
>
>
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
> or to administer your membership from a web page, go to:
> //www.freelists.org/webpage/si-list
>
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>
> List forum is accessible at:
> http://tech.groups.yahoo.com/group/si-list
>
> List archives are viewable at:
> //www.freelists.org/archives/si-list
>
> Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
>
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
