[SI-LIST] Re: Convergence problem after inserting detailed connectormodel
- From: Steve Corey <steven.corey@xxxxxxxxxxxxxx>
- To: si-list@xxxxxxxxxxxxx
- Date: Fri, 20 Feb 2004 16:53:25 -0800
Our passivity correction algorithms are rigorous, and we guarantee our
models to be passive. It's also worth pointing out that ensuring a
system's passivity at discrete frequency points doesn't ensure that it
will be passive when simulated in a variable time step simulator, since
the simulation algorithms have to interpolate between those discrete
frequency points and extrapolate beyond them.
To be rigorous you have to ensure that the equations underlying the
model are passive, and that the simulation algorithm will map a passive
system to a stable simulation.
-- Steve
TDA Systems MeasureXtractor is EDN Innovativion of the Year finalist!
http://www.reed-electronics.com/ednmag
-------------------------------------------
Steven D. Corey, Ph.D.
Time Domain Analysis Systems, Inc.
"The Interconnect Analysis Company."
http://www.tdasystems.com
email: steven.corey@xxxxxxxxxxxxxx
phone: (503) 246-2272
fax: (503) 246-2282
-------------------------------------------
Scott McMorrow wrote:
> Michael and Ray,
> First of all, I am not as smart as I sometimes sound. I had a good
> friend of mine, Rob Hinz, develop the code for passivity checking and
> passivity correction based upon some of my simple thoughts on the
> subject . We use what we call the "Wack a Mole" method for correction
> of non-passive s-parameters. You can imagine from that name what we
> might do.
>
> You will want to compute the minimum real eigenvalue of Identity matrix
> - Smatrix * Smatrix' on the complex data.
>
> In matlab this is done as follows
>
> mineigv = min(real(eig(eye(size(S))-S*S')))
>
>
> The matrix is passive the value is positive real
> THe matrix is conditionally passive if the value is zero
> The matrix is non-passive if the value is negative.
>
>
> The original reference is:
>
> R. A. Rohrer, "Lumped network passivity criteria," IEEE Trans. Circuit
> Theory, vol.
> CT-15, no. 1, pp24-29, March 1968
>
>
> At this time, no one has developed a mathematiclly rigorous method for
> passivity correction that I am aware of.
>
>
> regards,
>
> scott
>
>
>
>
>
>
> Mirmak, Michael wrote:
>
>
>>To echo Raymond (forgive the questions from the matrix-impaired):
>>
>>Scott,
>>
>>If I have an 8-port S-parameter measurement (S11, S28, etc.), from a
>>Touchstone or CITI file for example, would the given equation apply to
>>the complex data or just to eh magnitudes?
>>
>>More specifically, taking the equation literally, my eigenvalues are
>>calculated as:
>>
>>Identity matrix (8x8) - Smatrix * Smatrix'
>>
>>Where Smatrix is an 8x8 matrix of complex S-parameter values, Smatrix'
>>is its transpose, and the identity matrix is just that, an 8x8 matrix
>>filled with zeros on the off-diagonals and ones on the diagonals. Is it
>>still correct to say that a complex result from that calculation, if
>>having a negative real part, indicates non-passivity?
>>
>>More specifically still, I get an 8x8 matrix result of "eigenvalues"
>
>>from your equation. However, if I request eigenvalues from Matlab using
>
>>the same 8x8 Smatrix, I get a result which is 8x1 only. Why the
>>difference?
>>
>>Thanks!
>>
>>- MM
>>
>>
>>
>>-----Original Message-----
>>From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx]
>>On Behalf Of Raymond Anderson
>>Sent: Friday, February 20, 2004 12:26 PM
>>To: si-list@xxxxxxxxxxxxx
>>Subject: [SI-LIST] Re: Convergence problem after inserting detailed
>>connector model
>>
>>
>>Scott McMorrow wrote:
>>
>>
>>
>>
>>>One way to check this is to run the model in the frequency domain and=20
>>>compute the eigenvalues of the S-parameters at each frequency point=20
>>>with I - S*S'. If there are any negative eigenvalues then the matrix=20
>>>is not passive. Dr Liu has had some experience with this, since his=20
>>>software will simulate with standard SPICE models and with=20
>>>s-parameters. He has found that not only can s-parameters be=20
>>>non-passive, but also that lumped element models with coupling can be=20
>>>non-passive. The culprit is usually mutual L. A simple test for this=20
>>>in a multi-section connector model is to comment out all K elements,=20
>>>re-run the simulation and see if it is stable. If the simulation is=20
>>>stable then it is most likely non-passive. More formally it should be=20
>>>run through an eigenvalue check, which Dr Liu reports to me that he has
>>>
>>>
>>
>>
>>
>>
>>>done on a number of occasions. In that case, the model is bad and
>>>
>>>
>>
>>needs=20
>>
>>
>>
>>>to go back to the developer.
>>>=20
>>>
>>>
>>>
>>
>>Does anyone have any pointers to papers, books, or other info on how to
>>compute the eigenvalues of the s-parameters as described above. i.e.,
>>are
>>there any tools available or published matlab or whatever code available
>>to
>>automate the process?
>>
>>-Ray Anderson
>>Sun Microsystems Inc.
>>
>>------------------------------------------------------------------
>>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 technical documents are available at:
>> http://www.si-list.org
>>
>>List archives are viewable at: =20
>> //www.freelists.org/archives/si-list
>>or at our remote archives:
>> http://groups.yahoo.com/group/si-list/messages
>>Old (prior to June 6, 2001) list archives are viewable at:
>> http://www.qsl.net/wb6tpu
>>=20
>>
>>------------------------------------------------------------------
>>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 technical documents are available at:
>> http://www.si-list.org
>>
>>List archives are viewable at:
>> //www.freelists.org/archives/si-list
>>or at our remote archives:
>> http://groups.yahoo.com/group/si-list/messages
>>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 technical documents are available at:
http://www.si-list.org
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
