[SI-LIST] Re: why do I need mixed mode S parameters?
- From: "Brad Cole" <cole@xxxxxxxx>
- To: <kenvaz@xxxxxxxxxxxxxxx>, <vince_cavanna@xxxxxxxxxxx>
- Date: Wed, 15 Sep 2004 21:13:52 -0400
Ken,
I don't agree with your assertion about mixed mode s-parameters being valid
only for lossless devices. In fact, there derivation for mixed-mode
s-parameters is completely analogous to the derivation of single-ended
s-parameters. The difference is that the voltages and currents are defined
to be differential or common instead of being referenced to ground. The
derivation is prefectly valid for devices that are lossless or lossy, active
or passive, and reciprocal or non-reciprocal. The only assumption of all
s-parameters (single-ended and mixed-mode) is that the device is linear.
Looking at the mode-conversion (Sdc and Scd) terms can tell you something
about the susceptibility or likelihood of generating EMI, respectively. In
the frequency domain you can see bands where you may have problems of this
nature.
In the time domain, you can look at these parameters to determine the
location on the device where there is an asymmetry in the structure that may
be responsible for causing the mode-conversion terms to be non-zero, and
therefore be responsible for EMI issues.
The trick, of course, is in establishing a limit. In other words, how much
larger than zero can the mode-conversion terms be before you will have an
EMI problem. The answer to that question is less straightforward and would
make a great topic for a technical paper if not a disertation. (Any takers?)
As far as discrepancies with various simulators, one thing I have noticed is
that they are often cavalier about using the proper reference impedance for
the different modes. For example, if the single-ended data is referenced to
50 ohms, be sure that the differential data has a 100 ohm reference, and the
common data has a 25 ohm reference.
On the measurement side, the mode-conversion terms are typically very small
(< -40dB in the frequency domain) and ideally zero. To calculate them from
single-ended data you end up subtracting two pairs of large numbers from
each other. This requires the measurement system to have 1) high dynamic
range, and 2) very good accuracy. Having a true 4-port VNA and full error
correction is important. My experience has told me that 1) multiple 2-port
measurments will not yield very good data, and 2) mixed-mode data from
TDR/TDT is not accurate. (Shields Up)
Brad
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx
> [mailto:si-list-bounce@xxxxxxxxxxxxx] On Behalf Of Ken Vaz
> Sent: Wednesday, September 15, 2004 7:54 PM
> To: vince_cavanna@xxxxxxxxxxx
> Cc: si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Re: why do I need mixed mode S parameters?
>
> Vince,
>
> Regarding software which simulate mixed-mode s-parameters,
> Ansoft's HFSS can do this. One thing I have noticed about
> these simulations is that the single-ended as well as the
> differential s-parameter output files I can export from HFSS
> seem to be almost exactly the same, but when I view the plots
> in HFSS, the single-ended and mixed-mode s-parameters are,
> obviously, different. But using the Bockelman conversions on
> these s-parameters, in say MATLAB, I do end up getting the
> mixed-mode s-parameters that I see in the plots in HFSS. So
> I'm not sure how HFSS computes the mixed-mode s-parameters. I
> don't see why it should, but it seems like HFSS also uses the
> Bockelman conversions from single-ended to mixed-mode. If
> anyone knows if HFSS actually computes the mixed-mode
> s-parameters, please let me know. Also, Agilent's ADS
> computes mixed-mode s-parameters, but it does it too quick.
> I'm not sure of the accuracy of these simulations since it
> most likely doesn't solve for the fields first, as HFSS does.
>
> About the VNAs, the four-port VNAs available today use the
> same architecture as two-port VNAs, with with a single
> synthesized sweeper. For true four-port analysis we would
> need a VNA with dual synthesized sweepers, especially if we
> need to synchronize the two input signals as in the case of
> differential signaling.
>
> One assumption that Bockelman has used in deriving the
> mixed-mode conversion matrix is that the transmission lines
> (/DUT) is lossless. If the transmission lines (/DUT) are
> lossy, the derivation breaks down and the single-ended power
> waves used to stimulate the network cannot be combined to
> give the differential and common-mode power waves at each
> physical port. This in turn doesnt allow the conversion of
> the single-ended s-parameters to mixed-mode s-parameters.
> This really is a theoretical limitation more than a practical
> limitation, but I thought it would be worth mentioning.
>
> Ken
>
>
> On Wed, 15 Sep 2004 vince_cavanna@xxxxxxxxxxx wrote:
>
> > I have some philosophical questions about mixed mode S
> parameters that
> > I = have been struggling to understand as I re-enter the field of
> > signal = integrity and attempt to catch-up on some of the new =
> > measurement/analysis techniques. I would appreciate any insight you
> > can = offer.
> >
> > I understand mixed mode S parameters and can compute them from
> > standard =
> > (single-ended) S parameters or from a model - or the other way
> > around. = I can appreciate their usefulness in understanding how an
> > n-port, that = may have been designed to operate mainly under
> > differential stimulus, = responds to (reflects and scatters the
> > incident power) differential and = common-mode stimulus.
> >
> > What I am trying to understand is why I would ever want to
> use mixed =
> > mode S parameters in a time-domain or frequency domain
> simulation, and
> > = how to use them. I am also interested to learn what simulators
> > support = mixed mode S parameters directly, as using them in a
> > simulator such as = Hspice seems cumbersome. My approach
> today is to
> > simply use standard S = parameters directly.
> >
> > The "why" I really don't understand at all. With regards to
> the "how",
> > I = know of one approach but it is cumbersome and does not seem
> > worthwhile. = I would be interested to know if there are circuit
> > simulators that = handle mixed mode S parameters directly but most
> > important I need to = understand why I need them.
> >
> > One way to use mixed mode S parameters, that has been suggested on
> > this = mailing list, is to use the S element in Hspice, but
> > represented with = the mixed mode S parameters instead of
> the standard
> > mode S parameters, = and recognizing that the ports are conceptual
> > (differential and common =
> > mode) as explained in [ref1]. In order to interface the
> conceptual =
> > n-port to my circuit (which expects real ports) I then have to wrap
> > the = device with a circuit that converts the actual port
> waves of my
> > circuit = into the differential and common mode waves that
> need to be
> > applied to = the conceptual n-port. This approach should work but
> > seems cumbersome = and, more important to me, I don't
> understand what
> > I gain from it.=20
> >
> > The approach I described seems like a round-about way to attempt to
> > use = the mixed mode S parameters directly when they can easily be
> > converted, = with no loss of information, into standard mode S
> > parameters and used = directly with the S element of Hspice. Even
> > better I would prefer to get = standard S parameters for my
> components
> > so I don't need to do any = conversions at all. In my simulations I
> > prefer to see the physical ports = rather than the conceptual
> > differential port and common mode port = described in
> [ref1], and so
> > the most appropriate model for me seems to = be the standard s
> > parameters. I can easily compute the various = differential
> or common
> > quantities from the circuit if that is what = interests me.
> >
> > I also don't understand why I would need mixed mode S
> parameters of a
> > = device from a vendor when I can compute them from the
> single-ended S
> > = parameters. I do understand that there may be benefit in
> mixed mode
> > S = parameters that have been extracted using a true
> mixed-mode (pure
> > mode?) = VNA, but my understanding is that most VNAs
> available today
> > actually = apply single-ended stimulus and measure the standard S
> > parameters, and = then *compute* the mixed mode S parameters. That
> > means I derive no real = benefit from the mixed mode s parameters
> > other than the convenience of = not having to do any
> computations. I
> > don't consider this benefit = significant since the
> calculations are
> > quite straightforward and do not = suffer from numerical
> instabilities.
> >
> > I may be missing some fundamental aspect about the mixed mode S =
> > parameters that would explain their popularity and if so I
> would love
> > to = understand that.=20
> >
> > Vince
> >
> > [ref1]
> > Combined Differential and Common-Mode Scattering Parameters: Theory
> > and = Simulation David Bockelman and William Eisenstadt IEEE
> > Transactions on Microwave Theory and Techniques, vol 43,
> no. 7, = july
> > 1995 =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 FAQ wiki page is located at:
> > http://si-list.org/wiki/wiki.pl?Si-List_FAQ
> >
> > 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 FAQ wiki page is located at:
> http://si-list.org/wiki/wiki.pl?Si-List_FAQ
>
> 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 FAQ wiki page is located at:
http://si-list.org/wiki/wiki.pl?Si-List_FAQ
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:
