[SI-LIST] Re: SI Simulation of GHz signals
- From: Julian Ferry <julian.ferry@xxxxxxxxxx>
- To: "'clifford@xxxxxxxxxxxxx'" <clifford@xxxxxxxxxxxxx>, si-list@xxxxxxxxxxxxx
- Date: Mon, 26 Jul 2004 16:16:12 -0400
Hello Clifford
You have touched on some topics that are near and dear to me (and a few
others, I am sure). I think I can safely say that there are some simulation
methods appropriate for these speeds that are both robust and reliable. But
they haven't quite been reduced to the realm of "simple" yet.
Spice in its various flavors is highly accurate, assuming the models are
accurate. But as you have observed, run times can be so slow as to be
impractical. Simulators based on behavioral type models can provide much
more reasonable run times, and if care is taken in developing and
implementing the models, they can provide excellent accuracy as well.
IBIS is an approach into which we here at Samtec are putting some major
effort. But the official interconnect standard is so new that very few (if
any?) simulation tools will actually accept such models for input. Of
course, it is on the road map for implementation by most of the major
players, and we are committed to providing models that meet this standard.
We will be rolling them out starting in the next few months. In the mean
time, we provide what we call "IBIS like" models that have been optimized
for use in several simulators. You can find those at our website
www.samtec.com.
S-parameters work well for certain applications, but there are a few
downsides (like model flexibility) associated with their use as well. You
can expect to see more work from us in that area soon too.
Currently we have worked with some SI system consultants (most notably,
Teraspeed Consulting Group)to develop some combined simulation approaches.
In a nutshell, we use spice models where we can, and use a full wave 3d
field solver to generate s-parameters for structures that don't lend
themselves well to spice models.
We had a joint presentation with Teraspeed at DesignConn that described this
process. You can download a copy here:
http://www.samtec.com/reference/articles/pdfs/advances_in_modeling_interconn
ects_DesignCon_East_0404.pdf
We have also taken a slightly different approach to this problem. Since such
simulations aren't yet a simple cut and paste operation, we take this task
on ourselves for some general cases. Part of this process leads to what we
call "Final Inch". In the Final Inch, we design a PCB layout for a certain
connector under various signal/ground conditions. We develop models of the
connector, pcb break out region, and traces. You can license this design
from us at no charge, and we provide all the associated models (and test
data that validates them). You can cut and paste this design into your
system design, and cut and paste the models into your system simulation.
We have recently taken this process one step further and begun to provide
app notes for specific applications. Fortunately, one of the first
applications we chose was a XUAI application based on the Xilinx Rocket I/O.
You can find more information on that here:
http://www.samtec.com/reference/articles/pdfs/QXE-DP_FI_XAUI_5mm_appl_notes.
pdf
A list of other applications currently available can be found here:
http://www.samtec.com/reference/articles/articles.asp
Basically, we are trying to do as much of this simulation work as possible
on the front end. It just doesn't make sense for engineers the world over
to have to reinvent the wheel with some of this stuff. You might be able to
use our designs and models as is, or you might need to use bits and pieces,
with a few tweaks required for your particular application. But this should
definitely assist you in getting started.
If you have any questions, feel free to drop a line to me, or to
SIG@xxxxxxxxxxx We have a group of engineers who's job is to help you
choose an appropriate high speed interconnect, and to help you interpret our
data and implement our models.
Julian Ferry
High Speed Engineering Manager
Samtec, Inc
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On
Behalf Of Clifford van Dyk
Sent: Monday, July 26, 2004 1:32 PM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] SI Simulation of GHz signals
Hello
I would like to perform reliable simulation of GHz signals (up to
3.125GHz), specifically Xilinx RocketIO. I would like to include in the
simulation the effects of the following:
Driver->PCB trace (incl. vias)->connector->cable->connector->PCB
trace(incl. vias)->Receiver
The models that I have obtained from the various vendors are HSPICE
models. I have evaluated two of the most recommended S.I. toolchains:
Mentor G. HyperLynx and Cadence Spectraquest. Both tools use HSPICE as
the simulation engine, and essentially act as a front-end gui to HSPICE,
as well as extracting the PCB trace/via models. My experiences thusfar
with both tools have not been good. Anything but the most simplistic of
traces causes the tools to either crash or take rediculous time to
process (of the order of hours for even a simple net). A further issue
is that both of these tool vendors claim that the HSPICE simulator is
not necessary, and that the simulation can be performed without it, but
practically this is not the case, due to a lack of availability of
reliable models in anything other than HSPICE format. The conversion
from HSPICE to any of the custom modelling types is also, in my opinion,
non-trivial and potentially an extremely tedious manual process.
I believe that S-Parameter based simulation provides much faster
simulation, but again there is a lack of availability of S-Parameter models.
Can anyone recommend a method for simulating the above signals that is
simple, robust and reliable, or is the simulation of such signals still
premature? Coming from a HW design background, I am fairly new to S.I.,
but it seems surprising that there is no industry-standard modelling
type (equivalent of IBIS) that cable/connector vendors will provide, but
maybe this level of simulation is in its infancy, and S-Parameter models
will emerge as the standard?
Is the simulation of such signals entirely necessary? I am dubious about
the reliability of the results of such simulations, and I am wondering
whether it is not more practical to just take all the precautions
possible and hope for the best!
Please let me know if you have any advise, or a good solution to my dilemma!
Kind regards,
Clifford
------------------------------------------------------------------
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:
