RE: [SI-LIST] : IBIS or SPICE examples

About this list Date view Thread view Subject view Author view

From: Muranyi, Arpad (arpad.muranyi@intel.com)
Date: Thu May 10 2001 - 11:50:15 PDT


An ideal current source has an IV curve which is a horizontal line
on a current vs. voltage plot. In my experiments I proved it to
myself that the end of a transmission line sees the slope (derivative)
of the IV curve as the termination impedance. The impedance of
an ideal current source is infinite, in other words it looks
like an open to the transmission line.
 
My question is, what makes some people feel that it is a good thing
to have current source like drivers for high speed signaling?
Wouldn't it be better to have matched impedance terminations at
***both*** ends of the T-line which implies linear IV curves
such as a resistor's IV curve? Am I missing something?
Remember, the driver is also a terminator on the T-line while
it is driving...
 
Arpad Muranyi
Intel Corporation
=====================================================================
 
-----Original Message-----
From: C. Kumar [mailto:kumarchi@yahoo.com]
Sent: Friday, May 04, 2001 6:12 AM
To: Willis, Ken; 'Mark Alexander'
Cc: si-list@silab.eng.sun.com
Subject: RE: [SI-LIST] : IBIS or SPICE examples

I agree with Ken's excellent response. In fact most high speed drivers
should behave close to ideal current sources. Otherwise the device will be
in deep trouble.

However advanced modeling will require "structural" circuits. Specctraquest
has this ability with a comprehesive suite of controlled sources which we
use extensively.

  "Willis, Ken" <Ken.Willis@sycamorenet.com> wrote:

Hi Mark,

I have run into a couple of chip suppliers that indicate things the
other way around, which I found interesting. In this particular
case, I had put in a request for a model, and eventually got to
the model developer. His opinion was that for the high speed
(2.5 GHz) data rates we were talking about, the driver HAD to be very
linear and symmetric by definition. His take on it was that when
looking at these kinds of speeds, the driver model actually became
simpler and the interconnect model becomes much more detailed and complex.
He felt the result would be "interconnect-dominant" rather than
"device-dominant". He gave me the IBIS data I asked for and some
details on the package. I was able to model all this up, run a quick
pseudo-random data pattern down there, and get some eye patterns that
were very close to what we had in the lab. We wer! ! ! ! e able to track
down
some stubbing problems inside the receiving package that were closing
the eyes down a bit as well. So it seems in certain applications you may
be OK using IBIS style models for some pretty high speed applications.

A couple of important caveats (of course). Power and ground problems did
not seem to be an issue in this application. And the loss tangent values
are very important to get proper results. Also it is critical to model
vias in detail, including the stubbing that has been discussed at length
here. I used the spice subcircuit capability of SPECCTRAQuest
quite a bit for modeling complex passive stuff like packages, vias, etc.
Your results will be very dependent on how well you can model these things.
So it wasn't push button "out-of-the-box" tool usage, there was some
stuff to do under the hood, but it worked pretty well. At any rate, there
is another data point for you on behavioral vs. structural device mod! ! ! !
eling.
I'm sure there are cases in which structural modeling is mandatory, but this
didn't seem to be one of them. As always it will be dependent on your
specific
application, and your mileage will vary.

Ken

-----Original Message-----
From: Mark Alexander [mailto:Mark.Alexander@xilinx.com]
Sent: Thursday, May 03, 2001 11:48 AM
To: Todd Westerhoff
Cc: si-list@silab.eng.sun.com
Subject: [SI-LIST] : IBIS or SPICE examples

Todd et al,

Many of us are familiar with the fundamental differences between SPICE and
IBIS
-- one is difficult and highly accurate, the other is simple and moderately
accurate. What might be more useful in this discussion is information from
personal design experience about when you would use one versus the other.

For example, up until recently my department used IBIS exclusively for our
board
design and simulation. These boards involved single-ended signals running
from0 to 200 Mhz and differential signals up to 840 Mhz. However as we move
into
the world of high-speed serial channels at 3, 5 and 10 Ghz, the ice on the
IBIS
pond gets a bit thin.

Right now we're running both SPICE and IBIS simulations of our systems in
order
to make sure we're seeing everything we need to. Our concerns stem both
from
the general aspect of IBIS's approximate nature, as well as the specific
aspect
of package modeling. Detailed package modeling within the confines of the
IBIS
spec is difficult. We see differences in our simulation results (between
SPICE
and IBIS), though we're just beginning to draw conclusions as to what's
causing
them and how we can improve the IBIS model.

Others have commented that in order to model systems with a non-ideal power
plane, you have to use SPICE because no one yet makes an IBIS simulator that
can
handle non-ideal power planes. I'm not very deep in this realm! ! ! ! ...
can
anyone
comment?

-mark

Todd Westerhoff wrote:

> The answer to your question is: it depends.
>
> IBIS is actually a language for characterizing I/O behavior that has a
> defined structure and syntax. That's the long way of saying it models a
> defined subset of all possible I/O behavior. SPICE, on the other hand,
can
> model anything. Want to model viscous fluid flow through a pipe? Sure -
> come up with the correct electrical equivalent model, and SPICE can handle
> it for you.
>
> So, "it depends", means that you have to understand the interconnect and
I/O
> you want to analyze, and whether or not behavioral analog simulation will
> model enough of the effects properly to give you a reasonable answer. If
> IBIS provides an adequate answer, then by all means, use it. And, if IBIS
> won't give you the detailed answer you want, you still may be better! ! !
! off
> running IBIS up-front to get you in the ballpark, before you run a SPICE
> analysis. This is especially true if you're running pre-route analysis
and
> looking at a number of different scenarios. We regularly analyze hundreds
> and thousands of "variations" at a time using IBIS models, receiving
answers
> within minutes. You can imagine what the turnaround time would be if we
> were doing things differently.
>
> But for I/O structures that IBIS cannot address, SPICE may be your only
> choice. And for critical applications, it's almost always worth the
effort
> to look at structures with both forms of simulation, to make sure the
> answers correlate.
>
> I view simulators as tools. They help you get a job done, but you have to
> understand what their limitations are, how to use them, and to which
> problems they are best applied. Please don't expect a simulator or SI
too! ! ! ! l
> to give you an answer - because it won't. Its purpose is to give you data
-
> on which you can base design decisions.
>
> Todd.

**** To unsubscribe from si-list or si-list-digest: send e-mail to
majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****

**** To unsubscribe from si-list or si-list-digest: send e-mail to
majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****

  _____

Do You Yahoo!?
Yahoo! <http://auctions.yahoo.com/> Auctions - buy the things you want at
great prices

**** To unsubscribe from si-list or si-list-digest: send e-mail to
majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Jun 21 2001 - 10:11:54 PDT