Re: [IBIS-Users] Overshoot and Undershoot issue

From: Andrew Ingraham <a.ingraham_at_.....>
Date: Wed Aug 17 2005 - 08:55:09 PDT
> If let say, now no more in input pins but output / I/O pins, will the
> overshoot shows in the model data?

Overshoot is something that happens when certain kinds of signals are driven
toward a device.  It doesn't matter whether the pin is an input-only pin or
an I/O pin; and the model data can never show you the overshoot.  That would
be like a textbook telling you the hair color of the person reading it!

The details in the model can affect how much overshoot there is as a result
of one signal being driven toward the device, but the model itself doesn't
cause the overshoot, the external signal does.

Even if you are talking about an I/O pin that is driving, and causes
overshoot elsewhere, or even at its own pin when the reflection comes back
to it, the model can never tell you how much the overshoot will be.
Overshoot is the result of interactions between the signal source, all other
devices connected to it, and the interconnections (wires, traces)
themselves.  You can never know the overshoot without knowing all these
things and how they interact.

Now, it is possible for a model (or a spec sheet or data sheet) to include
information about how much overshoot can be tolerated by a device.  But
saying how much overshoot can be tolerated, doesn't say how much overshoot
actually happens.

When you asked your original question (your group found overshoot that is
not shown in the IBIS model data), I wasn't sure what you were asking.  Were
you saying that your group measured overshoot (with an oscilloscope) that
wasn't predicted by their simulations that used IBIS models?  If that is
what you said, there are a few possible reasons.

One, make sure to use very good lab practices and short scope probe leads.
It is easy for scope probes to cause ringing and overshoot to be displayed
on the scope, that aren't really there when the probe is disconnected.

Were the simulations and the measurements observed at the same exact points
in the circuit?  Often there can be bad overshoot at an IC pin, that isn't
there at the same IC's die pad.

If you did everything right, then it is possible that the IBIS model is
wrong, or not as good as it should be.

There can be some discrepancies due to the IC package model.  At least one
major simulator does not include the package that is in the IBIS model at
all; you need to explicitly add it yourself with a little subcircuit.  Also,
depending on the complexity of the package model, the simulator might use
something simpler.  (In other words, the IBIS model may be correct but the
simulator took some "short-cuts.")  IC package effects can have a big effect
on the overshoot that is seen at the external IC pin.

Regards,
Andy Ingraham


|------------------------------------------------------------------
|For help or to subscribe/unsubscribe, email majordomo@eda.org
|with just the appropriate command message(s) in the body:
|
|  help
|  subscribe   ibis       <optional e-mail address, if different>
|  subscribe   ibis-users <optional e-mail address, if different>
|  unsubscribe ibis       <optional e-mail address, if different>
|  unsubscribe ibis-users <optional e-mail address, if different>
|
|or email a written request to ibis-request@eda.org.
|
|IBIS reflector archives exist under:
|
|  http://www.eda.org/pub/ibis/email_archive/  Recent
|  http://www.eda.org/pub/ibis/users_archive/  Recent
|  http://www.eda.org/pub/ibis/email/          E-mail since 1993
Received on Wed Aug 17 08:56:14 2005

This archive was generated by hypermail 2.1.8 : Wed Aug 17 2005 - 08:57:44 PDT