RE: [SI-LIST] : SDRAM termination

About this list Date view Thread view Subject view Author view

From: Tom Dagostino (tom_dagostino@mentorg.com)
Date: Fri Nov 19 1999 - 09:29:15 PST


Andrew

Your advice is a little to general to be useful for this application. I
agree you need to use decent probes but just saying "passive reisitive type"
opens the door to most scope probes. It also limits what may work in this
case. What is needed is any probe type with enough bandwidth and low Cin.
Either passive or active probes will work. I would recommend over 4 GHz
system bandwidth (scope and probe) and less than 1 pF of input C. Good Zo
probes meet this requirement. Good ones will have less than 1 pF of Cin and
> 5 GHz bandwidth. Good active probes can also solve this measurement
situation. Passive Zo probes with 500 Ohms input R may place too much DC
load on some circuits and thus the higher R in of active probes may be
required.

Very importantly, you need to be sure the probing technique used matches the
signals of interest. Long ground leads (or, yes I have seen this) no ground
lead will cause poor bandwidth even with the proper probe/scope combination.

BTW, if you are using a Tek scope with 500 MHz bandwidth and getting 1.5 ns
risetimes you are not using the correct probe or have some other problem.
You should see about 700 psec risetime with a 500 MHz scope. The Tek scopes
I worked with had their bandwidth specified at the probe tip. You may want
to look at your probe compensation or the amount of ground lead length.

Tom Dagostino

-----Original Message-----
From: owner-si-list@silab.eng.sun.com
[mailto:owner-si-list@silab.eng.sun.com]On Behalf Of Andrew Phillips
Sent: Friday, November 19, 1999 5:12 AM
To: si-list@silab.eng.sun.com
Subject: Re: [SI-LIST] : SDRAM termination

"Greim, Michael" wrote:
>
  Have you checked the interface lines with
> a scope yet? What do they look like? Your description is at
> to high a level to solve right now.
>

Before you rush off and do this don't forget to make sure you are using
decent oscilloscope probes (e.g. passive resistive type) in a proper
manner - otherwise you'll probably load down the signal too much and
what you see is not what is really there - the circuit may even suddenly
work!

Also make sure your oscilloscope has a decent bandwidth otherwise your
signal risetime that you see on the screen will be baloney. I tried to
use a crappy 500MHz Tektronix scope recently to look at some signals
with 200psec risetimes - forget it! The scope itself had an effective
risetime of about 1.5ns.

BTW - a question to others since I've wandered into this area - since
signal risetimes are forever shortening how are we ever going to be able
to measure them properly without having to pay out grillions for the
latest high-bandwidth scope? Are there better options?

Check out one of Howard Johnson's columns for good info on scope probes:

http://www.sigcon.com/articles/straight/probes.htm

Andrew Phillips
Supercomputing Systems AG
Zurich, Switzerland

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

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


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Tue Feb 29 2000 - 11:39:01 PST