RE: [SI-LIST] : eye diagram in HSpice

About this list Date view Thread view Subject view Author view

From: Todd Westerhoff ([email protected])
Date: Fri Dec 22 2000 - 12:01:42 PST


Hi all,

We were talking about this thread over lunch today (we've had our fun with
Awaves eye diagrams as well), and realized that there is another, easier
option for a number of us: just load the waveform data into Cadence's
SigWave display and switch to eye diagram mode.

If you've used print statements in your Hspice deck to list nodal voltages,
then you can select File->Import->HSPICE in SigWave to read the data from
the .lis file. SigWave allows you to set the eye diagram period and
starting offset on the fly, and changes the display accordingly.

SigWave is included as standard part of SPECCTRAQuest. Starting SigWave
standalone doesn't chew up a SPECCTRAQuest license, so you can still use
SPECCTRAQuest for other activities while displaying HSpice results.

Todd.

-----Original Message-----
From: [email protected]
[mailto:[email protected]]On Behalf Of Greim, Michael
Sent: Wednesday, December 20, 2000 11:33 AM
To: 'Ricchiuti Vittorio'; [email protected]
Subject: RE: [SI-LIST] : eye diagram in HSpice

Hi Vittorio,

This has been discussed many times before on this
list, and yet it still remains a confusing issue.
I will try to keep what should be a simple topic
simple. First cut and past the following into your
favorite spice deck desiring an eye display

**************** cut here **********************

*****************************************************************
* Setting up the eye diagram display
* To display the eye pattern for the simulation, click on t
* under voltages in awaves and assign it to the x axis.
* Note that display period is twice the period of the freq
* specified below.

.param eyefreq=625e6 per='1/eyefreq' $ max toggle lvds rate
.param Tshift=1.55n
eT T 0 VOL='((TIME-Tshift)*(eyefreq/2))-int((TIME-Tshift)*(eyefreq/2))'
rT T 0 1k
.print tran v(t)

*****************************************************************

***************** cut here too *******************

What the above does is creates a variable that is used to
replace the x axis of your display. This variable essentially
goes from 0 to a time equalling your period and then resets to
zero. This allows the display to be re-written over and over
creating the eye. The eyefreq variable is your data rate or
twice your toggle rate. The above illustrates an eye for a
625 MBS signal, or 312.5 Mhz signal. The tshift variable allows
the eye to be centered in your display. A non-centered eye can
offer some pretty interesting displays.

If you are in avanwaves click on t or v(t) in the curves window
Now go down to the current x-axis window and click apply. You
have now replaced linear time on the x axis with your new variable.

double click on a signal that you wish to see and voila,
an eye pattern. If you are looking at a clock, this will
be pretty uninteresting. However, if you are looking at
the output of a psuedo random pattern driven network you
should have something that looks like what you see in all
the high speed articles.

I hope that this allows you to get up and running. In the
beginning I always knew the block of spice to add to my deck,
just not what to do with it after the simulation was complete.

best regards and happy holidays,

Michael Greim

'Making the world safe for digital signals anywhere'

-----Original Message-----
From: Ricchiuti Vittorio [mailto:[email protected]]
Sent: Wednesday, December 20, 2000 10:45 AM
To: [email protected]
Subject: [SI-LIST] : eye diagram in HSpice

Hi all.
I have some problems to display the eye diagram of a pseudorandom pattern in
HSpice simulations. The technique that Avant! appliction note suggests is
not so good in my opinion.
Can someone give me informations about this topic and a HSpice deck with
statements to display eye diagram ?

Regards
Vittorio

ing. Vittorio Ricchiuti
CAD support and SI engineer
Siemens ICN
Loc. Boschetto
67100-L'Aquila
ITALY
Tel: +39/0862/335880
Fax: +39/0862/335623
e-mail: [email protected]

**** To unsubscribe from si-list or si-list-digest: send e-mail to
[email protected]. 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
[email protected]. 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
[email protected]. 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 : Tue May 08 2001 - 14:30:30 PDT