[SI-LIST] Re: Scale option in Hspice
- From: "Kai Keskinen" <kalevi@xxxxxxxxxx>
- To: <Tracy.Barclay@xxxxxxxxxxxx>, <a.ingraham@xxxxxxxx>
- Date: Wed, 8 Jun 2005 17:51:01 -0400
Use of the scale option really messes up simulations using device models
when they both have option scale with different values. I have received
models that do not use the option scale and have models that use it. Why
can't we educate the semi-people not to give us SI people models with scale
in it?
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx
[mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Tracy Barclay
Sent: Tuesday, June 07, 2005 9:48 PM
To: a.ingraham@xxxxxxxx
Cc: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: Scale option in Hspice
Hi Andrew,
I may regret getting involved in this discussion but, I'll attempt to
clarify the usage of the scale option.
The global scale option will scale the length, width and area of all active
(semiconductor) devices in the netlist. You need to be careful about making
sure that there is only one .option scale statement in the netlist since
HSPICE will use the last .option scale statement found will be the one that
is used. For example,
* top level netlist
.option scale=1e-6
*
M1 1 2 3 3 nch l=10 w=20
:
.include 'new_subs.ckt'
:
.end
In the file 'new_subs.ckt'
* file=new_subs.ckt'
.option scale=0.5e-6
.subckt ra1 in out
Ma1 1 in 3 3 nch l=30 w=50
:
.ends
For this case, the value for scale will be 0.5e-6. This means that the top
level transistor M1 will be
M1 1 2 3 3 nch l=5u w=10u
instead of the intended
M1 1 2 3 3 nch l=10u w=20u
Another thing to be careful of is that any included files are consistent in
how lengths and widths are specified if .option scale is used. For example,
if you specify
.option scale=1e-6
and
M1 1 2 3 3 nch l=10u w=20u
The result is a transistor with length=10p and width=20p!
The scale option does not scale passive elements or controlled sources. To
scale these elements, you need to specify scale locally to each element.
For example,
R1 1 0 1k scale=10 $ R1 value is 10k
Tracy Barclay
CAE
Synopsys
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On
Behalf Of Andrew Ingraham
Sent: Wednesday, June 01, 2005 06:51
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: Scale option in Hspice
> Td: signal delay from transmission line, in seconds per meter
> L : physical length of the transmission line, in units of meters.
> Default =1
The addition of the Length parameter (and treating Td as a delay per unit
length) is something unique to HSpice. Fortunately they did it in a way
that makes it compatible with other SPICE programs.
It is my understanding that the SCALE option is intended to be used for
scaling of semiconductor dimensions, in particular, transistor and diode
lengths and widths. That way, people could create models where the units
were in microns (rather than meters), then add a SCALE=1u and HSpice would
interpret them as microns. Things would be far better if everyone
consistently worked with the same units, and nobody did nasty things like
using meters here and microns there. It doesn't require rocket science
(sorry, NASA). And HSpice would be much better if the SCALE option never
existed!
Since SCALE affects semiconductor dimensions, then you need to ask, was this
transmission line part of the semiconductor that was scaled, or not?
I did several experiments a while back, trying to determine which things
were affected by SCALE and which were not; and the results were not pretty.
Some things were, others were not, and there didn't seem to be a consistent
rule about it. The reason I did this, was because some idiot semiconductor
manufacturer sent me HSpice models for their IC that had a SCALE factor in
them, which totally screwed up the simulations I used them in (because SCALE
affects the entire simulation, not just their portion of it!), and I needed
to find a way to FIX their models.
My advice: stay as far away from the SCALE option as you possibly can.
Avoid it like the plague. If someone gives you something with a SCALE
factor in it, give it right back to them and insist that they remove it.
> ... I ask this " scale" question
> because I find out in some cases, scale option affects Td but not
> affects L.
If the HSpice folks had documented it in their manuals, you wouldn't need to
ask this question! Hello Synopsys?
Regards,
Andy
------------------------------------------------------------------
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
------------------------------------------------------------------
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:
