|
|
Magic Mailing List |
|
From: Jeff W. Sondeen (sondeen AT rcf-fs DOT usc.edu) Date: Tue Nov 27 2001 - 11:58:01 EST
Thanks, Tim, my script to parse mosis files fails often due to layer
name changes, anyway, from their t17z-params.txt (to be in release
2001b which should be out next Monday (12/3)):
MOSIS PARAMETRIC TEST RESULTS
RUN: T17Z VENDOR: AMI
TECHNOLOGY: SCN05 FEATURE SIZE: 0.5 microns
..
PROCESS PARAMETERS N+ACTV P+ACTV POLY PLY2_HR POLY2 MTL1 MTL2 UNITS
Sheet Resistance 81.5 101.9 21.6 1120 41.0 0.09 0.09 ohms/sq
^^^^ ^^^^
so it's SCN3ME_SUBM.30.r will say:
.subckt poly2Resistor d g s b W=1 L=1 Rsquare = 41.0
R d s 'L*Rsquare/W'
Rg d g 0
Rb b 0 0
.ends
...
.subckt phrResistor d g s b W=1 L=1 Rsquare = 1120
R d s 'L*Rsquare/W'
Rg d g 0
Rb b 0 0
.ends
/jeff
R. Timothy Edwards writes:
> Dear C.Connaway,
> I think you must have the basic principle down, which is to draw the
> poly2 resistors with the "rpoly2" and "pseudo_rpoly2" layers, which
> will extract the resistor as a transistor. Run ext2spice, then convert
> all of the transistor records containing "poly2Resistor" from "M..." to
> "X..." (e.g., change from a transistor to a subcircuit). Then append
> "SCN3ME_SUBM.30.r" to the SPICE deck. This ensures a correct value
> for the resistance, and ensures that the resistor is extracted as a
> component. Otherwise, if you just use "phr", the value of resistance
> is a "lumped resistance" value, good for estimating timing in digital
> circuits, but not useful for analog simulation and LVS.
> However, as you have noted, the "Rsquare =" is missing a value in
> the "SCN3ME_SUBM.30.r". You will need to insert the Ohms/square value
> for the PHR layer here. I will CC this email to Jeff Sondeen so he
> can update the file for future releases.
>
> Regards,
> Tim
>
> P.S.---sorry I'm late responding; power was shut off in my building on
> Friday and I had to shut down my computer over the holiday.
c.connaway writes:
> Im trying to make a resistor using the AMIC5N 0.5u High Resist Layer (HR overlap Poly2/electrode), after extraction and than subsequent flattening using ext2sim, the value of the resistor is nowhere near the 1000ohms/sq it should be. While monkeying around in the SCN3ME_SUBM.30.tech27 file I noticed some code that looks like this:
>
> extract
> style AMI0.5um(amic5)from:T01X
> cscale 1
> lambda 30
> step 100
> sidehalo 8
> planeorder well 0
> planeorder implant 1
> planeorder select 2
> planeorder cap 3
> planeorder active 4
> planeorder metal1 5
> planeorder metal2 6
> planeorder metal3 7
> planeorder oxide 8
> planeorder xp 9
> planeorder comment 10
> planeorder contact 11
> planeorder via1 12
> planeorder via2 13
>
> resist (ndiff,rnd,ndc,nsd,nsc)/active 82100
> resist (pdiff,rpd,pdc,psd,psc)/active 102500
> resist (nwell)/well 827000
> resist (rnw)/active 827000
> resist (pwell)/well 1
> resist (poly,fp,rp,pc,pc,nfet,pfet,fet)/active 26000
> resist (poly2,ecap,p2c)/cap 26500
> resist (phr)/cap 26500
> resist (m1,fm1,rm1,ndc,nsc,pdc,psc,pc,p2c,m2c,m2c)/metal1 90
> resist (m2,fm2,rm2,m2c,m3c,m3c)/metal2 90
> resist (m3,fm3,rm3,m3c,pad)/metal3 50
>
> contact ndc 4 50700
> contact pdc 4 99400
> contact pc 4 17700
> contact p2c 4 17000
> contact m2c 4 710
> contact m3c 4 810
>
> Is this defining the right values for the resistance of the poly2/high resistance layer that is available for the AMIC5N 0.5u process? Also what about the SCN3ME_SUBM.30.r file, does this play any role when using ext2sim or :extract ?
> I noticed that the phrResistor doesnt have anything defined for the Rsquare= line. Should I try to edit these files? Anything will help. Thanks.
>
> C.
>
> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
> <HTML><HEAD>
> <META http-equiv=Content-Type content="text/html; charset=windows-1252">
> <META content="MSHTML 6.00.2600.0" name=GENERATOR>
> <STYLE></STYLE>
> </HEAD>
> <BODY bgColor=#ffffff>
> <DIV align=justify>Im trying to make a resistor using the AMIC5N 0.5u High
> Resist Layer (HR overlap Poly2/electrode), after extraction and than subsequent
> flattening using ext2sim, the value of the resistor is nowhere near the
> 1000ohms/sq it should be. While monkeying around in the SCN3ME_SUBM.30.tech27
> file I noticed some code that looks like this:<BR><BR>extract<BR> style
> AMI0.5um(amic5)from:T01X<BR> cscale 1<BR> lambda 30<BR> step 100<BR>
> sidehalo 8<BR> planeorder well 0<BR> planeorder implant 1<BR> planeorder select
> 2<BR> planeorder cap 3<BR> planeorder active 4<BR> planeorder metal1 5<BR>
> planeorder metal2 6<BR> planeorder metal3 7<BR> planeorder oxide 8<BR>
> planeorder xp 9<BR> planeorder comment 10<BR> planeorder contact 11<BR>
> planeorder via1 12<BR> planeorder via2 13<BR><BR> resist
> (ndiff,rnd,ndc,nsd,nsc)/active 82100<BR> resist (pdiff,rpd,pdc,psd,psc)/active
> 102500<BR> resist (nwell)/well 827000<BR> resist (rnw)/active 827000<BR> resist
> (pwell)/well 1<BR> resist (poly,fp,rp,pc,pc,nfet,pfet,fet)/active 26000<BR>
> resist (poly2,ecap,p2c)/cap 26500<BR> resist (phr)/cap 26500<BR> resist
> (m1,fm1,rm1,ndc,nsc,pdc,psc,pc,p2c,m2c,m2c)/metal1 90<BR> resist
> (m2,fm2,rm2,m2c,m3c,m3c)/metal2 90<BR> resist
> (m3,fm3,rm3,m3c,pad)/metal3 50<BR><BR> contact ndc 4 50700<BR> contact pdc
> 4 99400<BR> contact pc 4 17700<BR> contact p2c 4 17000<BR> contact m2c 4 710<BR>
> contact m3c 4 810<BR><BR>Is this defining the right values for the resistance of
> the poly2/high resistance layer that is available for the AMIC5N 0.5u process?
> Also what about the SCN3ME_SUBM.30.r file, does this play any role when using
> ext2sim or :extract ?<BR>I noticed that the phrResistor doesnt have anything
> defined for the Rsquare= line. Should I try to edit these files? Anything
> will help. Thanks.<BR><BR>C.<BR></DIV></BODY></HTML>
|
|
|
|