|
|
Magic Mailing List |
|
From: cfk (cfk AT pacbell DOT net) Date: Mon Sep 15 2003 - 21:39:02 EDT
Dear Atif:
Let me take a stab at answering and perhaps others will help correct my mistakes.
Dear all,
I'm confused with why the design rules for the SCMOS technology are so. Please answer my following questions to clear these designs.
Why the
i) poly-poly spacing is 2 lambda
The poly-poly spacing is rule 13.5, I believe. At least that one shows an edge4way test of 2. See about line 5798 in scmos.tech27 or do a search on "MOSIS rule #13.5". Thats the technical reason. The real question I think you are asking is why is it 2 and not 1. The answer to that lies in the fact that Magic is designed to use a half-lambda grid for layouts. If you are designing for .25um using say the tech27 file SCN5M_DEEP.12.tech27, then calculations are in .12um and the process is for .25um. I believe the same thing is happening in scmos.tech27 and you just need to adjust your thinking, geometrically speaking, to a half-lambda grid.
ii) 1 lambda of diffusion and metal surrounding cut
Here's a case where the minumum feature size is being used which is 1 lambda or 1um in the default scmos.
A trick that I use to measure distances is to draw a box on the screen and they type ";box" on the tkcon window. When you do that, you will see the dimensions of the box in lambda and microns. Another time I'll tell you the story of the ";", but just use it and it will become obvious after a while.
iii) 2 lambda overhang of poly at transistor gate
I believe this is rule #16.7 and is searchable in the file in the DRC section.
iv) why ndiff to pdiff spacing is so large (10 lambda)
At about line 5445 in the DRC section is the calculation:
spacing ndiff,ndc/a,pdiff,pdc,a 10 touching_Illegal "P-type diffusion must be 10 away from N-type diffusion (MOSIS rule #2.3a)"
v) why metal-metal spacing larger than poly-poly spacing
At about line 5482 you will find "Polysilicon spacing must be at least 2 (MOSIS rule #3.2a)".
At about line 5635 in the drc section you will find "First-level metal spacing must be at least 3 (MOSIS rule #7.2) and at about line 5674 you will find "Second-level metal spacing must be at least 4 (MOSIS rule #9.2a)"
vi) why metal2-metal2 spacing larger than metal1-metal1 spaning
About line 5633 is "MOSIS rule #7.1" showing First-level metal width at least 3 and at about line 5674 you will find "MOSIS rule #9.2a" showing "Second-level metal width must be at least 3 (MOSIS rule #9.1)"
These questions are very common for any student of VLSi. I've consulted so many books and web links but unable to find the answer to the above questions. Please guide me.
Regards
Atif
Research Associate
|
|
|
|