Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

DRC Errors after No Geometry Violations

Status
Not open for further replies.

Mahmoud_Dagher

Junior Member level 2
Joined
May 12, 2018
Messages
20
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
168
I had no geometry violations in my SoC Encounter implementation, but when going to Calibre I have many DRC Errors. how to fix this ? I have to fix all these errors manually!?
 

It would be good to perhaps point out which errors you see. Are they related to the actual standard cell or the routing?

Typically, Encounter might not have full support for the following type of issues:

1) Wide metal and long metal strips running next to each other
2) too close to wide metal
3) DFM errors in the standard cells and vias

1 and 2 you have to fix manually or tell encounter to not route wires there. Add a routing blockage or something in that metal layer.
3 you should probably live with as many standard cells violate the DFM rules.

But again, would make sense to see the errors.
 

I had no geometry violations in my SoC Encounter implementation, but when going to Calibre I have many DRC Errors. how to fix this ? I have to fix all these errors manually!?

encounter is not a sign off DRC tool. it will generate false positives and false negatives.
 

It would be good to perhaps point out which errors you see. Are they related to the actual standard cell or the routing?

Typically, Encounter might not have full support for the following type of issues:

1) Wide metal and long metal strips running next to each other
2) too close to wide metal
3) DFM errors in the standard cells and vias

1 and 2 you have to fix manually or tell encounter to not route wires there. Add a routing blockage or something in that metal layer.
3 you should probably live with as many standard cells violate the DFM rules.

But again, would make sense to see the errors.


Thanks jjx, the errors are the same but repeated a lot,

1-Min width of metal =
2-Min spacing of metal=
3-Metal must extend past via by 0.035 on two opposite sides.
4-Min spacing of via=
5-Min width of via=
6-Via must be inside metal

What do you think, I didn't expect to have all these kind of DRC errors with No Geometry violations in Encounter!
 

Those all look like "geometry" violations to me. Maybe
your rules decks are not really the same. Or maybe a
checkbox in Encounter setup omits ignores key checks
by default.

You should be able to pick one error and compare its
real layout details with the blessed LGRP. Then you
would know which tool is giving you the false (good
or bad) reports.
 

Those all look like "geometry" violations to me. Maybe
your rules decks are not really the same. Or maybe a
checkbox in Encounter setup omits ignores key checks
by default.

You should be able to pick one error and compare its
real layout details with the blessed LGRP. Then you
would know which tool is giving you the false (good
or bad) reports.

Could you clairfy to me please what is the meaning of (compare its real layout details with the "blessed LGPR") and how to do it?
 

LGRP = layout groundrule package. You should have a
controlled ("blessed") document that says in plain text
(with pics) what each rule is.

So pick one flagged error and measure. Does, or does it
not, agree with the document? With the error report?

Secondary files such as your local, or foundry "golden"
DRC, should match the document. There has to be a
discrepancy somewhere between the four points
(document, layout, Encounter, foundry deck) and it's
on you to figure out which, and push back or fix layout
appropriately.

I guess there's another possibility here, that the
layout synthesis rules are the problem (and the Encounter
deck might match, but both mismatch the foundry checks).

At some point you have to take off your necktie and
crawl in the mud. The time is now.
 

LGRP = layout groundrule package. You should have a
controlled ("blessed") document that says in plain text
(with pics) what each rule is.

So pick one flagged error and measure. Does, or does it
not, agree with the document? With the error report?

Secondary files such as your local, or foundry "golden"
DRC, should match the document. There has to be a
discrepancy somewhere between the four points
(document, layout, Encounter, foundry deck) and it's
on you to figure out which, and push back or fix layout
appropriately.

I guess there's another possibility here, that the
layout synthesis rules are the problem (and the Encounter
deck might match, but both mismatch the foundry checks).

At some point you have to take off your necktie and
crawl in the mud. The time is now.

When I'm running Calibre DRC on one of the library cells, it also have DRC errors! How come? like the library calibre rules don't match with its own cells!?
 

When I'm running Calibre DRC on one of the library cells, it also have DRC errors! How come? like the library calibre rules don't match with its own cells!?

It is very rare that a standalone standard cell is DRC clean. Cells are meant to be used in an environment, i.e., surrounded by other cells, rails, taps, etc.
 

Library cells can often be put out with DRC errors in the
cell alone, which are eliminated when the cell is butted
to a neighbor or "endcap".
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top