Folded Polygons

IanD

I'm running into an issue when subtracting one layer from another, the result ends up with a polygon that represents two separate areas connected by their corners. I need to grab the smaller areas, (which are artifacts to eliminate), but the two areas are treated as one. I tried using '.raw' in the subtraction, but it still comes out as a single polygon. Any suggestions to break apart the polygons at their touching corners, or how to stop the subtraction from merging at touching corners?

tomas2004

Hello IanD,

I did a quick test in the DRC environment, and indeed I get the same result, both when marked as clean and raw... (bug???)

1) when doing layer operations via the menu Edit > Layer > Boolean Operations make sure the "Minimum Coherence" box is checked and it should work

2) if you would like to stick to a DRC script, you could do an additional undersize/oversize, for instance with -1dbu/+1dbu as a workaround, but this might lead to sizing side effects

Cheers,

Tomas

IanD

Thanks Tomas,
Not sure what you mean by an additional undersize/oversize, but I'm going to try 'min_coherence = true' in my DRC script

tomas2004

Hello,

You first size by -1 dbu which will separate the "connected" polygons, then size by +1 dbu to go back to original dimensions, now they should be still separated. be cautious for sizing effects though.

Cheers,

Tomas

Matthias

I tried the min_coherence = true approach. It works, but it is slightly inconvenient.

First, it's not a DRC feature, but available as a somewhat cryptic option inside the Region class. The latter object can be accessed as the "data" member of the DRC layer.

Second, it is short-lived, so you have to set it explicitly before the NOT operations.

Here is an example with a two-pass NOT:

l1 = input(1, 0)
l2 = input(2, 0)
l3 = input(3, 0)

# plain
(l1 - l2 - l3).output(100, 0)

# with min_coherence = true

# using Region#min_coherence = true
l1.data.min_coherence = true
l1.data -= l2.data

# note that after the -=, min_coherence is reset, so 
# we need to set it again
l1.data.min_coherence = true
l1.data -= l3.data

l1.output(101, 0)

Here is my test case:

and while the layer 100 result is merged, the layer 101 result is not:

Matthias

IanD

Thanks, yeah, end up using size -1,+1. Did the trick.