DRC allowed this???

[larzarus]

So, a while back I got a run of boards back and they appear to have a short in them somewhere.  😞  I finally got the time to check the layout files, and I came across this.....  

This is clearly an overlap.  Two of them, in fact. I'm gonna guess that it happened when a library got updated and those pads got larger.  The question is, why did / does DRC not catch this? 

I believe the board was made in Eagle 7, but Eagle 8 doesn't see this error either. In fact, the only errors DRC is seeing is some track width and stop mask errors (which are not a problem).  Otherwise, DRC says this is fine.  This is the first time I've seen DRC miss an overlap like this.  I'd really like to know what happened, as I now don't trust the DRC as a valid pre-flight check.

[edwin.robledo]

Hi larzarus,

At the bottom of the DRC screen, you will notice the option "Select", after selecting this choice draw a selection rectangle around the affected  'Pad', please post a screen shot of the DRC results errors/warnings.

Best Regards,

Ed

[larzarus]

OK, I did what you suggested and here's the result:

just a couple width warnings about the polygons.  Nothing about overlap or clearance.

[jorge_garcia · Accepted solution]

Hi @larzarus,

I hope you're doing well. Upon further review of the file, there is actually no issue here but that's not easy to tell at first glance. I've embedded a screencast below that explains why there is no violation and how you can really tell when there is or isn't one.

Please accept as solution if my post fully resolves or you issue, or reply with additional details if the problem persists.

Let me know if there's anything else I can do for you.

[larzarus]

Wow! subtle! The restring param difference was not intentional, but it makes sense after your explanation. 

Neat trick about the pad color.  My faith in DRC is restored.  🙂

After checking against the physical boards, indeed this is not the short.