The Mysterious Case of the Missing Bowties Part II

By I. M. Baffled


After receiving our boards and having them pass inspection I breathed a sigh of relief as this project was finally on the home stretch. Procurement sent the bare boards and the design data to our primary contract manufacturer (CM). This CM also has DFM software. For them, the obvious focus is assembly related DFM issues.

By now I’m sure you have figured out where this is going. Yep. After having our design for a week, the CM found their own list of DFM issues that they wanted us to address before resuming the tooling of our job for assembly.

This time, there were about forty items being flagged.

I sheepishly shuffled into my boss’ office once again, bearing further bad news. “The project is on hold again,” I said – Might as well just get to the point.

“Which one?” he asked. “The new one?”

“No,” I replied. “The old one.”

“THE OLD ONE?!?!? What the hell! I thought you had taken care of all those issues and put them to bed!”

“I did, kinda. I fixed all the fabrication issues but now our CM has a whole new list of their own assembly DFM issues.”

“You gotta be kidding me!”

“How many issues are there? Are they legitimate? How critical are they?”

Oh great. Now the grilling is coming three questions at a time.

All I could say was, “I’ll get back to you once I investigate.”

I pull back up the design, use my second screen to display the DFM report from the CM and step through the issues one by one.

The first issue the CM is complaining about is the footprint for some of the 0201s isn’t adequate to comply with IPC Class 2 standards. How was I supposed to know? My EDA library doesn’t have the pin contact area defined.

Next on the list was an issue having to do with the height of a connector. Apparently, as the board would go through the AOI machine on the SMT line, the height of the connector would shadow the solder joints on an adjacent component, resulting in an un-inspectable part.

And then the CM DFM report said one of our alternate parts for a 3-pin SOT 23 didn’t fit the pad properly. Really? I wonder why our component engineer didn’t catch that?

Once again, my fire-breathing dragon boss stomped over to my desk and lurched over my shoulder. As I continued looking at the issues, he interrogated me about each one. “Where are the darn bowties?!?!? I don’t see any bowties on any of these issues. Why didn’t our DRC or DFM software catch these?”

“I don’t know. They’re missing.”

“I can see they’re missing! Why are they missing?!?!?”

He wasn’t the least bit humored when I sarcastically replied under my breath, “It’s a mystery.”

I spent the better part of three days revising the design to address the issues raised by the CM. Unfortunately, many of the changes required had to do with footprints and placements, so there was a fair amount of re-routing needed. And of course, that meant a new spin going back to the bareboard fabricator on our dime.

The boards were finally built without issue and we shipped the product – six weeks late.

to be continued…

Leave a Reply

This article first appeared on the Siemens Digital Industries Software blog at