Perforated Tabs

Mike: Look a the Hadco DFM manual ( http://www.hadco.com/ )

Yano if you�re lookin� for uniformly smooth (not silky smooth) board edges, have you considered scoring to separate your boards? Scoring has: * Minuses (straight lines only, stress on solder joints close to score, additional operation at PCB fabrication, and edge of PCB is not smooth) * Pluses: (leaves sorta smooth surface, increase panel use, ease of separation, and better in assembly).

Look at scoring in the Hadco DFM manual also.

reply »

Hi Mike,

The Hadco design guidelines are really good - I was just there.

To expand on what Dave said, scoring is an option, but I've never had good luck with it, myself. The scores are either too shallow, and I can't break the boards easily enough, or they're too deep, and the boards fall apart in conveyors, reflow ovens, waves, etc. But you do get a nice edge, and sometimes, better utilization of the vendor panel.

On breakaway tabs, a few things to keep in mind when you are designing: - You can only break one axis at a time. So whether its edge rails, or extra material between boards, you can only break them in one direction at a time. Don't let the layout guys create breakaways that go around corners or in arcs. When looking at the layout, think about the order and direction you will be breaking in. - Proximity of components to breakaways. Minimum (absolute) is 20 mils, for thngs like discretes. Larger spacing for finer pitch. Routing should be at least 15 mils. When breaking, the board material has a tendency to flex and tear; you don't want to crack joints or expose copper. - Mask relief all around the B/A area. Don't need soldermask there;the tendncy is to relieve it around the holes like any other non-PTH. But, if you relieve the whole area, and you do have problems breaking, it's easier to see the fracture in question. - Spacing from break patterns to break pattern depends on board thickness. Thinner boards need more, thicker need less (duh). Also depends on weight of components and types of processing (power supplies going over the wave need more B/A's than pager cards in a reflow oven). At that point, I'd consult my board fab house for input.

Hope I didn't restate anything too obvious,

Chrys

reply »

Chrys & Dave:

Thanks for your replies. Scoring won't work in this case. There are some overhanging components and some 90 degree routes as well.

My experience with scoring has been similar to Chrys'. For large panels I also had some issues with planarity of the board after double pass reflow. This resulted in inconsistent conformal coating coverage in a subsequent operation.

Thanks, Mike.

reply »