A small batch of PCBs

September 16, 2012 12:30:23 PM CDT

What with other things going on, it's been a while since I've done any etching. I'm converting a bunch of my designs for SMT utility circuits over to a format that's suitable for publishing though, and one part of that process is actually building the things so I'm sure they work.

I use my own variant of the toner transfer process for etching, and while I do plan to write about that someday, it won't be today. The process is kind of involved, and there's a difference between "doing something" and "making of photographic record of something while you do it."

For now, here's a small gang of boards I made for an upcoming SMT tool.

I do several small boards at once for a couple of reasons.. first, larger boards are easier to hold, move around, and fit through the machines (I use a laminator to adhere the toner to the copper). Second, the process time for a small board is about the same as the time for a large board, so I end up getting more finished product in less time.

This is what a board looks like just after etching and washing:

A gang of boards after etching

They aren't perfect. there are a couple of thin traces and some spots around the edge where I didn't get all the secondary resist cleaned off (that's one of my variations on the usual toner transfer process), but they'll all work.

Here they are after the toner has been stripped off:

A gang of boards after cleaning

There's nothing like a fresh, clean copper surface and a bright light to make the errors leap out at you. Aside from the residual gunk along the edges and a few pads with rough edges, the damage isn't too bad though.

Counting from the left, there's a blip in board 2 that didn't get etched away, a pinhole in the lower left pad of board 5, and a thin trace at the bottom of board 7. The thin trace is the only one that might cause electrical problems, and a continuity test showed that there's still enough copper to do what needs done.

If I'd been more concerned about etch quality, I would have used a wider margin from the edge of the patterns to the edge of the copper-clad blank.

Here are the boards after getting a layer of corrosion protection:

A gang of boards after being 

The secret ingredient here is acrylic floor polish (Pledge, which uses the formula from what used to be Future). It's cheap, easy to get, and is specifically designed to form a smooth surface. I lay on a coating, let the excess drip off, then bake it dry on a hot plate. Takes about two minutes.

The coating isn't a true solder mask because the acrylic melts well below soldering temperatures. It does discourage solder from flowing all over the place once the copper gets hot, but it's mostly there to prevent oxidation.

Finally, here are the boards after separation, trimming, and cleaning:

A gang of boards after 

With any luck, I'll get the components on a couple of them tonight, and will be able to publish the circuit in the next day or two.

Random brain cookies:

I'd love to go out with you, but I'm attending the opening of my garage door.