Page 1 of 1

PCB etching

Posted: Mon Feb 16, 2004 12:00 am
by Kotlowy
what is your favourite/preferred method?

Posted: Mon Feb 16, 2004 12:04 am
by leeeeee
Get someone else to make it. 8^)=

Lee.

Re: PCB etching

Posted: Mon Feb 16, 2004 2:05 am
by 8BIT
Kotlowy wrote:
what is your favourite/preferred method?
ExpressPCB.com offers free schematic and layout software and has reasonable prices for double-sided plated through hole boards.

Daryl

Posted: Mon Feb 16, 2004 2:47 am
by Kotlowy
yes yes yes, i know there are services out there that will do it for you, but I'm interested in DIY.

Posted: Mon Feb 16, 2004 6:52 am
by GARTHWILSON
Quote:
but I'm interested in DIY.
For digital work, making printed-circuit boards is really highly impractical to DIY.

For analog circuits I used to start with single- or sometimes double-sided copper-clad PC board material, lay my circuit out with pencil so I could erase and correct, then drill it, and use paint with a really fine artist's brush to paint the traces with two coats. I always found the etch-resist pens to be woefully inadequate for preserving the copper where you want it while it got etched away everywhere else. After etching in ferric chloride (I think that was the stuff I got at Radio Shack), I'd use acetone to remove the paint, and drill the holes again just to get the paint out of them. The reason for drilling first is primarily so the pads and traces on the two sides line up. Of course without thru-plating, all component-layer pads had to be soldered on the component side, even if the same hole had a pad requiring solder on the solder side as well. After the initial cleaning of the copper, keeping my skin oils off it through the lay-out process and until after the etching was a bit of a challenge. The skin oils acted as an etch-resist too, so you could get copper finger prints shorting traces together. I suppose disposable latex gloves would be a workable solution.

What makes this method impractical for digital work is that it's so hard to make fine lines that go between IC pins, and digital is almost nothing but ICs. Another problem I ran into was that no matter how hard I tried to make sure I really got all the ferric chloride out after etching, a couple of years later a little corrosion would always appear at the edges of the traces. Maybe there's an easy solution I never found.

At a place where I worked in the mid-80's we used a clear plastic sheet material with two layers, a colorless one and a red one. We used X-Acto knives to cut the red layer to peel off between traces IIRC. There was a UV-sensitive liquid we'd put on a clean copper-clad board in a darkroom with a safelight, then put the board on a gizmo that would spin it to get the excess off and form a smooth film. After it was dry, we could put the ruby master on the board, turn on a UV light for a few seconds, and then put the board in the developer. After the developing process (whose details I no longer remember), we could put the board in the etchant. It was neat, but still not suitable for fine-line work.

My choice for the last 20 years for analog circuits is the solderable breadboards, primarily the one that is found under various names but Radio Shack sells as p/n 276-170, usually for two or three dollars. It looks like this:
RS_276-170.jpg
I've used scores of these for various projects over the last couple of decades. In fact you can see two of them in the first picture at http://www.6502.org/users/garth/projects.php?project=1 These two are not particularly good examples of the density you can get with this board; but what I like about it that I don't like about so many others is that this one has five holes in a row connected on each side of the center island where the ICs go, instead of the three that so many others have. It makes a big difference. You can actually do more with these than you can with the solderless breadboard equivalent pattern because A., you can also solder a few parts on the back, B., the holes are big enough that you can do the occasional two-leads-in-one hole trick, and C., you can cut a trace if you need two small nets on that row.

For digital construction though, it's tough to beat wire-wrap for DIY. You can put the sockets right against each other for a dense board, and done correctly, WW is 100% dependable. [Edit: See my WW Q&A at http://wilsonminesco.com/6502primer/WireWrap.html.] For minimizing problems with things like ground bounce in fast digital circuits, you can even get perfboard with plated-thru holes and a plane on each side so you can have power and ground planes, and solder chip capacitors for decoupling directly from the IC pins to the plane, minimizing inductance.

Fairchild's application note AN640 at https://www.onsemi.com/pub/Collateral/AN-640.pdf.pdf is a good one for understanding ground bounce and decoupling. AN680 at https://www.onsemi.com/pub/Collateral/AN-680-D.PDF addresses dynamic thresholds and noise margins, which ground bounce is involved in. AN393 at https://www.onsemi.com/pub/Collateral/AN-393-D.PDF deals with transmission-line effects. The newbie may think ground is ground is ground, and you may not intentionally have any RF transmission lines; but if you build with fast logic on a big layout with long wires especially with no ground plane, this stuff can bite you. 20 or 25 years ago some people in the electronics field went into digital thinking it would be simpler mathwise and they'd avoid some of the analog complications. But as digital got faster, it gradually turned into the worst form of analog—RF !

I could mention, just as a matter of interest, that LPKF Laser & Electronics http://www.lpkfusa.com sells what amounts to a milling machine made specifically to route your circuits from CAD data into a copper-clad board, mechanically, without using chemicals for the "etching" itself. Their top-of-the-line products will do 8-layer boards with .003" lines; but even their bottom-of-the-line model is probably a little beyond your budget, at about $10K IIRC.

Posted: Tue Feb 17, 2004 2:58 pm
by Kotlowy
hey, i'm not new to the whole 6502 wire wrapping thing. www.6502.org/users/alexis . I lost interest in it ages ago because it was just a pain to use. I put it on the web incase anyone else wanted to use the DMA section, which a few have.

I only wanted to use PCB's for the simple stuff, just for the sake of using PCB's I suppose.

Posted: Tue Feb 17, 2004 4:31 pm
by 8BIT
There has been some success in using the direct toner transfer method. This involves printing your layout onto a special paper (or transparency). Then using an iron to transfering it onto a blank pcb. The toner acts as the resist.

I've seen several kits advertised in magazines like Nuts & Volts. Here is a link to a document containing directions and paper properties.

http://www.fullnet.com/u/tomg/gooteepc.htm

Hope this helps!

Daryl

Posted: Wed Feb 18, 2004 4:14 am
by Kotlowy
yeah, jaycar (Australia) have them. Imported from the US. Uh... AU$34.95!!! But its 5 sheets re-usable. Are they any good now or are they still a little unreliable?

Posted: Wed Feb 18, 2004 3:20 pm
by 8BIT
Kotlowy wrote:
yeah, jaycar (Australia) have them. Imported from the US. Uh... AU$34.95!!! But its 5 sheets re-usable. Are they any good now or are they still a little unreliable?
I have not used them myself. If you read the article in the link I posted, there seems to be many variables. The quality of the laser printer, having your heat source (iron) too hot or cold, pushing too hard or not enough, holding it on too long or too short. Moving the sheet while ironing. Problems with removing the sheet when done. Even the etching process has a lot of variables... Using a good tank with heater and air bubbler was recommended.

Like anything new, there will be a learning curve. It may take several attempts before you find the process that works for you. The article made it sound like the finished boards looked pretty good...I only wish they had pictures of some finished products.

If you choose to try it, please come back and report your results. I for one would be interested.

Good luck!

Daryl