@drewfitz yeah, this design here is a two-layer one!
I'm not sure how you do that. I guess you just expose it twice, once on each side, and then do the etch step once?
The way this works: You have a blank PCB coated in a thin copper layer, and it's coated in a photoresist material. This material breaks down when exposed to light.
You then shine a bright (usually UV) light at it for a while, and the spots which got hit by light break down, while the other parts (which were in shadow) remain on the board.
You then use a corrosive chemical to etch away the copper layer, but it only etches away the parts that aren't covered in photoresist.
Normally you'd do this by printing or photocopying onto transparent paper, but I think they're saying you can just turn up your brightness/wait longer and blast the light straight through the paper.
And of course on one of the following pages, they had ads from companies who were happy to sell you the chemicals and equipment needed to make circuits this way.
The cool thing about this kind of photoresist PCB etching that may seem surprising if you know how circuit boards are designed today... you could totally do this without a computer.
Fundamentally you just need to come up with a photomask: that's some dark black ink on a transparency. You could use stencils and stickers and markers to make that, then turn it into a PCB. The only electronics you need is a bright light.
it's just neat to think that today PCB design is "you use this complicated program on a computer then send some files to a place that mails you back a bunch of PCBs", when 30 years ago it would be more "you carefully do some arts & crafts with stickers and then some chemical processes"
@foone you can still do that if you want to bang out prototypes fast! or if you just find it cool to make your own pcbs
i do find it pretty cool and what i recently came up with is to use a dlp projector with the color wheel removed and a high-power uv led instead of the lamp, a'la photolithography
i found a cheap 1024x768 dlp projector and i calculated that if i focus that to an area of 81.92x61.44mm i should get 0.08mm pixels which is pretty neat (for comparison a tqfp144's leg is ~0.22mm wide) but we'll see how that goes lol
@foone I always used the Letraset-style stickers and a special marker pen. You needed a drill press, too, unless you were doing surface-mount (which I don't think we'd even heard of).
@foone I'm pretty sure that you could use a resin 3d printer to do the photo-resist part of the process nowadays, cutting out a lot of the arts and crafts and just leaving you with the ferric chloride bath and interminable drilling.
@foone that's still how most people do it though, usually you just send off for a more cleanly/professionally made one later. You can go to any microcenter and still buy coated PCBs and copper etchent. I always preferred the "iron on" process to the photoresist process though as it's a lot cheaper than buying the photoresist coated boards. (glossy paper, laser printer, toner transfer to the copper coated board with heat, then dip).
I just mill test boards now though, way easier/cheaper.
@foone iirc 30 years ago I would have been using CAD (with parallel port dongle), stacking transparencies for better contrast, using a homemade light box and then sticking the PCB in a repurposed sandwich box full of ferric chloride 😆.
@foone I'm old enough to remember doing a couple of projects like this in school, and IMHO, it's a more civilised way to build PCBs for small projects;)
@vxo Did you ever see my Copy Pro Control board? I still made it in Kicad so I couldn't get to weird with it, but I did attempt to get as Weird as possible given the circumstances.
@foone pcb images printed in magazines often were printed in reverse so that they could be xerox-copied on to polyester film, which it barely stuck to, then the print flipped onto the copper board and toner transferred using a hot iron. then tape was used to mask spots the toner did not stick to. toner resists ferric chloride fairly well.
@foone See, I know nothing about the modern process (I knew I wasn't going to be a Hardware Person 30 years ago) but was passably familiar with the traditional process thanks to magazine articles...
@foone there were also these kinda neat rub-on transfers - The ones I had were intended to directly be etch resistant, but I'm sure you could use them as a mask just as readily. I've extremely vague memories of similar transfers being somewhat common in offices (for office purposes) at the time too.
@foone If you want to be stunned though look into what a gerber file defines and how that production process used to happen. Hint: It defines apertures for a moving UV light ;)
@foone if you print on thin printer paper it's transparent enough to UV to expose chemicals sensitive to actinic light, and toner is plenty opaque. Exposure might be more finicky because of lower contrast, and it probably wouldn't work at all with a dye-based inkjet printer.
@foone iirc one of the tricks was to add some sort of clear oil (like baby oil) to stick it to the PCB, which made the white paper go kinda clear.
the actual home etching process is super messy though. ferric chloride stains eeeeeverything, it's bad for your skin, and you have to neutralise it before disposal which also creates a lot of mess. I used to own a kit and used it only a couple of times because the cleanup is horrible.
@gsuberland@foone ferric perchloride is so old school and annoying. copper chloride is the way to go. in practice that just means hcl with a pinch of h2o2. when depleted it can be replenished with an air bubbler.
@foone it worked pretty well, i used to do this a lot. I usually xeroxed the layouts (to not cut into my precious magazine, or since they printed on both sides) and sprayed the paper with some orange-smelling oil to make it more transparent AND stick to the PCB's photosensitive layer. Exposed with a 300W Osram Vitralux for 4 minutes. Develop in NaOH, etch in FeCl3.
@foone I made a fair number of manually taped boards, but I never did the photoresist.
I barely remember laying down the tape though, so I can't even comment if it was hard or not. I feel it must have felt pretty trivial. But damned if I remember how I got the traces in the right places. I DO remember floating the boards in the acid bath and what happened if you left them too long. ;)
@foone I think it involved WD40 to make the paper more translucent, and ink side down on the PCB. A one time only use, of course. Elektor put them in the back of the magazine, separate from the articles.
@foone Huh. wonder if that was the original idea for them. I'd always thought it was for component ID/location. but possibly a way to replicate them o.o
@foone I vaguely recall doing this in school in the early 90s. You create the circuit on a transparent sheet, building it out of little black lines, circles for pads etc. Maybe they were stickers? I dunno. Anyway, somehow you put it on the board and shine light through it, them put the board in a vat of corrosive stuff which etched away leaving the desired circuit.
@foone I think I met something like this back in 1989 or 1990. A friend's dad was doing this with his own designs. He found that photocopies or laser prints on acetate weren't dense enough to block the UV properly, so you had to run it through the printer several times to get the ink density to make it work.
@foone He also had problems with drilling the holes in the board, so he was building an X-Y rig for a drill.
We spent a long night with an Acorn A440, me on the front end with the keyboard, him on the back end with the printer port and an oscilloscope, working out how to write a driver for this thing that ran off the !Draw files he'd use for the board designs, and would put the holes in the right place.
@foone Oh, I don't think that was Geoff's problem. He was definitely on the agricultural end of things with his designs. He just found it hard to put the drill in the right place every time. :catsmile:
@foone I recall drawing PCBs in the 1990s with CAD tools and giving them to a technician for the etching process (which involved handling corrosive products). I'm unsure whether he printed it on paper or on plastic material (like "transparency slides").
@MonniauxD Transparency slides seem to be the usual way to do this, but I guess you can use white paper if the light is bright enough and the ink is dark enough
@foone Oh, yes I was doing that when I was a teenager with my dad in the 80s. The difficult part once etched was drilling the holes for the components. That was very tedious. You could also purchase "décalcomanies" (sorry can't find an English word) of pads and tracks that you could transfer to the copper layer.
It's now so much easier to produce Gerber files and have the pcb manufactured II :)
@foone
Pah! That's the easy way.. I used to use a pen to manually, by hand, draw the tracks onto plain copper-clad boards. Also had letraset-type transfers to create tracks/pads, but never had much success with those. Was never able to afford the photo resist board. Dunk the drawn-on board into ferric chloride to etch off the uncovered copper.
Found one of the pens in my old component drawers. Must be over 40 years old, and still draws a line!
@foone A trick that I used back in that time (UV sensitized boards were too expensive for me) is to photocopy on ordinary paper the circuit, then put it on top of a bare copper clad board, use a nail to gently mark the position of each hole through the paper, use a pen to trace manually the tracks by copying the best I could the tracks and finally etch the board and drill the holes. You save the UV transfer project, but it quite error prone and slow.
@foone 15-20 years ago, UV photoresist was the “fancy” way of doing it in my uni in Brazil.
Usual one was, photocopy the design in the darkest setting to get as much toner as possible stuck to the page (thin paper worked best), then use a clothes iron to heat transfer it into a bare copper board.
Dissolve the paper in water, then throw the board into an FeCl3 solution to eat away the bare copper and leave the covered traces behind. Then scrub with IPA to clean it, and drill.