Ok, so slightly deeper dive in my first foray into DIY PCB fabrication in over a decade.
I don’t have a great history with making my own PCBs. My first attempts were in the 80’s with dry transfer pads, tape, and resist pens. Not only was it tedious and resulted in clunky layouts but I never had much luck with actually etching. I think the big problem was I never heated the ferric chloride I was using at the time. But the only boards I actually managed to get the copper to etch on - it also etched under my design leaving the boards useless.
In college in the 90’s I gave it a try again. This time I used pre-sensitized boards since I was working in a photo lab and had access to very nice (and in some cases highly specialized) enlargers, light tables, contact printing frames, laser printers with stocks of transparencies and other crazy imaging supplies. The lab was part of the Schools college of medicine and we were responsible for providing everything from headshots for professors to reproducing x-rays to slideshows to digital productions. We got to do a lot of neat crazy things and they had to be done JUST RIGHT since it was all medical training stuff. So I was able to get designs transferred to boards much more accurately. But I still didn’t have a lot of luck with etching - I did know to heat my etchant by this time…but still I tended to have a lot of undercutting.
In the early 2000’s I tried again since I now had my own laser printer and heard about the toner transfer technique. I had a hard time finding paper that worked well for the method though. I never really got a clean enough transfer to bother trying to etch anything.
Then I heard about alternatives to ferric chloride and got interested in trying to make some boards again…but…by then OHSpark existed and most of the things I was interested in building boards were available cheap so I never got around to it.
Now that I’m building all these modules I’m getting a little frustrated with strip board. I’m enjoying building with it. But having used EDA software for board layouts I’m a bit spoiled and find having to go from schematic to layout without the benefit of netlists and ratlines annoying. And I’ve had bad luck buying stripboard off Amazon (even though listed as stripboard and shown as stripboard in the photos both times I’ve ordered it what I got was just perf board with individual pads.) The stripboard from Tayda is nice…but their large board is still too small for a lot of @lookmumnocomputer and @EddyBergman’s layouts. A few of my modules I redrew existing designs just to make them more compact and get them to fit the boards I could find.
But I really want some PCB’s for bigger projects. The Clock Divider was kind of the last straw for me…I was able to fit it onto a Tayda large board…but just barely. And it should be a lot more compact. Just the limitations of stripboard made it bigger than I wanted. I could just order boards from JLC and such…and have (in fact just got 10 of @analogoutput’s buss boards today…I don’t need 10 but it was only a few bucks more to get 10 than 5 so I got more than I need so I can share.) JLC is fast and I’ve yet to be disappointed in the quality of any board I’ve got from them…and they’re cheap.
But I really don’t like having to deal with 4 extra boards on a lot of things. I can sell them…but I’m terrible about shipping stuff and just don’t like to deal with it if I can avoid it. They’re cheap enough I could just toss them…but I can’t bring myself to do that. And if there’s a mistake in the board that isn’t easily worked around…well then it was all just a waste.
So - decided I want to try making my own boards again.
The one big expensive item I have that makes this “easy” is a laser printer. And honestly it’s not a big expense. I gave up on inkjet and switched to laser around 2002. Ink is just too expensive and dries up too quick. Toner doesn’t go bad as it sits…and a refill lasts me YEARS and per-page is way cheaper than ink. Yeah…I can’t do color. But I seldom miss it. When I do need color I go to Kinkos or use the color laser we have at work. I get cheap $100 Brother lasers. My old one finally died after over a decade of use so in March of 2019 I got a new one. It’s a full muti-function with wifi and bluetooth so I can print from my phone and put the printer anywhere. It will scan to dropbox or google drive, prints quick, and still uses the same cheap toner cartridges my old printer did. And it still only cost me $100. The one thing it doesn’t have is a sheet feeder for the scanner. I still have an older canon inkjet that hasn’t had ink in it in over a decade that I keep around because it does have a nice sheet feeder that is handy when I need to scan a stack of documents.
But I’m getting off topic. The point is I much prefer laser printers over inkjets so I already have a laser.
I’m giving two methods a try. Toner transfer and UV dry resist. For the toner transfer I got some special paper designed just for it off Amazon. It’s only $8 for 10 sheets so not overly expensive. Plus they’re A4 sheets so you can fit a couple of boards on one sheet. But…you’ll either have to batch up a bunch of designs at once or cut it up into sizes you can sheet feed in your laser printer to make the most of it.
For the UV resist I ordered this roll off amazon $13 for a 30cm x 5m roll. Should last me a long time. And apparently it can be used both as a mask for etching and as a solder mask. So that will be fun to experiment with.
Today my bare PCB’s came (they were supposed to be here sooner but USPS decided to take them on a tour of Nevada before getting them to me.) and I was itching to try etching one. I hadn’t ordered any laser printable transparencies figuring I could just get them locally. But at walmart today doing my weekly grocery shopping I couldn’t find any. I did find a Scotch laminator for $21 which was cheaper than getting one off Amazon and may be helpful for some of these processes so I picked it up. I also stopped by Staples to look for transparencies but they wanted $75 for a box of 50 which was a bit ridiculous - apparently people are using them for DIY face shields so the prices are a bit inflated right now. Amazon had a box of 50 for $18 so I ordered those…but won’t have them until Tuesday.
So…with no transparency film the UV options were off the table for today. So let’s try the toner transfer.
I didn’t want to take time to draw up a schematic and board myself first so looked to see what may be out there than I’d like to build and has an open source PCB design available. I was going to try one of
North Coast’s projects since they share kicad sources…but…I can’t get them to open 
Kicad keeps saying my version is out of date to open their files…but I downloaded the most recent version so I’m not sure what’s up there.
Yusynth shares PCB designs…but he only shares JPG’s and PDF’s. But they looked to be about the size of the bare boards I had and I do want an ARP filter so… decided to give it a try. The JPG seems kind of silly since trying to get a bitmap to print at just the right size is a pain. So I grabbed the PDF…did a test print on normal paper and confirmed that it printed at the right dimensions since IC sockets lined up with the pads.
So - I did a print on the special transfer paper through the manual feed slot in my printer. Even got it setup the right way up first try. I then realized that my boards were slightly shorter than the design…oh well - I expected a bunch of fail on this first try so wasn’t overly concerned with getting things just right. I taped the board to the transfer paper and then sent it through the laminator it’s 5mil setting. For good measure I then flipped it over and sent it through again…then did it once more on each side before I lifted the paper to see how the transfer did.
Well, it didn’t do great. I wasn’t surprised. Usually you have to hack these laminators and slow them down for this. Instead I got out my iron and IR thermometer. Dialed it in at 155c and just ironed it manually. That MOSTLY worked…but a few of the edges still didn’t transfer well. I kept trying more heat and more pressure in those areas but just wasn’t getting a full transfer. Oh well…first test…don’t even know if my etchant will work and some of the ground traces are off the board so…let’s not overthink this.
I peeled the paper off all the way and took a closer look. Most of it actaully transferred well. In fact…better than I was hoping. This was when I realized that the Yusynth files weren’t very clean and even the PDF was a bitmap. You can see the noise in the print on normal paper:
This actually IS the file not my printer. It’s hard to tell when you look at it on screen but that noise is there…it’s just lighter and less visible. It’s amplified in my print since I turned up the density on the print as far as it would go to make sure I got a good solid mask. Oh well. First try…experiments…let’s see how this goes.
I went over a few iffy traces with a sharpie to see if it would work as a resist and help with the more troubling spots. It didn’t hurt…but I’m not sure it really helped much. I don’t think it was very effective as a resist.
For the etchant I was debating two options. The first is Muratic Acid and hydrogen peroxide…I actually have almost a gallon of it on hand from when we had a pool 2 years ago. But that’s a bit more hassle to be safe with and I wasn’t sure if I still had it. So I decided to try the slower gentler option of Vinegar and Hydrogen Peroxide.
The solution is pretty simple 1:1 Vinegar and normal 3% drugstore Hydrogen Peroxide - then add some salt. I couldn’t find details on how much salt…the advice tends to be if the solution doesn’t fizz and react to the copper add more salt until it does. Heat does help it work more effectively as well so as a quick and easy test I turned on the heated bed on my 3D printer to 85c and put the tank on there - then in went my board:
It took about 12-15 minutes. Near the end it looked like there were spots that still weren’t etching, but I decided most of it was just the “noise” in the print and that it was time to stop.
I pulled it out and washed it off with lots of fresh water. Then wiped it down with some acetone to remove the toner.
For the most part it looked like the etch was pretty accurate…the big problem was the quality of the print I used as a mask:
I got out my multimeter and started probing traces. Surprisingly none of the iffy looking traces failed a continuity test. I did find 3 small bridges…I couldn’t see them…but the meter doesn’t lie. A quick cut with an xacto took care of them:
There are two bridges in that image. The two sets of 3 pads for two transistors each had a bridge. The top row the two left most pads were bridged and the lower row the two right most pads were bridged.
Not much of a difference…but it probes out ok now.
So it’s not a great board…but it’s a far better first attempt than I expected given my past experiences with etching boards. This board probably could be used. I’ll use it to practice my drill technique but don’t think I’ll actually build it. Maybe I will…but not really planning to yet. We’ll see how the drilling goes once I find my PCB bits.
I’m really looking forward to trying the UV resist. And working from some better imagery to start with. If I can’t find some kicad or eagle files for a small modular PCB I’ll make time later this week to design my own for something. I have a day or two while I wait for my transparency film anyway.
But overall - it worked better than I expected. I’m also curious about this product: https://www.pcbfx.com/main_site/pages/products/toner_foils.html
Apparently it’s foil which bonds to toner and gives a better mask. So instead of worrying as much about getting enough density out of the laser you can use this to ‘beef up’ the transfer. They also make a white version for a poor mans silkscreen. (and I wonder if the green could be used as a solder mask…though I’m also looking at trying some UV curable solder mask.)
I also have another method to try…which is actually the method that got me interested in trying my own PCB’s. Using a resin 3d printer to “print” a mask on the board directly. I’m curious to see how that will work out.
