It is perhaps a surprise that the widespread adoption of CNC machinery in the home has not come from 3D printing or desktop mills, but as a quiet revolution in the crafting industry. CNC cutters for plastic or card have been around for quite a while now, and while the prospect of cutwork greetings cards might not set all maker pulses racing these cutters do have significant untapped potential in other directions. Perhaps you have to own a carburetor whose gaskets have been unavailable since the 1960s to truly appreciate that.
[James Muraca] has a KNK Force, something of an object of desire in the world of desktop CNC cutters. The computer inside the Force is a Raspberry Pi, so of course [James] set about investigating its potential for running his own software. His progress so far is on GitHub, a web interface through which you can upload and cut an SVG file, but his plans are more ambitious. He hopes to turn his machine into a complete PCB manufacturing station, able to both cut the PCB, and with the addition of a vacuum attachment to pick and place components.
The KNK Force is an interesting machine not just because it is powered by a Raspberry Pi. Its cutter head is a rotary tool with a Z axis, so it can perform more heavy-duty and complex cutting tasks than its competition. In addition it has a camera built-in, and it is this feature that [James] hopes to use in his PCB project.
We’ve covered plenty of cutter projects before, from projects turning CNC machines and pen plotters into vinyl cutters to using a cutter as a laser engraver and even cutting solder paste stencils with one. We look forward to further progress on [James’s] project.
i’ve always wondered about these, everyone is still cutting copper or using (less than ideal) conductive ink. why can’t these machines ‘write’ circuits with solder paste? you could then stick your smd components on and bake it.
i’m sure there is an obvious answer..
They can do that, and people are doing it.
Why isnt it more common? Because traditional DIY board methods work better in almost every way, are cheap, simple, quick to reproduce, create boards more suitable for prototyping development, easily can have SMT or through hole parts, and last but not least do not have crazy-high resistance per inch of trace.
In the future this may change. Go to google and google “3d printed circuits” and be amazed by the dozens of interesting and creative ways people are plotting circuit boards.
ah, i had assumed that solder paste would have low resistance. thanks for the info
Solder paste DOES have low resistance. But it’s also impossible to make a circuit out of that alone, since surface tension just makes it want to ball up. Anything that would keep it in shape (that is to say, anything it will readily alloy with) is itself conductive, and therefore useless as a substrate for a PCB.
The “conductive ink” stuff is just a suspension of conductive particulate in an adhesive. It’s inherently high-resistance due to poor contact between particles in the suspension. Theoretically, you could get around this by applying it as the usual suspension, then sintering the particles while burning off the rest of it, but I don’t know of any real successes along that route yet. Oxidation and the aforementioned surface-tension issue cause problems with it.
Honestly, I don’t know what all the fuss is about regarding making circuit boards from things other than the usual copper-clad FR4/CEM. While I see the appeal of a sort of “push button, receive board” approach and with the glaring exception of plated through holes being problematic for homebrewers, making the PCB is arguably the easiest part. Assembly is much more time consuming for anything that isn’t quite low on the component count.
“Honestly, I don’t know what all the fuss is about regarding making circuit boards from things other than the usual copper-clad FR4/CEM.”
Word.
It takes me 2x as long to drill all the holes and fiddle with the through-holing (where needed) than it does to laser print, iron, and etch up some copper. Would I skip the intermediate toner-transfer step if I could? Sure, but it would only save like 5 minutes, so whatever the alternative is, it’d better be simple, cheap, and fast. Firing up a 3D printer to make aqueducts for molten solder is none of those.
“Theoretically, you could get around this by applying it as the usual suspension, then sintering the particles while burning off the rest”
This is done in thick film hybrid circuits. But to sinter particles of good conductivity – generally metal – you need high temperatures, higher than FR4 can sustain. So this is done with alumina ceramic as a substrate. I am not sure if somebody could make aup a low temperature version of this process, which is useful on polymer substrate.
But as you write: making the traces, even with photo-etching is not the difficulty. It gets really difficult if you want plated through holes. And there is no cheap, easy and reliable DIY process for this.
Surface tension of molten solder is why people don’t do it. Draw a line of solder paste on a piece of bare circuit board (no copper) and put it on your hot plate or in your oven. See what happens.
Ah.. This also makes sense.. Add it to the list.
For most of the somewhat sucessful approaches to ‘printing PCBs’ people use ‘conductive glue’ type formulations, hence the high resistance I was referring to.
Good point.
There are some good articles here on HaD about people making ‘printed PCBs’
There is a PCB printer out, which uses chemically deposited silver (silver nitrate and a reducing agent like citric acid or ascorbic acid). But the traces are really thin and so have high resistance too. You can overprint several times, but the “ink” cartridges are quite expensive.
That might be improvable by printing solder resist onto the unwanted areas of the board, producing similar results to HASL.
I’m not clear on what he is doing here – is he taking a cladded blank board and cutting away the copper that isn’t one of the traces?
i think he is using the mqchines cnc milling ability?
Custom gaskets as a compelling use case?!? I guess how any gasket (including any size holes) can easily be produced from gasket stock using nothing but a small ball peen hammer must be a lost art indeed…
You’re absolutely right, and I have cut many gaskets that way over the years. But cutting a water pump gasket by hand is one thing, a carburettor gasket is another. For example the Solex on my Triumph Herald 948 has some extraordinarily convoluted sections of roughly 3mm wide card, and believe me as one who has tried that’s not a gasket that can be hand-cut by mere mortals. And I wouldn’t want a ball-been hammer anywhere near it :)