The people behind the PocketNC heard you like CNC PCB mills, so they milled you a PCB mill out of PCB. They announced their surprising new open source hardware product, a pocket sized 3-axis CNC machine entirely made out of FR4 PCB material, aptly named “FR4 Machine Shield”, at this year’s Bay Area Maker Faire.
UPDATE: The FR4 Machine Shield is now on Kickstarter
We know the concept from quadcopters, little robots, and generally things that are small enough to make use of their PCBs as a structural component. But an entire CNC machine, soldered together from a few dozen PCBs certainly takes it to the next level.
There is no doubt that 2mm thick fiber reinforced epoxy can be surprisingly rigid, although the Achilles heel of this method might be the solder joints. However, it looks like all load bearing, mechanical connections of the machine are supported by tightly interlocking “dovetail” finger-joints, which may help protecting all the solder connections from the strain hardening effects of continuous stress and spindle vibrations.
As you might expect, most of the wiring is embedded into the FR4 frame construction, and to squeeze the maximum value out of the PCB material, the motor driver boards interface via card edge connectors with the (currently Arduino based) controller board. In addition to the milling head, which features a brushless DC motor and a tool coupler, the team wants to develop heads for circuit printing, microscopy, pneumatic pick and place, hot air reflow, and 3D printing.
With all those cost-driven design choices, from the one-step manufacturing process of the frame and wiring to the dismissal of screws and nuts from the frame assembly, the “FR4 Machine Shield” could indeed become one of the cheapest CNC machine kits on the market. The team targets an introduction price of $400 during a Kickstarter campaign in June 2016. Can they deliver? [Gerrit] checked Pocket NC out at the Faire and ended up raving about how they run their business.
Enjoy their teaser video below!
The recursion level is just too deep for me.
Unfortunately, it looks like most of the FR4 pieces are too big to produce on this machine, so the best you can do is make a smaller machine with it.
That doesn’t make the machine useless though… you’d be able to make some of the smaller parts for this.
And then a smaller machine on that machine…
And you’re on your way to Heinlein’s story “Waldo”.
Matryoshka dolls!
Never did like Matryoshka dolls. They always seem so full of themselves! :P
why did I laugh at that?
[Gerrit] check[ed]?
Are there any real world technical specifications available?
You can buy 50/50 solder or you could even go a little higher in filler hardness, the limitation being the temp/heat that will cause de-lamination.
What gets me though is the (eh em) horrible setup of the stacked axis gantry. It looks like the distance from the y axis plane to the z axis plane is larger than the distance between the y axis rails. That means the forward-backward play at the tool tip is going to be greater than twice the play on the y axis rails. ouch
Has anyone done some software to calculate these geometric tolerances by design? or do I have to go “do the math”?
I’m not bagging out here, it’s a excellent idea but some refinement may be on the way.
Adding a grid of vias in the pours to anchor them should help with mechanical de-lamination. Additional corner braces pieces could be added along the edges to make them more rigid..
+1 on opening line…
Given the rigidity (or lack thereof) of FR4 in sheets of that size, it seems like there would be some torsion/flexion of some of the members when the machine is actually under load. Probably not hard to significantly reduce, but I wouldn’t want to run anything critical on it until I’d done several stiffening cycles.
Will be interesting to see how this develops… price doesn’t seem too bad, considering. I hope they’re successful with this, it could be a game changer for folks who’ve been otherwise unable to afford such a tool/hobby!
On the other hand, compared to CNC3020 at around $500 + some electronics upgrades at around $100, the $400 for a much smaller and less rigid machine seems quite a lot. But nice idea and maybe it will come down in price or be available as a kit.
Those aren’t dovetails, those are finger joints.
Matthias Wandel gets mentioned on here enough times you should know this by now.
2012(or 11?)
[youtube]http://m.youtube.com/watch?v=XjF1wh0BRD4[/youtube]
https://m.youtube.com/watch?v=XjF1wh0BRD4
Shhh, it’s not real unless it was made by someone in America.
/S
Live on Kickstarter through August 8!
https://www.kickstarter.com/projects/1090944145/fr4-machine-shield
All’s I can say is; I hope the do a better job at “Getting started and up and running then the pocketnc did”. We’ve had a pocketnc for two months and have not made a single cut with it. There fusion to pocket video example (your first project) is way to complicated 8 blade impeller. They make nice machines but the training really sucks IMO. Not to mention the convoluted software control pathway.
Sorry that you’ve had a tough time with your machine. We are aware that we could use more(and more simple) training and are working towards that while trying to keep our heads above water. We would love the chance to help you get up and going with your machine. I couldn’t figure out which one it is to contact you, but if you email us at info@pocketnc.com then we’ll do our best to help you get up and going.