Snap-together PCB Mill

[Jonathan Ward] came up with the MTM Snap, a snap-together pcb mill as part of the Machines That Make group at MIT.

We covered [Jonathan]‘s previous work made out of half-inch plywood, but the new iteration of his PCB includes a clever snap-together mechanism instead of screws and bolts. Although the MTM Snap looks a lot like3d printers such as a reprap, the similarities end with the off-the-shelf stepper motors. Instead of using motor drivers and control electronics from a reprap, the project uses custom stepper drivers, controlled by a bare Arduino.

We’re really impressed with the results of the MTM Snap compared with what is possible on a reprap-derived milling machine like a makerbot or wolfstrap. We’re thinking that’s due to the mass of this project compared to the printed ABS parts of the ‘common’ 3d printers, but any MEs are more than happy to correct that notion.

Check out the video after the break to see the machine in action and a great view of the snap-fit mechanism.

[ ?posts_id=5080926&dest=-1]

26 thoughts on “Snap-together PCB Mill

  1. I will be watching closely this project, it would be great to have my own CNC and having more options to choose is always wellcome!
    And this one seem ideal to create small PCBs
    I hope they will be offering soon a kit, at least the snap together frame :-)

  2. Wait…. what results? I watched the video and don’t see any. I just saw it moving around. How good is this thing? It has potential but where are the results?

  3. Stiffness would be the correct wording, not mass. This mill’s increased stiffness improves the mill’s accuracy when subject to cutting forces. More Stiffness -> Less deflection under the same loading.

  4. Neat toy. Probably does a decent job milling PCBs and foam for mold making/casting, but certainly don’t expect it to be able to do real machining. The rigidity just isn’t there.

  5. @Isotope: probably a good thing they don’t claim it can. It’s certainly the best PCB milling rig I’ve seen because of the clean design and portability.

    If they used a dense/heavy polymer and better motors they could do metals and wood too.

  6. personally i think this is dumb
    the difficult and expensive part of cnc is the precision, that means expensive drive mechanisms

    reprap is a failure because the whole point was you could build one and then help other people build their own, except nobody wanted to make parts and ship them for free to anyone.
    makerbot was a failure because it was an attempt to make a low cost cnc type thing but guess what it was expensive and it sucked.

    this kind of “open source” stuff is simply not suited to projects where the whole design is pretty simple an idiot could have made it, but then relies heavily on expensive parts that have to be bought or sourced second hand at still high cost.

    when someone can build a reliable, accurate, repeating drive mechanism for cheap then sure, perfect. until then these projects are doomed.

    anyone who gets fooled into paying money for what is essentially a box because it means they only have to snap it together instead of build it… well…
    call yourselves hackers?

  7. I’ll say it again, a bom of $658 is not what I’m looking for, as pff has said, the design is simple enough, the only bit that is remotely interesting is the clip together part but at that price it just isn’t worth the effort.

  8. i have been tinkering with various cnc designs and configurations. it is exceedingly difficult to get the cost low while maintaining good results. granted even the cheapest machine is better than all but the most skilled humans however it still costs hundreds.

    the big cost that cannot be mitigated readily by the diy community right now is proper steppers. they cost. even the drivescrews and bearings that can be technically manufactured at home would be hard to make precise on diy equipment and would not cost much less. the frame gets the most attention yet is just a small fraction of the overall cost.

    i, along with everyone else it seems, am trying to make a design that has a low price point at reasonable quality. the few that do bring products to market usually state things such as “Having precision linear bearings.” that are actually just bits of plastic with no bearings at all. worse, claiming the machine’s resolution is the same as its precision. most steppers can microstep down to .0000625″ and similar ridiculous numbers when the screws that they are connected to can only do .003″ per foot.

    all diy machines are crap, but people are trying to change it

  9. I think some of you are missing the point. They’re not trying to mass-produce high precession parts for nuclear reactors, they’re giving hobbyists a way to make a few parts when needed.

    Sure, a coat hanger may not be the height of design, but if I needed to crank out ten pcb’s once in a while, or make a custom case for an Arduino based project these are an interesting alternative to going to a machine shop and having project made for me.

    I’d like to see this device crossed with a Make-Bot. First the Make-Bot lays out the gross structure, and then the cutting tool comes in and refines the shape and cuts whatever holes are needed.

    Keep the total cost under $1000.00 and then folks will snap it up at the Craftsmen department at Sears. I sure would.

  10. Why would anyone buy a plastic mill for $700, when harbor freight has a cast iron one for $600? I’ve never understood these cheaply-built pseudo-machine-tools. Just buy the real thing, or build it yourself (see: Dave Gingery’s books). The whole idea reeks of software guys trying to design hardware.

  11. @Devin – Why would anyone buy a “cheaply-built pseudo-machine-tool” at Harbor Freight? I have seen new out of the box tools from them that are so grossly out of spec that they were dangerous. Spending hundreds on poorly cast iron is no better than spending it on plastic.

  12. @Drone

    kickstarter is the worst thing i have ever seen
    any CNC forum has hundreds of designs that are probably better than this one in some way or another, and none of them involve some whiny kid asking for money from people for some unnecessarily machined bar stock and some v bearings bolted to a panel.

    The slide is arguably the easy part. get back to me when someone makes a cheap linear drive

  13. I got my CNC from china brand new for about 900$ and it’s an excellent machine. It has rigidity and precision well above any plastic frame could deliver. It includes a very nice controller box that keeps everything nice and compact. The work size is also larger than this (about 12″ x 8″ x 2″). So why bother get a plastic one when I can get an aluminum one? They even sell smaller ones, for about 600$ and they still are one class above this one.
    I’m not against building your CNC. I’ve build wood CNCs that sucked compared to my new aluminum one. At the moment I’m building a 5 axis one just because I can’t find a commercial CNC that doesn’t cost a fortune. I’m just against building something almost as expensive as a commercial unit but much worse. If this was about 100$, then great, I’d get one too. But for 500$, there are better options.

  14. The cheapest pcb miller i found are “cmode cnc” (just search for that) for around 650 euro on ebay, not sure if that is any good, cannot find any reviews or user impressions about them.

  15. When all mills have been built, all repraps assembled and every stupidity tried out, man will notice that PLASTIC IS JUST NOT STABLE ENOUGH FOR MILLING ANYTHING EXCEPT CHEESE

  16. @Meinaart: search for CNC 2520 on ebay. That one is only $499. Mine is a little larger, I think.
    You could also search for SABLE 2015. It’s smaller but I think it’s more rigid. Also $499 :)

  17. I think this highlights that academia could probably improve the practicality of these designs. Of course this brilliant exercise points out that snap together is a good choice for making bigger parts out of smaller removal technology instead of 3D printers. This is a good point.

    However, the reprap community seems to obsess over “Vitamins” or non-printed parts. In reality they probably could print a bunch more stuff if simple non-printed materials were included. In this case sheet plastic, but think about angle-iron (the kind with holes). One could print parts that just clip in the angle-iron’s holes? This commonly available material is cheap and rigid.

    I work at an industrial site where they stock the somewhat fancier u-track used to hold electrical and plumbing to structures. They make everything out that stuff! Screws together, or snap printed plastic to it.

    What we need is better CAM software I think…

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.