Wednesday, June 5, 2013

On 3D Printing And Making Guns

A little diversion on the subject of making guns and stuff. 

In broad terms, there are two ways of making things: additive and subtractive.  Most of us have made things out of wood, even if it's just from a mandatory wood shop class we took in school.  Woodworking is a great example of a subtractive process.  You start out with boards, cut away portions to cut elements to size, maybe cut away portions to make strong joints, or thin the the stock down to size with a plane.  Of course, it started out as a tree, so a lot was cut away or thrown away to get the lumber you started with.  Sculpting, as most of it seems to be done, is another subtractive process.  As one sculptor said, start with a block of marble and cut away everything that doesn't look like (his finished piece). 

In contrast, there are additive processes, where you put down only what you need.  A rough example would be slip casting ceramics, where you pour a water-clay mixture into a mold, and form a layer of the ceramic on the inside of the mold.  You eventually remove the mold and are left with a thin layer of clay that is then glazed (or not) and fired.  Likewise, in lost wax casting, widely used in the jewelry industry, a wax mold is embedded into a ceramic that dries and sets up hard, then fired to produce a copy of the wax mold inside the ceramic, and molten gold (or silver, palladium, platinum...) is poured into the mold, putting down just the amount needed.   

Clearly, 3D printing is an additive process; the subtractive process is usually just called machining. 

Quite some time ago, I did a piece on The Futility of Gun Control, in which I linked to a piece by Popehat called "The Third Wave, CNC, Stereolithography, and the end of gun control".  In that piece, he talks about 3D printers and CNC machine tools and was really right on target.  As someone with the kind of low-end, home CNC machines he talks about, I really think the emphasis on 3D printing lately, and especially the bedwetting over 3D printing a gun, is a bit excessive. 

For $1300, your local Staples will get you into a Cube 3D printer.  I'm not even sure if that really gets you started, and I'm less sure you could build anything other than a single shot gun that might be as dangerous to you as someone you need to shoot.  For about two bills more than that price, you can home build a basic CNC RouterGizmodo just ran a piece on a Shopbot at $5000, with a work area of about 24 x 18 x 2".  Clearly intended for thin stock, like making signs or cutting things out of plywood, this is almost 4x the price of the 3D printer, and while it can make serious stuff, I don't believe you could finish an 80% lower on it.  Check out the video on this page of a couple that makes a cool triple bunk bed for their obviously delighted daughters.  They milled what looks like plywood or MDF and assembled it themselves. 
Note that the cool machine they show in most of the pictures in that article isn't this little $5000 machine, it's Shopbot's $40,000 five axis machine.  Personally, if I was going to drop $5000 on a ready made machine, I'd get one of these from Little Machine Shop:

This is a Sieg (Chinese) CNC machine with a work envelope that's 10.2" × 4.3" × 7.1"; smaller in XY than the desktop Shopbot (it's not designed for quarter sheets of plywood), but tall enough to cut that fire control pocket.  It will cut aluminum, steel, stainless, titanium, brass, and other really useful materials along with wood, plastic and wax (useful for models).   Its list price is $500 less than that Shopbot, but it weighs 275 pounds so it will cost a bit to ship it and get into place in your garage.  By the way, this is considered a mini-mill.  It's not a big machine by anybody's standards.

We can get into the particulars of machines another time.  The 3D printer, as I see it, has only one thing going for it: it doesn't require much from the user to create a piece that has been designed and is known to work.  The same thing can't be said for the CNC milling machine, which needs some knowledge of machine shop practice from user.  You need to know, for example, the way you clamp your work down can change its dimensions enough to matter.  The 3D printer, to the best of my knowledge, isn't terribly dangerous.  It can burn you if the plastic is hot, but a well designed machine is probably not going to let a user stick their fingers into places where they can get hurt.  A mill, on the other hand, is capable of cutting steel, so your finger isn't going to offer much resistance!  On the other hand, if you know it can take your hand off, you tend to keep that in mind, and there's no reason any adult who can cook on a hot stove, or shoot safely, can't use machine tools safely.

Finally, I have to say that no matter what you get, it's just the start.  For the CNC machines you'll need design software and very likely software to turn part shapes into the instructions for the machine.  For the printers, you'll need 3D modeling software, too.   Some of them seem to be shipping proprietary software with their printers. 


11 comments:

  1. While the thought of a turn key CNC is appealing, I think it would cost about a grand to convert a X2 mini mill to stepper motor driven CNC control. It may not be as pretty, but where's the fun in that?

    ReplyDelete
    Replies
    1. I just meant if you're going to drop $5k on a CNC setup, I think the mill has advantages over the Shopbot. I built mine from scratch, but it ends up that building the CNC becomes as big a hobby as building stuff with the CNC, and I'm sure some people would rather just use it. And to have someone to call when things aren't working the way they ought to.

      The drawback to that Seig is that it doesn't have a big work area. In fact, my hybrid Sherline/A2ZCNC is the same, if not bigger. OTOH, it has a cutting fluid system, which mine doesn't. A lot of trades to look at.

      Delete
  2. If someone does the path files and documents the workholding, you can learn enough in less than a week to actually turn out an AR15 lower on a CNC mill. Yes, it's harder, but in absolute terms, it's not that hard. After taking less than 10 hours of training total on a manual mill, metal lathe and Tormach CNC mill, TechShop is willing to let me use all of the equipment. Figuring out the path files and the workholding is a bit tougher, but once documented, it's easy to duplicate.

    Also, TechShop is okay with working on firearms as long as you don't do anything dangerous, illegal or stupid.

    ReplyDelete
  3. look at some plans at diylilcnc.org to make your own CNC.

    ReplyDelete
    Replies
    1. Cool little system. There are several places online that offer plans. Usually for a reasonable price.

      Delete
  4. So....tools like the Shopbot can easily cut shapes out of thin stock. Is there any reason why a laminated receiver wouldn't work? There would be some boring required, and perhaps some threading, but it seems that a CNC router, a drill press, a few taps and some rivets could produce a workable receiver that would hold up quite well for at least several hundred rounds. Or, am I missing something?

    ReplyDelete
    Replies
    1. Oops. Bad editing. Sometime before I posted this, I had a paragraph from my older post in here that had a link to plans for an an AR lower made from plates on Weaponeer forum. I edited it out thinking I shouldn't re-use so much old work, but I took out the good link.

      Bottom line, not being able to mill out the fire control pocket of a lower doesn't keep you from making an AR - or other gun. It just requires adapting the work to the tool.

      Delete
  5. Who needs a CNC mill? You can buy 60%-80% complete lower receivers and with a little drilling on a drill press you can complete them. See "More than 10,000 80% complete AR-15 receivers already sold" here:
    http://www.examiner.com/article/more-than-10-000-80-complete-ar-15-receivers-already-sold

    NOTE: i have read that Examiner.com is supported by popup ads, so I strongly recommend you have a popup blocker or script-blocker enabled.

    ReplyDelete
    Replies
    1. Anonymous,

      You might want to check out Graybeard's site a bit more, he has a link on here to the 80% lowers he has built.

      The point of buying a CNC mill is not just to build something simple like an AR lower, but other receivers that are much more complex.

      Delete
  6. Laminated receivers - Crosley Motor Company, familiar with constucting electrical equipment, built their original production automotive engine blocks from stacked stampings oven brazed together. A decade later the engines were popular in Y-class hydroplane racing. So it appears they were durable.
    Rather than stamping equipment used for transformer laminations, a CNC plasma cutter should do nicely, with some cleanup.

    There are hundreds, perhaps thousands of Fadal class machining centers with machining time available in home garages and small shops nationwide. With tooling already in place, a tool path code should provide lotsa chip producing activity. It's aluminum, after all. From a billet.

    itor

    ReplyDelete
    Replies
    1. In the magazine I subscribe to for home CNC, Digital Machinist, someone estimated around 10,000 home/garage CNC systems in the US. It could be a group of cottage industries producing all sorts of things, and there's a cottage industry to supply us.

      The same publisher produces a magazine called Home Shop Machinist. Its circulation is bigger, and back issues sell on eBay pretty close to cover price, which means they hold their value better than paper money.

      Delete