CNC Machines are not Magic

Today we’re taking a look at what is required to use a modern CNC machining center to reproduce a part designed for factory production on a bank of horizontal milling machines. It’s not as trivial as many people think –  CNC has some definite limitations. The part is from an RP46 top cover, a conversion to made the DP28/DPM light machine gun into a belt-fed weapon instead of using pan magazines.

26 comments to CNC Machines are not Magic

  • j

    nice video i did not think that making something that looks simple could be so complicated

  • Walter Mittey

    Assuming the final part discussed was the linkage to the “op rod/charging handle” – somewhat similar to a DShK feed assembly? Would suggest EDM for the finer cuts so there won’t be any tool chatter, but EDM = $$$$ – your idea of clamping is probably the best/cheapest way to go, though. You mentioned welding – one way to go would be to mill only part of the “actuator fork” and then weld the rest on. Dunno how well it would hold up to repeated battering, though.

    Hope it all works out without too much hair pulling.

  • Ownerus

    Could also cut a flat pattern and form the two 90° projections then machine that. The offset from the edge of the large fork would require somewhat thick material but still better than cutting from a solid billet. OR, cut a through slot for the large fork, weld in the roughed blank from the back side, finish machine the part, feed groove later. I think I’d prefer the formed blank because it should probably be heat treatable and therefore not as desirable for welding.

    Some of the original processes probably involved broaching in addition to horizontal milling machines.

  • Matt

    Powder metals and printed metals are the future. I’m sure it could be made with either of those methods.

  • Hrachya Hayrapetyan

    What if make the last part from two layers of metal sheet and fix the 3rd axis part between two layers by folded “wings”. Upper layer will have that groove cutaway and when it is welded or riveted on bottom layer it will form the cannelure. I am not sure if I could express my idea well enough …
    Well, it won’t be like the original, but it is another possible way to make it.

  • Anonymous

    Never realized the full complexity for a part like this, especially the angled surfaces. Then again, I’m sure the overhead for a CNC is much lower in the long run than the overhead for such specialized and numerous horizontal mills.
    I remember seeing a video inside Colt’s factory as they made 1911 pistols, and they too had a complex mix of CNCs cutting the basic frame and shape followed by machines edging on 70 or more years old with stiff jigs and numerous go/no go gauges.

  • Nathaniel

    You actually can do that inclined bit in the same operation as long as you have software that can handle 3D cuts. It would simply keep the bit level and raise it while moving it horizontally, like an escalator. Depending on your settings, the finish might be interesting, though. I would use a flat end mill to rough it and a ball end mill to finish it.

  • Nathaniel

    I definitely learned something about 1930s/40s machining methods.


    You need so many differnet positions, because you do not use a CAM program.
    Ι used to make programs for molds and other pieces with a CAD – CAM, and this pieces are not very compicated. Perhaps only one catch for each side.
    Of course it will take more time…

  • Tyler

    For the “slide” part, why not just plug weld the fork to the plate? I can’t see really well from the video, but it looks like there is enough room to do this, and it should be strong enough.

  • Keith

    Hi Ian,

    You’re hitting on a very interesting problem, and one possible way around it with the equipment and budget which you have available to you.

    Clearly, with a small shop and small batches, it would be absolutely counter productive to produce gangs of form relieved, multi tooth cutters for all of those profiles – that only works for a big factory with line after line of horizontal mills, making thousands of each part a day, it would take you on your own, years just to set up the tooling…

    Another approach is to look at how the likes of model steam train and clock makers work in their home shops.

    Small horizontal mills are very expensive for what you get, big ones in good condition are often pretty cheap.

    For small batches, you can make fly cutters, or even better, a single lip cutter to the required form.

    Assuming you start with a bar (or even better a tube, broached for the key) of suitable carbon tool steel on the lathe, the form is cut eccentric to the centre line of the bore, the bar is rough slit to form the cutting edge, and heat treated before finish grinding on the face of the cutting edge.

    One, single lip, form relieved cutter!

    clearly it will need a proportionately lower feed rate than its multi tooth counterpart, and there is a single shock load per revolution of the mandrel, so a “flywheel” on the mandrel would help smooth things out a bit, but those old horizontal mills have plenty of metal built into them to absorb years of such use. They were also designed to be scraped back into alignment when they wore out. There are a couple of scraperhands post at “practical machinist” who do very reasonably priced work.

    For some of the fine work, and for removing the big meaty bits from the angled part, it could be worth your while sending it out for cutting with a wire EDM. Beauty of that is, you can wire EDM hard stuff as easily as annealed, and it only takes minimal amounts of cleaning up to size (a few thou).

    I’ve an old book on form tooling design, and a more recent magazine article I’ll scan and mail to you.

    Good luck with the course and the project.

  • Keith

    Another possibility for some of the radii, is to grind your own single lip cutters for the vertical mill, with the correct radius on the end

  • Keith

    For anyone who’s interested in late 19th / early 20th century gun making in the Springfield Armoury, Colvin and Viall, produced a book length description of the manufacture and quality control of the US 1903 rifle, suitable for other manufacturers to prepare bids to make the rifles, should Woodrow Wilson’s plans to get the US involved in WWi bear fruit for him.

    Wilson got his wish, American troops died in battle, and he was able to use the “crisis” to enlarge the Federal government into all areas of life.

    Two links to free scans of the book (I think some pages don’t show on one version, but I can’t remember which)

    What the book is not, is an effective guide to making small numbers of rifles in a small shop. Ironically, the methods shown were’nt particularly suited to mass production in that era either! it seems that Colvin made some helpful suggestions, which were taken by the armoury staff as belittling insults

    The skills tooling and setups for small shop use are much more the realm of toolmaking and the jobbing machinist, and his amateur counterpart, the model engineer. has some really good old texts on toolmaking and ordnance manufacture from WWi and before which can be adapted for use in a small shop

  • Mg-42

    Part of the solution is to redesign the part for cnc operation and to remove some of the unnecessary angles. Some of the solutions is to use the older horizontal in the shop. the rest will be dealt with water jetting and welding. The problem will be solved, not necessarily to everyone’s satisfaction, but to a product that will work.

  • wyop

    Two additions you need to consider adding to your shop:

    1. A fourth axis for your mill.
    2. A wire EDM machine.

  • It’s a pity that much of the knowledge — this tribal knowledge — is lost. I once worked for a machine tool company. In its archives were thousands and thousands of old setups, and because the company was located in a center of gun making and a major contributor to the machine-tool wave of WWII, many of those setups were on guns: BREN guns, Browning MG, BAR, etc.

    The company, Heald Machine Company, had been run halfway into the ground, and then was bought by Cincinnati Milacron, who ran it the rest of the way into the ground, then backed it up and slammed it back in repeatedly with stumbling, whipsaw management. They were convinced that the Japanese machine tool firms were beating us on price, and they were, but we were also getting creamed on quality, a direct result of Milacron’s waves of MBAs “maximizing” profits.

    I staved off several attempts to destroy the archives, once on the grounds that all those old pictures were a fire hazard. I finally left and went back on active duty for a year or two (after seeing the dynamic of a company circling the drain and throwing off all its heavyweights by centrifugal force). In the end the company closed, and I assume the archives finally succumbed to their fate.

    It’s depressing to think of how often this scenario must have played out along the river valleys of NY, MA and CT in the 1960-90 time frame. There is a small museum dedicated to preserving machine tool history, but when I left the industry I lost track of them.

  • It’s amazing how intricate these machining processes can be. Most people don’t know all of the work that goes into manufacturing what appears to be a simple part or product. Thank you for the demonstration.

  • Mg-42

    1. A fourth axis for your mill.
    2. A wire EDM machine

    You sound just like the people that work in my shop. It is on the list. My major problem is that I have so little knowledge on wire edm’s. Any information or insight on wire edm’s would be helpful.

  • Floy

    There is a nifty museum of precision machining in Windsor, Vt. Worth the trip!

  • Great video, and a very good explanation of a complex machining process. Keep up the good work and make more videos!

  • […] that framework is fair game! And those limitations are steadily diminishing. Even CNC machines, which may seem like magic at first, have limitations that haven’t been overcome at the cost- and space-conscious individual […]

  • Jason Kiger

    Lost wax investment casting. You can have those parts cast in 4140 and need to only drill holes and lightly finish or touch them up. It eliminates the cost in machining and are just as strong as a milled part. I build jet engines for a living and in doing so we use investment castings extensively.

    • Chuck

      Jason I agree with the lost wax casting concept. However, that being said I just don’t think there is enough interest out in the market place to warrant the cost. How many could I sell at a profit and what would be my initial cost to do it. Maybe with the lowering costs of a 3d printer that could be used as a pattern for the casting. It is something to look into when I have the extra capital and time. Thanks

  • Jason Kiger

    I understand and your right. The market for this setup would be very slim. Would you consider selling me a set of 3D printed parts ?

  • […] has a great video post at GunLab on how what looked like a simple project to make a simple add-on, one that has already been proven […]

Leave a Reply





You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>