584,837 active members*
5,158 visitors online*
Register for free
Login
IndustryArena Forum > MetalWorking Machines > Benchtop Machines > 7 Axis Swiss Mill-Turn (DIY)
Page 1 of 7 123
Results 1 to 20 of 142

Hybrid View

  1. #1
    Join Date
    Jun 2014
    Posts
    25

    Lightbulb 7 Axis Swiss Mill-Turn (DIY)

    Hey everyone! I haven't posted here in 4 years, but I have something that I think you'll find interesting. I built a Swiss type CNC mill-turn center. This was a DIY project, 2k budget. What you see here is actually the 2nd prototype. The reason I'm posting this now is because I'm planning a crowdfunding campaign soon, I and need some feedback ahead of time. To summarize, I got into building this because I was manufacturing 3d printer hotends and the machining costs were kind of high. So I designed and machined the machine you see below.

    It's sized for a 6 inch chuck on the main turning spindle, and the swiveling milling head is a 30 taper. I built this thing as rigid as possible. The milling head spindle bearings are larger than on a bridgeport. The other bearings have a 100mm diameter OD. The machine base is a solid 4x4 square bar, just under 4 feet long. All linear axes are large boxways, I know the people around here seem to love linear guides, but I prefer my machines to have that old school stoutness that you can't get with fancy drawer slides. And no ballscrews either. I've extensively tested my linear axis motion; positioning repeatability is a few ten-thousandths, backlash is under 0.0006". Checked with a dial indicator. I wouldn't have been able to make this machine with $2,000 if I had to spend $2,000 on the rails and screws alone!

    The whole thing is controlled by a 3d printer control board. Saves a lot of money, and 3d printer tech has a much larger community that has overcome many of the glitches that still plague cheap CNC machine controls. This machine is compatible with NEMA 23, 24, and 34 steppers and servos, but I plan on using a lot of ODrive servos in the production version. The main turning and milling spindles already have them, they're closed loop systems that allow for rigid tapping and threading and stuff. Plus, they're approximately 10 times more accurate than steppers and ridiculously fast in comparison.



    I pity the foo who can't see these glorious GIFs. It may take a minute to load them if you have bad internet. There should be 6 of them directly below, and they should look just like a 30 FPS video if they're playing normally. The gifs are hosted on the Hackaday forums since they don't have restrictive file size limits over there.








    I don't think this forum allows post editing, so I'll have to fill in the details in subsequent posts. More pictures and videos to come. Ask any questions you can think of. Critique all you want, I'll try to respond to everything.

  2. #2
    Join Date
    Aug 2006
    Posts
    670

    Re: 7 Axis Swiss Mill-Turn (DIY)

    this...is..amazing...

    I'd love to see build pictures.

  3. #3
    Join Date
    Apr 2012
    Posts
    32

    Re: 7 Axis Swiss Mill-Turn (DIY)

    That's awesome! Do you have a build thread?

  4. #4
    Join Date
    Jun 2014
    Posts
    25

    Re: 7 Axis Swiss Mill-Turn (DIY)

    Here's the first bit of progress from last July. Let me know if you'd rather have the build thread pictures as GIFs or still images.


  5. #5
    Join Date
    Jul 2003
    Posts
    586

    Re: 7 Axis Swiss Mill-Turn (DIY)

    GIFS GIFS GIFS :banana:

  6. #6
    Join Date
    Jun 2014
    Posts
    25

    Re: 7 Axis Swiss Mill-Turn (DIY)

    Yes Mountaindew, the production version will have a hard anodized PTFE surface on all the boxways. Everything else will be color anodized, although I haven't finalized the color scheme yet. The plastic-on-aluminum boxways don't actually need oil or grease.

    Cmelo, the 3d printer controllers I've been using generally cost around or just under $200 and come integrated with 5 to 12 stepper motor drivers directly on the board. The controllers execute gcode and are easy to use with free software, firmware, and inexpensive touchscreen or control panels. I know mach3 is the standard on these forums but from my perspective, that stuff should have been retired in 2010.

    Skrubol, I was also concerned with that during the build process. I did a run in test a while ago and it turns out that even after spinning the main spindle at 2,000 RPM for an hour from a cold start, there is no noticeable temperature change in either the spindle or the spindle housing. It's a huge mass of aluminum with face contact everywhere, so heat is distributed quickly and evenly. I think the main problem in the future is if people let this machine sit in a garage or shed that fluctuates 30 degrees daily. That would mess with the tool offsets a bit.

    Hackish, I'm actually not making my own electronics for this project. I'm using off the shelf 3d printer controllers that are proven and working for thousands of people as the main control board. The power supply is a big 48v Meanwell unit, which has great protection features (comes with a 5 year warranty).
    As far as the costs of building this machine, I have a good idea of what it's going to cost. The cost of the aluminum and steel was around $1,200 for this prototype, but that was at retail with a fairly high markup. Buying tons of it at a time would cut the cost roughly in half. Your $10,000 estimate is probably close to what it would be if I outsourced this to job shops. I'll be making these myself with a single large VMC in a small rented industrial space. The overhead cost of this will be several thousand a month whether I make 1 machine or 100 machines in that month. The tooling cost of milling aluminum is inexpensive, energy cost is approximately 30 cents per 20 pounds of chips cut.

    BTW, I should mention that there will be a few variants of this machine at launch to span the price range. Since they all use common parts, there isn't much added complexity except when it's boxed and shipped.

  7. #7
    Join Date
    Aug 2015
    Posts
    142

    Re: 7 Axis Swiss Mill-Turn (DIY)

    I like ur build , is clean , and working .
    All parts are made from aluminum ?
    Can u process brass , or steel with ur build?
    What do u use for linear movement if u dont use ballscrews ?

  8. #8
    Join Date
    Apr 2012
    Posts
    32

    Re: 7 Axis Swiss Mill-Turn (DIY)

    Quote Originally Posted by Andreiir View Post
    I like ur build , is clean , and working .
    All parts are made from aluminum ?
    Can u process brass , or steel with ur build?
    What do u use for linear movement if u dont use ballscrews ?
    Ya curious as well if the box ways and the slides are all aluminum. If so I'd think they would gall and wear quite fast. Then also how you're getting such accuracy from regular acme screws if you're not using ball screws.

  9. #9
    Join Date
    Apr 2005
    Posts
    1268

    Re: 7 Axis Swiss Mill-Turn (DIY)

    Wow!
    That thing has more moves that a fat belly dancer! :cheers:
    What software are you using to drive all the axis? I can't imagine trying to keep up with all the available cutting planes!
    Great Machine!
    Very impressive!
    Thanks for posting.
    Bill
    billyjack
    Helicopter def. = Bunch of spare parts flying in close formation! USAF 1974 ;>)

  10. #10
    Join Date
    Jun 2014
    Posts
    38

    Re: 7 Axis Swiss Mill-Turn (DIY)

    GIFs GIFs GIFs:banana:

  11. #11
    Join Date
    Oct 2017
    Posts
    49
    Turning GIFs pls!

  12. #12
    Join Date
    Oct 2017
    Posts
    49

    Re: 7 Axis Swiss Mill-Turn (DIY)

    Also is that <2k for the just the aluminum or for the entire thing? Because if you managed to make all of that for under 2k i want to know where you're shopping lol.

  13. #13
    Join Date
    Jun 2014
    Posts
    25

    Re: 7 Axis Swiss Mill-Turn (DIY)

    To answer your questions:

    1) All parts are made from aluminum ?
    Nope. The spindles, shafts, and hardware parts are all steel. The big structural parts are solid aluminum.

    2) Can u process brass , or steel with ur build?
    Yes, it cuts anything softer than tungsten carbide. I have a steel and a brass part in the video, as well as aluminum and plastic.

    3) What do u use for linear movement if u dont use ballscrews ?
    Leadscrews with roller thrust bearings and long, zero-clearance, double adjustable, lockable anti-backlash nuts. More on these later.

    4) What software are you using to drive all the axis?
    Repetier host for now. I'll make a custom GUI for the production version of the machine.

    5) Also is that <2k for the just the aluminum or for the entire thing?
    Entire thing. The 6061 I bought was 3 to 4 dollars per lb, and comprised the majority of the build cost. In production this will be $2 -$2.50 per lb for bulk purchases.

    6) Ya curious as well if the box ways and the slides are all aluminum. If so I'd think they would gall and wear quite fast.
    The boxways are aluminum but there are acetal wear pads under each moving assembly. No metal on metal contact to gall. The production version will be hard anodized with PTFE.


    A few more GIFs;



    This last one is to prove my point about the linear repeatability of this machine. You can see from the needle that movement is smooth to the micron, and doesn't have much backlash. It's about as good as industrial machines in this aspect.

  14. #14
    Join Date
    Jun 2014
    Posts
    25

    Re: 7 Axis Swiss Mill-Turn (DIY)

    Arrrrrg I just typed out a 10 paragraph post addressing all of your questions and comments in detail, only to have it deleted because this website glitched when I was uploading a photo of stuff. It just makes the screen greyed out indefinitely when I try to link a photo.

  15. #15
    Join Date
    Oct 2017
    Posts
    49

    Re: 7 Axis Swiss Mill-Turn (DIY)

    yup. type in word and copy pasta

  16. #16
    Join Date
    Nov 2007
    Posts
    2151

    Re: 7 Axis Swiss Mill-Turn (DIY)

    Wow, dig those gifs also
    Looks like all it needs are some way covers, ptfe coating and lube system.

  17. #17
    Join Date
    Apr 2012
    Posts
    32

    Re: 7 Axis Swiss Mill-Turn (DIY)

    Definitely following this thread and look forward to learn a bit more about how it's all gone together. I hope the aluminum/plastic ways hold up just because I think that's a cool idea I haven't seen executed before. For medium duty it could very well hold up fine? I have no experience with building 3D printers so I think using parts to drive the motors from that community makes total sense from an economy of scale standpoint. Not sure how programming for something like that works to control a 3+ axis machine when 3D printers are typically 3 axis and don't have a spindle ect. You seem to have it all figured out but my experience has been more with conversions using Mach 3 so I am probably missing something simple.

  18. #18
    Join Date
    Jun 2014
    Posts
    25

    Re: 7 Axis Swiss Mill-Turn (DIY)

    jgwentworth and Hackish,

    I've tested several 3d printer boards; Smoothieboard with 5 steppers, Azteeg x3 pro with 8 steppers, and the production version machine will most likely use the Duet board with up to 12 steppers.
    3d printer boards also have other inputs and outputs, like lots of thermistors for temperature, heater outputs than can be used for air cylinder and solenoids, display screens, sd cards, lots of limit switches ect.

    Embraced and Ianagos,

    The spindles on this mill turn center use P5 grade (ABEC 5 equivalent) 15 degree angular contacts with indicators on the outer races.
    I picked bearing sizes such that they are also compatible with deep groove bearings and even tapered roller bearings.
    This is for forward compatibility, and it gives people the option of replacing bearings cheaply if something bad happens to the originals (at the cost of a couple tenths of runout).
    The bearings fit with a very light press fit on the shafts, no play allowed. The housings are blocks of solid 6061 aluminum with close tolerances. This eliminates the slight loss of rigidity from bolting a spindle cartridge into place.

    Neither spindle is hardened or ground on the prototype. The big headstock spindle started as a DOM tube with thick walls and a 2 inch bore. There is no need for this thing to be hardened.
    The milling spindle started as a solid 2.25" bar of 4140 pre-hard. Since both the taper and the bearing shaft was turned in one chucking, the concentricity of the assembled thing is excellent.
    On the production version, the spindles will be made in a CNC lathe with a large collet nose. If the spindles can be turned with tight enough tolerances, grinding won't be needed.
    For heat treating, I'm still uncertain. They could be roughed, through hardened, then ground. Or they could be ferritic nitrocarburized for a hard, rustproof skin without warping. Or they could even be PVD/CVD coated.

    BTW, the cost of the milling spindle head was about 200 dollars for the prototype. That's for the spindle headstock, spindle tube, P5 bearings, and BLDC motor combined.
    Runout at the tool tip is under 0.001", I can't measure any more accurately with the limited tools I have right now. I expect the production machine will range from 0.0002 to 0.0006" at the tool tip .

    Onto the "drawbar" and ATC mechanism. The drawbar on the prototype is literally just an M12 socket head cap screw. The bottom of the socket head sits on a roller thrust bearing sandwich.
    The production version will have an ultra low friction coating on the threads. With BT30 tool holders, there is no need for pull studs. The drawbar threads directly into the tool holder.
    5 foot-pounds on the socket head generates approximately 900 pounds of clamping force between the spindle taper and the tool holder, which is much more than adequate.
    There is no risk of the tool holder coming out during milling operations, and no pull stud to fatigue.



    The cross section below is a bit outdated, and it doesn't show the mechanism for stopping the drawbar relative to the spindle, but this should give you a general idea of what the inside looks like:




    Notice in the cross section photo that there is a spring and stop mechanism around the draw bar. This allows the drawbar to be pushed up by the tool holder before it engages with the threads.
    It's kind of hard to explain in words here, but the spindle ATC works by threading and unthreading tool holders into the spindle taper using the milling motor itself.
    This is inherently slower than an air cylinder and bellevile stack, but it's ultra compact and cheap in comparison. It also allows for 10 second manual tool changes without power or air.


    I know I'm forgetting something that I wrote in my first long response, but it got deleted and I can't remember it now.....

    Questions? Comments? Criticism? Let me know!
    Attached Thumbnails Attached Thumbnails milling spindle cross section prototype2.jpg  

  19. #19
    Join Date
    Feb 2018
    Posts
    92

    Re: 7 Axis Swiss Mill-Turn (DIY)

    I think it's very interesting and the machine looks very sexy. I remain unconvinced about the abilities of the machine on anything but the softest of metal. I'm not in any way trying to bust your balls, and I am not a professional machinist nor engineer. I ran a race car fab shop for a number of years, and based on my experience working with different types of metal I have difficulty picturing this thing doing anything that requires rigidity. I would love to see a video taking a reasonable cut with a boring bar on some steel.

  20. #20
    Join Date
    Feb 2015
    Posts
    31

    Re: 7 Axis Swiss Mill-Turn (DIY)

    I certainly don't want to take away from Generic Default's impressive work here, but...

    Quote Originally Posted by hackish View Post
    I think it's very interesting and the machine looks very sexy. I remain unconvinced about the abilities of the machine on anything but the softest of metal. I'm not in any way trying to bust your balls, and I am not a professional machinist nor engineer. I ran a race car fab shop for a number of years, and based on my experience working with different types of metal I have difficulty picturing this thing doing anything that requires rigidity. I would love to see a video taking a reasonable cut with a boring bar on some steel.
    I absolutely agree with this, and I'm an engineer who's done machine design in a few industries.

    The thing I'd be most concerned about is long term reliability, but the lack of hard numbers on capacity reinforces my suspicion that this isn't going to be taking serious cuts in anything. Saying that the spindle has larger bearings than a Bridgeport is the kind of thing that makes me question the amount of engineering that went into the design in the first place. Larger bearings can't compensate for the overall lack of rigidity in the machine, and aluminum is a poor choice for a small machine where every pound of mass is a plus.

    At the right price point (under $5k), I'd probably be willing to try one out once people had trialed it through its teething period. I definitely have aluminum parts that could benefit from the mill-turn setup.

Page 1 of 7 123

Similar Threads

  1. Replies: 0
    Last Post: 04-01-2016, 07:55 PM
  2. Replies: 0
    Last Post: 04-01-2016, 07:52 PM
  3. Mill/Turn Y axis Instead of C
    By FlorinMBV in forum FeatureCAM CAD/CAM
    Replies: 2
    Last Post: 03-29-2016, 07:07 AM
  4. Replies: 15
    Last Post: 04-14-2011, 05:33 PM
  5. Swiss turn material
    By scook179351 in forum CNC Swiss Screw Machines
    Replies: 8
    Last Post: 05-11-2008, 06:37 AM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •