Announcement

Collapse
No announcement yet.

Building a new Vertical Machining Center.

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Building a new Vertical Machining Center.

    Tired of missing opportunities for jobs due to not having a small, fast VMC available to take in the small parts and crank them out quickly. The machines in the shop are old, slow, but I was able to pay for them so they get work done. Time to expand.

    I can't outlay the 30-50K to purchase a good machine to do this with right now, and banks aren't loaning money at favorable terms in my area. I missed a few good ones this past summer, but I have layed in a good supply of parts and equipment over the last 10 years so why not build one now?

    I started modeling around the components on hand, digging through the scrap pile, and watching machinery sales and scrapyards for the missing components.

    A few weeks ago, I figured I finally had enough modeled to get started, and commenced work with the spindle mount and Z-axis column.

    Both are weldments of A36 plate, mostly waterjet scraps and sprues pulled from the scrap bins. I plan to build the whole machine this way, but some material was plasma cut...which was a mistake.

    All weldments are heat treated or will be heat treated at 630C for 1hr/25mm thickness, and allowed to cool in still air. It's not hot enough to anneal anything, but it is hot enough to remove the stresses and by removing it from the furnace after the heat and allowing it to cool, it does regain some of it's strength, similar to normalizing but not as much.

    Here's the spindle mounting block, which was welded up out of saw-cut bits of waterjet scrap:

    The spindle cartridge:
    Click image for larger version

Name:	shFmeOph.jpg
Views:	210
Size:	141.6 KB
ID:	7241990
    Getting started on the block. The mounting ring was turned and bored from a waterjet cut circle of 1.125" plate, and then a pair of plates were sawn and tack-welded together. The two plates were bored and faced, and a shallow counterbore cut to receive the prepped mounting ring.

    Click image for larger version

Name:	cUJtE2gh.jpg
Views:	183
Size:	155.5 KB
ID:	7241991

    The weld prep is just over half the width of the face, making sure that the mounting bolts for the spindle cartridge are going to be through the weld prep.

    Click image for larger version

Name:	jlyIuHxh.jpg
Views:	183
Size:	158.2 KB
ID:	7241992

    After the mounting ring was fully welded in place, the plates were set back up in the lathe, dialed in on the mounting ring face and bore, and the rough-bore was completed and a pair of counterbores were cut to receive a spindle tube.

    Click image for larger version

Name:	RV8QzSwh.jpg
Views:	184
Size:	177.1 KB
ID:	7241993

    After the main bore components were aligned and the mounting plate for the block to the Z-axis was partially welded in place, all internal welding was completed and the block was set up in the bridgeport to cut in a deep U-groove weld prep. Plenty of room to get a very deep penetrating weld.

    Click image for larger version

Name:	j7nBEPmh.jpg
Views:	181
Size:	131.0 KB
ID:	7241994

    Finished up the weld here, just showing off a bit.

    After that, side plates were welded in to complete the box, but the front was left open for access to the mounting bolts, and the drawbar operator release button.

    Click image for larger version

Name:	M4QMBMEh.jpg
Views:	184
Size:	195.4 KB
ID:	7241995

    Click image for larger version

Name:	jUC5LOnh.jpg
Views:	183
Size:	162.2 KB
ID:	7241996

    Fully welded and loaded into the heat-treat oven:

    Click image for larger version

Name:	IXd7mKmh.jpg
Views:	183
Size:	144.1 KB
ID:	7241997

    After heat treating, the scale and carbon was cleaned off and the part was set up in the mill. The stress-relieved weldment machined very nicely, and the machined faces have not moved at all during or in the weeks after machining.

    Click image for larger version

Name:	chzKsD9h.jpg
Views:	181
Size:	152.6 KB
ID:	7241998

    Click image for larger version

Name:	dB2iBoSh.jpg
Views:	182
Size:	189.3 KB
ID:	7241999

    Click image for larger version

Name:	dqtWpueh.jpg
Views:	183
Size:	134.8 KB
ID:	7242000

    Once the millwork was finished up, I set up the weldment on the boring table on the lathe and took some measurements. I need to make a new line-boring bar, to finish-machine this bore to 90mm+some clearance to match the spindle cartridge. I haven't decided if I want a snug slip fit, or a looser clearance. The tramming is handled in the backing plate, but spindle cooling is a concern. I am considering wrapping cooling lines around the spindle, as it's easy to do that now but not easy later.

    Last edited by Xnke; 05-04-21, 05:26 PM.

    #2
    On to the column.

    I started with some 25mm linear rails that I picked up in 2018, they've got some speckles but they're BLH25 rails, so easily replaced in the future. I have 660mm rails for the Z and Y axis, and 990mm rails for the X axis. Preloaded carriages to match.

    Pulled this 254mm x 203mm tubing for the column. It's 10mm wall thickness, and is good enough. More waterjet scrap gets added.

    Click image for larger version

Name:	C97twdAh.jpg
Views:	207
Size:	108.7 KB
ID:	7242003

    Click image for larger version

Name:	OyXapvch.jpg
Views:	184
Size:	194.9 KB
ID:	7242004

    Rail testing-no mounting or machining has been done yet.

    Click image for larger version

Name:	4eCtyBhh.jpg
Views:	181
Size:	226.9 KB
ID:	7242005

    After a few tweaks and some heat-warping to get it aligned, I stated welding the base to the column. The 1"-14tpi bolts are going to have to be changed out to socket-head cap screws...at a high price.

    Click image for larger version

Name:	RGiMZSWh.jpg
Views:	182
Size:	183.5 KB
ID:	7242006

    Just the root pass layed in so far. The column is set nice and square to the base, and the support braces are test-fit. A rigging ring will be welded to the top of the column for ease of handling.

    Click image for larger version

Name:	iEkGeXKh.jpg
Views:	184
Size:	152.9 KB
ID:	7242007

    Click image for larger version

Name:	Xj68b4ah.jpg
Views:	182
Size:	177.9 KB
ID:	7242008

    The column base has three 3/8-24 set screws in a triangular pattern, and there are three matching divots in the machine base plate. A trio of ball bearings will be trapped in the divots and the setscrew cups, and used to set the column square to the base, before it's epoxy grouted in place.

    Once the column weldment is finish-welded, and welding distortion is corrected, it'll go to the heat treater for stress relief, same as the smaller parts that I can do in my oven here.

    In the meantime, the battle that is getting a functional STMBL drive continues. The LV control boards are fine-but the HV control boards are giving me fits and so far, I've ruined two PCB's trying to get them going, and have had to strip and re-test multiple assemblies now to finally figure out what the problem is. Now that I have the problem defined, I stand a better chance of making the dang thing work.

    Comment


      #3
      Neat project. The CNC guy I used to work with built a bunch of his own VMCs, apart from one welded frame setup, he mostly converted old manual machines.
      Are you going to do something like an epoxy/granite fill of the column to help control ringing?
      I know there are guys epoxy/granite cast whole head assemblies for high speed spindle conversions of older machines - has always appealed to me as a technique.
      "Where can we get hold of a Vincent Black Shadow?" "Whats that?" "A fantastic bike," I said. "The new model is something like two thousand cubic inches, developing two hundred brake-horsepower at four thousand revolutions per minute on a magnesium frame with two styrofoam seats and a total curb weight of exactly two hundred pounds."

      Comment


        #4
        Yeah, the column will probably get partially filled after heat treatment, but before final machining.

        The advantages of epoxy-granite composite are only in damping, it's weaker, more flexible, and less rigid than the steel structure, so the steel column will get sandblasted inside and out after heat treatment and an internal form will make it so about a 40mm thick epoxy-granite layer will be bonded in. Maybe thinner-I need to update the solid models and try different fill ratios to get the resonant frequencies and damping coefficients where they need to be.

        It's entirely possible it won't need any fill, or that more than a certain amount of fill will be detrimental. Steel is so stiff compared to cast iron or epoxy-granite composites that the self-exciting frequencies are very high, and any filling will only lower those frequencies. If the self-exciting frequency isn't reached in normal operation, then filling the column can lower that frequency enough that the vibration of normal operation could excite the structure and cause chatter and broken tooling.

        If the machine is going to be made of epoxy-granite, the best way is an internal steel skeleton to position steel threaded inserts in the epoxy-granite matrix, and do a "Mineral Casting". Some extremely good, very expensive machine tools are built this way, and now some of the cheaper Chinese-built tools are also built this way. Syil, for example, builds epoxy-granite based machine tools.

        Comment


          #5
          Awesome work, why buy it when you can build it?!

          Comment


            #6
            wow this is next level
            www.holditflat.com

            Comment


              #7
              It's going to be a few months to a year to get it built, but that's just because I can't drop 2000$ on steel in a week for this. I am already doing that to feed the machines for my business and it doesn't leave much left over for me...and bills still gotta be paid.

              I am working on Youtube junk for this...I just really am not good at speaking to the camera, and when I watch the footage I just sound like a dumbass. (not that I'm *not* a dumbass, mind you...)

              Comment


                #8
                Originally posted by Xnke View Post
                when I watch the footage I just sound like a dumbass. (not that I'm *not* a dumbass, mind you...)
                That doesn't seem to stop at least half the youtubers out there...
                Imagination is more important than knowledge.

                Comment


                  #9
                  Would have thought you'd be $15k in the hole, plus time / labour once it's finished?
                  Definitely following though.

                  Comment


                    #10
                    Nah, you easily *could* spend 15 to 20K on building one of these, but you certainly don't have to, if you're willing to put in a few hours here and there. Yes, that's still a cost, but it's just like cars...

                    I've got all my costs recorded (because I can take them off my taxes) but I figure by the time I'm all in, it'll be around 5500$ plus time, and that's before things like tooling, tool holders, vises, etc. I already have the software for the other machine, and the controls are easy enough to build.

                    Click image for larger version

Name:	r1n3zsMh.jpg
Views:	91
Size:	160.4 KB
ID:	7243014

                    The above is a 2kW 3 phase servo drive, missing a few through-hole components. Total cost is about 60$, except for the smaller of the two microcontrollers on the bottom board...it was $1.32 each part last year, but this year they're 54$ each due to high demand and COVID shutting down ST's factories last year. It *does* take me 5 hours to hand-place all the components and solder them in place.

                    It's going to take some fiddling and fettling to get the servo controllers built and tuned up but it makes it significantly cheaper when I can buy literally any servos and plug them in, without having to dig up 1500$ worth of servo drivers to match each 500$ servo motor.

                    So far, the most expensive components (from a single-component cost perspective) have been the ball bearings. Not the linear rails, but the actual ball bearings used in the spindle and the bearings used on the ballscrews. (Yes, even my C1 class ballscrews were cheaper than the bearings that hold them in place!)

                    Comment


                      #11
                      Yeah they certainly aren't cheap at that size and spec, that's for sure!

                      Comment

                      Working...
                      X