Seig X2 minimill CNC conversion, Part 1: X/Y ballscrews

brought to you by:  Dave Williams
This page: www.bacomatic.org/~dw/tool/x2/xy/xy.htm
Main page: http://www.bacomatic.org/~dw/index.htm
Last Updated: 18 Sep 2008

Author: Dave Williams; dlwilliams=aristotle=net

Though CNCFusion's web site talks about installing the kit on an assembled mill, it's much easier to take it apart. In fact, you'll probably wind up taking it all apart several times. Fortunately it's easy to disassemble.

Mine was also full of grit and cast-iron swarf, which was even on the gears inside the head.

quandrant2005's videos on YouTube look as if all he had to do was tilt his machine back and install the CNCFusion kit. Mine wasn't nearly as simple. There appears to be considerable variation in the castings, and I'm happy that CNCfusion decided to go for maximum travel with a little fiddling if needed, but it was still a surprise when nothing seemed to fit.

This is more or less what I did for my installation:

1. remove electricals

(get stuff out of the way)

  • four socket head screws hold motor mount plate to main head casting

  • four Phillips screws hold the control box lid; four more hold the box to the casting

  • four Phillips screws hold the motor driver lid; three hold the box to the column

  • two Phillips head screws hold grounding wires from the driver box to the column

  • set all the bits off to the side

    2. remove column

    (you need to get under the base, etc.; it's easier to do if it's apart, you have to disassemble the table anyway, and it's easier to clean it this way)

  • it will help if the base is still bolted to the shipping platform

  • run the head about halfway down the column

  • loosen the big nut that holds the column to the base. Keep one hand on the column so it doesn't fall over. Remove the nut and thick washer.

  • slide the column away from the base, off the long stud. It is heavy and awkward to hold on to, and covered in grease. If the base isn't anchored to something it may want to slide around

  • lay the column on its back, taking care not to bend the feed levers

    3. disassemble base

  • remove X handwheel; it is double-nutted

  • loosen gib screws one turn

  • remove the left end cap from the table

  • loosen the set screws that hold the X-axis bracket

  • loosen the set screw that holds the leadscrew nut to the saddle

  • slide the table off to the right

  • the gib will fall out. Note: my gibs had nasty burrs where the screws rode. Though the gibs are some kind of steel, they're dead soft, so a file or sandpaper will work fine. If you use power tools, be cautious.

  • run the saddle to the back

  • remove the two screws from the front of the base, that hold the Y-axis bracket

  • loosen the single Allen screw that holds the Y-axis nut to the saddle

  • slide the leadscrew out

    4: clean

  • I used car engine degreaser and the garden hose, then blew the pieces with compressed air. A light spray of WD-40 to delay rust.

    5: clearance base casting for Y-axis leadscrew

  • get the Y-axis (short) leadscrew assembly out of its wrapper

  • remove the Lovejoy coupler

  • remove the spanner nut. Mine came from CNCFusion run down just snug; I was able to get it off with my fingers

  • slide the motor mount off. The ball bearings are loose and will fall out if you're not careful. Do not remove the ballnut.

  • slide the Y-axis leadscrew in from the bottom. The threaded end sticks through the front of the base casting.

  • put the back bearing in the motor mount and attach it to the base with the screws provided. Snug the screws down; don't over tighten.

  • the ballscrew is made as long as possible for maximum Y travel, but there are variations in the coring of the base casting. With some castings, the screw will bottom out before the motor mount is tight. My arbitrary decision was that .005" to .050" would be enough clearance. I wound up with .100" due to enthusiastic grinding.

  • the left side of the ball nut has the recirculating tube for the balls. On my mill, the ball nut scraped the rough side of the clearance slot, so I had to grind some there, too.

  • if you don't have enough clearance, file or grind the casting. Then you get to clean and oil it again.

  • if you have access to another mill you could save a lot of grinding

  • don't get too crazy clearancing the back of the base casting. More then 1/8" or so clearance and the ballscrew may be able to run off the last thread and drop balls out. If you want to be precise, check your screw and nut and see exactly how far the nut may go before you can lose your balls. Someday you might need that extra .100" or so of Y travel

    6. Y-axis leadscrew nut

  • make sure the threads on the end of the screw are clean, and the threads in the special spanner nut are clean.

  • put a drop of oil on the threads

  • CNCfusion's forum says you should be able to hold the ballscrew firmly in one hand and use pliers to run the nut down snugly against the ball bearing. This wasn't even in the same universe as the screw and nut I got. Using a piece of old leather belt to pad the jaws, I tried holding the screw with 10" pliers and still couldn't get enough of a grip to run the threads past the locking ring in the nut. I was leery of using Vise Grips, so I picked away at the plastic locking ring with a penknife until I could run the nut on. I figured I could use a drop of Loctite later if needed. You might try padding the screw with leather or wood and holding it in a bench vise to run the nut on and off a few times to loosen it up. My nut has plier marks all over it now. Surely there's a spanner wrench to fit the thing, somewhere...

    7. install Y-axis leadscrew

  • hold the short gib into the saddle casting and slide it onto the base. the gib screws go to the front on top and the right on bottom.

  • slide the saddle all the way to the back of the base

  • install the screw and motor mount. You may have to fiddle a bit to get the lug on the ballnut into the hole in the saddle.

  • snug the motor mount screws to the base

  • slide the outer ball bearing into outside of the motor mount

  • install the spanner nut. Those look like ordinary ball bearings, so they're not intended to take a whole bunch of side load. That's a 32-pitch thread, so "just snug" should do. If it loosens up under load you can always snug it tighter; if you ding a bearing race it will have to be replaced.

  • CNCfusion's web site says to run the saddle all the way forward and tighten the ballnut retaining screw. "All the way forward" means "as far as you can go and still get the Allen wrench between the saddle and the motor bracket." The purpose of running the saddle forward is to try to align the bearings and the ballnut. Run the saddle forward, loosen and re-snug the motor bracket screws, then loosen the ballnut bracket screw in the saddle and snug that. Now push the saddle all the way forward and back. If it gets stiffer at one end, you'll have to twiddle the ballnut bracket position until the saddle moves freely throughout its travel.

  • assemble the Lovejoy bits onto the ballnut and slide the motor shaft through the bracket into the Lovejoy coupler. Snug set screws, making sure the motor side screw is on the flat in the shaft.

  • there was .060 or so negative axial clearance in my Lovejoy couplers. That is, the motors would fail to touch the brackets by that amount, and tightening the mounting screws put a lot of load across the stiff plastic coupler crosses. I didn't like the idea of that much end load on the motors. I emailed CNCfusion and they said it wouldn't hurt anything. I still didn't like that much preload, so I used a 1/4" Forstner bit to put some simples in the crosses. Everything slid together nicely then


  • a ball-end metric Allen wrench would be a big help when installing the motors, or soldering or brazing the provided wrench to a piece of tubing so you can turn the wrench 360 degrees instead of 90 degrees at a time. I found a long ball-end wrench set at the local hardware store for $10. It was worth at least that much not to have so much hassle with the screws. You will probably run those screws in and out *many* times while fitting the parts to the mill.

  • it's easier to see what you're doing if you turn the mill base upside down

  • when wrestling the mill around the bench, try not to grab the Y-axis motor mount no matter how convenient it looks as a handle

    8. X-axis leadscrew

  • drop the X-axis leadscrew into the saddle. The threaded end goes to the right.


  • set the gib in the front of the dovetail. Make sure the dimples line up with the gib screws

  • slide the table on from the left to right

  • see if the ballnut contacts the underside of the table at any point. if so, the table will have to be clearanced. Mine would grate over rough bits and then jam about 2/3 of the way across, even without the screw being mounted in its end blocks.


  • the back and top of the ballnut contacted the table, but the most interference was the ball return tube on the front of the nut. I removed +1/8" of the casting to get room for the nut.

    It took half a dozen tries to get things to move freely. I was concerned about warping the table if I removed too much metal. Black spray paint was a convenient way to see where the nut was hitting. Since the ballnut is offset, the entire width of the table doesn't need to be ground to clear it.

    I had to take some off the middle to clear the ballscrew as well.


  • the C-shaped bit with X-/X+ goes on the right side, spacing the original thrust washer and bracket out

  • the rectangular motor mount block goes on the left

  • screw the Lovejoy coupler to the X motor, assemble the cross and inner joint, and feed them through the hole in the motor mount. The motor-side coupler must be tightened before mounting the motor unless you drill a hole in the mounting bracket to allow access.


  • tighten screws, making sure motor shaft flat is under the screw. The ballscrew doesn't have a flat. I recommend filing or milling a flat - that part of my screw was fairly soft, and the setscrew raised a large burr where it contacted the shaft. This made removing the coupler very difficult; I had to pad everything, put it in the big vise, and beat it out with a brass drift since I didn't have a puller small enough to do the job. I probably tightened the set screw too much, but a flat would still keep you from marring the shaft.

  • if you try to tighten the motor with negative clearance in the Lovejoy coupler, it can jam the ballscrew part of the coupler onto the screw too tightly to be removed easily. Remember, you can't just beat it off without damaging the ballnut.

  • there is very little clearance between the thrust bearing spacer (the spacer on the right side of the table) and the ballnut. On my particular machine, lifting the spacer up with my finger while tightening the through-bolts was enough to get by


  • unfortunately, on the left side of the table the Lovejoy coupler kit the saddle casting before the table moved all the way to the right. The saddle has to be relieved to accomodate the length of the coupler. I ran the table to the right until the coupler touched the saddle, ran a felt tip pen around the coupler where it touched the table, and disassembled the mill yet again.


  • Here we go! The table runs all the way to X+ now.
  • I used a carbide burr and abrasive rolls in a 1/4" die grinder and took +1/4" out.
  • Later, I saw some pictures of another CNCFusion conversion on the web, and it had a smaller diameter yoke on the X screw - the same one that was on my Y screw. If I had looked closely, I could have swapped the yokes and everything would have cleared. However, I now have room for helical flex couplers if I want to upgrade.


  • Finished installation from underneath.


    The X travel still seemed a bit rough in places. When I disassembled the machine to do the limit switches, I noticed some screw marks in the Y ballnut bracket. It was rubbing the X ballscrew.
    Some tape, some plastic bags, and a trip to the disc sander, and that problem is fixed easily enough...
    The CNCFusion Z-axis parts bolted right on without a hitch. Unfortunately, the ballnut and screw blocked access to the gib screws on the head casting. After looking at things for a while I decided not to use the CNCFusion bits as delivered.