X-axis Part II
5 Bears Home Homebrew CNC bench mill

In Part I, we finished machining the cast-iron table and are beginning to mount hardware more or less permanently. Need I say it again? alignment... accuracy... parallelism... etc. Errors in hardware installation can become magnified on the extremes of the table, so we want this stuff to be done right the first time. The finest motion hardware in the world won't help you if not correctly installed. For example, if the rails are not parallel to the working side of the table, machining a flat when the mill is complete will produce a taper cut, i.e. the stock will be thinner at one end than the other.

The completed end plate is secured to the table with the 4 ea 10mm SHCS. The screws are tightened slightly; then, the table is inverted on a granite plate, and everything lined up as much as possible, before final tightening. The granite surface plate ensures the end plate does not project above the table top when the work table is turned over once more to its final operating position.
The ballscrew is installed into the simple support end, and the driven-end bearing block is temporarily secured to the end plate with 4 ea. 1/4" SHCS. Note the brass wavy washers, which provide some pressure of the bearing block to the plate, yet allows lateral movement of the block assembly.
At this point, the end plate is quite secure; the rails and blocks are installed, aligned, and torqued. The simple support end (shown in the picture) for the ballscrew is secured tightly. The driven end of the ballscrew (off picture, to the left) is held only by the spring of the wavy washers. The ballnut is similarly held to its carrier, via washer pressure only, and may be shfted slightly.

It is time to secure the saddle plate. Note the planar accuracy cuts for the 4 blocks, and the central cut for the ballnut carrier. All four THK block cuts are in the same plane, while the depth of the ballnut carrier cut, while parallel to the block cuts, is of a calculated depth to allow tightening of the carrier cap screws within the limits of the through holes. The plate is dropped down on the blocks, carefully registered and aligned, and the 16 ea. 4mm stainless SHCS are sequentially tightened.

When the smoke cleared, the plate is secured to the blocks, and exactly 0.020" of clearance exists between the simple support end of the ballscrew, and the plate, as intended. CAD is a wonderful thing!

Note how thin the aluminum becomes at the top of the bearing retainer, as pictured... it is a tight fit between the iron table and the aluminum saddle plate.

The saddle plate, which is secured only to the way blocks, is moved by hand to a position as far over towards the simple support end as possible. In this picture, the driven end of the ballscrew is off-camera, towards 8 o'clock. This prepares the assembly for attachment of the ballnut carier to the plate.

Note that at this point, the simple end of the ballscrew, which has as its bearing carrier the clamp block, is firmly secured to the table. By mounting the nut carrier now, any vertical pressure on the ballscrew will move the driven end of the ballscrew, held only by the wavy washers.

Remember that the four screws securing the ballnut to the carrier are a bit loose? By alternately tightening and loosening these four, as well as the two heavy carrier-to-plate screws, we work the mounting of ballnut, to carrier, to plate, into a stress-free, accurate state relative to the ballscrew. All of this is done as close to the simple end of the ballscrew as possible.

Once the ballnut mount is tightened firmly, the sadlde plate is moved to the opposite end, the fixed end, where the servo mounts. At that end, the bearing block is free to move within the limits of the oversized bolt holes. Once the saddle plate arrives at the end, it will have physically moved the bearing block into the best position by itself. All that remains is to tighten the 4 bolts which retain the bearing block tot he end plate. Whew! Lots of fiddling, but its easier to do than describe.

To summarize the order of operations - the simple end of the ballscrew is secured to the iron table underside. With the nut next to the simple end, the ballnut is secured. The saddle plate is moved to the drive end, then the drive end bearing block is secured.

For each tightening operation with multiple screws, each SHCS is secured in an alternating pattern, gradually increasing the torque. Note the lower two SHCS have had their heads turned to a smaller OD. This was necessary to clear them with the two closest THK blocks. Doing so got me an additional 1/4" of travel. You need as much travel as you can get on a mill this size!

Done! A test with the manual crank handle showed smooth, flawless motion. Of course, I had to mount a motor and do an accuracy test! This was the first motion I had created via CNC in my own shop. Was it accurate? Find out!