The bottom of the Z axis screw has a washer that rubs against the bearing block, the top has another washer.  Lots of friction.

I had thought that the gas spring was way too big, but in fact, it was not.

I first made cuts and then did not quite realize that I was going to cut into the vise to make the hole for the bearing.  Either use a larger piece of aluminum (.75 is not enough) or use two pieces of sacrificial plate to start with.

Behind the screw is the U shaped slot that i cut to clear the bearings.  I used a flat mill, then use a ball mill to complete the job.  I ended up wobbling it a bit, making parallel passes down the slot.  The middle was made a tiny bit deeper.

The little blue box is the stepper motor tester.  Since it's unipolar only, and the motor is a 6 lead motor, I can use it to test this out.  Since a bipolar driver will be better, not at low end torque, but at higher speeds, I can see how the design works.  As is, the tester does 1 and 10 steps/second.  The power supply is giving me 12 volts for the tester, and the stepper is being run at 4.6 volts or so and 1.55 amps (if the meter manages to follow it) for two coils. 

So we're getting  170/200 rev/sec which is .855 RPS which is 51.3 RPM (way slow), and that's 31.3 RPM delivered to the shaft. 

That works out to 1.56 IPM. 

So that means that I'll have to get up to about 1700 PPS drive half step to get about 15 IPM, which is not too bad.

The 4-40 tap broke in the headstock mounting block.  I thought that was really special. 

So I moved the brass bar over a bit, and tapped for 6-32.

So far, it works, without much of a problem.  I will have to see how it works with better (or different) drivers. 

Things not to do: 

1) do not machine the shaft at all.  I think I will eventually replace that shaft.  I may or may not remake the bearing block, but it doesn't really need to be replaced now.  Main problem is that the pulley assembly and the shaft will wobble.  I may have bent the shaft a bit when I was working with it, or it may have just have been bent naturally.  I also had a loose fit on the top bearing, and that was within the bearing assembly itself, shaft to bearing hub.  This allowed slop. 

2) It's possible to move the strut mounting up a bit, that will work.  Another option is to move it down a trifle.  The main problem is that one of the screws is in the way of either the Z axis handle or the drive.

3) Probably better to make the belt drive lower and closer to the motor body and to the bearing block. 

Plusses:

1) I like the way the sensors are mounted.

2) the belt drive (All Electronics part number CGB-33).  The gears need to be mounted differently, but are useful as is.  The adaptors are relatively simple lathe work.  The ratio is 41 teeth on the large gear, and 25 teeth on the smaller one. 

3) I had enough sensors for the entire mill (I think).  All Electronics does not have any more of them.  They can be replaced by something generic.

I did a few more things to clean up the implementation, but not the design.  The first thing I did was to order another z axis screw from Little Machine Shop.  Since this did not have the unthreaded part narrowed down, it would fit the bearings much better.  However, I found at least one  thing that had to be done.  Because the shaft had been cut down in the original design, the upper bearing was off center, and could be assembled that way.  This was not a Good Thing ™. 

I have a centering indicator, so I put the bearing block back in the vise.  I added the little side plates as best as I could, and then  left the bottom bearing in.  I indicated from that bottom bearing's inner race the diameter.  That centered the shaft to the bottom bearing, the inside of which had to clear the 1/2 inch hole that I had bored.  This way, the concentricity of the half inch hole didn't matter as long as it cleared the shaft.  With the boring head set to the proper outer diameter for the bearing, I was able to make a pass down the block, and clean up the hole.  It was probably about .010 or so off. 

An 8 mm rod would run down the two bearings without a problem, so the two bearings were concentric.

I found that the washers that I had used were suitable for 1/4 inch stuff, not 8 mm stuff (the shaft diameter).  Adding a few 8 mm ID washers made up for the (slight, if at all) manufacturing tolerance, and the small amount that I had probably removed from the inner thickness of the bearing block.  Note to self: put locktite on the outer race of the bearing to help hold it in, but probably not needed.

With this addition, the bearing block was reattached, and the very slightly modified leadscrew was put in.  You'd think I hadn't learned anything, but I did modify the leadscrew a bit.  What I did was to drill and tap the end for a 1/4 - 20 thread, and put in a cap head screw.  Since the screw head diameter was bigger than the thread itself, I turned it down.  This is to give me a way of grabbing onto the end of the screw without messing up the threads.  Locktite can be used if needed.  When you assemble the top part, it screws down on the bearings and preloads them.  How well that will work is another matter.  However, they are cheap bearings.  If I absolutely have to, I can put in thrust bearings and either remake the block or just add spacers as needed.

Overall impression is that there's much less wobble, and the screw did not move from side to side.  It seems to be easier (by hand feel, nothing else) to turn the Z axis screw.

Overall last comment so far:  Do Not Turn Down The Lead Screw.  Second comment, you really have to watch the concentricity in the two bearings.  I'll remember that on the X and Y axis.