Rockwell 21-100 Spindle Rebuild

[machPete99]

Re-posting after server crash issue...
I recently acquired one of these vertical mills, and will be rebuilding and refinishing it as time permits. Most of it is
still in pieces, but I managed to just complete the spindle rebuild. I leveraged some ideas posted on the Yahoo Rockwell
Milling Machine group, which includes some improvements over the stock setup.


I am using a duplex set of Nachi Super Precision 7008CYDU/GL P4 angular contact bearings for the lower spindle bearings,
from ebay. These are provided in matched sets and are to be installed "back-to-back" ("DB"). The races had some markings
on them (the high spots?) which I positioned on opposing sides.

The original fit on the spindle was a bit sloppy, with about 0.001" clearance (for both old and new bearings). I used some
0.0005" stainless shim stock strips to tighten this up to a light press fit, and make everything concentric. I found that
you need to shim each bearing separately, and cut off the excess shim before it fully bottoms out with a utility knife, as
the bearings tend to pull the strips down with them. 1.5" PVC drain pipe makes a good mandrel for installing these. The
bearings were lubricated with Kluber IsoFlex NBU 15 spindle grease.

The bearing retainer (#90) was modified to accept a SKF 17897 oil seal (or equivalent, as used here). This seal is used in
automotive timing cover applications so should be up to the task. This requires turning a pocket about 2.283" diameter by
about 0.3" deep. I used a carbide boring bit as the material was fairly tough (half hardened?).

The upper spindle bearing (#85) was replaced with a new MRC R16ZZ bearing, which has rubber seals. I popped the seals and
repacked it with the Kluber grease. The original fit in the quill for this bearing was a press fit. I used some emery
cloth on the lathe to open this up to a barely sliding fit, so the wavy washer is able to do its job. The press fit on the
spindle shaft was kept, and 3/4" copper pipe used to drive it home.

After assembly, I measured the TIR at the R8 taper and found it to be around 0.0005". The spindle turns slightly firmly,
with no discernible play. Everything seems to be a good fit.

 

[machPete99]

See the picture for my bearing preloading jig, used for setting up the 6007 spindle drive pulley bearings.
This is used to take ordinary deep groove bearings and add some preload so there is no play when assembled.
The original bearings had a similar setup, but they custom gound the inner races a bit narrower to create the preload.

The (2) step bushings fit inside bearings and register on inner races, one being longer so it picks up both bearings.
I put (4) small (.020" diameter) solder pieces between outer races and clamped the bushings and bearings together, crushing the solder.
Bearings are consistently stacked with part number markings on outside.
Once unclamped you measure the solder, and that is about what you want to use as a shim between outer bearing races.
In my case it measured just under .013" and I used .012" brass shim stock, cut to fit between outer races, which seemed to work good.
When assembled with the shim there is a slight drag, with no discernable play or rocking.

The new bearing stackup is somewhat thicker than the original bearings (by about .015") so I had to machine the bearing pocket a bit deeper,
so the snap ring could be installed.

 

[machPete99]

(duplicate post deleted)

 

[machPete99]

Finally got the head put together, with new bearings in quill and drive pulley.
Its also been fully stripped down and repainted, need to tackle the upright and base next...

 

[extropic]

Nice, thorough write up so far.
I'm wondering where you got the idea to use the strips of shim stock on the lower spindle bearings as shown in OP pictures?
I'm curious if you've read about or seen it used before.
My concern is that the method results in the bearings running with about 2/3 of the inner race unsupported. If you read the text on page 54 of the manual linked, particularly paragraph 2, you should understand that a shaft with four .0005" lobes is well outside the recommended shaft roundness of .0001".
I'm sure that a number of options exist to improve the bearing/shaft fit, depending on you interest and available resources.

Timken Super Precision Bearings (MPB)

 

[machPete99]

The bearing shimming is my own idea, probably not standard/textbook practice, and was a compromise between providing *some* support and concentrically versus having .001 or so clearance with the potential for things to run off center and potentially bang around. Some form of split sleeve with full support would probably be a better arrangement but much more work to machine, and I really did not want to cut into the spindle. The spindle locknut thread diameter is almost the same as the bearing journal, which complicates things.

I believe that I staggered the shims between the (2) bearings, if that makes you feel any better

When I replaced the motor bearings I noticed that someone (possibly the original manufacturer) had punched dimples in the bearing pockets as well as the armature shaft to improve the fit. Lots of hacks are available...

 

[machPete99]

Another repair process... the motor step pulley had about half of the smaller pulley flange busted off. I machined the rest of if off and made a replacement with steel, heavily chamfered for brazing. I ended up preheating in the oven at 450*, tig brazing with silicon bronze, and final machining on the lathe with carbide tooling (silicon bronze is tough stuff). I did hit an area with some carbon, and might have gotten it a bit too hot as well, but it came out pretty good. I got a bit of brazing and some shrinkage on the end of the bore, cleaned up most of it on the lathe followed by a .0005 undersized spiral reamer.

 

[machPete99]

An update on this slow moving project...

Oil grooves were cut into the saddle and knee and holes drilled for zerk oil fittings and distribution.
NOTE: drill longest hole first as you may not get as deep as you'd like.
I modified a grease gun to make a pressure oiler by brazing the end tube shut.

I replaced all bearings in the motor and the mill. 5202 (double row angular) bearings were used in place of standard 6202 for one end of the X and Y axis. This requires turning down another 0.193" of bearing surface on the leadscrews. I test fitted the X axis so the leadscrew is located by the 5202 bearing on one end and the 6202 on the other end is not preloaded axially. The X/Y dials have around 0.008" of slop!

Most of the mill is together. I still need to purchase and set up aVFD before I can actually use it.

 

[Ulma Doctor]

looking good! nice work.
there are some very inexpensive VFD's on ebay nowadays.
i have purchased 15 inexpensive units in the last couple years, all of them are still functioning in different types of equipment.
if you need links, i can provide them for you

 

[machPete99]

Ulma, Thanks for the offer, but I am probably going to hold out for another Hitachi WJ200, like the one I used on my lathe:
https://www.hobby-machinist.com/threads/clausing-5914-vfd.60815/

They are not cheap, but seem to be pretty robust, and have lots of low end torque due to the "sensorless vector" capabilities.
Since this mill has no back gear I want to be sure it has at least some low end grunt.
I'll probably set the mill up a little simpler, with just a forward/off/reverse paddle switch and a speed control.