Small Epoxy Granite Filled "Watchmakers Lathe" Design Question

[hman]

Couple thoughts ...

(1) As far as Taig using aluminum extrusions, I'd strongly surmise that it's because a (custom) extrusion is a net or near net shape. So they can use it easily. Yes, it's "ringy" (my non-sophiosticated hillbilly engineering terminology). So they fill it with the E/G.

(2) I've filled the column of my Harbor Freight 44991 mini-mill with E/G. Just for fun, I decided to add some "long distance" stiffness by including 1" drywall screws and some ~½" 4-40 screws in the mixture (along with glass beads and aquarium gravel). Didn't do any sophisticated testing of the final result, but it does seem to have helped with rigidity and vibration damping. I used West Systems epoxy with the slow hardener, in hopes that the long cure time would allow air bubbles to escape. When mixing it , I stirred slowly, trying to reduce introduction of bubbles. Once poured, I used a vibratory engraver (with an extension) to agitate the mix. A bunch of bubbles did float to the surface before the epoxy cured. And I was pleasantly surprised by how well the epoxy wetted out the filler materials.

 

[shermama96]

Hats off to you @shermama96 , I like your choice of career path . Your observations about the E/G fill I find very interesting because it reminded of when Warner & Swasey was experimenting with welded up bed frames and then filling them with a type of concrete . They proved that their lathe could produce a finish that could negate grinding because of the fact there was no vibration . I would love to see a picture of your lathe when it's done , Are you planing on a thru hole spindle ?

For the spindle I plan to use an ER-16 collet chuck that I’ll cylindically grind to the proper bearing fit.

The physics dept at my school has an ancient wire edm designed for burning out deep holes. Instead of annealing the through hardened collect chuck to drill its center out only to re-harden it, I’ll burn out the center of the collect chuck to get my through spindle.

 

[JohnG]

I've seen similar construction used in industry for large, high precision machine tools--8" x 8" x 1/2" wall steel tube 20' long for a base with laser interferometers to guide the cutting head. The machine was going to grind and polish steel rolls used to make super calendared paper. Here's what I remember.

The steel was all normalized before fabrication. The top block which was machined to be the precision way was also normalized steel and was continuously welded at both bottom corners to the base tube. The difference between fabricating a composite beam like this with intermittent fasteners and continuous adhesion is night and day. The machine screws you plan on using will do nothing to improve the bending moment of this beam but will provide some shock resistance which epoxy will lack. West System epoxy gets my vote for adhesion to steel, but this isn't my specialty. I always just had base fabrications welded.

I'm not sure about filling the core. You may already have looked at the construction of One Way lathes which seem similar to me. They rely on the rigidity of a large steel tube, and that simple formula worked well for the machine builders I worked with when I built tools for industry.

 

[shermama96]

I've seen similar construction used in industry for large, high precision machine tools--8" x 8" x 1/2" wall steel tube 20' long for a base with laser interferometers to guide the cutting head. The machine was going to grind and polish steel rolls used to make super calendared paper. Here's what I remember.

The steel was all normalized before fabrication. The top block which was machined to be the precision way was also normalized steel and was continuously welded at both bottom corners to the base tube. The difference between fabricating a composite beam like this with intermittent fasteners and continuous adhesion is night and day. The machine screws you plan on using will do nothing to improve the bending moment of this beam but will provide some shock resistance which epoxy will lack. West System epoxy gets my vote for adhesion to steel, but this isn't my specialty. I always just had base fabrications welded.

I'm not sure about filling the core. You may already have looked at the construction of One Way lathes which seem similar to me. They rely on the rigidity of a large steel tube, and that simple formula worked well for the machine builders I worked with when I built tools for industry.

I was hoping to braze or weld the composite beam together but as you said a normalizing cycle would be essential afterwards, especially with CRS. I don’t have an oven even close to large enough to run normalizing cycles though. The machine epoxy and fasteners seemed to be my next best move.

 

[whitmore]

I'm a second year mechanical engineering student, and as a summer project I've designed and plan to build a small lathe .
-I have read several threads on this site about different epoxy granite mixes... I have some cheap general epoxy resin from ebay that I planned to use, but I am worried the performance of the E/G will suffer dramatically due to the poor quality of epoxy..

That oughtn't be a major concern. If you expect vibration to be damped, it must be absorbed (not reflected) by the
bed, so a material that is soft and low density (the epoxy) is less effective than something with similar density to steel and
granite. If it were my project, I'd consider plaster of paris or hydraulic cement (both are intended to set up without
any volume change). Do you plan to make any test samples? Even just a phonograph cartridge and dropping a BB
on a sample can tell you a lot by how it rings or thuds.

Most of what you pay for in an adhesive is not the quality that matters for this application. Crack resistance, because
the whole thing is under pressure while it cures, and has a steel skin, is one of the irrelevancies. You don't require
tensile strength.

 

[JohnG]

OK. Makes sense. I can see that the hollow steel tube gives you a high moment of inertia, but I've also found hot rolled steel tube to be unstable for small, precision machined pieces. It's full of internal stresses that relieve themselves anytime you cut it. If you do use it, orient the weld in the least important face, most likely the bottom, and cut it as little as possible. I'd mill the top face flat, drill & tap for the mounting screws, grind it flat, then not do anything else to the tube, and count myself lucky to get one true reference surface. My antivirus software is blocking your PDF, so I can't see if you've already thought of this. On the other hand, you might just start with a 2" x 2" cold rolled bar, particularly if you need to work off more than one face.

 

[Asm109]

I'm worried grinding the CRS tubing straight will be difficult as it flatten out of the magnet and then return to it's original bowed state.

Thank you in advance.

That is why you lay the tube on the chuck and then place shims under the middle of the tube so it does not distort when the magnet is engaged.
Shim it, clamp it, take a light pass. Release the chuck, flip the part, re shim, cut the second side.
Release Flip back to first side and take another cut. Rinse, Repeat until it is as straight and parallel as you require.

 

[shermama96]

OK. Makes sense. I can see that the hollow steel tube gives you a high moment of inertia, but I've also found hot rolled steel tube to be unstable for small, precision machined pieces. It's full of internal stresses that relieve themselves anytime you cut it. If you do use it, orient the weld in the least important face, most likely the bottom, and cut it as little as possible. I'd mill the top face flat, drill & tap for the mounting screws, grind it flat, then not do anything else to the tube, and count myself lucky to get one true reference surface. My antivirus software is blocking your PDF, so I can't see if you've already thought of this. On the other hand, you might just start with a 2" x 2" cold rolled bar, particularly if you need to work off more than one face.

This is the issue I'm running into now, I can't seem to source steel tubing that would be ideal for this application.

The hot-rolled steel has large corner radii that would compromise some of the parts functionality, its wall thickness is non-uniform, it lacks straightness in every sense, and its composition is usually less than perfect.

On the other hand I can't even find thick walled CRS square tubing. Even if I could, I'm worried about the dimensional stability of the cold rolled steel, although it only needs several small holes drill in it. Since the foundation for this machine's accuracy is the bed, I'm worried even thermal cycling in the winter could induce some slight movement. I don't have the facilities to stress relieve the steel before grinding either so welding sections up is out.

I don't want to make the bed from solid solid steel either because I'd like to get the vibration damping properties for the epoxy granite and I don't want to make it out of solid cast iron because of cost. This is also supposed to be a unique learning project so I wanted to try some unique and less conventional methods to make the thing instead of just milling it from solid.


If anyone has any ideas I would love to hear them. You've all been very helpful so far.

 

[whitmore]

This is the issue I'm running into now, I can't seem to source steel tubing that would be ideal for this application.

The hot-rolled steel has large corner radii that would compromise some of the parts functionality, its wall thickness is non-uniform, it lacks straightness in every sense, and its composition is usually less than perfect.

Purity of composition is no problem (it's gonna be packed with rocks, you know), and if the project requires corners that aren't
in the structural tube, get some corner-like stock and weld it on. Normalizing stressed material (and welds) is basically the
same thing, just a few slow ramps up and down in temperature. An earthen pit and ten pounds of charcoal would
be useful. Dry ice, too.

Would two C-sections welded to a square, or a couple of ell-sections welded on as rails, be an improvement?

Straightening and sizing afterward is always going to be a problem, probably there's a surface-plate in your future. Or,
maybe you can mount the bed between centers and rough-machine it well enough on a mill? At least you could
get things parallel that way.

 

[shermama96]

I was wondering if anyone has experience drilling granite on a millling machine. Can it be done with specialized tooling without the hammering action of a rotary hammer?

I’m thinking about using solid granite for the machine base and need to drill some holes for the dovetail and for the rack gear mounting fasteners. I’d be drilling relatively shallow holes for 8-32 and 4-40 threaded inserts.