Small Epoxy Granite Filled "Watchmakers Lathe" Design Question

[shermama96]

Any time I see the word "Cornell", all I can think of is Andy Bernard saying "I went to Cornell. Ever heard of it?"

I posted about my little CNC lathe here: https://www.hobby-machinist.com/threads/scratch-built-cnc-micro-slant-bed-lathe.62966/

At the moment, it is on hold because I have been focusing more on my new mill, but I will get back around to working on it eventually. I have been working on a design for a tiny CNC mill to go with it. Maybe one day I will make that too and have a pair of "large" machines and a pair of tiny ones.

My friends bust my chops every time I mention it by throwing office references in.

The little CNC lathe is very impressive, really looks great. I like the idea of JB welding the headstock up. Although I'm sure it will never come apart, I'm way too OCD and would image it delaminating every time I took a cut. What material did you turn the spindle out of. I was thinking about some pre-hardened 4130 for mine but wanted the ability to run collets in it and don't trust I could turn/grind a ER collet seat. I think my favorite part is the 250 dollar limit, which is pretty low to get a fully functional lathe with a tool changer and everything.



I didn't get very far with my vertical mill design. Once I started calculating cost I realized this might be the greatest financial move as a self-supporting college student. Would love to see a yours if it starts coming together.

 

[shooter123456]

The JB welded head stock was one of the cost saving measures. Using the 4 smaller pieces from scrap yard cast aluminum cost about $7 I think. The alternative was using a single piece of billet, but sourcing it would require a 12" piece and that was $75+ shipping (or something around there). I am not an engineer, so I may have this all wrong, but JB weld claims to have a shear strength of 3960 PSI, and each face has about 10 in^3 of contact, so to separate any 2 of the plates would take about 40,000 lbs. Even if I managed to get 2.5% of that, I don't ever imagine the spindle on a 60ish lb machine ever producing 1000 lbs of force to pull them apart. Also, the way the spindle is designed, the preload on the bearings will be forcing the plates together, which would make it even more difficult to separate them.

The spindle on that little lathe is made of Mysteralloy from the recycling center. Cutting it for the ER collet really wasn't too bad. It may have been smarter to use the lathe to cut the taper once it was done, but just used the big lathe and some prussian blue to get the taper close.

I do wish I had an engineering background so I could know if those sorts of calculations were anywhere near correct, but for the most part, I just have to go with overbuilding and hoping it works out...

Here is my design for the tiny milling machine. I tried to design it around components I have found inexpensively to keep cost down and using the dimensions of materials I can find at the recycling center near me. The idea was for it to be really small so I could put it in a spare bedroom or basement, and move it so I can still work on stuff when I don't have access to my larger machines. It obviously needs a lot more work before I start doing anything with it though. Thinking something like a BT15 spindle around 15,000 RPM.

Next to my PM25 for some scale.

Also, "Its pronounce Kernel and its the highest rank in the military." "ITS PRONOUNCED CORNELL AND ITS THE HIGHEST RANK IN THE IVY LEAGUE!"

 

[hman]

Like any epoxy (actually any adhesive), surface conditions will affect JB Weld's bond strength in a big way. I've had great success with sand blasting the mating surfaces. Then either rinse off any stray sand with plain water and let dry "naturally," or else just blow off what you can with clean dry air.

Sandblasting does two things - it removes any trace of oily residue and other contaminnts. And it gives the surface a "tooth," with additional area for the epoxy to adhere to.

I've not done any testing of this theory (sandblasted surfaces vs conventional cleaning methods), but the technique is cheap and easy. And in theory, it should be pretty effective.

 

[shermama96]

Decided to bore the headstock through on the 4 jaw out of a big chunk of grey cast iron square stock. Drilled and bored close on the mill and then cleaned up with a boring bar in the lathe. Once I got within two thou or so I switched to a 36 grit grinding wheel on the tool-post grinder. Had to cobble together an extension to clear the 4 inches but it seemed to work okay.

The bearings will fit in with very light taps with a brass hammer which is perfect because the recommended housing tolerance is 0 to 7 tenths over. I think there is a slight waist towards the center because the last quarter of the bearing, the press really takes up.





In order to ensure that the dovetail was parallel to the spindle axis, I turned up a pin to be a close fit with the headstock bore by tapering the first half of the pin length with a coarse bastard file and leaving the far half a light press. I could then indicate on the pin along the top and side. Each was within 4 tenths along the the 4 inches length.





Milled the headstock and the clamp as an assembly:

 

[NortonDommi]

Also considering putting the mixture into a vacuum, such as a lab bell jar, if you access to one large enough. Whatever plan you select, it's probably worth doing a test cast of the mixture first.

Making a simple vacuum bag and using the vac-sealer you bag all your freezer stuff with is a good idea. Old fridge/freezers are a good source of free vacuum pumps.
Most Epoxies can also be thinned slightly with a solvent,(Acetone is often used),also temperature besides speeding cure also lowers viscosity.