Quirky Turn Mill Build

I can use CAD, and even draw with ink on drafting film (ancient, I know), but sometimes I like to just build something, and work out the inevitable planning mistakes miscalculations in real 3D. For erasers I have torch, hand grinder, hack saw and files.

A couple weeks ago I decided to stop just thinking about it, walked down to the shop, grabbed a couple pieces of channel iron, square tube, and some iron pipe and started welding up a base and rails for a turn mill I had in mind. It felt great to actually do something!

The capacity was supposed to be 12" x 24, but I can already see that failures of visualization may reduce that an inch or two.

The square tube was for frame support, the pipe for rails set floating in collars made of larger pipe welded to the channel iron. The rails will be positioned by bolts through the collars.

I was thinking about making the mill convertible as either as turn mill, or a flatbed, but again, visualization limitations at my age may limit what I actually achieve. Turn mill is essential, flat bed optional.

For linear bearing housings I thought I'd try making castings using the lost foam process. I normally make wooden patterns when casting, but I got inspired by the great pix and commentary on BuildYourIdea.com and wanted to try this method out. The advantage is that relatively thin and complex castings can be made this way, without the need for draft -- that is, tapering the sides of patterns so they can be drawn out of the greensand.

It took a couple tries to work out what the bearing housings should look like. Well, drew it out in Sketchup actually, so not very spontaneous after all, then converted to a fake airfoil.dat file so I could use Jedicut to cut it out of foam on my homemade 4 axis hot wire cutter. Final version shown below. Also a test of the fit on the frame rails. (the pink versions were not final, I went with the blue, which have bosses for accepting screws and an upright edge for the gantry box):

This is one of the foam patterns coated with drywall compound that has been diluted to the consistency of thinnish yogurt. The long square section bits are intended to be sprues. The part is dipped and then hung out to dry. It takes a day.

And here is a first casting of the linear bearing housing. As you can see it's a little tight on the pipe rail. That's because I didn't allow enough clearance on the foam pattern. I could probably bore it out further, but will probably adjust the CNC code for a little more clearance, cut more patterns and re-cast.

A second casting attempt failed because the metal wasn't hot enough to completely fill the mold. Metal mus be very hot for casting this way, because of the thin part, and because the metal must also flash out the foam before filling the mold.

I do find that the lost foam process is slower for multiple parts than conventional wood pattern casting. You have to re-make the pattern for every failure, which includes making sprues and attaching them, dipping the pattern and drying it for 24 hours. In contrast, a wooden pattern can be re-used as soon as it is pulled from the drag, and you can pour multiple parts at once if you have enough flasks, and the crucible capacity for it.

The pour is also relatively unpleasant for lost foam by comparison. Flames and smoke erupt, and the metal looks dirty. In conventional casting, it's relatively quiet, and odor free, and the metal is bright and shiny.

You can cast iron? Gasp! How do I get you on my buddy list? :beer:

Er, isn't that just aluminium - 'the metal looks dirty' probably explains why it looks less like aluminium and more like iron?
Looking forward to seeing the rest of this machine in any case.

Quote:

---

Originally Posted by Jonathan

Er, isn't that just aluminium - 'the metal looks dirty' probably explains why it looks less like aluminium and more like iron?
Looking forward to seeing the rest of this machine in any case.

---

Sorry, yes, just aluminum so far, though I have gathered everything needed to cast iron, if I build a new furnace, and would very much like to.

Once I was testing some clay refractory samples I'd made up, and also some real charcoal I'd made (as opposed to barbecue briquets), and just for the heck of it, I stuck a foot long piece of 1/2" rebar (concrete reinforcing rod) into the top of my aluminum melting furnace through the vent. I was running pretty good heat , because after about 10 minutes the bar started to rapidly sink down into the furnace! I stopped the blower, and pulled the bar out because I didn't really want a pool of molten steel in the bottom!

Later when I opened it up there was a nice big iron nodule. So I'm sure it can be done with wood charcoal, as it was historically.

Btw that casting is supposed to house skate bearings, it's not meant to be a sliding fit, as it is now -- don't know if that was the reason cast iron was considered?

I think lost foam casting of the type I did wouldn't be possible with iron -- I can imagine big problems there from mold fragility and flammable gas. And considerable danger. I'd want to stay with regular greensand castings. Maybe others have done it, I don't know. But it doesn't appeal to me.

Also, I plan on keeping the traveling parts as light as I can, so aluminum is first choice for this part. The steel is in the base.

I cut 6 foam patterns out today so I have spares in case I get some that don't pour well I've also decided to pour them as 2" sections instead of 4" -- and that should make filling the molds easier. I'm also re-considering how to orient them in the sand and where to put the sprue. I'm going to try the first one with the flat base uppermost, and maybe a 1" diameter sprue instead of the four smaller 1/4" square sprues. I think the metal will fill well, but I do wonder whether I will get a shrink cavity on the flat base. Guess we'll find out.

I've been very influenced by the vertical orientation threads, so I'm going to try that on this one. I've also decided that this is going to be a pure turn mill, rather than a convertible flat bed/turn mill. So I will only use two linear axes and one rotary, rather than the 4 axes I first envisioned.

Because of these changes, I'm considering a cantilevered carriage, rather then a gantry, like a conventional lathe. But since the mill's X axis will be flipped vertically, the carriage (Z axis) will be on top, rather than extending in the direction of the operator, as it is on a lathe.

I was going to box the gantry from pieces in the original configuration, but now I'm thinking about making a single casting -- probably from a regular pattern in greensand (not a lost foam process casting) I'm not sure yet. Just some ideas floating around in my head.