Making a part from brass bar

[Matty F]

I was asked to make a bell part from brass. Just one was needed. I imagine a CNC mill would do this but I don't have one or know anything about CNC. So I just carved it up on the mill and finished with a file and a bit of a polish!

The finished job resting on the specification drawing:

the holes using the mill. I usually find that a drill press wanders all over the place, and even the mill isn't good enough without using a centre drill first: the shape with the mill. I had to cut the other side off with a hacksaw and grind the rest to shape: I expect someone will tell me a better way to do this! The 110 year old original was cast in a foundry. And it was a bit weak and broke after 110 years of bell pulling and brazing repairs.

[Bob Minchin]

Nice job as ever Matty!

You used a method that was appropriate to your tools. Nowt wrong with that.

Bob

[Matty F]

I wonder how long it would take to set up a CNC machine to make one part. And would it be economic to have a CNC machine sitting idle most of the time?

[Andy Dingley]

Can you find someone with CNC? Here in the UK, "hobby" CNC is everywhere these days, whether it's mills, routers, Makerbots or lasers. If you can find a local group like a Hackspace or a Dorkbot chapter, there may well be people there who can assist.

Another technique is casting. Brass (even easier, bronze) casting is workable as a DIY process, if you can afford to clean the surface up afterwards. Lost foam or Petrobond sand can give goood results for little effort.

[dennis]

You could have used a 3D printer to make a pattern and cast the part. I think the way you did it was better.

[Andy Dingley]

Assuming that you've previously worked on profiling cuts in 3/8" brass, so that you know the feeds & speeds to use, then it will just take you the time to draw out the profile (in Cambam or Auotcad), which will be about the same time as a neat pencil sketch, then 5 minutes of dull typing and moving files between machines on your desk and workshop. Then probably 1/2 hour drinking tea while it cuts it (home CNC isn't quick, as you can't afford a machine that will do heavy cuts)

Last thing I made - easy stuff to machine, but it would just have looked nasty if I'd hand-engraved all that text.

And would it be economic to have a CNC machine sitting idle most > of the time?

Like that's going to happen.

[Thomas Prufer]

the engraving with a firm colored wax, dripping it in with a lighter. Scrape even using the straight edge of an offcut, polish with a rag and a bit of oil, and there's your extra color...

Thomas Prufer

[The Natural Philosopher]

Lucky you had a milling machine... first job as a prentice was to carve solid metal with hacksaw and files.

You CAN cast these using 'lost wax'

Make a wax master. Candlewax will do.

Use something - maybe even plaster of paris, or cement and sand - to surround it with a couple of pegs to form inlet and outflow holes.

dry mould thoroughly in the oven, If wax comes out who cares?

Cast with molten brass or bronze, then break up casting mould and finish with surface milling and file and emery, and open up holes to correct finals.

[brass monkey]

We knew it as traffolyte.

[The Natural Philosopher]

You can certainly make CNC machines for smallish beer using small and held power tools to do the cutting. Use those on thin plywood a lot..not sure you could carve metal though.

Drawing the part up in 3D cad would take me about an hour or two.

Not sure how that would translate to 3d CAM files tho.

Could for sure make a 'path following' style plotter file easily enough.

leaving a couple of tabs to make sure the part didn't fall out of the sheet.

[Andy Dingley]

I do a similar thing for brass - engrave lines, fill with coloured shellac, then polish.

[Andy Dingley]

Bollocks. You've never actually tried this, have you?

I can't do this. I can't get the burn out of the wax to work well enough to produce a good part at the end of the day. It's harder than it looks. Vague handwaves of "candlewax", "plaster of paris" or even "cement and sand" and "dry it in the oven" are _not_ going to be good enough.

[Tim Lamb]

In message , Matty F writes

You could have done this on the metal cutting bandsaw I have just scrapped:-)

Nice job!

regards

[The Natural Philosopher]

google 'lost wax' or 'cirie perdu' for the correct way.

[The Natural Philosopher]

[dennis]

Did that at school. You even had to do it to a few thou using a file to pass the exams.

Then you had to machine a part with bits done to a few thou.

All following a drawing.

[dennis]

Well you could always dip it in ceramic slurry, then put it in the kiln to harden it. The wax will melt away in the kiln. You need to bury the ceramic in sand to support it while pouring the metal. Works well enough in industry so it should be OK.

[Matty F]

Actually I don't need a CNC machine, just a metal cutting scroll saw or similar. My bandsaw has a one inch blade and it's too hard to hold odd shapes.

Thanks!

[Andy Dingley]

It won't. What actually happens is that most of it melts out, but some of it soaks into the investment. When you make the pour, the extra heat of the metal burns the remaining wax, causing outgassing and thus blowholes in the surface of the piece. The investment can't be completely impervious, or else that gives its own problems with trapped gas from the melt. So an effective wax "bake out" has to be just that, a _bake_ out, not merely a melt out.

[dennis]

The kiln is so hot the wax will evaporate, it is only a hydrocarbon.