[richard5000]

Hi. I'm a newbie on a learning curve.

I have some questions, I hope I'm doing right asking them in this topic area.

Imagine a 2" square brass plate 3/8" thickness. And that I wanted a single character/font to stand out in relief, say 1/32th" (or about 1mm) in height. To achieve that relief, one can have the part cast (as in sand casted), or CNC machined. Of course, one might end up, depending on the character, milling out 80% or more of the front surface area.

Okay:

* I presume that milling to a depth of say, 1mm, is not engraving, but CNC machining?

* If I wanted to make 50 of the items a week, would you CNC machine the items or get them casted?

I like the idea of getting a CNC machine, but realize that casting might be a no-brainer from a cost perspective.

I wonder if anyone else here has faced the issue of choosing between casting or CNC machining and what was learned.

Thanks. Rich

[awerby]

This seems like a situation that could go either way. Mastering them on the mill makes sense if you've got one; you can make a pattern board for sandcasting pretty easily. 50/week is small for a casting run, and fairly reasonable for machining. On the other hand, if you wanted 500 of them, the economics of casting would prevail, I'd think, and you would have enough of them to do something else with your mill for the next 10 weeks.

Yes, milling to a depth of 1 mm around a letter is called Face Milling.
I don't see CNC milling as antagonistic to casting; it's a tool that makes it easier to do many casting-related tasks.

Andrew Werby
www.computersculpture.com

[marcusCA]

How many total will you do?
If this is only a series of 50 I would probably mill them all and be down with the batch.

It is about how long it will take for a piece x 50 x number of week running, compared to the casting capabilities you have (+ eventual polishing time).

Marcus Minor
Application Engineer Artclip3D Software
ArtClip3D | 3D CNC Software for Router - Wood Carving Software - V-Carving - Signmaking

[Geof]

What type of finish do you want? Sand casting brass is going to give a rough finish, poor corner definition and will require considerable finishing around the perimeter and back. CNC machining once you have the program will be very rapid and you will be able to chamfer all the corners and get a machined surface that will require very little finishing.

[richard5000]

It's the issue of surface finish that currently feel I need to grasp because I don't have experience. Let's just cycle through this in a bit more detail.

Lets say that I'm making a house number and, as I mentioned, I'm face milling to leave the number in bas-relief. Also, that in the end, the whole thing will be chrome-plated and look smooth and shiny.

Now, I'd likely have to smooth the surfaces. Logically, it would be, I think, a trivial matter to smooth the back, sides and the surface of the number itself. But, how would the field around the number itself be finished smooth? Where I've milled out the material. Would I put into the CNC machine a polishing tool? (Sorry for being so ignorant! :-) )

Common sense tells me, after milling I'd insert a second tool to polish the field around the number that is in relief. What kind of tool?

[richard5000]

Let's now imagine that it's sensible to sand cast the product rather than CNC it.

If I purchased a CNC machine, my output would be a pattern.

Now, it seems to me, that it would make sense to make the pattern of metal if anything.

So, I'd be polishing the pattern.

Then, I'd get my casts - but that process might introduce surface irregularities, so that would involve polishing the casts. You would be back to using a CNC machine to polish the field around the number at least and maybe redefining the number outline.

All true I think if I were to want to make a house number that looked precisely engineered. Which of course you may not in practice.

[beone]

At 2 x 2 I would get between 8 and 10 cast per mold using my standard 8 x 12 flask, so 50 isn't even a days run. I usually run 10 to 15 flasks per session, two sessions a day if needed. With sand casting you can get a nice "pebbled" type of finish that can be very nice by itself with a little wire brushing. File to get to a flat "mirror" type finish. With cnc you get a nice finish that is very dimensionally correct(needed for a house number?) but you willstillneed to file, sand, or polish to get the "mirror" finish. Also, cnc is not particularly fast. Casting will be far faster if the finish and tolerances are acceptable.
I love cnc and foundry is my passion. I built my cnc'd in order to make patterns for my foundry. Which I use for a particular item depends on volume, finish and tolerance.
Dave

[Bob La Londe]

Lets think about this for a minute. Would a combination process perhaps be the fastest? Machine a steel mold. Cast your part, and then if need be you could make a couple fixtures to finish machine the cast item.

Its sounds like a setup to make and sell brass house numbers. I highly doubt that he has any house numbers with 50 of all the same numeral. So... He is going to need a minimum of 10 numerals, and maybe the alphabet as well. Given a potential master set of 36 (maybe more with a few bits of punctuation) simple sand casting might be a lot faster and cheaper to get started.

"Lets say for example" isn't going to cut it. (pun intended) We need to more to give optimum advise.

[hjl4]

I agree, Bob your point well made.
Myself, I would have the molds made.
Their are many tricks you can use to minimize the amount of time you will spend polishing or sanding the surface down. Sand casting is tricky on smaller parts.
Here's what I do to get super smooth finish in Aluminum. After years of doing this at work, I can tell you it works well.
Before pouring, (at Start of day and after you start the furnace up), heat your steel mold to 300~500C. Depends on how thin your part is. Mix some porcelain or ceramic dust or even plaster of Paris with water into a very thin solution, almost like milk in the case of plaster, and spray the mold cavity with it using a paint sprayer.
What happens is the water( the carrier ), evaporates leaving behind on the face of the molds, a thin coating. Again the thickness of this coating depends on many other factors. But anyways, this coating prevents checking, veining and cracking of the mold face by insulating it.
It also allows the liquid metal time to flow into small crevices without losing too much heat thereby solidifying too quickly. Plaster of Paris leaves a very fine surface finish, so I estimate you would only have to buff the imperfections out.
Just my thoughts.

[Bob La Londe]

That's kind of interesting. Lost wax casting often doesn't use "sand" anymore, but a very fine porcelain or plaster dust. I know a jeweler who does a lost wax process for certain types of custom pieces in an ultra fine media like that in the classic lost wax in sand method. Except for the sprue, the vents, and obviously stone settings he gets pieces requiring very little hand finish work that way. He cuts the wax blanks on a high speed 4 axis CNC machine in his workshop in the back. He's not using a Taig or a MaxNC. LOL.

[lccsuppliers]

investment casting(Lost wax casting ) might be the best solution for smaller parts. thanks.

[zeox]

for 50 pcs. of standard sized house numbers, with a different number on each. I would consider obtaining a slab/sheet of aluminum/brass/bronze and milling them in a single operation
For 50 identical pieces (same number on all) I would 3D Print or CNC mill a casting model, and cast them. If I had the choice, I would prefer 3D printing a casting model that incorporates maybe 10 pieces with a common sprue and gating sytem, and cast them 10 at a time.