Newbie- Complete Newb: 4 axis feasable?

[temetvince]

Alright, I find cnc extremely fascinating but I'm also completely new to the whole idea. So if you use any lingo at all without telling me what it means, I'll be lost.

I don't think a 3 axis machine will cut it for what I want to do. Get it? Cut it? Sorry.

I want to make 3d figurines. In general, I care more about fully 3d models and less about 2.5d wood engraving. To start with, I would like chess pieces (and I would like it to be very a very accurate machine in case I use it for anything else besides gaming pieces.) Eventually I would like a machine that would help me build a big sterling engine (solar powered to fuel my home), but that is probably a long ways away and it by no means has to be this machine.

Will a 4 axis make anything I would ever need to make, or would I basically need a 5 axis? I sure hope not, as from what I could ascertain they are almost impossible to build and get software for. Also, just as a side question, can a 3 axis machine be used to cut half the chess piece, and then flip the wood completely over, and use it to cut the back of the chess piece?

I don't mind paying for plans if they are good, but whatever it is will need excellent documentation as I'm so new to all of this. Any recommendations? Also, any idea on about how much building a good 4 axis will cost me? I've found them on ebay for around a thousand, although they could be crap, I don't know.

Thank you for you time.

[awerby]

I want to make 3d figurines. In general, I care more about fully 3d models and less about 2.5d wood engraving. To start with, I would like chess pieces (and I would like it to be very a very accurate machine in case I use it for anything else besides gaming pieces.) Eventually I would like a machine that would help me build a big sterling engine (solar powered to fuel my home), but that is probably a long ways away and it by no means has to be this machine.

[I guess the major question is, are you more interested in carving 3D models or in building machinery? If you want a machine to carve chess pieces, something like the Taig mill with a 4th axis attachment would get you going with that for a reasonable cost, assuming you could also purchase the appropriate software. I'm not sure that would be the same machine to make all the parts for your big Stirling engine, but it would be able to do some of them. On the other hand, you could easily spend the next year trying to build a machine yourself that would do the same thing, but which might or might not actually work. You'd certainly learn a lot in the process, though.]

Will a 4 axis make anything I would ever need to make, or would I basically need a 5 axis? I sure hope not, as from what I could ascertain they are almost impossible to build and get software for.

[You can do a lot with a 4-axis CNC machine, but certain undercuts may be unreachable. You can usually carve those in by hand afterwards, but if it's important for you to have the whole thing carved by machine, coming up with a CNC mill with more axes would be one solution. If you put a trunion on your 4-axis mill, that would make 5 axes, and it would increase the amount of your part you could reach, although it also would require a different sort of fixturing, or refixturing, in order to mount it on the trunion. You could alternatively add swings and tilts to your spindle, making a 6-axis machine, but there are few software programs able to write that sort of code.]

Also, just as a side question, can a 3 axis machine be used to cut half the chess piece, and then flip the wood completely over, and use it to cut the back of the chess piece?

[Yes, that's done commonly. The trick is to locate the part accurately enough when you flip it over so that everything stays in registration. One way to do that is to use locator pins on the axis of rotation. ]

I don't mind paying for plans if they are good, but whatever it is will need excellent documentation as I'm so new to all of this. Any recommendations?

[Sorry, I've seen numerous plans for 3-axis routers, but none that put much effort into integrating a 4th axis, much less a 5th or 6th.]

Also, any idea on about how much building a good 4 axis will cost me?

[That could be almost any amount of money, but it would be difficult on your first attempt to build a better machine than the ones that have been proven in the marketplace for less money than they cost, especially if you value your time at all. If you're really good at this sort of thing, have a well-equipped shop and a lot of persistence, you could probably come up with something that would work adequately for a few thousand dollars, but you'd still need a couple thousand more for CAD and CAM software.]

I've found them on ebay for around a thousand, although they could be crap, I don't know.

[You'll find out when you get it home...]

Andrew Werby
ComputerSculpture.com — Home Page for Discount Hardware & Software

[KOC62]

Welcome to the club....

You have ambitous goals. Because there is so much to learn, it is wise to start simpler.

If you can imagine holding the cutter tool in your hand, and then carve that piece of wood on the table into a chess piece you could imagine what problem(s) you might encounter. Limit your hand movements to the same motion as the x, y, z axis. The challenge is to design a machine that mimics your hand carving, with your hand motion restricted to the x, y, and z axis. When done as far as the cutting tool can go, you rotate the wood and repeat the above process - until your object is finished.

Yes you can carve chess pieces with a 3-axis machine, but perhaps not as easy as a 4-axis machine. Also realize that you cannot undercut an object as you may do with a 5-axis machine (within limits).

Consider starting with something like microCarve's designs to learn the steps with the goal towards a 4-axis machine.

[temetvince]

Thank you both for the replies. So in short, by goals were a little lofty. I am more interested in 3d modeling. I assume I could use my first cnc machine to help me build better machines for building the Stirling engine later.

It sounds as if a 3 axis would be best.
Reasons:
1. There are many well documented plans.
2. It is simpler (relatively speaking).
3. It will actually make chess pieces, although it will require extra work.
4. It will help me eventually get a 4 axis machine, both by a. making the parts and b. giving me knowledge gained from building and using the 3 axis machine
5. It will be cheaper. Big plus.

1. Would building a 3 axis machine with well documented plans, and using that machine to produce chess pieces be a much better, and lower aimed, goal?
2. If so, what plans are best recommended?
3. How much would the software cost to make the chess pieces (generally speaking). Is there any free software?
4. Do I need to know how to program in any computer languages?

I'm sure I'll learn as I go, but it would be nice to at least understand a bit more before embarking. I suppose I could always simply buy a 3 or 4 axis machine. Are they cheap enough that I could not build a home built one for the price of buying one? I am eyeing that microcarve KOC62 mentioned, but I am open to other companies as well.

[KOC62]

1) Yes. 3-axis is a good start.
2) MicroCarve designs are a good start. He makes small sturdy units and has plans available.
3) Not sure what you mean. The cnc needs g-codes of the object to cut it. Are you thinking of the 3D software to create the image? Then its file format needs to be converted to g-code. I haven't done any of this so I'm not able to advise.

I can't tell you what is "best" because there are so many variables that need to be researched, or known.

You could start with a kit, like Probotix Fireball V90 ~$600 plus electronics. Again, you need to decide if the building of a cnc machine is important or if carving chess pieces is more important. Perhaps both? Only you can answer such questions.

MicroCarve sells a fixed gantry type and claims good accuracy for a small design. The Probotix is a moving gantry, a bit larger than microCarve. Shipping costs may deter some purchases and needs to be considered. Finally, its the pocket book and floor space that often decides what you are limited to do. Add dust collection systems, noise reduction systems and you may re-evaluate the entire project.

[temetvince]

I guess I meant about converting the object to g-code. I'm assuming I could use blender to make the image, but after I have the 3d file I wouldn't know what to do with it. I know there's some info at the cnc linux website, so I might do some sniffing around there.

Ultimately the chess pieces are the most important, although I do think building one would be fun, although not necessary by any means. I just assumed it would take more time but be cheaper.

What's the main difference between fixed gantry and moving gantry, from a usability standpoint? I don't really need complex features unless the will affect they products and things I can produce. And I have a lot of shop space.

It would be nice to have a cnc machine that could finish the chess pieces well enough that they wouldn't require a significant amount of sanding to smooth them out.

[ger21]

1. Would building a 3 axis machine with well documented plans, and using that machine to produce chess pieces be a much better, and lower aimed, goal?

I think you'd be happier with a 4 axis machine.

I cut this on a 3 axis machine, rotating it 4 times and cutting each side. As was mentioned, alignment is critical.

3. How much would the software cost to make the chess pieces (generally speaking). Is there any free software?

Do you already have modeling software, or will you need that too?
You'll need a machine controller. Mach3 running in windows is $175. EMC running in Linux is free.

Once you have your model, you'll need CAM software. The best inexpensive 4 axis software is probably DeskProto.
I used MeshCAM for the part in the pic. But it only does indexed 4 axis, meaning it will rotate for as many sides as you want, but won't rotate while cutting.

I'd recommend the same as the others, either the Microcarve, or the Fireball. The one issue you need to be aware of, is will a 4th axis fit on either machine??

What I really recommend you do, is spend a month or two (at least) here reading as much as you can, before buying anything.

[KOC62]

A CAD software package designs the object, as per your Blender program. The output of CAD (e.g Blender) will be the input into a CAM software. Here you need to know if the CAM software can use the CAD output file. Some use stl as a CAD output file. Hence the CAM software needs to be able to read the stl file and convert it into g-code. The CNC machine will need software like Mach3, EMC2, and the like to read the g-code and thus control your stepper or servo motors.

The fixed gantry type is more solid due to less moving parts. Hence it tends to be more accurate because of its mechanical stability. The moving table needs a longer axis so that the cutting tool can reach any desired area of the table.

The moving gantry is less stable (relatively speaking) since it has to be driven, plus weight is a factor - i.e. moving mass has to start and stop. Alignment is more critical to maintain than a fixed gantry. It generally also takes up less space than a similar table size fixed gantry.

Many builders use a rough cutting tool first to get rid of most of the material and use a smaller tool with very small passes to finish it. A well made machine would not require sanding of the finished part. Much of that process is determined by correct techniques - e.g. feed rate and proper cutting tool with its appropriate RPM. These requirements also dictate the minimum spindle power required and perhaps speed control. The material used to make the object has its own criteria for proper cutting tool selection. e.g. type of wood, aluminum, brass, plastics, etc.

On this forum, both types have been successfully used but each has its own pros and cons.