[mkzr32]

Hello.

I'm currently taking machine shop classes and will have access to a CNC mill, and CNC lathe, along with an assortment of other metal working tools, and would like to build a 5 axis machine while I have the chance.

I've studied most of the information on cncroutersource.com and looked through this forum, along with a few others, but I'm still wondering a few things. Is it even feasible to build a machine like this, even with these tools? What parts can I realistically build at the shop, and what materials should I use if any of the one's I've listed are good for this sort of thing? Are there any good designs out there that I could start from to build a machine like this?

I'd like to be able to cut, and shape Styrofoam, wood, aluminum, steel, and possibly make PCBs and cut graphite. It will likely be used mostly on wood and aluminum, and I could sacrifice the steel, graphite, and PCB capabilities. I'd like it to be fairly large, although the size limit (Along the X, and Y axis.) without moving pieces on the mill seems to be about 16 inches, so I may wind up settling on something about that size. If possible I also want it to be able to handle a spindle and plasma cutter, if possible, and may eventually want a 3d printing attachment, again, if possible. I'm leaning towards either an MDF, or T slotted table. If practical I'd like to be able to operate the 4th axis separately as a manual late, have a work cooling system, and be able to disassemble some of the major parts for moving.

I'll have access to 1080, cold, and hot rolled, steels along with 6065 T5 aluminum, at the same price the school gets them at.

[Benonymous]

Constructing the machine will be a challege but the follow on is to get software to run it. As far as I know, there are only really expensive commercial/industrial cam packages or EMC2 freeware that runs under Linux. Also, unless you're planning to build a very substantial machine strength and stiffness wise, you'll have to lose the steel milling part of your spec. Knee mills and their ilk are made from cast iron for a reason, the tool pressures from steel milling are very high compared to wood, plastic or aluminium.
There are several DIY 5 axis theads on the Zone, do a search and read through them.

[CarveOne]

Milling steel is best done with a machine that is specifically designed for steel, like a Bridgeport style milling machine. They have relatively short X, Y, Z axes and cast iron construction for specific reasons aimed at accuracy and smooth finishes.

A CNC router machine of any size is a whole different animal. In your post, you have described at least three separate machine designs in one description. I would put the circuit board machine in a different category of specialized small machines specifically designed for accuracy in cutting very narrow traces and fine pitch component pads with very small carbide bits.

People have experimented with cutting steel on DIY wood router machines for the fun of trying it but they don't do it on a regular basis because the results aren't acceptable, it takes too long to accomplish, and it's too hard on a router and on the machine.

For a 5 axis CNC machine for milling steel you would be better off to put a specialized head on a used Bridgeport milling machine. You will also get better finishes on aluminum with that converted milling machine.

Wood and aluminum can be cut on machines that are what most of us are building in the DIY wood router machine forums. There are many sizes and styles of machines in this category. These are where many of us start, learn CNC, and then progress onwards to other CNC interests.

You can build these machines from just about any material, but there are always cost versus effectiveness issues to deal with. Make it as stiff as you can for what you can afford to pay. Consider the first machine as a learning effort, and buy good quality motors and drivers that can be reused on your next machine build. Don't skimp too much on these.

CarveOne

[rowbare]

Instead of getting bogged down in trying to build a 5 axis machine which you might not be able to finish while you have access to machines, why not build a decent 3 axis machine then use it to build another.

bob

[knau]

I would agree with Bob. Besides, it's much easier to build second time around and especially when you know what are the things that (will) go wrong on your first build.

[awerby]

While it would be nice to have all these capabilities built into a single machine, that doesn't seem very feasible. Also, every extra articulation always reduces rigidity, which you will need, especially for working with metals. Think of this first machine as a test-bed, for your skills as well as your projects. If you start out with a relatively simple 3-axis machine, you can add a 4th rotary axis later, or a trunion, or mount the spindle on a couple of rotaries attached to the Z axis for tilts and pivots. Try to scale the initial project down to a reasonable size, so you actually get something finished in the time allotted. Don't let visions of the ultimate machine get in the way of building an adequate one.

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

[mcphill]

5-axis software is getting cheaper. You can get full 5-Axis capable SprutCAM from Tormach for under $1500:

SprutCAM for PCNC

[jwolin]

Totally agree with building a 3 axis machine first. You will get lots and lots of experience that you can apply to other builds. I am on my fourth machine now and contemplating the "perfect" 5th machine (we all know what that means :-)

Also you the famous CNC toolkit can help you with the CAM side of things. Software is always the killer. The CNC Toolkit - Creative Toolpath Control

[mkzr32]

Thank you all very much for the replies.

On the program side: is there any reason I couldn't combine CNC Toolkit with GMAX to run it? If not I'd gladly install Ubuntu on a computer with EMC2.

On the machine side: wouldn't steel be better for the frame of a steel cutting machine than cast iron? I'm under the impression that a steel frame could be stiffer. Also, would it be possible to build a very solid frame from the materials I listed, and upgrade the rest of it later? or is there far too strong a chance of screwing up for even that to be possible?

As far as older threads go: I've done quite a bit of searching and only came up with a couple of incomplete threads, but will do more. Thank you for pointing me back to that resource.

Also I'm sorry for my placement of this thread. I simply saw "DIY" and jumped on it. Not really realizing it was specified for wood routers.

Lastly, are there any good CAD files or schematics for basic parts like ACME screws and nuts, gears, rails, blocks, etc. out there? And are those parts even worth building, cost or quality wise?

[noobCNC]

I havent gotten to learn GMAX and CNC toolkit yet myself although I have done enough to install it and get it to start properly. When you start GMAX you have to load in the toolkit. You do this by going under MAXscript and choosing run script. What I did was I also put CNC toolkit into the startup folder so it loads automatically.

Steel as a material generally has better mechanical properties than cast iron (e.g. its yield, ultimate tensile and shear strength), but the Young's Modulus is around the same most of the time. Material choice is one thing but more important is knowing the overall stress that will be applied to the machine and the ability of the machine parts to more than sustain those applied stresses before engineering failure occurs, is whats most important. Material properties is one thing, but designing the mechanical parts properly to a suitable strength to sustain the same level of loads, you can get away with using other less ideal materials.

I suppose since cast iron is easier for production if one so chooses and then they can square off and machine/flat the critical parts, the bridgeport machines could get away by creating parts with cast iron as long as the machine is capable of withstanding the machining stresses so that none of the materials that makesup the machine experiences stresses to the point of yielding, then it is ok. Engineering failure is critical , but for maching I guess machine deflection/bending and vibrations are also critical as it affects the quality of the machining, so other than downright failure by yielding they have to consider maximizing machine rigidity to reduce elastic elongation of the machine and minimizing vibrations. Essentially, Stainless steel allows you to create a strong machine with less volume of material since the material is better to begin with or you can use less ideal stuff like cast iron where the parts are generally larger to attain the same strength and rigidity, both can achieve the same level of capability.

[rowbare]

While many alloys are stronger, cast iron does a much better job of damping vibration. Mechanical Properties of Gray Iron - Damping Capacity

Using more cast iron to get equivalent strength further improves the system's damping.

bob

[mcphill]

Can you do ....

Absolutely!

Will it take you a few years to figure it out?

Likely...

If you have tons of time, patience, and great computer software skills, you can get there at a very low cost (even free). If not, you will need to make some investment, and even then it will be a very difficult learning curve. Even 3-axis CAM has many people stumped, going to 4, 5, 6+ axes is orders of magnitude more difficult. I would support all the other voices that you should start with 3 axes. If you feel you have mastered that, add a 4th. When you retire and have tons of time on your hands, you will be ready for the 5th.

Not to discourage you, but the trip you are starting on is VERY long. With a perfect machine, the software is very hard to learn. With perfect software, building a machine is hard. Building a machine for the first time, with no knowledge of the software (and assuming you are doing it "alone"), is a heroic effort.

All that said, dive in and have fun! I have, and love it so far.

[mkzr32]

Thank you for the further information. But does anyone know if are there any good CAD files or schematics for basic parts like ACME screws and nuts, gears, rails, blocks, etc. out there? And are those parts even worth building, cost or quality wise?

[awerby]

It's not worth trying to build yourself. And building your own linear rails, leadscrews, etc. would make for some very challenging projects, even if you were well-equipped for it and knew what you were doing. I'm sure you'd learn a lot by attempting it, but you might find yourself running out of time.

There are a few vendors who have CAD libraries of their parts catalogs on their websites; here's one: Reid CAD Library

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

[rocketflier]

This video may interest you.

[ame=http://www.youtube.com/watch?v=b-y03hhCCME&feature=player_embedded]UMC 10: Scratch Built CNC with Automatic Tool Changer - YouTube[/ame]

[mcphill]

I hate you for posting that

Any chance those are for sale?

[CarveOne]

I hate you for posting that

Any chance those are for sale?

According to the screen credits at the end of the video they are for sale and it gives an email address. Watch the last 30 seconds.

CarveOne

[PaulRowntree]

I think that the linear slides he has used on his sliding door panel look better than my X axis ...

[rocketflier]

I hate you for posting that

Any chance those are for sale?

Well, thank you very little,..

The thread I saw that in the poster said he had tried to get a hold of the guy and hasn't heard anything back.

A quick search {on the utube link} shows a thread over at home shop machinist where the price is discussed:
David Decaussin UMC-10 [Archive] - The Home Shop Machinist & Machinist's Workshop Magazine's BBS

Note, he uses weldments (machined steel tube sections) rather than cast iron which is why I suggested looking at it. There have been many a fine machine built here with similar methods and/or with epoxy concrete for damping.

[edit] Here's the thread I found the video in. Go through it and follow the links within it. If you have access to tooling, the design, time and cost of components are the biggest hurdles.
http://www.cnczone.com/forums/vertic...including.html

[noobCNC]

I think that the linear slides he has used on his sliding door panel look better than my X axis ...

LOL. I wish I had a better budget too!