[robotmickey]

Hello all,
Having fooled around with lasers and 3D printers for a few years now, I figured it was time I graduate to some sort of CNC mill/router. I have built go karts and gadgets and the like all my life and the temptation to make parts out of usable material is far too tempting. Though I have not built either my laser nor my 3D printer I feel I'm familiar enough with the components to at least draw something up.

The first picture is an electric wagon go kart I built while still in high school. The next is a just a little cardboard throwing star off the laser.

I haven't put toooooooo much thought into the actual dimensions of the machine though this is close to the size I'm looking for. Mostly still conceptual at this point. I'm looking for some validation from people who have a better idea than I do. (why I'm here!) Let me know if this shape/component combination is something worth pursuing. I've done some reading, forgive me if I ask a frequently answered question.

Design Objective:
-Cut aluminum relatively quickly.
-cut steel? (is this a dream?)
-Desktop-ish size
-$1500 range budget
-no frills, hopefully easy to use

So far:
A fixed gantry configuration made of welded 2"x 2" by 0.25" wall square tube. My plan was to weld the pieces together in such a way that I can fill the whole frame with sand (i.e. drilling large holes to 'connect' the members before I weld them together so the sand can fill the entire volume of the frame). I chose this configuration as a compromise between rigidity and build volume while maintaining a reasonable "desktop" size. At present, the spindle is the generic 2.2kw model I found on GrabCad though I am interested in a belt driven brushless setup I've seen some members use. I've heard the high speed spindles are much too fast for some materials. Rails are 20mm square type, ball screws are 1605 type. Steppers are 425oz-in nema23s (mounting remains undecided as you can see in the model). I made the frame more 3D printer shaped (cube) as I would like ability to slide poly-carbonate panels into grooves just on the interior of the machine to create a chip/coolant enclosure. Many of the routers I've seen online cut using only air or the fogbuster style sprayer. Is there a reason more builds don't use typical CNC coolant? In the picture I have the machine sitting on an angle-iron frame that I will enclose to catch chips and coolant. I think I like the Gecko G540 4-Axis all-in-one controller. I've heard good things about it, but it is a little spendy.

Travel is roughly:
x:19"
y:8"
z:4"

Questions:
I know every one of these questions is a huge can of worms to get into, with many things that factor into the answers.
1. Do you think this type of fixed gantry configuration and geometry is rigid enough to mill aluminum relatively quickly? (mostly: whats your overall opinion of how it looks)
2. I have no intuition about the size of the steppers required. I figured that this is a smaller machine and I wouldn't need much more than 425oz-in? Is this fair?
3. Talk me into or out of the Gecko all-in-one thing. I see significantly cheaper options that appear to the same thing. At this level of machine, is it worth it?
4. Is cutting steel on a machine like this a pipe dream? With a different spindle maybe?
5. 1605 ball screws large enough? Or would stepping up to the 20mm be worth it?
6. Asked above: Is there a reason more builds don't use typical CNC coolant? At least the ones I've seen...
7. What your concerns with my design/component selection so far. What are some things you think I may have overlooked or not thought about?

Thanks for your help in advance!

[arizonavideo]

Lots if things here to talk about but to start with the basics first.

You can do a weldment and be just fine and sand filled or glue and sand is a good idea. You want to make a frame that will allow you to adjust it so you can tram the mill.

If you want to make a lot of things chip management with water cooling or a blower and a path for chips to go is a good idea.

16mm screws are fine and 425 oz steppers are good. The Geckos 3.5A are fine.

One thing to look at with a high speed spindle is your cutting speed. The 425 oz steppers are fairly high inductance so a 50V system they will not be that fast. You will be spinning your ball screws and steppers way past the point of max power at 150 IPM if you want to do 3D work with small bits in aluminum. If you don't plan on using let say larger then 38" cutters most the time which for a router is about normal then going to 10mm pitch screws might be a good idea.

Steel will be quite hard to do. I would say anything larger then a .25" cutter would be too much for that type of mill. 2.5" tubes would help and wont cost much more. 16mm rails are fine too. If you can find them get med or high preload rails.

I would change a few things on the design but lots of things would work.



For small cutters at high speeds you will want to have almost zero backlash in all the moving parts so good AC adjustable end bearings and re packing the ball nuts for no backlash will need to be done.

I have a 4' double nut 2.5 TPI screw I could let go for really cheep. After I get my shop up and running I could do up the ends.

[handlewanker]

Hi.........., a 19 X 8 X 4 is a mill configuration and will handle more materials than a gantry any day......plus the fact that you can have a belt driven spindle that will mill metals like steel as opposed to just engraving them..
Ian.

[robotmickey]

Thanks for the info!
Educate me on the inductance topic a little please. My assumption is the 425oz stepper is wound to produce more torque, sacrificing rpm/v?
Could you see a case where I could choose smaller/faster steppers, using the same 1605 screws to increase the speed of the machine , and not see a significant decrease in performance anywhere else?

[rocketflier]

I've been picking up .25 2x steel drops ($0.79/lb) for just such a thing. I'm thinking more along the lines of

[NIC 77]

A couple more threads of interest:

http://www.cnczone.com/forums/mechan...464-cnc-2.html

http://www.cnczone.com/forums/epoxy-...cnc-forum.html

Only a few DIY machines I've seen cut aluminum really well. Surprisingly, the two best ones are actually moving gantry machines. Cutting steel isn't a dream on a DIY machine necessarily, it's just rare. It's a dream with what you have designed so far I think.

Personally, I'd use epoxy granite or concrete to fill the gantry with instead of sand.

2" x 0.25" tube will distort much easier than larger tubes will when welded. Larger, thicker tubes, filled with epoxy granite or some other similar thing is your best hope of cutting aluminum well.

I'd personally use the 20mm screws instead but the kind of linear rails you use will have a bigger impact 20mm square should be OK. For the ballscrew, 10mm lead will get you faster speeds and I believe are a better choice.

I don't think 4" Z travel is enough because what happens when you want to mount one or two vices on there? Well now you can't.

Geckos are good drives (from personal experience).

Leadshine are good drives too (so I hear).

1500 total is hard to make a good machine for.

Good luck.

[arizonavideo]

I like that frame a lot.

The cost to go to 2x3" or 2.5" tubes is not vary much.

All the non moving arts can be as large are you want so why not go to 3" .25 wall for them

[handlewanker]

Hi, if you're happy with the design concept I don't think you would go far wrong by having the Y axis as a short travel and the X axis as the long one.....that would give it quite a bit of bracing to resist deflection etc...........I wouldn't bother filling the tubing, just make it as heavy a gauge as you can find........over engineering the base frame won't make it any stronger to resist cutter forces.

All the force resistance is concentrated at the back of the frame that supports the crossbeam so this is one area you can make as strong as possible while leaving the front open for access.
Ian.