5 axis toughts

[TimothyD]

Hey guys,

i am designing a large format gantry CNC (3500mmx2000mmx35mm working envelope) The design would need to be able to cut thin sheets of marble (about 3-5 mm thick) using sawblades (so i at least need a 4th axis on the head) and be able to mill aluminium. Speeds don't need to be super high (i'm not expecting super fast cuts anyway). The machine would feature a chinese watercooled ATC spindle. 5axis capabilities would be nice since this allows a huge amount of work to be saved with sawblades (angled cuts and so)

I was wondering if anyone ever thought of using a trunnion table mounted to the Z axis as a "cheap" way to get 5 axis capabilities. I would of course need an adapter to my spindle but thats easily made. I was thinking of trying a cheap chinese trunnion table and just upgrade the steppers to some stronger ones. The other option is building my own 5 axis head, or selling a kidney to buy one. Please give me your thoughts on this mad idea


Similar Threads:

• Open Source : 4 axis and 5 axis print with a 3 axis milling in order to have a 5 axis
• Unwinding the rotary axis ( 4 axis rotary axis milling fanuc control)
• Need Help!- 4th axis rotation on x axis chaging to Y axis in simulation
• 5-axis, 4-axis, and 3-axis CNC Router manufacturer

[awerby]

I'm wondering what you do with these extremely thin sheets of marble once you've cut them. They seem like they'd be too fragile to be good for much. I don't see why you'd need a CNC machine to cut this stuff, and using a circular sawblade and a trunnion table seems a really difficult way to go. Why not cut it on a tile saw or a diamond bandsaw, and save the CNC machine for aluminum cutting (which is an entirely different task)?

[TimothyD]

The standard marble slabs are about 1500mmx2500mm and are glued to honeycomb aluminium. this makes them strong enough to build stuff with them. The machine would be used to make designer tables. the cuts need to be at a precise angle and location. We want a CNC to automate production as much as possible (and we would like to prototype stuff faster)

Aluminium cutting is not the main focus of the machine, but it would be a handy feature to have. If the tooling and speeds are provided. i think a machine able to machine marble will be strong enough to slowly mill aluminium

about the trunnion table idea:


this is basically a trunnion table with a spindle on it if i look at it

[Docop]

Hi
Pretty great design, as myself too are planning on this.So far a 60:1 could work out, but I do have a small 1.5kw spindle.. But on your design, you are quite bigger. Will you need more than 400oz on the stepper ?

[hanermo]

Forget it as drawn.
You are about 100x from being stiff enough.
The width makes a huge lever arm at 1000 mm pwr 3.

The height makes a huge lever arm at == 400-450 mm from metacenter pwr 3.

Then multiple 1000 pwr 3 x 450 mm pwr 3 => ??
And then you need less than 0.02 mm vibration at tooltip to prevent instant chatter and broken tools.

Some examples.

Drop the vertical rotational center by 50%. 8x more rigid.

The Z axis box should be thick steel, about 15-20 mm. Minimum openings possible.
Lowest axis should be about 100 mm thick, you can use thin-section bearings.
Use a a turned, ground, polished precision shaft, assemble in place, final fix the bearings.
Use a slight preload via a spacer from a shimpack, 2-10 microns, on one of the bearings.
So you just feel a tiny slight drag on the 5 th axis.

Then when you have a totally rigid, precision assembly, saw off the center part of the axis.
This gives you the only cheap, easy practical way to align the 2 bearings accurately enough.

Aligning 2 bearings, far apart, in x,y,z and xn, yn, zn is non-trivial, to under 10 microns and high angular accuracies.

4th axis, on top.
Make it 1D-1.5D-2D thick vs diameter of shaft 5.
So about 150 mm thick, and about 0.5-1D of free length of 5th axis.
So you want about 200-300 mm diameter on the rotational part on top.

Both will need preloaded, perfectly rigid, rotary controls.
Use 2 worms, preloaded against each other, or precision planetaries at 10:1 or less (more have extra backlash).

[TimothyD]

this isn't my design it was just a quick google search to illustrate my point of the movements being similar to a trunnion table, Thanks for the input tho

[awerby]

So these thin slabs of marble are glued to the aluminum honeycomb before you cut them? I wonder how well the diamond saw blades will deal with that - have you tried it by hand yet to make sure the aluminum doesn't clog up the blades? If the slabs are that large and aren't mounted to anything, they'll be difficult to handle without breaking them, particularly when they are stuck down to the bed of the machine with water suction. I don't think those cheap Chinese trunions will be rigid or massive enough to mount your circular saw arbor on, but you could probably get something built that was (following Hanermo's advice on its construction).

[Docop]

But, as from the previous picture design, a belt system could work, if it's just for wood. The load requirement will be quite much more lower.

And is 1 idler wheel might be helpfull to keep the belt stretch and then, kept out any possible slippery ?

[TimothyD]

it would be ideal if we could cut the marble with one sawblade and then cut the honeycomb with another one. But our current saws dont have any problems cutting thru the aluminium. The slabs are quite strong, and would be held in place with a vacuum table. another solution i was thinking of was using a pneumatic Saw like this:

and mount it as an second Z axis.

this would allow me to use a rigid 3 axis system for milling and a 5 axis system with a sawblade to saw stuff.

[3space]

@hanermo: Great answer. It's nice to have people like you who take the time to provide a thoughtful response. Thanks for your input.