[glintid]

I've designed a CNC moving table i'm looking to build in the coming year for 3d foam work primarily for themeing. I may cut some harder materials but it's not high precision work. A tolerance of 1 or 2 mm is fine seeing most things will be sanded and coated with epoxy or other hard coatings.

Here's a brief description of the design. ALL COMMENTS WELCOME!

The work size is 3m(X) x 1.5m(Y) x .9m (z) and it runs on dual V wheels on 1" angle iron@ spot welded to the rails at 45 degrees. I'll use stepper motors on all axes and rack and pinion on X and Y with a ball screw on the Z axis.

I initially tried to adapt the Mechmate design to have a deep Z but found I was stretching the concept. The moment arm and inertia of the Z axis would have tended to tip the Y car up. I needed to capture the Y car solidly with 4 dual-vees.

The Moving table seemed to preset a simple solution and I could then ignore the weight of the gantry. Actually, the heavier the better. The only moving piece is the Y carriage and Z axis.

In keeping the moving table separate and relying on bolting it to the concrete slab, simplifies the steel structure. Each upright has 3 bolts to the slab and the moving table frame 6. It complicates installation a bit but cuts down on steel work.

I've uploaded a few images of the sketchup prelim. model.

David

[jsheerin]

Hi,
If you put a plate between or on the back of your y axis beams, you would significantly reduce deflection. Since that's not a moving assembly, the additional weight should not be a big negative.

[glintid]

Thanks jsheerin, that makes a lot of sense. The greater rigidity may even allow me to use lighter Y axis cross members.

[FandZ]

Suckers going to be big. I like it! I've seen a large foam cutter recently. I'll try and find a link. Concrete is seldom flat, How do you plan on adjusting your table and your gantry to get them perfectly leveled? I'm just worried when you go to bolt everything down you may end up with a little bit of twist here and there in the fame that could cause other issues.

[glintid]

FandZ, yea, it's going to be big, but as they say in my other business, film, "go big or go home!"

I intend to embed bolts with epoxy or other fixing system into the slab. The retainer holes in the steel plates will be over sized to allow for some lateral adjustment. I will then capture the plate between two bolts, upper and lower to allow for vertical adjustment. The method is, level one side in the X direction, them level the other in the Y direction with a level and reference straight edge. This is the system used to bolt structural steel columns and light standards to concrete slabs. I'd install the fixed gantry first and then use the Z axis in the lowered position to align the moving table.

It all sounds so easy when I write it. I expect it will be a critical slow process in real life.

[FandZ]

Sounds like you got it all worked out I'm sure it'll progress nicely. Here's a link to that other big build I was talking about.
http://www.cnczone.com/forums/showthread.php?t=96031

[glintid]

FandZ, thanks for the link. It seems to be a light system that certainly gets the job done. I like the simplicity of their X axis tracks; a 2x2 tube at 45 degrees.

[glintid]

I neglected to add above that I may add a second set of mounting plates to the uprights but much lower down, closer to the moving table that would allow for only 30or 40 cm of Z travel. Then, if I start to do a run of "flat" jobs (lower Z), I can lower the cross member and negate some of the inaccuracy that comes with the longer Z axis.

It would be a bit of a job to move it down and realign things and I would only do it it makes sense at the time.

I could raise the table I suppose but that seems to be a bigger job or it would add too much weight.

[FandZ]

You could also make a solid table that would mount onto the base of the moving table. Maybe something out of uni-struts or extruded aluminum so it'll be light enough for you to lift.

[glintid]

It's been a month or so since I posted the early design for my 3D foam cutting machine. I've worked through most of the details. My next step is draft the parts and assembles in detail to have them laser cut.

I've been very close to the thing and am looking for some engineering input. I did a rough calculation of the deflection on the main Y beam and it looks like about .015 inches assuming 150 pounds mass on the y car and z axis assembly. Seeing I don't need really high accuracy because I'll be doing primarily 3d objects for themeing, that seems fine.

In short, I have R and P on x, y and z axis with one NEMA 34 7.2:1 on y and Z and 2 steppers on the X (moving table).

This design grew out of a Mechmate, believe it or not! and I've used Gerald's tried and true drive systems.

Any and all input is appretiated. Any engineering knowledge is limited and confined mainly to static, structural loads (buildings) so if there are any dynamics issues glaring I'd be glad to hear about them.

[ReignMaker]

I like this design, different than any I've seen on here, I'm no engineer but it looks solid to me. To answer FandZ's question about leveling, metal shims between the concrete and frame should get that thing as level as you want it.
One thing that I see that I would look into is mounting the x rack under the moving table rather than the beam, fixing the motor upside down from what it is now. The only reason for that would be to hide it completely, not necessary at all since chips can't get into it anyway. Would be difficult to mount the pinion though.

Can you tell me how you determine the travel on the x-axis? I would like to design a large moving table with as much travel as possible, but not to the point of losing accuracy.

[jsheerin]

One thing you might think about adding are supports for the gantry beam on the vertical beams. I found these helpful when I assembled my gantry. If you plan to assemble the entire gantry on the floor and then lift it into the vertical position, they're probably not necessary. Other than that, looks nice.

For Reignmaker, the x travel is the length of the rails - the width between bearings. So if the bearings were very, very short and very close together, the x travel would essentially be the length of the linear rails. However, you'd probably want the bearings around 1/4 of the width in from the edges of the table if you used two bearings, so in that case travel would be length of the rails - 1/2 width of the table. So typically you will need longer linear rails than if you used a moving gantry, but you can overbuild the gantry and not have to worry about moving it around. However your x motors will have to move the entire weight of the table and the part. In glintid's case (cutting foam), that's probably a good trade off. For large metal pieces it might not be as good an idea.

[glintid]

One thing you might think about adding are supports for the gantry beam on the vertical beams.

That's a good idea jsherin, It would hold the beam as I bolt it together.

[glintid]

One thing that I see that I would look into is mounting the x rack under the moving table rather than the beam, fixing the motor upside down from what it is now.

Can you tell me how you determine the travel on the x-axis? I would like to design a large moving table with as much travel as possible, but not to the point of losing accuracy.

Placing the rack on the moving table is a possibility I explored earlier on but decided against it fore some reason I can't recall now . I'll look at it again...I think it may have had something to do with the track extending past the table. Having the motors stationary certainly reduces the weight on the moving table and solves the dragging cable problem.

As for the X axis length, I built the table size I needed and then moved it to the extremes so the bit covered the whole table. I then extended the x rails to accommodate the movement. I'll get the exact length when I draft it in Autocad but the Sketchup model is pretty close, within mm's.