Newbie- Building a machine with a sliding table AND gantry

[bvz]

Longtime lurker, first time poster:

I was wondering whether it is possible to control a system with a moving table AND a moving gantry. Specifically, does the "normal" software that I am likely to use allow for this kind of an arrangement?

I am asking because we are considering building our own cnc router system, need one that can handle 4' x 8' sheets, but are severely limited in work space. A machine that has a 4' x 4' table would fit in our workshop. A 4' x 8' machine (with the associated extra length to obtain a full 4x8 cutting surface) will not. Naturally we would have "extensions" that would support the working surface while it was running (and we would have/need 2+ feet of clearance on either end while using it), but it would be pushed in the corner and folded up when not in use.

I figure if the software can handle this kind of an arrangement, then we can get a full 8' of coverage on the x axis simply by moving both the gantry AND the work piece itself. An added benefit would seem to be that the drive screws (is that the right term?) would be shorter and therefore avoid some of the issues I have been reading about here when they get too long.

We've got a talented group of people here (including some over the top programmers if that is useful, as well as machinists and welders) but none of us have ever done any cnc work.

Stupid idea? Flash of brilliance? Insanely complicated? Any other comments?

Thanks for your time.

[FandZ]

Originally Posted by bvz Longtime lurker, first time poster: I was wondering whether it is possible to control a system with a moving table AND a moving gantry. Specifically, does the "normal" software that I am likely to use allow for this kind of an arrangement? I am asking because we are considering building our own cnc router system, need one that can handle 4' x 8' sheets, but are severely limited in work space. A machine that has a 4' x 4' table would fit in our workshop. A 4' x 8' machine (with the associated extra length to obtain a full 4x8 cutting surface) will not. Naturally we would have "extensions" that would support the working surface while it was running (and we would have/need 2+ feet of clearance on either end while using it), but it would be pushed in the corner and folded up when not in use. I figure if the software can handle this kind of an arrangement, then we can get a full 8' of coverage on the x axis simply by moving both the gantry AND the work piece itself. An added benefit would seem to be that the drive screws (is that the right term?) would be shorter and therefore avoid some of the issues I have been reading about here when they get too long. We've got a talented group of people here (including some over the top programmers if that is useful, as well as machinists and welders) but none of us have ever done any cnc work. Stupid idea? Flash of brilliance? Insanely complicated? Any other comments? Thanks for your time.

Have you thought about a vertical 4x8 machine? I've seen a few builds that fit nicely along a wall and protrude about 2 to 3 feet. Kind of like a lop sided triangle. You use counter weights to off set the weight of the gantry. It works.


To drive the 8ft section you probably want to look into a rack and pinion or a belt setup.

Ahren/cncrouter parts just finnished designing his rack and pinion setup that he will be selling.
http://www.cnczone.com/forums/showth...t=77372&page=6

Chris

[gooberdog]

This could be doable though it is bordering on over complicated. You would probably want the gantry to move in unison with the table. As the gantry moves 1 ft +X the table moves 1 ft -X. The program would need to see this as 2 ft +X. If the gantry and the table were riding on racks driven by a common gear, that would provide the movement you need. Rigidity to the extensions might be your big problem.

[bvz]

A vertical cnc router. Wow. Hadn't even considered that. Very cool idea.

It seems that there might be a number of advantages to this kind of a setup, in addition to the floor space saved. You might, for example, be able to get away with longer drive screws (lead screws? I have to get my terminology down) because they are no longer going to bow due to gravity. The entire X axis (which is now vertical) would also be pre-tensioned against the screws (assuming your counter weight is marginally less than the gantry weight) so no backlash there.

Of course, fixing material to the machine is more complicated. And the stepper motors will most likely have to be stronger on the X axis. And gravity will be trying to pivot the entire gantry along the Y axis so that will have to be much stronger.

Still, a very intriguing idea Chris. You say you have seen exampled of this? Would you have any links? I'll try googling it in a moment.

gooberdog, I hadn't considered the rigidity of the extensions. That's a good point. Though I think that should be addressable. I think the harder part would be to get the gantry and the table synced up so they both move in unison like you described (and I was originally thinking). That said, maybe I could somehow have the gantry do one half of the piece, then manually slide the table (along machined runners) to a stop that sets it up for the second half of the piece. Hmmm. Still sounds very complicated, especially for someone who has not done this before. The vertical cnc router idea sounds like it might be worth exploring more.

Thanks to the both of you.

[RubberChicken]

Hi, I'm a bit of a lurker too but your question has prompted my first post.

I have a similar problem and am about to build a 4x4 that can do 4x8.

I was inspired by the shopbot buddy with the powerstick

http://www.shopbottools.com/shopbot_buddy.htm

Cheers,

Paddy

[Hirudin]

I don't think syncing up the 2 moving axes would be much of a problem. After all, CNC machining is all about many axes moving in unison. I figure you would slave the table axis to the gantry axis (or vise versa). Make sure the two parts move in opposite directions*. Then in the motor setup you would just use half the steps per unit.

* Something that can be easily done several ways: in software; by manually changing the wiring to one of the motors; or you could use a screw with left hand threads (probably the least desirable as left-hand screws are harder to find).

[ger21]

Typically, moving the table requires more room than moving the gantry. Are you sure it'll fit with the moving table?

[Riceburner98]

Originally Posted by Hirudin I don't think syncing up the 2 moving axes would be much of a problem. After all, CNC machining is all about many axes moving in unison. I figure you would slave the table axis to the gantry axis (or vise versa). Make sure the two parts move in opposite directions*. Then in the motor setup you would just use half the steps per unit.

^^^ Exactly what he said. Before I found a 32" slide, I had planned out a 9" X axis gantry and a 9" X axis moving table for my benchtop machine. You should be able to get the full 8', but it will be a bit bigger than 4' in size. (plus the stock overhanging the machine) I use Mach3, but wasn't even going to let Mach know there were 2 moving parts involved. I would just let both (stepper) motor drivers run off the same step and direction signals, but wire the motors so they move opposite each other. Then tune Mach with half the steps per unit, as above. Mach sends out 1 unit, the moving bits move apart by 1/2 a unit each. The table would be mounted higher than the frame rails, so it would ride above them and clear on full moves. When 'parked' in the center it would be close to 4' x 4' in size.

All that said, the vertical machine sounds nice for a cramped area, depending on how far your Z / spindle sticks out. One screw will 'sag' the same as it will on a horizontal machine, just in a different direction. The screw that's now sitting vertical now won't sag, but it will see the entire weight of the axis moving parts (like being a huge Z axis), so a counterbalance / gas strut would likely be advisable to keep it neutral. That way the same motors can be used.. Clamping could be an issue if you're cutting multiple smaller parts from a 4x8 sheet though, as they would have a tendency to 'fall' rather than sit on the table after being cut free.

[Hirudin]

It sounds like the plan is to keep the machine in its most compact size when it's not in use (in the corner) and pull it into the open when they're running it.

[HuFlungDung]

Something tells me that moving the table and the gantry at the same time (in opposing directions) is going to be problematic when it is time to cut arcs.
I'd be tempted to do a full sheet with two programs, and move the gantry manually (ie., don't tie it into the cnc as a working axis), and lock it down in one or the other position. Use a different workshift for each position.

[Geof]

Originally Posted by HuFlungDung Something tells me that moving the table and the gantry at the same time (in opposing directions) is going to be problematic when it is time to cut arcs.....

Maybe not. If you pretend you are sitting on the spindle and watching the table move on a regular gantry you really have no idea whether it is the spindle (gantry) moving left or the table moving right. Connect the two to a single drive motor using opposite pitch ballscrews so they move in opposite directions and the programming should be completely normal. The downside would be that the positioning error will be doubled.

[Riceburner98]

Originally Posted by Geof ... and the programming should be completely normal. The downside would be that the positioning error will be doubled.

Exactly. Running should be transparent, but homing and manual (ie: hand crank) positioning could get interesting. I had looked at using 1 motor with a cable / pulleys to move both axis in opposite directions, but that got ridiculous pretty fast! Best / simplest to avoid 2 moving synced parts, but definitely possible if necessary. Rapids would theoretically be twice as fast, assuming both axis could run wide open.