Machine base (aka table)

Sounds like a plan. Just going to do a good strong simple table like you have in the instructions and go from there. Thanks!

What size lumber are you using here? You happen to have any better pics?

@David Level and flat are two different things. A floor can be level, but have enough surface features to throw off your table. It can also be flat, but not level, which if you’re using a level on your rails, will make your table top not perpendicular to the z carriage. My living room floor is dead flat, but not level. I have to shim one side of some shelves almost an inch to make them level. Room was originally going to be the garage, so the concrete slopes to one side. My 4 foot level doesn’t rock on the floor though.
Table = Legs and frame holding the top.

I can post a few later when I get home, but they’re not really all that necessary. In my pic I’m using a mix of 2x6 for the outer pieces and 2x4 for the inner and vertical pieces. But that’s overkill for a MPCNC, since you shouldn’t build it this big (and believe me, I tried).

Just head down to Home Depot and get:

From there the table design should be pretty obvious, and the Simpson ties should let you test fit it all together before screwing anything down. When you actually screw it down though, make sure you have it all clamped and straight. If you don’t any clamps long enough, just get ratcheting tie-downs (http://www.homedepot.com/p/Husky-1-in-x-12-ft-Ratchet-Tie-Down-4-Pack-FH0829/206802316). That’s what I used, and it worked wonderfully.

1 Like

Would a decent way to ensure the top frame of the table is flat, be to construct it on top of the MDF, screwing it together while it’s on it? I’m picturing constructing the table upside down, using the MDF to make everything flat.

How flat is 3/4" MDF in general?

I used lumber largely because it was cheap and thick, but I concede it’s not going to be perfectly flat. I used shims to make it as flat as I possibly could for the pieces that hold the rail supports though.

MDF on its own is extremely flat, but it will adhere to the surface if you screw it down and will no longer be flat. I had grand plans to build an interlocking plywood torsion box table, but got impatient (it was going to be a lot of cuts). That would be the best possible option, though.

1 Like

Also, remember that most of what you’re cutting isn’t going to be perfectly flat either. You’ll have to either surface the cutting area or correct it in software. I’ll be experimenting with the latter in the next few weeks.

Ahh. So even if I built the table upside down on MDF, it will still conform to the lumber as soon as I screw it down, unless the lumber is perfectly flat. So I guess at this point the lumber would have to be selected and possibly be machined somehow to be as flat as possible or use shims.

I’m looking at expanding my 2’x 2’ existing MPCNC to be more like 52" x 26". You say “you shouldn’t build it this big, I tried” Do you think ill fail at the 52" x 26"

Thanks so much for the detailed list!

[size size=“{24}”]@David. Would the reinforced design be constructed like this?[/size]

The design looks good, assuming you’re using strong ties at the corners. If not, you’ll want your outer structure to kind of wrap around the legs. The ties make it really easy to square up, so I would recommend using them :slight_smile:

As for expanding the MPCNC to 52" x 26": I would really not suggest making an MPCNC that size, and also not recommend making one with differing X and Y dimensions. The tubing just isn’t rigid enough. I don’t want to push other designs on this site, but if you want something that large PM me and I’ll provide more info.

1 Like

Why not differing X and Y axis? I think that is the better way to make a large machine, if it isn’t rigid enough a midspan support is easy to add.

What machine do you recommend?

I was thinking along the lines of what Ryan just mentioned, that it would be much more stable to have a long and short axis than two long ones. I’m pretty committed to MPCNC at this point since I have all the parts and hardware already. I just need to build a table so I can expand it a bit. I realize there are going to be issues to deal with in going larger, but they don’t seem insurmountable. As Ryan also mentioned, many here put the mid supports to deal with longer reaches.

@Ryan
What I found is that the the longer axis bends while the shorter doesn’t, resulting in greater error in one direction than the other (squished circles, for instance). I was using 1" SS tubing at 4’ x 7’, but even the 4’ segment would bend noticeably.

As far as what I would recommend… It depends on what you’re trying to achieve. Here’s what I designed as a way to maximize reuse of my MPCNC parts: (Gigantic Bastard CNC by spiffcow - Thingiverse). It worked pretty well, but I have since dismantled it to make room for one I designed from scratch. That one is still a WIP, but so far none of the pieces have any of the rigidity issues that I had with the oversized MPCNC or the hybrid I posted above (It’s much larger at 9.5’ x 5.5’, but if you’re interested you can follow it here: http://www.openbuilds.com/builds/gargantua.5353/). I have you to thank for the idea for the long axis tubing rails on both designs :slight_smile:

If I were to build a quarter-sheet machine like we’re talking about, I would probably either a.) do something with OpenBuilds extrusions (e.g. Ox, Sphinx, C Beam Machine, etc.), or b.) roll my own using the relatively inexpensive supported rails and ball screws that are starting to enter the US market.

I’m not trying to dog on the MPCNC – people are obviously getting some good results, and it’s an ingenious design given budget and size constraints. But deflection increases proportional to the square of the length of a span, and that puts a pretty abrupt limit on the dimensions of this design.

So just having experiences it, I think I should mention that my intuition about having one longer axis was the same as your’s. I was wrong. One axis moves, while the other kinda wiggles. On small holes, you get ovals because the spindle stays in places while the long tube bends when the steppers move across each other. It’s not the top to bottom deflection you need to be concerned with, it’s the lateral deflection.

There is a huge difference from a 7’ to a 52", 52 is almost 2 times as rigid without any supports with a 26" axis supporting it things will be okay, but not ideal.

Anything above 36" I always recommend the LowRider, it was designed for that very reason. The lowrider will run circles around the mpcnc above 36" everytime. I have this in the size page and specs page. A big MPCNC can be done, I would prefer a lowrider but nor harm done try it out, if you need more than make a lowrider with the parts.

On top of that I am not sure the end goal of the build. It can easily be made large enough to carve surfboard blanks, foam is easy at almost any length, plotting, vinyl cutting, wood burning, etc. Too many variables I try not to speak in absolutes because of it.

I’ve only glanced at the lowrider so I have no clue, are you saying I could take the MPCNC parts and table and turn it into a lowrider with not too big of a hassle?

I should point out that the 48" spans bent noticeably, just not as much as the 7’ span. It might work for laser cutting or something I guess, but the problem is that the long span has no way to add supports to it. You can support the sides, but you can’t support the center. Even without load it would tend to bounce as it moves.

The electronics are the same, you will need some other parts, you can cut some of them out yourself once you assemble the mpcnc.

Pretty sure Barry is cutting with a 4’x4’. Like I said too many variables including conduit type and rigidity. And anything above 3’ I have it on the size page that I recommend the LowRider. I don’t feel I have misrepresented myself or the machines capabilities anywhere. In all size threads you can see me jumping in and recommending smaller.