Hello. I am researching a build, and I see that 2’x2’x81mm (Z) is recommended for accuracy. I can see myself wanting to intricate 3D carvings on solid woods, but what if i wanted to do, for example, something door sized, and I have a 2’ setup?
Is it possible to invert this machine and re-mount it from the top (rigidly of course), so that I could slide something large underneath it larger than 2’ wide, and still be able to CNC it?
Thanks for any ideas or advice! I started with and ender 3 3D printer making COVID faceshields, then quickly added a K40 laser and larger printer; now that COVID continues, looking to get something fun to whittle away the dark winter in the garage tinkering…
A full door carve would take a really long time. Like, maybe a week. Maybe a few weeks.
Flipping the feet up might be possible. I would have to think really hard to see if the stock parts could be turned in a way that would leave the gantry right side up.
Upside down the X/Y heights would invert, but that’s ok, they would basically rotate 90° and be the same.
You may lose a bit of work area. Since the trucks would still be right side up, the one core clamp with the cutout for the bolt would still be OK.
Increasing the number of parts between the bed and the corner mounts is going to affect rigidity, particularly since it would be VERY difficult to make the top all one piece, without making the legs long enough (Again, rigidity) to accommodate the core, Z axis motor, plus tool, and whatever cable management you need.
Still, even with those solved, I would not do it.
The feet, leg clamps and corner pieces are not really intended to hold the weight of the machine. The reason that the legs are so effective is that gravity is on our side. As soon as you invert the machine, the feet and leg locks must support the weight of the machine + stresses from movement, and an inch or so of clamping plastic to steel just isn’t strong in resisting those kind of forces. Upright it works because those forces are pushing the steel further into the plastic, instead of trying to pull it out. I’d expect all sorts of levelling gremlins in the machine in an inverted position, particularly operating with loads and vibration.
The strength of the plastic layer adhesion also becomes a problem. The feet are designed to not depend on any single layer, but the leg locks will fail if ANY single layer around the leg tubes from the end of the steel to the clamp fails. the feet become similarly stressed above the tallest of the foot clamps.
All in all, this ends up being a bad idea.
Make one axis 3’ will affect rigidity less and still allow you to do your door (in pieces) or of course a lowrider can probably take the whole door at once.
Dan, Jeff,
Thank you for the thoughtful replies and your insight. I had thought about a door being done in pieces also - good to know that will work in practice as well.
I have an old family credenza from the 1800s with inlaid engravings that is missing a piece. out of it. LIDAR scanning, meshing fixing, and CNC’ing a replacement flip down desktop (and keeping the old one) may be my first project.
I am fairly happy with 3D prints that have 0.2mm layer height and 0.6mm wide nozzle. If I make my machine 3’ in the width dimension with 3.25" Z, how accurate could I expect it to be with 1" OD structural tubing (looks like available is 0.065" wall… would love to get thicker if it exists).
My Primo is 25" by 37" work area. I am using 1" .065" wall tube (DOM, not stainless) Thicker wall exists, but it’s law of diminishing returns.
The accuracy is as good as you build it. I get a little bit of dip in the center, as I have not yet completed my torsion box for the bed.
I am very happy with the accuracy of the Primo. I am not looking for a machinist’s CNC, I’m not building engine components, or doing precision work. This is a very accurate woodworking tool. I have not used mine for aluminum, though I’d like to, once my CAM-fu is up to the task.
If my Primo weren’t much better than my 3D printer, I would be very disappointed. I am not. Of course, like anything else, it can only be as good as you make it.
The recommendation in the calculator is aiming for better rigidity, which then impacts accuracy. The smaller footprint has been successfully used by folks moving in to machining metal, where rigidity is very important. Many people have larger MPCNC’s, and the current Primo design is reputed to be more rigid out of the box than the previous Burly (which is what I’ve got, so I can’t provide an informed opinion). As suggested, the Lowrider was designed with large flat panels in mind, but you could do a (right-side up) MPCNC that was wide enough to slide a door through, then carve the panel in multiple jobs, advancing the panel through the machine for each portion of the carve.
I think your best bet here, would be to make the mpcnc able to sit on the door, clamp it in place, and lower the gantry through where the table would be. Basically, have a table that’s only edges with a giant hole in the middle to carve through
Hi Dan, did you choose 25" x 37" work area to allow for 1/2" either side of a 24" or 36" work piece? I hadn’t considered that, but I like you work area sizing.
That’s exactly what it was for. My old tablesaw could manage exactly 24" and I was constantly running into problems because I wanted something with an interior dimension of 24", so having just a little bit more was always an issue.
I needed 49" of length to accomodate the 37" dimension, but since MDF is sold in 49" by 97" sheets, it worked out okay.
hm so it wouldn’t be the entire surface. Just decoration, like a little filleted rectangle cut into the door, with some kind of raised design. May never do that, but still I"m just thinking ahead to what I will want to do. emulating an “old world” carving style is definitely on the dream list, but using my head to design it, instead of brute force and skill. If that takes a much more expensive machine to do time efficiently, then maybe this would bet me started. If I had say a 18"x12" board I wanted to turn into a carved eagle, about how long would it take using a DW660 on this?