Hi Everyone, this is a pretty amazing community, I can't wait to get started on my machine. I have wanted a CNC Router every since I saw the first DIY MDF machines years ago and after learning about open builds I think I am in a position to build what I want. I am looking to build a 1500mmx1500mm machine that utilizes as many off the shelf parts as possible. Because my interest is in larger projects such as furniture and speaker cabinets eventually I would like to be able to load full 4x8 sheets of plywood so I am trying to design something that can be built upon or partially recycled as I gain a better understanding. My plan is to use C beam for the X, Y and Z rails with the X axis being driven by CNC Router parts R&P gears and drives and the Y axis being driven by a 1/2 5 start Acme screw and the Z by 3/8 or 8mm Acme. I was hoping that someone familiar with openbuilds hardware would look over what I am planning and let me know if I have made any obvious mistakes and how rigid it appears before I order enough of the parts to partially mock up the machine to better design the R&P X drives and the Acme Y. Below are a few shots from Sketchup, note they do not show a bed which will be a torsion box entirely below the X rails so I can keep the Risers as short as possible and keep most of the cutting area around the center line of the x carriage (is this a good idea it seems like it would limit the effect of deflection from the x axis).
I've built a CNC router combining OpenBuilds C-Beam and CNC Router Parts R&P. Take a look at some of the photos and the cad link I posted here: Rack & Pinion Here is some advice: Any axis that you decide will be acme screw driven should use an 8mm ACME screw and the fittings from the OBPS. If you choose a different size, you will need to make or find a lot of custom fittings. The CNC Router Parts store fittings for ACME screws, for example, are designed to attach to a different gauge of T-slot rather than the 20mm-base stuff that V-Slot is. So your best bet is to stick with all OBPS for those axes. Then later, if you want to upgrade to a larger screw, you will have a good idea of what kind of custom parts you will need and be able to design them. I'd make both the X and Y axes into R&P drives. You will be sad if one of your two large axes moves at 1/4 or 1/8 the speed of the other one. And while increasing the cross section of an ACME screw can increase the effective length, a 1500mm span is likely going to put even a 1/2" diameter screw at the edges of its capabilities. I know that a 1m span with an 8mm screw was definitely a stretch. Conventionally, the Y axis is the two rails attached to the table., the X axis is the elevated axis perpendicular like the cross bar in an 'H'. Let me talk about your X axis first. It is hard to see your X-axis, but I do see a couple of potential issues. First, you have gantry plates on both sides but they don't seem connected. Every screw that goes through the wheels of one gantry should connect (using spacers) through the other wheel and then through the other gantry plate. This gives you the most solid connection you can get (at least when you have two separate gantry plates). Second, the ends of your acme screw seem to go through the vertical extrusion on either end. This will require some custom drilling, and more importantly, you will need to add some kind of pillow block on either side. A pillow block is a support at the end of the shaft which has a bearing inside. This supports the shaft and lets it rotate freely. Although I am not an engineer, I've read that when your shaft spans a long distance, it is better to have two pillow block supports at the end with a bit of distance between them to better support it. So in this scenario, you would likely want one pillow block on either side of the extrusion and then attach your motor to one of the ends after it moves through the pillow block. You will also need lock collars to prevent the shaft from moving laterally. On your Y-axis, I'd swap things. Put the rack teeth inside the center of the C-Beam and the wheels outside. Given the weights and stresses, the wheels being further apart will provide better stability. I'd also try to center the load across a larger base. This leads into the other and more major problem. The way the R&P works from CNCRP is that their assembly is attached to your gantry plate in two locations. One location is a solid connection, but is a pivot on a bearing. With just this connection, the rack gear is allowed to rotate 360 degrees freely. The other connection is with a long screw and a spring. The long screw and spring have to pull the gear against the rack so that the gears consistently mesh. The spring lets you adjust the tension. But the point is to keep the gear on the rack countering gravity and countering the forces when the gear moves. But not to make it so tight that there is excessive wear. Take a look at the photos of my build to see how this works. On both the X and Y axis, I think your best bet is to use custom lasercut gantry plates like I do if you are using a R&P system. This is not only more compact, but it will likely be easier to maintain down the road. On a previous build, I did something similar to what you did with vertical V-slot connecting to horizontal V-slot and while it seemed easier on paper, it was harder to assemble and harder to tweak my design later on. A good way to prototype custom gantry plates is to remember that the outer shape doesn't matter very much. The gantry plates exist to provide a set of holes in the proper geometric relation to each other. So get a piece of plywood (or plastic) of roughly the right size, print out a perfect to-scale diagram of your gantry plate, tape that paper to the plywood, drill out the holes, then start putting screws in for testing fit. You can ensure a 100% perfect fit gantry plate with low turnaround times if you need to tweak things. And then at the very end you only go to the laser shop once and only pay the fee to get them custom cut once. You mention considering upgrading to a 4'x8' at some point in the future. If you want to upgrade, your best bet might be to attach 2 C-Beams end to end and use a bunch of quad makerlinks to securely attach them. Make sure there aren't any discontinuities in the rail at the point of connection. You can also abut two CNCRP rails together and they should tesselate perfectly. So then at that point, all you need to do is make sure that the now double-length rails are supported all the way along their length. I'd only do this to expand the Y-axis where you can easily add supports rather than the X-axis where you cannot. -Jonathon Duerig
Thanks for all the advice and great look machine too! I did not realize that there would be that much of a difference in RP and Acme speeds. I had planned on mounting the Acme between a pair of tapered roller bearings so I could load it a bit (but didn't take the time to illustrate it along with quite few other things) but it looks like it doesn't really matter because it doesn't fit the end goal anyway. Any thoughts on using the SMW3D OX kit as a basis? I'm sure it will be enough to get me started, just not sure if I'd be better off designing from scratch or using it for a while and modifying as the need arises?
The Ox doesn't really seem rigid enough to be a great 4'x8' machine. And I'm not sure my own design would fare much better at that scale. As the spans get larger, belts stretch more and beams have to be bigger or more well-supported to maintain rigidity. On the other hand, if you start out with an existing design, you are much more likely to end up with a working router at the end of the day. And you will also gain a lot of insight into what exactly you want out of your CNC router. My first CNC was from a kit and although none of the pieces have made it into my latest design, I learned enough from it to design my own that fitted my needs perfectly. And I was able to later reuse most of the parts. So if you have the money and want to start at a 4'x8' machine, you might just get a CNC Router Parts kit. If you have less money and want to get a working machine that lets you figure out CNC machining, get an OX kit and recognize that you will probably have to modify it to make it stiffer if you want to scale it up. Designing your own 4'x4' or 4'x8' machine as a first attempt will be quite tricky, though. So going with an existing design is likely to be a smart move. It might also be fun to try to make yourself a Maslow. It is likely to have less precision and be harder to use as a 'production' machine, but the plans are out there and it is much cheaper than the alternatives to try it out. Most of the investment would be your time.
