Introduction: Low Cost DIY CNC Router - Part Deux
OK, here it is, an update to my hobby CNC router build. As good as the initial design was, I started thinking of ways to improve it not to long after I started using it. Another driving factor in this rebuild is that I continue to receive requests for a plans set. I really didn't want to do plans of the initial design because there really wasn't a plans set to begin with, and secondly, IMO the design was a bit to complicated for what I thought a low cost easy to build DIY router should be. With this rebuild, I will try to rectifying some of that.
- Simplify and improve the Z axis motion
- Remove any metal machining requirements
- Use a less fiddly v-bearing adjustment setup
- Standardize on one thickness of plywood for all parts
- Develop a plans set along the way
The main objective of the rebuild is to simplify the z-axis motion. This is the motion that moves the router up and down. In the initial design, I built a cool system where the entire gantry was moving up and down. The design required two separate lift mechanisms and motors, one on each end of the gantry. This worked OK but it turned out to be slow and occasionally one side could go out of alignment of the other. It was also more complex than it really needed to be. With this rebuild I am replacing that setup with a more traditional z-axis mechanism that is located on the router carriage (with my own design twists of course).
The first design required some machining here and there (mainly lathe work). This was because I was adapting material that I already had in stock. With this rebuild, I am standardizing on 12mm bearings and shafting across the board, which has removed any machining requirements.
In the first design, I utilized the elastic properties of wood and came up with an easy and robust system of tightening the v-bearings onto the angle iron linear rails. This system works great but it is fiddly in its implementation in that you have to precisely align the positions of the v-bearings to the particular angle iron for that rail. This required mounting the top v-bearings of the assembly and then placing the angle iron and lower v-bearings in position and then drill the mounting holes for the lower v-bearings. Not that simple. In addition, the method doesn't have a large range of motion which may cause issues with different manufacturers angle iron offerings (not all angle irons are created equal). To address this, I've come up with a cam system that has a 3/8" adjustment range which will be more adaptable.
In this redesign, I am doing all the plywood parts in 3/4" material. In the first build, most of the major components were built up from two layers of 1/2" furniture grade birch plywood. Furniture plywood is great stuff. It's made from many layers of birch hardwood. It's strong, stable, and machines nicely. The problem? It's most likely not sitting on the shelf of your local home store. To simplify a design also means to simplify material procurement. So I decided to stick with plywood I could get at the big box home centers. My plywood type of choice is hardwood plywood with as many ply's as possible. I guess you could build a precise CNC router out of soft pine plywood but hardwood ply is what you want for a build like this. In the home center plywood aisle, there's not much in 1/2" that's worth considering so I jumped up to 3/4" which had more and better offerings. For this build, I found a 3/4" 7-ply luan plywood that comes in easy to transport 2x4' precut sheets. I used this 3/4 luan in the gantry of the initial build and it's working well so I'm going with that across the board in this rebuild. The luan is nice because it light, straight, and void free, but you can use any 3/4" material. The more ply's the better obviously.
Plans set? Yes, I'm developing and working on drawings before I make a cut. Hopefully and eventually, these will be offered as a plans set.
Step 1: The New Carriage
Out with the old, in with the new.
So the idea here is to scrap the old gantry dual lift system and replace it with a single z-axis lift located on the router carriage. This simplifies the gantry end plates but the carriage? Yep, we are adding complexity there but I think I've come up with a simple and elegant solution. To make room for the z-axis lift, the x-axis (left and right motion) motor assembly is moved behind the gantry in a separate module. On the z-axis lift, I'm really proud of what I came up with. There are only five major components, the spindle clamps, the two angle iron rails, and the top bearing plate.
Router/spindle mount: Note that I'm sticking with the v-bearing and angle iron rail solution. Thinking on how to simplify this setup, I figured I could use the angles as a structural component of this assembly. The two angles are secured in place by the router compression clamps and the top bearing plate. I designed this assembly so that the angles are automatically and accurately positioned during assembly. The jack/lifting screw is a 10mm dia. 2mm pitch acme threaded rod with a matching bronze acme nut commonly available from CNC parts suppliers. Inside the the top bearing plate is a 6200-2RS 10mm deep groove ball bearing good for a 250 pound thrust load. The threaded rod continues up and is attached to the z-axis motor via a 10mm to 1/4" clamping shaft adapter. I am using no flexible coupling here. The z-axis motor is solidly clamped to the jack screw. The jackscrew is the only thing holding that motor up. To keep the motor body from spinning, there are two screws that run up and snuggly fit into holes in the motor plate. This setup gives a zero backlash between the motor and jackscrew and leaves enough give in the mount to allow the motor to spin freely. It also has the benefit of a lower parts count and less fiddly setup of the motor mounting.
Carriage plate: The carriage plate is glued up from two layers of 3/4" plywood. This makes for an assembly that is 1 1/2" thick. Pound for pound, plywood is stronger than steel. When you have a 1 1/2" thick piece of it, you have one strong piece of material that makes for a decent hobby grade CNC part. Another advantage of such a thick base plate, is that the 3/8" v-bearing mounting bolts are held very rigidly in place. The thick plate also gave me the space to imbed the v-bearing cam adjustors and a recess for the z-axis jackscrew into the meat of the carriage plate.
V-bearing adjustment: Not pictured all that well (but highlighted in the videos), I came up with an eccentric cam design that I was able to machine easily (can be handmade also) and assembled from some threaded rod and nuts. The adjustor gives an adjustment range of around 3/8". This new design is an improvement in that it's easier to adjust the v-bearings and the design lends itself better to a parts kit (if I ever want to offer one).
X-axis motor module: To make room for the z-axis mechanics required moving the x-axis motor and reduction out of the way. I accomplished this by relocating it to behind the gantry as a separate module. The module is mounted to two of the carriage v-bearing mounting bolts that were extended to behind the gantry. The motor module is designed so that it can be accurately positioned to the x-axis roller chain independently from the router carriage. I am using the same 3.55 reduction ratio as in the original design. I know a reduction adds complexity but in this case, it is worth it. With the micro-stepping set to eight, this router has a resolution of .0005" and with the new 560oz Nema24's motors, this router doesn't stop for anything. I can easily make single pass cuts in 3/4" material at very reasonable speeds and accuracy for a plywood hobby cnc machine. Twice so far, I've miss-judged locations of stock hold down screws and managed to mill chunks of screw top off without the router skipping a beat.
Dust collection: If you build a cnc router, do yourself (and your lungs) a favor. Your first project should be a decent dust collection setup. The dust collector on the Mk1 version worked very well and was quite unique. Since the entire gantry and router carriage assembly was moving up and down, I came up with an articulated design that always kept the dust shoe on top of the work piece. That setup worked wonderfully and captured 99.999 percent of the dust and shavings (ok, maybe 98% ;-). I didn't need the articulation on the rebuild as the gantry and router carriage are now fixed in the z-axis direction. So I have a new design, based on the old design but using a snap on/off setup built out of pvc pipe. Take a close look at the pics and you can figure out how it's built. I have a section of 1 1/2" pvc that the 1 1/4" pvc of the dust collector piece snaps into. So far, the setup is working as well as the original and it's easy to adjust up and down, and to remove for bit changes.
Step 2: Assembly!
I didn't take any assembly pics of the rebuild. What I did do was video it! These are my first goes at the ubiquitous YouTube how-to video. I hope you like them!
Step 3: The Next Step?
Now that I've improved the router carriage and z-axis assemblies. The next logical step is new side plates and gantry beam. I'm starting to work on that and who knows? Maybe in less than a years time, I'll have something to post?
If you made it this far, I thank you for putting up with my ramblings.
Participated in the