CNC Router v1.0, The Project

CAD output for the CNC router project.

For a while I’ve been thinking about building a CNC machine from scratch. Not that I could not find one that I wanted, I just wanted to build one. It was just an idea. Until I bumped into Graham Ham’s web site and his CNC router. His postings of the entire process were so well detailed that it inspired me to get started (not to mention to clean up my workshop as it usually looks like a cyclone went through it when compared to his). And started I did. My design is loosely based on his design, taking in consideration my abilities and work limitations and the sizes of the main components I had and/or could find.

The main limitation, aside from the fact 2 years ago I thought milling machines were used to turn wheat into flour, is the size of stock material I can machine. My current CNC mill is a miniature Sherline mill (where it all started) and even with some modifications, which allowed for extended X and Y travel, it cannot handle widths larger than 140mm. I have a larger manual mill but some of the parts will have to be done through numeric control. Not to mention that even the larger mill, a Sieg SX3 (labeled Grizzly G0619) is also limited at 150mm of Y travel. This is all after all, just a hobby setup in my basement. This machining limitation is obvious when you see the compound side posts that hold the gantry. I just could not machine one single 250mm x 390mm piece of aluminium plate. I had to make it using pieces no wider than 140mm.

Stone age meets high-tech. Ball screws and linear rails wrapped on newspaper and hand-woven sisal rope.

Following Graham’s ideas, I found and ordered a set of Chinese ball screws and linear rails off eBay. I bought it from an eBay seller called “Solar Jean”, who sells an interesting combination of nail art stickers and CNC machine parts. I fail to find the relationship but I assume there is one in there somewhere. The items arrived in less than a week and it all were in fine shape. One thing to keep in mind, and a lesson I learned, don’t trust engineering drawings from these Chinese components. I had started my project based on these “documented” dimensions just to find they were all quite off once the parts arrived. The parts are OK and overall well done. Just not to spec. I had to measure everything and adjust all the plans accordingly. The fact that the parts arrived wrapped with hand-woven sisal rope should tell you something.

If the link above no longer finds Solar Jean, you can always look for “ballscrew set” directly on eBay and see what comes up. These are pre-machined sets of ball screws and matching linear rails. Makes life quite easier than trying to source these things individually. But that also means your machine’s size will depend on the combination of sizes you find. If you want something specific, you can always email the vendor and they will cut and machine the parts to your specification.

Speaking of plans, I designed the entire machine using a solid modeling tool before cutting the first piece of metal. Software, not hardware, is my background. This allowed for a very accurate design with very tight tolerances. It also allowed to experiment. Using the solid modeling program’s constraints and analysis tools, I was able to see what would or would not work before any real machining started.

This is the beginning. In the following weeks I will try to describe the process through which I will go to get this project going. I chose this “blog” format because it was really simple to install and maintain. Let’s see where this will go. And to Graham, thank you!

Comments are closed.