New CNC router

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.


Well-Known Member
HMEM Supporting Member
Jul 8, 2009
Reaction score
For the last couple of years I have been playing with the idea of designing a CNC router based on a Bosch Colt compact router, because I happen to already have the router. I wanted a CNC router that could handle wood, plywood, plastic - within reason, and MAYBE non-ferrous metals. I was initially thinking of making an MPCNC, for those of you that are unfamiliar with this that stands for Mostly Printed CNC. I liked some of the ideas he had when he designed this thing but I always thought that the Z-axis set-up was a little iffy. It works just fine, but there seems to be so many ways it can get out of tram.

I really liked his guide rail idea, so I started there and worked on my design. My idea was to design a CNC router that would initially have the conduit guide rails, ala MPCNC, some 3D printed parts and some plywood parts., that was about 2 years ago. About a year ago I got to the point in the design where I needed to make some decisions so that I could finalize the design. My initial desire was to make the working volume 48"x24"x6", for you metrically inclined folk that's about 600mmx1200mmx150mm. That never changed, so that's the working volume.

I waffled between NEMA23 motors or NEMA34 motors for some time. I decided that at this size, and with my intended use, I would NEVER need to use "Ludicrous" speed on this machine and I could manage with lower cutting loads - so NEMA23's won out. With the motors decided on, that allowed me to pick the stepper drivers. I didn't buy the cheapest Chineesium's that I could find, but I've still all my arms and legs - so I didn't go the other way either.

How do I make it move, belt drive or lead-screw? While belt drives are faster, they've got their own problems, mainly that I couldn't find the parts that I wanted in the size that I needed. Remember I'm not trying for "Ludicrous" speed so a lead screw is doable. So now we're up to the ball-screw or Acme screw decision... HOLY CRAP they want how much for that length, Acme screw it is. And, I happened to find 3/8"-16 Acme screws in 6 foot lengths for a very reasonable price. I know, I know, you're going to say that small of a diameter lead-screw at that length is going to whip and I agree - under normal conditions. However if you keep that screw under tension the whip is drastically reduced. There's a video on YouTube showing an 8mm lead-screw that is 1.5m in length, both before and after being put in tension. My lead-screws are designed to be in tension.

The CNC controller opened up a whole 'nuther can of worms. What do I go with, Mach something, Linux, Acorn, GRBL, something else? I asked what other people were using. I eventually settled on using a DDCS v3.4 standalone 4 axis controller, because it's small and could easily be transferred between any CNC machines that I may build. It didn't set me back THAT much, so if I screwed up I won't regret it TOO much.

I've got the 3D design done, and the 3D printed parts have been printed, all 151 hours of them. The Y&Z axis combo is 4 big parts that took about 40 hours of combined printing time. Right now I'm working on the 2D drawings for the bed, and the gantry sides. My 3D software isn't real friendly when it comes to sharing with other drawing packages, or printing 2D layouts - it's easier just to re-draw them.

I've got a lot of other crap going on, so this will be an on again off again build. I'll try to upload a couple of screenshots of the 3D model and maybe a family shot of the printed parts. Everything else is just BIB now so it's not too interesting. (Bits Inna Box)

Last edited:
Hi Don.

Looking forward to your CNC build. I built a CNC router in 1999 with the same footprint you are talking about. I have a 3.5 HP router mounted on it. My lead screws on the X & Y are ultra-smooth .75 inch. I think using a .375 lead screw is a big mistake. Besides being wiggly, the working load on the drive nut will be very high. My lead screws are probably a bit heavy. At the very least i would use .5" and would say .625" would be perfect. I am still using the original drive nuts with no signs of wear.

Over the years I have replaced the bearings in my router (Porter Cable) three times and my brushes once. Its on the bench again now for new bearings and brushes. And I just received a SX2 mill head with R8 spindle that I will be adapting to my table in the coming weeks for nonferrous metals.

Mark T

I realize that most of the decisions I made have been kind of on the "Lightweight" side. I'm looking at this thing as being a permanent work in progress, where no part of it is such an investment that I'll regret throwing it out to try something better.

I've attached a couple of files, the first is a screen-print on the 3D model, the second is a 3D PDF of the model - which a lot of people won't be able to open for some reason. I think it has to do with the security settings on the computer.



  • CNC.jpg
    76.5 KB · Views: 317
  • CNC.pdf
    1 MB · Views: 346
I can open it (Adobe Reader DC) after allowing 3D for the document.
It is a bit too much left-centered, so zooming into details is not always possible as the part is leaving the frame.
- found it - holding left AND right mouse button TOGETHER shifts the drawing wherever you want.
Thank you!

Building a CNC router is not overly difficult however, you need to decide early on just what you intend to use it for and then design around that requirement. I built mine to machine everything from mild steel, aluminium, brass, plastic and I have also cut carbon fibre. I have included a couple of photos and you will note that the gantry is made from 3/4 inch plate aluminium with lots of bolts holding together. After five years of use cutting the above materials, the only change I am making is to replace the spindle motor with one that can be programmed to run at a slower speed e.g. 500 to about 1500 RPM. To this end, I am trying out a motor salvaged from a mobility scooter which is about 400 watts. My current spindle is 1500 watts not by design as it was given to me.

The CNC control using Mach 3 is quite straight forward and the drivers are quite cheap with plenty of info around. I have converted a lathe as well as several other machines to CNC with out any problems.

Below is a general view of the router. Photo taken during the build.
Screen Shot 2020-09-01 at 11.48.16 AM.png

Cutting some 6mm mild steel.

Another view of the current spindle which has water cooling.


There are many ways to go about the build and everyone has their own ideas however, if you have a question, you only have to ask. My router has been the most useful machine that I have built for the workshop and is in regular use. I machine mainly aluminium altho just recently there have been several cast iron components for my Fairbanks Morse R80 engine.

Good luck with the build.

Bruce Weir-Smith
Western Australia
Bruce - That is one seriously beefy router that you built. My planned main use for this will be cutting wood and plastic. I wanted the working area to be big enough that I could use this router to cut the parts for the next generation of this router. If I can convince it to cut aluminum, then it will eventually start looking like your router, even though I'd be more tempted to call yours a mill.

Brian - I liked the MPCNC guide rail design, so I started my design there. Other than that, this thing is a conglomeration of the bits and pieces that I liked best from just about every CNC router build I watched on YouTube. It's not built yet, but I've got all the parts printed that I need. And most to the other parts have been gathered, I think I still need a power supply or two - and some more bearings for the guide rails, and some 4 conductor shielded cable for the motor leads, and.....


Latest posts