Thread: Homemade CNC Router

I get quite a few enquiries from this thread and others on this forum so there is definately interest in CNC builds in Australia. This thread has 22,000 hits so that in itself is an example of the interest.

On the CNCZone there is probably more Australian CNC builders per capita of any other country. I also think that the machines built are first rate and a credit to everyone that has overcome the supply issues we face here.

Collectively the Aussie builders have a wealth of knowledge and a willingness to help others out with builds and supplier links. Our friends in NZ also have an active interest in CNC and they would also benefit from a forum relative to our location. Even if you do not want to build a machine the price of CNC ready machines is reducing with a new mill from the Sieg factory just released. I predict it will not be long before commercial gantry woodworking machinery is within the reach of the hobbyist as well.

Bob (Crocky) asked the administrators and also offered to moderate a CNC forum but I haven't heard how this was accepted. Bob would do an excellent job. Neil of course has final say and I respect his decision on how he handes it. I feel confident that we could quickly fill a couple of threads with machine pictures and of the work we do.

It's amazing this thread has had no posts since mid November but in that time has had nearly 4,000 hits.

There definitely is a lot of interest out there in this subject.

Greg

Hay guys i am new to this forum and just starting my 4x8 machine, i paied for the plans from the internet and was very dissapointed but have made mods and think i am on to of it now. i am using 460oz/in steppers with dedicated drives ( 3 off) breakout board but will have to locate / build a power supply around 400 w at 48 volts. if any one want to share my infomation or plans i am happy to help out

Tony

Tony have you run into Rod yet?

I reckon it would be a good move to make sure you do.

Greg

Originally Posted by Greolt

Tony have you run into Rod yet?

I reckon it would be a good move to make sure you do.

Greg

Yes I would as well he got me started and I built one of his designs, the grand master of CNC Mentoring and tutorials

Hey Greg,

Amazingly your on here doing CNC of all things.

Attached is a picture of my proof of concept machine for those interested.


/Michael

Hi Tony,
I would be very interested to see a photo of your progress.
Just a heads up on the power supply I would be surprised if you need 400 watt power supply for 460oz steppers. That is more than 8 amps per motor.
You do not have to allow for all three motors drawing full power at the same time. You can get a switching power supply quite economically but building one using a toroidal is probably better.

Hi Michael,
It looks more than a proof of concept machine to me. A lot of work has gone in to that and I see you are using top quality components for motors and drivers.

Hi all,

I started reading this thread before I began my build. Now that I'm almost done I thought I'd post some happy snaps.

I have based my build on the Solsylva belt drive table: http://solsylva.com/cnc/belt_drive.shtml




The machine has the following travels: X=1640mm; Y=760mm; Z=140mm.

The original design was modeled and tweaked in a CAD program before building for two reasons:

1. The Solsylva plans have lots of words describing the components and drawings of discrete parts, but no assembly drawings. This drove me batty - I had no idea what component went where until I built a virtual table, then everything clicked into place

2. American stock and components specified in the plans made using Australian stock very difficult. Tweaking sizes made everything go smoothly. Mostly. I bought things like belts and pulleys from sdp-si.com and econobelt.com before I redesigned the machine. My X belts are too short because I got too ambitious with the X travel, and the Z belt is too short because the gantry beam is wider than specified in the plans. These are small inconveniences.

The table is constructed from 3mm steel angle, bolted together. It flexes, but not as much as you would think. The 20mm NB pipe rails on the X axis add quite a bit of stiffness. The table deck is 140x45mm lumber. 12mm gaps between planks act as T-slots.



A $50 Ozito plunge router seems to work pretty well. 20mm flexible electrical conduit keeps cables tucked away.



Ugly, ugly, ugly. But it works for the time being. Computer and controller sit on top of a cardboard TV box until I get going with my first plywood project: a little workstation to house the controller box and to put the laptop on (with a big red stop button).

The controller is the Xylotex 3-axis drive box with 425 oz.in. steppers. The computer is a second-hand Dell Latitude D600 laptop. I had doubts about whether this would be the right choice of computer - I'd read that the parallel ports in laptops weren't always up to the job. But the risk paid off because this setup seems to work really well. The Ozito router is plugged into a power board with switches behind the laptop and turned on and off from there.

The laptop runs the Ubuntu Linux live CD with EMC2 from http://www.linuxcnc.org
I can't recommend this setup highly enough. EMC2 and the Axis user interface are fantastic and uncluttered (unlike Mach3) and Ubuntu Linux is very, very nice, and very easy. And everything is free. Which is a definite bonus.



My first real project on this thing wasn't even a routing project - I used it as a mill and created a proper clamping bracket for the router out of left-over 6mm aluminium angle and plate. The original setup (in photo) used band clamps, but this wasn't too nice. I bought some 1/16" carbide end mills off ebay (bottom of picture), a 1/4" to 1/8" adaptor bushing from McJing tools (the little round thing above the 5 broken bits - these things are so bloody fragile , fired the thing up with a prayer and let it go for a couple of hours. The result can be seen in the pictures above: a neat clamp that holds the router tightly, and a place to put a vacuum cleaner nozzle too.

So, that's it. It's definitely not perfect and will always probably be a work in progress, but it works. Having the flexibility to cut small alminium parts as well as sheets of plywood and other large things on the same machine is good. Now all I need are some projects, and the enthusiasm to make them.

Mark

Hi Mark,
Like the way you have approached it. Every machine I look at there is something different and your belts and gapped planking stand out.
Job well done and I hope you get years of use out of it.

Mark,
Your machine looks great ... I am planning something similar, about 1500 x 900 and was looking at belt drive only for the X axis (Solsyla style) and screws on the other axes. Others have suggested rack and pinion in lieu of belt, so am currently weighing up the pros and cons (and cost!). How have you found the belt drive ... any stretch etc? When you are milling your aluminium, what pass depth are you using and feed rate ... I seem to be too agressive @0.2mm!
Cheers,
Alan

Rod: Thank you for the words of encouragement. Building the thing was a hobby in itself. I'd always wanted a CNC machine after using a machining centre at uni. I'll get years of use out of it when I figure out what to do with it

Alan: As far as I can tell, this belt drive setup works really well. I haven't investigated what sort of stretch I get out of the belts, but If you've ever seen an L-sized belt you'd see that they're pretty stiff. At a guess, I'd say that any cutting aggressive enough to stretch the belts any considerable distance would cause the table and aluminium components to flex mightily before then. There are two L-belts on the X axis (one on each side of the rails) and one wrapped around the gantry for the Y axis.

The belts and pulleys for all axes worked out to cost about US$335, including postage. I don't know how that would compare with a rack and pinion setup. It would have cost US$104 less (yes, 31% less) if sdp-si didn't have such a stupid system - they sent my package via UPS without quoting first. Buggers. And it only arrived about two days before the econobelt order, which the nice man sent via USPS for $13.50.

Milling aluminium with little bitty 1/16" cutters is a steep learning curve. Lucky these things only cost about $20 for 10 off ebay. To cut the clamp out of 6mm plate in the pictures above, I could just about get away with a feed of 250mm/min at a cut depth of 0.2mm. The band clamps didn't hold the router in very good alignment, which meant that going around corners (for the circular parts) put undue strain on the cutter. I snapped a couple before I learned to ease the speed back.

Maybe bigger cutters are the go from now on, in aluminium anyway. I've got a 1/4" end mill and an 8mm ball nose end mill for carving things (the Ozito router comes with an 8mm collet as well as the 1/4", which is handy).

Good luck with your build and your cutting!

Mark

Hi Gents,
having pawed thru the cnc zone site I came across this one. I am looking into building a machine with a view to expand to a 16 foot length to machine foam core for wind turbine blades. I build the big ones 2MW) for a living and would like to eventualy build my own much smaller unit. Am investigating construction, software, drivers, 3d design software etc. I worked at RMIT Aero for a few years and had a lot of exposure to CAD CAM etc. Now I just need to do it on my own budget. Any guidance greatly apprecated
Andy Ross

Hi Andy and welcome to the forum .

Anything is possible 16ft thats about 4m could it be done cant see why not for what your looking at doing, I have started small and have learnt a huge amount from the journey so far I will look to start to acquiring parts in the next couple of months for a larger build myself.

Feel free to start a new thread with your questions we all will try to help as much as we can we only ask that you do the same with others down the track.


Willy

Hi Andy,
16 foot is huge.
Cosidering the cost of rails, etc I think the only affordable (for me) construction would be the Mechmate design.
Try
www.mechmate.com
While I would encourage anyone to build a machine I think you may want to at least cost this type of work being contracted out. The software alone can run into thousands and maybe tens of thousands of dollars and unless you have heaps of work you might not break even. Anyway I am sure you will look into it or have already.

A machine like you see on this forum will run about $3,000 for linear rails and zero backlash drives. It used to be cheaper but steel has gone thorugh the roof lately and the second hand market for parts is being driven by an inceased demand in Hobby CNC machines.

Anyway welcome aboard and ask away. As Sean said probably better to start a new thread with a few of your ideas to kick around.