My path into CNC.
#4
I have related bits and pieces of my limited experience with CNC and thought some of y’all might benefit from pieces of it.

About 10 years ago, I bought a bench top manual mill from Grizzly with the intention of converting it to CNC, but as luck would have it, exactly the same model CNC mill came available from a private seller nearby for just a little more than I thought it would cost to convert the one I had, so I jumped in the truck and brought it home.

At that time, the software offerings for CAD were either expensive, or limited or both.  Also then, you needed to buy separate CAM software to generate G-code and then not all the CAD was compatible with all the CAM and very few of the CAM packages had post processors to generate g-code that was compatible with Mach3(the control software on my machine)

I played around with Google sketchup and while it was quite capable of drawing well, I used it infrequently enough to require relearning with each episode.  Then getting it to play nicely with CAMBAM and generate g-code that then still needed to be hand edited to some extent, I found myself defaulting to using the manual machine and the CNC mill just sat for years.

About 2 years ago, I decided to sell the CNC mill off and regain some space, and off it went to a good home.

Not long after I came across an estate sale that listed a CNC router. 
This router was a shapeoko XL that was in running condition.  While I was at the estate sale, I also bought a Piranha CNC that was also in good running condition.  Laying off in the floor, was a dewalt trim router with what appeared to be the broken gantry from an x-carve.  For $20 it came home with me, with plans to use the router and trash the broken Gantry.

Home I came with my haul.  I found that the carbide create software to draw 2d from shapeoko was not only easy to use, it also had lots of tutorials on youtube from the likes of Myers Woodshop.
Not only was it easy to draw, now the software that you drew in also functioned as the CAM and generates the g-code as well.  Also their software has baseline feeds and speeds defaulted in.  This is another great benefit to a beginner.

At home, I found the pirhana to be less than Ideal due to the need for CAD that generated post processed g-code compliant with its needs, so I sold it for a little more than I paid for it.

The shapeoko got some use, and I found myself looking at the broken gantry.  I looked up x-carves website and found that having the gantry end plates,  entire Z axis and all motors was enough to start a build.

I ordered a few hundred dollars worth of parts from x-carve, but stopped short on buying their controller.

I found enough information on youtube and some forums to build a controller based on 4 tp6600 motor controllers (cost about $10 each), and arduino uno, 36v power supply and downloaded the free open source GRBL software to run the Arduino Uno.  So for about $100 I had a more robust controller than the x-carve ships with.
Universal G Code Sender (also free) was great to interrogate and tune GRBL on the arduino Uno and it was a great G code sender as well.  

I like to tinker, so when I saw another x-carve 1000 without controller for a few hundred dollars, I bought it as well, built a controller and sold it for a few hundred more than I had into it.
Another x-carve found its way into my hands without the controller so I did the same with it.
(I am not addicted, I recently passed on one locally and put someone else onto it)

By this time, I was tinkering with cutting aluminum with the x-carve and the shapeoko wasn’t getting much use due to my liking universal g-code sender better than their proprietary g-code sender( could have probably used UGS with the shapeoko as it’s GRBL based as well, but I never tried it.

I got the hankering for a more robust machine.
I saw locally a 4x4 plasma cam that was advertised as not working.
I knew when I picked it up that I wasn’t likely to be able to repair its control system, but though I could probably use it as at least a table and basis to build a router.

Once it was home, I decided I was okay with its bearings on steel for the Y axis but I wanted something better than that for the X and Z.  I bought a ready made Z axis with linear bearings and ball screw for the Z but the X required some fabrication.  I mounted a 1/4” thick x 6” plate the length of the X axis and mounted 2 linear slides on the front of that.  I used the x-carves to make motor mounts to adapt Nema 23 closed loop stepper motors to drive it and had a functioning router.  It needed some tuning with additional bearings to take out play in the Y axis, but with rack and pinion drives, it works quite well. 

While the plasma cam conversion was in progress, I convinced myself to order the ball screw conversion for the bench top mill that ‘started this mess’ as my wife says.  

The Bench top mill got its conversion and now has closed loop steppers and Mach3 driving it as well.  It proved to be very useful in drilling very well aligned holes for mounting plates for things like the Z axis to the linear bearing trucks.  Lining up and drilling 16-20 holes in precise locations really is reason enough to have CNC in your shop.

During my conversion of the plasmacam to router, I had also purchased some 80/20 that I had planned to use to modify the Y rails on the plasma cam.

My mother used to say that if I found a button, I would sew a vest onto it.  True to that, I decided that rather than modify a machine that was ‘good enough’ I would use the 80/20 and go ahead and build another machine significantly more robust than anything I already had.

So I built a table based largely on Avid CNC’s design and used their motor mounts for nema 34 motors.
For that machine I decided to try the centroid acorn controller, its definitely a very capable controller and was easier to setup than Mach3, but I really don’t like the inability to probe for the corner of a workpiece without buying an additional extension for the software (maybe this function is available, I just haven’t found it). One of the benefits of Mach3 is that it has been around in the hobby market long enough, that there is a tutorial for just about anything you would want to do and macros available for probing.  
I don’t have any plans to build another router although I may change the controller on that one out at some point if I can’t find the functions I want with Centroid Acorn.

Once that machine was up and running, I bought the metal to build a Plasma table.  For that one, I am using a UCCNC AXBB-E combined ethernet controller and breakout board along with their software to run the g-code.  I really like it and will probably convert the mill to that controller and software after I have gained some more familiarity with it on the plasma.
I have a torch height controller on order from Price in Ireland and will need to upgrade the plasma cutter from the Hobart 27i thats currently on there to a machine that has a CNC port (for the torch height control) soon.

I like woodworking, and have built several things over the years, all the way from furniture to cabinets, but what I really like to do, is tinker with tools.  I am much happier buying a jointer that doesn’t work and getting the motor up and running than buying a brand new jointer, or buying a used Sawstop and making my own T-square fence to go with it.

In the last couple years, the count, through the shop, is:
One bench top mill CNC sold
One pirhana CNC Sold
One Shapeoko XL sold
3 x-carves sold
1 plasmacam converted to CNC router currently for sale locally.

1 4x4 CNC router Table based on Avidcnc design no plans to sell it
1 4x4 CNC plasma table no plans to sell it either
1 Bench top CNC mill with no plans to sell it either
Also have a metal working lathe but don’t plan to CNC it.

What I have found, is that CNC is a slippery slope, and it is certainly significantly more attainable today than it was as recently as 8-10 years ago.  With free software like Carbide create, Universal G code sender, and GRBL on an arduino uno, you can build a very low cost controller that is capable of doing great work.  Pair that will a platform like the x-carve with its belt drive, or for more money a ball screw. 
For about the cost of a sawstop cabinet saw, you can be into an entry level CNC.

For a little more somewhere between 4500 and 5000 you can build a 4x4 that would be at home in a production environment running rapids fast enough to break 1/4” bits if your not careful.

The challenging part, is that you almost need a CNC to build a CNC due to the precision of some of the holes and their alignment.

Hopefully my ramblings will be of some benefit if your thinking about getting into CNC.

I am certainly not an expert on CNC either programming or building, but I am evidence that if you can work wood, you can probably build and run a CNC.

Duke
Reply
#5
Thanks for the read.
Very diverse place we have here on Woodnet.
Reply
#6
That's quite a story.  I'm working on my cnc mill right now.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)

Product Recommendations

Here are some supplies and tools we find essential in our everyday work around the shop. We may receive a commission from sales referred by our links; however, we have carefully selected these products for their usefulness and quality.