—Comment by AmericanFabricator on video showing a CNC machine milling a part.
To make these rod logic contraptions out of brass, I figured it would be a good idea to have a CNC mill so that I could precision-cut whatever I needed. A “real” CNC machine is too expensive, so I decided to go the conversion route.
CNC Fusion sells conversion kits to convert Sieg X1, X2, or X3 mills to CNC. Sieg is a Chinese manufacturer of machines, and the X1, X2 and X3 mills are rebranded, renamed, and sold elsewhere, for example by Harbor Freight and Grizzly. The low prices and good quality of these mills make them a target for customization, and there are a few business devoted to converting the mills to CNC control (Syil America being another).
X1 is called a Micro-Mill, X2 is a Mini-Mill, and X3 is a Small Mill. I decided to go with the X2 from Harbor Freight, SKU 44991, called in their catalog the “Two-Speed Variable Bench Mill/Drill Machine”, which is why Google and good search-fu is your friend. I have a Harbor Freight under half an hour from my house, so that would save on shipping costs. They didn’t have it in stock, so I special-ordered it and it was in two weeks later. I took it home in its big wooden shipping crate, which I delighted in opening with a crowbar. I half-expected a cursed antiquity to be inside.
I decided to go with CNC Fusion for a conversion kit because they seemed to be more hobbyist-oriented than Syil. The disadvantage is that I would have to purchase motors, a controller for the motors, and software separately.
For the motors and controller, I went with HobbyCNC’s 3-axis Pro driver board package with 305 oz-in stepper motors. Sadly, they don’t include a case or an AC transformer, so I had to buy those separately.
So where are we at?
|X2 Mill with 20% coupon! w00t!
In Delaware! No sales tax! Pwned!
|$ 391||$ 0|
|HobbyCNC Pro driver package||280||16|
|Case from Mouser||31||11|
|Transformer from Allied Electronics||63||14|
|CNC Fusion Kit #4 with ball screw and preloaded ballnuts||679||0|
|Total with shipping||$ 1485|
And then there was building and installing everything. The motor driver assembly went very quickly and without any issues. I downloaded a trial version of Mach 3 CNC software, and was able to successfully control the motors from my PC.
Next, the X2 conversion. I found an excellent set of videos on YouTube produced by Richard (quadrant2005) in the UK to use as a guide. All the axes installed without a problem. However, I found that upon controlling the axes via computer, the screws tended to stick in places and the ballnuts sometimes had a grinding sound. This is an indication that the ball screws are not exactly aligned.
For the X and Y axes, I loosened all the supporting screws and slowly tightened each screw slightly, in round-robin fashion, one after the other, rotating the axis by hand so that the screws were as tight as they could be without causing any sticking or grinding. That worked. I was able to measure the backlash of each axis to be at most 0.0005 inches, which was my goal. It sounds easy, but it took several hours to tighten, rotate, tighten, rotate, loosen, rotate, until I finished.
The Z axis, however, was another story. I could not get it aligned no matter what I did, and the grinding in the ballnuts sounded more serious than on the other axes. I removed the screw assembly entirely, and there was no grinding, so I knew the screw was good. Once back on the machine, the grinding began again. No amount of fiddling helped (and I spent about eight hours at it spaced over several days). This meant that there was some fundamental flaw in the way the assembly was mounted to the mill.
I had heard it was possible that because of low quality control at the manufacturer, the mill’s dimensions are sometimes not consistent. I suspected this could be the case, since two of the screws supplied by CNC Fusion were too long, indicating that the holes in my mill were not drilled as deep as expected (I hacksawed a bit off the end of each screw, which worked).
So I measured the mounting holes on the mill, and compared them to the mounting holes on the Z axis assembly. I found that one of the holes was misaligned in one axis by 0.05 inches. Now, this doesn’t sound like much. It misaligns the screw by around 0.2 degrees. But over the course of a 12-inch screw, that adds up to 0.05 inches that the screw presses into the ballnut, which could cause grinding and sticking.
That’s not to say that something else might be wrong as well.
Cursed antiquity, indeed.
The kind folks at CNC Fusion promised me that they would help me make the Z axis work, even unto making a new assembly to fit my mill. That is very classy of them, and I commend them for it. The only downside is that they are a very small family-run business, and they are very busy with day jobs, so sometimes it can take some time to respond. But I’m patient. I felt panicky when I first came to the realization that this thing just wouldn’t fit, but not anymore.
More updates after I get the modified Z axis assembly.