From seeing this and other things that you have posted, I should have realized that you would have thought of the extrusions. I am impressed by your knowledge and craftsmanship. Someday I am sure you will make a big mark on society. Well done.
I did originally look at using 8020 extrusion, as we use it frequently for prototypes on my robotics team. I only didn't use it because it is significantly more expensive, and has to be ordered either online or from places who usually sell huge orders o big companies, and will be unlikely to sell small quantities without significant markups. eBay and other auction sites sell it at a reasonable price, but it's still generally beyond the ~$7 I spent on steel rod and 2x4.
I do plan to make some significant upgrades at some point, like using extrusion or linear bearings.
I do agree that though the resolution and precision are fair, and the accuracy is what falls due to a lack of rigidity. I believe it would not be considered cnc, because though computerized and has a numerical readout, the actual operation is handled by a person. There are no steppers or servos other than the tool. Because the threaded rod produces a very strong an accurate motion, the precision is within a thousandth, though once a load is applied the accuray falls a bit.
I'm also not entirely sure what you mean by a Campbell story line though...
This is a very interesting and clever bit of work which I could never invent. However, I do not think you would have to go to steel for the rigidity you seek. We regularly build fixtures from aluminum extrusions that are ridgid and stable. They come with T-slots for which you can purchase clamping components and accessories to built very complex frame works.
Since this is actually CNC rather than human-skilled machining (in spite of the title), I can think of both computational and electronic corrections that could get at least CLOSE to that kind of precision (.001 with a wood frame would be nearly magical though, wouldn't it?). Feedback loops, appropriate sensors, and some skilled coding...
John Duffy, are you SURE you've not fallen through a wormhole from a '50's Heinlein or a '30's Campbell story line?
I've used it a bit, and noticed, especially at first, a lot of movement. Originally, there were no steel rods, only the angle aluminum. That did not work well, so I added the rods for support, though I hope to upgrade them to actual linear bearings (or at least brushings) and at larger diameter rod in any future version. The stiffness of most of the device is okay, but the improvised vise and Z axis generate most of the looseness. I do plan to make another version at some point that at least uses more precise parts and a better design, if not better materials.
Great job. I just wanted to clarify the impression recieved from the introduction.Stiffness is the key to accurate machining. You will realize what I'm talking about after you've used it some. Keep up the good work.
I see how it seemed that i meant this as a replacement, so I'm sorry for the misunderstanding. This really is not a replacement for a true mill so much as a way for people who cannot afford a well-constructed mill. In my other comment I mentioned making a smaller, sturdier mill as opposed to this design. This was made as it was for easy fabrication and building without accurate tools or strong materials. It was also designed for soft materials, and thus lower side loads.
So yes, it will have a low accuracy if you try milling anything beyond soft wood at low speed, but that's really what it's made for.
Good work, John. I sent this link to my 13 year old son. I think it's partly young folks like you (maybe you specifically, regarding LED flashlights:) making things like this that inspire my son to dabble in things mechanical and electrical. He restored a 50+ year old industrial air compressor earlier this year and his most recent project involved rewinding the secondary of a toroid transformer to get the 36 Volts he needed to power a 50W LED module.
He made videos of that and more and put it in his YouTube channel. Look for YT user XfmXz if you'd like to see what he's been up to.
I'm sure he's picked up an interest in tinkering from me, but I think seeing other teenagers doing cool stuff on their own inspires him and removes any reticence caused by the "I can't do that because I'm just a kid" kind of feeling.
I'm sorry to rain on the parade, but commentors have confused resolution for accuracy. With the construction there is no way to achieve 0.001 accuracy on the X & Y axis'. It is not stiff enough. Perhaps the mechanism totally unloaded could achieve that accuracy, but a mill involves side loads on the tool and it's supports in order to perform the task. I am still impressed with the iniative of building it. It is no substitute for a Bridgeport.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.