While I definately agree, that in most scenarios, it's better to remake something that already exists, as a.saji said, there are also many cases where it's better to totally remake something. With this design, for example, the X axis has very little connection space to Y along the Y direction, which allows it to tilt easily, and mostly, the Z axis is small, insecure, and can only hold light tools. While it could be redesigned, it would be much faster and easier to compleatley remake it, from the ground up, based on what I learned from making this. I have revised and redesigned quite a bit on this one already as well. Originally, it used a triangle side, similar to a reprap, and an X axis that moved the Z. Then, it was revised similar to the current version, but with no steel rods to act as guides, then a weird Z axis, and so on. This revision gave me the best performance I could, with my time and resources, obtain from this general design.
@William: Yes in a way but I will not agree on that completely since it all depends on what the requirements is. If the requirement needs modifications and does not cost much, then its always good to edit the existing but if the requirement is huge and costly then should go for a new one indeed.
I do agree that it's possible, just way more difficult and time-consuming than it would be worth in this case. With this design, other factors will compromise accuracy far more than threaded rod being worn down by nuts, or manufacturing tolerances in it. It's far from perfect in either case, but just isn't a significant factor in this specific design. I wouldn't want to ask the people reading this to contribute parts for projects, because for one, I get paid way too much for these in the first place, and if I end up not finishing a project or it's not approved by Design News, (though not finishing it is more likely, I have at least two bins of semi-abandoned projects at any given time). Also, I am waiting a little while to redesign or rebuild the mill, at least until my 3D printer gets here (a makibox), as that will allow better tolerances and more complicaed parts, though only small ones.
Once make something more substantial, using the "right" materials otherwise, and with a stronger, more stable design, I will look more seriously into leadscrews and/or ballscrews.
John, it is indeed a challenge to align the shaft pieces accurately enough. No question about that. That is where he concrete fixturing comes into play. The supporting surface is made using one longer section of angle stock, giving you a groove with 45 degree sides and perfectly aligned. The gap for working on them is formed with modeling clay, regular clay, wax, or somcthing else to provide clearance below the rod sections when the casting is turned with the goove up. Then it is just a matter of setting the rods so that there is no thread offset. That can be done with a scale or a nut or really good eyes. I recommend using a scale to set the end to end gap. And if not welding, then brazing, which is a lot more like soldering. And any excess simply gets filed away. Not a simple method, but fairly cheap and it does not require much in the line of expensive resources.
And you may be able to send out an apeal to the Design News readers for a contribution of some materials. Presently I don't have access to any, but 25 years ago I could have got you a two foot section of leadscrew that was only bent a little. It can be straightened with lots of patience, wood blocks, and a medium hammer.
Actually, I've got two such motors from american science and surplus. I was originally going to use them to drive the standard threaded rod (remove/cut off the leadscrew), but I found better motors since then, and I'm working on milling functions with those. The leadscrews are only a few inches long, and I'm not sure if welding would yield a very straight or accurate leadscrew. Plus, I don't have a welder. With a precision jig, welder, lathe (there can't be any addition material from the weld in the screw itself) and a VERY straight saw, it might work, but in my case, threaded rod is really the only viable option.
I do plan to use the motors with leadscrews, as you mentioned, for a smaller mill, probably a CNC for making circuit boards or something. That requires a similar setup to the milling function motors, which I'm still working on.
And while reusing things that would otherwise be thrown away (like the seat motors) is great, I often find that projects which call for something that's "easy to find at a local junkyard/surplus store/thrift store" rather annoying when its something rarely thrown away, or found at such stores. For personal projects, I use these sorts of materials all the time, but when I put something online (here or on dedicated how-to sites), I try to use only materials avaliable retail, even if its slightly sub-optimal.
John, One handy source of acme threaded leadscrews is in automotive power seats. For many years the for-aft screw has been several inches long. If you are able to weld pieces togather end to end that could be a reasonable choice, although the nuts nay be a bit loose, a preloaded pair could solve that problem as well. An added bene fit is that the motor drives are alredy assembled to the screws, at least in some cases.
Of course the validity of this suggestion depends a lot on the price of used seat bases.
Actually, some scrapyards may yield really useful treasures in the areas of parts. And adding recycling to the list of a projects values is always a positive thing.
I did originally considder using concrete, as I read about that somewhere, though I believe it was actually used as recently as WWII for mills and lathes near the front, so they could quickly and cheaply make machines for repairing guns and the like. I ended up not using it because it requires forming/molding, which would be significantly more difficult than just cutting some 2x4s. I did considder that threaded rod is definately not the best option, a leadscrew designed for this obviously is. The problem is, they're fairly expensive, and this was solely for the purpose of having something that requires no unusual/salvaged parts, and costs as little as possible.
The accuracy is good, but only if the material is cut very, very slowly, though the precision is (as seen in the video) very good, even for a "real" mill. The arduino is used to read rotary encoders on each axis, and display that information on a tv taped to the top of the mill. That is why the mill is very precise, though that precision is compromised under any noticable side load.
toolman, while improving the accuracy would be a challenge, and using aluminum extrusions would be one option, there does exist an alternative that I have heard about. It was used on some older Chinese machinery, which is making the framework out of good concrete, with steel inserts to accept the bolts and support the bearings. The equipment that my acquaintance had seen had probably been produced in the i900 to 1910 era, and it was still usable around the late 1930's. A similar technique has been used for some machines made in the USA, although those machines had a much larger proportion of much better quality steel. The other very real challenge to maintaining accuracy is the threaded rods, since that form of thread wears faster, and the normal threaded rod stock is quite ssoft steel.
I did not grasp where the arduino came into the picture, but it was quite an interesting thing to see the claimed accuracy and then the video showing the wider cut due to vibration. Resolution does not equal accuracy or repeatability.
But the project does show a great deal of initiative and creativity, and a lot of determination.
A good source of the various extrusions would be industrial surplus houses and scrap dealers. These may also have complete assemblies available as well. And pricing is usually closer to the scrap value of the material than the new price.
In many engineering workplaces, there’s a generational conflict between recent engineering graduates and older, more experienced engineers. However, a recent study published in the psychology journal Cognition suggests that both may have something to learn from another group: 4 year olds.
Conventional wisdom holds that MIT, Cal Tech, and Stanford are three of the country’s best undergraduate engineering schools. Unfortunately, when conventional wisdom visits the topic of best engineering schools, it too often leaves out some of the most distinguished programs that don’t happen to offer PhD-level degrees.
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.