(no subject)

[fivemack]

I've recently been trying some woodwork. If I want a cube of wood 33 millimetres on a side, I buy a stick of wood hat has been planed down to 33 millimeters on a side with square corners, I set up a mitre saw, I make two pencil marks 33mm apart on the edge of the wood, put the wood on a flat table and extend the marks across the side of the wood using a square resting on the table. I make sure the blade is in the same position relative to the mark each time I cut. I clamp the bar of wood to the back of the saw housing in two places, and I cut twice. If I've been careful the cube is 33mm rather than 34mm.

I believe that a competent metalworker could make a metal cube 33.00 millimeters on a side, I've read about Fabry-Perot etalons which are essentially glass blocks with one dimension precisely 33.0000 millimeters. Where would I look for information at the boringly-detailed level of accuracy that I gave for woodwork of how these extra orders of accuracy are obtained?

I suppose I'm looking for information at a grade that would satisfy a six-year-old's sense of recursive questioning - yes, you measure it with a micrometer, but how did you make the micrometer and ensure it was accurate. I guess this is a one-term module taken in the first year of a mechanical engineering degree, but what's the best textbook for that course?

Comments

[randwolf.livejournal.com]

I'm confused. Are you asking about wood, or about precision in some other materials?

[fivemack.livejournal.com]

I'm asking about precision in general; I have the dim belief that, if your tolerance is much under a millimetre, wood is not the material to work in because it's isotropic and has complicated microstructure in the way that brass doesn't. I'd have thought that asking for a wooden cube 33.0 millimeters on a side would be vastly inconvenient and 33.00 millimetres actually impossible, whilst the same cube in brass or in hardened steel would be at least possible if quite expensive. Whenever I read of sub-micron precision, the material seems to be glass or some exotic ceramic.

The particular bits I know that I don't know about are how you hold things still when machining them, and how you would set the bed of the milling machine to be 33.2 millimetres below the bottom of the face-mill held in the chuck of the milling machine, but I am sure there are also areas where I don't know that I'm ignorant.

[randwolf.livejournal.com]

Oh, I see. Now, this is a designer talking, rather than a machinist, so don't take this for gospel.

Fine-grained hardwoods can be shaped to .3mm or so, and it's not that hard, but I think even hardwoods expand and shrink a bit with humidity, so you'd have to be careful how you used the finished part. In general I would mark, do a test cut so that I could see how much material cutting would lose, cut outside the line, and fine-sand to finish. At that level of precision, an accurate ruler and a magnifying glass is enough, though you might want calipers and a micrometer.

33.00 mm is probably pretty easy in hard metal, but I'm not sure about brass--it might not hold the shape; bronze is harder and might do better. At that level of precision, I believe micrometers and calipers are still used, but I think optical technologies have become more common. Machinists and engineers handbooks give information about cutting precision for various tools--the handbooks will tell you, for instance, how much additional material a particular bit or laser-cutting technique will remove.

At the highest levels of precision, optical interferometry is a typical method of measurement. I think the reasons ceramics are used for high precision is that they're easier to work than metals, they hold shapes slightly better, and they have low coefficients of thermal expansion.

[mobbsy.livejournal.com]

Searching Amazon for the word "machinist" yields many possibly useful books.

Looking less at mechanical engineering degrees, and more at what people doing NVQs use be helpful too.

[ellarien]

I have no knowledge of woodwork or metalwork, but the mention of cutting cubes of wood caught my attention, because my father made some attempts at that. Are you by any chance trying to make burr puzzles or soma cubes or 3d pentominoes?

[fivemack.livejournal.com]

I was initially thinking of making a Menger sponge (http://en.wikipedia.org/wiki/Menger_sponge), but that turns out to be particularly difficult; at the moment I've reached a frustrating stage in making what's basically a Greek-key design (eighteen 45-degree mitre joints to have a line wander around a cube in an interesting way).

Grandad made me a burr puzzle and some 3d pentominoes when I was very small, and I'm not quite sure where they are now, which is a pity.

[randwolf.livejournal.com]

Oh, dear. Problem made for additive technology, I think.