How did people make more precise instruments using less precise instruments?

107 viewsEngineeringOther

I was thinking about it for a while: The instruments to produce things and to measure things improve in precision as the technical progress goes on. For example modern machine tools allow us to produce things with tolerance of 1 micron or less. But this machine tool was made with something And that thing was made with something too. And since the tech level was lower back then you’d have to make a more precise machine tool using a less precise one. How is this possible?

In: Engineering

4 Answers

Anonymous 0 Comments

If you know the goal and doing something with hands you will get various result, ie an axle with tolerances +-1 mm. Some of them will be better, some of them worse. But some will be 0.1 mm. By hand picking you take better parts and build a more precise machine.

Anonymous 0 Comments

There are several ways to make precise things with less precise starting materials. You can make a flat surface using three rough starting surfaces, and you can use the flat surfaces to build a lathe that can make precision round surfaces. With precision round surfaces you can make vacuum pumps and seals, and in a vacuum chamber you can use physical or chemical vapor deposition to deposit atomically precise layers of metals.

The youtube chanel Machine Thinking has several great videos about such topics. I can especially recommend “The origins of precision” (https://www.youtube.com/watch?v=gNRnrn5DE58) and “The 1751 machine that made everything” (https://www.youtube.com/watch?v=djB9oK6pkbA).

Anonymous 0 Comments

Well there is a need for a standard for calibration, then it is scaled.

For example if you have a ‘standard’ meter length (in the old days, this was a metal bar stored in a national standards lab). Things like an inclined plane (or screw) provides a means to scale movement. In crude terms, make an inclined slope with a 10:1 ratio, now every 1 m moved laterally, moves you 10cm vertically. If that 1m is calibrated then the 10cm is similarly calibrated. Take that 10cm, do the same now you have 1cm calibrated. And so on. If you wind that inclined plane over a cylinder, what you get is a screw. Now you can say, “so many turns of the screw, moves the shaft of the screw up and down a certain amount”.

Anonymous 0 Comments

If you go look at the hinges on your kitchen cupboards, you’ll find that they have a few screws that allow you to adjust the hinge up and down, inwards and outwards. The holes they drilled to screw in the hinges can be fairly sloppy because you then have those adjustments to make sure the doors line up. That sort of adjustment exists in most machines, which allows you to fine tune your machine to a better tolerance than it was produced to. That’s crucially important: the individual pieces are produced to a lower standard than what you achieve at the business end of the machine!

Whatever adjustments you made, though, you still have some amount of imprecision. Where is that imprecision coming from? Perhaps you have way too much vibration somewhere and you need some way to dampen it with a spring, or add bracing to the machine so it becomes more rigid. Perhaps the problem comes from the way you’re operating the machine and you can change how you use it. Or maybe you just identify which components have the worst tolerances, and then you fabricate however many you need to get a “perfect” one.

Sometimes, improvements come from completely different schools of engineering. Car engines benefitted massively from electronics becoming cheap enough for electronic fuel injection. Sometimes pure science gives you a breakthrough. All sorts of instrumentation became much more precise when lasers became available.