So I'm looking into how stuff is made from scratch, starting with a surface plate which has been lapped flat using the three plate method. But almost everything I read on the subject on "the history of precision" and such stops there and glosses over the rest by simply stating that it's possible. How does one go about making perfect 90 degree corners from simply a flat surface?

To clarify, when I say "from scratch" I mean going from no tools at all to a perfectly square block. I'm asking about the history of machining, how the first precision machines were made.

If anyone has any good resources on the subject which explains it in further detail I would appreciate it very much :)

  • $\begingroup$ Do you mean steel ? do you mean "Jo" blocks ? $\endgroup$ Jan 15 '21 at 0:30
  • $\begingroup$ I mean any material really, but yes, steel is usually what precision machines are made out of. "Jo" blocks/gauge blocks are only to measure length, right? Or are they also perfectly square? In that case, yes, the question could be phrased as "How do you make gauge blocks starting with nothing". $\endgroup$ Jan 15 '21 at 0:38
  • $\begingroup$ Check out the tools used by the romans 2000 years ago for surveying roads... then work from there. $\endgroup$
    – Solar Mike
    Jan 15 '21 at 6:16

The "three plate" method has been used as the basis for precision manufacturing/inspection. This is a decent summary, it's pretty ingenious and only requires dye and abrasives. This will teach you everything you need to know and then some, "Foundations of Mechanical Accuracy".

  • $\begingroup$ I would be very interested in that video :) After some soul searching I've realized that straightness is easy to achieve, since it's just the intersection of two planes, so if you make to neighboring sides of an object flat, anything sliding along those two surfaces will travel straight. Still not sure how to get 90 degree corners tho :S $\endgroup$ Jan 15 '21 at 12:48
  • $\begingroup$ See my updated answer $\endgroup$
    – jko
    Jan 15 '21 at 13:02
  • $\begingroup$ Oh, so they just "did it by hand", checking the surface and straight edge for flatness using an indicator and just went to work scraping it flat until it was straight and flat? :S Thought there was some... less labor intensive way of doing it, like the three plate method but for straight edges and such. Or did I miss something? :S $\endgroup$ Jan 15 '21 at 13:20
  • $\begingroup$ A flat plane is a "straight" edge, but it seems you are asking about parallel edges (2 sets of straight edges in parallel planes). The precision you desire dictates the amount of labor required to achieve it. But once you've achieved an acceptable flat plane as a reference surface, determining flat, parallel and square surfaces goes much faster. Look up how you can determine squareness of a table saw sled to get an idea of how to make "square" blocks from scratch. $\endgroup$
    – jko
    Jan 15 '21 at 14:01
  • $\begingroup$ Am I misunderstanding what a straight edge is? :S From the looks of the pictures and how they describe it in the text a straight edge is a line, the line formed when two flat surfaces intersect, right? So a flat plane itself has an infinite amount of straight edges on it, but it does not define any of them. $\endgroup$ Jan 15 '21 at 14:43

there is a whole genre of "square block" videos on YouTube by various machinists. The standard tool seems to be the lathe.

But really, you are asking how to set up a machine to be perpendicular.

apropos, video for your edumatainment

  • $\begingroup$ You misunderstand my question, it's about the history of precision. When I say "From scratch" I don't mean "From a lump of metal using modern tools", I mean, from no tools at all to a perfectly square block. $\endgroup$ Jan 14 '21 at 23:53
  • $\begingroup$ @Beacon -- good question! tag the post for "metrology", lets see what we get. For what its worth, when tramming the mill, the guy in the video constructed a 90 deg angle from just the rotating axis of the spindle. But in ancient times they might not have had axisymmetric objects. I suspect strings and pendulums were involved. $\endgroup$
    – Pete W
    Jan 15 '21 at 0:18
  • $\begingroup$ Thanks, I was looking for more tags but I was only looking for "precision" and stuff like that which didn't exist. $\endgroup$ Jan 15 '21 at 0:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.