1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Primitive methods to make simple geometrical solids

  1. Dec 15, 2012 #1
    Hi guys. I am old person but not very old. I have a qustion that is very important for me to be answered. How can you cut simple geometrical shapes from a irregular solid using simple methods only? By simple geometrical shapes I mean cuboid, cylinder, pyramid and sphere. stuff like that. You are not allowed to use any modern technology like lathe or anyting made using modern technology. I wonder this because at some time in history somebody must have done this somehow. I mean they did not find a lathe machine while walking in the forest right? For example I tried to cut a 25 cm long cylinder with diameter 3 cm out of irregularly shaped wax(it was big wax) and believe me whatever I did I could not make a perfect cylinder with these dimensions. I know it is impossible to make perfect cylinder but at least something close to it, for example with 1% dimensional error. modern machines do it as low as 0.0001% error. so im not asking much. what i did was about 10-20% error. using hand cutting and eyes as a feedback you can only do that much. imagine yourself in ancient time with no moder technology around and try to figure out how to do this. you can only use atuff available at that time like stones, fibers, cloth.... and shape a wax (or other easily cuttable solid like wood, butter, wetted soil) as a close to perfect cuboid, cylinder and sphere. its like sculpting, but a simple geometrical shapes rather somebodys face. this was long but only way to explain you what i want, and i will be very very happy if you find something. please reply if you do.
    Last edited by a moderator: Dec 18, 2012
  2. jcsd
  3. Dec 15, 2012 #2


    User Avatar

    Staff: Mentor

    Some remarkably accurate techniques have been around for a very long time. For example, a potter's wheel makes quite serviceable cylinders; and grinding three surfaces against each other produces very good flats.

    But a lot of it is just plain craftsmanship and practice... any klutz can make dovetail joints in wood with a router and a jig, but watching a pro doing it with a handsaw and chisels is a revelation. You could do worse than studying up on 18th century shipbuilding and woodworking to get a sense of what can be done.

    An exercise from the first day of machine shop, back in the good old days: Using nothing but a hacksaw and files, make a cube of aluminum 1" on the side. I don't know if they still do that.
  4. Dec 15, 2012 #3
    Hello Oldperson and welcome to Physics Forums.

    Is there some visual reason for the excessive font size you employed it makes things difficult for others.

    You will find all you require in

    Mathematical Models by

    Cundy and Rollett

    You might like also to look at the other books listed, I can't vouchsafe any of their credientials though.

    http://www.amazon.co.uk/s/ref=nb_sb_noss?url=search-alias=aps&field-keywords=mathematical models cundy and rollett

    go well
  5. Dec 15, 2012 #4
    There's no reason to exclude the lathe from your tool box because it is an exceedingly old technology. I'm sure the Babylonians and Egyptians had something like this:

    Last edited by a moderator: Sep 25, 2014
  6. Dec 18, 2012 #5
    That font size is my mistake, i didnt know it would be that big, sorry about that.

    In order to make a good turning you parts öust already be good. for example in that video the guy turns good cylinder and cuts some its part. if the cylinder that you turn is not nearly perfect then your cylinders rotation will be bad and the cuts you make on it will be bad too.

    its like a machine makes a part. if machine is not geometrically accurate then the part also will not be geometrically accurate. so there must have some other prectical method that guaranties geometrical accuracy.

    i mostly wonder how to make a flat surface, cuboid and cylinder. any ideas?
  7. Dec 18, 2012 #6
    A flat surface is made from scratch by "The Rule of Three," which says that if three flat surfaces all mate with each other, then all three must be flat.

    Three surface plates labelled 1, 2, and 3, are cast. 1 and 2 are pressed together with a layer of ink or blueing between them. Where the blueing has been pushed away you know there is a mismatch. The high spots are then ground or scraped down. Then the plates are retested. The next batch of high spots is then ground down.

    Next plates 2 and 3 are tested and the high spots ground down. Then plates 3 and 1 are tested and ground. Then we're back to plates 1 and 2. The cycle is repeated over and over until each plate matches with the other two with no appreciable high spots showing when they are pressed together with the ink.

    It's a long and tedious process but an experienced metal scraper can make three plates this way that are flat to within a very tight tolerance.

    It takes three plates because if you only use two you can get them to match with one slightly concave and the other convex, which you don't want. This isn't possible with three separate plates. If all three match, all three are flat.

    By the same method excellent straight edges can be made, as can the ways of a lathe or mill or any metal shaping machine. This is how machines were first made "geometrically accurate".

    The ancients did not do this that we know of, although there was nothing preventing them. (I think something like this must have been used on those stone walls in South America where the stones are fitted together so well they claim you can't slip a knife blade between them. We have no record that they did, though.) This technique didn't start being used in Europe until people started seriously exploring converting wood lathes to turn metal. It was soon clear the lathes themselves had to be constructed of metal and that hand held tool bits didn't make for good cuts in metal. Precision ways were the result of the need to feed the tool bits both along the workpiece and also into it for deeper cuts. Precision metal lead screws were developed.

    "Cuboid" objects are generally cut on the milling machine, one flat surface at a time. You are copying the flatness of your ways onto the workpiece.

    Turning a sphere is a matter of having the tool bit mounted in a holder that can both rotate and be incrementally fed into the work.

    Last edited by a moderator: Sep 25, 2014
  8. Oct 1, 2013 #7
    Reply to comment #2:

    You are suggesting to use potter's wheel which is a man made something already. No difference than using modern lathe, both are cheating.

    Reply to comment #3:

    I have had a look at that book, that guy builds his models using already available precise man made something like flat sheet and etc. i guess it was on page 14-15. So again cheating.

    Reply to comment #4 and #6:

    video: the guy uses already man made metal tool and the wood seemes like to be cylindrical already before use, so cheating again.

    rule of theree: what about cuboid ? as for paint, I think i could use some blood( naturally available)

    cuboid: using milling machine is cheating.


    Any other ideas on How would make precision simple geometric solid objects like cuboid, cylinder, sphere, pyremid, cone ( but especially im interested in cuboid and cylinder) ( precision requirement is 0.1 mm error ) if you were on an island without any modern technology having only natural materials around like: wood, stone, sand, soil, clay, wax, animal parts, water, iron, copper ( the solids listed are irregularly shaped ). the cuboid and cylinder must be made out of any solid listed above. ????

    By the way, this question is really very important for me, see even a year after I am back to it. Have done a research for one year but haven't found anything. Thanks.
  9. Oct 2, 2013 #8
    The idea is not to use already man made tools.
  10. Oct 3, 2013 #9
  11. Oct 3, 2013 #10
    Your criteria as to what is "cheating" appear to me as capricious and arbitrary. There used to be a guy on TV who made wooden furniture. He started each episode with a log, rough cut with an axe, and turned to round with a "lathe" which was a rope and a bent sapling acting as a return spring. He did use hand chisels and wooden mallets. Are the axe & rope cheating? How about the chisels? If so, I guess you're asking how to make finished pieces with your teeth and bare hands? Chipped flint hand axes?
  12. Oct 3, 2013 #11


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    You are replying to a thread that last had any activity in Dec. 2012, in case you didn't realize it. There's a good chance that the train has left the station.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook