Hi Keith,
It is true for any machine that uses asymmetric leverage, geometric
primitives and geometrically simple bearings.
An example, think of an asymmetric lever with point contact on a bearing and
the long end sweeping over a linear surface with errors. Then the short end
will generate a motion of shorter length but with smaller errors. The
leverage is the opposite sense from how people 'normally' perceive it. The
opposite techniques is used with levers using light beams to create more
sensitivity of output.
Another example is a lathe. A lathe with shafts and bearings that are not
so precise and not so geometrically perfect can be used to produce round
items that are more perfect - providing the headstock shaft is pushed firmly
in a single direction. I remember very early on in my life someone turning a
component to an accuracy of 2/1000 of an inch with a lathe that had bearings
with 10-15/1000 inch of error in them.
Flat is often harder to do than round or spherical. Until recently,
precision flat surfaces were made by hand by making two of them. They were
both cut roughly flat, covered with ink, rubbed together gently, and then
the high points carefully scraped off by hand using a small metal scraper.
The process was repeated until the required flatness was achieved. I did a
small one - it teaches patience! Once you have two precision flat surfaces
you can use them at a distance apart to use the leverage effects to get more
accurate flat surfaces through machining techniques.
A slightly different geometrically based technique produces spherical
surfaces - lenses. Simply take two glass blocks. Put abrasive between them
and rub in a circularish fashion. Eventually one goes convex and one
concave. Accuracy of sphericity and surface finish depends on abrasive size
and relative random even-ness of forces.
If I remember right this is covered in the literature of the field of
'history of machine dynamics'. Its also found in the technical end of the
alternative technology literature around the 1970s.
Cheers,
Terry
===
Dr. Terence Love
[log in to unmask]
Tel: +61 (0)8 9305 7629
Fax: +61 (0)8 9305 7629
Mob: +61 (0)4 3497 5848
===
-----Original Message-----
From: PhD-Design - This list is for discussion of PhD studies and related
research in Design [mailto:[log in to unmask]] On Behalf Of Keith
Russell
Sent: Tuesday, 15 May 2007 5:02 AM
To: [log in to unmask]
Subject: milling and flatness
All these stories about starting points has brouight back to mind an
argument that the milling machine was a vital invention in that it was the
first machine to be able to produce a higher level of precision in its
outputs than the level of precision in its component parts - that is, the
milling machine could produce a better milling machine.
I cannot find a source/resource for this argument.
any help?
keith russell
oz newcastle
|