On 2013-11-25, Leon lcb11211@swbelldotnet wrote:
On 11/25/2013 1:00 PM, Swingman wrote:
On 11/25/2013 12:52 PM, Leon wrote:
Understood, I was talking more about the actual drawn dimension having
limitations to their resolution.
Like this:
https://picasaweb.google.com/lh/phot...t=d irectlink
No, can you think/tell me what that converts to as a fraction off the
top of your head? LOL My calculated industries calculator can't
either. ;~)
I like to see fractions, on my drawings, that I can actually come close
to reproducing. ;~)
That is a difference in measurement techniques and machine
design. Most machine tools have dials which read on 0.001" (or finer
for some machines), but in decimal format anyway.
Way back when, machinists worked to 1/128" at best (the Vernier
calipers would measure to that, while scales were marked to 1/64" at
best. But then, to make a running fit in a bearing, they would use
inside calipers and outside calipers to transfer measurements from one
to the other. Bore the bearing hole, take the measurement with an
inside caliper, transfer that measurement to an outside caliper (by
closing one onto the other by feel -- these had and have no markings)
and then machine the shaft to fit the bearing by slowly removing metal
until the outside calipers just slide over it with the right "feel".
(the calipers will spring a bit, so you need to learn what the right
feel is.)
These days, you purchase the shaft, measure it with a micrometer
to be sure that it is what it is claimed to be, (in decimal fractions of
an inch), make a trial bore with the cross-feed dial on the lathe
zeroed, measure the bore it produced, subtract that from the desired
size, divide by two (since most machine's cross-feeds are calibrated in
radius, not diameter) amd for rough work, just set it and bore. If you
need more precision, you approach the final cut in finer cuts, so set
that your last cut will be the same depth as the others, measure as you
approach it to be sure.
And (on a regular lathe, if you want even finer precision, you set
up a toolpost grinder, set the compound at an angle which gives you
1/10th the measurement infeed (5.7392 degrees, but you are likely to
only set it near to 5.75 degrees given the accuracy of the compound's
built-in protractor, and sneak up on the final dimensions. At last with
surface grinding, you don't have the degree of spring that you do with
normal turning.
And -- if you need even more precision, you bore and grind to
just under size, and then use a roller burnishing tool to mash the
surface down to a smoother finish at the desired measurement.
Or -- you use lapping to get that final finish and dimension.
The above is how *I* would approach greater and greater
precision on my machines.
This is how it could be done on a manual machine -- especially
one in a home hobby workshop (such as mine). CNC changes the game
somewhat. But -- the whole time you are working with tools and
instruments which read and are set in decimal factions of an inch, so
there is never a need to convert something like your 144.531250" to
144 & 17/32", and you never *think* in fractional inches. If you did, you
would be reaching for a calculator all the time. Maybe you buy your
shafting in fractional sizes, such as 0.500" or 0.375" or 0.125". Yes,
these are fractional sizes, but you *think* of them in decimal inches.
BTW The conversion with my scientific calculator (HP 15C) is done
with no problems -- discard the integer inch part, multiply the
decimal faction by the largest likely denominator (64), see that
it reads an even number, so multiply by two (converting to 32nds
instead of 64ths and get an odd precise integer number, so you
are there. Then add back the integer part of the overall
dimension once you have your fractional part right.
There are *some* digital calipers which will read in both
decimal factions of an inch, and in the nearest fractional inch size --
but you are unlikely to find a machinist using one of these for the
fractional readings -- which are, after all, just a "nearest fractional
size", not a "true reading", or you would wind up needing it to display
at least down to 1/1024th of an inch (to be close to the metalworking
basic of 1/1000" -- in some fields called a "mil" -- such as in the pin
layout dimensions for integrated circuits in electronics -- useful for
designing printed circuit boards. Some few of us got into metalworking
from the electronics field (as did I), but we seldom mention "mils" as
it confuses those measuring in mm (Millimeters -- a very different unit.
I've seen these "fractional reading" digital calipers, but never
been tempted to buy them. I just don't *think* in fractional inches
most of the time. Some few places, it is convenient. 16 Ga steel is
very close to 1/16", so I can convert that to 0.0625" and be close
enough to tell 16 ga from other sizes. (And no, that does not work
anywhere else, as the larger the gauge number, the thinner the metal.
This is related to how it is formed, progressively rolled thinner and
thinner, so it is just a lucky crossover point -- and where the limits
of my sheet metal brake and shear happen to be, so it is easy to check
whether I should try the sheet metal in those tools or not.
Enjoy,
DoN.
P.S. Not sure why I am bothering to post in this cross-posted
argument, but at least it is metalworking related, not
political. :-)
--
Remove oil spill source from e-mail
Email: | (KV4PH) Voice (all times): (703) 938-4564
(too) near Washington D.C. |
http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---