[time-nuts] Gear ratios, inch vs meter, tooling, and how technology has made life easier

[Jim Lux]

On 12/16/11 8:00 AM, Don Couch wrote:
> The idea that conversion to metric would require replacing all of the machine tools (lathes, mills, etc) is a myth. Any U.S. machine shop has walls and toolboxes covered in conversion charts, converting drill, screw, wire, sheet sizes from one crazy measurement to another. One single additonal conversion chart, inch to metric, and you can keep using your inch machines on metric projects.
>
> My mill has inch lead screws. I added a low cost digital readout with a little button to show inch or millimeter movements, and now I do everything in metric. No problem.
>
> Don Couch
>
20-30 years ago (when Space Shuttle, a notable user of inch/pound, was 
being designed and built) this was not the case.  An awful lot of stuff 
was done with machine tools that were designed and built in the 1950s or 
earlier, and things were done with gear trains to set the feed rate, etc.

All those gear ratios were designed to produce that 20 threads/inch, 
etc. and changing gears *is* a big deal: once you've got the shaft to 
shaft distance fixed, there's only certain ratios that "work", unless 
you want to start making gears with weird shaped and sized teeth. That 
*is* a big tooling expense.

  This is still the case for a lathe. Sure, you can do "electronic gear 
boxes" but then you need multiple motors, rather than just driving 
everything off one.  And having it geared means that even if the motor 
slows down, the ratio of along to around stays fixed.

With CNC, life is easier.

Sometimes we forget that not too long ago, there was no such thing as a 
"low cost digital readout", and CNC was fairly rare.


Today, though, a "big bang" conversion would be a lot easier.  It *is* 
just a matter of pushing a button, since the "gearing" is done 
electronically for the most part. Ditto for displays.  My car displays 
temperatures in F or C with a push of a button.   Back in the 70s when 
we last ventured forth into metrication, it would have been much tougher.




And to keep this tangentially timenuts related..  I ran into this whole 
"not just any gear ratio is possible" when trying to make a Mars Clock 
back in 2003.  There was some guy making Mars wristwatches (a 
fashionable accessory for the folks doing MER ops). So I thought about 
getting a clock and some gears from Small Parts or Boston:  I talked to 
a friend who had made a couple Orreries (which have similar needs); and 
learned some practical experience about gear trains.

I took the easy timenuts way out:  Buy a cheap battery clock with a 24 
hour movement, rip the 32.768kHz crystal out, drive it with a HP3325B 
set to slightly less to match the Mars day length, locked to the 
Hydrogen maser on lab, thereby producing the world's most accurate Mars 
clock.  Sure, there was this piece of coax running from the 3325 under 
my desk out the door to the clock out in the hall, but hey, it worked.