The making of vintage precision instruments

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
Sign up Log in
14 replies · 2K views
DrDu
Science Advisor
Messages
6,423
Reaction score
1,004
Ever when I see one of these beautiful brass scientific instruments, like microscopes, telescopes, goniometers etc. from the 19th century, I wonder how they have been actually made and whether I could learn how to do this. I think that most of gears used in the construction of these pieces were rather simple and mostly manual.
Are there some resources on the topic?
 
Physics news on Phys.org
I know of at least one field that still (at least recently) practices precision manual metalworking.

In Sweden, I learned that to become certified as a master clock/watch maker you must pass a final exam. The exam is to make a clock or watch starting with nothing more than a solid block of brass.

Perhaps you can find Internet forums for clockmakers that might help you.
 
anorlunda said:
I know of at least one field that still (at least recently) practices precision manual metalworking.

In Sweden, I learned that to become certified as a master clock/watch maker you must pass a final exam. The exam is to make a clock or watch starting with nothing more than a solid block of brass.

Perhaps you can find Internet forums for clockmakers that might help you.
Yes, that's exactly what intrigues me. In former time, watchmakers could make a gear with little more than a file. But if you look into internet, most people think it is impossible to make a gear without a CNC mill.
 
DrDu said:
Yes, that's exactly what intrigues me. In former time, watchmakers could make a gear with little more than a file. But if you look into internet, most people think it is impossible to make a gear without a CNC mill.

I'm sure the day will come when Internet users believe that the Internet existed for all of human history, but it was just slower in ancient times. :rolleyes::rolleyes:
 
Was thinking that one way to attack this is track down people who build scientific instrument reproductions, and ask them how they do it, and how it had been done historically. http://radio-guy.com/index.html is a collector of everything old and oddball rather than a builder, but he may have a line on folks who are.

Beautiful microscope reproductions from Science Heritage Limited.
 
DrDu said:
But if you look into internet, most people think it is impossible to make a gear without a CNC mill.
If you want to get technical you can push sticks into holes around the outside of flat disks and have something that performs the function of a pair of bevel gears.
 
https://www.craftsmanshipmuseum.com/Wilding.htm. Have a look at his work and read some of his clock construction books .

Superb craftsman but also an excellent teacher .
 
Early in my career (almost 50 years ago), I worked for Hamilton Watch Co. in Lancaster, PA. I was always astounded at the craftsmanship of the older employees there. They had steady hands, good eyes, and could do absolutely astounding work.

One day, I sent a simple sketch to the model shop to get a support made for a simple experiment I wanted to run. It was just a foundation block with about 4 holes in it and no real need for high precision. I got the finished piece back a few days later, and the man who brought it to me said, rather apologetically, "You did not put any tolerances on the dimension, so we just worked to the usual." I told him I was sure that would be fine, but then curiosity got me, and I asked, "What is the usual?" In a very offhand manner, he replied, "Oh, half a tenth." That would be 0.00005 inches.
 
  • Like
Likes   Reactions: Asymptotic
Dr.D said:
curiosity got me, and I asked, "What is the usual?" In a very offhand manner, he replied, "Oh, half a tenth." That would be 0.00005 inches.

Correct me if I'm wrong or misunderstanding but, a tenth is 0.1" and half a tenth would be 0.05".
I can't imagine a clock would need down to 100 thou tolerance either. Maybe I'm wrong?
 
RNickl said:
Correct me if I'm wrong or misunderstanding but, a tenth is 0.1" and half a tenth would be 0.05".
I can't imagine a clock would need down to 100 thou tolerance either. Maybe I'm wrong?

You are looking at this mathematically, but that is not the way things worked in the mechanical watch and clock industry. One tenth of an inch is a huge distance compared to the size of many watch parts, so the unit distance was usually considered in terms of thousandths of an inch. The "a tenth" becomes one tenth of a thousandth, or 0.0001. Half a tenth is then half of that amount, or 0.00005 inches.

I recall vividly one part dimension that was 80 millionths of an inch (I don't recall the tolerance). This was a thickness on a hair spring.
 
RNickl said:
Correct me if I'm wrong or misunderstanding but, a tenth is 0.1" and half a tenth would be 0.05".
I can't imagine a clock would need down to 100 thou tolerance either. Maybe I'm wrong?
The mil (1 mil = 0.001 inch) remains a common expression in U.S. machine shops, and fractional parlance is as @Dr.D described in post 13.

Large equipment may also require rather exacting tolerances, although not as critical as those in mechanical watch movements. For example, a rule of thumb for maximum shaft misalignment in high power, high speed applications is one-half mil (0.0005"). Another is the maximum bar-to-bar height variation (MBTB) spec for DC motor commutators of under 1/5th mil (0.0002 inches). Sparking, and other high/low bar manifestations occur if variation goes much above 0.3 mil (0.0003 inch).
 
I've seen Victorian-era advertisements for 'instrument lathes'. No bigger than a A4/Legal page, we'd call them 'model making lathes', but they could do wonders. Patience, appropriate tools & materials, hand-read Vernier dials and gauges, exquisite design to make the most of their possibilities...

Before them, you'd have copper & brass smiths who could hand-form sheet, rod and bar into shapes you'd swear were cast.

FWIW, one of my BILs apprenticed as a traditional copper-smith. Years later, he was flown out to the Gulf to re-work the bearing lubrication on a new super-tanker whose engine had its complex 'loom' fitted back to front. The oil went in the 'low pressure' end of the cylinder block instead of the 'high pressure' and, even with the ship 'in ballast', half the bearings were running hot.

Snag was the ship's engine had gone in as a module, then the accommodation section plonked on top. To fit a new 'loom' as-is meant cutting the ship in two. Or, sending a very skinny, very limber guy with an old-fashioned knack...

So, for a week, he wriggled into gaps around the engine and, with simple hand-bending tools, crafted a new loom 'in-situ'. He did such a neat job that, to put it politely, the Lloyd's Inspector who signed-off the repair was flabbergasted...
 

Similar threads

Cleaning copper connections for high precision instruments
  • marcel
  • · Replies 22 ·
  • Chemistry
Replies
22