WOSTEP, Gibri Hamilton Isochronism and Timing


NAWCC Member
Feb 5, 2007
The book by Gibri (subject of an earlier post) combined with my schooling to motivate me to see how I could apply everything I learned when there was no customer pressure.

OK. I have been using a Hamilton 970 (1901) 21j 16s for a lot of self demonstration. I did change the mainspring from .18 mm to .17mm and improved the rates. But I wanted to see how far I push this.

The 970 is one of the highest grades watch produced by Hamilton. It was the pendant set forerunner of the 990 but included finish details like bevelled regulator spring, gold wheels, etc.

The factory spec for this watch 10 seconds per day across 5 positions.

I have now completed a self-demo using a .16 spring (Elgin 571BWR) which is even longer than the 620 mm long .17mm spring I used earlier

Gives 48 1/2 winds or 24 full turns of the crown. Max wind is over 330 degrees amplitude in horizontal. Very surprising result, but the spring length has a lot to do with final power output. Still, this is more amplitude than I got with the recommended .18mm spring.

All measurements taken with a Vibrograf MU 700. This reports the results digitally so there is no graphic interpretation involved like the paper tape machines.

This took about one week almost full time so I am not suggesting this should be commercial practice nor that there are many customers who would pay for the result. It is more an experiment in how far these 120 year old instruments can be pushed.

Better than watching TV anyway.

OK. I wanted to find the flat part of the curve at around 260 degrees vertical.

It took 8 turns to get 240 degrees vertical.
At 8 turns I had 240 Amplitude PU rate was +10
At 12 turns I had 260 Amplitude PU Rate was +9
At 14 turns I had 260 Amplitude PU Rate was +9
At 16 turns I had 260 Amplitude PU Rate was +9
At 18 turns I had 280 Amplitude PU Rate was +8
At 20 turns I had 300 Amplitude PU Rate was +6
At 24 turns I had 310 Amplitude PU Rate was +2

So I backed off to 18 turns and had the following result:

Amplitude Horizontal 310
Amplitude Vertical 270


DU +5
DD +2
PU +5
PR +4
PL 0
PD +10

The result of 5 seconds difference across 5 positions exceeds factory specs. Even the 6 position rate is within the factory limits for the 5 position specification. Pendant down has a lot to do with the pinning point at the collet which explains why it was often ignored (and explained as unimportant when worn in a pocket on a chain).

I would expect the run time to be very nearly doubled given the number of turns for full wind. Of course the last rates would be horrible.

So now I have to wear it for a week and compare it daily to the USNO network lag adjusted clock.


Staff member
NAWCC Member
Jan 12, 2017
New York State
Even though I find this post very interesting I believe it belongs in watch repair. It's more about the technical
side of watch repair(and regulating a watch Isochronism and Timing) more than talking about the pocket watches themselves.


Forum statistics

Latest member
Encyclopedia Pages
Total wiki contributions
Last edit
E. Howard & Co. by Clint Geller