In horology, “isochronous” refers to the timekeeping principle of pendulums noted by Galileo in the late 16th century. He observed that the oscillation period of a pendulum is constant, regardless of the angle of the swing.

This later led to the "swinging" balance wheel which oscillates more than a few times a second to regulate the rate of the unwinding mainspring through the Minute wheel and associated mechanics.

In both cases the swing is "driven" by the force "F" to the clockwork or watch train by the springs or pulley suspended dead weights acting on an escapement mechanism meant to maintain the "isochronism" or constant RATE and degree of swing.

In a tall case pendulum clock, the variations in Gravity can seriously affect its accuracy making it very ineffective for navigation as the gravity variations around earth are too great for accuracy without calibration to each location.

So came the clock, chronometer and watch. I (and countless others) have already spent too much time thinking about this (and probably talking about it here too much as well ). . . .

So my point; After TRULY seeing a watch accurate enough so that I could experiment with how to make it "hold time" within a few seconds a month, I think I understand one narrow area of how to maintain pocket watch accuracy.

Since cleaning, re-staffing and regulating a 16s 21J mdl 99 Waltham Crescent St on April 1 (no kidding) I have been observing its behavior to winding cycles, once and twice a day. If I wind it every 12 hours, it will lose about 1/2 sec a day, and if I wind it every 24 hours, it gains a 1/2 sec/day. So every week I cycle the winding back and forth, and it keeps within 5 sec of actual time.

The effect will change over time, but the causes are only the "isochronism errors" caused by variations of mainspring tension on the train and escapement. When the tension is lower, there is a slightly lower "kick" and a slightly faster cycle.

It is really nice to have a movement so accurate that you can actually see the results of this type of isochronism. I am certain this is why Hamilton warned their dealers that each watch has to be calibrated to the (habits of the) user.

If you have a really accurate watch, try this experiment and see what it brings.

You can see that through the course of this little experiment since 4/1/2010, this "worn old Waltham watch has kept within 5 sec accurate for over 10 weeks.

FUN!

This is the best 21J version because the motor barrel is Jeweled. Waltham added a pair of cap jewels to the escape wheel to make it 21. I used the micro-regulator for the last error correction when I timed it.

thanks David. there is so more to these watches than meets the eye.

Great post, Dave! I don't think many people realize how much science there is behind the motion of a watch or clock movement. Isaac Newton would be proud!

Great stuff, Dave. I'm going to try this with my Vanguard.

Dana