Since the condensation subject keeps coming up, I made a little spreadsheet to roughly estimate water vapor diffusion through a seal into a watch. The interesting result (to me, at least) is that watches are kind of on the edge of developing condensation issues after a few years, even if none of the seals fail catastrophically, just from diffusion through the crown gasket. I'm assuming the crown gasket still seals with the crown pulled out, which has to be true at least for dive watches, otherwise condensation issues in cool water would develop immediately, since the air in the watch volume would be at the relative humidity of the surroundings after a short time (I'm guessing minutes) with the crown pulled.
You are excused from further reading if you have a number allergy....
Assumptions:
permeation coefficient for water vapor through Viton: K= 100x10^-15 m^2s^-1hPa^-1 (I picked a value that seems to be mid-range)
crown gasket area A: 10 mm^2 (3mm diameter o-ring, pushed into a square cross section of 1mm by 1mm)
crown gasket thickness d: 1mm
water vapor saturation pressure at 25C: 20 hPa (from this follows that 50% relative humidity at 25C corresponds to 10 hPa)
water vapor saturation pressure at 10C: 8.2 hPa
watch volume: 5 cm^3 (watch is 1cm high, 2cm radius; 50% of volume taken by movement)
Results:
Starting with a completely dry watch, and storing it at 50% relative humidity at 25C,
after one year you have 6% relative humidity at 25C in the watch (and 15% at 10C), and after
eight years you are at 40% RH @25C and 100% RH (=condensation) at 10C.
If you start with 25% RH at 25C in the watch (still a relatively dry assembly area), then it takes less than four years to reach the 100% RH @10C point.
Conclusion:
If you assume a larger seal area, or a smaller watch volume, or a higher storage humidity, or a lower water (=crystal) temperature,
condensation will appear earlier.
So, there does not seem to be a lot of margin, and having the watch assembled with dry gas inside could double your time until condensation can become an issue.
Click to expand...