gerald.d
140
Not sure how much that would help?
Jul 06, 2020,21:00 PM
As far as I'm aware, silicon escapements and "technological" advances are far more about ensuring reliability of movements between longer service intervals, and I'm not sure how much benefit they would provide in a standard COSC test that is performed in a very controlled environment (no chance of any magnetizing) over just a couple of weeks. Yes, Rolex do an incredible job pumping out all those movements at +/-2 seconds per day, but that's an accuracy statement, not one of precision, which is what COSC measures (a watch that performs on alternate days at plus and minus 2 seconds per day in a COSC test would get a very poor test result compared to one that consistently performed at +5 seconds per day).
It's great that Seiko went to the effort of getting these watches certified, but to pick up on Patrick's observation above, it is worth pointing out that even the regular Grand Seiko Standard of the early 1970's surpassed the COSC requirements of the late 1990's (and of course that continues to this day).
It is likely that the regleurs at Grand Seiko in the 1960's and 1970's were more accomplished than those in the late 1990's (no disrespect to Ohira-san - I'm pretty sure he would agree on this point).
As an example, I just dug out the certificate for one of my Grand Seiko "Firsts", dating from 1961, and its results are as follows (the Credor from 36 years later in parentheses for comparison) -
Mean daily rate in the five positions: +0.7 (1.5)
Mean variation: 0.2 (0.7)
Maximum variation: 1.0 (1.4)
Maximum difference between mean daily rate and any individual rate: 3.3 (3.9)
Variation of rate per 1 degree centigrade: -0.33 (-0.16)
Rate resuming: -2 (-0.5)
At the time this watch was tested, the Swiss Chronometer standard was of course even looser than COSC is today.
By the end of the 1960's, we were of course in a whole different league altogether with the VFA's. I tested a 6186 once - although unable to do the temperature testing, I ran it for two days in each of the 6 positions (because why limit yourself to 5?), and it returned the following daily deviations (can't recall the positional order, sorry) - 0,0,0,0,2,1,0,0,2,2,0,0.
Kind regards,
Gerald.