1 - 1 of 1 Posts

· Administrator
1,098 Posts
Horology 101 - deciphering a COSC certificate

<!-- BEGIN TEMPLATE: facebook_likebutton -->

<like href="" font="tahoma" layout="standard" show_faces="false" width="300" action="like" colorscheme="light" data-share="true"></like><!-- END TEMPLATE: facebook_likebutton -->

by <!-- BEGIN TEMPLATE: memberaction_dropdown -->


<!-- END TEMPLATE: memberaction_dropdown -->

Published on 12-19-2009 10:58 AM

Number of Views: 1421

COSC is short for Controle Officiel Suisse des Chronometres, which translates to Official Swiss Chronometer Control. Founded in 1973, it is the conglomeration of several formerly independent testing agencies. The Director's headquarters is in La Chaux-de-Fonds and the testing offices are located in Bienne, Geneva, and Le Loche. If you own a COSC certified watch you probably have seen the paperwork that came with it but may have questions about it's meaning. For demonstration purposes we'll use the certificate from an ETA 2892 that's powering a Tiffany's Mark stainless steel Coupe model:
Font Parallel Rectangle Number Document

Let's take a look at what it all means. At the top left is the certificate number [10670108] and to the right is this particular movement's serial number [7403]. Underneath the certificate number is the movement's category [1] meaning it is a spring balance oscillating caliber larger than 20 mm in diameter and greater than 314 square mm in total surface area. The categories are explained on the back of the certificate in French. Bonne chance à lire! Next to the category is the type of movement. This one is an automatic winding with center sweep seconds hand [Automatique Seconde centre]. Immediately to the right is the movement's dimensions - diameter [25.60] and thickness [3.60] in millimeters.

The center section is the day-by-day testing results:
Font Material property Rectangle Parallel Number

The first column on the left is the 24 hour period of the 15 day test.

The second column is the temperature that day's test was performed at. 8˚C = 46˚F; 23˚C = 73˚F (called room temperature); 38˚C = 100˚F.

The third column tells us the position the movement was tested in that day. There are 2 horizontal positions (laying parallel with the ground):

CH = dial up

FH = dial down

and 3 vertical testing positions (hanging perpendicular to the ground):

3H = crown up (3:00 pointing up)

6H = crown left (6:00 pointing up)

9H = crown down (9:00 pointing up)

One other, 12H or crown right, is not used in the procedure.

The fourth column is the daily rate difference. Each day is measured individually against an atomic radio signal clock, and has no bearing on the next day's results.

The fifth column is the variation between the first day's and second day's results per position. Each of the 5 positions is tested for 2 consecutive 24 hour periods at room temperature, and any difference is recorded here.

For example, on day 3 the movement gained +2.7 seconds in the crown up / 3H position. On day 4 it gained +3.0 seconds in the same position at the same temperature. That makes the variation 0.3 seconds between the 2 days. Day 11, 12, and 13 are each one-day tests in the dial up (CH) position at the 3 different temperatures - column 5 isn't used here since there is no second day to compare the first day to. The only test condition that caused this particular movement to lose time was the cold temperature / dial up test on day 11.

Just below the chart is the date the testing process was completed [22/11/2003] November 22nd, 2003.

The bottom third of the certificate is the summary. This is the left half of the bottom third:
Font Rectangle Material property Parallel Number

The first, Mean daily rate in the different positions, is the average of the scores of the first 10 days. It's simply calculated by adding the 10 results together then dividing the total by 10. To pass, this score must be within the +6 to -4 second daily rate that's the familiar COSC standard. This movement actually had an average gain of +4.06 seconds so it is rounded up to +4.1 seconds.

The second, Mean variation, is also referred to as the rate deviation. This is calculated by comparing all the scores of the first 10 days. The Mean variation can be off no more than 2 seconds positive or negative to pass as this shows how consistent the movement is. The biggest variation was from day 6 [+5.3] to day 3 [+2.7] for a net of 2.6 seconds, and the smallest variation was between day 4 [+3.0] and day 8 [+3.0] for a net of 0.0 seconds. This movement only varied +0.5 seconds per day on average.

The third, Maximum variation, is the largest difference between the 2 days results in any one position. This measurement can be off no more than 5 seconds positive or negative to pass. This movement's worst variation was in the CH / dial up position between day 9 and day 10 at only +0.8 seconds.

The fourth, Difference between flat and hanging positions, is calculated by subtracting the average of the rates of day 1 & 2 from the average of the rates of day 9 & 10. This score must be within +8 to -6 seconds to pass. For this movement the day 1 & 2 average is +4.75 which gets rounded to +4.8 seconds, and the day 9 & 10 average is +4.7 seconds, so 4.8 minus 4.7 = the final score of 0.1 seconds.

Here is the right half of the bottom third:
Font Rectangle Parallel Circle Number

The first, Greatest difference between the mean daily rate and any rate from the first 5 positions, is the biggest discrepancy between the Mean daily rate (the first result on the left half of the bottom third explained above) and any of the first 10 test days. To pass, this score can be no more than 10 seconds positive or negative. This movement's largest difference was 1.3 seconds, which is the difference between the Mean daily rate score of +4.1 seconds and day 3's score of +2.7 seconds (remember that the figures on the certificate are rounded, so even though this is a 1.4 second difference the actual non-rounded off figures differed by only 1.3 seconds).

The second, Variation of rate per 1˚ centigrade, is how much difference the movement has for every degree of temperature change. There are 30˚ difference from the cold (8˚) to the hot (38˚) test temperatures. To pass, this figure can be no greater than 0.6 seconds positive or negative. The result from day 11's cold test is -0.2 seconds; the result from day 13's hot test is +1.4 seconds. -0.2 minus +1.4 = -1.6, then divide 1.6 by 30 to get 0.0533 which is then rounded to 0.05 seconds per degree. *Here's one flaw in the testing of this movement! Notice that when the test went up 15˚ from 8˚ to 23˚ the movement changed 4.5 seconds (-0.2 up to +4.3 seconds). Then it dropped from +4.3 seconds at 23˚ to +1.4 seconds at 38˚ for a change of 2.9 seconds. If you add the two fluctuations it's 7.4 seconds which equals 0.2467 (rounded to 0.25) per degree - this is still within parameters but is much different from the score according to the COSC calculation method.

The third, Rate-resumption, is the difference between the average mean daily rate of day 1 & 2 and the daily rate of day 15. This must be less than 5 seconds positive or negative to pass. This movement's average mean daily rate for day 1 & 2 is +4.8 seconds (rounded up from +4.75), and the day 15 rate is +2.8 seconds. It lost two seconds so the result is shown as -2.0 seconds.

This article was originally published in forum thread:

Horology 101 - deciphering a COSC certificate
started by
View original post
1 - 1 of 1 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.