Steyr
136
No need for PhD, just count the black lines
Apr 06, 2022,21:42 PM
...9 lines in between two same-colored lines, you multiply with the oscillation duration and you get the measured value. If all measured values equal 1.8s (9 black lines=9 osc.x 0.2s) while events durations are from 1.62 to 1.98, this means that this watch cannot make the difference between 1.62 and 1.98. This is is a 0.36s gap. Much more than a tenth.
Too complicated ? Another demonstration for a young kid. You want to measure a 1.49cm length with a centimeter graduated ruler. You will have to choose between 1.or 2 centimeters. So the uncertainty is 2 centimeters. This is greater than 1cm.The measure cannot be centimeter accurate with a ruler graduated every centimeter.
If you have a half-centimeter graduated ruler, The uncertainty will be between 1.5 and 1 cm : 0.5cm. It is bellow 1cm so you can measure accurately the length with a 1cm accuracy.
To measure centimeters, you need a measurement tool with twice the frequency you want to measure.
Ho wait that's pretty much what is stated in Shannon Nyquist theorem "If a function x(t) contains no frequencies higher than B hertz, it is completely determined by giving its ordinates at a series of points spaced 1/(2B) seconds apart. A sufficient sample-rate is therefore anything larger than 2B samples per second."
You may check the "uncertainty of measurement" in your favorite search engine as well
This is when you realize that the Journe centigraphe is based on a 3Hz oscillator :-D
Do you remember if you had it in your hand or the videos on the internet, you understand why the 1/100th indication seem to "slow down" on 6 specific graduations
How would you measure 1/100th when obviously, some values are more "important" ?
6 graduations, 3 Hz, oscillation/alternance, you get it ?
Going back to the clutch. The El-Primero chronograph wheel has 360 teeth. Wlutch wheel is 150. When the gears engages, you have a 2.4° angle possibility (360/150). 1 turn for the chrongraph hands is 60x10=0.6°. So depending on the positions of the gears, you have a +/- 0.2 seconds. So the measured value is already 0.4s accurate. Not even counting the frequency bias, explained above. So you can understand that the measured value of an El-Primero can be as low as 0.4+0.4s=0.8s accurate Not counting the beat error that is less impacting with 5Hz frequencies.
About the beat error. Take a look at the mikrograph.
Do you seen these Tic......Tac..Tic......Tac..Tic.....Tac..Tic......Tac..Tic Can't you see the difference in between the oscillation and the alternance importance with theses uneven durations ? What can you really measure with a hand that goes faster then slower, faster, slower ?
You seen, only counting multiplications and basic observation here ;-)