WatchUSeek Watch Forums banner

1 - 9 of 9 Posts

·
Registered
Joined
·
11 Posts
Discussion Starter #1
About 3 months ago a set the watch to loose 1,5seconds per day on the wrist.

About two weeks ago we had a heatwave here and for some reason it's running at +0,5s but the heatwave is over.
Can temperature have this lasting effect or are the 7s26c's temperamental lile this?
Nighttime rates are on the right.
I didnt drop or magnetise it.
15413532
 

·
Registered
Joined
·
103 Posts
Those are terrific numbers despite changing.

A new 7S26C from the Seiko parts department is essentially $30. They turn a profit at that price. So I set my expectations for all my 7S26s (I have over a dozen) way lower than you are achieving. I think you should be very pleased and not concerned in the slightest.

Great job charting it too !
 

·
Registered
Joined
·
11 Posts
Discussion Starter #3
Being new to watches but havong some experience in data I try to plot it in a way I can understand. This is one of the plots I make to try to make sense of it.

15415327
 

·
Registered
Joined
·
11 Posts
Discussion Starter #5
Sure,
I check the time before putting it on and off, and record in which position it rests. I then can get the rates in the 6 positions and on the wrist.
The bottom two graphs are sorted so you van see the distribution of the rates. The errorbars are made on the time of the measurement, if you measure only an hour the error is pretty big but if you measure 16 hours it's small.
The top two graphs may be a little more familiar as they are more like a timegrapher, the blue dots are "corrected", this means that I calculate the deviation it would have if it was never put to rest. So if resting face up would add 9.11 seconds a day, and it was resint 7 hours, I correct for 9.11*7/24 seconds.
The left middle graph is the same but I do not modulo 40 there so the true deviation is better visible. On the right middle the separate deviations are plotted, with the small dots before correction for the rate, the big dots after.
 

·
Registered
Joined
·
428 Posts
Sure,
I check the time before putting it on and off, and record in which position it rests. I then can get the rates in the 6 positions and on the wrist.
The bottom two graphs are sorted so you van see the distribution of the rates. The errorbars are made on the time of the measurement, if you measure only an hour the error is pretty big but if you measure 16 hours it's small.
The top two graphs may be a little more familiar as they are more like a timegrapher, the blue dots are "corrected", this means that I calculate the deviation it would have if it was never put to rest. So if resting face up would add 9.11 seconds a day, and it was resint 7 hours, I correct for 9.11*7/24 seconds.
The left middle graph is the same but I do not modulo 40 there so the true deviation is better visible. On the right middle the separate deviations are plotted, with the small dots before correction for the rate, the big dots after.
You can use the WatchCheck app for getting measurements as frequently as you want. This way you don't have to make corrections for 24hr deviation. Then you can extract the data and plot them as you like. I don't know if there is a way to insert error though. But I don't think you need it, for this application at least. When you have many measurements you can change to the summary page to see how your movement performs. Like in the picture below.
15415502

This is the summary for my SKX007 from the 30th of March to the 23rd of August that I had it on daily.
 

·
Registered
Joined
·
11 Posts
Discussion Starter #9
Ok, this is getting ridiculous. It's gone into less than +0.2 day on the wrist, -0.05 crowndown.

Unbelievable.
15453103

And if we assume a gaussian distribution for the different positions we get a nice cumulative probability density functions as well:
15453104


So this is what I was talking about when plotting as I want.
 
1 - 9 of 9 Posts
Top