Watch Experiment

[QuickMick]

Ok... this came up in 'seamanship' here:

http://www.sailnet.com/forums/seaman...tml#post675167

so i was wondering.... how accurate are our timepieces. So what I was thinking is that for those who want to participate we calibrate our watches to noon GMT on Mon (to give people time to read this) or mark your local equivilent... let them run for....uh i dunno two weeks? and see how you measure up for the +/-

ill set my timex expedition (the one with dual time/compass), my tommy bahama 'relax' dual analog/digi and my vestal tide watch.

maybe we can get a handle on the accuracy vs price of various models...

suggestions/improvements welcome to try and make this semi scientific...lol

bill nye the science guy signing off.

[cb32863]

You realize that if we do Noon GMT that here in the Central time zone, that is 6am local....... 4am on the west coast. If we do 6PM GMT then that may be a bit more manageable.... Or even Midnight GMT. Just throwing it out there....

[noelex77]

If serious about celestial navigation you need to know the error in your timepiece, but what is more important than absolute accuracy is a consistency. A watch that gains 4 seconds a day is still fine if you have measured the error and if it is a consistent reproducible error.
A watch that gains some days, looses others, depending on temperature etc will be inaccurate for celestial navigation.
So the watch that’s closest after 2 weeks, even if it was spot on, may be still a poor choice, with poor consistency that happens to have evened out over the 2 weeks.
Checking it every day for 2 weeks would be a better test. The winner (for accurate celestial navigation) would be the watch with the most consistent result.

If looking at the imaginary worst case when the GPS system is wiped out (extremely unlikely IMHO) its worth remembering many things like a nuclear pulse that would wipe out the GPS system will also take out quartz watches. So a mechanical watch may be your best back up even if its inherently less accurate.

[Hesper]

I suspect a nuclear pulse is going to take out a hell of a lot more than just my quartz watch.

I've actually done this kind of test with several of my watches - a Tag Heuer, a Seiko, a Nautica (don't know who actually made it) and another whose brand name escapes me at the moment (a nice titanium model). All of them kept time to the second over a period of about a year. I admit I didn't check them every day, but periodically over that time. No difference between them.

[AdamLein]

I have just synched my watch with USNO Master Clock and will be participating in this experiment!

I will probably report back once a week with my average absolute errors for each of the preceding seven days, inclusive of the day of report.

[cb32863]

As for the above mentioned worse case scenario.... I mean really? How about we just think about what would happen if our GPS batteries died......

[cb32863]

Originally Posted by AdamLein

I have just synched my watch with USNO Master Clock and will be participating in this experiment!

Just did the same with my Seiko, will be interesting to see how it holds up.

[SVAuspicious]

Cool. I'm in.

[zedboy]

In with my (quartz but analog) timex.

Prediction: they will all be within their setting error (less than a second of drift).

[scottyt]

i would compete here but my watch would win. it is rated at 2 seconds a month error on its own, but the real problem is it up dates by the national radio signal. so it gets reset daily, or worse case every 2 days. casio g shock for the win

funny side story
my grandfather worked for raythoen many moons ago. he bought one of the first electronic watches. it had a guarantee to be accurate to something like 8 seconds a month. well he tried to return it under warranty, it was off 10 mins in a month, the store refused to take it back. they said the watch was accurate but the clocks he was checking it against had to be off . he then informed them it was checked against a nuclear clock at his work. they took the watch back.