From: Bill Anderson
Sent: October 19, 2017 9:32 PM
To: Bob Goethe
Subject: What watch is it?



What watch do you use whilst doing celestial navigation, and why?  Here is the reason for my question.  I have a solar powered Citizen Chronograph, a gift from my employer in 2007 after 25 years of service.  It is a thing of beauty, is solar powered, has all the bells and whistles, requires 4 PhD's to figure out how to make it work (this is the truth) and it weighs a tonne (which is a slight exaggeration).  Perhaps it is even accurate... an added bonus is it being analog, I like sweep hands.


The 4 PhD's, friends of mine who also got the watches, sat down one lunch hour to figure out how to set them up.  I think they may have done it, but don't recall.  Joel and I this summer set about to do the same, and while it looked like we had it figured, it turns out that there was a missing step of a total reset at first, so the date and month are out of whack.  For about $60 one can get a Casio resin watch that will show either local time or another time zone (UTC) with the press of a button.  Some even have a radio link to WWV.  


So, I ask, as those delightful folk in Casablanca did, What watch?


From:                              Bob Goethe

Sent:                               October 20, 2017 8:18 AM

To:                                   'Bill Anderson'

Subject:                          NASA Time Lesson: What watch is it?


Ah, my padwan learner, you have asked the right question.  And by it I can see that you are reading in your celestial materials.



There are several ways to get accurate time.


What I use is a Casio WaveCeptor watch.  For $42 CDN you can get the time accurate to within milliseconds.


It syncs by radio every night with the National Institute for Standards and Technology (NIST) WWV radio station, as long as you are in continental US or southern Canada.  It works on Vanc. Is. and in Edmonton.


You read about “incredible accuracy” from this or that $5,000 Swiss watch...but the truth is that it cannot compare with the accuracy of this $42 watch from Japan.


It supports two time zones.  I keep one on local time, and another on UTC...which removes one possible source of error when I am taking my sextant sight.  If I don’t need to calculate MST + 7 = UTC, then that is one less thing to go wrong.


A second method is to download an app (free in the Android world; probably free in the iPhone world) that tells you “Internet Time” or “Atomic Time” or “syncs your time” – all of these amounting to the same thing.  Here is the one I have:



The server gets a confirmation packet back from my phone, to indicate when its time packet was received.  From that, it can determine network delay.  Cut that in half, and you get the time delay for sending me the original packet.  It passes this back to the app, which adjusts the time for network delays.


In the end, it means that my cellphone is showing a very accurate “atomic time” here...and shows that my phone’s clock is 0.477 seconds fast.


But when taking a sight, I can read directly from the phone if I am on shore.



A third method is phone WWV at 303-499-7111.  My cellphone plan has unlimited calling, so this is convenient.  However, I find that this is invariably ½ second slow compared to my Casio watch...which is a consequence of the time it takes for the cellphone network to flip the signal through its various switches.


But if you know about that half second, you are OK.


A fourth method is to go to, which applies the same network delay measuring as my cellphone app, and so gives you very accurate time.  Hmmm...right now, that site is not coming in for me. My guess is that somebody is hitting it with a DDoS attack.


The Canadian equivalent is from the National Research Council at  There is a discussion there about network delays.




In my case, the ClockSync application on my cellphone, my Casio watch, and the NRC Official Time web site all agree with each other to the limits of my eyes being able to determine synchronicity...which is more than good enough for celestial navigation, where time is measured to the nearest second.



A fifth method is to use a shortwave radio.  You can get WWV at 2.5, 5, 10, 15, and 20 MHz.  The NRC has a station called CHU.  CHU transmits 3 kW signals on 3.33 and 14.67 MHz, and a 10 kW signal on 7.85 MHz.


The last method uses one of the above in combination with your solar Citizen Chronograph, and that is to use your Chronograph as a “hack watch”.  “Hack” in this case is not a derogatory term, and goes back to the 1800s for the watch you compare to your authoritative source, and then you take it on deck with you to do your sextant sight.


If you are able to stop/start the second hand, and can sync it directly with an authoritative source, I believe that is traditionally called “hacking the watch”.


But you don’t need your hack watch precisely synched, as long as you know how its time relates to UTC.  If you determine that your hack watch is 1 minute 18 seconds fast, then you take your sight with your sextant and hack watch, and subtract 1:18 to get UTC.



On my Hawaii trip, I was out of range of WWV for almost the entire time (my Casio did sync with WWVH in Hawaii for a couple of days).  But I had already “rated” my watch.  That is, before going sailing, I had turned off the daily sync by radio with WWV, and simply observed its rate of change over a 30 day period.  What I determined  is that my Casio, if it does not get a daily radio sync, will gain 0.18725 seconds per day.


So once I lost access to WWV, I could look at my watch’s indicator as to the last successful sync, calculate the number of days that had elapsed, and apply that number times 0.18725 and figure out how many seconds my watch was off...and subtract those number of seconds from my watch’s UTC zone reading to get true UTC.  I would redo this calculation each day, adding yet another 0.18725 seconds to my watch’s error.


(This also told me that my Casio, when it DOES synch at midnight, is extremely accurate at 1 AM, and could be off by 180 milliseconds or so by 11 PM.  I don’t USE that knowledge, as I don’t apply corrections of 0.2 seconds to my sights.  I have at least that much error just in switching my eyes from my sextant down to look at my watch.  Getting time within 1 second is good enough, and is actually a bit of a practiced skill, that you and I will work on together when you start taking sights.)



If you become a time junkie like me, we will go into how you can determine the difference between solar time (called UT1) and UTC.  They can differ by up to 0.9 seconds theoretically before a leap second is inserted.  Practically speaking, the world never lets UT1 and UTC get more than 0.6 seconds apart before inserting a leap second.



Your What Watch? Question put me in mind of an essay I wrote 25 years ago.  I am attaching it.



Bob Goethe


"The only noble thing a man can do with money is to build a schooner."
            Robert Louis Stevenson, 1889




This message was sent using recycled electrons.