Chronometers

[webcraft]

Before we start . . .

Yes, I'm paranoid, maybe one day the US will switch off GPS or a solar flare will fry the system. My idea of caution is a main GPS set with two hand-held GPSs in reserve, with the ultimate backup wrapped dry and snug with a small mountain of spare batteries - AND a sextant in case! If (as many of you do) you think the sextant is obsolete, then fair enough - but that is not the intended purpose of this thread. (Lets start another one for that).

Now I have a sextant I need to learn to use it. I also need a chronometer. This is a subject on which most astro-nav websites seem to have little to say, in spite of the popularisation of John Harrison's story in Longitude.

How accurate does it have to be to achieve a theoretical accuracy of 1nm for a daily position? How accurate to achieve +/- 10nm? And how expensive . . . are there fairly basic digital timepieces that would do?

For coastal sailing and up to a few hundred miles offshore timepieces synchronised to the atomic clock at Rugby are available - I believe they use LW radio signals. Can SSB be used to receive time signals in mid ocean with any degree of reliability?

I know there's still a few of you out there using sextants - what chronometers do you use, and what sort of accuracy do you achieve? (Assuming an accurate metal sextant).

- Nick

BlueMoment.Com
http://www.bluemoment.com

 

[Sinbad1]

Nick, for a chronometer you can use almost any timepiece. What makes it a chronometer is not the accuracy but that the inaccuracy is known! Buy a simple battery operated clock, either digital or with large sweep second hand. Listen to the radio four/two pips at the same time every day and keep a note of the error, ie it will be fast or slow X number of seconds. You will find that the error is cumulative. It will not be fast one day and slow the next.

Then you learn to take a sight. From sight to noting the time will take a specific number of seconds. Measure this and deduct from the recorded time. This deduction, assuming you take the sight from the same location on your boat, will not vary much on a daily basis.

Now some theory. We assume that the sun travels one nautical mile in every 4 seconds. You thus need to note the time of your sight within that accuracy.

Accuracy of position using a sextant is something which improves with practice. However, all a sextant sight gives you is a single position line. So, at 1030hrs you take a sight and and the time and it will give you a position line (assuming you are in the Nth hemisphere) which runs SE to NW. You note your speed and direction and then at noon you take another sight. This is a doddle and needs no clock/timing. This is the sun at its local zenith and the good girl will hang up in the sky in the same position for 5-6 mins to let you practice before it begins its slow descent westward. This will give you an East-West position line.

Move your earlier position line forward to the noon position line and hey presto...a fix (sort of)

Now all sorts of errors can creep in, ie inaccurate timing, swell changing your height of eye, currents making the distance travelled under or over read. You reduce these errors by taking an afternoon sight and moving the noon position forward.

It is fun and immensely satisfying. Accuracy with practice will be 4-5 miles.

Then you get cocky and take star sights. Same procedure at sun-up and sun-down. But instead of having one position line you take 6 stars and you have 6 position lines from all around your horizon. You will now find that you can get accuate to within a mile or less and any sight/position line which is incorrect will immediately be apparent.

As for time signals. You can get the world service on any short wave transistor radio and check your chronometer daily by the hourly time pips.





l

 

[webcraft]

Thanks,

That's helpful, simple and concise.

Now I've read it, can't imagine why I thought it was so complicated - or why I forgot about time signals on shortwave!

Is a clock's error always cumulative, irrespective of (eg) atmospheric conditions? I guess a quartz mechanism is likely to be relatively unaffected by ambient conditions.

As regards timing of sights - one method I though seemed simple is to keep the chronometer down below and press a separate stopwatch at the time of taking a sight. Then go below in your own time to work out the sight and simply deduct the time on the stopwatch from ships (chronometer) time to get the time of the sight.

- Nick

BlueMoment.Com
http://www.bluemoment.com

 

[Chris_Stannard]

Be careful in your choice of timepiece. The whole point of the original competition, which Harrison won, was a chronometer with a temperature compensating mechanism so that the error rate did not vary with changes.

The standard for chronometers was that the gain should be no more than 6 seconds a day, and that it should be consistent. The error was found by taking daily reaadings, but off your GPS will do the same job. Then when every thing fails you can rely on the watch by extrapolating the error.

I have a SEIKO quartz watch which gains 1 second a week and seems unaffected by change in temperature. It cost about £100. For me that seems good enough but remember that uncertainty will prevail.

One old mariners trick was to sail to a latitude and then sail along it to make a landfall. If you are in the middle of nowhere and worried about missing a small target this is not a bad trick. Lines of cloud often indicate land as the moist air is forced over them, that is why the Maoris called New Zealand the Land of the Long White Cloud. And birds often indicate that you are getting near to land.
Another trick is if the coast is all low lying and you are worried about hitting a port aim off about 30 miles and then you will know which way to turn to get there.

You are unlikely ever to need this but I hope it is of interest

Chris Stannard

 

[DoctorD]

Quartz - accuracy of

Quartz watches are affected by temp. That's why in electronics we have TCXOs (Temperature compensated crystal oscillators). In fact for super accracy the crystal oscillator is placed in a small oven to give it a constant temp - I won't give you the acronym for that type of oscillator in case I get a reputation for geekiness over and beyond my already dodgy reputation. Having said that I agree with all the comments on accuracy and cumulative error made. Most quartz watches will have a cumulative error that will increase or decrease at a regular rate.

 

[HaraldS]

Re: Quartz - accuracy of

good wrist watches do pretty wel, since the are temperature stabilized to body temperature, provided you keep wearing them.

 

[Gordonmc]

Just one note of caution on use of the Rugby atomic clock... it was out of action for quite a period just last week.

 

[Sinbad1]

Nick

I had a nice seiko watch with a stop watch mechanism. I took the sight, clicked the stop watch and wandered down to the quartz clock and took the time as well as clicking the stop watch again. Simple and effective and you don't panic if a flying fish suddenley hits you in the eye as you are heading below......ahhh those were the days.

 

[jollyjacktar]

Cheap Wall Clocks

Nick, I have tested and used some of those cheap wall clocks that come in a plastic housing. The accuracy can sometimes be adjusted quite easily and I have had them accurate within a few seconds in a month. Easier to read with failing eyesight. Costs bout 2 pounds. Not a bad bargain and Captain Cook would have welcomed one on the Endeavour.

 

[Paulka]

I wear through the year a 5 years old "diving" Swatch, which never varied more than a few seconds between both yearly summer/winter time changes, and I use this whenever I take a sighting.
I think the errrors due to the reading of the sun's height are far more relevant than the error on time, once I know by how much my Swatch is "wrong".
Basically, one error of one second, gives you an error of one Nm.

I love my sextant!

Good luck.

Paul

 

[Sinbad1]

sorry Paul, but an error of 1 sec is 0.25 of a nautical mile. For seconds of time is 1 mile due to the sun moving through an arc of 15 degrees in 1 hour.

I love my sextant too, despite the dust which has arrived subsequent to my GPS

 

[HaraldS]

This is actually the biggest possible error that would be time induced.

You would have to be at the equator and have the sun either exactly west or east of you to move the LOP this far from your actual position.

The time induced error is zero with the sun due south or north of you, as this would be a noon latitude equivalent.

At other times the LOP will be off the actual position by an amount between zero and the max of .25 miles per second of time error.

With the chronometer being consistently off, all LOP's will have the same error in longitude, whic induces a longitude error into the derived fixes which is a quarter minute of longitude per second of error. Depending on where you are that is less than a quarter mile. At LAT 60 it is half that amount.

 

[trout]

I was a navigating officer in the MN during the seventies when the chronometer was a beautiful brass instrument in a purpose built baize lined box. It had to wound the same amount of turns daily. The second mate would have dyed his hair for a quartz watch I am sure. To find the Chrononmeter error it was a simple matter of listening to the BBC World Service and getting the "pips". A book was also kept which showed the rate of error.
I am sure a good quality moderately priced quartz watch would more than fit the bill along with a good quality radio receiver.
Good luck I wish I was out there with you Sextant in hand oh how I miss it!

Regards,

Graham