maxi77
Well-Known Member
One could always have a stab at checking the error in one's steering compass....
We always checked the bearing of the sun at sunrise and sunset, and woe betide for any OOW who failed to note the result in the log
Sextant - celestial almanac?
- Thread starter Cardo
- Start date
We always checked the bearing of the sun at sunrise and sunset, and woe betide for any OOW who failed to note the result in the log
oldvarnish
Well-Known Member
I very much agree that practice is the key.
On the subject of time (and apologies for having posted this before) but GPS time can differ from UTC by considerable amounts. For example, when I looked at the website below at the time of posting, the discrepancy was 15 seconds, which is quite a lot in astro-nav. Someone, perhaps can tell us how much. Might it be 15 miles?
http://leapsecond.com/java/gpsclock.htm
Just noticed this at the bottom of the above page
"Global Positioning System time, is the atomic time scale implemented by the atomic clocks in the GPS ground control stations and the GPS satellites themselves. GPS time was zero at 0h 6-Jan-1980 and since it is not perturbed by leap seconds GPS is now ahead of UTC by 16 seconds."
wadget
Well-Known Member
Yes thats something I forgot to mention. Of course as the error is known you can either adjust your watch as nessesary or calculate it as a time error, as is shown in some books.
You need to check your GPS with this though as many will calculate for the error by taking away the 16 seconds automatically.
As to how much error 15 seconds will make, I'm not really sure. I guess it would depend what you are taking a sight of as they "move" at different speeds.
Conachair
Guest
Max 4 miles at the equator on a solstice? Or when suns declination is same as your latitude? Is that right?
John Barry
Well-Known Member
15 seconds will give you up to a 3¾ mile error. You'll get the maximum error if the body is East or West of you, dropping to nothing at meridian passage.
Earth spins at (an average of) 360°/day = 15°/hour = 15'/minute = ¼'/second, and GHA changes at that rate (give or take a bit).
For a body East or West you can convert this to distance by multiplying by cos(dec), but only for a very short time difference. To do it properly you should solve the intercept for the new time (assuming you're doing Sumner / StH reductions), as the azimuth will change too.
Last edited:
JayBee
Well-Known Member
Errors in time only affect the longitude of a fix, provided the declination of bodies observed do not change significantly. In practical terms this only applies to the moon. An error of one second in time shifts position lines one quarter of a minute of longitude to the west, if fast and the same amount to the east if slow.
It is a bit misleading to express these errors in miles, as a minute of longitude on the earth's surface is everywhere less than one nautical mile, except at the equator. Think of the meridians converging towards the poles, with the distance between them decreasing to zero at 90 degrees north and south.
It is a bit misleading to express these errors in miles, as a minute of longitude on the earth's surface is everywhere less than one nautical mile, except at the equator. Think of the meridians converging towards the poles, with the distance between them decreasing to zero at 90 degrees north and south.
laika
Well-Known Member
leap second difference between GPS and UTC is, I believe, transmitted as part of the GPS message. The standard nmea GPS sentences put out by receivers should be UTC, not GPS time. Time taken from nmea alone without referencing a PPS signal won't be microsecond-accurate but I'd hope my plotter would be displaying time accurate to a couple of seconds.
Last edited:
oldvarnish
Well-Known Member
That's interesting. If what you say is correct, and I don't doubt it, the time shown on the screen of our GPSs is not, strictly speaking 'GPS time' but UTC. I'll try and remember to check it against the GTS next time I'm on board.
To add even more complexity, perhaps I should get the GTS from a DAB radio!
John Barry
Well-Known Member
I have two DAB radios and their propogation delay is about 2 seconds apart - the best one being about 2 seconds after the analogue radio. It does make the pips go on for a long time if they're all on. Mind you, my phone takes its time 'from the network', and it's about 2 minutes out.
And while we're on the subject of picky definitions of time, this started with a questions about astro almanacs, and they are referenced to UT, not to UTC.
The difference is small (always within 0.9s, and normally within 0.5s), but UTC is an atomic clock based time, and is kept close to UT (which is really UT1 and is what we used to call GMT) by adding/subtracting leap seconds. UT is the rate at which the Earth spins.
If you want to get really picky you also need to look at TDT (Terestrial Dynamical Time) which is the time reference for the movement of stars / planets / etc. It's linked to UT by the value ΔT, which is currently 66.664995 seconds (ish). TDT is the time frame used to generate the almanac raw data, which is then converted to UT (based on predicted values of ΔT) for publication. ΔT varies with the Earth's rotational irregularity.
oldvarnish
Well-Known Member
Sorry John, but I am such a careful navigator that I will need a more precise figure than that, he said, tongue in cheek.
John Barry
Well-Known Member
It's worse than that - the figure I gave is wrong. I've just noticed that I have 2009 in the year, not 2012 (I was looking at some 2009 ephemerides last night).
The current value (as at 12:12:12 today 12/12/12) is 68.0288 - nearly 2½ seconds more. The current RoC is about 0.00128s/day, but it varies, and this is a prediction anyway...
Good luck with your precision navigation.