|
|
From: Brooke Clarke (no email)
Date: Mon Jun 06 2005 - 18:36:07 EDT
Hi Peter:
I think there are ways to work around clouds. Since the clock can keep
good time it also knows the date. Note that this is much different from
a sextant where the sighting is at an instant. In the simplest case it
needs to record a full day's sun.
I have a mental picture of one of the geochron clocks, like at:
http://www.geochronusa.com/new_geo/
http://www.geochron.co.uk/
If you draw a horizontal line through your location then the shape of
the light curve would vary with latitude but it's timing is directly
related to longitude. The peak (or more properly the center) of the
curve would tell you your longitude and the width would tell you your
latitude.
Now that I have this picture I can see that the accuracy depends on the
shape of the sun terminator line on the geochron. When it's more of a
square shape (equinox) the latitude accuracy suffers but when is more of
a sine wave (solstice) the accuracy is fairly good.
Have Fun,
Brooke
Peter Fogg wrote:
>>From: Brooke Clarke
>>
>>Suppose that a clock that keeps UTC time to within 1 second has a photo
>>sensor with an analog output to a computer. The clock is indoors but
>>the room has a window to the outside so although the clock is in the
>>shadow of the Sun it does see a daily brightness curve.
>>
>>How accurately can the clock determine where it is in the world?
>>
>
>
> It will have a hard time working out its longitude.
>
> As to latitude, presuming the variance due to weather can de discounted,
> could the question be reformulated as: how accurately can the principle of
> an annalemma be used for determining latitude?
>
> Or another variation, could the times of sunrise and sunset be used over a
> period to determine latitude?
>
> As you may have noticed, these aren't so much answers as more questions. The
> general idea sounds plausible.
>
>
>
-- w/Java http://www.PRC68.com w/o Java http://www.pacificsites.com/~brooke/PRC68COM.shtml http://www.precisionclock.com
|