[time-nuts] finding time astronomically.
albertson.chris at gmail.com
Mon Jan 23 16:12:33 EST 2012
> I can think of two general scenarios here.
> If you planet has air you will need to know how it refracts st
> One is where you "lay the iphone on the table" in a fixed position. One
> could use the internal accelerometers to determine "level", but I don't
> think you could tell orientation, unless, perhaps, you can see circumpolar
> stars? That is, by watching the movement of the stars/planets through the
> field of view over some hours, could you figure it out? Or is there some
> fundamental ambiguity.
No, you can point to any location and you can (in theory) figure out
where it's pointing given that you have a large enough field of view
to see many stars at the same time. You can make a fixture easy
enough, just some epoxy and a large boulder. I used lag bolts
onto my garage roof and it worked more than good enough.
If you can choose, straight up is the best aim point. Refraction is
not much of an issue and there is less air to look through. But
looking at the equator means there is less field rotation and the data
is easier to reduce. We looked at the equator because we did not want
to deal with image rotation. Motion blur is minimize down there too.
But if you want to know "absolute time" then you need more. Looking
at any random but fixed location will get you the period of the
planet's ration to about a mSec with cheap equipment but to get
absolute time you need to measure the aim point relative to the local
meridian. That is not as easy. Star with a protrator and a plumb
bob. That is what I used. But to refine that you need a good
source of time and for the purpose of this exercise we don't have
that. Only the plumb bob which means "a few seconds of error". maybe
an precision level can do 10X better?
> (obviously, you can trivially see the moon/sun)
> The other scenario is where you get an inexpensive camera (webcam, or
> perhaps some slightly better point and shoot) and build a precision mount
> (so you DO have accurate knowledge of sensor orientation and position) Could
> you, perhaps over time, do an insitu calibration?
> I suppose any of these techniques is going to have issues with the
> uncertainty in when the image is actually captured (e.g. there's probably
> 10-100 ms you're not going to get away from).
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to
> and follow the instructions there.
Redondo Beach, California
More information about the time-nuts