3 ms·
> As the coverage area is a sphere in reality You mean a circle projected onto a sphere. Or the intersection of a cone with a sphere's surface. Or I don't ev
by mlyle 6y ago
> As the coverage area is a sphere in reality
You mean a circle projected onto a sphere. Or the intersection of a cone with a sphere's surface. Or I don't even know what you mean.
Latitude and longitude are just a crummy coordinate system.
The reason why a given point has more satellite coverage is simply because the circles are on the northern and southern extents of the satellite's movement more of the time. It really is that simple. You're scrunching yourself all up with map projections, etc. If you want, we can go all the way through the spherical trig and derive the exact distribution.
Or you can just eyeball a crummy desmos notebook of the 1D version of the problem: https://www.desmos.com/calculator/u2s44maasf https://www.desmos.com/calculator/u2s44maasf
> it should look stretched east/west on a https://en.wikipedia.org/wiki/Mercator_projection https://en.wikipedia.org/wiki/Mercator_projection
Jeez dude. Just no. Yes, it would be sqooshed on e.g. an equirectangular projection, but not a Mercator. You linked the Wikipedia article. It's right there at the top of the page.
"As in all cylindrical projections, parallels and meridians on the Mercator are straight and perpendicular to each other. In accomplishing this, the unavoidable east–west stretching of the map, which increases as distance away from the equator increases, is accompanied in the Mercator projection by a corresponding north–south stretching, so that at every point location the east–west scale is the same as the north–south scale, making it a conformal map projection."
https://en.wikipedia.org/wiki/Tissot%27s_indicatrix https://en.wikipedia.org/wiki/Tissot%27s_indicatrix
That's why the satellite footprints on the Mercator projection map that you shared are (correctly) drawn as circles that are getting bigger in apparent size, instead of ellipses.
At the school that I now teach at part-time, we explain this in 5th grade.
> As to the US vs North America again I was trying to say different things. The equator pass through South America therefore all of North America benefits. However, only the continental US benefits enough for a 400 satellite network vs 1200 to provide 24/7 coverage.
OK, which is already modelled in the study that you were responding to. But you implied that Starlink could further move satellites towards CONUS beyond what's modelled. And now you're handwaving about orbital dynamics when I've been mentioning that inclination improves coverage at the northern and southern extremes since the beginning of our conversation. What are you looking to accomplish, exactly? At some point you're just embarrassing yourself.
I was part of a team retained by Starlink doing early feasibility study of optical satellite to satellite links, deriving control system pointing requirements based on relative movement between SV's in various choices of inclined orbit. It's somewhat cringe that you keep trying to explain (wrongly) to me how this works.
P.S. There's one minor effect we've not discussed. It doesn't increase coverage, but it makes individual satellite passes longer. The Earth rotates in the same direction as the satellite is apparently moving near its northern/southern extremes, which lengthens satellite pass times.