4 ms·

The circles in this map are really deceptive. They're two fixed sizes depending on altitude with no apparent relation to actual coverage. Here is my attempt at

by gpm 6y ago

The circles in this map are really deceptive. They're two fixed sizes depending on altitude with no apparent relation to actual coverage. Here is my attempt at estimating actual coverage, based on a 25 degree angle from the horizon https://i.imgur.com/poj8cae.png https://i.imgur.com/poj8cae.png I've included a screenshot of the code I modified so people can play with it themselves. I made a slight simplification in my math (explained in the comments in the screenshot), but I think it should be close enough. Edit: Pictorial explanation of simplification https://i.imgur.com/XtADkmh.png https://i.imgur.com/XtADkmh.png

Dylan16807 6y ago
They're confusing, but I wouldn't say "really deceptive". The big circles seem to be satellites that are at final altitude, showing range equal to altitude, so slightly lower than 45 degrees. The small circles are satellites that aren't there yet, and the size is meaningless.
- arijun 6y ago
  I would say really deceptive, or perhaps just plain wrong. The circles are, in fact, intended to convey coverage and left me pretty confused about the huge gaps everywhere. From the about section: > Dashed lines show the approximate coverage with very conservative viewing slant (approx 60 degrees above horizon, where 90 means vertically overhead).
walrus01 6y ago
The circles in the map match what a number of offline (TLE based) standalone Linux and Windows satellite pass prediction tools will show you. They assume that each ground location is on a flat plane with no hills or mountains, and the size of the cone extending from the satellite to the ground is the full coverage of everything the satellite can see (such as if you had a magical laser on the satellite capable of illuminating any ground location visible from the satellite). Such as if you had a theoretical, nonexistant ground terminal that was capable of communicating with a satellite at only 1 degree look angle elevation above the horizon. Of course in reality this is not the case and most systems that communicate with LEO satellites will have minimum elevation look angles of 15-20 degrees above the horizon. The circles do, however, provide a useful approximate scale of how much ground area is visible at any given time from the point of view of a satellite. It's not intended to be perfect.
- Dylan16807 6y ago
  > Such as if you had a theoretical, nonexistant ground terminal that was capable of communicating with a satellite at only 1 degree look angle elevation above the horizon. No, no, that is not even close to true. Measure the big circles. They have a radius roughly the same as satellite altitude. From an altitude of 550km, the area the satellite can see all the way down to the horizon is wider than the contiguous united states, east to west.
gpm 6y ago
Made a fixed map without the above approximation. It just assumes the earth is a perfect sphere. https://droid.cafe/starlink https://droid.cafe/starlink

by gpm 6y ago

Dylan16807 6y ago

They're confusing, but I wouldn't say "really deceptive". The big circles seem to be satellites that are at final altitude, showing range equal to altitude, so slightly lower than 45 degrees. The small circles are satellites that aren't there yet, and the size is meaningless.

arijun 6y ago
I would say really deceptive, or perhaps just plain wrong. The circles are, in fact, intended to convey coverage and left me pretty confused about the huge gaps everywhere. From the about section: > Dashed lines show the approximate coverage with very conservative viewing slant (approx 60 degrees above horizon, where 90 means vertically overhead).

walrus01 6y ago

The circles in the map match what a number of offline (TLE based) standalone Linux and Windows satellite pass prediction tools will show you. They assume that each ground location is on a flat plane with no hills or mountains, and the size of the cone extending from the satellite to the ground is the full coverage of everything the satellite can see (such as if you had a magical laser on the satellite capable of illuminating any ground location visible from the satellite). Such as if you had a theoretical, nonexistant ground terminal that was capable of communicating with a satellite at only 1 degree look angle elevation above the horizon. Of course in reality this is not the case and most systems that communicate with LEO satellites will have minimum elevation look angles of 15-20 degrees above the horizon. The circles do, however, provide a useful approximate scale of how much ground area is visible at any given time from the point of view of a satellite. It's not intended to be perfect.

Dylan16807 6y ago
> Such as if you had a theoretical, nonexistant ground terminal that was capable of communicating with a satellite at only 1 degree look angle elevation above the horizon. No, no, that is not even close to true. Measure the big circles. They have a radius roughly the same as satellite altitude. From an altitude of 550km, the area the satellite can see all the way down to the horizon is wider than the contiguous united states, east to west.

gpm 6y ago

Made a fixed map without the above approximation. It just assumes the earth is a perfect sphere. https://droid.cafe/starlink https://droid.cafe/starlink