Thanks for sharing the paper, it seems my intuition on size was bad but also my intuition on max satellite density for LEO was wrong.