4 ms·

Maybe it's easier to keep things orbiting in space than floating in the air? The earth atmosphere is not as predictable.

by Grue3 10y ago

baq 10y ago
my thoughts exactly, and also no problem with keeping the batteries charged overnight while flying, no FAA/other countries' counterparts issues, etc.
- TeMPOraL 10y ago
  > no problem with keeping the batteries charged overnight while flying Satellites in LEO have night and day cycles too, albeit shorter. But you have more problems with keeping batteries cool during the day and warm during the night, and also each additional gram you want to add to a satellite costs a lot of money thanks to the magic^Wtyranny of rocket equation - so you can't just spam solar cells and batteries like you could on an aircraft. > no FAA/other countries' counterparts issues, etc. Space is pretty regulated too; you can't just lob a piece of junk into whatever orbit you want and call it a day.
  - tim333 10y ago
    > just spam solar cells and batteries like you could on an aircraft. Things like the Titan drone are mostly solar cells and batteries. It's not like you can increase the % of mass allocated to that because it's pretty much maxed out.
  - jpm_sd 10y ago
    LEO day/night cycles are less than 2 hours, typically. An order of magnitude shorter, in other words. Satellites don't really have to expend any energy to stay "up", and thermal regulation is much easier in a vacuum than it is in the stratosphere.
TeMPOraL 10y ago
It's easier to keep things in space, but it's much more difficult to get them there because you have to use a huge rocket to make the thing go fast enough to stay on orbit. If you don't need that speed and the ability to circle the whole globe, high-altitude planes seem like much cheaper options.
- rtkwe 10y ago
  You'd also need a much larger number of the solar powered drones to cover the same area as a satellite. They don't make it clear but I'd guess one of the main issues was getting the radio and networking gear light, small, and low power enough to work in their drone design with enough throughput to make the network vaguely acceptable while covering a large enough area to make the number of drones required reasonable.
  - TeMPOraL 10y ago
    OTOH having to have more drones gives you benefits of economies of scale when you move past testing phase; you can mass-produce a platform and drop the price significantly. Satellites, on the other hand, are mostly one-of-a-kind constructions, produced in single or double-digit numbers. (Also with a fleet of drones, failures are cheaper and less dangerous.) There are lots of factors affecting economics of both solutions. I guess one of these days I'll have to make a spreadsheet with back-of-the-envelope calculations to sort this all out, but my gut feeling for now is that solar drones are a viable replacement for a lot of current satellite uses.
    - rtkwe 10y ago
      Making more would help but there's still limits on how far down it could possibly drive down the price on this large complex drone. To me the largest issue is simply power, these drones need to store a lot of energy just to fly overnight. Add in the power requirements for all the radio gear, it'll need to operate overnight too to make the service particularly useful, and battery weight can quickly spiral outside of what can reasonably be done by a small company trying to build a small plane. There's also 2 hard competing design goals solar panel area and aerodynamics. The first wants broad flat surfaces to place as many panels as possible on while the second is pushing you towards long thing wings and bodies to reduce drag. I'm not saying these are impossible things to build but I do think it's more in the realm of large aeronautic companies rather than a small division of google.

Maybe it's easier to keep things orbiting in space than floating in the air? The earth atmosphere is not as predictable.

by Grue3 10y ago

Maybe it's easier to keep things orbiting in space than floating in the air? The earth atmosphere is not as predictable.

baq 10y ago

my thoughts exactly, and also no problem with keeping the batteries charged overnight while flying, no FAA/other countries' counterparts issues, etc.

TeMPOraL 10y ago
> no problem with keeping the batteries charged overnight while flying Satellites in LEO have night and day cycles too, albeit shorter. But you have more problems with keeping batteries cool during the day and warm during the night, and also each additional gram you want to add to a satellite costs a lot of money thanks to the magic^Wtyranny of rocket equation - so you can't just spam solar cells and batteries like you could on an aircraft. > no FAA/other countries' counterparts issues, etc. Space is pretty regulated too; you can't just lob a piece of junk into whatever orbit you want and call it a day.
- tim333 10y ago
  > just spam solar cells and batteries like you could on an aircraft. Things like the Titan drone are mostly solar cells and batteries. It's not like you can increase the % of mass allocated to that because it's pretty much maxed out.
- jpm_sd 10y ago
  LEO day/night cycles are less than 2 hours, typically. An order of magnitude shorter, in other words. Satellites don't really have to expend any energy to stay "up", and thermal regulation is much easier in a vacuum than it is in the stratosphere.

TeMPOraL 10y ago

It's easier to keep things in space, but it's much more difficult to get them there because you have to use a huge rocket to make the thing go fast enough to stay on orbit. If you don't need that speed and the ability to circle the whole globe, high-altitude planes seem like much cheaper options.

rtkwe 10y ago
You'd also need a much larger number of the solar powered drones to cover the same area as a satellite. They don't make it clear but I'd guess one of the main issues was getting the radio and networking gear light, small, and low power enough to work in their drone design with enough throughput to make the network vaguely acceptable while covering a large enough area to make the number of drones required reasonable.
- TeMPOraL 10y ago
  OTOH having to have more drones gives you benefits of economies of scale when you move past testing phase; you can mass-produce a platform and drop the price significantly. Satellites, on the other hand, are mostly one-of-a-kind constructions, produced in single or double-digit numbers. (Also with a fleet of drones, failures are cheaper and less dangerous.) There are lots of factors affecting economics of both solutions. I guess one of these days I'll have to make a spreadsheet with back-of-the-envelope calculations to sort this all out, but my gut feeling for now is that solar drones are a viable replacement for a lot of current satellite uses.
  - rtkwe 10y ago
    Making more would help but there's still limits on how far down it could possibly drive down the price on this large complex drone. To me the largest issue is simply power, these drones need to store a lot of energy just to fly overnight. Add in the power requirements for all the radio gear, it'll need to operate overnight too to make the service particularly useful, and battery weight can quickly spiral outside of what can reasonably be done by a small company trying to build a small plane. There's also 2 hard competing design goals solar panel area and aerodynamics. The first wants broad flat surfaces to place as many panels as possible on while the second is pushing you towards long thing wings and bodies to reduce drag. I'm not saying these are impossible things to build but I do think it's more in the realm of large aeronautic companies rather than a small division of google.