It’s not a bad analysis but seems to overlook some important details on both SpaceX and Airbus sides of the table.
- that the next SpaceX rocket using methane instead of RP-1 will reduce the amount of CO2 per unit thrust. As well as the fact that making carbon neutral methane is much easier than making RP-1 grade kerosene biofuels means potential future impact to SpaceX launch costs may not be as large as he expects.
- that this is a prototype aircraft which has no long term lifecycle data on things like thermal cycling on the airframe or the solar cells, or motor lubricants and could result in a lifespan much shorter than necessary to make up for the difference in cost. Using the quoted $500000 cost and 5 year lifespan, Starlink operates Ming cost per satellite is $11.42 per hour, we’d still need the serviced area per drone and per satellite to do a proper comparison but effectively the total cost to provide equivalent coverage to 1 starlink satellite by the drone has to be close to or cheaper than $12 bucks an hour to make it competitive with starlink in a 24/7 internet service provider role, the aircraft industry has a lot of red tape that could make that prohibitively difficult once you factor in the rest of the operating and lifecycle costs.
- that while this can fly above the clouds and thus avoid most of the weather troubles of lesser aircraft, it now faces the wrath of upper atmospheric lightning, (https://en.wikipedia.org/wiki/Upper-atmospheric_lightning) more correctly referred to as a Transient Luminous Event (TLE) these are upper atmospheric plasma effects formed above thunderstorms. So while it wouldn’t have to worry about the normal reasons that make flying in a thunderstorm a bad idea, it may be negatively affected by these, such as when a stratospheric scientific ballon mission just spontaneously released its payload over a Texas farm in 1989 due to a large high voltage current causing damage to the electronics that controlled the release mechanism (http://stratocat.com.ar/fichas-e/1989/PAL-19890605.htm). This might necessitate ongoing operations to navigate around and avoid weather that could reduce revenue due to time not servicing customers, increase operational costs due to additional flying, and possibly capital costs in the case of requiring spare aircraft to make up for any losses due to such upper atmospheric electrical phenomena such as Sprites (https://en.wikipedia.org/wiki/Sprite_(lightning)), Jets, Elves, Trolls, and Ghosts which are not as well understood as other thunderstorms related atmospheric effects, the sort you and I would normally call “weather” and that pilots and aircraft designers are used to. So while rain fade (https://en.wikipedia.org/wiki/Rain_fade) affects Starlink (anecdotal example https://old.reddit.com/r/Starlink/comments/nz611h/fyi_starli...), the loss of signal due to this does not pose a hazard to the satellite, which means there’s another increase in the potential operating cost variability compared to the relatively fixed costs of Starlink operations.
Oh absolutely, I was assuming that SpaceX have factored average storm activity over the solar cycle into their on orbit life estimates, since other than orbit decay that’s the other big factor in the on orbit lifespan of a satellite and factoring this in is normal practice.
And I’ve been super keen on these sorts of high altitude aircraft stuff for years. The entire “atmospheric satellite” concept is brilliant and has a lot going for it. I really want more progress on things like the high altitude platform that’s part of the Airship to Orbit architecture being slowly developed by http://jpaerospace.com
I was specifically trying to keep my criticisms to this analysis of the hypothetical Airbus drone competitor to Starlink rather than the idea that something can compete with them or the idea of high altitude platforms in general.
5 comments
[ 3.2 ms ] story [ 23.1 ms ] thread- that the next SpaceX rocket using methane instead of RP-1 will reduce the amount of CO2 per unit thrust. As well as the fact that making carbon neutral methane is much easier than making RP-1 grade kerosene biofuels means potential future impact to SpaceX launch costs may not be as large as he expects.
- that this is a prototype aircraft which has no long term lifecycle data on things like thermal cycling on the airframe or the solar cells, or motor lubricants and could result in a lifespan much shorter than necessary to make up for the difference in cost. Using the quoted $500000 cost and 5 year lifespan, Starlink operates Ming cost per satellite is $11.42 per hour, we’d still need the serviced area per drone and per satellite to do a proper comparison but effectively the total cost to provide equivalent coverage to 1 starlink satellite by the drone has to be close to or cheaper than $12 bucks an hour to make it competitive with starlink in a 24/7 internet service provider role, the aircraft industry has a lot of red tape that could make that prohibitively difficult once you factor in the rest of the operating and lifecycle costs.
- that while this can fly above the clouds and thus avoid most of the weather troubles of lesser aircraft, it now faces the wrath of upper atmospheric lightning, (https://en.wikipedia.org/wiki/Upper-atmospheric_lightning) more correctly referred to as a Transient Luminous Event (TLE) these are upper atmospheric plasma effects formed above thunderstorms. So while it wouldn’t have to worry about the normal reasons that make flying in a thunderstorm a bad idea, it may be negatively affected by these, such as when a stratospheric scientific ballon mission just spontaneously released its payload over a Texas farm in 1989 due to a large high voltage current causing damage to the electronics that controlled the release mechanism (http://stratocat.com.ar/fichas-e/1989/PAL-19890605.htm). This might necessitate ongoing operations to navigate around and avoid weather that could reduce revenue due to time not servicing customers, increase operational costs due to additional flying, and possibly capital costs in the case of requiring spare aircraft to make up for any losses due to such upper atmospheric electrical phenomena such as Sprites (https://en.wikipedia.org/wiki/Sprite_(lightning)), Jets, Elves, Trolls, and Ghosts which are not as well understood as other thunderstorms related atmospheric effects, the sort you and I would normally call “weather” and that pilots and aircraft designers are used to. So while rain fade (https://en.wikipedia.org/wiki/Rain_fade) affects Starlink (anecdotal example https://old.reddit.com/r/Starlink/comments/nz611h/fyi_starli...), the loss of signal due to this does not pose a hazard to the satellite, which means there’s another increase in the potential operating cost variability compared to the relatively fixed costs of Starlink operations.
Edit: Also, for handwavy blue-sky thinking about concepts and comparisons like these, there is much more than this article, like here:
(·) https://en.wikipedia.org/wiki/Atmospheric_satellite
And I’ve been super keen on these sorts of high altitude aircraft stuff for years. The entire “atmospheric satellite” concept is brilliant and has a lot going for it. I really want more progress on things like the high altitude platform that’s part of the Airship to Orbit architecture being slowly developed by http://jpaerospace.com
I was specifically trying to keep my criticisms to this analysis of the hypothetical Airbus drone competitor to Starlink rather than the idea that something can compete with them or the idea of high altitude platforms in general.