The D and A in EBITDA are more than usually important for SpaceX. Starlink satellites have an expected lifetime of 5 to 7 years, so they have to pay back their costs very fast. Also, this means that the existing fleet of satellites isn't the barrier to entry it might otherwise be. An entrant with a superior produce and moderately deep pockets can launch its own and wait while the capital stock of Space X falls out of the sky.
At a failure rate of 9%/year, a quarter of the chips would fail over 3 years, more than 1/3 over 5 years. And, as you point out, no way to replace them. That alone should doom the concept.
And there's a lot of micro meteorites floating around that will ding the solar/cooling arrays, degrading function over its life.
And the high performance electronics are super sensitive to cosmic radiation, and they can't be hardened without giving up a LOT of performance. There's a reason that NASA still uses fairly primitive electronics in space applications...
The D and A in EBITDA are more than usually important for SpaceX. Starlink satellites have an expected lifetime of 5 to 7 years, so they have to pay back their costs very fast. Also, this means that the existing fleet of satellites isn't the barrier to entry it might otherwise be. An entrant with a superior produce and moderately deep pockets can launch its own and wait while the capital stock of Space X falls out of the sky.
At a failure rate of 9%/year, a quarter of the chips would fail over 3 years, more than 1/3 over 5 years. And, as you point out, no way to replace them. That alone should doom the concept.
It would be nice if links to prior articles pointed to the Substack version, not the Medium version. For example: https://www.planetearthandbeyond.co/p/musks-most-moronic-idea-to-date
My apologies, I'm not sure why I haven't been doing that... I will get onto this in the new year.
Agree 100% with this
I knew that the "put data centers in space" wasn't a good idea, but it's really a TERRIBLE idea!
And how would a 4 square kilometer data center not be running into all kinds of space junk?\
Not to mention what happens if it falls out of the sky?
Thanks, Will, I just love reading your articles!
And there's a lot of micro meteorites floating around that will ding the solar/cooling arrays, degrading function over its life.
And the high performance electronics are super sensitive to cosmic radiation, and they can't be hardened without giving up a LOT of performance. There's a reason that NASA still uses fairly primitive electronics in space applications...
Be careful with your math. The cost of solar panel is rather $3 per watt or less, not $300. So, this cost is rather insignificant.
Residential solar is $3/W, commercial is $1/W and utility scale is about $0.70/W . I would expect space rated solar is running about $50/w
That is for terrestrial solar, space solar is usually 3 orders of magnitude higher (and they are far more efficient per square meter of surface area).
They are not putting Chinese panels into orbit.
Thanks for explanation. Indeed, space solars are completely different. I didn't expect that price difference is so big.
Perhaps Musk did the same mistake and in his business plan used price of panels that he has on his houses' roofs 😀
It is more likely Elon was high when he came up with the idea