There is a berm in front of the tank farm. Probably not enough, and I don’t have a picture handy, but it may have reduced the worst of the damage.
ETA: Picture of the berm:
According to the Starship User’s Guide, Starship can put 21 tons into GEO, or 100+ tons into LEO, without refueling or disposing of the rocket. So if it can put a kg in LEO for say $50, it can put it in GEO for $250.
This is an excellent resource, by the way, but it’s already a little bit dated:
The problem with serious math here is that we are dealing with a lot of speculative numbers. Garbage in, garbage out. But if you want to work out the economics of a space elevator, here are some things to consider:
-
The cost of capital. A space elevator would cost billions - probably $10-$20 billion minimum, and take many years to complete. Capital costs are the bane of huge infrastructure projects.
-
The energy cost of lifting the elevator to orbit in the first place.
-
Maintenance costs of a 24,000 mile long structure under incredible tension, plus all the moving parts. How do you inspect a 24,000 km long ribbon?
-
Safety Engineering. How do you fix a problem like a stuck car 12,000 miles up? How to you egress passengers in an emergency? How do you mitigate collisions with space junk?
-
Systems and people. How many people are required for the day-to- day operation of the thing? How big is the space station at the top, and how much to maintain it? ISS costs several billion per year. How does a heavy car actually climb a ribbon that far, how fast does it go, etc. If your car goes straight up at 60 mph, it’s a 400 hour trip to GEO. How fast do the cars go, how big are they, and how do you care for the people in them? Or is this cargo-only? If the trip takes hundreds of hours, you will have to have dozens of cars on the tether at any given time, I guess.
Then you have to figure out the useful life of the elevator, how many tons it will lift over that life, and get the fixed cost per ton and add it to all the variable costs.
No, but it’s interesting to look at the issue to see if it’s even worth pursuing a space elevator in the future.
Here’s a good study on space elevators from 2014:
From the section on ‘economic benefits’, which illustrates how much Starship changes the equation:
That $250/kg figure is likely low. For one thing, it’s in 2000 dollars. For another, there are always unknown unknowns in new technologies that tend to raise costs.
But even if that number is realistic, it shows that such a space elevator makes no sense in a world with Starship. Starship should be able to easily beat $250/kg. For a space elevator to make economic sense, I think it would have to reduce that by at least an order of magnitude, but Starship might beat that as well over time.
Space elevators were exciting back when the Shuttle cost $35,000/kg to LEO and other rockets cost $10,000. But now that SpaceX is already launching for $2500/kg or less, and Starship will lower it by at least an order of magnitude, the space elevator concept seems almost archaic.
I think you’re missing the point. If SpaceX told them to be there, then either the people at SpaceX are sociopaths (unlikely) or they legitimately thought it would be safe (likely). In which case something quite unexpected happened. I’ve see a lot of talk online about how much concrete seemed to be kicked up. A lot more than is typical suggesting that the launch pad was insufficient. There has been some speculation as well that some of the flying concrete may have hit the rocket, which may have contributed to the failure.
You’ve excluded a giant middle: SpaceX knew it was an unsafe location, as it was well within the hazard radius, and would not have allowed humans within that radius for any reason. However, they did allow remotely operated cameras from various media orgs covering the event. They knew there was the possibility of damage, but it gets them a better shot, and it’s just equipment so who cares. The damage might have been more than what SpaceX expected, but was nevertheless less than the worst-case estimates they used to calculate the evacuation radius.
I’d bet that if SpaceX had told the media folks that they could place their cameras as close as they wanted and had provided the media folks a contour map of likely camera survival rates, most media folks would have put their cameras well inside the “no-hope of survival” ring entirely get better pictures. And been happy with the results they got.
Wi-fi or cellular-enabled vid cameras now are cheap enough that businesses, and at least some amateurs, can consider them disposable. Or at least expendable for a big enough payoff of images recorded before they get eaten.
Sam is quoting 25753-58722_-tyson_sparks-may_3_2014_1128_am-_sparks_final_paper.pdf
I’d like to know how this $250 per Kg estimate is derived. Unfortunately the citation from the paper that quotes this seems to be unavailable?
Let’s do some back-of-the envelope math. We’ll start with earth escape velocity to simplify the kinetic vs potential energy calculations: energy for escape velocity is obviously somewhat more than required for geosync orbit, so we’re being a bit conservative here. (Not interested in LEO, we’re talking about real space travel here).
From wolfram alpha, earth escape velocity is 11,180 m/s. Kinetic energy of 1Kg at this velocity is 62.72 MJ, or 17.422 KW/h.
The retail cost of energy in the US seems to average about $0.2 per KW/h.
That’s retail, so we’re being a bit conservative again.
So the energy cost of 1Kg to geosync looks like about $3.5 per Kg.
I realize this is incredibly oversimplified. There will always be engineering inefficiencies, and one has to amortise the cost of building the elevator, which can’t even be considered with current materials and technology.
But in principle, it seems that a space elevator could be far cheaper than any rocket-based system.
If we could only build one…
If you want to go by energy cost, Starship can lift 150 tons to
LEO on about $500,000 in fuel, So about $3.50/kg. Same as the space elevator.
But that’s not the measure. There are capital costs, operational costs, insurance, maintenance, repair, etc.
Let’s say a space elevator lasts for 30 years. It costs $30 billion to build, so a billion per year has to be earned just to pay back capital costs, not including interest. Lets say you put four climbers on the elevator at a time, so one reaches orbit every day. At a billion per year in capital cost recovery, each car sent to orbit costs $2.8 million in capital costs. With a 13 ton capacity, that’s $218/kg just in capital costs. Adjust that up and down however you want. The paper uses $10 billion and 122 launches per year.
For your energy calculation, don’t forget friction and motive efficiency for the climbers, which would be substantial. Also, how do you power it? 24,000 mile power lines are not really feasible, and cables pulling it are totally out. The paper you linked suggests laser power, but lasers are very inefficient and you’ll be lucky to get 10% of their input power converted to climb power. Better multiply your energy needs by ten.
Then there’s the climber itself, which takes days to get to GEO under any reasonable strategy. If people are in it you have to carry substantial and heavy life support. Even if not, the climber will be heavy. The article suggests 7 tons for the climber, with 13 tons of cargo. So for each ton of cargo you have to lift more than half a ton of dead weight. Better increase energy requirements by 50%. So $3.50 becomes $35 due to transmission inefficiency, and becomes about $50 per kg of cargo lifted because of the dead weight of the climber. Not including friction.
And we haven’t begun to consider maintenance and operational costs. Nor the cost of having to move the payload to its actual intended orbit.
In comparison, we can send 50-150 tons at a time in a reusable steel can, to any orbital inclination or altitude. And do it multiple times per day. Which do you think is likely to be more cost competitive?
The space elevator concept was designed to get launch costs down from $10,000-$35,000 per kg, which is what it was in the pre-SpaceX era, down to $250 or so. That would have been revolutionary in 2000, but Starship is almost certain to get us below $250/kg if it works at all. There’s just no reason to invest the kind of money it would take to build a space elevator, even if we had the materials.
If we build space elevators in the future, my guess is that they would be almost exclusively for throwing mass out of Earth orbit entirely - planetary probes, supplies for Mars, etc. And we’ll likely see one on the Moon before we see one on Earth. In 1/6 gravity and a vacuum the problem is much easier.
Agree with all your math & logic.
A quibble though …
I’d suggest that if a space elevator is built on whichever body, its mechanically useful lifespan will be centuries, not just 30 years. Which really alters the terms of payback on construction costs. During that time it will need maintenance and may need or get upgrades as e.g. materials science progresses to build better tethers. But the real value, the counterweight up there, the tether connecting it, and the embodied effort of assembling that, will last a very long time.
Whether its economic life extends for centuries depends hugely on how the rest of space tech and the space economy develops. So it may be obsoleted economically with lots of mechanical life left in it.
I understand your overall thrust is that even before it’s buildable, much less actually built, space elevators have already exceeded their economic life due to Starship. And I don’t disagree with that contention.
I’m not sure about that. The payments on a 30-year loan are only a bit worse than on an infinite year loan. Even if you expect the thing to last forever, it still has to make payments on its $10B (or whatever) capital cost. Just like money, stuff-to-orbit is worth less in the far future (and not just because it might be obsoleted by something).
I specifically ignored LEO, as I said, we’re talking real space travel here. GEO or escape velocity is a LOT more expensive.
And we’ll likely see one on the Moon before we see one on Earth <<
The moon isn’t really a good candidate, it doesn’t rotate fast enough. More likely an electromagnetic accelerator (as in Heinlein’s Harsh Mistress)? Mars, perhaps.
As I said, this is all theoretical anyway until we produce the necessary unobtanium. 
I suppose that may be true: once we’ve got all the solar power stations (or whatever) up there that we need, perhaps lifting more stuff isn’t needed so much.
Of course, someone may come up with a real TOE at any time and then we just build spindizzies. 
While that may be true, I meant it in a much more conventional way. $100/kg would be an amazing launch cost today, but no way am I paying $100/kg now for a satellite that will be launched 50 years from now. That mass-to-orbit is worth less the farther out you go. But the space elevator has to be paid for upfront, even if most of its upmass is many decades out.
Without refueling in orbit, Starship can lift approx 40 tons to GEO instead of 150 tons to LEO. So about three times as expensive. If Starship can hit $100/kg to LEO, then it’s $300 to GEO, which is still in the same range as a space elevator, without the massive up-front construction cost and risk.
The thing is, the current market for LEO launch is much, much bigger than the GEO market. A space elevator would not be competitive at all for LEO flights, and probably not for GEO either.
Maybe if Musk’s Mars colony gets established there will be enough demand to warrant elevators for flinging material to Mars.
Not exactly. There are very few things for which LEO specifically is desired. There’s a market for GEO, and a market for “anywhere in space”. The only reason that latter market is all in LEO right now is that, right now, LEO is the cheapest “anywhere in space”.
Cheap is good, but low latency is better. Starlink (and its upcoming competitors) would still be in LEO even if GEO were cheaper.
Perhaps counterintuitively, the atmospheric drag is also a win. It makes the volume self-cleaning.
It’s tidal locked to the earth. No way to spin this.
It’s spinning once per month. Slow, but not zero. The cable would have to be longer, but is under less tension due to the lower gravity.
I don’t think there is any stable selenosynchronous orbit, though. The Earth’s gravity rules it out.
There might be some viable tether-based approaches, but they wouldn’t resemble a ‘traditional’ space elevator, I think.
Poor moon - so far from God, so close to Earth
Those Space 1999 guys had it right.