Elon Musk quote: SpaceX will lean in big on the Moon

[meekGee]

It's how you used it. It didn't occur to me that someone would know the reference and so thoroughly miss the message. 

It's satire. Self deprecating.  Stopping the sermon to have a chuckle. Lighten up...

[Vultur]

The only exception would be if lowering the prices increased demand enough to make up for it. E.g. if at $100/kg you got more than a factor of ten increase in business. Even if it did, a sudden price drop would not instantly increase business that much. That kind of increase would take years to happen. Never mind the factor of 100 to make it worth dropping the price to $10/kg even if the launches cost SpaceX nothing!

That's a big problem for Starcloud or Google Suncatcher - but not for potential AI data centers hosted by SpaceX itself on Starlink satellites.

(I actually don't think it's at all guaranteed that  SpaceX will ever face real, direct price pressure from competition. Chinese launch systems are not direct market competitors due to geopolitics ... and within a single market, economies of scale may be sufficient to make launch essentially a natural monopoly.

The US government and certain other large customers (e.g. Kuiper) will ensure SpaceX doesn't have a true orbital launch monopoly, but that doesn't at all guarantee a genuinely price-competitive system.

The profit margin from launch in a world with relatively cheap launch may be insufficient to justify the investment for a "Starship-killer" system. Though it might happen anyway, as essentially a personal project for someone like Bezos with multi billions to burn and no real need to see return on that investment.)

[Greg Hullender]

The reason for going to orbital PV is that nobody else has a path for installing 100 GWatt per yr. In the 10-year term.  Not terrestrial PV, not nuclear, not the US, not China.

Nor in space either. Not soon, anyway. If a thing is impossible, then it won't happen.

Impossible how?
I mean, I get it's ambitious, but how is it impossible?

(Not the relative pricing argument, the deployment rate.)

Let me quote from the SpaceCloud article:

It is, therefore, conceivable that 5 GW of compute could be deployed with fewer than 100 launches, with a similar number of launches required for the combined solar/radiator modules of Starcloud’s design.

So if 5 GW takes 200 launches, 100 GW will need 4000. Per year. In the next five years!

At the risk of repeating myself, I'll repeat myself: :-) If a thing is impossible, then it won't happen.

[Greg Hullender]

That's a big problem for Starcloud or Google Suncatcher - but not for potential AI data centers hosted by SpaceX itself on Starlink satellites.

Fair enough, but see my immediately previous post about needing 200 Starship launches to put a single 5 GW data center in LEO. (Never mind trying to put it on the moon.) And that includes solving the robotic assembly and maintenance problems, plus working out station keeping for a 20,000 ton object, and all the other niggling details.

[InterestedEngineer]

That's a big problem for Starcloud or Google Suncatcher - but not for potential AI data centers hosted by SpaceX itself on Starlink satellites.

Fair enough, but see my immediately previous post about needing 200 Starship launches to put a single 5 GW data center in LEO. (Never mind trying to put it on the moon.) And that includes solving the robotic assembly and maintenance problems, plus working out station keeping for a 20,000 ton object, and all the other niggling details.

It's not a 20,000 ton object, it's 20,000 ton cluster about 1km in diameter, with each element of the cluster around 5-10t.  Think of each satellite as a 42U compute rack, interconnected by multiple Tbps optical links.

Station keeping is ion thrusters, just like Starlink. 

If each "rack satellite" is 10t that gives us 2,000 satellites in the cluster.

Just do do the math, at 20,000t at 5GW is 0.25kW/kg.   Current data center racks are running roughly .02kW/kg so that's gonna be a big improvement in power density.

Starlink is running roughly 2kW per 1000kg or .0002kW/kg.

So seems a tad optimistic on the power density (by tad, I mean an entire order of magnitude)

[envy887]

Spinlaunch has already produced operational mass driver capable of 5000mph. Lunar escape velocity is 5325mph.

You definitely don't want to have to deal with the constraints Spinlaunch imposes.    An EM linear accelerator would be the way to go.

What constraints?. Bulk water and metal shipments don't care about high Gs involved. Delicate items like humans will use landers.
Best to run launchers in pairs, energy recovered while slowing one down to reload after a launch can be used to accelerate another to launch RPMs.

Is Spinlaunch still dropping the counterweight? AFAIK most of the energy will be lost with the counterweight.

If the counterweight is 10x the mass at 1/10th the radius, then it carries only 10% of the energy.

ISTM that rotating tethers would be ideal for balancing energy flow if there is nearly equal mass going up and down - no hypersonic catches required, just a zero altitude, zero instantaneous velocity docking.

[meekGee]

The reason for going to orbital PV is that nobody else has a path for installing 100 GWatt per yr. In the 10-year term.  Not terrestrial PV, not nuclear, not the US, not China.

Nor in space either. Not soon, anyway. If a thing is impossible, then it won't happen.

Impossible how?
I mean, I get it's ambitious, but how is it impossible?

(Not the relative pricing argument, the deployment rate.)

Let me quote from the SpaceCloud article:

It is, therefore, conceivable that 5 GW of compute could be deployed with fewer than 100 launches, with a similar number of launches required for the combined solar/radiator modules of Starcloud’s design.

So if 5 GW takes 200 launches, 100 GW will need 4000. Per year. In the next five years!

At the risk of repeating myself, I'll repeat myself: :-) If a thing is impossible, then it won't happen.

What's impossible about a tower launching 10/day, and 10 active towers?

That's 3000/month.

Ambitious, for sure.  But that's what this system is being designed for.

You see jetliners queue up at the end of the runway and depart every couple of minutes. That used to be unthinkable too. Who would even need that?!

This here is 100x slower rate per tower/runway.

Not impossible by any means.

[DanClemmensen]

What's impossible about a tower launching 10/day, and 10 active towers?

Not to the same datacenter. Works for ten datacenters in different planes. For each tower, you will also need a catch tower to avoid launch/catch contention, about three Ships per plane, and single shared REALLY busy Booster.

[Action]

The reason for going to orbital PV is that nobody else has a path for installing 100 GWatt per yr. In the 10-year term.  Not terrestrial PV, not nuclear, not the US, not China.

Nor in space either. Not soon, anyway. If a thing is impossible, then it won't happen.

Impossible how?
I mean, I get it's ambitious, but how is it impossible?

(Not the relative pricing argument, the deployment rate.)

Let me quote from the SpaceCloud article:

It is, therefore, conceivable that 5 GW of compute could be deployed with fewer than 100 launches, with a similar number of launches required for the combined solar/radiator modules of Starcloud’s design.

So if 5 GW takes 200 launches, 100 GW will need 4000. Per year. In the next five years!

At the risk of repeating myself, I'll repeat myself: :-) If a thing is impossible, then it won't happen.

What's impossible about a tower launching 10/day, and 10 active towers?

That's 3000/month.

Ambitious, for sure.  But that's what this system is being designed for.

Greg seems to be misreading the proposed timeline.  Everything I can see in that SpaceCloud article seems to be years away and assuming a mature Starship or equivalent vehicle, which as you point out, is planned to achieve this but not right away.  The white paper says "launch cadence will not be a bottleneck long term."

I may be wrong, but I don't see where he's getting the idea that we're supposed to be launching 100 GW/year in five years.

But maybe I misread Greg.  "Impossible" is a strong term for "something that may be ten or fifteen years out."

[TrevorMonty]

What's impossible about a tower launching 10/day, and 10 active towers?

Not to the same datacenter. Works for ten datacenters in different planes. For each tower, you will also need a catch tower to avoid launch/catch contention, about three Ships per plane, and single shared REALLY busy Booster.

The launched containers only need beacon and simple gas RCS to stabilize it. Use space tugs to retrieve and deliver to orbital processing centre.  Ideal local for data centres is around EML1 as deliveries from earth don't need to deal with lunar gravity well. Space around EML1 is clean for now and station keeping is low compared to earth orbit. Also no reflective heat from earth for the radiators to contend with.

[meekGee]

What's impossible about a tower launching 10/day, and 10 active towers?

Not to the same datacenter. Works for ten datacenters in different planes. For each tower, you will also need a catch tower to avoid launch/catch contention, about three Ships per plane, and single shared REALLY busy Booster.

I think 1 catch tower per 1 launch tower is an obvious requirement.

The planes is indeed an issue if you're only going to DD SSOs.  But we don't know the orbital plan.  That's an important issue though.

Mind you Musk's 100 MWatt/yr matches this very crude estimate, and he did mention high orbit rather than DD SSO which might be for this exact reason.

When people say "impossible", I expect a capital I. Not some "wait that's too large", especially when within the design parameters of the system...  Maybe "over ambitious" should apply instead.

[Coastal Ron]

I thought there was an NSF thread or conversation that was looking at mass drivers on the Moon, and they determined (or came to the conclusion) that in reality they really don't work as well as we thought.

For instance, whatever mass you launch is either going to be semi-orbital (which means you have to have a catcher in orbit that can snatch the payload before it falls back to the Moon) or it leaves the orbit of the Moon, in which case you have to go chase it somehow, but it is difficult to have a "catcher" prepositioned (something to do with unstable Moon orbits or??).

In any case, I think the burden of proof is on those that think we can "manufacture" anything on the Moon, much less something as high tech as solar cells in large volume (as well as everything it takes to install those panels so they can be useful), at any point in near future. I think it is more likely we could have a small Earth-supplied colony on Mars well before we could figure out solar cell manufacturing on the Moon.

Blue have already demostrated they can manufacture a solar cell from regolith (artificial). Extracted silicon by electrolysis and then produced cell from it.

Sure, Blue Origin says:

Using regolith simulants, our reactor produces iron, silicon, and aluminum through molten regolith electrolysis, in which an electrical current separates those elements from the oxygen to which they are bound. Oxygen for propulsion and life support is a byproduct.

I've never doubted the materials needed were on the Moon, my whole point is that refining the needed material from raw lunar material, and then transforming that material into finished products, takes a factory.

Not like some box that you can carry, but here on Earth a solar cell manufacturing facility for high volume production would be the size of an American football field. So not only do you have to ship all that high tech equipment to the Moon, then install it and calibrate it, but how do you maintain such delicate machinery? Ooh, and the chicken and egg challenge is how do you power such a factory if it can't yet product the solar cells needed to power it? 

And no, if we can't fully automate a factory here on Earth, then we won't be able to do it on the Moon. Which is why I think a lot of people are not understanding how hard it will be to build solar cells on the Moon.

Spinlaunch has already produced operational mass driver capable of 5000mph. Lunar escape velocity is 5325mph.

If you look at what I wrote, I didn't dispute that mass could be thrown into space. My post talked about the challenge of CATCHING whatever was launched. How do you stop something that is traveling X kph? What happens to the kinetic energy? How exact does the launch need to be in order for the payload to be caught? THAT is what I remember NSF members talking about, and how they thought that would be extremely difficult.

[TrevorMonty]

I thought there was an NSF thread or conversation that was looking at mass drivers on the Moon, and they determined (or came to the conclusion) that in reality they really don't work as well as we thought.

For instance, whatever mass you launch is either going to be semi-orbital (which means you have to have a catcher in orbit that can snatch the payload before it falls back to the Moon) or it leaves the orbit of the Moon, in which case you have to go chase it somehow, but it is difficult to have a "catcher" prepositioned (something to do with unstable Moon orbits or??).

In any case, I think the burden of proof is on those that think we can "manufacture" anything on the Moon, much less something as high tech as solar cells in large volume (as well as everything it takes to install those panels so they can be useful), at any point in near future. I think it is more likely we could have a small Earth-supplied colony on Mars well before we could figure out solar cell manufacturing on the Moon.

Blue have already demostrated they can manufacture a solar cell from regolith (artificial). Extracted silicon by electrolysis and then produced cell from it.

Sure, Blue Origin says:

Using regolith simulants, our reactor produces iron, silicon, and aluminum through molten regolith electrolysis, in which an electrical current separates those elements from the oxygen to which they are bound. Oxygen for propulsion and life support is a byproduct.

I've never doubted the materials needed were on the Moon, my whole point is that refining the needed material from raw lunar material, and then transforming that material into finished products, takes a factory.

Not like some box that you can carry, but here on Earth a solar cell manufacturing facility for high volume production would be the size of an American football field. So not only do you have to ship all that high tech equipment to the Moon, then install it and calibrate it, but how do you maintain such delicate machinery? Ooh, and the chicken and egg challenge is how do you power such a factory if it can't yet product the solar cells needed to power it? 

And no, if we can't fully automate a factory here on Earth, then we won't be able to do it on the Moon. Which is why I think a lot of people are not understanding how hard it will be to build solar cells on the Moon.

Spinlaunch has already produced operational mass driver capable of 5000mph. Lunar escape velocity is 5325mph.

If you look at what I wrote, I didn't dispute that mass could be thrown into space. My post talked about the challenge of CATCHING whatever was launched. How do you stop something that is traveling X kph? What happens to the kinetic energy? How exact does the launch need to be in order for the payload to be caught? THAT is what I remember NSF members talking about, and how they thought that would be extremely difficult.

Should end up in orbit if velocity and trajectory is correct. Container will need radio beacon so it can be located and picked up by.

[Greg Hullender]

Greg seems to be misreading the proposed timeline.  Everything I can see in that SpaceCloud article seems to be years away and assuming a mature Starship or equivalent vehicle, which as you point out, is planned to achieve this but not right away.  The white paper says "launch cadence will not be a bottleneck long term."

I may be wrong, but I don't see where he's getting the idea that we're supposed to be launching 100 GW/year in five years.

But maybe I misread Greg.  "Impossible" is a strong term for "something that may be ten or fifteen years out."

The original claim was that something was needed in the next five years to handle the power load for AI, which would supposedly be needing 100 GW/year by then. I think it's impossible for that to be done in Space in that amount of time. The SpaceCloud article was offered as evidence that this could be done. But, as you say, SpaceCloud isn't proposing to do anything like that.

What's impossible isn't that it could never happen. It's just impossible for it to happen in a timeframe that's material to AI datacenter requirements.

[Greg Hullender]

What's impossible about a tower launching 10/day, and 10 active towers?

That's 3000/month.

Ambitious, for sure.  But that's what this system is being designed for.

You see jetliners queue up at the end of the runway and depart every couple of minutes. That used to be unthinkable too. Who would even need that?!

This here is 100x slower rate per tower/runway.

Not impossible by any means.

Someday, maybe. But this is not going to be happen in the next five years, and, by that point, the AI power crisis will have been resolved some other way.

[Greg Hullender]

It's not a 20,000 ton object, it's 20,000 ton cluster about 1km in diameter, with each element of the cluster around 5-10t.  Think of each satellite as a 42U compute rack, interconnected by multiple Tbps optical links.

Station keeping is ion thrusters, just like Starlink. 

If each "rack satellite" is 10t that gives us 2,000 satellites in the cluster.

I'm pretty sure that's not what's proposed in the paper! The elements of the cluster are clearly intended to be rigidly connected as close together as possible to minimize latency. Can you point to anything in the paper that says otherwise? Also, from the other sources, elements in a data center need to be within 50m of each other, so a 1 km diameter sphere would make for a terrible data center!

Just do do the math, at 20,000t at 5GW is 0.25kW/kg.   Current data center racks are running roughly .02kW/kg so that's gonna be a big improvement in power density.

Starlink is running roughly 2kW per 1000kg or .0002kW/kg.

So seems a tad optimistic on the power density (by tad, I mean an entire order of magnitude)

Much of the mass is for solar panels and cooling, but that wouldn't explain why they thought they needed that much power. Do you think SpaceCloud just made a gross error in their calculations? If the actual power requirements for datacenters are so low, why is everyone so worried about it?

[Coastal Ron]

Spinlaunch has already produced operational mass driver capable of 5000mph. Lunar escape velocity is 5325mph.

If you look at what I wrote, I didn't dispute that mass could be thrown into space. My post talked about the challenge of CATCHING whatever was launched. How do you stop something that is traveling X kph? What happens to the kinetic energy? How exact does the launch need to be in order for the payload to be caught? THAT is what I remember NSF members talking about, and how they thought that would be extremely difficult.

Should end up in orbit if velocity and trajectory is correct. Container will need radio beacon so it can be located and picked up by.

By what? How will that work?

What size payload mass is being launched, what speed will it be going when it is "caught", how much fuel does it take to operate this system, etc., etc.

See, this is the "Then a miracle occurs" part of this that everyone seems to think will be easy, yet no one seems to have thought it out... 

[Twark_Main]

It's how you used it. It didn't occur to me that someone would know the reference and so thoroughly miss the message. 

It's satire. Self deprecating.  Stopping the sermon to have a chuckle. Lighten up...

Great joke. I'm in stitches. However the giant gaping logical hole in the argument remains.

Remember Elon's Law: "The requirements are always at least a little bit wrong."  If the assumption about limitless AI demand is wrong, it would be nice to know that now rather than later...

[Twark_Main]

That does seem problematic. I think Starship will drastically reduce launch cost, but $10/kg is a stretch. That might be doable for a really mature Starship-type system (as mature as cars or airliners are today), where fuel cost is a significant part of the lifetime cost, but not anytime soon.

So much depends on this assumption that SpaceX will offer launch services at cost (and at the low-end of what even Elon has estimated for those costs) that it's worth harping on this a little. SpaceX will not offer customers prices that are much lower than what its competitors charge! This is elementary economics. If competitors are charging $1000/kg, then SpaceX does better to charge $999/kg even if it only costs them $10/kg. Then they pocket the huge margins and apply that money towards their Mars project. They'd be stupid not to, and although Elon may be a fool, he's not that kind of fool!

Tell us, which launcher is $1/kg more expensive than F9? If that really is SpaceX's pricing strategy, then they should be doing it right now, correct? 

The other possibility is that the (economically "irrational") goal of the company really is to advance space technology as fast as possible, not just maximize RoI. As Musk used to constantly reminds us, if all he wanted to do was choose a venture with the best risk-adjusted RoI, then starting a space company is probably the worst choice imaginable.


Remember, the same "fool" admitted that opening up the Tesla charging network would probably hurt Tesla economically, but it was the right thing to do to advance the company's mission. So yes, apparently he is that much of a "fool."


I don't think people know quite what to do when a company actually does work toward its stated mission, not just amoral profit maximization.... 

[stormhelm]

Good luck..
I`ll believe it when it happens if ever before the Chinese plant their flag..