Shenzhou-10: China launches next manned space mission

[PeZook]

Yeah, it was more "hey guys we bought this here massive rocket and it's awesome, but what, are you gonna tell me our guys will just walk around and plant a flag? We gotta make better use of all this hardware than that!"

Still, NASA had lots of stuff planned for post-J missions, but alas - bombing yellow people to spare them the disease of communism was way more important!

And then you got the Shuttle which was supposed to be cheaper but wasn't (keeping the Saturn V would've been so much more economical )

[energiewende]

Mass production - The US 70mm rocket is an unguided piece of ammunition that has been produced to number millions. Each rocket costs $600 (inflation adjusted) and weighs 6.2kg, for a cost per tonne of $95,000.

http://asafm.army.mil/Documents/OfficeD ... //ammo.pdf page 3

A Falcon 9 costs costs $54m and weighs 330t, for a cost per tonne of $160k.

So assuming that Falcon 9 is of no greater complexity than piece of disposable ammunition I can throw over my head, mass production cannot conceivably even halve costs. Interestingly SpaceX claimed in 2005 when it was first announced Falcon 9 would cost $27m/flight - interestingly almost exactly the same as the Hydra 70 per tonne in then-dollars.

There's no free lunch here. The cost of space is anchored tightly to the cost of dumb rockets, not special engineering processes, billion dollar wrenches and NASA bureaucracy. Space has to get >10x cheaper to be practically rather than just scientifically interesting, and for that to happen, so need a lot of other things.



Scientific interest - Apologies, I meant manned space missions. All the examples you gave were of unmanned, which I grant are useful (and much cheaper).



Re-fuelling satellites, etc. - Come on, it's just silly. Remember you still have to launch the fuel to the depot, then either launch the fuel from the depot to the satellite or launch the satellite to the depot and back again. This only gets really interesting if you can manufacture fuel in space. I grant there may be some saving in building less reliable rockets to refuel unmanned devices, etc. - but why not just fly unmanned rocket to an unmanned satellite and skip the (manned?!) depot station.



Space vs stopping Communism - Saving 50m South Koreans from the horrors of Kim Il Sung's gulags isn't just more important, more useful, and more morally creditworthy than the space program, it's more of all those things than practically anything the US government does or has done since. If you think Vietnam was different, you're wearing the winner-writers-history goggles.

[someone_else]

So assuming that Falcon 9 is of no greater complexity than piece of disposable ammunition I can throw over my head, mass production cannot conceivably even halve costs.

This. Can you please explain what kind of broken logic allows you to assume that the level of complexity of a Falcon 9 is comparable to a glorified firework (where it's "engine" is a fucking metal tube with solid rocket fuel and its "nozzle" is a fucking metal cylinder with a conic hole in it) that is not even sub-fucking-orbital?

Seems to me that this is another example of "Huh let's compare the ISS to CERN".

Scientific interest - Apologies, I meant manned space missions.

You are shifting the goal posts, this may cause others to call you bad names. Anyway, what I linked still applies . Also this has some more links. Men can do better than machines at collecting data and stuff, and this is beyond any doubt. We didn't send men only because costs were too fucking high to do it without the US gov to pay the bills, not because an unmanned probe has better performance.

you still have to launch the fuel to the depot,

One rocket sends in LEO a big metal tank filled with fuel in the same orbit of the orbital tug. That's the "fuel depot". Just a glorified metal tank. The one with mating equipment, pumps and whatnot is the tug.

launch the fuel from the depot to the satellite or launch the satellite to the depot and back again.

There is no "launch", only stuff that goes from surface to orbit is "launched". Orbit-to-orbit movement is not called like that.

The tug docks with the depot and refuels itself (as it is cryogenic, the satellite's own maneuvering systems use another kind of fuel, more on that later), that then mates with the satellite (launched in the same LEO orbit as the propellant depot, so it's not hard to get there with minimal delta-v, and no, we can do that since Mercury missions and Shuttle did it every time with ISS so it's not hard).
Then does what it was supposed to do, brings the satellite to its orbit by burning fuel.
Then detaches the sat that will spring to life and start to do its job.
Now, lo and behold, it decelerates by areobraking and manages to return to LEO without using a lot of fuel, where it is ready to do its job again.

In case it wanted to refill a satellite's own tanks, it's even easier, as now the payload is just the fuel tank and pumps for the satellite, which is going to be tiny compared to a full satellite, so fuel expenditure will be far less.
When it arrives to the sat it mates with it and pumps the fuel in its tanks, then detaches and aerobrakes to LEO again.
Such payload will likely piggy-back the ride with another bigger payload on a rocket inbound to LEO, so it won't cost a lot.

why not just fly unmanned rocket to an unmanned satellite and skip the (manned?!) depot station.

I never said it was manned. The tug is actually a mildly modified Centaur upper stage. The Centaur is used as the last stage of a multistage rocket, the one responsible of placing stuff at the right orbit, it has automatic celestial navigation systems and is smart enough to do everything a tug needs to, the guidance from earth will come only during the close approaches and refuelling operations where men will tele-operate the beams and tooled arms. Rockets only carry stuff to LEO.

Space vs stopping Communism

This has nothing to do with Vietnam nor alleged moral reasons of saving people from evil Communists, let's not derail.
The space race was a pissing contest between USA and USSR. USSR started it by sending Sputnik and then the first man in space, then the first probes on the Moon and on Venus. It was not just a mere achievement of a nation, or of science, it was a statement that the Communist way of life was inherently better than the US way of life.
Or the US read it as such. That's Cold War mentality, pretty sick shit.
The US had to prove that it was better than the Soviets.
So they decided to go to the Moon. The Soviets did fail by a few months, but were in the race as well. When the US planted the flag they scrapped the program and kept everything secret ("us? The moon? Nah, never ever thought about it"). Today it is no more, so we know what Soviet Moon Missions were supposed to be.
It was a "virtual" war, they couldn't go to war with one another (nukes or not it would have been a suicide even if it wasn't completely idiotic from a logistics standpoint), so they resorted to these pissing contests (and all the smallish wars on border countries like Vietnam or Afghanistan, yes even back then that shithole was contested), and trying to develop even better weaponry than the other, in an escalating circle that eventually dealt a fuckton of damage to both anyway.

[energiewende]

So assuming that Falcon 9 is of no greater complexity than piece of disposable ammunition I can throw over my head, mass production cannot conceivably even halve costs.

This. Can you please explain what kind of broken logic allows you to assume that the level of complexity of a Falcon 9 is comparable to a glorified firework (where it's "engine" is a fucking metal tube with solid rocket fuel and its "nozzle" is a fucking metal cylinder with a conic hole in it) that is not even sub-fucking-orbital?

Seems to me that this is another example of "Huh let's compare the ISS to CERN".

Space rockets being more complex than Hydra 70 reduces the scope for saving, ie. more of the higher cost per tonne must be inferred to be inherent rather than a result of shorter production runs.

And I see no good answer why the ISS is a better use of money than funding CERN for 100 years (especially since it's only 50 years old); Im not sure such a claim is seriously defensible anyway.

You are shifting the goal posts, this may cause others to call you bad names. Anyway, what I linked still applies . Also this has some more links. Men can do better than machines at collecting data and stuff, and this is beyond any doubt. We didn't send men only because costs were too fucking high to do it without the US gov to pay the bills, not because an unmanned probe has better performance.

I don't think that is true, eg. why do we not have a man with a porthole to collect data from the LHC, or men with binoculars in planes rather than radar? Machines have much higher spatial and time resolution, can accept and store more data, don't get bored, don't blink, in fact it's really difficult to think of any way machines arent far superior or data collection. The most advanced telescopes don't use manual data collection on earth either.

you still have to launch the fuel to the depot,

One rocket sends in LEO a big metal tank filled with fuel in the same orbit of the orbital tug. That's the "fuel depot". Just a glorified metal tank. The one with mating equipment, pumps and whatnot is the tug.

launch the fuel from the depot to the satellite or launch the satellite to the depot and back again.

There is no "launch", only stuff that goes from surface to orbit is "launched". Orbit-to-orbit movement is not called like that.

The tug docks with the depot and refuels itself (as it is cryogenic, the satellite's own maneuvering systems use another kind of fuel, more on that later), that then mates with the satellite (launched in the same LEO orbit as the propellant depot, so it's not hard to get there with minimal delta-v, and no, we can do that since Mercury missions and Shuttle did it every time with ISS so it's not hard).
Then does what it was supposed to do, brings the satellite to its orbit by burning fuel.
Then detaches the sat that will spring to life and start to do its job.
Now, lo and behold, it decelerates by areobraking and manages to return to LEO without using a lot of fuel, where it is ready to do its job again.

In case it wanted to refill a satellite's own tanks, it's even easier, as now the payload is just the fuel tank and pumps for the satellite, which is going to be tiny compared to a full satellite, so fuel expenditure will be far less.
When it arrives to the sat it mates with it and pumps the fuel in its tanks, then detaches and aerobrakes to LEO again.
Such payload will likely piggy-back the ride with another bigger payload on a rocket inbound to LEO, so it won't cost a lot.

You've got a depot to store fuel, so before one can store fuel it first must be sent to the depot. Then, it is indeed launched into a higher (or lower, both cost energy) orbit. I'm struggling to understand what the whole depot stage is even for - you could just shoot low-reliability rockets directly to satellites to refuel them. The reason this isn't done is that it's more economic to simply replace them, especially as by the time they run out of fuel all the systems they are carrying tend to be obsolete anyway.

There is scope for reasonable disagreement on this and many other policy issues but I think most of your ideas are a result of basic misunderstandings.

Space vs stopping Communism

This has nothing to do with Vietnam nor alleged moral reasons of saving people from evil Communists, let's not derail.

I agree, I don't want to derail either, and I overegged the argument too. But my point is, if you see nothing wrong with manned space missions on a cost:benefit basis the Vietnam War should not only have been funded, but funded dramatically more than it was. Sure it caused a lot of damage, to innocents as well as the guilty, but it at least had a large potential good payoff. Manned space, not so much.

[someone_else]

more of the higher cost per tonne must be inferred to be inherent rather than a result of shorter production runs.

Again bullshit. According to what are you assuming it must be inherent to the device?

Mass production always reduced the cost of making complex stuff (by investing in expensive automation of the assembly lines). It's the entire point of it since the Industrial revolution. decreasing the cost of stuff that before was available only to the very welathy.

And I see no good answer why the ISS is a better use of money than funding CERN for 100 years (especially since it's only 50 years old); Im not sure such a claim is seriously defensible anyway.

It's not better, it's different.
Space endeavours are a different field than particle physics, so whatever one can discover the other cannot.
Also, for the sake of reversing the question, let's see if CERN actually deserves funding for the next 100 years.
Please name some of the great discoveries that impacted the life of us all that came from CERN, since you pulled it out.
I already listed stuff for space exploration (of which ISS is a non-trivial part even if you keep denying it).

I don't think that is true, eg. why do we not have a man with a porthole to collect data from the LHC, or men with binoculars in planes rather than radar?

Again trying to twart my words. I'm getting tired of this.
Probes stopped sucking donkey balls only in recent times. Up until the 1990 sending a man was better as far as data-collection ability goes. Then of course he will have to operate instruments and not just stare at rocks with his eyeballs you dumbfuck, we are a tool-using species. We have lost quite a few probes due to dumb software errors and other crap that would have not happened with a man in charge.
Even now, for exploration of anything farther than the moon, where light-lag is significant, it's better to have men closer to the probes to ensure a full-telepresence.

You've got a depot to store fuel, so before one can store fuel it first must be sent to the depot.

No, the "depot" is a fucking fuel tank, that is launched full of fuel as payload of a rocket. It's just called "depot".

you could just shoot low-reliability rockets directly to satellites to refuel them.

Because the cost would be the same as putting a new satellite in orbit. The reusable orbital tug is reusable for a reason. It is heavy and the equipment it uses to mate and refuel satellites is expensive.

The reason this isn't done is that it's more economic to simply replace them, especially as by the time they run out of fuel all the systems they are carrying tend to be obsolete anyway.

And you base this bullshit again on what?
Please provide reliable sources backing that for commercial satellites.

Commercial satellites are replaced only in case of system malfunction OR when they end the fuel and can no more be kept safely in their orbit (yes, they need fuel to keep their orbit and avoid the odd piece of orbital trash).
Why? Because it costs a fuckton of money to replace them.
Read this.

But my point is, if you see nothing wrong with manned space missions on a cost:benefit basis the Vietnam War should not only have been funded, but funded dramatically more than it was.

Stop derailing the thread. This is irrelevant speculation that does not belong in this thread. Start a new one if you really want to discuss this (and get spanked by the board's military experts).

[energiewende]

Mass production - You're not-quoting the important bit again. Hydra 70 is already as simple as it gets and is already mass produced. It therefore represents some approximate minimum cost per tonne. It seems reasonable that super mass production (1000s or 10,000s of units) might get a 1/3 or so cost saving, which is nothing like what is needed. What you plan to do with those 1,000s of rockets is another matter.



What has CERN done for us anyway, huh? - The invention of the World Wide Web. Since you asked for one, and insisted on practicality (I make no such claims for manned space, I'd happily accept an important theoretical discovery). There are many more.

Is this any surprise? No not really; any lab that employs 6,000 people is going to make more discoveries than a lab that employs 6 people just by the law of large numbers. ISS could alternatively the fund the entire German scientific establishment for 50 years. The ISS is directly competing with mass employers of scientists and PhDs that have each produced a dozen or more Nobel Prizes in Physics.

Actually even your own signature seems to agree about the "realism" of manned space:

"Stereotypical spacecraft are pressurized.
"Less realistic spacecraft are pressurized to hold breathing atmosphere."



Refuelling satellites - This is a sales brochure for a start-up, written by its directors. "Reusability" of the tanks aside, the reason this is an obscure start-up rather than an established practice is that you're burning 100t of fuel to deliver 100kg of fuel to keep a 1980s radio transceiver going for another 10 years.

[Jub]

Refuelling satellites - This is a sales brochure for a start-up, written by its directors. "Reusability" of the tanks aside, the reason this is an obscure start-up rather than an established practice is that you're burning 100t of fuel to deliver 100kg of fuel to keep a 1980s radio transceiver going for another 10 years.

That's still cheaper than putting up a new satellite and that fuel can come as extra cargo on an already scheduled flight that has some spare capacity. Even with satellites not designed for refueling you could still make an add-on thruster pack with on board fuel supply and dock that. It's still cheaper than a new satellite and could even maneuver itself to the target with relatively simple guidance.

This sort of innovation could be big if properly invested in. It also means that companies could possibly get a twofer on new satellite launches, they put a new one up and refuel one or more additional satellites with any free cargo space. That makes the overall utility of the launch higher and, in a roundabout way, makes launches more cost effective.

EDIT: Also it seems like you think that space flight is worthless when in reality any space launch is a step forward even if it's just sending supplies to the ISS. Unlike physics, which has been practiced for centuries, space travel is still less than a hundred years old and manned space flight even younger. We're in the early years still and that means great costs for sometimes limited gain.

However in the future mining space and colonizing it will be important. Resources are abundant among the stars and the sooner we start getting out there the better. If that means spending on a pork project like the ISS until somebody grows the balls to push further that's still a worthy thing.

[someone_else]

Hydra 70 is already as simple as it gets and is already mass produced. It therefore represents some approximate minimum cost per tonne.

Being that a liquid-fueled rocket, it's better if you find an actual liquid-fueled rocket to base your comparisons on.
What about the V2? There is an interesting essay about that, read it here.

What has CERN done for us anyway, huh? - The invention of the World Wide Web.

An important step of the evolution of Internet into what it is now, but it was a step of an incremental progress that happened in large parts outside of it. Ever heard of Arpanet? Internet History

Is this all? That could have been done by any university. It didn't require any of the CERN's labs and high-tech equipment.

Is this any surprise? No not really; any lab that employs 6,000 people is going to make more discoveries than a lab that employs 6 people just by the law of large numbers.

I like this approach to science MORE NUMBERS MEAN MORE SCIENCE HUR HUR.
By this logic ITER and fusion research would have discovered a butt-load of stuff by now (there are literally hundreds of thousands of people involved and the costs are staggering). Yet, they didn't. And it's at least 30 years that fusion technology is still "more than 20 years away".

Also, what makes you assume that what happens on the ISS only occupy only 6 people again? They are just there to execute the experiments that labs around the world send or ask them to do. They don't process the data acquired (the scientists back at their labs on Earth do and integrate that with their own experiments). The ISS's crew has more of a lab assistant's job, than "whole research team". So if you want to do a comparison you might need to get an idea about how much people is actually involved.

"Reusability" of the tanks aside,

It's not the tank that it is reusable, it's the orbital tug. You know what is an orbital tug? It's a small rocket that remains in orbit forever and moves around the orbits without ever landing. Did you look at the pic in the wikipedia page about the Centaur? That's what it looks like.

The tank can be reusable or not, and it will matter little, as it is just a few tons of iron. The tug has engines and the mating equipment and it is the one doing the job, and is reused.

you're burning 100t of fuel to deliver 100kg of fuel to keep a 1980s radio transceiver going for another 10 years.

Yeah, all satellites are prehistoric crap none really uses anymore.

Commercial satellites have a shorter service life than 10 years or even 30 like in your example (and mostly due to the size of their fuel tanks, limited by the max payload deliverable to their current orbit limited in turn by the rockets we have atm). Most of what is up now is less than 15 years old and was put there with additional allowance for growth (as you cannot simply double your satellite fleet on moment's notice).

Anyway, that is just one of the possibilities. The other is using the tug (the rocket in space) to move the satellite to its orbit, and allow the cost reduction as I said a few posts above. Which does make economic sense even now.

Also companies like say Iridium or any satellite telecommunication company or the sat TV companies have launched a bunch of satellites, and if you notice, they did launch quite a few spares.
If a system like this was in place, they could save tons of money by sending up less spare satellites and purchasing the services of these tugs only when needed. Which cost less and aren't a fixed cost (that is you pay only if a satellite needs assistance)

Another thing that this allows is the re-purposing of old satellites, they can be sold by their previous owner as "used" to someone else that will re-purpose them.

the reason this is an obscure start-up

There is also another company (not a startup) having a try here
NASA listing good reasons here
And is also testing other relevant technology.
And some posts above I linked the DARPA's Phoenix project.

You see? It's a conspiracy! They all want to do it. Stop them! Save the Kittens!

EDIT: forgot this:

Actually even your own signature seems to agree about the "realism" of manned space:

My sig is about realism in fiction, as the last part demonstrates:
Realistic spacecraft are pressurized because they are flying propellant tanks. And with cryogenic fuels like the average rocket, any fuel tank is pressurized.
Even if the vehicle is unmanned, it's still pressurized because its mostly fuel tanks and fuel tanks are pressurized.

[someone_else]

addendum:

this is an obscure start-up

ViviSat is formed by ATK aerospace and US Space LLC Which are two companies operating in aerospace (ATK makes the Orion's escape tower among other things), and the second provides satellite-based "voice, data and video" services to the US government.

I would not call it "obscure startup". They aren't newcomers in the scene nor cashless nerds.

[Starglider]

By this logic ITER and fusion research would have discovered a butt-load of stuff by now

Fusion research has resulted in numerous discoveries in plasma physics (relevant to astronomy, industrial processes using plasmas and yes, space engines) and technological spin-offs in high-power lasers, exotic optics (e.g. for advanced semicondutor litho) and applied nuclear science (e.g. cheap neutron generators).

[PeZook]

Mass production - The US 70mm rocket is an unguided piece of ammunition that has been produced to number millions. Each rocket costs $600 (inflation adjusted) and weighs 6.2kg, for a cost per tonne of $95,000.

http://asafm.army.mil/Documents/OfficeD ... //ammo.pdf page 3

A Falcon 9 costs costs $54m and weighs 330t, for a cost per tonne of $160k.

So assuming that Falcon 9 is of no greater complexity than piece of disposable ammunition I can throw over my head, mass production cannot conceivably even halve costs. Interestingly SpaceX claimed in 2005 when it was first announced Falcon 9 would cost $27m/flight - interestingly almost exactly the same as the Hydra 70 per tonne in then-dollars.

There's no free lunch here. The cost of space is anchored tightly to the cost of dumb rockets, not special engineering processes, billion dollar wrenches and NASA bureaucracy. Space has to get >10x cheaper to be practically rather than just scientifically interesting, and for that to happen, so need a lot of other things.

Oh my God, are you kididng me?

You have just compared the price of a solid-fuel rocket motor (yes, that's right: 500$ is the price of the MOTOR, not the entire Hydra rocket) with the MISSION COST of the complete (and fuelled!) space rocket. The hell was that?

Now, an actually meaningful comparison would see how rocket motors on both sides size up, and that's (funnily enough!) the exact opposite of your conclusion. The RS-68 costs 14 million per engine ; J-2X is projected to cost 25 million, SSME - 30-40 million ; RL-10 - 38 million. So we're talking a price-per-ton of several million.

These are all the "classic" engines - note how the RS-68 and RL-10 engines in a Delta-IV are actually the MAJORITY of the cost of each 140 million launch. Hell, engines in the Centaur upper stage alone cost something like 70 million ; We'll come back to that in a moment.

Now, the Falcon 9, which uses mass production techniques you think are irrelevant, managed to reduce the cost of its engines to below 1 million. That reduced engine costs to just around 16% of the total launch cost.

Now, with mass production comes another benefit, that is amortization of fixed costs. Out of every launch right now, about equal parts of the mission cost are in procedures and hardware. With four Falcon 9s per year, these costs don't amortize very well ; With more launches, the fixed costs get amortized, and thus price per launch goes down. Elon Musk seems to think 500$ per kilogram to LEO is completely achievable ; So far he's been reasonably good at delivering, even if the Falcon 9 is actually about as cost effective as the Soyuz right now.

Scientific interest - Apologies, I meant manned space missions. All the examples you gave were of unmanned, which I grant are useful (and much cheaper).

Again you've ignored me pointing out that if "made major discoveries" is the standard, then nearly all earthside labs are completely useless, too. If the scientific community thought the ISS was useless, there would be no need to sift through all the proposals for experiments that are flooding NASA and have to be prioritized before every mission.

You also never admitted that your original position was wrong, BTW. Shifting goal posts is not looked very favorably upon here.

Re-fuelling satellites, etc. - Come on, it's just silly. Remember you still have to launch the fuel to the depot, then either launch the fuel from the depot to the satellite or launch the satellite to the depot and back again. This only gets really interesting if you can manufacture fuel in space. I grant there may be some saving in building less reliable rockets to refuel unmanned devices, etc. - but why not just fly unmanned rocket to an unmanned satellite and skip the (manned?!) depot station.

Funny that NASA does not think fuel depots are pointless then.

The point is that the refuelling tug would be reuseable, which means you ONLY have to launch the fuel, rather than an entire upper stage every time (such as the massively expensive Centaur) ; And fuel is really cheap, so it can be launched on cheap, light rockets that do not require massive ground facilities to prepare and service - if you lose a payload of fuel, well who cares?

Plus of course you can piggyback fuel missions on rockets that would've never went above LEO anyways, rolling in fixed costs for two missions into one.

Space vs stopping Communism - Saving 50m South Koreans from the horrors of Kim Il Sung's gulags isn't just more important, more useful, and more morally creditworthy than the space program, it's more of all those things than practically anything the US government does or has done since. If you think Vietnam was different, you're wearing the winner-writers-history goggles.

First of all, Korea was irrelevant to the space program. It was over years before the first ever satellite was orbited. As for Vietnam, yeah, I'm sure the Vietnamese would've been way better off under a brutal dictatorship aligned with the US than they were under a brutal dictatorship aligned with the USSR

Still, cancelling Apollo did almost nothing to help with the war anyways ; The reality is that the program just did not cost that much per year over the 11 years it was running, especially since all the R&D was already paid for when it was cancelled. So of course Congress later decided to fund the Shuttle which meant developing entirely new hardware instead of producing tried and true heavy launchers with the capability to loft 5 times more the useful payload...

[Simon_Jester]

Oh my God, are you kididng me?

You have just compared the price of a solid-fuel rocket motor (yes, that's right: 500$ is the price of the MOTOR, not the entire Hydra rocket) with the MISSION COST of the complete (and fuelled!) space rocket. The hell was that?

That's actually interesting, though- if he's pricing the rocket engine on the dumb rocket, versus the cost of the Falcon launch...

Hey, they actually come out within a factor of 1.7 in terms of cost per ton!

This would strongly support the idea that when you build a rocket, you're paying a flat price per ton because the weight of the engines scales more or less linearly with the size of the rocket. It's entirely possible that the cost (per ton of rocket) of the rocket's engines is simply a flat multiple of the tonnage.

And that all the other costs of the Falcon rocket, which total around 70% extra over and above that estimated engine cost, come from all the other stuff that's being paid for like fuel, transportation to launch site, and the people who run the actual launch.

Since both types of engine are being mass produced, or are supposed to, that may actually give us a useful insight. It's surprising that you can scale up the rocket by a factor of 50000, and the cost per ton of rocket remains roughly constant.

It would be very impressive if economies of scale allowed us to beat this "constant multiple" effect.

The point is that the refuelling tug would be reuseable, which means you ONLY have to launch the fuel, rather than an entire upper stage every time (such as the massively expensive Centaur) ; And fuel is really cheap, so it can be launched on cheap, light rockets that do not require massive ground facilities to prepare and service - if you lose a payload of fuel, well who cares?

Or on really massive but not man-rated rockets. That could save money too, I suspect. Building a rocket to NOT be man-rated saves a lot of money, because it means you worry less about making sure the engines are 99.9% reliable and so on.

[Beowulf]

The point is that the refuelling tug would be reuseable, which means you ONLY have to launch the fuel, rather than an entire upper stage every time (such as the massively expensive Centaur) ; And fuel is really cheap, so it can be launched on cheap, light rockets that do not require massive ground facilities to prepare and service - if you lose a payload of fuel, well who cares?

Or on really massive but not man-rated rockets. That could save money too, I suspect. Building a rocket to NOT be man-rated saves a lot of money, because it means you worry less about making sure the engines are 99.9% reliable and so on.

It gets better than that. You don't even have to have the same reliability as is required by satellite launchers. You're just tossing fuel up. The only people who care about whether it makes it up is you, as the depot service company. And then you're pretty much entirely concerned about cost to orbit, not whether any particular payload makes it up, because they're all essentially the same.