How excited should I be about Skylon?

[PakledHostage]

Reading comprehension, much? The satellites, like I said where you precisely quoted.

You don't need to be such a [diminutive of Richard] about it. While Nibb may disagree with you, he has been polite in his responses to you.

[Kryten]

Reading comprehension, much? The satellites, like I said where you precisely quoted.

There aren't that many satellites, that's what we've been trying to tell you; the figures they've given can't be hit even if Skylon gets 100% market share.

EDIT: Let's get some numbers here. You say 37.5 Skylons are needed to make up the research costs; at 200 flights each, that's 7,500 flights. In the last few years, the flight rate has been about 60-80 per year; if we're generous and assume it's 80, then that many Skylons are enough to fulfill market demand for 94 years.

Edited December 4, 2014 by Kryten

[m4inbrain]

You say 37.5 Skylons are needed to make up the research costs; at 200 flights each, that's 7,500 flights.

That's not what he's saying.

And I just said they don't even need that much business to break even in 10, the 200 flights is ten times the break-even.

That's what he's saying. From the exact posting that you were quoting. I don't care much for Skylon itself, but you and that other guy are either not able or not willing to actually read what your opposite is saying. I don't even, honestly.

[Kryten]

Yes, they could break-even for the operators at however many low number of flights, but that's not how operators work. They aren't going to buy far more aircraft than are justified by the demand just so the manufacturer makes up it's development costs.

Edited December 4, 2014 by Kryten

[m4inbrain]

Yes, they could break-even for the operators at however many low number of flights, but that's not how operators work. They aren't going to buy far more aircraft than are justified by the demand just so the manufacturer makes up it's development costs.

I agree there. But: we're assuming the current situation in terms of launches here. I wouldn't say that it'll triple or quadruple with the reduced launchcosts - but you will have a measurable increase in launches. I won't say "it's enough", but for an economic analysis, there's way too many holes in yours and nibbels.

edit: including for example "knowledge". A lot of new stuff has to be invented/tested, and i'm pretty convinced that there's money there too. So i actually doubt the 37.5 units in the first place.

Edited December 4, 2014 by m4inbrain

[Kryten]

I agree there. But: we're assuming the current situation in terms of launches here. I wouldn't say that it'll triple or quadruple with the reduced launchcosts - but you will have a measurable increase in launches. I won't say "it's enough", but for an economic analysis, there's way too many holes in yours and nibbels.

It's needs to increase 10* to justify the breakeven-for-REL level, that's the issue.

[Frozen_Heart]

I don't understand why the dev cost of Skylon is so much higher than something like the Falcon 9R.

Also Reaction engines moves at a tiny fraction of the speed that SpaceX does. Its just so inefficient. It was about 5 Years between Falcon 9 starting development and first launch. Skylon has been in development for about 15 years and they have managed to make 1 precooler for the cost of the entire Falcon 9 development.

[m4inbrain]

It's needs to increase 10* to justify the breakeven-for-REL level, that's the issue.

Care to share what numbers you used? I'm especially interested in what you'd assume would be the launch costs for a "generic satelite X". Basically, what the customer pays to get his satellite where he wants it to be.

edit: and i certainly hope you didn't use the 6.6 million rubbish that was stated in the thread earlier.

I don't understand why the dev cost of Skylon is so much higher than something like the Falcon 9R.

I assume it has to do with the fact that the Falcon is something entirely different. The Falcon is basically an improved design of something that already exist. Skylon is something that hasn't really been done before, all in all (closest thing in my mind would be the STS, but the technologies there are not compatible). SABRE plays a big part in there for example, where spaceX just used older (albeit improved) designs.

Edited December 4, 2014 by m4inbrain

[Kryten]

REL are proposing to be a manufacturer, not an operator, so the individual launch costs aren't actually relevant for them. The dev cost and the planned cost per skylon were given upthread; REL needed to sell 37.5 to make a profit on their development. Given Skylon is supposed to be able to make 200 flights, that translates to them requiring a market demand of (37.5*200=) 7,500 launches. Even at highest-ever market demand (1967, 173 launches), that's requiring people to buy enough skylons to fulfill 44 years of market demand; it's not plausible they'll have a long enough service life or the operators enough patience for this to actually happen.

[m4inbrain]

REL are proposing to be a manufacturer, not an operator, so the individual launch costs aren't actually relevant for them. The dev cost and the planned cost per skylon were given upthread; REL needed to sell 37.5 to make a profit on their development. Given Skylon is supposed to be able to make 200 flights, that translates to them requiring a market demand of (37.5*200=) 7,500 launches. Even at highest-ever market demand (1967, 173 launches), that's requiring people to buy enough skylons to fulfill 44 years of market demand; it's not plausible they'll have a long enough service life or the operators enough patience for this to actually happen.

It's either me, or you are jumping from operator break-even to REL break-even as it fits your argument.

Let's clear up. You say: REL needs to sell 38 (there's no 37,5) units to make a profit on their development. I say: rubbish. You don't even need to be an economist to see that. And i assume that i talk to an intelligent person (considering where this thread is located), so i also assume you pretty much know that it's rubbish. But to be clear: you have a company, developing a vehicle based on technology that doesn't yet exist. The UK government already showed interest in SABRE. Do you actually think that all REL has to offer is 38 units of their product? They develop new technologies, in case of the SABRE, pretty marketable ones too. That's why i said earlier, i doubt that they need 38 units to break even. They would need less. How much less i can't actually tell, but i at least know that i can't.

If you now go ahead telling me "launch costs irrelevant for REL" after you just told me:

Yes, they could break-even for the operators at however many low number of flights, but that's not how operators work. They aren't going to buy far more aircraft than are justified by the demand just so the manufacturer makes up it's development costs.

Which is directly tied to launch costs. You didn't give me an estimate, so i assume you just used the 6.6mil figure. And yes, obviously it's important and directly tied to RELs break-even. It's actually the single biggest questionmark. How fast can a skylon amortise itself for me? Can it be done in 10 launches? 20? How many? Based on that, companies would buy skylon. You didn't give me any number to look at there, so i'm pretty unsure what to think about that.

edit: and yes. They would sell more, if they amortise quick enough. That's how an open market works. Without knowing how quick they make their money back, after how many launches, years - or in your case, even dismissing it alltogether, you have zero basis for your argument.

edit2: to be even clearer: i actually don't think that skylon will be feasable at all. I'm just a bit annoyed by how much rubbish gets stated as facts. Like the 38 units, the 6,6mil launch costs, etc - these are numbers that aren't even properly predictable by full fledged economists, let alone in a science forum. These also are the numbers you'd need to know to dismiss skylon based on an economic reason (next to others, but for starters).

Edited December 4, 2014 by m4inbrain

[Kryten]

It's either me, or you are jumping from operator break-even to REL break-even as it fits your argument.

It's you, I'm deliberately contrasting the two. Operator financial case; fine and dandy, assuming REL's launch costs are accurate. REL? Screwed in any realistic market.

Let's clear up. You say: REL needs to sell 38 (there's no 37,5) units to make a profit on their development. I say: rubbish. You don't even need to be an economist to see that. And i assume that i talk to an intelligent person (considering where this thread is located), so i also assume you pretty much know that it's rubbish. But to be clear: you have a company, developing a vehicle based on technology that doesn't yet exist. The UK government already showed interest in SABRE. Do you actually think that all REL has to offer is 38 units of their product? They develop new technologies, in case of the SABRE, pretty marketable ones too. That's why i said earlier, i doubt that they need 38 units to break even. They would need less. How much less i can't actually tell, but i at least know that i can't.

They've shown no interest in marketing these technologies outside of Skylon; the only other thing they've ever proposed is a hypersonic airliner using roughly the same technology.

If you now go ahead telling me "launch costs irrelevant for REL" after you just told me:

Something about operators, not REL, yes.

Which is directly tied to launch costs. You didn't give me an estimate, so i assume you just used the 6.6mil figure. And yes, obviously it's important and directly tied to RELs break-even. It's actually the single biggest questionmark. How fast can a skylon amortise itself for me? Can it be done in 10 launches? 20? How many? Based on that, companies would buy skylon. You didn't give me any number to look at there, so i'm pretty unsure what to think about that.

I decided to be generous, and assume Skylon would be bought by all launch companies and be used to replace all existing launchers, hence no need for such calculation. That it's problematic even then should tell you something about how optimistic these projections are.

edit: and yes. They would sell more, if they amortise quick enough. That's how an open market works. Without knowing how quick they make their money back, after how many launches, years - or in your case, even dismissing it alltogether, you have zero basis for your argument.

REL don't make any money back through launches, they intend to fulfill the role of a manufacturer only. I don't deny that if this works it could well be a good deal for an operator, it's REL I have issues with.

[Iskierka]

REL have no intention to be manufacturer for anything but heat exchangers. Rest of engine and airframe are being left to bigger companies, and bigger companies are taking interest, so apparently they know more than you.

[Kryten]

So their income per skylon will be considerably lower than I've assumed. That makes their position far worse.

[NERVAfan]

Dividing the price of orbital launches by a factor of 10 doesn't mechanically multiply the demand by 10... or 100 or 1000. Demand will be driven by need, not price, and there simply is very little need for space.

But if the price drops enough, new space industries - new "needs" - will probably appear. I agree Skylon probably won't be funded in the current environment... but "the current environment" may not last much longer. If SpaceX gets first stage reusability working, and if it saves money, then other companies/agencies will be looking for reusable vehicles.

[Iskierka]

So their income per skylon will be considerably lower than I've assumed. That makes their position far worse.

Where is their own income a concern? The concerns are making back overall development cost, regardless of who footed the bill. And with bigger companies to foot the initial costs and get things going, as well as set up markets, that only improves chances of making back development cost. While they do want to make money from it, REL's goal is revolutionising space access, not to become sole billionaires, so it's perfectly acceptable to sacrifice a majority of the manufacturing to get it done.

[magnemoe]

I don't understand why the dev cost of Skylon is so much higher than something like the Falcon 9R.

Also Reaction engines moves at a tiny fraction of the speed that SpaceX does. Its just so inefficient. It was about 5 Years between Falcon 9 starting development and first launch. Skylon has been in development for about 15 years and they have managed to make 1 precooler for the cost of the entire Falcon 9 development.

Fast planes are far more complex than rockets, this is also true in KSP.

Engine is an totally new type, the airframe is also special.

For falcon 9R they had the luxury to keep improving designs.

the reusable is not a lot of new technology either. They need the first stage to be able to use rcs to reorient and restart, copy from upper stage.

Moon and Mars landers does powered landing. Their main problem has been to keep it stable in atmosphere. solution is to use grind fins on the top this is used on bombs.

And they can do the tests as part of normal paid missions.

[Kryten]

Where is their own income a concern? The concerns are making back overall development cost, regardless of who footed the bill.

How are they supposed to do that without sufficient source of income? You've said yourself, this is not going to be successful unless it causes a 'revolution' in the market. Why would anybody sane invest in that?

[shynung]

But if the price drops enough, new space industries - new "needs" - will probably appear.

Not necessarily. The demand for space access is inelastic. Those who need it really need it, those who don't rarely cares.

Before continuing this line of discussion, I'd advise reading this old thread first. Otherwise, feel free to add anything else not covered there.

[NERVAfan]

Not necessarily. The demand for space access is inelastic. Those who need it really need it, those who don't rarely cares.

(my bold)

How do we know that? IIRC there never has really been a dramatic drop in launch prices.

I agree that current industries (eg comsats) aren't going to buy 10x more launches if launches are 10x cheaper. What I'm arguing is that totally new things will quite likely appear.

EDIT: Look how many cubesats have been launched in the ~10 years since cubesats were invented. There is quite a bit of interest in space; the prices just make it out of reach for most purposes.

[Iskierka]

Current industries would change in response, but how is unpredictable. Cheaper and more reliable launches, that don't have multi-year-long queues mean many companies may take more risks and develop cheaper satellites that are less-than-perfect, and be willing to replace them, or start using swarms with higher data bandwidth or observational capability and redundancy rather than very precise high reliability. Quite what change will occur is unknown, but even current industries will become elastic, even if only by a small factor to exploit the updated costs.

Interferometry is becoming interesting for getting very good pictures, with the ability to launch multiple sats I could see ground observation being replaced by a few smaller units orbiting in formation. Might finally be able to improve on the current resolution limits. (IIRC around 0.5m/pixel)

[shynung]

How do we know that? IIRC there never has really been a dramatic drop in launch prices.

I agree that current industries (eg comsats) aren't going to buy 10x more launches if launches are 10x cheaper. What I'm arguing is that totally new things will quite likely appear.

EDIT: Look how many cubesats have been launched in the ~10 years since cubesats were invented. There is quite a bit of interest in space; the prices just make it out of reach for most purposes.

Well, I think SpaceX may have been responsible for the beginning of cheap space launchers era. True, the prices they offered wasn't very dramatic, but it's enough to turn heads.

Though, at the moment, I'm presuming that any complicated space operations (say, asteroid mining) will require a lot of mass in orbit. Whether upcoming affordable heavy launchers (like Falcon Heavy) will make such things profitable, I haven't enough data to have neither an idea nor opinion.

Regarding cubesats, I think it has more to do with what a cubesat can do compared to a more conventional satellite. If a single cubesat can be used as a GEO comsat, then a reduction in launch prices, in the form of smaller rockets, would have a significant effect. But alas, they are only little boxes. As for now, there are only so much equipment that we can stuff into them.

Edited December 6, 2014 by shynung

[NERVAfan]

Well, I think SpaceX may have been responsible for the beginning of cheap space launchers era. True, the prices they offered wasn't very dramatic, but it's enough to turn heads.

I think what will really be the game changer is if/when they succeed at reusing the first stage. Not just for what it does to their own prices, but what will others try to do to compete? That's when I can really see something like Skylon development getting funded; in a world with a half-reusable F9 and FH, there's not much point in developing a plain old expendable rocket.

Regarding cubesats, I think it has more to do with what a cubesat can do compared to a more conventional satellite. If a single cubesat can be used as a GEO comsat, then a reduction in launch prices, in the form of smaller rockets, would have a significant effect. But alas, they are only little boxes. As for now, there are only so much equipment that we can stuff into them.

That's kind of my point. Cubesats are incredibly limited, yet there have been a huge number of them in a comparatively short time. If larger satellites could be launched for cubesat-like prices...

EDIT: I don't expect any of this to happen quickly. Creating new markets will take time, as will developing 2nd generation (post-Falcon) reusable LVs. But 15 or 20 years down the line, unless SpaceX fails completely at reusability, the picture will be very dramatically different IMO.

[Nibb31]

I think what will really be the game changer is if/when they succeed at reusing the first stage. Not just for what it does to their own prices, but what will others try to do to compete? That's when I can really see something like Skylon development getting funded; in a world with a half-reusable F9 and FH, there's not much point in developing a plain old expendable rocket.

Why not? The economical viability of reusable rockets hasn't been demonstrated yet. Again, they only make sense if you have a high flight rate and fast turnaround.

SpaceX has been cutting costs through mass production. If the launch market doesn't expand dramatically, they have to choose between maximizing production capacity (lowering costs by building more expendable units) or going reusable (reusing stages but each unit is more expensive).

And I don't think that a 10 or 20% lower price tag is going to make space more affordable to thousands of new applications that will expand the market dramatically.

That's kind of my point. Cubesats are incredibly limited, yet there have been a huge number of them in a comparatively short time. If larger satellites could be launched for cubesat-like prices...

Cubesats are typically disposable educational payloads launced by schools and universities. Even if they could afford a ticket on a Falcon 9, those institutions don't usually have the funding or facilities to build larger payloads.

The only new market on the horizon is LEO constellations for high-speed global communication. While that will offer a few juicy contracts for launch providers, it's not going to justify going from 50 launches per year to 500.

The fact that there is no predictable market for fast turnaround means that Skylon is a non-starter, whichever way you look at it. "Build it and they will come" doesn't work. It's more like a 6-lane "bridge to nowhere".

Edited December 6, 2014 by Nibb31

[NERVAfan]

Why not? The economical viability of reusable rockets hasn't been demonstrated yet.

Right, but if SpaceX succeeds, it may be quite soon. And given the time lag to develop rockets, starting developing a fully expendable rocket, even now much less after they demonstrate recovery & reuse, is betting that they will fail.

SpaceX has been cutting costs through mass production.

And through vertical integration/bypassing traditional aerospace suppliers with higher costs, IIRC.

If the launch market doesn't expand dramatically, they have to choose between maximizing production capacity (lowering costs by building more expendable units) or going reusable (reusing stages but each unit is more expensive).

Well, they are definitely following the reusable path.

And I don't see why it requires a huge launch market expansion immediately; the expendable -> reusable transition will be a gradual thing. (The current plan is that only the 1st stage will be reusable on the Falcons; full reusability will wait for the next rocket). SpaceX's long term plans do indeed require a gigantic expansion of the launch market, but that's actually pretty plausible at the prices they seem to be expecting for later vehicles.

And I don't think that a 10 or 20% lower price tag is going to make space more affordable to thousands of new applications that will expand the market dramatically.

No, it wouldn't, but I think they are projecting much much bigger price decreases than that. Probably not immediately, but as reusability develops.

Cubesats are typically disposable educational payloads launced by schools and universities. Even if they could afford a ticket on a Falcon 9, those institutions don't usually have the funding or facilities to build larger payloads.

Sure. I'm not disputing that.

What I am talking about is the potential system where you can launch a 10kg or 20kg (or, in the long run, 50kg or 100kg) smallsat for the current price of a 2U cubesat.

The only new market on the horizon is LEO constellations for high-speed global communication. While that will offer a few juicy contracts for launch providers, it's not going to justify going from 50 launches per year to 500.

In the very short term, that is probably true. 15 or 20 (or even 10) years down the road... there will probably be more. (Planetary Resources, Bigelow, etc. - plus stuff that hasn't even started yet.)

[shynung]

And through vertical integration/bypassing traditional aerospace suppliers with higher costs, IIRC.

Actually, it's more through mass production than vertical integration. SpaceX's Falcon 9 rocket has almost identical engines throughout the stages. The lower stage has 9 Merlin engines, the upper stage has 1 Merlin with a nozzle extension. Since a single rocket requires 10 engines, they can mass-produce the Merlins and cut costs by economies of scale. Furthermore, using only one engine design cuts R&D costs, since the engineers have only one engine type to worry about; most of the tooling can be specified for this engine alone.

The only other engine SpaceX makes are the Draco engines used in Dragon spacecrafts, which are simpler hypergolic thrusters mostly used as OMS/RCS.