Quote from: AncientU on 11/14/2017 07:59 pmWhat Ed is advocating is that everyone should go back to the 'sensible' world where all rockets were thrown away after each use... sure, that makes perfect sense to me! Certainly not! That is not my point of view, even if you wish it were. I am offering a surrogate for figuring out where the crossover point might be for reuse payback. I agree that such a crossover point exists, and I'm not suggesting that SpaceX or others might not be closing or even crossing that gap soon. I merely suggested that they haven't done it quite yet. - Ed Kyle
What Ed is advocating is that everyone should go back to the 'sensible' world where all rockets were thrown away after each use... sure, that makes perfect sense to me!
Quote from: edkyle99 on 11/14/2017 08:54 pmQuote from: AncientU on 11/14/2017 07:59 pmWhat Ed is advocating is that everyone should go back to the 'sensible' world where all rockets were thrown away after each use... sure, that makes perfect sense to me! Certainly not! That is not my point of view, even if you wish it were. I am offering a surrogate for figuring out where the crossover point might be for reuse payback. I agree that such a crossover point exists, and I'm not suggesting that SpaceX or others might not be closing or even crossing that gap soon. I merely suggested that they haven't done it quite yet. - Ed KyleThe difficulty is that using only the F9 statistics to determine an economics-only 'cross-over point' assumes other value factors equal zero (or one). I've pointed out that there is difficult-to-quantify, but still important value to reusability as demonstrated with F9 for both FH and BFR, also for flight rate, etc. Since the corporate goal is not 'closing the business case' -- but getting a viable transportation system to Mars going -- economic analysis is too simplistic. Life was simpler when a launch system was designed and built 'only' for launch, but that is not the case for F9. It, more than anything, is a proof of concept for a quite different developmental goal -- it needs to be evaluated against the goal for which it was built.BL: It doesn't matter if/when economic 'cross-over' occurs. As a development program, it is essentially mission accomplished, 'business case' closed.To tie this back to Blue Origin, it doesn't matter where/when the 'cross-over' point for New Glenn is reached. The financing is in place; the goal is 'millions of people living and working in space.'
Lowering the cost to orbit, 24-hour reusability, colony on Mars, millions of people living and working in space are visionary goals for a future that could be radically different than the one that is/was settling for the status quo in space -- rare, expensive, and 'hard' (so send us lots of money). Fully and rapidly reusable rockets are a key to that goal -- which could also fail to materialize, even with gas-n-go rockets -- but failure to try to build these rockets guarantees that the future envisioned is impossible. Expendable rockets, even those which have business 'cross-over' points already achieved (if any actually do*) could never begin to move us into that future.Expendable rockets were built for a domestic business, basically covering for Shuttle's failure to provide low cost to orbit and the future that could be possible. Those expendable rockets wouldn't have existed if Shuttle succeeded, and they will be road kill if/when the (fully and rapidly) reusable rockets are fully realized.* Are either Atlas V or Delta IV at their respective cross-over points? Boeing/LM spent several billions developing these vehicles... How about Ariane 5? Will Arine 6 ever 'cross-over'? Angara? How about Vulcan/ACES? SLS? Delta II, Soyuz, Proton probably have crossed over, but each is planned for retirement.
Quote from: AbuSimbel on 11/14/2017 09:53 pmThe thing is that, imo, you only have to really consider two parameters: refurbishment costs and the higher manufacturing cost of the 'overbuilt' reusable system.You can't just ignore economies of scale.An extreme example: You have a rocket that can be reused 10x, but there's only demand for 1 launch per year.Consequently you will manufacture a single rocket every 10 years. That rocket is going to be bloody expensive. Imagine what a car or a computer chip would cost if only one were produced every 10 years. If the production cost of a rocket drops by x% with every doubling of the production rate, you can realize big savings from getting the production rate up from 1 to 2 to 4 to 8 etc.Bottom line: With 1 launch per year reusability will not pay off, of that I'm 100% certain. With 50 it might, because you still get to make 5-10 rockets a year, which gives you reasonable economies of scale (assuming 10-5 uses).
The thing is that, imo, you only have to really consider two parameters: refurbishment costs and the higher manufacturing cost of the 'overbuilt' reusable system.
Quote from: Oli on 11/15/2017 12:34 amQuote from: AbuSimbel on 11/14/2017 09:53 pmThe thing is that, imo, you only have to really consider two parameters: refurbishment costs and the higher manufacturing cost of the 'overbuilt' reusable system.You can't just ignore economies of scale.An extreme example: You have a rocket that can be reused 10x, but there's only demand for 1 launch per year.Consequently you will manufacture a single rocket every 10 years. That rocket is going to be bloody expensive. Imagine what a car or a computer chip would cost if only one were produced every 10 years. If the production cost of a rocket drops by x% with every doubling of the production rate, you can realize big savings from getting the production rate up from 1 to 2 to 4 to 8 etc.Bottom line: With 1 launch per year reusability will not pay off, of that I'm 100% certain. With 50 it might, because you still get to make 5-10 rockets a year, which gives you reasonable economies of scale (assuming 10-5 uses).I have to apologize. The above describes an experience curve which is invariant to scale, so if both expendable and reusable rockets have the same curve, doubling the flight rate will reduce the cost of both by an equal percentage.sThe flight rate is thus irrelevant to the cost ratio, i.e. to the question of whether reusable vehicles are less/more costly. One has to introduce fixed costs to make the cost ratio dependent on the flight rate.And you people should read my posts better .
And you people should read my posts better .
Quote from: AbuSimbel on 11/14/2017 09:53 pmThe thing is that, imo, you only have to really consider two parameters: refurbishment costs and the higher manufacturing cost of the 'overbuilt' reusable system.You can't just ignore economies of scale.An extreme example: You have a rocket that can be reused 10x, but there's only demand for 1 launch per year.Consequently you will manufacture a single rocket every 10 years. That rocket is going to be bloody expensive.
Imagine what a car or a computer chip would cost if only one were produced every 10 years.
Fixed costs are mostly per year of operating the factory and keeping employees on payroll, not per rocket built.If you are paying for a factory that can build ten rockets per year, 10x reuse for a constant ten flights a year won't really change much. What that level of reusability does give you is the option to keep the same factory and potentially fly 100 times a year, which is a huge improvement.Instead of linear growth proportional to the rocket production rate, you get quadratic growth proportional to the rocket production rate times the reuse rate. If you want to put a million people into space by launching 10 000 times per year, making 50 rockets per year and reusing them 200 times, is a lot more realistic than building factories to make 10 000 rockets per year. At small launch rates quadratic growth gives no advantage over linear growth, but at launch rates that are required to fully support an off-planet civilization, it's necessary.What is ultimately needed to colonize space is a high flight rate. Reuse is extremely helpful for that. You don't pursue reuse to stick to the status quo, you do it to enable a hundredfold growth of operations. If you can't find customers for your higher flight rate, you become your own customer by making your own constellations. Or, if you have pockets as deep as Bezos, you keep guarenteeing that this capability will continue to exist as you wait for other entities to plan around it.
Quote from: Nilof on 11/20/2017 02:20 amFixed costs are mostly per year of operating the factory and keeping employees on payroll, not per rocket built.If you are paying for a factory that can build ten rockets per year, 10x reuse for a constant ten flights a year won't really change much. What that level of reusability does give you is the option to keep the same factory and potentially fly 100 times a year, which is a huge improvement.Instead of linear growth proportional to the rocket production rate, you get quadratic growth proportional to the rocket production rate times the reuse rate. If you want to put a million people into space by launching 10 000 times per year, making 50 rockets per year and reusing them 200 times, is a lot more realistic than building factories to make 10 000 rockets per year. At small launch rates quadratic growth gives no advantage over linear growth, but at launch rates that are required to fully support an off-planet civilization, it's necessary.What is ultimately needed to colonize space is a high flight rate. Reuse is extremely helpful for that. You don't pursue reuse to stick to the status quo, you do it to enable a hundredfold growth of operations. If you can't find customers for your higher flight rate, you become your own customer by making your own constellations. Or, if you have pockets as deep as Bezos, you keep guarenteeing that this capability will continue to exist as you wait for other entities to plan around it.The factory cost argument is only true in the extreme example where the factory is under-utilized, and in reality, if any of these players are doing that bad, they're not in the game any more.In reality, even a factory for reusable rockets will be well-utilized. It's just the you'll need only one factory to support hundreds of flights, not several.I am amazed that people still think that you can compete with a reusable rockets by making cheap expendable ones.
There is a third business strategy to consider that could ameliorate the issues of high vertical integration/low flight rates. Outsource as much of the production of your RLV as possible. Many of the parts of an RLV could be sourced externally or bought off the shelf, avoiding the expense of maintaining huge facilities or standing armies. This would be helpful in variable or low production rates. Basically, the Orbital Sciences strategy.Although no company has directly tried it for RLVs, new entrants to the launch market are mostly developing small mass produced RLVs. It's worth considering that SpaceX had its demo ITS LOX tank made by Janicki Industries and that DC-X used existing RL10s, avoiding the gargantuan expense of developing new engines.
I read a discussion here about the two companies recovery ships and which approach was better. But this seems an area where cooperation and/or third parties could benefit all. I doubt either wants to maintain a naval fleet of ships, barges and tugs. If we get to the point where stages are landing in ocean on a weekly basis, that's where I see a general recovery service operating. Thoughts on that?