Alternative to reusable

[todofwar]

Is all this money spent on reusability really worth it? Honestly, I understand the sentiment that if we had a fully reusable system or SSTO we save money, but not too long ago I had a long argument with a friend over why they don't build a new set of space shuttles. In the course of that I looked into the shuttle system in quite a bit of detail. After realizing just how expensive it is to get any launch system launch ready again, you realize that any reusable system is not going to save you nearly as much money as you think it will. 

I say why not go the other way, stop trying to reinvent the wheel and settle on three launch designs (light medium and heavy) and let them pay for their design costs over the next fifty years. Make as many parts as possible interchangeable between them, and pump them out of an assembly line. Bring mass production to the space launch world. Invest all that RnD money into getting the manufacturing process as cheap as possible. Stop trying to build perfect ferraris and start going for the reliable subaru. Granted, you would never want to use such a system for manned launches, and you will need to incorporate the cost of failed launches into the price you charge for succesful ones. Still, I think overall it might be possible to beat reusable systems at cost this way.

[SargeRho]

Yes, it probably is. The Space Shuttle is a bad example of a reusable spacecraft. Because it really isn't. It's a refurbishable spacecraft. It also had a lot of stuff it didn't need.

Big Dumb Booster is one way of doing it, but given the performance requirements, I'm not entirely sure that it's actually viable.

[max_creative]

Falcon 9 is reusable. New shepherd is reusable. SSTOs will be reusable. Spaceship one and two are reusable. Has anyone considered fly back boosters? They work in ksp and are actually really efficient in career mode. 

[Red Fang]

In the current market, Big Dumb Boosters would never work, however new expendable launchers tend to keep being more and more cost effective with each generation, which is more or less in line with your 2nd paragraph. Wether reusable launchers will ever get to a point of  displacing expendables or opening new markets by significantly lowering launch prices, remains to be seen in the upcoming decades. 

[Shania_L]

The Soyuz fanclub is that way

As ever, the Russians beat you there, despite modernisations Soyuz is the same launcher/capsule as they were launching in the 60s, this is the essence of what you are suggesting.

[todofwar]

24 minutes ago, Shania_L said:

The Soyuz fanclub is that way

As ever, the Russians beat you there, despite modernisations Soyuz is the same launcher/capsule as they were launching in the 60s, this is the essence of what you are suggesting.

This is getting at half my point, and definitely proves that we already have perfectly good rockets. But space is still expensive, because even the Soyuz is pretty pricey to build even though its design costs were paid off long ago. That's why I think you can develop a sound design, something that gets you to orbit with a high success rate, and then spend all that money currently going towards new launch systems on manufacturing tech. Automate as much as possible, streamline the testing process, get it to a point where a crew of five guys can build your booster in a month. Is that hard? Very. Is that impossible? Maybe, but reusable is also hard yet people are throwing money at it.

[Three1415]

The problem with your mass-production idea is that there simply isn't enough demand to justify the infrastructure necessary to benefit from economies of scale. I doubt there are more than 100 or so launches a year, and those represent a very wide variety of payloads with very different needs rocket-wise. As such, companies like SpaceX are focusing on reducing the costs of launches on a per-unit basis instead, which at this point is much more effective.

Add on to that the fact that even a Big Dumb Booster is mind-numbingly complex--real-life rocketry is far harder than what you see in KSP (even with RSS and RO), requires extensive testing and quality control, etc., to the point where the "Five guys and a rocket" will probably never be possible--and this means that simple industrial-style production might never be viable. 

[Kerbart]

6 minutes ago, Three1415 said:

(...) real-life rocketry is far harder than what you see in KSP (even with RSS and RO), requires extensive testing and quality control, etc., to the point where the "Five guys and a rocket" will probably never be possible--and this means that simple industrial-style production might never be viable. 

Exactly! The Tyranny of the Rocket Equation simply means that the margins on any launch are razor thin. If you're going to build in safety factors to get around (relatively—by aerospace standards) sloppy manufacturing then you will simply get not into space.

[todofwar]

If you make a set of designs you always use, and have most components interchangeable, you can have some benefits of scale. The infrastructure would be very expensive to build yes, but you allow a fifty year repayment plan (probably the real reason it won't happen: no one has that kind of patience in the private sector) and then you have made space incredibly cheap. I'm not denying this plan is full of problems that need to be worked out, and the "five guys and a rocket" goal is probably 100 years away when we have straight up robots out of that Will Smith movie (I need to read that book...) but in the shorter term, I still think it might be feasable to cut down your labor costs thorugh more automation, your testing through more streamlining, and probably several other measures. 

Let's do a comparison: reusable needs the ability to refit rocket engines for a second launch, which the shuttle tried and failed to make cost effective, and ultimately you only get so many launches out of any one booster no matter what your design. You either invest the money in getting that refurb cost as low as possible and your number of launches per unit as high as possible, or you take that money and throw it at manufacturing infrastructure and tell people they get three or four options for a launch vehicle and they can deal with it. It could very well prove more cost effective depending on what your payback timeframe is on that initial investment. 

[max_creative]

Some ideas:

• first stage is a SRB to save cost. 
• every stage has a computer and is designed to withstand reentry so it can be recovered. 

[todofwar]

2 minutes ago, Kerbart said:

Exactly! The Tyranny of the Rocket Equation simply means that the margins on any launch are razor thin. If you're going to build in safety factors to get around (relatively—by aerospace standards) sloppy manufacturing then you will simply get not into space.

If Kerbal, even RO, actually simulated rocket designing it would be a rather boring game, NASA has teams of engineers working on this full time for a reason. But I'm not saying sloppy manufacturing, I'm saying smart manufacturing. Cars have to work for ten years with hundreds of thousands of miles on them and we mass produce those no problem. Rockets have to work for ten minutes. Granted, they have to work flawlessly for those ten minutes, but there has to be a way to lower the cost of manufacturing without sacrificing reliability too much. 

I'm not arguing this is doable right away, I'm arguing it's no less achievable a goal than going for SSTO spaceplanes or reusable VTVL rockets. I would imagine at first you maybe bring costs down by 5%, not really much. But you use that extra cash to continue to refine and improve, slowly you work down your costs per launch. It's not as sexy to design a better robot arm in a factory than a better VTVL booster, but it might actually be the better investment in the end.

[Bill Phil]

4 hours ago, max_creative said:

Falcon 9 is reusable. New shepherd is reusable. SSTOs will be reusable. Spaceship one and two are reusable. Has anyone considered fly back boosters? They work in ksp and are actually really efficient in career mode. 

Falcon 9 isn't reusable yet. F9R hasn't been reused and won't be for a few years.

[RainDreamer]

Exchangeable parts would require industry standardization of all components in making rockets - like how computers all use the same kind of USB port, which would be difficult since not all rockets are made equal due to mission requirement, along with a lack of demand from the low number of launches.

[fredinno]

7 hours ago, SargeRho said:

Yes, it probably is. The Space Shuttle is a bad example of a reusable spacecraft. Because it really isn't. It's a refurbishable spacecraft. It also had a lot of stuff it didn't need.

Big Dumb Booster is one way of doing it, but given the performance requirements, I'm not entirely sure that it's actually viable.

Big dumb  booster is a bad idea, the Proton shows that no one likes a rocket with a low reliability rating, because it increases insurance costs, even if the rocket is dirt cheap.

6 hours ago, max_creative said:

Falcon 9 is reusable. New shepherd is reusable. SSTOs will be reusable. Spaceship one and two are reusable. Has anyone considered fly back boosters? They work in ksp and are actually really efficient in career mode. 

Yes, people have. It just never comes to frutition, most people in the industry look at reuse and decide it isn't worth it.

https://web.archive.org/web/20090523152328/https://www.space.com/businesstechnology/technology/nasa_mars_booster.html

5 hours ago, max_creative said:

Some ideas:

• first stage is a SRB to save cost. 
• every stage has a computer and is designed to withstand reentry so it can be recovered. 

1st stage SRBs aren't always a good idea, especially when it comes to humans, and/or the larger SRBs.

Ares 1 is a testament to that, and is a good example of a SRB rocket gone wrong.

Ares I had a huge 1st stage SRB due to the high payload needed, along with the lack of extra SRBs to negate the core SRB vibrations, this resulted in a very bumpy rocket not very safe for a human to ride on.

And every stage having a computer and designed to withstand reentry increases cost enormously. The reuse will have to save a lot of money if you do that (and the sea landings reduce the viability of reuse)

2 hours ago, RainDreamer said:

Exchangeable parts would require industry standardization of all components in making rockets - like how computers all use the same kind of USB port, which would be difficult since not all rockets are made equal due to mission requirement, along with a lack of demand from the low number of launches.

The problem I think, is that rockets are distributed across several nations, and thus this increases cost due to the fact each rocket has less launches.

But this is unavoidable, people want their own rocket launch system for national security reasons...

[RainDreamer]

1 hour ago, fredinno said:

 

The problem I think, is that rockets are distributed across several nations, and thus this increases cost due to the fact each rocket has less launches.

But this is unavoidable, people want their own rocket launch system for national security reasons...

Even without the national security reasons, we still have trade secrets policy from private sector corporations that prevent sharing of technology due to competition. A reason that SpaceX do not file any patent for their technology and instead keeping as trade secret is because Elon thinks the Chinese competitors can simply copy them.

Perhaps until we have a large enough space craft industry that it make sense to have have numerous manufacturers specialized in different parts like in KSP, instead of having each manufacturer making a whole package from engine to fuel tank to everything else, we won't see exchangeable parts soon. It just doesn't make enough profit for them. Its like an Apple product and its accessories. Gotta go for a full package, no compatability with anything else.

Edited April 8, 2016 by RainDreamer

[magnemoe]

7 hours ago, todofwar said:

If you make a set of designs you always use, and have most components interchangeable, you can have some benefits of scale. The infrastructure would be very expensive to build yes, but you allow a fifty year repayment plan (probably the real reason it won't happen: no one has that kind of patience in the private sector) and then you have made space incredibly cheap. I'm not denying this plan is full of problems that need to be worked out, and the "five guys and a rocket" goal is probably 100 years away when we have straight up robots out of that Will Smith movie (I need to read that book...) but in the shorter term, I still think it might be feasable to cut down your labor costs thorugh more automation, your testing through more streamlining, and probably several other measures. 

Let's do a comparison: reusable needs the ability to refit rocket engines for a second launch, which the shuttle tried and failed to make cost effective, and ultimately you only get so many launches out of any one booster no matter what your design. You either invest the money in getting that refurb cost as low as possible and your number of launches per unit as high as possible, or you take that money and throw it at manufacturing infrastructure and tell people they get three or four options for a launch vehicle and they can deal with it. It could very well prove more cost effective depending on what your payback timeframe is on that initial investment. 

This is a lot how some Russian rockets is used. 
Note that it has its downsides, 50 years is so long you will get technological surprises, you also end up with using standard parts who is non standard after 50 years, 
Part of the reason that building an saturn 5 replica is stupid. 
Most of the modern rockets require far fewer people to launch than the 1960 versions. 

The shuttle did most thing wrong, you want to save weight on the upper stage, adding an 1 kg on upper stage is equal to 5-10 on first.The shuttle had to bring 30 ton extra to orbit,
It was so bleeding edge an pushed the envelope a lot so it required a lot of maintenance between launches, thing formula 1 car rater than pickup.
this increased cost who reduced number of launches who increased cost of each launch. 

[fredinno]

16 minutes ago, magnemoe said:

This is a lot how some Russian rockets is used. 
Note that it has its downsides, 50 years is so long you will get technological surprises, you also end up with using standard parts who is non standard after 50 years, 
Part of the reason that building an saturn 5 replica is stupid. 
Most of the modern rockets require far fewer people to launch than the 1960 versions. 

The shuttle did most thing wrong, you want to save weight on the upper stage, adding an 1 kg on upper stage is equal to 5-10 on first.The shuttle had to bring 30 ton extra to orbit,
It was so bleeding edge an pushed the envelope a lot so it required a lot of maintenance between launches, thing formula 1 car rater than pickup.
this increased cost who reduced number of launches who increased cost of each launch. 

Only thing is that the Soyuz was upgraded and modified significantly over time.

[KSK]

10 hours ago, todofwar said:

Is all this money spent on reusability really worth it? Honestly, I understand the sentiment that if we had a fully reusable system or SSTO we save money, but not too long ago I had a long argument with a friend over why they don't build a new set of space shuttles. In the course of that I looked into the shuttle system in quite a bit of detail. After realizing just how expensive it is to get any launch system launch ready again, you realize that any reusable system is not going to save you nearly as much money as you think it will. 

I say why not go the other way, stop trying to reinvent the wheel and settle on three launch designs (light medium and heavy) and let them pay for their design costs over the next fifty years. Make as many parts as possible interchangeable between them, and pump them out of an assembly line. Bring mass production to the space launch world. Invest all that RnD money into getting the manufacturing process as cheap as possible. Stop trying to build perfect ferraris and start going for the reliable subaru. Granted, you would never want to use such a system for manned launches, and you will need to incorporate the cost of failed launches into the price you charge for succesful ones. Still, I think overall it might be possible to beat reusable systems at cost this way.

A lot of that sounds like steps that SpaceX have already taken, which is why the F9 is comparatively cheap even without taking re-use into account.

Small number of designs geared for mass production - Check. They don't have a light launcher but they do have a medium one (Falcon 9) and a heavy one (Falcon Heavy) under development, supposedly due for a (much delayed) maiden flight this autumn.

Going for Subarus rather than Ferraris - check. The Merlin engine is deliberately not a cutting edge engine in terms of raw performance (kerolox instead of hydrolox, relatively simple combustion cycle) but is geared towards ease of manufacture and acceptable performance for it's price.

Interchangeable parts - check. The Falcon 9 upper stage tanks are essentially a short version of the lower stage tanks. Same diameter, same materials, storing the same propellants, churned out on the same tools. The lower stage tanks do have the unavoidable extra complexity of needing to feed nine engines rather than one. Speaking of engines - the Merlin-vac used on the Falcon 9 upper stage is (I believe) essentially the same as the Merlin engines used on the lower stage, with (relatively) minor nozzle modifications to optimize it for vacuum use.

I imagine there are other examples but those are the ones I'm aware of.

Commonality - check. The Falcon Heavy lower stage is basically three Falcon 9 cores strapped together, taking a leaf from the Delta IV design book. The F9 core was (allegedly) designed from the outset with that in mind - time will tell how successful it is.

 

[InsaneDruid]

3 hours ago, fredinno said:

Big dumb  booster is a bad idea, the Proton shows that no one likes a rocket with a low reliability rating, because it increases insurance costs, even if the rocket is dirt cheap.

Proton-M has a reliability of 91% in 86 launches (BTW, it had 100% for 13 launches in a row after being introduced).

SOYUZ 2.1A/B has a reliability if 92% in 38 launches.
 

EDIT: Source: The Annual Compendium of Commercial Space Transportation: 2014
 

Edited April 8, 2016 by InsaneDruid

[magnemoe]

2 hours ago, fredinno said:

Only thing is that the Soyuz was upgraded and modified significantly over time.

True, however this don't make it an standard rocket anymore, then it become an standard system, yes its cheaper and safer to modify than start over. 

[Scotius]

Fifty years ago we got to LEO using multi-staged, expendable launchers. Five decades later (aside from Apollo episode) we are still in LEO, using multi-staged, expendable launchers. Simply put - we stalled.To move forward again, we need new ways of doing things - thats what SpaceX, Virgin, New Shepard, maybe Skylon in the future are doing. It's in human nature - if not for that curiosity and drive for innovation we would still use sail ships and horse-drawn carriages. They were 'good enough' back then after all

[todofwar]

27 minutes ago, Scotius said:

Fifty years ago we got to LEO using multi-staged, expendable launchers. Five decades later (aside from Apollo episode) we are still in LEO, using multi-staged, expendable launchers. Simply put - we stalled.To move forward again, we need new ways of doing things - thats what SpaceX, Virgin, New Shepard, maybe Skylon in the future are doing. It's in human nature - if not for that curiosity and drive for innovation we would still use sail ships and horse-drawn carriages. They were 'good enough' back then after all

Fair point, but we have tried and failed for 50 years to improve, when we could have been turning those efforts to manufacturing technology instead. At the end of the day, getting to space is hard because of the ridiculous physics involved just in terms of energy. Engines that can start more than once are difficult, engines that can throttle are difficult, mostly because they go through such extreme stresses during launch. I still think any engine that has burned its way to space is going to need all kinds of refitting to be used again, which is what killed the shuttle in the end. 

I don't disagree with what people have said here about the issues that will be faced, but as has been pointed out SpaceX has managed to get costs down but doing some of the things I mentioned, now if they cut out their reuse program and poured all that RnD into an assembly line that can pump out a Merlin engine on the cheap, they could probably see similar savings to what they would get with successfully reusing boosters. 

[Scotius]

So, essentially it's: "Making more of the same" vs. "Trying something new". Both methods of approach have their good and bad sides. Personally, i prefer innovation. I was born too late to see man landing on the Moon. I don't want to die as a person born too early to see a man landing on Mars.

[todofwar]

11 minutes ago, Scotius said:

So, essentially it's: "Making more of the same" vs. "Trying something new". Both methods of approach have their good and bad sides. Personally, i prefer innovation. I was born too late to see man landing on the Moon. I don't want to die as a person born too early to see a man landing on Mars.

I would say its "making more of the same better and cheaper" vs "risking it all on something untested and expensive". Nothing has topped the Saturn V in terms of payload, so we really haven't advanced at all by that metric (yeah yeah Falcon Heavy, not done yet so it doesn't count). In terms of launching we have all the tech we need to reach Mars (or Venus) or anywhere else we like. There are plenty of other challenges associated with such missions, and we should absolutely be investing in how to solve those. But in terms of launch vehicles, I think it's time to shift focus a little. 

[InsaneDruid]

1 hour ago, Scotius said:

Fifty years ago we got to LEO using multi-staged, expendable launchers. Five decades later (aside from Apollo episode) we are still in LEO, using multi-staged, expendable launchers. Simply put - we stalled.To move forward again, we need new ways of doing things - thats what SpaceX, Virgin, New Shepard, maybe Skylon in the future are doing. It's in human nature - if not for that curiosity and drive for innovation we would still use sail ships and horse-drawn carriages. They were 'good enough' back then after all

50 years ago we got to LEO, could stay there for some days living in the place of a car's front seat pooping in a plastic bag we brought along with us. Now we live and work in LEO for half a year routinely and over a year without a problem on a international space station. We are still using multi-staged, expendable launchers, for they are reliable, proven technology in a rather risky field of exploration. We aren't stalled at all.

SpaceX, Virgin, New Shepard etc don't move us forward (That much as your text may suggest). They (try to) make it cheaper. Virgin and New Shepard aren't even there where we were 50 years ago and they wont be for a LOOONG time. Maybe ever. If we go to mars, we go there on a multi-staged, expendable launcher.