What should NASA actually do with SLS?

[clongton]

Annual fixed cost for launching rockets into space is the sum of:
the maintenance of the manufacturing and launch facilities
supplying utilities to those facilities
maintaining an efficient workforce at those facilities

Fixed costs for a 1-rocket system (SLS)
● maintain 1 manufacturing facility
● maintain 1 launch facility
● supply utilities to 1 manufacturing facility
● supply utilities to 1 launch facility
● maintain workforce for 1 manufacturing facility
● maintain workforce for 1 launch facility

Fixed costs for a 2-rocket system (Ares-I/V)
● maintain 2 manufacturing facilities
● maintain 2 launch facilities
● supply utilities to 2 manufacturing facilities
● supply utilities to 2 launch facilities
● maintain workforce for 2 manufacturing facilities
● maintain workforce for 2 launch facilities

I think it's pretty clear that the fixed costs for a 2-rocket system are going to be close to 2x the fixed costs for a 1-rocket system.

Having covered this let me simply say that the SLS is going to be NASA's launch system. I would suggest that we not debate any more the wisdom of that choice. The fact is that the choice has been made. It's done. Now it's time to see what we can do with this system, which just happens to be the topic of this thread (hint, hint).

[Robotbeat]

...
Having covered this let me simply say that the SLS is going to be NASA's launch system. I would suggest that we not debate any more the wisdom of that choice. The fact is that the choice has been made. It's done. Now it's time to see what we can do with this system, which just happens to be the topic of this thread (hint, hint).

Chuck, if Ross were to have this attitude, DIRECT would never have happened.

Even if the bad choice is inevitable as you say, it's better to be fully aware that it is a bad choice than to shut our eyes and ears.

[clongton]

...
Having covered this let me simply say that the SLS is going to be NASA's launch system. I would suggest that we not debate any more the wisdom of that choice. The fact is that the choice has been made. It's done. Now it's time to see what we can do with this system, which just happens to be the topic of this thread (hint, hint).

Chuck, if Ross were to have this attitude, DIRECT would never have happened.

Even if the bad choice is inevitable as you say, it's better to be fully aware that it is a bad choice than to shut our eyes and ears.

Ross' attitude (and mine as well as others) was not that Ares was a bad choice, it was unsustainable and unworkable. SLS is neither. SLS is both sustainable and workable. I don't agree that it was a bad choice. Ares would never work. SLS can. That's the difference, and Congress has decided to move ahead with it.

[Robotbeat]

...
Having covered this let me simply say that the SLS is going to be NASA's launch system. I would suggest that we not debate any more the wisdom of that choice. The fact is that the choice has been made. It's done. Now it's time to see what we can do with this system, which just happens to be the topic of this thread (hint, hint).

Chuck, if Ross were to have this attitude, DIRECT would never have happened.

Even if the bad choice is inevitable as you say, it's better to be fully aware that it is a bad choice than to shut our eyes and ears.

Ross' attitude (and mine as well as others) was not that Ares was a bad choice, it was unsustainable and unworkable. SLS is neither. SLS is both sustainable and workable. I don't agree that it was a bad choice.

It's a matter of principle. Not everyone agrees with your view. I really hope you're right, but there's no reason to stop discussing alternatives or whether the decision really was wise or not.

If the answer to this thread is "not much," then that's not off-topic.

[sdsds]

#pragma crystal_ball on

Although SLS and Orion must be capable of ISS missions, commercial and international partners will meet ISS crew and cargo transportation needs.  Orion will fly with crew to LEO only twice:  once not involving ISS and once docking with ISS.

Beyond that, all the crewed Orion missions will be to destinations beyond LEO.[1]  And the political pressure to get there before 2020 will be intense!  Yet the funding will be tight and the piece that's going to get squeezed will be the full-sized SLS upper stage (SLSUS).  Years before 2020 NASA will identify the shortfall, but will be unable to get adequate SLSUS development funding when it would actually help, i.e. while Orion, the core and the boosters are also under development.

Seeing the pickle they are in, NASA will reluctantly adopt the Delta IV cryogenic 5 m upper stage (DCUS) or a close variant thereof as an interim solution, and use it for an "Apollo 8 in Hi Def" mission in 2021.  This is going to hurt though, because funding that otherwise would have gone to SLSUS development and integration will instead be diverted to integrating DCUS with SLS and Orion.  Thus SLSUS deployment will be even further delayed!  Whereas SLSUS might have been available by 2024 without DCUS on SLS, its first availability will be 2030 or later once the interim DCUS solution is chosen.

So there will be an entire decade, from 2021 through 2030, when every Orion flight will be on SLS and every flight will use DCUS for Earth departure.  Congress will fund 1.5 SLS launches per year, but only 1 out of three will be crewed Orions, so that means (10 * 1.5) / 3 = 5 Orion missions will depart LEO on DCUS.

#pragma crystal_ball off

For that reason, I think the capabilities of this stack deserve considerable analysis.  Other than lunar orbit and EML-1 or -2, and assuming its mass in LEO can be as much as 70 t, can the stack provide delta-v sufficient for any other destinations?

EDIT to add footnote [1]:  if China makes certain structural changes to its HSF program, NASA might agree to an LEO Orion/Shenzhou mission patterned on Apollo/Soyuz.

[mmeijeri]

Huh? With hypergolic refueling at L1/L2 you can go pretty much anywhere you want. I thought we agreed on this.

[Robotbeat]

Good question, sdsds. It's quite important because launching Orion with a DIVUS could be replicated with just rendezvous with the DIVUS (launched separately) in LEO... but I digress.

DIVUS plus Orion, starting out in LEO, could get to EML1/2 and back, but not (I think) to LLO and back (correct me if I'm wrong!). Thus, it would probably focus the lunar lander to be capable of starting at EML1/2 and going to the Moon's surface and back.

Long-duration Centaur may also be an important piece of the puzzle...

I'm not sure that an Apollo-8-redux would really have much support compared to something like an Apollo-11-redux (or Apollo 15). But yet again, I digress.

[mmeijeri]

Getting into LLO may be barely possible, but returning would require refueling. And L1/L2 is a better staging point anyway. Refuel a lander there and we can do what Constellation set out to do and then some.

[alexterrell]

But what's the optimum flight rate?
[...]
I might guess 4 flights per year, but I doubt if anybody knows.

Right.  In most optimization techniques you start by establishing a cost function and a set of constraints that limit the solutions you are going to consider.  Then (often) you pick a starting point inside the constraint bounds and begin "seeking" improvement, essentially by examining the slope of the cost function at your starting point and moving "downhill."

Right now though, the question is whether NASA can find any starting point that's within the constraints.  Is there any flight rate that can be achieved in a way that will satisfy Congress?

I assert NASA needs to design an SLS that will fly vehicles, "At least X times per year and at most Y times per year, at the direction of Congress" and then build that system.  I suggest values of X=1.5 and Y=3.0.  Building a system where e.g. Y=4.0 doesn't help if Congress never funds more than 1.5 flights/year.

(Apologies in advance to those who actually understand nonlinear programming optimization techniques.)

Right, though normally you end up with a result which is sensitive to unknown inputs, who's main benefit is to tell you which inputs you need to manage carefully.

Without yet knowing in detail what SLS is, my starting point would be to start off at a rate the shuttle delivered comfortably (4 flights per year), because I know the infrastructure can handle this, and then ruthlessly strip out costs - primarily those currently spent on Orbiter Maintenance.

I hope congress does not fund X or Y flights. At most, congress should say "Launch Operations are to cost $Z per year". Then, the mission for the head of SLS could be "Maximise Y within the constraints of $Z, and certain LOM/LOC variables. "

[sdsds]

With hypergolic refueling at L1/L2 you can go pretty much anywhere you want.

We agreed on this.  We might even agree on a few ways NASA could get hypergolic propellants to L1/L2.  But my crystal ball says they aren't going to attempt that, at least not until an L1/L2 station has been established as a place to marshal the elements of missions headed beyond.  And in this scenario, the mission modules would be in competition with the SLSUS for funding during a time when NASA is also flying Orion/SLS missions.  Any mission or station modules are going to have to be built on an extraordinarily tight budget!

(Furthermore NASA will likely also be developing a "demonstration" hydrolox depot and operating it in LEO sometime between 2021 and 2030.  That consumes additional funding during that period without actually contributing towards exploration before the 2030s.)

launching Orion with a DIVUS could be replicated with just rendezvous with the DIVUS (launched separately) in LEO...

Yes, if the rendezvous were "prompt" enough and/or the departure stage loiter time could be extended sufficiently.  It's a nice coincidence, but it does look like the margins would be slender!  What if you added in an Ariane-launched module and a European crew member?

[Robotbeat]

With hypergolic refueling at L1/L2 you can go pretty much anywhere you want.

We agreed on this.  We might even agree on a few ways NASA could get hypergolic propellants to L1/L2.  But my crystal ball says they aren't going to attempt that, at least not until an L1/L2 station has been established as a place to marshal the elements of missions headed beyond.  And in this scenario, the mission modules would be in competition with the SLSUS for funding during a time when NASA is also flying Orion/SLS missions.  Any mission or station modules are going to have to be built on an extraordinarily tight budget!

And that's the dilemma that we're in. This is why some folks aren't that excited about SLS without an included (measurable) budget increase.

(Furthermore NASA will likely also be developing a "demonstration" hydrolox depot and operating it in LEO sometime between 2021 and 2030.  That consumes additional funding during that period without actually contributing towards exploration before the 2030s.)

Just because it's a demonstration depot doesn't mean it can't be used!

launching Orion with a DIVUS could be replicated with just rendezvous with the DIVUS (launched separately) in LEO...

Yes, if the rendezvous were "prompt" enough and/or the departure stage loiter time could be extended sufficiently.  It's a nice coincidence, but it does look like the margins would be slender!  ...

That is true, but small performance improvements could bring the margins back up.

[sdsds]

DIVUS plus Orion, starting out in LEO, could get to EML1/2 and back, but not (I think) to LLO and back (correct me if I'm wrong!).

These mission plans require a sharp pencil and crisp module mass and propulsion Isp values, which I lack.  That said, I agree that entering and then leaving a low, circular lunar orbit might be beyond what this stack could accomplish.

I do however have high hopes for: (a) an Orion capable of longer duration missions than Apollo, and (b) a clever mission planner who finds a 3-body trajectory that gives "Orion 3" much of the political value Apollo 8 gave.  More concretely, there might exist a trajectory where Orion performed a propulsive maneuver at a high perilune, which sent it into a trajectory just shy of actual orbit around the Moon, but which (with another propulsive maneuver at some critical moment many days later) sends it back on a low perilune pass where a third maneuver gets it headed to atmospheric reentry.  On the low perilune pass we get "Apollo 8 in Hi Def" bread and circuses value.

[Robotbeat]

DIVUS plus Orion, starting out in LEO, could get to EML1/2 and back, but not (I think) to LLO and back (correct me if I'm wrong!).

These mission plans require a sharp pencil and crisp module mass and propulsion Isp values, which I lack.  That said, I agree that entering and then leaving a low, circular lunar orbit might be beyond what this stack could accomplish.

I do however have high hopes for: (a) an Orion capable of longer duration missions than Apollo, and (b) a clever mission planner who finds a 3-body trajectory that gives "Orion 3" much of the political value Apollo 8 gave.  More concretely, there might exist a trajectory where Orion performed a propulsive maneuver at a high perilune, which sent it into a trajectory just shy of actual orbit around the Moon, but which (with another propulsive maneuver at some critical moment many days later) sends it back on a low perilune pass where a third maneuver gets it headed to atmospheric reentry.  On the low perilune pass we get "Apollo 8 in Hi Def" bread and circuses value.

And sitting there at EML2 or EML1 are the following vehicles:

1) Lunar lander (slightly increased performance for EML1/2 rendezvous)
2) Small hab and EDS stage (for NEO mission)
3) Larger hab and EDS stage (for Mars orbital or Phobos mission)

If you take the Mars orbital or Phobos route, when you reach Mars orbit, two more vehicles are met there: A Mars lander to go to the Martian surface and a large in-space stage for going back to Earth!

What we need next are spacecraft!!!

[alexw]

Annual fixed cost for launching rockets into space is the sum of:
the maintenance of the manufacturing and launch facilities
supplying utilities to those facilities
maintaining an efficient workforce at those facilities
Fixed costs for a 1-rocket system (SLS)
Fixed costs for a 2-rocket system (Ares-I/V)
I think it's pretty clear that the fixed costs for a 2-rocket system are going to be close to 2x the fixed costs for a 1-rocket system.

    Chuck, SLS core-only (and STS) looks more like a two-rocket system; sustaining the SRBs apparently have an annual cost tolerably comparable to sustaining one of the EELVs, and the SRM workforce is entirely separate from the USA workforce. Ares-I and -V are at least a three rocket system (SRB, AIUS, AV-core), all of which are dedicated rockets with little in common with anything else (leaving aside the question of how to classify the AV-EDS).
     Compare this to something like Falcon 9, which is just one rocket in two pieces.
     -Alex

[clongton]

Annual fixed cost for launching rockets into space is the sum of:
the maintenance of the manufacturing and launch facilities
supplying utilities to those facilities
maintaining an efficient workforce at those facilities
Fixed costs for a 1-rocket system (SLS)
Fixed costs for a 2-rocket system (Ares-I/V)
I think it's pretty clear that the fixed costs for a 2-rocket system are going to be close to 2x the fixed costs for a 1-rocket system.

    Chuck, SLS core-only (and STS) looks more like a two-rocket system; sustaining the SRBs apparently have an annual cost tolerably comparable to sustaining one of the EELVs, and the SRM workforce is entirely separate from the USA workforce. Ares-I and -V are at least a three rocket system (SRB, AIUS, AV-core), all of which are dedicated rockets with little in common with anything else (leaving aside the question of how to classify the AV-EDS).
     Compare this to something like Falcon 9, which is just one rocket in two pieces.
     -Alex

That's a stretch. Is the Space Shuttle called a 2-rocket system? Is Atlas-V called a 2-rocket system? Is Ariaine called a 2-rocket system? And those are just for the solid boosters. What about Soyuz? Is it called a 2-rocket system? It has boosters that are not like the core. None of them are called 2-rocket systems yet all of them have boosters that are different than the core. Please note my use of the qualifier "system" in my original post. SLS is not a 2-rocket system just because it has strap-on boosters. The difference is whether or not the boosters also fly by themselves as their own rocket. Ares was a 2-rocket system because it involved launching 2 different rockets, that had no other mission but to fly together as a system. NASA had to build and maintain facilities and workforce for 2 completely separate "flying" rockets.

[Warren Platts]

Look, the thing to do is focus on what the mission is going to be. Clearly, despite the words of Buzz and our beloved President, the Moon continues to loom large. And to do the Moon on a cost-effective basis entails extensive surface operations, and to really maximise your bang for the buck in that regard, we want permanent manned presence (and correct me if I'm wrong, but didn't one of the congressional resolutions mandate "permanent" manned presence beyond LEO?).

So if you're going for a permanent station grounded on the Moon, you've got to go for the polar regions: not just for the ice that our best science says is there, but also because of the relatively benign thermal environment and extended solar power hours. To go for the poles, you need an L1/L2 fuel depot (manned space station at L2 is expensive luxury we cannot afford--we already have ISS, and then we'll have the lunar station).

Since we're going for the poles, we should go for combined basic lunar science/astronomy/ISRU rocket propellant program. (The anomalous polar crater floors are mostly out of line of site with Earth, and thus ideal for building radio antenna arrays without interference from Earth--no need to build on the dark side.)

To do all this in a semi-timely manner, we'll need around 700 tons of propellant to LEO per year.

This is where the SLS could really shine, given that it actually gets built: it's main job should simply be lofting ACES-71 space tankers. The surfeit of propellant this would provide would obviate the need for Orion. The ACES-41 DTAL lander/ascender would be the BLEO transfer vehicle. It's roomier than Orion in any case. If refueled at L2, fully propulsive reinjection into LEO wouldn't be a problem, especially if and when ISRU propellant comes on line.

In this manner, it is possible to fuse the SLS into the ULA commercially-based lunar architecture plan. The most common objection I've seen to the ULA plan is the "you've got to be kidding: 30 launches per year!?!" objection. The SLS could get all the propellant the ULA plan calls for in a mere 10 launches: yet 10 launches of SLS per year is much more than the current plan of 1 or 2 per year. If we can get the launch rate up, OV-106 says we can get the per/kg cost down to $5K/kg or possibly even much less, down into the $2K-$3K/kg range.

Thus, what ULA calls an "aggressive" program would only be about 4 or 5 launches of Delta IVH: 3 manned landers + 2 cargo landers per year. Actually, the ascenders could be reused, so you wouldn't have to make 3 of those year after year. Once ISRU propellant came on line, the main DTAL stage could also be substantially reused.

Since there would be no Orion, humans could simply ride up to orbit to the ISS using Dragon, Soyuz or the new Boeing capsule. Once at ISS, they could transfer to the ascender stage mated to the ACES DTAL.

This could all be done for $7-$8B per year, which would leave a couple of billion leftover for ISS to keep it flying forever.

Piece of cake...

[Lampyridae]

DIVUS plus Orion, starting out in LEO, could get to EML1/2 and back, but not (I think) to LLO and back (correct me if I'm wrong!).

These mission plans require a sharp pencil and crisp module mass and propulsion Isp values, which I lack.  That said, I agree that entering and then leaving a low, circular lunar orbit might be beyond what this stack could accomplish.

I do however have high hopes for: (a) an Orion capable of longer duration missions than Apollo, and (b) a clever mission planner who finds a 3-body trajectory that gives "Orion 3" much of the political value Apollo 8 gave.  More concretely, there might exist a trajectory where Orion performed a propulsive maneuver at a high perilune, which sent it into a trajectory just shy of actual orbit around the Moon, but which (with another propulsive maneuver at some critical moment many days later) sends it back on a low perilune pass where a third maneuver gets it headed to atmospheric reentry.  On the low perilune pass we get "Apollo 8 in Hi Def" bread and circuses value.

Sounds like a good idea. SLS may wind up having more oomph than we bargained for anyway, given DIRECT's conservative margins.

I think it would be expedient to develop a Wal-Mart lander for two. Send it up on a 2nd J-130 DIVUS, use the US as a crasher stage. But getting back to EML-1 might be a problem with regards to endurance and a hypergolic ascent stage...

[alexw]

Annual fixed cost for launching rockets into space is the sum of:
the maintenance of the manufacturing and launch facilities
supplying utilities to those facilities
maintaining an efficient workforce at those facilities
Fixed costs for a 1-rocket system (SLS)
Fixed costs for a 2-rocket system (Ares-I/V)
I think it's pretty clear that the fixed costs for a 2-rocket system are going to be close to 2x the fixed costs for a 1-rocket system.

    Chuck, SLS core-only (and STS) looks more like a two-rocket system; sustaining the SRBs apparently have an annual cost tolerably comparable to sustaining one of the EELVs, and the SRM workforce is entirely separate from the USA workforce. Ares-I and -V are at least a three rocket system (SRB, AIUS, AV-core), all of which are dedicated rockets with little in common with anything else (leaving aside the question of how to classify the AV-EDS). Compare this to something like Falcon 9, which is just one rocket in two pieces.

That's a stretch. Is the Space Shuttle called a 2-rocket system?
Is Atlas-V called a 2-rocket system? Is Ariaine called a 2-rocket system? And those are just for the solid boosters. What about Soyuz? Is it called a 2-rocket system? It has boosters that are not like the core. None of them are called 2-rocket systems yet all of them have boosters that are different than the core. Please note my use of the qualifier "system" in my original post. SLS is not a 2-rocket system just because it has strap-on boosters. The difference is whether or not the boosters also fly by themselves as their own rocket. Ares was a 2-rocket system because it involved launching 2 different rockets, that had no other mission but to fly together as a system. NASA had to build and maintain facilities and workforce for 2 completely separate "flying" rockets.

    I'm thinking in terms of your comment above about "the sum of:
the maintenance of the manufacturing and launch facilities ... [and] workforce".
   In that light, yes, STS is something like a two-rocket system. Ariane V almost certainly is as well -- but at least her SRBs come from and overlap costs with the French SLBM production effort, whereas apparently ATK's SRB production have very little to do with ATK's Trident production. (Hence the whole red herring about SRBs for national defense, etc.) For Soyuz, I don't know for certain, but aren't her stages mostly just variations on a theme? For Atlas V, it's a little harder to say, but historically I imagine that the Colorado and San Diego production were quite separate (after all, both Centaur and Atlas flew without each other.) I don't know much about Decatur, but nowadays it really sounds like they are condensing something like five separate vehicle teams down into just three: Atlas core, Delta core, Common Upper Stage. SpaceX (and eventually -Heavy) is a clear-cut counterexample: the same facility and overhead build the core, the LRBs, and the upper stage (plus the Falcon I core and upper stage), start to finish: 1 rocket group. It sounds like Orbital is also something like a single unified operation.
   I'm not seeking to define it based on whether a vehicle has strap on boosters, and I don't think your proposed definition of whether the strap-ons also fly on their own is the essential point, either. SLS's boosters could also be used to launch small payloads (and Ares I-X did!), but whether they are or not doesn't change the essential fact that SLS requires maintaining two completely different rocket production facilities with thousands of workers and nothing in common: Utah and Michoud, building rockets that overlap with noone else. In that sense, SLS is a two rocket system.
   -Alex

 

[dlapine]

Alexw, the rest of us define it by the number of launches required. No gray areas there.

Did anyone put out proposals for asteroid defense?

I know that we have already some coordinated efforts to identify potentially threatening objects; is there any possibility about doing something with 70MT to orbit, and maybe propellent depots? This is blue sky, and we'd better not need it before 2017, but still, is there anything we reasonable do about it with SLS capability?

[guru]

This is a fairly inefficient mission plan that uses the SLS to send two astronauts to Mars for a few days.  It's called BasicMars, but it is just a revamping of an earlier concept I had that I called Mars To Go.  A pdf overview can be found in the main thread about halfway down the page:

http://forum.nasaspaceflight.com/index.php?topic=23169.81