Apollo: The Case for the Saturn C-3

[GovernorGeneral]

Apollo: The Case for the Saturn C-3

The Saturn V is arguably the most famous rocket ever designed, built, and launched into space by humanity. It was selected from among the options of the Saturn Family because of its ability to execute a Lunar Orbit Rendezvous mission in a single launch. During its history, it had a 100% safety record with no human fatalities. In cases with partial failures of single engines, other engines were able to pick up the slack and complete the mission.

For all its success, the Saturn V was nevertheless expensive to operate, and its launch capabilities vastly exceeded that necessary for the post-Apollo missions. The Saturn V could put 140 metric tons into Low Earth Orbit. Compare this to the approximately 20 tons for the Atlas V and 29 tons for the Delta IV Heavy of later decades.

During the Apollo rocket selection process, Direct Ascent was discarded because of the huge launch mass requirements. Earth Orbit Rendezvous was discarded because of the prohibitive number of launches required and because the mechanics of landing and relaunching a direct ascent vehicle from the Moon remained. Lunar Orbit Rendezvous was selected because it required a smaller launch vehicle than DA, but far fewer launches than EOR.

What if another approach was chosen: Earth And Lunar Orbit Rendezvous?

The Saturn C-3 had almost 1/3 the payload capacity of the Saturn V; 45 tons versus 140 tons, but required less than 20% of the launch mass, meaning it was in some sense a more efficient launch vehicle.

The combined mass of the Apollo Command and Service Module and Lunar Modules was 44.6 tons; Saturn C-3 could get that package into low earth orbit, albeit barely.

The S-VIB stage consisted of 13.5 tons of dry mass plus 109.5 tons of fuel mass. Of the fuel mass, only 67% was required to put the CSM-LM stack onto Trans Lunar Injection (33% having been used to finish the Saturn V’s LEO burn); 87 tons of dry mass and fuel mass combined for the Apollo TLI. In theory, two Saturn C-3 launches could assemble the Earth Departure Stage in low earth orbit with a 3 ton buffer. The problem then is how to design an Earth Departure Stage in two 45 ton components whose orbital assembly is not overly complicated.

Proposal:

Take a Saturn C-3 rocket, remove the payload, attach a guidance and docking interface and a detachable aerodynamic nose cone to the top of the S-IVB stage. Launch two of these into low earth orbit and dock them remotely.

Option A:

Top up the partially emptied tanks of one S-IVB from those of the other. Discard or de-orbit the tanker. Designate the refueled S-IVB as the Earth Departure Stage.

Launch the CSM-LM stack using a third Saturn C-3. Perform transposition and docking to extract LM from the launch stage. Rendezvous and dock with the Earth Departure Stage. Put a second docking attachment on the bottom of the LM with structural strengthening struts of some kind around LM, for the TLI burn. Assuming these problems are resolved, execute TLI burn in an eye-balls out configuration with astronauts facing backwards. En route to the moon, CSM-LM jettisons EDS.CSM-LM executes Lunar Orbit Insertion burn as historically.

This requires some mass to be dedicated to the remote refueling and LM strengthening, but the refueling takes place without astronauts right next to it.

Option B:

Leave both tanker stages tethered to await the CSM-LM.

Launch the CSM-LM stack using a third Saturn C-3. Rendezvous and tether with the Earth Departure Stage. Astronauts attach hoses to transfer fuel from tethered tankers to refill the CSM-LM’s S-IVB launch stage. Jettison tankers. Execute TLI burn as historically with the CSM-LM-S-IVB stack intact.

Theoretically more mass efficient, but in practice I suspect there would be all sorts of problems, especially with orbital refueling by hoses of liquid hydrogen and liquid oxygen. A bunch of half full fuel tanks spinning around their combined center of mass on tethers and then smashing into each other if something goes wrong; e.g. a rogue RCS thruster fires at the wrong moment, comes to mind.

Result:

Same TLI payload capability, but spread across three launches of much smaller rockets than the Saturn V. NASA ends the Apollo program with a more useful / less overpowered launch vehicle than the Saturn V, but with significantly more payload capacity than any of the OTL successors, and with less complexity and more reliability than the Shuttle. NASA’s manned spaceflight vehicle remains the proven Apollo CSM and its likely iterative successors.

The 1970s, 1980s, and 1990s all see far more capable and massive and/or faster deep space probes launched on Saturn C-3s. TTL has a larger 1976 Viking Lander which is an Apollo Command Module with the guts ripped out and replaced with the probe and a propulsive landing system to augment the parachutes.

Overall, the American space program post-Apollo focuses the launch vehicle work on improving and refining the Saturn C-3, perhaps with strap-on boosters if such a capability was required. This streamlined, economical approach leads to improved reliability and less duplication of effort. The fortune sunk into the Shuttle can be spent on a space station, better deep space probes, or even Venus and Mars flybys.

How practical is this Kerbal-esque Saturn C-3 “cunning plan”?

 

[Chewbacca]

The Saturn V was designed for Mars missions. It was Von Braun's vision to go to Mars as soon as the moon was conquered. Besides, ANY space program advancements besides those of OTL (such as they were) depended on the Vietnam War ending before 1968 (or never taking place).

 

[e of pi]

The general idea is an interesting one--I've explored similar EOR/LOR profiles in a few timelines including one not yet posted and either EOR/LOR or LOR/LOR seem to be keystone elements in current real space planning for the cancelled Constellation or the new Artemis program.

The devil, of course, is in the details. The first one you need to consider in a three-launch architecture with multiple EOR steps is hydrogen boiloff. I don't have a citation immediately near to hand, but the loiter time of a fueled S-IVB in LEO with hydrogen aboard for TLI was measured in single-digit days. In Gemini, NASA would eventually once manage a 9-day turnaround off the same pad, but in 1961 when the mode decision is being made, that capability is much less clear. Managing three C-3 launches in...probably about 5 days means, likely, three launch complexes. Managing two uncrewed tanker C-3 launches, getting them to autonomously dock in space, and then get a crew to the scene before a week has passed from the first launch is also a very complex task with that timetable. The crew is last to launch, so if there are launch failures, docking issues, or other friction causes the mission to end up with boiloff or other complications leading to a mission abort you can do so with the crew never leaving the ground...but you'd still wasted at least one and maybe two C-3 launches.

Second, automated rendezvous and docking is a new task you're adding in, and not a trivial one. By the early 70s is was solved, but betting on computers to do what in 1961 human pilots were still working out in simulators as a critical element of the project (and with little margin for error on timing) is another challenge.

A C-3 lasting into the 70s, perhaps with modifications for recovery and reuse, would be a tremendously capable vehicle for a station program, but that's a pretty weak argument for an American program under the immediate need for an assured way to get to the moon. LOR required a single rendezvous and docking, handled by crew, and thus a manageable risk. Even accepting that, the Saturn V ended up barely capable of the final stack even though it should have had substantial margin--Apollo and particularly the Lm ended up needing every available pound to hit the lunar goal within the abbreviated timeline. While the C-3 might be a more useful rocket later, it would be the wrong pick in 1961/62 given the existing OTL requirements.

 

[GovernorGeneral]

The general idea is an interesting one--I've explored similar EOR/LOR profiles in a few timelines including one not yet posted and either EOR/LOR or LOR/LOR seem to be keystone elements in current real space planning for the cancelled Constellation or the new Artemis program.

The devil, of course, is in the details. The first one you need to consider in a three-launch architecture with multiple EOR steps is hydrogen boiloff. I don't have a citation immediately near to hand, but the loiter time of a fueled S-IVB in LEO with hydrogen aboard for TLI was measured in single-digit days. In Gemini, NASA would eventually once manage a 9-day turnaround off the same pad, but in 1961 when the mode decision is being made, that capability is much less clear. Managing three C-3 launches in...probably about 5 days means, likely, three launch complexes. Managing two uncrewed tanker C-3 launches, getting them to autonomously dock in space, and then get a crew to the scene before a week has passed from the first launch is also a very complex task with that timetable. The crew is last to launch, so if there are launch failures, docking issues, or other friction causes the mission to end up with boiloff or other complications leading to a mission abort you can do so with the crew never leaving the ground...but you'd still wasted at least one and maybe two C-3 launches.

Second, automated rendezvous and docking is a new task you're adding in, and not a trivial one. By the early 70s is was solved, but betting on computers to do what in 1961 human pilots were still working out in simulators as a critical element of the project (and with little margin for error on timing) is another challenge.

A C-3 lasting into the 70s, perhaps with modifications for recovery and reuse, would be a tremendously capable vehicle for a station program, but that's a pretty weak argument for an American program under the immediate need for an assured way to get to the moon. LOR required a single rendezvous and docking, handled by crew, and thus a manageable risk. Even accepting that, the Saturn V ended up barely capable of the final stack even though it should have had substantial margin--Apollo and particularly the Lm ended up needing every available pound to hit the lunar goal within the abbreviated timeline. While the C-3 might be a more useful rocket later, it would be the wrong pick in 1961/62 given the existing OTL requirements.

Those are good points. I wonder if it is possible to uprate the C-3 enough with boosters so that the S-VIB has enough fuel for the CSM-LM’s TLI, meaning only one EOR is needed. I suspect the number of boosters to effectively double the LEO payload would be prohibitive.

That begs the question of, “why not the C-4?”, which is, while better than the Saturn V post-Apollo, still overpowered.

Maybe the better way is to posit that NASA uses the Saturn V, but builds one of those every year instead of the Shuttle.

 

[e of pi]

Those are good points. I wonder if it is possible to uprate the C-3 enough with boosters so that the S-VIB has enough fuel for the CSM-LM’s TLI, meaning only one EOR is needed. I suspect the number of boosters to effectively double the LEO payload would be prohibitive.

That begs the question of, “why not the C-4?”, which is, while better than the Saturn V post-Apollo, still overpowered.

The number of boosters isn't necessarily prohibitive--with Titan-sized solids (then still under development, off the shelf by the time Saturn V derivatives were considered in the later 60s), you can easily double the performance of the core, which gets you where you need to be. Two launches is a lot easier to manage than three. However, as you point out, it's still a question of why not simply use the C-4...and the C-4 is still oversized for the OTL post-Apollo requirements. Besides, once you have the C-4, growing the S-IC from four engines to five is mostly a matter of putting an additional engine in the center of the existing cross-shaped thrust structure, as Von Braun did to create the C-5 historically, and gives even more margin to enable EOR or LOR as options. It's worth noting that when the Saturn V was picked, the final mode decision wasn't even made yet, though early discussions and studies on LOR had occurred.

Maybe the better way is to posit that NASA uses the Saturn V, but builds one of those every year instead of the Shuttle.

Sadly, one Saturn V can't do the job of nearly a dozen Shuttles a year, even if the theoretical summed payloads work out to less than the Saturn V's LEO launch capability. Too many different orbits, too many different missions, too much required payload fairing volume...and a single unit produced per year has to incorporate the total cost of the entire "standing army" of production and launch teams, meaning the cost per flight may be eyebrow-raising, as we see with SLS.

This kind of thinking on a way to save heavy lift in the 70s and 80s after Apollo is a lot of what lead @Workable Goblin and I to create Saturn 1C and Saturn Multibody for our Eyes Turned Skyward timeline--have a core 20 to 30-ton capability that can serve Shuttle-type roles, and then create a heavy clustered rocket from those cores that can serve the occasional station heavy lift mission until a lunar mission requirement re-opens--a way to get large build rates but avoid overcapacity issues. Arguably, though, it's got a lot of the worst of both world in needing much of the development cost of a new rocket (new F-1 powered monolithic lower stage tanks and thrust structure) without even much potential for cost reductions that Shuttle was at least at the start believed to offer.

 

[marathag]

And is fuel transfer even possible in microgravity? Even just for staging ullage solids were burned to settle the contents of the tanks before ignition

 

[Workable Goblin]

And is fuel transfer even possible in microgravity? Even just for staging ullage solids were burned to settle the contents of the tanks before ignition

Well, the Russians have done it quite a number of times with their Progress tankers. But doing it with cryogenics is harder and although it's definitely possible it's not necessarily easy. Just another critical thing on the path...

The C-3 makes the most sense in a situation where the United States is following something more like the original Apollo plan, where they were going to build a space station and do some circumlunar flights before doing a lunar landing sometime maybe eventually. In that case, there's little reason to go for the Saturn V, since there's plenty of time to figure out how to do everything with smaller launch vehicles before you actually do the mission, and the smaller vehicle is more useful in the interim. This probably requires that the Space Race have been "won" by the United States early on, though.

 

[e of pi]

And is fuel transfer even possible in microgravity? Even just for staging ullage solids were burned to settle the contents of the tanks before ignition

You don't need a lot of acceleration for settling, so there's proposals for things like using a small amount of propellant ullage gas for tiny thrusters (like the solids, but even less thrust and pulling from the main supply) or rotating the tanks along their axes to settle with centripetal/centrifugal forces. It's solvable...but as @Workable Goblin said, it's another complication. You might see some attempt to study it under Gemini.

 

[Rickshaw]

100% safe? What was Apollo 1 mounted on? It lost three crew.

 

[e of pi]

100% safe? What was Apollo 1 mounted on? It lost three crew.

Not the rocket's fault.

 

[Rickshaw]

Not the rocket's fault.

Still sitting on top of the rocket. You cannot claim that the rocket was 100% safe. It wasn't.

 

[e of pi]

Still sitting on top of the rocket. You cannot claim that the rocket was 100% safe. It wasn't.

To be a little crude about it, every Saturn I, IB, and V that launched worked. The rocket which was under Apollo 1 at the time of the fire was not responsible for or effected by the disaster. It was destacked at LC-34, restacked at LC-37, and later launched the Apollo 5 mission which was the Lunar Module's first flight (uncrewed) in space. If I got shot while driving a Toyota, that's not the Toyota's fault for being unsafe.

 

[Athelstane]

Sadly, one Saturn V can't do the job of nearly a dozen Shuttles a year, even if the theoretical summed payloads work out to less than the Saturn V's LEO launch capability. Too many different orbits, too many different missions, too much required payload fairing volume...and a single unit produced per year has to incorporate the total cost of the entire "standing army" of production and launch teams, meaning the cost per flight may be eyebrow-raising, as we see with SLS.

I agree (of course) with your entire post, though this section stood out to me for pondering, and not just because NASA never managed a dozen Shuttle flights in a year - obviously, they were thinking for a long while that it *was* feasible - but more because it begs the question of what NASA would actually need a dozen Shuttle flights for in the first place, since so many of the flights it *did* fly were borderline makework science missions . . . but then once again, we're looking at Shuttle here with the benefit of some hindsight, since we know it wasn't viable or sensible for all of U.S. government (let alone commercial) payloads.

No, the more interesting question is the Saturn V side of the question. It strikes me that two launchers per year is the absolute minimum that could justify the program, not just for amortization and launch cadence safety but also because by the 70's NASA would want to be using some kind of Applications hab like AES, ALSS, or even LESA to get its money's worth out of lunar sorties before long, and that means a second launch. But I also think Bob Gilruth was right to think NASA was playing Russian roulette with every Apollo lunar mission, and it would not have been long before it found a loaded chamber. The 70's probably really was just too soon for a sustained lunar program.

 

[Athelstane]

To be a little crude about it, every Saturn I, IB, and V that launched worked.

Well, Apollo 6 would have been an abort if crew had been aboard....

Still, it's a niggle. It's still amazing to me just how successful the Saturn family ended up being. Certainly had nothing to do with Apollo 1.

 

[Rickshaw]

Three dead Astronauts would disagree. It was a Saturn rocket underneath. Therefore, you cannot claim that all Saturns were 100% without casualties.

 

[Polish Eagle]

Well, Apollo 6 would have been an abort if crew had been aboard....

Still, it's a niggle. It's still amazing to me just how successful the Saturn family ended up being. Certainly had nothing to do with Apollo 1.

IMO, Apollo 6 should be counted as a loss-of-mission due to LV failure. The point was to demonstrate TLI, the S-IVB failed to restart, so the primary mission objective was not fulfilled.

Anyway, back to OP's proposal:

There is a potential Option C: Daisy-chained upper stages. If you put a docking collar on each end of a stage, you can fire them off in series. This eliminates the need to develop microgravity cryogenic propellant transfer technology.

I think NASA could be sold on an architecture that requires two same-day Saturn launches--they did same-day launches for Gemini-Agena, and the Skylab 2 crew was supposed to follow their space station into orbit the day after launch, so rapid range turnaround isn't the problem. It would require at least LC-39C for redundancy, in case of a vehicle failure on the pad. (EDIT: Unless maybe LC-37 could be reconfigured to support the slightly-bigger rocket? It's a factor-of-two increase of thrust, but that's not as bad as the factor-of-five Saturn V imposed).

Conceivably, you could extend the lifetime of stages on-orbit by not being so insistent on LH2--go with propane or methane, for example, or hydrazine/N2O4, at the cost of needing a lot more launch mass and rendezvous events per mission (though you can extend spacecraft lifetime to weeks or months at a time, so a single failed rendezvous doesn't have to kill the mission). But I think that would require too much change in the NASA mindset at the time.

Though I wonder if you couldn't circumvent the problem by not rendezvousing the LM and CSM until you are already in LLO--that is, launch the LM first and let it hang around in a lunar orbit until the crew catches up with it a few days later in their CSM. This eliminates the LH2 storage issue, since each launch burns its LH2 within hours of start (or, at most, single-digit days if you use an LH2/LOX stage for LLO insertion--Douglas proposed S-IVB mods to do just that IOTL). After TLI, the delta-v for LLO is about 900 m/s--a lunar module with a partially-fueled Titan Transtage could perform that maneuver, if you can put 23 tonnes of payload through TLI. That's about halfway between C-3 and C-4. Maybe an uprated C-3, with OTL's S-IVB in place of the S-IV? A two-launch LOR-only mission.

 

[Workable Goblin]

Three dead Astronauts would disagree. It was a Saturn rocket underneath. Therefore, you cannot claim that all Saturns were 100% without casualties.

I really do not think they would disagree. There was not one shred of evidence that the Saturn IB had the slightest effect whatsoever on Apollo 1's fire, which would have happened just the same if they had been doing the same test with the same capsule but in a warehouse instead of on a rocket. It makes no sense whatsoever to include Apollo 1 in the Saturn IB's safety record, because the fire had nothing to do with the rocket and doing so just obscures the actual point of looking at rocket safety records (i.e., what are the dangers associated with the rocket, as opposed to with the capsule or with the ground equipment or what not).

In any case, the OP was referring to the Saturn V, not the Saturn IB. Apollo 1 was certainly not mounted to a Saturn V when it had a fire. There were no fatalities during a Saturn V-launched mission.

 

[Rickshaw]

I really do not think they would disagree. There was not one shred of evidence that the Saturn IB had the slightest effect whatsoever on Apollo 1's fire, which would have happened just the same if they had been doing the same test with the same capsule but in a warehouse instead of on a rocket. It makes no sense whatsoever to include Apollo 1 in the Saturn IB's safety record, because the fire had nothing to do with the rocket and doing so just obscures the actual point of looking at rocket safety records (i.e., what are the dangers associated with the rocket, as opposed to with the capsule or with the ground equipment or what not).

In any case, the OP was referring to the Saturn V, not the Saturn IB. Apollo 1 was certainly not mounted to a Saturn V when it had a fire. There were no fatalities during a Saturn V-launched mission.

Ah, finally an answer to my original questions, "what was under Apollo 1?"

Now we know it was a Saturn 1B, you can claim nearly what ever you like about the Saturn V.

 

[sonofpegasus]

Is it not correct to say that the Saturn V rocket had a 100% safety record but the Apollo capsule did not. The fire in Apollo one could have happened whether it was on to of the 'stack' or not if the same tests were being carried out. That I think is the important differentiation here.

 

[Athelstane]

Ah, finally an answer to my original questions, "what was under Apollo 1?"

Now we know it was a Saturn 1B, you can claim nearly what ever you like about the Saturn V.

Well, we can claim the Saturn IB was flawless, too, because as WG says. it had absolutely nothing to do with the AS-204 fire. Seriously, you can read the report. Th Saturn wasn't even fueled. The fire would have happened just as readily, as WG says, if the capsule was stationed inside a warehouse for the plugs-out test.

This contrasts with an accident like AMOS-6, where the Falcon 9 second stage exploded on the pad, thus destroying the payload. Not an in-flight failure, but certainly a failure which happened on the launcher. But that is not what happened with Apollo 1.