Sputniks... an Alternate Space Race

[Puget Sound]

I wonder what exactly the President meant by that. Did he mean, private enterprises should negotiate with NASA for leasing/renting launch and ground support facilities, or that they should not only purchase rockets and build their own satellites, but also construct their own parallel infrastructure--pads, tracking, the whole thing--wherever they felt like, out of pocket?

JFK was actually economically conservative, so my guess is that he wanted to promote private enterprise in space.

 

[Shevek23]

JFK was actually economically conservative, so my guess is that he wanted to promote private enterprise in space.

Indeed the Democrats in general, and in particular those who rise to the highest levels of government, are and always have been staunch supporters of private enterprise.

The question is, what exactly does supporting private enterprise mean? Does "supporting private enterprise" mean staying completely out of private business and letting the chips fall where they may? Or do we in fact get a better outcome, from the point of view of there being more enterprise and more freedom in the sense of there being sharper competition that spurs established players to greater efforts and allows access to the markets for innovators and gives the public greater freedom of choice for superior products, with a bit of strategic intervention in various forms here and there? A tax credit here, a subsidy there, a federal agency to carry out research and the phases of development that don't appear immediately as sure winners to private investors, and perhaps an organized cartel of some kind or other under government auspices to draw in investors, give some assurance of returns and access to markets in a situation where it all seems dubious?

In fact American business history is rife with such forms of government "interference." This is one of many reasons why I personally look quite wryly on the alleged "magic of the marketplace" and the claims academic economists so often make that the "pure" market solution is always the superior outcome for everyone. In fact I doubt there is any mindless mechanism that can guarantee good results; the market of course is not mindless but the product of many minds, but the jam-ups possible are all too obvious, and pragmatically speaking American politicians of all stripes have often jumped in to smooth the path of the economic machine forward.

...
Comsat development is not my field of expertise (if, indeed, I have any), so in the tradition of Marvel Comics, you get a No-Prize if you can make my take on events plausible.

Well, I think I gave it my best shot already, pointing out that contrary to US right-wing mythology, the Democrats generally are not covert agents of Communism and are deeply committed to the ideology of private enterprise.
I guess it's not implausible that the young President might well listen to lobbyists arguing that private enterprise should best meet privately opted demands for new kinds of services, rather than bogging them down in bureaucracy.

Still a dumb move I think; I suppose AT&T might have caught him on a busy day and he made a snap decision he found himself committed to, while the bad consequences of it boiled down to the absence of a service unknown before in either timeline, and delayed with respect to OTL--but they don't know what they are missing ITTL. And even if JFK avoids an assassin's bullet in this timeline and wins a second term, he will be out of office before the times when OTL ordinary Americans started to observe the benefits of comsats in action; if anyone grumbles, most likely in the pages of columns in SF magazines like Analog or Galaxy, about their slowness in development it won't be until the 1970s. So pragmatically speaking the President probably won't lose much sleep over it.

I have to wonder if the delay of their development leads to a knock-on delay in the development of military comsats, for lack of stimulation. Presumably the Soviets will be somewhat faster to develop them and that might substitute quite well for the stimulus of the absent civil developments.

But that implies that the gap OTL between Soviet unmanned satellite capabilities and US ones will be much narrower, or might even favor the Russians.

As I said, it was a dumb decision on Kennedy's part in part because of considerations of national security.
...

I think the X-15 platform was a dead end, developmentally. The Air Force already had a number of designs which looked like the X-20 very early on, so it made sense that they would stick with it (factoring in lessons learned from Crossfield's flight and OOC the fact that I have lots of pictures of X-20 and X-20 derivatives and no pictures of this hypothetical X-15c)

Aside from the lack of images, because no one was after all designing pointy-needle spaceplanes OTL, I think you're probably right. After all the Soviet Spiral aka MiG-105 while it looks less like a Shuttle than the X-20 (which really isn't much like one either, my impression it did came from the angle of view in the painting, the tip of the delta wing looks like the thicker blunt nose of a Shuttle in that perspective) also looks nothing like an X-15, nor would its grown-up hypothetical operational descendants, nor other rival Soviet advanced reentry vehicle concepts.

But I was a bit infatuated with my notion of a Shuttle alternative that is strictly for hauling people, not cargo, up and down, and I was visualizing it as an expanded X-15. I should be visualizing it as an expanded X-20 I guess!

And not too expanded at that--one mode of operations planned for the Dyna-Soar series was indeed to use it as a "shuttle", by substituting a 4-astronaut passenger pallet for the standard recon or weapons gear intended for direct military operations. Assuming that wasn't an unrealistic goal, the Dyna-Soar, as a transport, would already have been just 2 passengers short of the upper target of 7 for a Shuttle launch. So just a slight upgrade might allow it to meet and exceed Shuttle capacity as far as hauling people up and down went. I have some question as to whether the X-20 really already would have had sufficient capability for 5 astronauts realistically, but at any rate it could surely take up several of them, counting the pilot.

... Artemis is a fitting name, I think. Moreso than Apollo.

I quite agree!

EOR is actually much heavier than LOR. You need to send enough fuel to land the equivalent of the Apollo CSM *and* LEM and then blast the CSM off again.

But it doesn't require tricky rendezvouses around the moon. Before Gemini, we didn't realize such things weren't that hard.

Another advantage of EOR is you get much longer lunar stays since you can carry more provisions with you on the heavier spacecraft.

The way you deal with the possibility of launch failure is to have a fuel-refrigeration station in orbit to keep the fuel stable until a crew can be launched.

I wasn't sure if would be a lot heavier; on one hand you need to send a heavier lander stage (integrated with the Command/Service stage instead of a separate lander ascent module) to take the full weight of these return stages all the way down to the Moon, then the Command Module or some separate lunar-launch rocket has to have the fuel to lift those two modules back off the moon, and the heavier lander stage with its extra fuel for landing and takeoff has to be sent to Luna in the first place. But with OTL LOR, we after all needed that separate ascent module which is streamlined out of the EOR design completely, so I figured most of the mass difference might come out in the wash.

A big difference, aside from the orbital rendezvouses being closely supervised from the ground in Earth orbit, is that even though the total mass to be sent Moonward is higher, the load is broken down into two or three or more separate launches from Earth, so the rockets to get those separate loads up are now that much lighter and the development path toward them is that much shorter. Another bonus, probably not considered (and offset by the extra logistics and risks involved in many launches rather than one big one) is that the economics of just about any enterprise improves with increased volume. If an Artemis mission requires say three launches, then for a given number of planned moon missions, a smaller, cheaper, probably standardized booster is being produced in triple the numbers of the OTL Saturn V, and being handled with equipment and facilities that have to be doubled or trebled but are individually smaller and simpler. In some ways the investment is larger, in others smaller, but for a given investment you have larger numbers of rockets produced and handled. Also, after the Artemis program succeeds, if there is a sudden plug-pulling as OTL, it will be easier to persuade political leaders to keep some kind of program going on the basis of the smaller standard rocket, much closer to the right size for modest orbital missions and space probe launches and so forth.

I suppose you'll eventually tell us just how many launches an Artemis mission requires. I've been thinking three--launch a somewhat bigger than OTL translunar injection stage into orbit, then launch the Lunar Lander/Ascender (or simply fuel tanks for the SM, to replace a separate Ascender rocket and to be bolted on to the sides of the SM during assembly rendezvous) and finally the last launch is a Command Module/Service Module package pretty much to OTL standards. Perhaps the packages won't break up into thirds so neatly like that though! Anyway the procedure would be to first pick up the Lander stage, doing a rather tricky backward dock. Then link up to the translunar rocket. If the thrust of that rocket is not too high--a long slow burn, saving on engine mass--perhaps the CM can dock to it nose-on and be driven to the Moon backwards at a low acceleration? Aside from the nose grapple I'm picturing support struts that would swing out of stowage to lock to sockets on the SM, to make this upside-down load of CM/SM/LL less awkward! Anyway backwards is how the ship has to approach its Moon landing anyhow. Clearly this plan involves good rear-view mirrors, or more sophisticated versions of same like periscopes or closed-circuit TV!

On the comsat issue, I think AT&T was willing to trade government subsidy (which is what Intelsat was) for monopoly, and then realized it just wasn't profitable. But they didn't care because as long as no one else could afford it, AT&T could keep doing what it did on Earth and not have to worry about competition from space.

I think this sort of short-sightedness in the pursuit of monopoly is sadly quite plausible.

And I obviously agree with you there!

 

[OutsiderSubtype]

I just wanted to put in some encouragement to keep this thread going. I love plausible alt-Space Races.

 

[Shevek23]

Say, can EOR with a combined CM/SM/Lander that takes all astronauts down to the Moon rely on just one engine, the one in the service module?

Then adding on the Lander part in orbit would be a matter of strapping on a "belt" of landing legs, fuel tanks, and other storage needed to get to and operate on Luna. The fuel tanks would feed the Service Module engine for the descent. Then when the manned return elements need to launch from the moon, most of this rack would be abandoned except for some more tanks of fuel that remain strapped and linked to the SM; if the latter's engine had enough thrust and fine control to land safely on the Moon, it surely has plenty of thrust to lift the CM/SM/Ascent fuel racks off it again and go for injection back to Earth.

In fact those tanks that stay with the SM and get boosted back to Earth could contain most of the fuel the SM needed OTL, since it is all needed in these later stages and relatively little of it would be needed to accomplish the rendezvous with the Lunar Lander package. That will still be a tricky operation since whether or not we are using the SM engine for everything or instead using extra engines in the Lander stage, the CM/SM composite must back in so the astronauts are not upside-down throughout their stay on the Moon!

So overall--with a lighter SM due to not having to store much fuel in it, and the elimination of two engines needed in the OTL LEM, and the elimination of the LEM habitation/ascent module, the downside is mostly a need for somewhat more elaborate stuff--optical gear of some kind to facilitate otherwise blind dockings; a more elaborate, perhaps a bit more powerful, and more critical SM main engine; a need to rely on successful dockings where all links and braces and so forth lock into place and work reliably. But mass-wise it seems maybe we have generally reduced it, except for the fact that the CM/SM combo is after all heavier than the LEM ascent module (not quite as massive as the OTL versions though, given a lightened SM) so we have to ship more fuel to the Moon.

I have to admit, trying to visualize how a shrunken CM/SM combo maneuvers to link up with this Lunar descent platform frame is a little tricky, especially if it has to be launched in a collapsed, folded form as seems quite likely.

 

[e of pi]

EOR and direct ascent are fundamentally similar, in that both will require roughly the same mass in low earth orbit to start, the only change is if that mass comes up in one big launch or several slightly less huge ones. LOR is fundamentally different because of the huge mass savings involved with a lander. Remember that to carry a kilogram of earth return fuel to the moon, into orbit, to the surface, and back costs roughly 30 kg in earth orbit, whereas a kg of return fuel just to the moon and into LLO only costs 3 kg in LEO. How big of efect does this have? It's a biggie.

To illustrate, let's compare an EOR mission with a 12 ton (dry) CSM, a 1.9 ton (dry) descent stage, a 0.5 ton ascent stage (essentially drop tanks for the CSM engine), 3 tons of payload to the surface (rovers, instruments, ect), and 0.2 tons of payload back (samples taken from the surface). Compare this to a similar LOR mission. Same CSM, same descent stage dry mass, same surface and return payload, but a 2.2 ton seperate ascent stage.

The burns in the mission are as follows: TLI (3.1 km/s), LOI (900 m/s), lunar descent (2400 m/s), ascent (2200 m/s), and TEI (930 m/s). These burns will be used with both missions. For completeness and personal interest, I ran the numbers with both hypergolic space stages (ISP 270 s) and methalox (ISP 350 s). In both cases the TLI stage was a hydrolox stage with ISP 400 s.

Most of the numbers for stage dry masses and burn delta-vs are from Apollo, the only one not from it is the ascent stage mass for EOR (no need for separate hab, it's just tanks).

With these figures and the mission plans, the fuel needed can be back-calculated, starting with TEI's required fuel, then adding that to the structural mass for the previous burns, and so on back through the mission to derive initial mass needed in LEO.

EOR/Direct Ascent:
Methalox: 207 metric tons
Hypergolic: 349 metric tons

LOR:
Methalox: 88 metric tons
Hypergolic: 115 metric tons (roughly Apollo-level for roughly Apollo numbers, so a good check.

So, yeah. Even if you launch in two launches, your "land the whole stack" mission will actually require a bigger rocket than LOR. This will be true no matter how you futz with surface payload or samples returned. Now EOR/LOR makes some sense (like Constellation, multiple launches to assemble a stack of departure stage, capsule, and lander), but pure EOR with no lander is just a bad plan in my book, almost 3x bad.

 

[e of pi]

Guh. This is what I get for trying to lay out a spreadsheet in a hurry. Forgot to account for the TLI stage dry mass (SIVB: 14.7 tons) and the ISP should be 421s, not 400. I also dropped the surface payload to only 0.5 tons instead of three to better match Apollo's OTL mission. (I always forget just how little payload to surface they had to play with.) All other values were unchanged. Anyway, this makes the kg/kg for TEI fuel brought to LLO via the lunar surface 24, not 29, and 2.97 not 3 for LLO without the detour to the surface (lander lands, CSM stay in orbit), but as you can see, the story is still the same, with the lander option winning in a landslide.

Revised IMLEO data, including launchers required:

No lander:
Methalox: 230 metric tons IMLEO (2 Saturn Vs or 5 Falcon Heavy-equivalent)
Hypergolic: 367 metric tons (3 Saturn Vs, 7 Falcon Heavy-equivalents)

Lander:
Methalox: 101 metric tons (2 Falcon Heavies, 1 Saturn V)
Hypergolic: 124 tons IMLEO (3 Falcon Heavies, 1 Saturn V)

 

[Shevek23]

So it looks like another volley of the Idiot Ball then!

In order to decide that EOR without LOR is the way to go, either someone is making the extraneous decision I was trying to get as an unexpected bonus--the USA is going into the space launch business in a big way, permanently, and so it makes sense to invest in lots and lots of launch capability to be used continually, realizing economic value per unit by means of a high volume of operation.

Or, since this it is very unlikely anyone in the ruling circles of the nation is going to want to buy that pig in a poke, no such commitment can be assumed (there's no way to lock it in, other than someone going ahead and making the investment and then pointing to it later to silence critics--that didn't work so well OTL, now did it?) but NASA deems LOR at such a distance from ground control a risk too uncertain to take. Frankly if they are so worried about that, they ought to worry about EOR too; what good does Mission Control looking over their shoulders do the astronauts? And if MC can help get them out of a jam, presumably an extra couple seconds of lightspeed delay over the radio is not going to pose any fatal impediment. It's not like there are Space Tow Trucks on call in orbit during the Artemis mission!

The point being, if they can solve EOR they should not be afraid of LOR either.

EOR can still be an option along with LOR; the point would be to not have to develop any Saturn Vs. Vice versa if everyone on Earth is terrified of committing to rendezvous of any kind, they have no choice but to have a Direct mission, and that, I guess from your numbers, means a launch rocket with the capabilities of two Saturn Vs! (Or more!) It would be cool of course. But rather than call it a Nova perhaps they should call it a Supernova? Or perhaps the Godot; because of how long everyone would be waiting for it!

So it seems that if we want a plausible Artemis mission with multiple EOR launches, the plan should be for LOR as well.

BTW I was thinking they'd want not just two launches but three or more. Enough launches and the launch rocket can establish a standard capability for general use; enough practice with those things and efficiencies in the assembly and launch process can be developed.

Say, smaller rockets might better lend themselves to horizontal assembly and delivery to a launch pad, where the rocket can be swung up, fueled, and then launched, having been hauled at considerably less risk on a somewhat less extreme vehicle than the OTL Crawler. Being horizontal, it can be shrouded for protection from the elements while being transported; it can be hauled faster because there would be no risk of the thing toppling over; unexpected windy conditions would not threaten it. Faster hauling and quicker assembly address some of the fears Von Braun had OTL about the downside of multiple launches for a given mission, that part of a mission might go up and then unexpected bad weather scrubs the later launches. Clearly you still can't erect and fuel a 500-1000 ton rocket stack in the middle of a Florida lightning storm or a gale! But with quicker logistics (and more numerous pads, more easily available for smaller rockets--they have to be kept at a certain minimum distance from each other in case one rocket blows up on the pad, and the smaller the rocket the less that distance has to be) you can cut the interval between one launch and the next down considerably, making the window for unexpected bad weather narrower, and can make progress hauling the rockets through bad weather in the expectation the storms will clear by the time it is time to erect and fuel the ship for launch.

So--EOR is not a bad approach, if it is combined with LOR, and why not, with all that EOR experience they'll be getting?

It's also good for laying groundwork for assembly of modular space stations and the like.

 

[SAVORYapple]

totally agree with shevek

The problem is not how much you need to lift into orbit to get to the moon and back, the problem is getting a cost effective system in place. I doubt there is a market for 150 metric ton satellites, or 20 metric ton space probes. Therefore, the most logical way to go is with EOR. EOR cannot be combined with LOR, since the cost of developing two launch vehicles will be prohibitive, if we’re talking about going into space in a big way. Super heavy launch systems are all very good for monolithic space stations and LOR, but their uses stop there. A further advantage of EOR is that since the rocket is smaller, it is cheaper to develop. Overall costs might be more expensive, but a rocket that can lift 20 metric tonnes to orbit has far more civilian applications.

 

[e of pi]

The problem is not how much you need to lift into orbit to get to the moon and back, the problem is getting a cost effective system in place. I doubt there is a market for 150 metric ton satellites, or 20 metric ton space probes. Therefore, the most logical way to go is with EOR. EOR cannot be combined with LOR, since the cost of developing two launch vehicles will be prohibitive, if we’re talking about going into space in a big way. Super heavy launch systems are all very good for monolithic space stations and LOR, but their uses stop there. A further advantage of EOR is that since the rocket is smaller, it is cheaper to develop. Overall costs might be more expensive, but a rocket that can lift 20 metric tonnes to orbit has far more civilian applications.

Lookupshootup, you appear to be have a fundamental understanding of this situation. There's not one question here (Do we launch in multiple parts via EOR and then land the whole stack OR do we launch in one shot but with both a lander and an orbiting return craft?), but two independent ones.

1. Do we land our whole stack on the moon, or just a specialized landers
AND
2. Do we launch our stack in one shot or assemble in LEO.

Question 2 was the one NASA was looking at IOTL before they started considering LOR, and the issue it faces is that you are stuck with same IMLEO figure, it's just a question of whether you're launching that daunting number (367 tons!) in one chunk or two chunks or three or whatever. EOR seemed best under that analysis, since it meant building multiple smaller rockets (though still huge, they were talking more than 80 tons!) instead of one giant F-off rocket, resulting in just the benefits in infrastructure and per-launch cost you're talking about. However, when LOR came to their attention (actually from several sources roughly at the same time, contrary to the usual mythology), they realized that with LOR, they were within the window of a 1-launch flight. To minimize complexity, they went with LOR/single-launch.

Examining the decisions made by Project Constellation shows how these two answers are independent of one another, depending on outside criteria not the answers to the other question. For Constellation, they knew a lander architecture would have the benefits I already showed in my earlier posts, so much so that a full-stack landing and ascent wasn't ever really considered to my knowledge. But the scope of the project (multi-week stays, four crew, big rovers, big everything. Apollo on steroids). This needed much more mass, something like 200 mT in LEO, not Apollo's 120ish. Even Mike Griffin and his administrators could see that a 200 mT LV wasn't going to happen, so Constellation went with an EOR assembly via two launches (one Ares 5, one Ares 1), but still used LOR. The DIRECT underground alternative proposed a similar two-launch EOR/LOR, but with 2 roughly-100-ton rockets instead of two different rockets.

A lunar flight actually lends itself to breaking down into several chunks for launch: there's the EDS, the lander, the crew return vehicle/CSM, and there's all the fuel. Constellation did Lander/EDS/fuel on Ares 5, crew and capsule on Ares 1. DIRECT considered several options of parting things out (see the bottom of this page for some diagrams that show it well), including an interesting one where the fueled lander and fueled capsule are sent to lunar orbit separately by two EDSs, which apparently actually had a slight net reduction in weight (and means a cargo lander could be sent with the same mission architecture).

I'd say either a two or three launch architecture, with an IMLEO target of Apollo-equivalent for an Apollo-equivalent mission. This is either a 60 or 40 ton rocket, respectively. If you have orbital fuel transfer capability (like a fuel depot), then you could part it up more since you can launch the hardware separate from the fuel to fill the tanks--the dry mass of Apollo's entire system was only 30ish tons, with the biggest chunk being the 14.7 ton empty mass of the SIVB, so theoretically you could launch your empty SIVB on one 15-ton vehicle, then all the rest of the hardware on another, then 5 or so launches of fuel to fill the tanks. Finally, the crew launch.

This is 7 launches, though, so you'd better be pretty darn good at launching many times in short succession--I'd imagine something like SpaceX's hanger and transporter-erector, but with several side-by-side leading to two or three pads. Prep all the rockets for the mission at once, rolling them out to pads for wet dress rehearsals and such as needed, then you launch one every two days or so. Even if you can pull that off, you're talking about putting fuel transfer on the critical path, which means it'll have to be proved. With today's tech, I'd be more okay with this, but in Apollo's day I don't think it was ever even on the table.

What would be really interesting would be a Falcon 9/Heavy sort of thing, with a 20-ish ton 1-core capability, and 60ish ton capability in a three-core heavy design. You get the high-rate production benefits, but can launch very large single elements when it comes to it. With those options, my plan would be basically:

1. Lander launches on heavy #1
2. EDS launches (fueled) on heavy #2
3. Crew launch with CSM on single-core.

Total IMLEO available is about 140 tons, a bit more than Apollo had OTL. Your cores could be based on one F1 each if that engine hasn't been butterflied.

 

[Shevek23]

totally agree with shevek...

Well, not unless there's a typo here:

EOR cannot be combined with LOR, since the cost of developing two launch vehicles will be prohibitive, if we’re talking about going into space in a big way.

If there's no typo there then I don't quite follow. I thought my point above, which may well be in error but is what I perhaps foolishly meant to say, was that the two rendezvous modes can be combined. From what e of pi tells us, if we insist on using EOR to enable a direct lunar landing and ascent, we're either going to need Saturn V's, in quantity, or legions of smaller vehicles, to assemble a much more massive vehicle for transLunar injection. Since we are piecing together a ship from multiple payloads to orbit anyway, why not make it a LOR modular mission? It's a hassle and a risk to be Legoing together a spaceship with no space station to serve as a platform, but I'd think that if a LOR ship is that much lighter there'd be fewer steps involved than assembling a more massive one, even if much of what we are snapping on is just clusters of fuel tanks.

Let's see, the OTL LOR system was 30 tons for the CM/SM, 14 for the LEM, and the Saturn V third stage (=Saturn 1B 2nd stage, called the S-IVB) was 120 tons on launch. Breaking all this down into 15 ton units, we have 2 plus 1 plus 8 ! Clearly the S-IV is the sticking point here; trying to break its integrated fuel tankage into clusters that fit into 15 ton loads would introduce a fair amount of parasitic mass, not to mention complexities that are extra vulnerable possible points of failure! And who exactly would be methodically piecing together all this?

Actually of course the S-IVB would not mass 120 tons for transLunar injection; much of its propellent would be used up in the course of the third launch stage to put the assembly into parking orbit. Piecing things together from two separate Wiki pages the third stage's share of the orbital burn was 150 seconds while the translunar injection was 350 seconds, so actually a translunar booster that only has that job, having been launched to parking orbit as cargo, would presumably have to have 7/10 the fuel. The OTL stage had a bit over 105 tons of fuel, so say the Artemis injection rocket now needs 71 tons. Call its all-up mass after achieving orbit 90 then.

So, 135 tons all up, plus any extra that comes from splitting OTL integral units into modules. Well, let's make it a round 140 tons, and think in 20 ton units--we can presumably slim down the Service Module by hauling 10 tons of its 18 tons of fuel for the main engine on one of the other launches and then snapping it on during Earth-orbiting vehicle assembly. 5 of those tons might go up with the LM load. That leaves 5 more launches out of 7 to assemble the translunar injection rocket. Perhaps we can offset some of the inefficiency of the modular assembly of numerous tanks by successively blowing them off as they are exhausted.

Or, we can think in 5 30 ton units, or 4 34 tons, or 3 44 tons, or finally two 68 ton units. The bigger the units the fewer the launches and assembly steps and modular inefficiencies, but the bigger the standard launchers have to be.

Even with two launchers, each half as capable as the Saturn V, we are quite awkwardly subdividing the translunar rocket, as its fuel alone would mass more than 68 tons; the 45 tons of the spacecraft won't fill out the other load so we are carrying along fuel in a penny packet to tack on to the main injection rocket. Well, this can be done most naturally as a second stage.

Still a 68--round up to 70!-- ton to parking orbit capability is embarrassingly high for any more routine use. Designing a half-Saturn V might beg the question of why not design a whole one and be done with it! The closest thing to a natural division of the load to my eye is to have three launches, each sending up 50 tons (kicking it up to allow for modular inefficiencies and then hopefully if we have any margin left over, Artemis might be a bit bigger a spaceship in terms of useful payload) to assemble a 150 ton all up Lunar vehicle. That's 50 for the manned modules, and two loads to snap together possibly by remote control from the manned unit standing off to the side to make the translunar rocket. Which the manned units could again most easily join "upside down"!

Or maybe, snap two parallel rockets each with integral fuel tanks and their own engines together side by side, and hook the CM onto the "bottom", with the two injection engines angled out a bit so the blast doesn't hit the manned ship, and let the manned elements hang from the two rockets between the engines? Or something like the Shuttle arrangement, with the manned units snugged onto one side of the parallel joint between the two parallel units riding sidesaddle?

Something to consider here is that since the S-IVB stage used two cryogenic propellents, liquid hydrogen and oxygen, and since LH needs to be kept especially cold, there is a time limit as the propellants, the hydrogen especially, will be boiling off in the sunlight while someone maneuvers them into linking up. Aside from the arguments involving extra parasitic mass associated with breaking the translunar engine into even smaller segments, presumably the more pieces there are the more time it will take to assemble them and hence the more waste of propellent, also the more steps there are the more uncertain the timetable is to complete the assembly process. I think maybe three steps, first launching the manned unit then two parallel (or serial, snapped one on top of the other with the manned unit then backing on to a link up with this stack) injection rockets at the same time or in very quick succession might not be a lot more time consuming than the 2 orbits the standard Apollo mission profile provided for OTL. Perhaps, if we can do multiple launches of many smaller rockets simultaneously, we can break it down into 6 25 ton units--first the CM/shrunk SM, then the LM to be scooped up on the nose as OTL (but braced there, because it is going to be under thrust, and a fuel line for its share of the SM fuel has to be hooked up too), then 4 simultaneous launches of 4 translunar rockets, to be hooked onto the sides of the SM and braced to form a collar of rockets--I think that's the lower limit of reasonable masses to be pieced together. Subdividing them further suggests breaking up the main mission modules themselves and I think that's going too far!

Alternatively instead of LH-LOX engines, the translunar rockets can be fueled with the less efficient but non-cryogenic (but toxic!) fuels used in the Titan rockets and others, these have also the virtues of being denser hence needing less volume, and being storable we buy more time for a more leisurely assembly of the craft. But being less efficient--well, it's getting late and I'm getting confused, but we might well need a fifth and perhaps even sixth injection module, another 25-50 tons. But what's two more launches?

The Wiki article on the Apollo program says as many as 15 launches were contemplated in some EOR schemes; the stub article on EOR says the rockets involved were to be "half" the size of the Saturn V (!) which would imply no more than three launches I'd think.

Even using a less efficient (but much more compact and less fragile, a major consideration for this laborious assembly project) storable fuel mix, I don't think we'd need more than 8 launches. But if we have to launch each one one at a time, that's a lot of orbits for the Artemis crew to be making as they come up--half a day or more. If the plan is to launch up to say six translunar thrust units before the two manned units go up, use ground control to remote-control the six into say two clusters of three each for the manned craft to assemble by attaching first one than the other to the SM sides and then using some kind of actuator link them with braces, after first spearing the LM, stashing it on the nose and bracing it there for these maneuvers--well that might save some orbital time and get the task down to something we'd wish on three men in a working day.

This scheme of mine then relies on having 25-ton to orbit boosters, which strikes me as a reasonable capability for future orbital missions and just right for a lightened Apollo-type CM/SM, suitable for orbital missions. Or a 25 ton spaceplane of some kind, or 25 ton cargo packets.

Having a space station of some kind, even a very small one, to serve as the construction platform for this EOR assembly, strikes me as another good idea. Say, one pieced together from two or three specialized 25 ton modules, manned by an Artemis CM mission, that successively remote-controls in each launched translunar module, slides it down tracks to the assembly of the prior ones forming a ring, then when the ring is all assembled, slides it back to the tip of the station. Meanwhile the second to last launch sends up the LM, which the station crew remote-controls in to stick on the other end of the station. Now the CM/SM stage of the moon mission comes up last, backs up to the station tip with the engine ring, gets secured to a mounting there, the station slides the ring to its attachment position on the SM, releases the assembly, now both station and spacecraft rotate around so the LM is ready to link onto the nose of the CM. The station crew helps make sure the linkages and braces are all deployed and secured correctly, then the station and spaceship maneuver away from each other, and when well separated the translunar rockets do their thing!

----

So then, we'd need: one each space station requiring 2 launches to send up;
One station assembly mission, involving a lightened CSM unit for orbital work to go up to supervise linking the station modules together, man it to check it out, and then perhaps immediately prepare for an Artemis mission to be sent up shortly if all goes well there.

3 launches so far. Maybe 4 if the task of facilitating the moon shot is too much after putting the assembly station together.

Six successive launches of translunar propulsion units, assuming storable fuels; each one is remote-controlled to dock with the station and get assembled. Total 9.

One LEM launch, 10.

The CSM launch, 11.

Once the Lunar mission is on its way the station team can come down, as there is little they can do to help the moon mission if something goes wrong.

Subsequent missions will require 9 each launches, 8 for the moon mission and the first one for the station crew.

Thus we need to launch a total of 225 tons to orbit for each moon shot, as opposed to OTL 135, but that's mainly because of using storable fuel and needing a human crew on the assembly jig station. If we don't need an assembly station, or if it can be remote-controlled, we can eliminate one of those 25 ton units; if we can slim down the translunar rocket assembly by using liquid hydrogen because we can speed up the assembly process, we can eliminate 2 more, and that would bring us down to just 15 more tons than OTL, which seems reasonable to me.

The standard booster, instead of having to raise 135 tons to orbit, only has to lift 25 and so should scale down from 3000 tons to a bit over 555 tons on the pad, call it 560 or even 600 if we suffer from some inefficiencies or are locked in to a lower state of the art by freezing the design some years earlier than Saturn V was OTL. By that same token, these rockets ought to be ready for testing and then use that much earlier.

 

[e of pi]

Shevek, take a step back and look at what you're proposing. Not only are you proposing requiring 9 launches per mission, but you're talking about putting modular assembly (not just component rendezvous and docking but honest-to-god modular assembly and integration), fuel transfer, space stations, and so much more on the critical path just to cut your rockets down to 25 tons each. I will refrain from saying what I really kind of want to, since it'd get me kicked for a week, but instead say this: it's only the moon. Keep it simple, keep it doable.

The way NASA is thinking about this, the commercial viability of their rocket is not enough of a driving concern to let all this complexity seep in just to get it. Sometimes you have to step back from your biases, take a hard look and give up a feature you really like because it's implausible, anachronistic, or impossible to justify from the real requirements. I had to do this in ETS, actually--the rocket I described at the end of the post above was originally going to have a place in the TL, but it ended up not getting picked because when we sat down to create the situation that lead to it...there were better choices within the frame the people in the TL would be thinking of. So we let the consistency of our TL over-ride our plans, and truth and I are still juggling some of the fallout. Here, I think you have to accept that a 9-launch space-station-based mission is not going to get the nod just so that the rocket is commercially viable later.

 

[SAVORYapple]

Examining the decisions made by Project Constellation shows how these two answers are independent of one another, depending on outside criteria not the answers to the other question. For Constellation, they knew a lander architecture would have the benefits I already showed in my earlier posts, so much so that a full-stack landing and ascent wasn't ever really considered to my knowledge. But the scope of the project (multi-week stays, four crew, big rovers, big everything. Apollo on steroids). This needed much more mass, something like 200 mT in LEO, not Apollo's 120ish. Even Mike Griffin and his administrators could see that a 200 mT LV wasn't going to happen, so Constellation went with an EOR assembly via two launches (one Ares 5, one Ares 1), but still used LOR. The DIRECT underground alternative proposed a similar two-launch EOR/LOR, but with 2 roughly-100-ton rockets instead of two different rockets.

I was looking at astronautix.com the other day, and came upon some very interesting CEV proposals. Constellation assumes a four man crew and LARGE and HEAVY surface lander for long stays. A CEV proposal by Raytheon was for a three metric ton three person CEV, launched with a 17-ish metric ton booster stack, and a second launch of a lander with same booster stack….two launches totalling 40 metric tonnes, and you can still get to the moon, with the number of people that crewed Apollo capsules…..I don’t see why we need massive super-rockets, which have HUMUNGUS development costs and need large pad infrastructure. So I don’t think either NASA or the DIRECT proposal (now embodied in the SLS) would be very cost effective.

A lunar flight actually lends itself to breaking down into several chunks for launch: there's the EDS, the lander, the crew return vehicle/CSM, and there's all the fuel. Constellation did Lander/EDS/fuel on Ares 5, crew and capsule on Ares 1. DIRECT considered several options of parting things out (see the bottom of this page for some diagrams that show it well), including an interesting one where the fueled lander and fueled capsule are sent to lunar orbit separately by two EDSs, which apparently actually had a slight net reduction in weight (and means a cargo lander could be sent with the same mission architecture).

The ultimate question is how much modularity you want. A massive LOR stack is not modular. Sending the lander and capsule into lunar orbit is. So is EOR, but on a much more complex scale. Btw, what’s your opinion on Mars For Less???

I'd say either a two or three launch architecture, with an IMLEO target of Apollo-equivalent for an Apollo-equivalent mission. This is either a 60 or 40 ton rocket, respectively. If you have orbital fuel transfer capability (like a fuel depot), then you could part it up more since you can launch the hardware separate from the fuel to fill the tanks--the dry mass of Apollo's entire system was only 30ish tons, with the biggest chunk being the 14.7 ton empty mass of the SIVB, so theoretically you could launch your empty SIVB on one 15-ton vehicle, then all the rest of the hardware on another, then 5 or so launches of fuel to fill the tanks. Finally, the crew launch.

I don’t think there is a market for 40-60 metric tonne satellites either. One fundamental problem of launching one large empty lunar injection booster is that it will take up a lot of volume.

 

[Shevek23]

...I will refrain from saying what I really kind of want to, since it'd get me kicked for a week...

Sorry, I was just thinking with my brain engaged to "mouth," which unfortunately is how I tend to do it.

I agree, 9 launches is too many! But if you pare it down to say 3, you've already built a rocket that is too big for most other uses--might as well finish the job and build a Saturn V. That's what Von Braun decided anyway.

As it is, 3 launches is still kind of nightmarish, however you pare it up.
It's pretty obvious why OTL once the economies of LOR were appreciated, they never looked back at EOR. So I'd think the only reason to favor EOR would be if a 3000 ton pad assembly weight rocket seemed beyond anything they could assure. If they go for smaller boosters I'd think they'd go for much much smaller ones, and that implies lots of launches.

After all, I'm not the guy who came up with needing 15 launches! That was someone on paid NASA staff!

Presumably that was for assembling a direct lunar descent/ascent mode vehicle, which would be double the mass of LOR, hence in the ballpark of double the number of launches I put out as an extreme case, implying that that guy too was thinking of rockets in the range of 25-40 tons to orbit.

...Sometimes you have to step back from your biases, take a hard look and give up a feature you really like because it's implausible, anachronistic, or impossible to justify from the real requirements. ... Here, I think you have to accept that a 9-launch space-station-based mission is not going to get the nod just so that the rocket is commercially viable later.

I'm just trying to explore the limits of the problem here. I think I've learned something from it.

And more from your more nuanced strategies. Which however still involve some rather dangerously fussy measures to work.

 

[e of pi]

I was looking at astronautix.com the other day, and came upon some very interesting CEV proposals. Constellation assumes a four man crew and LARGE and HEAVY surface lander for long stays. A CEV proposal by Raytheon was for a three metric ton three person CEV, launched with a 17-ish metric ton booster stack, and a second launch of a lander with same booster stack….two launches totalling 40 metric tonnes, and you can still get to the moon, with the number of people that crewed Apollo capsules…..I don’t see why we need massive super-rockets, which have HUMUNGUS development costs and need large pad infrastructure. So I don’t think either NASA or the DIRECT proposal (now embodied in the SLS) would be very cost effective.

SLS really is not what Direct was calling for, or at least it won't be once they get started "evolving" away from the 70 ton initial to the monster rocket Congress decided they needed.

The ultimate question is how much modularity you want. A massive LOR stack is not modular. Sending the lander and capsule into lunar orbit is. So is EOR, but on a much more complex scale.

Yeah, I actually like the concept of sending the lander and capsule with separate translunar burns, since it means you can do orbital missions with just the CSM without wasting capacity and ditto for unmanned cargo landers. Not to mention that if you then launch the two translunar stages seperate from the capsule/lander, you get a simple low-assembly 4-launch architecture.

Btw, what’s your opinion on Mars For Less???

Never looked terrifically hard at it. I'll take a look at it when I have time, but it might not be for a while. My gut reaction is "oh god, not another Mars Direct variant," but that's likely not entirely fair.

I don’t think there is a market for 40-60 metric tonne satellites either. One fundamental problem of launching one large empty lunar injection booster is that it will take up a lot of volume.

I don't actually have a huge issue with NASA building a huge commercially inviable rocket if it's what the job requires and it'll get used enough (this is why I'm cool with Saturn V--it was what was needed for the job they wanted to do of a 1-launch LOR mission). NASA isn't a rocket company, it's an exploration and research agency, and sometimes that means specialty hardware.

Besides, a 4-launch mission built around 35-40 ton rockets flown two or three times a year would involve a flight rate of 8-12 rockets, which is enough production and ops to really get into the knee bend of amortizing fixed costs. If you can get the cost down to a couple hundred million per flight (not out of reach if it's well-thought out and you design for ease of operations, which is much easier on a smaller rocket than Saturn V or even the full Shuttle stack), I think there'd be a few interested NASA unmanned missions that would design around it, so there's some additional flights. For commercial, depending on the price point it's possible you could make it economical by dual-manifesting or other such tricks.

Falcon Heavy proves you don't have to be uneconomical just because you're big: at $100 million per flight for 53 tons, it's actually better $/kg than Falcon 9 or any other rocket on the market. You could actually just buy a Falcon Heavy, use it to launch the same 20 ton bird you'd put on Ariane or Proton, and still come out ahead in terms of cost. Sure, you'd be wasting half the potential capability, but if you got launched what you wanted for a lower cost, who really cares?

I agree, 9 launches is too many! But if you pare it down to say 3, you've already built a rocket that is too big for most other uses--might as well finish the job and build a Saturn V. That's what Von Braun decided anyway.

As it is, 3 launches is still kind of nightmarish, however you pare it up.
It's pretty obvious why OTL once the economies of LOR were appreciated, they never looked back at EOR. So I'd think the only reason to favor EOR would be if a 3000 ton pad assembly weight rocket seemed beyond anything they could assure. If they go for smaller boosters I'd think they'd go for much much smaller ones, and that implies lots of launches.

Well, as I said above, a four-launch mission can be done fairly easily if you inject the lander and crew to the moon separately. Kind of LOR/LOR instead of EOR/LOR. The only "assembly" is the docking of complete, fully fueled stages with no requirements for fluids transfer or anything. If your rockets are 40 tons (and as Falcon Heavy is showing, that may not be as uneconomical as you think), then you can put a pretty nice surface payload down for your money.

And more from your more nuanced strategies. Which however still involve some rather dangerously fussy measures to work.

Neither the two-launch EOR or four launch LOR/LOR I've proposed really involves anything more complex than simply docking two spacecraft. If that's your definition of dangerous and fussy, I'd like to hear your thoughts on the ISS assembly process.

 

[neopeius]

Neither the two-launch EOR or four launch LOR/LOR I've proposed really involves anything more complex than simply docking two spacecraft.

You smart folk make me all nervous. It's going to be a two-launch EOR. NASA chose to go with the full C-5 configuration.

 

[e of pi]

You smart folk make me all nervous. It's going to be a two-launch EOR. NASA chose to go with the full C-5 configuration.

Yikes. There's going to be very tight margins with only 260-odd tons available in LEO, it'll be interesting how they make that work. And the mission won't be more capable than OTL Apollo, it'll be the same or less capable in order to meet margins. Why not do EOR/LOR or LOR/LOR? What makes them not trust Lunar rendezvous but they do trust the same process in Earth orbit?

 

[SAVORYapple]

You smart folk make me all nervous. It's going to be a two-launch EOR. NASA chose to go with the full C-5 configuration.

yikes!!!! two launch EOR is gonna be MASSIVE INDIVIDUAL ROCKETS!!!
lets just hope that a moon base makes the C-5 config worth it

 

[e of pi]

yikes!!!! two launch EOR is gonna be MASSIVE INDIVIDUAL ROCKETS!!!
lets just hope that a moon base makes the C-5 config worth it

As I understand it, C-5 is the OTL Saturn V. So they have 2 rockets, each with 127 metric tons to LEO. According to my earlier analysis, this is going to be a very mass-tight mission plan, and that's without taking into account how to actually split it into two launches. Regrettably, I don't know how von Braun was calling for it to be split up. Either way, this is essentially getting all the benefits of OTL Apollo at only 2x the cost and complexity. Nice economics!

 

[SAVORYapple]

As I understand it, C-5 is the OTL Saturn V. So they have 2 rockets, each with 127 metric tons to LEO. According to my earlier analysis, this is going to be a very mass-tight mission plan, and that's without taking into account how to actually split it into two launches. Regrettably, I don't know how von Braun was calling for it to be split up. Either way, this is essentially getting all the benefits of OTL Apollo at only 2x the cost and complexity. Nice economics!

do I see sarcasm??? they'll need double the pad infrastructure as well, as well as probably double the workforce

 

[e of pi]

do I see sarcasm??? they'll need double the pad infrastructure as well, as well as probably double the workforce

Well, they have double the pad infrastructure: the VAB has 4 high bays, there are 2 pads, 3 MLPs, 2 crawlers...they have the space to dual launch Saturns, if not necessarily the 4 per year launch rate they peaked at IOTL. But yeah, the workforce to operate it and the added costs of rockets will be pretty heavy in cost.