sevenperforce

For reusing an Artemis component like Boeing's ascent module, there are four basic possibilities. Capsule Reuse. In this approach, the only thing that is carried over between missions is the actual pressure vessel, backup power, LS, and associated hardware. All propulsion, propellant tankage, propellant, and primary power elements are delivered to Gateway and mated to the capsule at the start of the sortie; this carries the capsule down to the lunar surface (with or without any number of additional stages) and returns it to LOP-G before being jettisoned and discarded. The advantage here is that you have the safety assurance of a brand new propulsion unit for every launch and you do not need to develop any new docking systems or propellant transfer; the disadvantage is that it is the least mass-efficient approach. Capsule and Propulsion Reuse. Here, the capsule includes reusable propulsion and RCS, but the tanks are discarded at the end of every mission. This requires the development of dockable/mateable tanks which can expend their propellant and then be jettisoned. It is advantageous because you do not need to develop propellant transfer and you do not throw away perfectly good engines after every mission so you save mass. It is also the most mass-efficient approach, in a way. The difficulty in developing such tanks is the primary disadvantage here, though end-of-life considerations for the engines are another concern. Descent Module Refilling. In this version, the entire capsule is reused, along with propulsion system and tanks, and the ascent module's tanks are refilled from excess capacity on the descent module. This has the advantage of only requiring a single connection event during mission construction (structural and for prop transfer) but does require full-scale propellant transfer development. It also is less mass-efficient with respect to the descent module, which is then required to carry the weight of empty tanks. If the descent module is multistage (e.g., if there is a transfer stage as well), then using the transfer stage props is ideal because that is jettisoned before descent. Logistics Module Refilling. This seems to be the version that NASA favors, though I don't know why.

Sounds good to me. Two launches plus SLS+Orion. Might even be able to land the side boosters, particularly if New Glenn can throw it into an eccentric orbit. The SLS fairing is 8.4 meters to New Glenn's 7 meters...hopefully the Boeing lander isn't too big. Of course once New Glenn is flying, its hydrolox upper stage would also have very good performance as a naked tug. Making the Boeing lander's ascent stage stage reusable would also be very mass-efficient.

I confess I am very fond of Boeing's lander design. Putting the airlock on the descent stage and the engines on the sides allows it to comanifest flatpacked cargo to the lunar surface or act as an independent cargo lander that doubles as a surface asset (showers, anyone?). It also permits the ascent stage to be limited just to what is needed for return. Dragon 1 and 2 both use the trunk to mate to the tank structure that the PAF bolts to. With Starlink I think they may have braced the sats against the payload fairing.

Given the need for engines and COPVs and egress, I don't think it helps all that much. Yeah, reusing the ascent module doesn't make much sense. Reusing the capsule, on the other hand, does. Strip down a Dragon 2, put a docking port underneath for mating and for an access tunnel to an airlock in the landing module, and send it to LOP-G. Then you merely launch a new landing module with each mission.

We don't even have to guess at the burn, because Constellation. The J2-X on the EDS would have downthrottled to 1,048 kN for the TLI burn, pushing a stack that at burnout would have massed roughly 86 tonnes, for 1.24 gees through the same NDS that Dragon 2 has. Also keep in mind that the NDS already has pyros for contingency undocking, which mean you already have a "decoupler" option when constructing a lunar descent/ascent vehicle.

You run into some problems with the structural and aerodynamic limits of the stage during ascent, here. The PAF simply cannot handle very heavy, tall payloads, Starlink notwithstanding. The fairing support is independent of the payload support. And that is a lot of hypergols. Also, where would the engines go?

All of those involve crew onboard the vessel being docked. Progress docks to the crewed ISS, Apollo is obviously crewed. Hence why I said docking EU would be trickier than anything we've done before. You're intentionally conflating the two examples I gave. Progress, Soyuz, Salyut, and Dragon 2 are all examples of autonomous, crew-agnostic dockings, which you seem to think is impossibly challenging. Apollo was an example of building a rocket assembly in space, which you said has never been attempted. The very earliest autonomous dockings (Kosmos 186 + Kosmos 188, Kosmos 1443/Progress 23 + Salyut 7, Progress M1-5 + Mir) were wholly uncrewed, because that is easier and less risky than doing it with people on board. Well, the last time we built a rocket assembly in space, burnout acceleration was 0.3 gees. Given that the Common Berthing Mechanism has a diameter 222% of the Apollo Docking System, I'm sure we could find a way to beef it up to handle 0.74 gees (minimum downthrottle of MVac to 364 kN, total stack burnout mass of 49 tonnes).

If it doesn't need to take re-entry heat, then it makes more sense to keep a composite fairing.

The cold gas thrusters have sufficient placement authority for roll control and ullage, so that will be enough for holding position. I would imagine that the docking sequence could potentially involve a slight "spin-up" once the vehicles were perfectly aligned, before final close. The upper stage would need more bottled nitrogen, of course, as well as more helium. Hence the same Frankenstage design as the FH demo. Merlin Vac can downthrottle to 39% of max thrust. Dry mass of S2 is 4.5 tonnes...let's say 5 tonnes after Frankenstage mods. Reduce payload to 44 tonnes (the same TLI throw as Saturn V) to ensure sufficient residuals and account for increased dry mass and possible boil-off during rendezvous. At minimum throttle that's just 7.4 m/s2. Not that it would be a problem anyway...we have routinely seen S2 burnout with much smaller payloads. No idea. SpaceX advertises 3.5 tonnes direct to Pluto, so I assume it can get 3.5 tonnes direct to Pluto. You need 8.2 km/s, with the Oberth effect to get a Hohmann transfer to Pluto. The only way to get that much dV out of the upper stage is if it has nearly 75% of its propellant remaining at LEO SECO-1; the only way to get to LEO while burning only 25% of S2 propellant is to stage at around 1,031 m/s short of LEO. How SpaceX achieves that is up to them.

50 tonnes to LEO is a good starting point for orbital assembly, though. If we had multiple vehicles with flight heritage throwing 40-50 tonnes to LEO whenever we needed it, we could have a moon base already.

Your upper stage has double the dry mass it should. The Merlin 1D masses 470 kg and the extended reinforced carbon-carbon nozzle is heavy, but not greater than the mass of the engine itself. I would estimate the MVac at 550 kg tops. Tankage is under 4 tonnes. A good dry mass estimate for the upper stage is 4.5 tonnes...maybe 4.7 if you go with a longer-duration "Frankenstage" for BLEO ops. dV from LEO to LOP-G is actually 3.63 km/s. But yeah, your dry mass is messing you up.

Progress pushes ISS all the time. Cygnus did from the unity module, as well. I mean, it was a fair point: you can't just slap an IDA onto the end of Europa Clipper, mate to a half-depleted Falcon Heavy upper stage, and fire up the MVac. It's a little more complicated and we definitely need to develop it. Especially if Artemis.

Almost everything we’ve docked by has involved crew in some fashion onboard the vehicle. Soyuz to the ISS, shuttle to Mir, Apollo to Skylab, Gemini to Agena, CSM to LM. Almost everything? It's 100% the opposite. Unmanned Progress spacecraft autonomously dock themselves to the ISS several times every year without incident. There have been far more unmanned dockings than manned dockings, and Soyuz dockings are automated. Apollo would beg to differ, unless docking one vehicle to another and then using the first vehicle's engine to push the connected stack doesn't count as building a rocket assembly in space. Also, this is exactly what the Artemis plan calls for: building rocket assemblies in space. We need to develop it for Artemis anyway. It would have to be newer and stronger, yes. Again, these are things we need to develop. If Apollo could devise such a guidance system with less software than a smartwatch, I'm sure we can today. Especially since we have to, because Artemis. Until 2030 (at the soonest SpaceX can compete in lunar space) NASA’s SLS is the only vehicle ready for BEO flight. SLS is not ready for any flight, let alone sending any competitive payload BEO, and I seem to recall Falcon Heavy sending something BEO two years ago. If distributed launch with eyeballs-out burns was good enough for Constellation, it should be a no-brainer for sending Orion and crew to the moon now.

Wow.

So the component attaching the pilot parachute to the main failed. Someone's getting fired.

Paint the kerosene tank with black stripes and barrel-roll the stage during transit. Kerosene will be kept warm by solar heating. You'll lose about 0.2% of your LOX per day to boil-off, according to NASA's GR&As for lunar mission planning, but that's not terrible (especially since the majority of the burn happens within a few hours of launch). The stage would need a few of the same modifications that the first FH launch had (extra nitrogen tanks, extra helium tanks) but again, pretty minimal. Exactly. Cheaper than a whole freaking SLS for sure.

It's a tricky proposition. As @tater always points out, an expendable Falcon Heavy cannot actually inject a 64-tonne monolith into LEO; its PAF alone wouldn't be able to handle it, and the upper stage has structural limits below that. What Falcon Heavy can do is put 64 tonnes of usable mass into LEO, as a sum of the actual payload and the propellant residuals, intended for a BLEO restart. The question for distributed launch is the amount of residuals for any upper stage when launched without payload, but no rocket companies quote "naked" residuals. However, since virtually all rockets quote GTO payload, that's the apples-to-apples comparison I chose. Using the quoted GTO payload (26.7 tonnes at 2.27 km/s) and reverse-calculating gave me 44.8 tonnes of LEO residuals when launched without payload. However, SpaceX also quotes Falcon Heavy's payload to Mars (16.8 tonnes at 3.6 km/s) and payload to Pluto (3.5 tonnes at 8.2 km/s). Getting 26.7 tonnes from LEO to GTO would require 29.5 tonnes of propellant residuals, which makes enough sense. Getting 16.8 tonnes from LEO to Mars Transfer Injection would require 39.8 tonnes of residuals. Still tracking. However, getting 3.5 tonnes from LEO to a direct Pluto transfer, at the same 348 s specific impulse as everything else, would require a whopping 80.4 tonnes of residuals!! This suggests that when flying Falcon Heavy fully-expendable with a 3.5 tonne payload, core staging occurs just 1,031 m/s shy of LEO. If you imagine launching FHe with a max-GTO payload but dropping it at staging, then you end up with 44.8 tonnes of residuals at LEO, which is what I used for the tables I've built. If you imagine launching FHe with a max-MTI payload but dropping it at staging, then you end up with 48.9 tonnes of residuals at LEO, which is slightly better. However...and here's the ridiculous part...if you launched a 3.5 tonne Pluto payload on an expendable Falcon Heavy and dropped it at staging, you would hit LEO with 81.3 tonnes of residuals. If my math is right...holy crap. That's enough to throw up to 47.8 tonnes of distributed-launch payload to TLI. It's enough to deliver 38.3 tonnes of distributed-launch payload direct to LOP-G. It's enough to deliver a 26.9 tonne two-stage lunar lander to LOP-G, load it up with astronauts, and then drop it off in LLO. See above...evidently not a problem. And of course you can make it even better by sending the reusable ascent cabin/module to LOP-G ahead of time.

Docking in LEO is not harder than docking in cislunar space. What is the difference between staging distributed launch in LEO vs staging distributed launch at LOP-G? We need to develop it regardless. Europa Clipper is going to be spending years in space anyway. What's the harm in throwing it to LEO a few months early?

They really mean "which we pretend requires the use of the SLS" but otherwise.... Distributed launch for Europa Clipper would be so fffffing easy. I haven't done the math in a while but IIRC you could toss it into LEO on a reusable Falcon 9 and send it direct to Jupiter, faster than SLS could, with a DIVH or Atlas V companion launch. Under $200M in launch costs.

In a converging-diverging nozzle, the fluid velocity must increase with increasing pressure (Bernoulli), but the fluid density must decrease at the throat with increasing velocity (Venturi). Initially the mass flow goes up with pressure, but eventually Venturi overcomes Bernoulli and the mass flow decreases because the density at the throat drops precipitously. When fluid velocity reaches Mach 1 at the throat, you have a steady state choked flow where Venturi and Bernoulli are dancing together in sync.

Yes, mass flow will be lower when you have choked flow at the throat.

Indeed. I just don't know enough about "Centaur Heavy" to make a good estimation of BLEO performance. Very broadly, the 2016 version of the BFR was quoted at 550 tonnes expendable or 300 tonnes reusable. Reusability included upper stage landing prop reserves, so reusing the lower stage and dumping the upper stage would probably have given 360 tonnes or so reusable. So expending a methalox core could be estimated as a 50% boost to capacity. Falcon 9 is quoted at 8300 kg to GTO expendable or 5500 kg to GTO reusable, which is also about a 50% boost. However, New Glenn reserves less propellant since it doesn't do a boostback burn or entry burn. Then again, sending 5500 kg to GTO would not involve a boostback burn either. If we give New Glenn a 40% payload boost for expending the first stage then it would probably be able to put 63 tonnes in LEO. Not quite Falcon Heavy, but close...and New Glenn's hydrolox upper stage would outperform Falcon Heavy's kerolox for TLI purposes. Orion's injected lunar mass is 26.5 tonnes, so if it was launched to LEO on an expendable New Glenn we would expect it to have around 37 tonnes of hydrolox residuals. If it can push over 435 s isp and we assume upper stage dry mass of 5 tonnes then it can absolutely do TLI. That would be nice. Wikipedia says "ACES is now expected to have the same tanks as Centaur V, but with the possible addition of two more RL-10s and IVF" but it is uncited so who knows who put it there. Ah, I see it now. From one of ULA's graphics, the difference between Centaur V and Centaur Heavy:

It's ridiculously important and I don't know why we haven't spent all the money wasted on SLS developing autonomous spacecraft assembly, propellant transfer, and ZBO tankage. Vulcan 6 is a 6-solid-booster methalox core with the two-engine Centaur V stage. Vulcan Heavy is a Vulcan 6 with a "Centaur Heavy" stage in place of Centaur V. I made decent estimates of Centaur V's dry mass and performance. If Centaur Heavy is ACES, and ACES is Centaur V with IVF and two more RL-10s added, then it's straightforward enough, I suppose.

I give a naked DCSS launched on DIVH the capacity to send 12.6 tonnes to TLI or 9.8 tonnes to LOP-G, whereas the numbers are 14.8 and 11.5 for Centaur V on Vulcan 6. I did not do the math for Vulcan 6 Heavy because not enough is known about Centaur Heavy.

By my numbers, a naked Vulcan 6 upper stage can throw just shy of 15 tonnes to TLI or can deliver 11.5 tonnes to LOP-G. Roughly comparable to the performance of Falcon Heavy with only the central core expended. Vulcan 2 beats Atlas V 551 for LEO but not for BLEO because the Atlas-Centaur has lower dry mass. This is all when you launch payload into LEO and then do a second launch with the naked upper stage to perform TLI or other BLEO work. I didn't bother to calculate Vulcan's single-launch throw to TLI, but given that Vulcan 6 can deliver 13.3 tonnes to GTO (2.27 km/s) or 6.0 tonnes to GEO (4.04 km/s), I'm guesstimating its throw at 9-10 tonnes to TLI (3.2 km/s). If your want your cryogenic upper stage to deliver something all the way to LOP-G, its capacity is actually better than capacity to GEO, because Oberth.