sevenperforce

That's gotta be bare-bones. The Apollo ascent module was 4.7 tonnes wet and had less dV. The LockMart lunar lander concept uses an Orion pressure vessel but I wonder whether that's ideal. Certainly simplifies some things. The real issue is that you just don't know exactly what the purpose of lunar exploration is, which tends to complicate discussion of what your lander should look like.

What's a notional wet mass for an ascent stage (reusable or otherwise) that goes from the lunar surface to the Gateway? EDIT: It's also annoying that we literally have zero steps toward a manned capsule for lunar surface ops.

Even for larger vehicles, the question is worth asking: What is the mass of wheels vs reserved propellants for landing? If it flew the same profile, then instead of plummeting straight down, it glided (steeply) as a lifting body... Nothing chnages except they trade landing props and a burn for a runway. It's not just the wheels; you do need a certain minimum L/D ratio to make horizontal landing possible, setting aside the mass of the landing gear. Starship would likely need double the "wing" area to make it work, plus more control surfaces. My concept would use hot-gas thrusters for auxiliary attitude control on entry and descent as well as for the terminal landing burn.

I want to see a man-rated lander "truck" with dual engines and a central cylindrical tank that can be vented and used as a wet workshop/airlock. Could be used to deliver either crew or cargo.

Per hull mass, yes. But not internal volume per cross-section. Enterprise is clearly not mass-limited. The Enterprise's warp core is in the lower module, while the warp nacelles are up high. Presumably the nacelles need some minimum separation for efficiency or warp field stability. The conventional impulse engines are in the center behind the CoM. We don't know anything about making artificial gravity, so this could be any sort of solution. If your artificial gravity has a gradient then you want to make your ship as flat as possible. Who knows how warp works? Outer mold line. Basically the "shape" of the ship. What if warp fields turn out to send conservation of angular momentum haywire, causing roll orientation to go wack? The ship would tend to roll in the warp field and rip itself apart due to the extreme gradient. So you'd want an oblate toroidal field to stabilize.

It's less than half the length of a F9 first stage, so yep, that's the plan. The square-cube application assumes that wet mass is dependent on volume, which scales with the cube of height, while dry mass is dependent on surface area, which scales with the square of height. Starship has a structural fraction of 3.4% of its GLOW (3520 tonnes). A 201-tonne vehicle has 6.7% the mass of Starship, so we assume 6.7% of the volume, 14.8% of the surface area, and 38.5% of the length. If this vehicle had the same skin thickness as Starship, it would have a dry mass of 17.8 tonnes, but its structural loading is only 6.7% as high, so we should reduce skin thickness. Doing so by 50% gives us a dry mass of 8.9 tonnes. So yeah -- build it out of 301 stainless. Only do two fins (in the front), which saves some weight, and have them double as forward landing legs. Use methagox thrusters for auxiliary attitude control during entry as well as dual-thrust-axis landing.

Then why make the saucer instead of a cylinder? Because you may well want a shape which maximizes internal volume while minimizing cross-section. You can make a cylinder as long as you want, of course, but you quickly end up with fineness and bending moment problems. If you have a cylinder, the cross-section grows with the square of increases in diameter; if you have a constant-ratio saucer, the cross-section grows linearly with increases in diameter. There could be other considerations...for example, if you have antigravity (as in Star Trek), it may only support a single plane within your OML, which would be more convenient if you have a few broad levels than many small levels. Additionally, there may be a polarization component of your warp drive, where "pushing" something in and out of warp is much harder if the object is very long than if it is wide but not high.

Ever since Elon's interview with Everyday Astronaut, I've been kicking around the utility of employing some kind of altitude-compensated Raptor...not for Starship, per se, but for something smaller. Elon claims Starship will cost on the order of $2-3M per launch, which I think is rather extraordinarily low, but even at $20M (twice his highest estimate) it's still game-changing. Yet presumably there will be many instances where you simply do not need 100 tonne or fifty people delivered to, say, the ISS. If Falcon 1 was the original Tesla Roadster, Falcon 9 is the Model S, and Falcon Heavy is the Model X, then Starship is clearly the Tesla semi-truck. But no one wants to drive to dinner in a tractor-trailer. Like Tesla, SpaceX needs a newer, better Roadster. I decided to play with a Raptor-based SSTO using GNOM-style air-augmentation and altitude compensation, with the dedicated goal of developing a vehicle with the same capacity as Falcon 9+Dragon 2. With a fixed vacuum-sized nozzle integral with an inlet duct, the Raptor would have air-augmented altitude compensation as well as full expansion in vacuum, without the pain and suffering of an aerospike. I went with a single-engine design. I assumed: T/W ratio at liftoff: 1.2 (low because air augmentation will boost T/W ratio rapidly once moving) Average air augmentation after liftoff (Mach 0 to Mach 1.75): 18% Average air augmentation during boost (Mach 1.75 to Mach 5.5): 100% Average air augmentation during supercruise (Mach 5.5 to Mach 12): 36% Structural load multiplier (essentially, decreased skin thickness): 50% Using the above assumptions and the application of the square-cube law, I came up with a single-Raptor vehicle massing 201 tonnes at launch, with a structural mass of 8.9 tonnes and 181 tonnes of methalox, and 10.8 tonnes of payload margin. 9,939 m/s total dV. Definitely enough to get to LEO and back with a crew of six.

It depends on how warp drive works. If there is some sort of drag-limited warp drive effect, you may well want a shape which maximizes internal volume while minimizing cross-section and sharp angles, in which a saucer with big-ass engines attached is optimal.

Gas-gas spark ignition is very reliable. Same principle as a gas stove.

Elon says has said the methane-gox hot-gas thrusters for Starship will be ten-tonne-class.

I am fond of laser sintering bricks and building around structures with robots, myself.

Odd. The typically-reliable figures on wiki give 1.31 km/s for both LOI and TEI. **digging** Looks like NASA's GR&As for cislunar ops assume going to LLO will involve a burn into HLO first, for an HLO LOI of 508 m/s and another 520 m/s to get down to a 100x100km orbit. So that's where the ~1.3 km/s figure comes from.

Huge problems with impulse density, unfortunately. If you really want to use methane+GOX for some reason, you can always use an electric pump and electric resistance heaters to prime pressure-feeding bottles from a liquid reservoir, but then you have a limited burn duration. Might as well just go ZBO hydrolox tanks with the BE-5 instead. Transit boiloff is not prohibitive. 0.35% per day with hydrolox by NASA's current GR&A's. If you have hot-gas thrusters then you can just route any boiloff into the RCS bottles for a pseudo-ZBO approach. What you want to avoid is having hydrolox in loiter; that's where the third-of-a-percent losses start to accrue. For a disposable lunar landing architecture I have always preferred using the TLI stage for LOI and deorbit, then as a crasher stage to scrub off everything but the last 300 m/s of the landing burn. You can then do a single-stage lander. Although DLOR has its attractions.... Closer to 1300 m/s, so 2600 m/s dV. And that's only braking itself into orbit, not braking in a lander. Orion is too heavy, its service module too light, and SLS is too puny.

Regardless of how you get headed to TLI, isn't it more efficient to dump your expended LOI tankage on the moon than to drag it back with you to EOI? Hmm, I guess it depends. EOI and LOI both cost 1.31 km/s while lunar landing costs 1.87 km/s. So perhaps keeping that expended tankage on your command module is the better choice, unless you have a way to drop it somehow. Using the TLI stage for your LOI burn (and for your lander deorbit) probably beats both.

Word on the street over at SNF is that the strip weld reinforcement band on Mk1 is where the lower bulkhead and engine mount transfers force to the outer mold line.

Orion was sized for the EOR-LOR lunar landing architecture of Constellation, where the Altair lander's descent module would perform the lunar capture burn and leave Orion's SM topped up in low lunar orbit. Strictly speaking, it's a more efficient architecture than Apollo. The reason the Apollo CSM performed the capture burn in the 60s and 70s was that the AJ10-derived SPS was grossly oversized (it had originally been planned for lunar direct ascent) and so they threw more propellant onto the SM.

Not without Block 1B. It's a shame the SM is so lame.

SpaceX can just send up a cargo Starship and say "Our proposal is to meet Orion in LEO, dock it inside our cargo bay, land on the Moon, and come back."

I don't see why not. Methane-oxygen has a great combustion cycle.

I suppose they could launch something to orbit now. If they fired the capsule escape motor at MECO that's another 625 m/s, and then New Shepard's capsule can carry up to 410 kg, which is plenty to get into a cubesat into orbit if they used their BE-2 engine.

Ah, good point. That could prove problematic. If you want to burn radial, great, but if not then you're going to have a great deal of trouble "collecting" all that hydrogen. Of course, relative to 8.9% c, 700 km/s might as well be standing still.

Elevators are not needed, as humans are extremely strong for their weight on both the moon and Mars. Even if someone was ill they could be readily carried up or down a ladder by a fellow crew member with just one arm. A powdered-regolith laser sintering printer for building structural blocks would be one of the very first machines to have up and running past life support considerations. The need for a flat floor does tend to complicate things, as it cuts the usable volume considerably. Plus, laying Starship on its side would mean all the elements in the upper cabin which are ordinarily intended for a single axial orientation would be 90 degrees off. Not an issue for a cargo Starship, of course, but it's still a consideration. I consider it more likely that an end-of-life Starship would be converted into a series of habs by slicing it into sections horizontally. You have three different bulkheads which will provide nice domed roofs, after all. Then the nose-cone crew section can simply be "planted" into the ground as one of several hab structures.

Notionally, it would do 99% of the burn on the three Vacuum Raptors at close to minimum throttle, then ignite one of the core SL engines for the actual landing. All three core engines can gimbal through the CoM and hot-gas thrusters would be used to correct attitude. I calculated over in the horizontal-landing thread that you need nine and a half hot-gas thrusters to land, factoring in cosine losses. Completely doable.

At 30 km/s relative to Sol (at least near Terra), your magnetic ramscoop need only be 68,326,751 km in diameter...a little more than the diameter of Uranus.