sevenperforce

Not a lot different. Humans can handle low air pressures (to a point) without difficulty, as long as the partial pressure of oxygen remains the same. John Glenn's capsule flew a pure-oxygen mix at 5.5 psi, which is just barely more than a third of STP. I don't think there was any great change in what he could hear, or how his voice sounded. There might be a change due to air density, though, since the chemical mixture is very different. https://www.ncbi.nlm.nih.gov/pubmed/8128881

Someone who has no idea what he is talking about: Can it be done? It's not clear how the second stage, which reaches a much higher altitude than the first stage, would be returned to Earth. It may use some combination of the Merlin engine and parachutes, but so far the company has not specified how it will attempt such a recovery. Bahahaha. Silly people. No, it's not the altitude that's the problem; it's the velocity. And no, the Merlin engine would not be used, other than potentially for the deorbit burn. And yes, SpaceX already has an animation of how they would attempt to recover it.

Sarcasm level: missed.

I'm torn on whether I prefer "near-SSTO plus a small expendable second stage" or "near-SSTO plus a small recoverable parallel booster". Perhaps the former for cargo, the latter for crew. Unrelated... Thinking back about possible mini-ITS concepts for a single fully-reusable second stage integrating Crew Dragon with the Falcon 9 s2. Not that SpaceX would pursue it, of course; just as an idea. It occurred to me: by the time the second stage fires, it's out of the atmosphere...so engine thrust doesn't actually have to be along the long axis. What if you had a cluster of aerospike engines oriented perpendicular to the long axis of the vehicle? The burn to orbit would take place with the vehicle oriented normal to prograde. This way, the same engines (or a subset of them) can be used for propulsive landing.

There are numerous instances in which you may need an escape system, each with their own complications. 0/0 pad abort. You've got to go up very fast and get enough horizontal velocity that you come down, you don't land in fireball. You also need a way to come down without going splat. MaxQ abort. You've got to get clear of a fireball in air too thin to breathe, while you're already going supersonic, avoid tumbling uncontrollably beyond the aerodynamic limits of your vehicle, and then come down without going splat. On-orbit Lifeboat. For instances where your crew cabin is incorporated into your launch vehicle, you need a way to leave your launch vehicle behind on-orbit if its primary TPS or its structure is damaged during launch or on orbit (e.g., Columbia). Then you need to re-enter without burning up, and come down without going splat. Landing snafu. If your vehicle is coming down too hard/fast, then see 0/0 pad abort as above. Ejection seats work for the 0/0 pad abort and the landing snafu, but not for the MaxQ abort or the On-orbit Lifeboat. An ejectable capsule/cabin with its own pusher engines will work for a MaxQ abort, but it needs to be chuted down, and it will only work as an on-orbit lifeboat if it has its own TPS. And it may not work for a 0/0 pad abort or a landing snafu, depending on its orientation to the rest of the vehicle.

Spacex has already tested an egress system? By egress, I mean physically entering and exiting the vehicle. Ejection seats work very well for aircraft. They would work well enough for landing a spacecraft on Earth, but less so during launch, and not at all on orbit.

Hah! Yes, this is true. But things were a bit different for the LM. For one thing, the height-to-width of the LM was less than 1, so tip-over during egress was never a possibility. Plus, it wasn't the terminal stage; during landing, the ascent stage could have aborted to the command module at any time. When you're landing on Earth, you kind of want to be able to reliably get out of your vehicle if there was an emergency. Being at the top of a tower is one thing; being at the top of an unstable tower still partly filled with a bucket o' dV is another thing entirely. In the Dragon 2, if the LAS engines fail you can pop the chutes and let the legs crush...it'll be a bumpy landing but you'll be okay. Anyway I'm not concerned so much about the engine integrity. Using one set of engines for launch, launch abort, and propulsive landing (e.g. ITS) is fine and dandy, but even 100% reliable engines can't save you from an RUD if you're landing on your tail and you start to tilt.

Ooooh, I see. We were comparing apples and oranges. I was talking about a tailsitter landing being bad for crew with something like the ITS, where you don't have any separate escape system at all. If a mini-ITS pulled a CRS-5, CRS-6, or Jason-3, everyone dies. Even a Thiacom-8 is a very, very bad day. It's safe enough to use LAS engines as landing engines for a capsule like the Dragon 2, because the capsule will not RUD even if a leg breaks or it tips over. A mini-ITS would not have any escape system, which is why I said I didn't like the idea of landing it on its tail.

I think it's generally agreed that suicide burns/hoverslams are more precise and exacting than aborts, even if the g-forces of the latter are more severe. In an abort, you don't really care how much thrust you have, as long as it is enough to get you out of harm's way. In a hoverslam, you've got to have precisely the right amount of thrust at precisely the right amount of time so that your legs touch down exactly when your engines shut down. Any residual lateral velocity and you break a leg or tip over entirely. Tipping over in a capsule isn't a big problem, because even if you did happen to lose a leg partially (Thiacom-8) or completely (Jason-3), you'll still be mostly upright. In the unlikely event that you come in at an angle like CRS-5 or tip over completely like CRS-6 (which almost certainly won't happen since the height-to-width ratio is so low), you end up rolling, which won't be much fun for the occupants but at least the capsule itself will probably remain intact. But when you're thinking about landing a mini-ITS on its tail, with the crew perched on top, it's a lot more risky. A Thiacom-8 problem, and you're teetering on the top of a tower that's rocking back and forth while you hang on for dear life. Jason-3? You're dead. CRS-6? You're dead. CRS-5? You're so very very dead. And even in an ideal case, you're still perched on top of a tower with no way to get out on your own. I wouldnt say there's no way to solve that problem, but you are right in that an alternate heat source is something that needs to be engineered around. Raptor and Merlin both have high specific impulse for their propellant type and high TWR because their chamber pressures are very high. The ITS methalox thrusters are technically an autogenous pressure-fed cycle, which works fine in the ITS because it is using the Raptor as its heat exchanger, but wouldn't work otherwise. You could go to helium pressure-fed cycle, like the Kestrel, but that really hurts engine efficiency and TWR while driving up the dry mass of the stage. A hot-gas expander cycle could be used, but that's also inefficient and doesn't scale up very well.

The methalox thrusters for the ITS will be hot-gas pressure-fed from the autogenous pressurization tanks configured with the Raptors. The Raptor engine cools itself by cycling subcooled CH4 and LOX through heat exchangers. The fluid is boiled by the heat exchanger and used to pressurize dedicated pressure vessels inside the tanks (with the hot oxygen gas being stored inside the LOX tank and the hot methane being stored inside the subcooled CH4 tank). These pressure vessels are each then vented to their enclosing tank to maintain ullage pressure to the turbopumps. The same pressure vessels are tapped to run the methalox thrusters for the RCS system on the ITS spaceship (and, for that matter, the reaction thrusters on the ITS booster). Without the Raptor and its heat exchanger, then, there's no way to pressurize the tanks for the methalox thrusters.

I'm not sure how solving egress solves tip-over. If you're perched at the top of a top-heavy, unstable, smoking tower, things can get very messy very fast. If a landing leg fails, you're dead no matter what egress plans you may have had. Slowing down quickly means higher peak heating which means more thermal stress. Unless you're approaching the problem from a different angle than I'm imagining. I'm not sure putting a Dragon heat shield under the payload adapter will work all that well. Keeping center of pressure behind the center of mass would be virtually impossible. The key to successfully re-entering a whole stage from orbital velocities is to use as much of the cross-section as possible. Stages are fluffy. Strengthen it biaxially and make it enter normal to prograde.

To be fair, the Dragon V2 and the New Shepard capsule both land in a reclined position halfway between "back" and "butt". In any case, I'm not so terribly concerned about the seat orientation; I'm more concerned about landing stability and what comes after. If you're landing anything larger than a capsule, going tail-first means your crew is stuck at the top of a smoldering, potentially unstable tower. There's a lot that can go wrong.

Okay, so let's say they have an extra mass budget to play with. How can it be done? An inflatable heat shield under the payload adapter? Superdracos and fold-down legs on the sides? TPS paint on one side? This is exactly how the planned crew Dragon will land... eventually... Well, the planned Crew Dragon isn't landing on its tail so much. It's a capsule, so it's more like it's landing on its large, well-padded rump. What reasons are those? Two primary reasons: tip-over and egress.

The burn is at the perigee of the intended geostationary transfer orbit. The Falcon 9 first stage places the second stage in a lofted trajectory, since the MVac has a TWR lower than 1 at staging. So the second stage burns horizontally even while it continues to its initial apogee and starts to drop. The first burn of the second stage is carefully throttled so that the stage reaches a roughly circular parking orbit. The GTO is a Hohmann elliptic transfer between that LEO parking orbit and a geostationary earth orbit; its perigee touches the LEO parking orbit and its apogee touches the GEO orbit. Since the maneuvering bus on the comsat has a TWR far, far lower than 1, it uses the entire duration of the Hohmann transfer to slowly raise its perigee from LEO altitude up to a GEO circularization.

Slight change of topic: Musk is yet again making noises about second-stage reuse. Any idea how they'd want to try and manage this, or what modifications S2 would need in order to permit reuse? What kind of payload penalty would be involved? As a related off-shoot from the now-closed ITS-bashing thread: While SpaceX is probably not going to do this any time soon, I'd love to see a cogent design for a fully-reusable combined second stage+Crew Dragon which lands propulsively but doesn't compromise safety. It's a tricky problem. You can't just do a scaled-down ITS because the MVac cannot be used in the atmosphere...in fact, it would be ripped to shreds if it is exposed to the atmosphere at all. And landing on one's tail isn't the best way to land crew, either, for multiple reasons.

That would be part of payload design, not stage redesign. The ISS modules launched by the Shuttle were designed for the Shuttle. An ISS clone launched by the Falcon family would be designed for the Falcon family. But that's not really relevant. There are two ways to do it. The first is for each module to launch with a maneuvering bus, which is something virtually all comsats launched by SpaceX already have. The second way, which is simpler but a little more time-consuming, is for the second stage to place the module in the correct orbit but not decouple it, and simply loiter until Crew Dragon arrives. The second stage would use its cold gas thrusters to hold position while the Crew Dragon docked with the module, and then it would decouple. The Crew Dragon would then be used to maneuver the module to its final destination. Of note: you could do a combination of the two approaches. For example, you could launch the Canadarm module and the first hab module using the Crew Dragon as a bus, then launch the next couple of modules with a robotic maneuvering bus capable of decoupling independently. Then one maneuvering bus could remain attached to the growing station at all times while the other did the task of retrieving modules from unmanned Falcon launches. The maneuvering bus would use SEP, which would save the Crew Dragons from having to expend all their hypergols shuttling modules around. It would be a big job, but it could be done. Definitely more cheaply than the ISS via Shuttle program.

The payload fairing on the Falcon 9 is wide enough to loft ISS modules. It's also large enough to launch a folded Canadarm. The ISS could have been assembled using repeated Falcon 9 launches (alternating crewed and uncrewed) more rapidly and more cheaply than it was with the Shuttle. And that's if it was being flown expendable. Flying reusable, the cost savings would be astronomical. I understand. These potentially are: 9 engines, 2 tanks, legs and avionics. Musk said that the body, tanks, and engines were original.

Person-sized. Yeah, the reuse of the F9 first stage is closer to the reuse of the SRBs than to the reuse of the orbiter. Though, on that note, the reuse of the F9 first stage is vastly better in every way than the reuse of the SRBs. The SRBs were ditched in the ocean, towed to shore, and completely rebuilt inch-by-inch.

My gut says that the actual, physical refurbishment done to this stage was significantly more than the cost of testing and certifying a new stage, but significantly less than the cost of a new stage. If you include the development costs (all the testing, inspections, and refurbishments done to prior landed stages), then it's certainly more than the cost of a new stage.

In the technical webcast, one of the SpaceX employees said "Go for launch, go for age of reflight." Thought that was a fantastic touch.

Technically the circularization burn was completed while the booster was landing; the second burn is at the perigee of the geostationary transfer orbit. The grid fin actually did catch fire...and by catching fire, I mean the aluminum started to burn with the oxygen. At least that's what was implied during the presser. They are switching to titanium for Falcon 9 Block 5. I imagine that in the event of a grid fin loss, they've already programmed the opposite grid fin to fold closed, or for the other two to adjust their angle to maintain nearly the same control authority. It would be dicey, though. After they do the inspections (and, potentially, some investigative test-fires), they're presenting it to the Cape as a gift. SES is also getting some pieces for its board room.

Not a linguist either, but I did stay at a Holiday Inn Express last night. Did I date myself? I think I just dated myself. Craft is the preferred term to encompass everything. "Vehicle" works for anything other than a space station (launch vehicle, landing vehicle, transfer vehicle, ascent vehicle, etc.). "Module" can refer to the individual segmented components of a craft, some of which may be independently operable (crew module, boost module, propulsion module, landing module, excursion module, service module, command module, hab module, docking module, etc.).

I have to imagine that "handrail" would get pretty toasty in an abort. Which, again, would be the least of anyone's worries.

As proposed, it's a sub-GEO orbit.

The engines and a lot of the mechanisms are at the bottom, so I don't imagine it will be significantly top-heavy. Height-to-width is about the same as the Dragon 2, actually. That's not to say I like it. I don't. Egress out of something like that is just awful, and if one of the legs fails, there's not even a slim chance of survival for anyone. But making a horizontal-attitude vertical-landing Mars-capable lander that can still somehow manage to SSTO from the martian surface when fully fueled is a very, very tall order. But then you are dealing with multiple re-entry events instead of just one. Ideally there would be a way to use the crew module's TPS to help return the engines and use the crew module's LAS engines for landing, but have the engines and tankage break away in a launch abort. But configurations for that are hard to come by. Recoverable third stage doesn't make much sense to me. An orbital tug is more likely. If GTO launch frequency was much higher, I can see the advantages of an orbital tug that can grapple and push comsats from LEO to GTO and then aerobrake+burn to recircularize in LEO. But I don't think launch frequency is high enough to support that.