sevenperforce

Although, realistically, any orbital speed collision will wreck you.

Incidentally (and not to SpaceX fanboi), this is yet another reason why liquid-based crewed launches are intrinsically safer than using SRBs with human cargo. Even though the LES tower was jettisoned before failure, the core stage engine (I refuse to call it the second stage on principle) shut down nominally and the crew capsule separated using only small liquid thrusters. Liquid boosters are more complex, but I'll be damned if they don't fail in delightfully clean and safe ways. Even CRS-7 incurred 0% damage to the capsule despite utter lack of LES.

Dragon 2 has no EVA capability whatsoever. If NASA wanted to repair Hubble, they could push money to SpaceX to accelerate Dragon 2, launch an unmanned Orion to a Hubble-aligned orbit on Delta IV Heavy, and send a crewed Dragon 2 up to the same orbit on Falcon 9 B5 before the first SpaceX ISS mission. Orion and Dragon would mate and move into a Hubble intercept. Previously-trained astronauts would use tethered EVA from Orion to repair Hubble, using both Orion and Dragon 2's thrusters for stationkeeping, and then return. The sheer complexity would make the Soyuz-Apollo joint mission look like a cakewalk held at a park during a picnic.

I see what you did there.

From everything I've seen, S1 sep always comes before LES jettison. Glad the crew is ok.

Looks to be a 70-75 second test fire, by my count. Mach diamonds not nearly as pronounced as on the devscale engine. Suggests that they really are going all-out with sea level engines, rather than designing a compromise nozzle with a converging-diverging-converging bell like the SSMEs. The fact that they expect to be able to do a lunar free return (roughly 2.8 km/s out of LEO) without refueling, combined with the fact that they expect upwards of 100 tonnes to LEO in cargo configuration, can provide some degree of guesstimate as to how much "payload" they expect to comprise in a dozen passengers and their cabins, etc.. The mechanics of that flip are going to be challenging. Makes more sense now than it did before, though.

Ugh. I agree with you and disagree with the article. "...astronaut David Scott intentionally left a simple plaque the size of a beer coaster with the names of astronauts and cosmonauts that perished in the space race on the lunar surface. Charles Duke left a family portrait encased in plastic. Astronaut Alan Shepard decided to drive a pair of golf balls across the surface of the moon. Apparently for kicks, astronaut Edgar Mitchell hurled one of the support rods of a solar wind collector like a javelin through the lunar atmosphere. All these astronauts peed into sacks called urine bags then discarded them on the moon. (It reminds me of guys at a bachelor party in New Orleans.) Twelve men have walked on the moon, and zero women. It shows. The moon has become the solar system’s largest bro playground." Moralizing, self-aggrandized, self-righteous claptrap.

I did test the biconic, four-brake entry in KSP. It works quite well for how much limited control I was able to exercise. KSP's SAS engine doesn't understand what's happening, so the surfaces don't really work via SAS. I had to bind each control surface to an action group and toggle that way, and this meant no fine control authority. Forward and aft flaps are both elevons set perpendicular to the axis of the vehicle. Execute deorbit burn, then use RCS to rotate around to prograde, then SAS hold. Pitch up to radial-out as you cross 70 km. Set all flaps to maximum lateral extension (launch position for forward flaps; ~45 degree deflection from landing position for aft flaps). During descent, hold AoA at approximately 75 degrees and damp roll. Use the following controls: To pitch up, feather the aft flaps dorsally, away from landing position, and feather the forward flaps ventrally to maximum lateral extension. To pitch down, feather the forward flaps dorsally and feather the aft flaps ventrally to max lat ext. To cancel clockwise roll, feather both port flaps dorsally and hold both starboard flaps at max lat ext. To cancel counterclockwise roll, feather both starboard flaps dorsally and hold both port flaps and max lat ext. Care should be taken to avoid dropping AoA below 40 degrees. At roughly 40 degrees, the aft flaps come out of stall and begin to develop lift, kicking the tail up and turning the whole vehicle into a lawn dart. You have NO differential braking authority without a high AoA. Entry works best if you maintain forward flaps at maximum lateral extension and use the aft flaps for primary pitch control. This allows some degree of combined pitch/roll authority, because you can control pitch with the aft flaps while damping roll with the forward flaps. Yaw authority is possible if you lower AoA to around 50 degrees, so that the aft flaps develop stall lift (still no laminar flow) and then brake differentially. For port yaw, maintain lateral extension on the port forward and starboard aft flaps but feather the other two dorsally; this will point the small control surface lift vectors in the direction you want. For starboard yaw, reverse. With four differential braking surfaces, the landing flip is a breeze. Extend the forward flaps and feather the aft flaps to full dorsal position, which pitches you up into a body-lift stall and orients the engines retrograde. Throttle up, and all control surfaces stop interacting with the airstream as you complete the landing burn. Rotate the aft flaps forward into landing position and land.

Work has been nuts so I just now got a chance to come here and comment on the presentation (which I was lucky enough to view live).... WOWWWWWW AHHHH! Okay, got that out of my system. I must say I was completely sure that those trapezoidal panels were something fancy to protect the engines or act as a collective nozzle or something...and no, it's aft cargo. SMH. Aft cargo is neat, though. Competes neatly with Blue Moon, if you think about it. "Hey, we can deliver up to 4.5 tonnes to the lunar surface!" "Oh, that's nice. We can deliver 5 tonnes to the lunar surface. Twelve times. In twelve different places. Reusably. While carrying tourists." Also teases the possibility of a separate abort assist motor if NASA is hesitant about abort TWR; you could replace several of the pallets with solid motors to give the initial kick away from a RUDing first stage. On to the control surfaces. This makes a LOT of sense, and it explains why I was never able to get my BFS clones to re-enter properly. The four flaps do NOT have laminar flow over their surfaces during entry, so they do not provide any meaningful lift. Rather, they function as airbrakes. Elon made several comparisons to a skydiver, and that's exactly how it will work. A skydiver falls prone, with all four limbs extended to act as brakes. By pulling in or extending limbs in pairs, the diver can induce pitch and roll, and if skillful, yaw. I didn't see how the BFS could have three-axis control with only two control surfaces, but with four differential braking surfaces, it's fairly straightforward. It sounded a little odd when Elon made statements about "yeah it is riskier but I like the aesthetics," but on review I think that's looking more holistically. The risk is development risk, not operational risk; it's risky to force landing gear and control surfaces to develop as a single unit because if you can't do it then you've wasted a lot of time. And the aesthetics probably include stuff like simplicity, functionality, and so forth.

Visually, we see little evidence of such effect in the Falcon 9 launches. Then again, inward gimbal on skirt engines is very limited so they may not be able to gimbal inward enough. As far as I understand the fluid dynamics...which is not much, admittedly, but probably about as much as anybody here...it's a valid approach. Might squeeze out an extra 2-3% more thrust. Keep in mind that a full-size engine bell is only 5% more thrust and carries with it a bunch of extra dry mass. That's something completely different.

Such a configuration would really decrease the maximum thrust of the stage. Dev Raptor, SL nozzle, SL: 567 kN Dev Raptor, SL nozzle, space: 612 kN Dev Raptor, Vac nozzle, space: 633 kN Raptor, SL nozzle, SL: 1,700 kN Raptor, SL nozzle, space: 1,834 kN Raptor, Vac nozzle, space: 1,900 kN You know, looking at this, I don't think they need vacuum nozzles at all. The Raptor has such a ridiculously high chamber pressure that its SL pressure losses are really small, and it's only about 5% underexpanded in vacuum.

Nah, you misunderstand. The Raptor they've shown tests for was a scaled-down Raptor: So the development scale Raptor is a third the thrust of the full-size Raptor. Presumably, then, you could connect a dev scale Raptor to a full-size Raptor's engine bell assembly and you'd end up with a vacuum-optimized dev scale Raptor.

The dev scale Raptor is about the size of a Merlin, albeit heavier with much higher thrust. An MVac is roughly the size of a SL full-size Raptor. I wonder if they decided to mix 3 SL full-size raptors with 4 dev scale Vacuum Raptors and use the same engine bells for everything.

So, little-known fact: "aerospike" actually refers to a "plug" nozzle, the point being that you do not need a full-length spike because gases entrained at the base of the plug act as a stagnation zone which generate a spike effect. The spike is made of "air" (gases), thus aero-spike. If the outer engines are all gimbaled inward, then perhaps the same effect is possible, with the exhaust plume of the core engine acting as the virtual aerospike. Extensible nozzles a la RL-10 save length but not diameter. The BFS is diameter-limited.

They don't want a complete engine redesign.