sevenperforce

I'd wager the available interior space has increased slightly. My "at-a-glance" assessment was like-as-not colored by the fact that I can only see the top half of the rocket, which probably messed with perception.

I pulled up the source image and measured the pixels. Unless my pixel-spotting is WAY off, that's a 5.28-meter fairing.

That DEFINITELY looks bigger.

If you allow one expendable Falcon 9 flight for Dragon 2, you'd have more than enough margin to add a propulsion pallet to get out of lunar orbit and back to Earth, and to do so without man-rating FH. F9 with Dragon 2 then takes the role of Ares 1 with Orion in the Constellation mission profile, and you merely need to develop a lander with enough propellant in its descent stage to do the lunar insertion burn and drop it on a Falcon Heavy with two-core recovery; launch vehicle costs are then just $187 million. Unfortunately, with the 311-second isp on Apollo-class hypergolic engines, you'd only be able to allow a lander assembly massing 5.31 tonnes post-LOI, which wouldn't work. If you used something like Blue Origin's Blue Moon vehicle, which is presumably built around a vacuum-optimized BE-3 engine, then you could manage an 11.1-tonne lander assuming isp in the range of 425 seconds. EDIT: Of course, with an isp of 425 on your lander, you could afford to simply make the Dragon 2 stock, and use the lander's ascent engine for the lunar exit burn, which saves you $30M in launch vehicle costs and drives your allowable LM mass up to 14.8 tonnes. Should be plenty; even if it was single-stage, it would still be able to manage a dry mass of 4.36 tonnes. That's if you want to trust the crew's lives to a hydrolox engine.

Was this treating NASA-sponsored dev for Dragon 1 separately? SpaceX has incurred dev costs for Dragon 1, dev costs for Dragon 2, dev costs for Falcon 9, and dev costs for Falcon 9 Block 5. How much of these dev costs were passed on to NASA? Also, as I understand it, NASA flights cost more than commercial F9 flights because NASA is paying SpaceX for an ISS delivery service, which includes the spacecraft, unlike commercial F9 flights where you are paying only for the cost of the launch vehicle and you supply your own spacecraft.

Then there are program costs. The supposed marginal launch cost of Shuttle was in that range, too, and they use a 500 M$ figure for the launch cost of SLS. we all know that those figures are nonsense, however. With a commercial provider, the dev costs only matter to the extent the customer pays them. In the case of NASA, they will have paid a couple billion for Dragon 2, plus 100 M (whatever) a flight. I think they are contracted for 4 flights (6 counting the first 2 tests). That leaves them pretty close to 500 M$ per D2 flight in total cost. Any crew missions going forward would drop that cost. CST-100 more like twice that price. Not sure what you're arguing. $566 million per launch was the Saturn V launch vehicle price per mission. Program costs were $2.5 billion per launch ($33 billion, 13 launches). And Falcon Heavy was not developed with outside funding.

Glad you caught this on your own; I would have felt bad pointing it out. Moon missions can be done in one of three ways: lunar orbit rendezvous (Saturn V/Apollo), Earth-lunar orbit rendezvous (Constellation), or joint lunar orbit rendezvous. ELOR assembles an Apollo-style stack in LEO, while JLOR sends crew and lander to the moon in separate TLIs. If we baseline the mass of the command module and lunar module at the ones used in Apollo (28.8 tonnes for the CSM, 16.4 tonnes for the LM), then the cheapest way to do ELOR is to send both of them up together on a 2-core-recovery FH, send crew up on an RTLS F9+Dragon, and then send up another 2-core-recovery FH for the TLI stage. If you do JLOR, then the LM needs to brake itself into low lunar orbit, meaning we shift about six tonnes of propellant off the CSM and onto the LM descent stage, bringing the CSM down to 22.8 tonnes and the LM up to 22.4 tonnes. This brings both vehicles within the direct TLI capability of 2-core-recovery FH, meaning you would still need two of those but no additional booster for TLI. Of course, you still need to either man-rate Falcon Heavy or do an Earth rendezvous with a separately-launched Dragon 2 before the TLI burn. To do it more cheaply, you'd need to drive down the mass of the modules or somehow incorporate Dragon 2. The Saturn V cost $566 million per launch in inflation-adjusted dollars, so reducing launch vehicle costs by half is not a bad start. Obviously development costs for the spacecraft are an issue; I was just looking at launch vehicle capabilities.

What happens now? Does it reset? If you have a 12-minute burn and you want to break it into three periapsis kicks (as you should), so you set your node for three orbits ahead and start burning, your "dV remaining" counter goes up, rather than down, and your navball goes haywire.

A cursory look will show that a single Falcon Heavy with the core expended can deliver a payload 4X the launch mass of MSL direct to TMI. Red Dragon would have posed no problem for side booster recovery. The Apollo lunar stack was 45.9 tonnes. This could be lifted in a single launch by FH with the core expended, or two launches with three-core recovery. Expending the core is cheaper. Of course, you'd need another F9 RTLS launch to get the astronauts up to the stack if FH wasn't man-rated. Then a single core-expenditure tanker launch docks to the nose for TLI. Apollo for $252M.

"Falcon Heavy (booster recovery x2)" is the Falcon Heavy core expended with two side boosters landing maximally downrange on OCISLY and ASOG. Adding a third ASDS would not help; if the two side boosters are landing maximally downrange, the core is going to be going far too fast for recovery. Any added margins would need to be reserved for a longer boostback burn. The recovery x2 values (expending the core and landing the two side boosters on droneships) are taken from Elon's statement on Twitter that this approach is only a 10% performance penalty in comparison to expending all three. I wasn't really able to tackle the accuracy of that quote, but it is not beyond reason; the amount of propellant that the side boosters will need to reserve for ASDS landing is really quite low in comparison to what it does to lifting the whole core and upper stage stack. A 10% payload penalty means that staging occurs around 230 m/s slower. 230 m/s for the whole stack equates to many times more for a nearly-empty booster...back of the envelope would say 800-1000 m/s per booster, which is plenty for entry and landing burns. Recovery x3 is based on the quoted performance of Falcon Heavy; it is advertised at $90M for up to 8 tonnes to GTO, which is lower than the maximum GTO performance of an expendable Falcon 9. So yeah, there's a huge jump. It makes sense, though, because the core is going so blindingly fast at staging that it needs to reserve a lot of propellant, either for a boostback or for a longer entry burn. They will never do a 3x RTLS flight, because it would drop performance down into the Falcon 9 ASDS range.

And now, on a completely non-BFR-related topic.... Ever since Falcon Heavy's test launch, I've been working on building a table of Falcon family payload performance. I've come up with a pretty robust and (I think) accurate model, and so I've got payload and performance for a wide range of destinations. I haven't seen anything this complete or this useful anywhere online. Comprehensive Derived Falcon Family Payload Table (metric tonnes) Vehicle LEO (tanker only) LEO (payload) GTO (2.27km/s) TLI (2.73km/s) LLO (4.04km/s) GEO (4.33km/s) TMI (4.30km/s) price (USD) Falcon Heavy (expendable) 96.91 63.80 26.70 25.17 16.68 15.22 16.80 $150M Falcon Heavy (booster recovery x2) 87.22 57.42 24.03 22.65 15.01 13.70 15.12 $95M Falcon Heavy (booster recovery x3) 23.53 18.11 8.00 6.66 3.65 3.12 3.17 $90M Falcon 9 (expendable) 24.99 22.80 8.30 7.74 4.25 3.65 4.02 $92M Falcon 9 (ASDS recovery) 17.04 13.30 5.50 4.50 2.14 1.71 1.75 $62M Falcon 9 (RTLS recovery) 11.74 9.41 3.51 2.70 0.85 0.52 0.56 <$62M Underlined values are values actually provided on by official SpaceX sources; italicized values came from statements made publicly by Elon. All other values are produced by the model. RTLS performance values were estimated by carefully comparing staging velocities for RTLS missions to staging velocities for ASDS missions; this is likely an underestimate of RTLS capabilities because SpaceX has probably not pushed RTLS to its limits in missions to date. Falcon Heavy with three-core recovery may also be underestimated for LEO performance, since Falcon Heavy's TWR is so much higher than Falcon 9's, but it should be accurate for BLEO destinations. LLO and GEO delivery assume extended restart capability for the MVac and do not account for propellant boil-off; this must be factored in. LEO tanker performance assumes that the payload is nothing but an International Docking Adapter (mass: 526 kg), so that it could be docked to a vehicle waiting in LEO for an ejection burn. For ASDS payloads below the maximum payload, SpaceX sends the ASDS only part of the way out and uses the extra margin for a boostback burn to save money and time. ASDS performance above assumes no boostback burn and maximum downrange ASDS position.

Saturn V was 10 meters; BFR is 9. BFR will be substantially cheaper to build, in inflation-adjusted dollars, than the Saturn V. Remember that even if SpaceX flew the Falcon family fully-expendable, they'd still undercut their competition at every level. They build sturdy rockets cheaply. NG has a reusable upper stage? And once you have a launcher with a 150-tonne payload, you can start talking about a third-party payload delivery service that would buy the BFR launch slot for ~$10M, pack on a dozen tonnes of hypergolics, and sell fifty smallsat slots at $500k each, with promised delivery to whatever orbit you desire.

IIRC, highest ablation will actually be Mars entry from Earth TMI, because the Martian atmosphere is more tenuous and so peak heating will be way higher. Well, if you can deliver your 5 tonne payload direct to GEO, then aerobrake to LEO and transfer 40 tonnes of propellant to a tanker, then land, for operating costs of $4-5 million...why wouldn't you?

Well, it does have wings. But the key difference is that it doesn't have to glide down deadstick to a landing; instead, it glides down to landing approach, then pitches up, stalls, fires its engines, and lands on its tail. If it works, it will be AMAZING to see. The BFS's fueling connections are on either side of the engines, where it connects to the BFR. They'll use the same connections to transfer prop on orbit, and a relatively small rover can drive underneath, raise a fuel connector probe, and "dock in" to fuel it that way on Mars. Mars SSTO is not too hard, not if you have high-energy propulsion. A single BFR, or a single BFS? The booster alone would certainly have SSTO capabilities, just like Falcon 9 already does. Cargo BFS could do a zero-payload SSTO flight with enough reserves for deorbit, but probably not enough reserves for landing. It would have more margins if it had more SL Raptors, since its low loaded TWR would be killer on liftoff.

No, nothing like that. The BFS reaches orbit with significant fuel reserves. It deploys its payload, fires for a deorbit burn, conducts a biconic re-entry, and stall-flips to land on its tail and landing legs. That's the reason for those winglets, actually. It's possible for the BFS to do what it needs to do on body lift alone, but not if it has to deal with entry on both Mars and Earth with a range of entry velocities. Winglets add enough aerodynamic control to handle that wide range.

I'm not 100% on this, but my understanding is that SpaceX itself actually volunteered for a 5-flight man-rating process because that was easier and quicker for them than going through an extensive modeling and analysis program. Other launch providers change the launch vehicle configuration from flight to flight (different number of boosters, etc.) to handle changing payload requirements, but SpaceX has a single vehicle configuration and meets payload requirements by adjusting reuse (from RTLS, which is very inexpensive, to boostback-ASDS, which is slightly more expensive, to no-boostback-ASDS, which is more expensive, to expendable). It costs SpaceX virtually nothing to man-rate Falcon 9 Block 5 this way. I don't think there is any man-rating standard for private spaceflight. What "massive engineering problems" are you thinking of, exactly? Lifting-body entry has been tested extensively with the Shuttle. SpaceX already has firsthand experience with blunt-lifting-body entry. Supersonic retropropulsion is figured out. BFR is smaller-diameter than Saturn V. This makes me think you haven't even paid attention to the actual plans. The first few BFSs will be one-way and unmanned, and will carry a rudimentary ISRU rover and associated equipment that will be craned down to the surface. The Progress spacecraft has routinely refueled the Space Station through fluid transfer connections around the perimeter of the docking ring. It's been done. SpaceX has a solid plan to make it more effective. But don't forget: even without the Mars aspirations and orbital refueling, the BFR/BFS system would be the most cost-effective LEO access vehicle in history. Cheaper per-flight than Falcon 1.

Fix the maneuvering nodes so that it is possible to maintain a single node while employing a series of periapsis kicks.

Falcon Heavy is nowhere near man-rated. If they put off man-rating FH in favor of working on BFS, they can still do their cislunar tourist flight and be farther along toward their goal. They need five consecutive Block 5 F9 flights for man-rating. Same with Red Dragon. Could they have landed a Dragon 2 on Mars propulsively? Absolutely. They would have needed to rebuild the fuel tank system and do a few other things of that nature, but it absolutely would have been possible. But why spend time and money on a dead-end when they can devote resources to building something that can actually land people on Mars?

I'm not sure what you mean by "ramjet", but I am pretty sure you haven't made a stationary ramjet, because stationary ramjets cannot exist. Hydrogen peroxide (H2O2) is a common chemical which is available commercially in low concentrations (1-3% H2O2, 97-99% H2O), but can be used in high concentrations as a low-efficiency rocket propellant. It's just water with an extra oxygen atom crammed into the molecule; that extra oxygen atom is unstable and has a great desire to suddenly and energetically depart. High-concentration hydrogen peroxide (often referred to as "high-test peroxide" or HTP) has the advantage of producing a hot mixture of steam and oxygen gas at a volumetric ratio of 2:1, which means it could be suitable for starting a ramjet. A ramjet functions by directly compressing supersonic airflow, slowing that airflow to subsonic velocity, and burning it with a fuel; if you use an HTP rocket in front of your ramjet, it could temporarily use the supersonic exhaust stream as its oxidizer source until it got up to speed. What you're looking for is most likely a rocket-combined-cycle engine. Google it and read up on it until you can't read any more.

If I understand your pseudocode correctly, this merely says that warp is limited to the lowest warp currently being requested by all other players...so unless everyone requests at least a warp of x2, there's no warp at all? I don't think this is a good solution. Players 3 and 5 should be able to execute a docking maneuver above Laythe while Player 2 warps from Eve to Duna and Player 4 warps around Kerbin to get a good Mun transfer point and Player 1 is afk. The trouble arises in defining "stable situation". What if Player 2 has timed his ascent to rendezvous with a previously-launched craft in LKO? He reaches orbit and turns off his engines, ready to coast to the rendezvous, and suddenly warp kicks on, and he whips past the rendezvous before he can say "wait".

Without trying to be in any way rude or dismissive... If you need to ask these kinds of questions, you should not attempt to construct any engine in real life. We do not want you to blow up.

Not sure this is the right forum. Putting a pulsejet in front of a ramjet will not provide the necessary ram effect for the ramjet to function, because the pulsejet exhaust will be oxygen-poor. You can, however, have something like a peroxide monopropellant rocket in front of your ramjet, to provide a high-speed flow of oxidizer for your ramjet to combust with.

Because lots of people will die.

Aerospikes would be useful. A plug nozzle is good, too. Draggy. Another way to make a stand-out entry would be to advance a different entry and descent approach than standard heat shield. One problem with an inflatable heat shield is that it's not inherently stable; if you play with them in KSP, you know how easy they are to flip. What if you could use that to your advantage? Instead of an inflatable heat shield, suppose you advance a deployable rigid heat shield that comprises two triangular panels, forming (in essence) a glider shape. For high-altitude entry, you could have a passively stable configuration that would become progressively more unstable as the atmosphere thickened. Finally, the whole thing would flip, after peak heating, and it would look like an ultralight airfoil for guided descent. Then it would jettison and you would land propulsively.

Give me a spacesuit and I will be on the second flight. But yes, it looks like Fairing 2 and Big Net are go for launch! Does anyone have eyes on Fairing 2 yet?