sevenperforce

You're not alone there; the Imperial system sucks. Anyone in hard science feels the same way. I tuned in a few moments after liftoff; now to go back and watch that.

The SpaceX rep said they would inspect and prep this booster for reuse on this year's launch manifest; I don't know whether that was just a general statement or something specific.

She's down! What a magnificent, magnificent flight.

Is it just the lighting, or do those grid fins look different? Elon said they were swapping them out for titanium. Love seeing the cold gas thrusters from a distance; it's way more puffy than I would have imagined. I think this is the first time they've maintained a feed of telemetry on the first stage. I love that; I am sure we are going to see a lot of new analysis for how the re-entry works. Coming up on entry burn in a few moments. We are going to get to see the rocket tail-first during the entry burn...really really good image. Only 80 km up. Oh wow. That is...magnificent. Good god that is an amazing sight. Landing burn started!

WHOA holy mother of god that video of MECO, separation, and boostback ignition was incredible. This is the first time they have been able to maintain ground video the entire time.

Precisely. Keeping it there gives a central thrust column, if nothing else.

I'd have to dig a little to be sure, but I'd guess that the dry mass of the solid booster core is so low, comparatively, that it's still quite fine.

Aerocapture of the manned landing vehicle itself seems pretty doable. EDL on Mars requires heat shield and propulsion as a matter of course, so if you can manage direct entry and propulsive landing, you should also be able to manage aerocapture and circularization. Then, as I outlined previously, you can take advantage of propellant transfer (assuming you solved this problem before, in order to refuel your vehicles for TMI) to beat the rocket equation. Using aerocapture to get your return prop into Martian orbit, then transferring it back to the tanks of the bus you used to insert the transfer hab while you execute your EDL, seems very advantageous. And this allows partial or total reuse, which is nice if you can do it for cheaper than the alternative.

As I understand it, the MRO used its engines for the initial insertion, establishing an elliptic Mars orbit, then adjusted its periapse to gradually trim off orbital energy one orbit at a time to adjust. It was excellent for the mission requirements, of course, but it's very different from a much more aggressive "true aerocapture" where a single pass through the atmosphere moves you from a hyperbolic orbit to an elliptic one. MRO needed no special heat shield, while a true aerocapture requires a heat shield and high-gee resistance. A Martian sample return could be launched in 2020, if enough money was poured into it. But right now we don't even have a hab that could make the journey manned.

Then, of course, we must compare the mass of that whole aeroshell and hinged stow assembly against the fuel that would be required for simply burning to orbital insertion. Aerocapture (using the atmosphere to insert from a hyperbolic swingby to an elliptic orbit, then circularizing once around) is a tricky thing; I have done it plenty of times in KSP but doing it IRL is rather challenging; I don't believe it has ever been done. Aerobraking to EDL, yes, but no actual aerocaptures into orbit.

Daggone LOX tanks. If temperature at the first-stage LOX vent was out of the expected range, does that mean more stubborn LOX/COPV issues?

I guess it's 1.5-body. That is, large objects like planets and moons are on rails, but everything else reacts accordingly. Doesn't account for satellite mass.

I, for one, suspect uprating of the Merlin 1Ds in anticipation of Block 5.

And here...we...go!

That's not too kerbal for reality at all. My proposed mini-ITS upper stage with two dev Raptors? Precisely the transfer vehicle I was considering to use for this. 141-tonne propellant tank, two 1100-kN engines, 130 cubic meters of payload space, and auxiliary thrusters capable of landing 40+ tonnes on Mars after a lifting-body re-entry. Perfect. Propellant transfer port on the side or wing; docking port on the nose. Who cares if you have to switch back and forth? We have been docking spacecraft with the ISS for decades now; it requires some doing, but it is a solved problem. All the ISRU proposals I have seen are ridiculously limited, because you have to somehow integrate a separate crew capsule. Why on Earth (or Mars, as the case may be) would you want to do that?? Make the ISRU unit self-contained and reusable, and send it up to Martian orbit to refuel your lander. Your lander needs fuel to land, after all, and it is a lot easier to do an RCS-assisted docking and fuel transfer in orbit than it is to try and do a fuel transfer via EVA. You're going to need those high-specific-impulse, high-thrust chemical propellant engines to get off Mars, so why not use them for your transfer? I'm sure NASA has thought of docking upper stages, but consideration of propellant transfer might be a little low on their priority list, and reversible propellant transfer moreso. I haven't run all the numbers, but back-of-the-envelope estimates suggests at least ten times more TMI propellant is needed for a straight chemical Mars Orbit Rendezvous than would be required for an aerobrake-assisted Mars Orbit Rendezvous.

Crap, I totally forgot about this. Strapping people to continuously exploding bombs and blasting them up so fast that they miss the Earth when they fall back down is, by definition, rather risky. But doing it Shuttle-style is one of the poorest ways to do it.

If you have a Martian Orbit Rendezvous mission architecture (the Mars equivalent of the LOR used by Apollo) and the ability to transfer propellant on orbit, there's a neat trick you can use to beat the rocket equation. Launch your service module, your expandable hab, and your lander/crew module into LEO, fuel them all, and dock them together. Use your service module's engines to perform the TMI. Continue along your lazy transfer to Mars, making full use of your nice roomy hab. However, immediately before reaching Mars, physically transfer all your consumables from the hab into the lander. Disconnect the lander from the hab, and use the lander to aerocapture into low Martian orbit while the service module uses the last of its fuel reserves to do an orbital injection burn, still attached to the hab. Once both the lander and the service module are in orbit, they can rendezvous and dock again. The lander will then transfer all but a few days' worth of consumables into the hab, and transfer all but the bare minimum of its fuel into the service module's tanks. The lander then disconnects, enters, and lands. After the mission is complete, it returns to orbit with the last of its remaining fuel. It does a rendezvous with the hab and service module. All the remaining fuel from the service module is transferred back to the lander's tanks, and then the service module is jettisoned. The lander propels itself and the hab back to Earth with the same engines it used for Mars ascent. In this way, you get the advantage of having an expandable hab on both legs of the journey, but you don't have to carry nearly as much propellant, because the propellant used for your return trip enters Mars orbit via aerocapture, but stays in Mars orbit in the service module's tanks while the lander executes its mission. Same with consumables.

The BEAM's expansion is a onetime affair, so you'd need a complete redesign. Also, getting an expandable module inside an aeroshell is a tricky affair. The expandable module will have to be docked to a pressurized module while expanded, but will have to be deflated and removed from that module in order to be stowed inside an unpressurized cargo bay, but one which can be protected during aerobraking or re-entry.

Transfer vehicle size is a tricky thing. There's a minimum amount of physical activity space you're going to need to keep the crew from literally going crazy for any mission longer than a couple of weeks; the activity space is not necessarily going to go up very much based on mission duration. You're also going to need a certain amount of pressurized space to store consumables; this is going to increase significantly with mission duration. Going expandable, like BA330, is the easiest way to get a big activity space. But you can't aerobrake an expandable module, and aerobraking is the only way you can do it without essentially doing a whole Mars Cycler. My thought is that you build the activity space into something that can aerobrake, and use expandable modules (like BEAM) to store outgoing consumables. That way, you can doing a free-return on the outgoing leg, taking your sweet time to get to Mars, and then dump the expandable modules and aerobrake at Mars. You can then take a high-energy, low-transit-time return with as little mass as possible. The mini-ITS concept I've been working on has a crew cabin volume of about 130 cubic meters. I was thinking they would go two at a time, each with two crew and a couple of BEAM modules docked in to store pressurized consumables, on a slow outgoing trip. Once at Mars, they'd dump the BEAMs, aerobrake, rendezvous, and transfer all but the barest fuel and consumables to the one remaining in orbit. The other vehicle would enter, land, complete the mission, and return to orbit. Both would refuel in orbit, using a previously-sent tanker or a previously-sent ISRU ship, and head home on the highest-energy transfer they could manage.

How much dV do you need for the TMI, and how much dV do you need out of Martian orbit? Trying to figure out whether it would be better to go ISRU-ahead or tanker only.

Wait, how are you getting 9 km/s? I'm getting 10.2-10.6 minimum for the round-trip, assuming aerobraking direct from a minimum-dV Hohmann transfer at both ends. If fully-refueled in LEO, my mini-ITS can deliver 69 tonnes to Mars orbit via aerocapture and 47 tonnes to the Martian surface with propulsive landing. One mission profile I like is to put fuel in Martian orbit, either by sending a tanker along or by sending an ISRU-equipped lander ahead to land on Mars, make the fuel, and return to Martian orbit. That should be able to permit speedy transfers both ways.

Why thank you, I agree that it is pretty much perfect. I was already planning on three variants: one with a crew cabin and no unpressurized cargo bay, one unmanned with a cargo bay/fold-away fairing, and one that is just a tanker. The math works really ridiculously well. Speaking of which, do you have a procedural parts mod, by any chance? I only have the demo, obviously without mods, so as much as I'd like to build a mockup of the mini-ITS, I cannot. I suppose I could do it in Google Sketchup, but then I couldn't do any simulated EDS goodness.

SpaceX lists the Earth re-entry as 12.5 km/s or a bit more, but with peak acceleration at 3 gees. Mars entry is lower, at 8.5 km/s, but you end up with up to 6 gees due to the thinner atmosphere. The ITS presentation had a lot of these numbers provided -- delta v vs payload. So you can start there. One of the issues is mission duration.

And launch frequency. And the requirement of a crew. And availability.

If we're limiting ourselves to Kinetic Projectile Weapons delivered via re-entry, what are the advantages and disadvantages of using elliptical vs circular orbits? If you use a significantly elliptical orbit (e.g., Molniya), you can do plane changes and adjust your re-entry trajectory for very little dV, allowing you to hit a wide variety of targets. You also can have your platform's closest approach over friendly territory. But this significantly increases your lag from strike order to impact, making its military applications more limited.