This is the one I'm really interested in seeing. In theory it is the culmination of all the things they will have learned from landing boosters back. Given the thermal damage to the guide fins on SES-10 (another one that wasn't 'supposed' to be recoverable) I expect there will also be an upgrade there some how.

I found it interesting that the latest piece on New Glenn showed the booster landing with winged guide fins. I'm not sure that will work given NASA research into hyper-sonic flight with airfoil wings. If they can pull it off that will be a good data point as well.

At lunch the other day we got into a discussion as to whether a 'landable' booster could also continue on to orbit. Specifically if, after the second stage had separated and headed on, the first stage could re-light and put itself into its own orbit. I couldn't find anywhere that suggested what the mass of the booster was at separation. Would be a fun kerbal program.

SpaceX announced that the plan for the currently aluminum-with-thermal-paint grid fins was to make them out of titanium.

https://www.nasaspaceflight.com/2017/03/spacex-new-spaceflig...

Thanks for that, I did not realize they had described the redesign publicly.

If you want to put the first stage in orbit, then eliminate the second stage engine, and plumb the second stage fuel into the first stage tanks. Replace the central first stage booster's bell with a wider vacuum-optimized bell. Hopefully, the weight saved by eliminating the second stage engine would make up for the reduced efficiency of that central engine, during atmospheric flight.

Easy peasy in KSP. No idea if it is possible IRL.

Running a vacuum-optimised engine in atmosphere means the exhaust is overexpanded, which causes severe flow seperation and chaotic flow. Even small amounts of overexpansion are harmful to the nozzle; overexpansion to vacuum-type ratios is likely to cause an Rapid Unscheduled Disassembly.

I believe the main problem here is the difficulty in pumping fuel around, rocket fuel injection is not a trivial matter

source: also played lots of KSP and looked into why asparagus staging is more difficult IRL

They've made some amazing progress! However, a landable booster can't put itself into orbit -- its engine cutoff is always well under 3000 m/s (today's was about 9800 kph according to the YouTube broadcast, or about 2700 m/s, and that was with no fuel margin left for landing), but about 7800 m/s is required for orbit.

Elon's tweet today (https://twitter.com/elonmusk/status/669132749500887040) was that a booster with no payload could put itself into orbit. And that certainly supports your statement. It clearly would not be possible to launch a second stage and have the booster stage make it all the way into orbit. But it isn't as far out of whack as it might be.

Tsiolkovsky is very clear about the propellant mass fraction needed for single stage to orbit, and it is unachievable with current materials, i.e. you can't build a useful SSTO using Al-Li tanks.

https://en.wikipedia.org/wiki/Single-stage-to-orbit#Design_c...

The first stage has a fuel mass fraction of 95.5%. Propellant densification and high thrust to weight ratio engines buy you lot. That should be enough to make orbit.

SSTO with a significant payload then, presumably.

Yup. Even carbon composites don't really work out (minimal payload) until the rocket is very large (see: ITS)