Starship (BFS) Engineering Thread (#dearMoon edition + Musk Stainless Update)

[meekGee]

The #dearMoon campaign will continue to eclipse (!!!) the rest of the program, and we don't have a pure BFS engineering thread, so here goes:

Musk and friends have completely rewritten the EDL book.  Musk was really proud of his little joke where the "horizontal stabilizer" is just a leg... But the most telling quote was how it was all about asking the right questions - a very first principles approach that resulted in the "sky diver" EDL model.

Falling absolutely belly first, with four barn-door actuated surfaces (and some body attitude no doubt) providing active stability and trajectory control during almost the entire descent, followed by a very brief propulsive touch down.

Other than that, I captured:

- Fin/leg combo
- Aft cargo for easy ground access
- Fan shaped cruise PV
- Vista windows
- 6 tiers of port hole windows
- Full ECLSS

Did I miss anything?

Both capture of engineering information and speculation based on such information welcome below

[nacnud]

Identical engines and bells on booster and spaceship.

[speedevil]

Landing delta-v at 260m/s or so empty (or nearly so), using around six tons of propellant.
From here.
Landing on earth goes from mach 0.33, to 0 in a  burn of 16 seconds, from an altitude of  800m.
As it's nearly vertical, the times are small, and any horizontal component has minimal cosine losses, a linear approximation seems reasonable:
160m/s in gravity losses. 100m/s in delta-v.
260m/s in total.

The IAC2017 one in contrast  was around 700m/s (for Mars obviously).

Assuming 85 tons dry weight, and 340s ISP, that comes out to only six tons of propellant to land.
260m/s in 16 seconds is 16.25m/s^2, or 138 tons average throttle - around 70%.

This would imply maximum single-engine landing weight with this landing timing is about 125 tons.

Savings may be made if accuracy at aerodynamic phase termination is high enough.
Remembering the RCS that is beefy enough to cope with 60mph winds.

[speedevil]

But the most telling quote was how it was all about asking the right questions - a very first principles approach that resulted in the "sky diver" EDL model.

Falling absolutely belly first, with four barn-door actuated surfaces (and some body attitude no doubt) providing active stability and trajectory control during almost the entire descent, followed by a very brief propulsive touch down.

Also mentioned in passing was the possibility of interrupting entry in the middle, going to LEO, and then later landing.

This greatly simplifies trajectory design and eliminates concerns over weather at your landing site being an issue when you're coming in from >>LEO.
It also basically eliminates concerns about accuracy of that first phase and having to have it right to hit your landing site.

Similar issues arise on Mars, and the benefits are also similar - but the atmosphere is not as well understood.
If it could be better understood - and perhaps practicing this sort of entry would help somewhat, concerns about landing in a dust storm or timing your departure to exactly hit a narrow landing window go away.

[Rocket Science]

Elon uses the terms wings/fins interchangeably including for the canards in the presentation...

[ThereIWas3]

F9 does not do the belly-first re-entry, but a "re-entry" burn instead and falls tail-first the whole way with just the "landing burn" at the end.  (Ignoring the "boost back" burn that only happens sometimes, to reverse direction)

A big difference though is that the F9 first stage is not even going at orbital speed, much less interplanetary arrival speed, and KE 　〜 mv^2.   Without doing the real math, it would seem to me that doing a propulsive-only EDL with BFS would take a lot more propellant than the sky-dive EDL.  It would be interesting to see whether the 6 tons BFS would use is more or less than whatever F9 uses.  I would not be surpised to learn BFS uses less.

[speedevil]

It would be interesting to see whether the 6 tons BFS would use is more or less than whatever F9 uses.  I would not be surpised to learn BFS uses less.

It is far less.
For a kilometer a second (entry and landing burns), though F9 is about a quarter of the weight of BFS, it uses perhaps twice the propellant.

[nacnud]

Elon uses the terms wings/fins interchangeably including for the canards in the presentation...

I guess there isn't a name for these aero features yet. Need a clever combo of elevator, rudder, airbrake, leg, plus a few more probably

[niwax]

Elon uses the terms wings/fins interchangeably including for the canards in the presentation...

I guess there isn't a name for these aero features yet. Need a clever combo of elevator, rudder, airbrake, leg, plus a few more probably

Especially since the third one apparently has no aerodynamic purpose und is only there to keep symmetry and house the leg

[nacnud]

Elon uses the terms wings/fins interchangeably including for the canards in the presentation...

I guess there isn't a name for these aero features yet. Need a clever combo of elevator, rudder, airbrake, leg, plus a few more probably

Especially since the third one apparently has no aerodynamic purpose und is only there to keep symmetry and house the leg

Elonevator

[envy887]

Elon uses the terms wings/fins interchangeably including for the canards in the presentation...

I guess there isn't a name for these aero features yet. Need a clever combo of elevator, rudder, airbrake, leg, plus a few more probably

It's purely an airbrake. Extending the front brakes slows down the front more, pitching the nose up. Extending the rear brakes slows the tail and pitches the nose down. Extending the right brakes causes a roll the the left, and vice versa.

They only need RCS to control yaw. Pitch and roll is controlled by differential braking.

[Rocket Science]

Elon uses the terms wings/fins interchangeably including for the canards in the presentation...

I guess there isn't a name for these aero features yet. Need a clever combo of elevator, rudder, airbrake, leg, plus a few more probably

Especially since the third one apparently has no aerodynamic purpose und is only there to keep symmetry and house the leg

That's it's sole purpose on this application and a vertical stabilizer is not needed in these EDLs profiles including for Mars as it would be aerodynamically "blanked-off". Shuttle had a large one for the long time spent in hypersonic cruise down to subsonic landing. RCS, flying wings (tail) (stabilator) and canards in this application...

[Ludus]

Identical engines and bells on booster and spaceship.

Plus the comments that in later iterations/versions vacuum versions might be swapped in and cargo stores removed.

A greater emphasis on the idea that this is a system that can be designed to optimize different things in different versions.

[philw1776]

Raptor now a 200 tonne class engine
300 bar
Raptor upgraded


Downgrade to 100 tonnes or slightly more to LEO from 150 tonnes

I speculate that now there will be <31 engines on stage one BFR.  Guessing 29 or 28.

[ZachF]

Elon uses the terms wings/fins interchangeably including for the canards in the presentation...

I guess there isn't a name for these aero features yet. Need a clever combo of elevator, rudder, airbrake, leg, plus a few more probably

It's purely an airbrake. Extending the front brakes slows down the front more, pitching the nose up. Extending the rear brakes slows the tail and pitches the nose down. Extending the right brakes causes a roll the the left, and vice versa.

They only need RCS to control yaw. Pitch and roll is controlled by differential braking.

They could do yaw with the wingbrakes too.

[ZachF]

Raptor now a 200 tonne class engine
300 bar
Raptor upgraded


Downgrade to 100 tonnes or slightly more to LEO from 150 tonnes

I speculate that now there will be <31 engines on stage one BFR.  Guessing 29 or 28.

I bet the engine number is the same, and liftoff weight is now >5,000 tonnes.

[ThereIWas3]

I am interested in how they design the couches for this thing.  Enroute everything is weightless so an ISS-style sleeping bag is sufficient and folds away nicely.   Using the BFS as a long-duration shelter on Mars requires a flat real bed, though with probably less foam than would be needed on Earth.  Inflatable?

The tricky part is EDL which although it only lasts a few minutes out of a possibly months or years long mission, has to be done right to prevent injury.  From entry-interface to over-the-target (Apollo Lunar landing called this the "low gate") acceleration is to the side and probably more than 1G.  Then it pivots to tail-first and fires the engines so now acceleration it toward the tail at something over 1G.   But you can't use the same flat bed arrangement as is used for landed sleeping because a flat bed is not sufficently supportive.

Something like a modern hospital bed might work, though those are heavy.  Maybe a custom "bed liner" like Soyuz uses, you get from a closet and fasten to the bed top?  Maybe that inflates too.  But then you still have to deal with the 90 degree pivoting.  And the direction of the sideways acceleration is different relative to the walls depending on where your compartment is located around the circumference!   Augh!   Trying to avoid some Rube Goldberg device or something from Thunderbirds...

The pilots can have fancy reclining couches with affixed screens and controls but all the passengers do not need anything so elaborate.

Maybe passengers are not in their compartments at all during EDL, but in racks of special purpose chairs installed down the central core, all aligned the same way.

[Rocket Science]

Elon uses the terms wings/fins interchangeably including for the canards in the presentation...

I guess there isn't a name for these aero features yet. Need a clever combo of elevator, rudder, airbrake, leg, plus a few more probably

It's purely an airbrake. Extending the front brakes slows down the front more, pitching the nose up. Extending the rear brakes slows the tail and pitches the nose down. Extending the right brakes causes a roll the the left, and vice versa.

They only need RCS to control yaw. Pitch and roll is controlled by differential braking.

I stated in the previous the thread including an attached paper that canards are also used not just for lift, pitch, but for roll as well in this application...

[philw1776]

Raptor now a 200 tonne class engine
300 bar
Raptor upgraded


Downgrade to 100 tonnes or slightly more to LEO from 150 tonnes

I speculate that now there will be <31 engines on stage one BFR.  Guessing 29 or 28.

I bet the engine number is the same, and liftoff weight is now >5,000 tonnes.

Doubt it.
That would be a downgrade of mass ratio efficiency, given that this rocket delivers ~100t to LEO not 150t.
Arguing against myself, the new winged BFS now probably masses above the old 85t

[envy887]

Elon uses the terms wings/fins interchangeably including for the canards in the presentation...

I guess there isn't a name for these aero features yet. Need a clever combo of elevator, rudder, airbrake, leg, plus a few more probably

It's purely an airbrake. Extending the front brakes slows down the front more, pitching the nose up. Extending the rear brakes slows the tail and pitches the nose down. Extending the right brakes causes a roll the the left, and vice versa.

They only need RCS to control yaw. Pitch and roll is controlled by differential braking.

I stated in the previous the thread including an attached paper that canards are also used not just for lift, pitch, but for roll as well in this application...

Are the actually canards though? The flow over them will be stalled for 100% of the relevant flight regime. I can't think of any canards that operate 100% stalled.