A question on BFS/BFR gaseous prop supply system

[John Alan]

I'm trying to wrap my head around the required on board equipment for BFS/BFR...
Things that are installed into the bare airframe and must work...
One item which puzzles me a bit, is providing prop for the planned control thrusters..

I'm led to believe they are pressure fed gaseous meth/lox... right?   

NASA's project Morpheus demo'd these... found they are doable...
Spark ignition... they found them to work great under computer control (PWM control of spark and prop flows)
Wide range of thrust from cold gas to max fire conditions...

BUT...and this is where I get stuck...
Those thrusters are fed with gaseous Methane and Oxygen at ~2500psi (in the Nasa project anyway)...

SO...
BFS/BFR have ~50psi tanks of COLD liquid prop... Metric tons worth...

How will SpaceX convert that liquid supply into a high pressure gaseous supply kept topped up in onboard COPV's... ready to supply the thrusters at a moments notice?
Will they use this same supply regulated down to pressurize the main tanks as needed?

Conversion system used?
1)Heat exchanger and Compressor?
2)High pressure liquid Injector feeding a Boiler?
3)Other?

All options require some electric power I think... Some more then others... 

If option 2... would the boiler be electric or prop heated? or both as a backup source?

Looking for feedback on how you think they will do this function on BFS/BFR...
Speculation and opinion welcome on topic... as to how they may do this function... 

On later edit added...
IIRC...
Marine LNG import terminals use giant air or water warmed heat exchangers to get a gaseous methane form and then big turbine compressors to feed medium pressure delivery pipelines...
Local CNG plants then use high pressure compressors fed from the gas system utility...

Airgas plants get LOX semi tanker deliveries and use a smaller version of that with series screw or multistage piston compressors to refill welding cylinders...

But how in the world do you lightweight, miniaturize, and flight rate either process.... 

[Semmel]

I am also quite curious about these thrusters and hope Elon answers a related question at the reddit AMA. If that happens.

[John Alan]

Ok.. Yes..
I am aware that they could (while a main engine is running) tap off 250bar (3626psi) mains in the engines and refill a bunch of COPV's in a WARM location (not in the prop tanks)...
And that the booster could just use a GSE source to fill big enough COPV's (again in a warm location) to make the RTLS booster launch and landing run...
And in theory, BFS could be topped up before it was hoisted onto the booster...
Note... I did not indicate any GSE to refill BFS on pad... because they would not want to need a high pressure connection at the stage joint...

SO... one solution is... you gotta fire a Raptor engine on the BFS to refill the COPV's...

Ok... That's fine for early use on this system...
But send a BFS for a long term space flight or sitting on the Moon or Mars a while...
You gotta make high pressure prop from low pressure liquid... without firing up a Raptor...
I think... 

[speedevil]

Ok.. Yes..
I am aware that they could (while a main engine is running) tap off 250bar (3626psi) mains in the engines and refill a bunch of COPV's in a WARM location (not in the prop tanks)...

That gets you really uncomfortably hot propellant, with added water and CO2, which will condense out in many conditions, and may add all sorts of fun issues. You almost certainly need a heat exchanger to chill the hot gas, which will make the water fall out, before running it through pipes, and it's a mess.

A small electric pump could work, perhaps even with a internal combustion engine to power it. This might also be nice for surface ops in darkness, and when you can't put out the solar panels.

Gas thrusters can be run at low pressure, though having to run the thruster at low power for a few seconds to warm up a liquid heat exchanger might be inconvenient.

[IainMcClatchie]

A high pressure gas/gas methalox thruster operates with supercritical fluids (it's not really gas/gas).  No need to heat all the propellant, and you can run a low power high pressure electric pump on liquid, perhaps driven from a battery for BFR.

Methane becomes supercritical at 667 psi, 191 K.

Oxygen becomes supercritical at 877 psi, 154 K.

So long as the combustion chamber is higher pressure than 877 psi, neither propellant will convert from supercritical to two-phase after it leaves the injector face nozzles.

There will be a start-up transient as chamber pressure builds.  One way to avoid two-phase during the startup transient would be to electrically pre-heat a dozen milliliters of propellant right before the control valve while ready but not firing.  However, this leads to a different transient as the propellant temperature at the injector nozzle changes, especially since the amount of hot propellant will vary with the amount of time since the last firing.  I think this effect might not be too bad.

The size of these thrusters is gigantic.

When the BFR comes down it needs sufficient lateral thrust to overcome crosswind at the top.  Half the crosswind impulse can be taken care of with the gimballed engine at the bottom.

Suppose a crosswind of 30 knots = 15 m/s against the 551 m^2 BFR cross-section at sea level.  That's 75 kN of drag.  The top thrusters must cancel at least half of that... a single Kestrel wouldn't quite do it.  Propellant feed rate to the thrusters would be around 12 - 15 kg/s.  A pump that could keep up with that might be 70 kilowatts output and 100 kilowatts electrical, but presumably the duty cycle would be low and a 10 kilowatt pump would do.

[envy887]

BFR/BFS are going to need pressurized GOX and CH4 for more than just the RCS. All the functions that hydraulics, N2 and He perform on F9 and Dragon have to be replaced: Turbine spin-up, main tank pressurization, landing tank pressurization, pneumatic separation pushers, landing leg deployment, grid fin control, etc. Some, like fairing/bay door opening might be converted to electromechanical actuation, but I think most will still be pneumatic.

As for the RCS, I think each cluster of RCS engines will have a local set of high pressure tanks, fed by a high pressure plumbing system running throughout the vehicle much like the current He and N2 plumbing on F9. These will be pressurized in each phase of flight as follows:

1) On the ground, GSE will supply high pressure GOX and CH4 to fill the tanks.

2) During ascent, the Raptors will provide high pressure low volume GOX and CH4 feeding pressurant to the main tanks. This high pressure will keep the RCS tanks topped off if the RCS is used during ascent (e.g. for ullage during separation and restarts).

3) During cruise phase, some tank boiloff will be compressed into the high pressure RCS tanks by electric pumps. If they need a heat source to generate more boiloff, they can run the compressed gas through an intercooler to pull the heat added in compression back out and use it to boil more LOX and LCH4. The same compress+intercool system can also be tied into the ship's radiators for an active zero-boiloff prop management system. This allows boiloff to be controlled to only meet the rate needed for RCS.

4) During descent and landing, the Raptors will again provide high pressure GOX and CH4 to top off the RCS tanks and press the landing tanks.

[Semmel]

A good start envy but cant be the entire story. On descent, the thrusters need to fire quite a lot to keep the correct orientation despite the delta-wing. And that is before the main engines start for the landing burn. The thrust needed for this phase is the highest of all phases. For navigation during cruise phace, not much thrust is needed. During ascent, gimbling takes care of everything. Only during atmospheric reentry and maybe for precision landing, the thrusters need firing extensively. Thats where the highest demand of a heat source for gasifying the Methane and Ox is. Are cooling lines through the heat shield base an option?

[envy887]

A good start envy but cant be the entire story. On descent, the thrusters need to fire quite a lot to keep the correct orientation despite the delta-wing. And that is before the main engines start for the landing burn. The thrust needed for this phase is the highest of all phases. For navigation during cruise phace, not much thrust is needed. During ascent, gimbling takes care of everything. Only during atmospheric reentry and maybe for precision landing, the thrusters need firing extensively. Thats where the highest demand of a heat source for gasifying the Methane and Ox is. Are cooling lines through the heat shield base an option?

Aerosurfaces will probably do most of the control during descent, and I doubt they will gasify prop on demand - though they will have to regen cool the RCS thrusters, that could be a significant source of heat. Maybe someone can run the numbers and see if phase change cooling can both keep the thrusters cool and supply them with enough fuel to keep firing.

Otherwise I imagine they will regen cool with cold GCH4 and just store enough gaseous propellants to complete the entry and landing.