Starship Methox RCS Thrusters

[Semmel]

I have not read or heard a lot about the steering thrusters of BFR. They are a new type of engine in their own right with its own development project, control and fuel supply. The list below is from memory and I would love to update them to something more substantial.

What we know:
* Methane/Oxygen fuel
* Pressure fed

What we assume:
* gas-gas combustion
* max thrust >50kN

What we dont know:
* min thrust
* ISP
* mass
* reaction time (important since they should be used for attitude control)
* control precision
* how the liquid CH4/O2 is converted to gas (preburner would not work)
* Number and location on BFR

This thread is intended to be an analogue to ITS Propulsion – The evolution of the SpaceX Raptor engine and I would hope for a technical discussion on the engines and their plumbing.

[speedevil]

I have not read or heard a lot about the steering thrusters of BFR. They are a new type of engine in their own right with its own development project, control and fuel supply. The list below is from memory and I would love to update them to something more substantial.

What we know:
* Methane/Oxygen fuel

What we assume:
* Pressure fed
* gas-gas combustion

What we dont know:
* max/min thrust
* ISP
* mass
* reaction time (important since they should be used for attitude control)

We know some of these.
It's been stated landing in 60KM/H winds is OK. Elon on twitter

This allows you to come up with a figure close to 15 tons/axis for BFS, somewhat more for BFR (to cancel the wind velocity and land vertically with no gimbal)

This more or less works with the observed thruster diameters on IAC2017 model - I have not gone over the 2018 images looking for them.

Min thrust of a gas thruster can be very low, if you operate it in cold gas mode, plus, the whole reason for going gas/gas was 'minimum impulse bit'. Elon on Reddit

If the thrusters are a little more powerful than this, 30 tons - they enable lunar landing and takeoff with enough fuel to ascend (50t) and a minimal payload, and perhaps even land horizontally.

Have they thought of this and realised it would be handy - who knows.
Oversizing them this way from the 15 tons calculated above for 60km/h to 60MPH  winds would considerably improve ability to ignore most winds.

[Slarty1080]

We know some of these.
It's been stated landing in 60KM/H winds is OK. Elon on twitter

This allows you to come up with a figure close to 15 tons/axis for BFS, somewhat more for BFR (to cancel the wind velocity and land vertically with no gimbal)

This more or less works with the observed thruster diameters on IAC2017 model - I have not gone over the 2018 images looking for them.

Min thrust of a gas thruster can be very low, if you operate it in cold gas mode, plus, the whole reason for going gas/gas was 'minimum impulse bit'. Elon on Reddit

If the thrusters are a little more powerful than this, 30 tons - they enable lunar landing and takeoff with enough fuel to ascend (50t) and a minimal payload, and perhaps even land horizontally.

Have they thought of this and realised it would be handy - who knows.
Oversizing them this way from the 15 tons calculated above for 60km/h to 60MPH  winds would considerably improve ability to ignore most winds.

Another benefit of more powerful thrusters higher up on the BFR would be to reduce the amount of debris that will be blasted into the surroundings and the engines during take-off and landing.

[Semmel]

I have not read or heard a lot about the steering thrusters of BFR. They are a new type of engine in their own right with its own development project, control and fuel supply. The list below is from memory and I would love to update them to something more substantial.

What we know:
* Methane/Oxygen fuel

What we assume:
* Pressure fed
* gas-gas combustion

What we dont know:
* max/min thrust
* ISP
* mass
* reaction time (important since they should be used for attitude control)

We know some of these.
It's been stated landing in 60KM/H winds is OK. Elon on twitter

This allows you to come up with a figure close to 15 tons/axis for BFS, somewhat more for BFR (to cancel the wind velocity and land vertically with no gimbal)

This more or less works with the observed thruster diameters on IAC2017 model - I have not gone over the 2018 images looking for them.

Min thrust of a gas thruster can be very low, if you operate it in cold gas mode, plus, the whole reason for going gas/gas was 'minimum impulse bit'. Elon on Reddit

If the thrusters are a little more powerful than this, 30 tons - they enable lunar landing and takeoff with enough fuel to ascend (50t) and a minimal payload, and perhaps even land horizontally.

Have they thought of this and realised it would be handy - who knows.
Oversizing them this way from the 15 tons calculated above for 60km/h to 60MPH  winds would considerably improve ability to ignore most winds.

Ahh nice piece of information. Assuming the 15 tons (say, 150kN) is emitted at both ends of the stage, which is logical to be able to allign the centre of thrust with the centre of mass. Then you need at least 2 thrusters to have a combined thrust of 150kN. They will not be exactly equal since running the system without contingency is dangerous. So I would guess that the system would be able to give at least 200kN and is distributed over 4 thrusters, two at the top and 2 at the bottom. Probably more but this gives an upper bound for the max thrust of 50 kN.

[DAZ]

Reading to this thread brings up a question. These gas- gas thrusters would seem to be relatively easy to build. So my question is, could you modify the Raptor engines to have a gas-gas mode and would it be useful? I know the efficiencies would be much lower, what I'm thinking of is giving you more finesse / control especially in lower gravity Landings.

[speedevil]

Reading to this thread brings up a question. These gas- gas thrusters would seem to be relatively easy to build. So my question is, could you modify the Raptor engines to have a gas-gas mode and would it be useful? I know the efficiencies would be much lower, what I'm thinking of is giving you more finesse / control especially in lower gravity Landings.

You would need additional gas injection systems into the raptors, valving, distribution pipework, redesigned cooling, more high pressure gas storage, and ...
It is not in principle impossible, but it would add considerable complexity and probably weight, for not very much obvious benefit.

If you really, really wanted to, you could shove a few RCS class thrust pods in the cargo bay area, but there is little point - as the existing ones all over the vehicle work pretty much the same.

[Semmel]

As far as I know, raptor already operates in gas gas mode. But the problem is the turbo pumps and preburners, so they are made for high sustainable thrust. Not precise thrust control and little control puffs. Adding gas gas injection side tracking the turbo machinery has little benefit. For steering its useless, because it's at the end of the rocket with little to no Leaver arm. For acceleration its too small thrust. There is really no reason to do this.

[docmordrid]

Elon Musk, Reddit AMA

https://www.reddit.com/r/space/comments/76e79c/i_am_elon_musk_ask_me_anything_about_bfr/

The BFS will have methalox RCS thrusters for spaceship attitude control. (See the three dark dots at the bottom of the spaceship.)
>
The control thrusters will be closer in design to the Raptor main chamber than SuperDraco and will be pressure-fed to enable lowest possible impulse bit (no turbopump spin delay).

[Semmel]

Elon Musk, Reddit AMA

https://www.reddit.com/r/space/comments/76e79c/i_am_elon_musk_ask_me_anything_about_bfr/

The BFS will have methalox RCS thrusters for spaceship attitude control. (See the three dark dots at the bottom of the spaceship.)
>
The control thrusters will be closer in design to the Raptor main chamber than SuperDraco and will be pressure-fed to enable lowest possible impulse bit (no turbopump spin delay).

Thank you doc, I updated the opening post.

[DigitalMan]

Since there is going to be another AMA in a week or two has anyone thought about getting questions together on these (and other) issues to see if some of them could be answered?

There are a lot of great questions posed by folks around here.

[speedevil]

What can we say about the methalox RCS for SS (and possibly SH)?

At the beginning of time,Elon stated of the RCS.

The control thrusters will be closer in design to the Raptor main chamber than SuperDraco and will be pressure-fed to enable lowest possible impulse bit (no turbopump spin delay).

We can estimate the thrust with the further tweet

Yes. All-weather. ~300km/h high altitude winds. ~60km/h ground winds. It’s a beast.

Assuming the drag coefficient of SS worst-case sideways is 1, 60km/h, and 450m^2, this comes out to around 10 tons drag.

It seems likely that if you were to want to land in this crosswind, you would want a RCS capable of at least 20 tons per axis, or at most 10 tons per cluster of RCS jets.

At an exhaust expansion of around 2 bar, this puts the approximate diameter of a nozzle at 0.5m^2 - some 80cm. (seperate nozzles needed for the three or four axes per cluster)

Assuming you want to be able to brute force an unexpected wind throughout the last 5 seconds of landing and come to a pinpoint landing - with an ISP of 330, and 10000kg thrust, this comes out to 5s*30kg/s of propellant - 150kg.

The tank volume at 4500PSI would be around a cubic meter total, though this would need to probably be at each cluster of nozzles.
In addition, you would need a high pressure gas feed line (to slowly refill over minutes).

Not igniting the gas gives you an ISP of around 70, not 330, so you have a thrust of around 2 tons.
Allowing very small flows would be simple, even in the absence of vernier thrusters.

Composite pressure vessels would bring the mass of each cluster (with propellant) to at least 300kg.

How the pressure vessels are filled is an interesting question - in principle there are many ways:
Cryogenic liquid reservoir and a rapid heating to go from ~50PSI liquid to ~4000PSI gas - possibly in a multichamber device to eliminate liquid pumping.
Compressor running from the tank gas volume.
...
With the revelation this is coming pretty soon 'Mk4' or so - it will be interesting to see the development.

[sanman]

Doesn't RCS particularly need good response time and precision?

How does Methalox measure up for these, compared to cold gas thrusters?

[TheRadicalModerate]

Are you sure the low-altitude 60 km/h is the limiting factor, and not the high-altitude 300 km/h?

Just for some starting numbers, my cheesy model for the F9 gives:

Altitude: 16.7 km
Airspeed: 540 m/s
max-q: 23.4 kPa
max drag force: 165.6 kN

(Anybody know what the average numbers really are?)

A sudden perpendicular change of 300 km/h (83 m/s) generates an effective AoA of arctan(83/540) = 8.7 deg.

This is about where my patience (and competence) for working out even the induced drag, to say nothing of the lift, fails me.  But 8.7 deg seems like a lot--likely more than simple gimbaling is going to be able to correct for.  Seems like some thruster support will be necessary.

[Norm38]

There already is a gas generator system, autogenius pressurization if the main tanks.
What pressure is that, and can it be tapped to fuel the thrusters?

[ThomasGadd]

One reason to consider the lower altitude a limiting is you have less time to react. 

[speedevil]

There already is a gas generator system, autogenius pressurization if the main tanks.
What pressure is that, and can it be tapped to fuel the thrusters?

~3 bar - 45PSI.
So probably not very useful.

[speedevil]

One reason to consider the lower altitude a limiting is you have less time to react.

I was assuming it in the original post, as there is a _long_ phase from 20km down where you can correct for errors over a long period, and then be prepared to light the main engine with some error that needs correcting.
In the last five seconds of landing, you have limited if any capability to cope with gusts without moving sideways or landing at an angle - without a very capable RCS.

[gin455res]

There already is a gas generator system, autogenius pressurization if the main tanks.
What pressure is that, and can it be tapped to fuel the thrusters?

~3 bar - 45PSI.
So probably not very useful.

This is the tank pressure, what about the pressure in the system that is pressurizing the tank?
Might this be used to pressurize a seperate tank to a higher pressure?

[livingjw]

Copied this from General Engineering thread:

- From inonepiece: Elon said that at around mark3/mark4 they'd likely replace starship cold gas thrusters with methalox methox driven from high pressure CH4/O2 cylinders.  Does "high pressure" mean COPVs?  I was looking forward to a non-COPV future for SpaceX since they've been responsible for several major problems...

My guess would be somewhere around 500-1000 psi. Reasonable pressure fed rocket pressure, but pressure used will depend on optimizing the RCS/OMS system including, in particular, its recharge subsystem.

For Moon and Mars launches, must have a recharge subsystem which takes cryo propellants from header tanks and pumps, gasifies and transfers propellants to high pressure gaseous RCS/OMS tanks. This will probably be done using electric pumps and heaters. You definitely want to pump liquids, not gases.

Years ago my team worked on a rechargeable gaseous RCS system. It had 1000 psi tanks of gaseous Lox GOX and CH4. As the gases were used the pressure and temperature of the gases go down.  When the tank pressure dropped below a set limit, an electric pump and heater were used to to pump liquid and gasify propellants from the main tanks into the RCS tanks. As long as you had liquid propellants and electrical power, you could maintain your high pressure gaseous system. It could even be completely offline for years and be recharged before launch from Mars or the Moon.

RCS and OMS are probably going to both be gaseous CH4/GOX because they need to both operate in zero g, and operate instantaneously. It will probably be an integrated RCS/OMS system.

I think they will make use of a lot of the tech they developed for draco/superdraco. Very similar requirements from a pressurization, plumbing perspective. Biggest difference is the need for fail safe ignition, since CH4/GOX is not hypergolic. I would expect to see redundant ignition systems in each rocket as well as redundant rockets. 2 OMS, ~12 RCS.

John

[Barley]

Is oxygen gas now considered lox?