Water injection into SpaceX raptor engine during first 10s

Water injection could increase jet engine thrust by as much as 64% by increasing mass flow and somehow decreasing exhaust speed end temperature. All those effects should be very welcomed for initial 10s of Starship launch, till it clears the tower. How feasible is a contraption that could inject water spray at 350bars pressure into external ring of 20 fixed raptor engines nozzles that could travel with Starship till the height of the launch tower, and then it would detach ? In this idea, the raptors will not need to be modified, and the additional weight of water will not add to the Starship weight since it will be supported from the launch tower. Admittedly it would be a very complex proposition. At the 10s mark booster fuel level dropped by more then 10%, if water injection could save 50% of that fuel gain will be significant.

alas666 said:

Water injection could increase jet engine thrust by as much as 64%

That's quite a claim. How come everybody is not doing this right now? Seems like a no-brainer...

Also, please always post links to supporting references when you make claims like this on PF. Thank you.

Simplicity suggests that, at the moment when you need the most reliable engines, you should not change the steady-state mode of operation.

If it was successful, the 10 seconds could fall to 5 seconds, and it would be unnecessary.

alas666 said:

Water injection could increase jet engine thrust by as much as 64% by increasing mass flow and somehow decreasing exhaust speed end temperature. All those effects should be very welcomed for initial 10s of Starship launch, till it clears the tower. How feasible is a contraption that could inject water spray at 350bars pressure

What is the basis of the claim if 64% increase in jet (rocket) engine thrust? What is the specific impulse of one's concept?

I would suggest one look at the properties of water (supercritical, or superfluid conditions) at 350 bar (and at temperature). Water is very corrosion/erosive at hot supercritical conditions. How much water must be stored? What is the impact of that mass (water and storage system)?

One has to bring the water up to temperature. Consider the enthalpy of room temperature water and what impact that has on the propulsion system.

alas666 said:

Water injection could increase jet engine thrust by as much as 64% by increasing mass flow and somehow decreasing exhaust speed end temperature.

I don't know whether this was tested for rocket engines, but indeed, you might get more impulse with the same energy by accelerating more mass to lower speed.

The problem is, that in case of rockets you brings the fuel with you, and the rocket equation can make even small ballast into big problem. So the choice is:
- you bring only fuel (energy)
- you bring water instead of ? amount of fuel (energy).

I won't say that no way for it to work, but you need to do the math.

berkeman said:

That's quite a claim. How come everybody is not doing this right now? Seems like a no-brainer...

Also, please always post links to supporting references when you make claims like this on PF. Thank you.

In turbojet engines water injection permits large fuel increase to gain extra thrust by lowering temperature and preserving engine from damage. See more discussion here: aviation. The gain of thrust is paid by much lower overall efficiency as 4 times more fuel is burned to gain 64% thrust.

For the water ingested rocket takeoff water could not be stored on board, but could be supplied from launch tower up to the clearing point. SpaceX have invested in deluge system that protects pad and instead of spraying bottom of the tower I propose to attach water spray directly into the raptor nozzle where produced steam could be converted to extra thrust and protect the launch facility at the same time. And yes, it would be a difficult contraption, most likely unfeasible design, since it would need to move up with the Starship and then detach.

alas666 said:

The gain of thrust is paid by much lower overall efficiency as 4 times more fuel is burned to gain 64% thrust.

An aircraft takeoff, from a limited length runway, may benefit from 64% increased thrust, and ignore the extra fuel used.
I cannot see how a rocket can benefit from 64% more thrust, if that requires burning 4 times the mass of fuel. There is no hurry, or need to clear the tower four seconds faster.

alas666 said:

In turbojet engines water injection permits large fuel increase to gain extra thrust by lowering temperature and preserving engine from damage. See more discussion here: aviation.

A link to Stackexchange is not a valid reference at PF. Please see the Rules link (under INFO at the top of the page) for a list of valid peer-reviewed journals.

Rocket engines are not jet engines, that's the first issue. You might get a somewhat higher thrust from injecting water, but it's not worth it.

* You can't carry the water with you - that would make things worse.
* The quick disconnect is designed to disconnect at takeoff, you can't have water transfer while the rocket is already ascending. If you redesign the quick disconnect to stay attached significantly longer, why would you use it for water? Replenish the methane and oxygen.
* The booster quick disconnect can pump something like a 1 tonnes of propellant per second but the outer booster engines burn 13 tonnes per second. You want to add ~50 tonnes of water per second which also have to flow through the same connection that handles 1 tonne/s now?
* A water injection system would make the engine far more complex. That's a larger failure risk, and it's extra weight that you need to carry all the way through the booster flight.

All this is optimistically assuming the engines can handle the water injection at all, and I'm not sure if that is the case.

I did some quick calculations, assuming 650 kg/s of MO is supplemented with H20 up to 1400 kg/s
Looks like 40% of thrust is possible with 2:1 water to MO ratio.
Unlikely such a gain is possible without Raptor redesign given 3 times mass flow increase.

alas666 said:

Looks like 40% of thrust is possible with 2:1 water to MO ratio.

What does that mean? Only 40%, or a 40% boost?

Your first graph does not show the reference thrust, with fuel instead of water.

Your second graph has no units on the axes.
Numbers without units are meaningless.
What is Ve ?

Baluncore said:

Your first graph does not show the reference thrust, with fuel instead of water.

The x axis goes to 0, i.e. no water. Using that as reference OP calculated 40% higher thrust values when dumping a lot of water into the engine.

Baluncore said:

Your second graph has no units on the axes.

Propellant flow in kg/s as given, v_e is the exhaust velocity in m/s. Not sure about the temperature. Fahrenheit in the combustion chamber would make the no water temperature plausible, but why would you use that unit?

alas666 said:

I did some quick calculations

Most idea-owners here never does that, so:


Even if some fine tuning may be needed, with that attitude you are in a club with a few members only...

Next is, to calculate the time needed to clear the tower with increased thrust, and then the LEO capacity change that short time may bring.

Honestly, I expect that to conclude the thing but we'll see.

mfb said:

Propellant flow in kg/s as given, v_e is the exhaust velocity in m/s. Not sure about the temperature. Fahrenheit in the combustion chamber would make the no water temperature plausible, but why would you use that unit?

Temperature is in K. I found 5000 estimate, but used slightly different numbers to get me 230t thrust to match initial Raptor spec.

Baluncore said:

Your second graph has no units on the axes.
Numbers without units are meaningless.

propellant rate [kg/s], Ve [m/s], Temp [K]

Rive said:

Next is, to calculate the time needed to clear the tower with increased thrust, and then the LEO capacity change that short time may bring.

I checked launch video again and in first 3 sec there is no motion, so time to clear 146m tower is about 7s. with thrust at 40% on outer 20 engines I see following gains: time 1.26s sooner, speed 14m/s faster, mass 37t heavier. That will result in 36t more cargo at staging 150s later, which amount to ~3% of fully loaded Starship spacecraft. I know, not much enticing.

Just to compare SpeaceX deluge system is using up to 1355t of water, over maybe 15 sec. The proposed water injection rate would amount to 30% of that.

alas666 said:

I know, not much enticing.

Actually, not that bad for a sudden pet idea.

Compare to just elevating the launch pad by 1 tower height? Hint: Dirt's cheap (and they don't bother).