Engine cycles

A liquid rocket engine has to get propellant from a low-pressure tank into a chamber at up to 300 times that pressure, and it has to find the power to do so somewhere. The engine cycle is the answer to that question, and it determines almost everything else about the engine.

The problem

Chamber pressure is worth having. Higher $p_c$ allows a larger expansion ratio within a given nozzle diameter, which raises specific impulse, and it raises thrust per unit chamber area, which shrinks the engine. Modern staged combustion engines run at 20 to 30 MPa.

Feeding a chamber at 30 MPa from tanks is not something tank pressure can do. Tank walls scale with pressure, and a tank pressurised to 30 MPa would weigh more than the propellant inside it. So the engine needs a pump, the pump needs a turbine, and the turbine needs hot gas. Where that gas comes from, and what happens to it afterwards, defines the cycle.

Pressure-fed

No pump at all. Tank pressure alone feeds the chamber, so the tanks are the pressure vessel and chamber pressure is limited to roughly 0.7 to 2 MPa.

The engine becomes extremely simple: no turbopump, no turbine, no gas generator, no rotating machinery whatsoever. With hypergolic propellants there is no igniter either, so a pressure-fed hypergolic engine can be reduced to tanks, valves, an injector and a chamber. Restart is trivial and reliability is exceptional.

The penalty is tank mass, which is unacceptable for a launch stage and perfectly acceptable for an in-space stage where the delta-v is modest. The Apollo Service Propulsion System and the Delta II second stage AJ10 both used this cycle, and essentially all satellite propulsion still does.

Gas generator

Divert 2 to 5% of the propellant into a small gas generator, burn it fuel-rich to keep the temperature within turbine material limits, drive the turbine, and dump the exhaust overboard.

This is an open cycle: the turbine exhaust never reaches the main chamber and never contributes useful specific impulse. The penalty is 2 to 5% of $I_{sp}$, visible as the dark secondary plume alongside the main exhaust.

In exchange the cycle is straightforward. Turbine pressure ratio is set by dumping to ambient rather than by chamber pressure, so pump discharge pressure only has to exceed chamber pressure by the injector drop. Chamber pressures of 7 to 12 MPa are routine. The F-1, Merlin, Vulcain 2 and RS-68 are all gas generator engines, and the cycle remains the default choice when simplicity matters more than the last few seconds of specific impulse.

Staged combustion

Burn all of one propellant plus a portion of the other in a preburner, drive the turbine with the resulting hot gas, then inject that gas into the main chamber where combustion completes. Nothing is discarded.

This is a closed cycle. Every gram of propellant contributes to thrust, so the specific impulse penalty of the gas generator disappears, and chamber pressure can be pushed to 20 to 30 MPa.

The cost is severe. Because the turbine exhaust must be injected into the chamber, the pumps must raise pressure well above chamber pressure to cover both the turbine and injector pressure drops. Pump discharge can exceed 50 MPa. The whole system is a tightly coupled loop in which turbine power depends on chamber pressure and chamber pressure depends on turbine power, which makes startup a carefully choreographed sequence rather than an event.

Oxidiser-rich staged combustion is a Soviet speciality that Western programmes long considered impractical, because hot high-pressure oxygen attacks most metals. The solution was metallurgical: specific alloys and burn-resistant coatings developed over decades. When the RD-180 became available after 1991 it was significantly ahead of anything the United States had flown.

Full-flow staged combustion sends all the fuel through one preburner and all the oxidiser through the other. Both turbines then run on gas that is either entirely fuel-rich or entirely oxidiser-rich, at lower temperature for the same power, and crucially the turbopumps handle only one fluid each, which removes the interpropellant seal that is a classic failure point. Raptor is the first full-flow engine to fly.

Expander cycle

Use the fuel itself as the working fluid. Cryogenic hydrogen or methane is routed through the regenerative cooling jacket, where it absorbs heat from the chamber wall and expands into a gas, drives the turbine, and is then injected into the chamber and burned.

There is no preburner and no combustion outside the main chamber, so turbine inlet temperature is low and the cycle is exceptionally benign. The RL10 has flown since 1963 with an outstanding reliability record.

The limitation is thermodynamic and hard. Available turbine power comes from heat picked up through the chamber wall, which scales with surface area, while thrust scales with chamber volume. As the engine grows, area grows as the square and volume as the cube, so power available per unit thrust falls. This caps closed expander cycles at roughly 300 kN of thrust, which is why every expander engine is an upper-stage engine.

The expander bleed or open expander variant dumps the turbine exhaust overboard instead of burning it. That decouples turbine pressure ratio from chamber pressure and removes the scaling limit, at the cost of a small $I_{sp}$ penalty. Japan’s LE-5A and the much larger LE-9 use it.

Electric pump-fed

Drive the pumps with electric motors powered by batteries. There is no turbine, no gas generator and no hot-gas plumbing at all; the engine is a pump, a motor and a chamber.

The cycle removes the hardest part of engine development and enables very fast throttle response and simple restart. The penalty is battery mass, which is carried the whole way and is dead weight once discharged, so the approach only closes for small vehicles with short burn times. Rocket Lab’s Rutherford is the flight example, with battery packs jettisoned in flight as they deplete.

Tap-off

Bleed hot gas directly from the main combustion chamber to drive the turbine, then dump it. There is no separate preburner or gas generator, which removes hardware, but the turbine must tolerate gas at main chamber conditions and the tap-off point must be positioned where the gas is cool enough to survive. Blue Origin’s BE-3 uses this cycle.

Comparison

References

• Sutton, G. P. and Biblarz, O. Rocket Propulsion Elements, 9th ed., Wiley, 2017, chapter 6.
• Huzel, D. K. and Huang, D. H. Modern Engineering for Design of Liquid-Propellant Rocket Engines, AIAA, 1992.
• Sutton, G. P. History of Liquid Propellant Rocket Engines, AIAA, 2006.
• Sutton, G. P. “History of Liquid-Propellant Rocket Engines in Russia, Formerly the Soviet Union”, Journal of Propulsion and Power, 19:978, 2003. On oxidiser-rich staged combustion.