Expander cycle — hydrogen-cooled nozzle drives turbine

WHY does this cycle exist?

WHAT problem are we solving? A rocket engine needs to force propellant into the combustion chamber against enormous chamber pressure. This needs pumps. Pumps need a turbine to drive them. A turbine needs a hot, high-pressure working gas.

Where do you get that gas? Two bad options and one clever one:

Gas generator cycle: burn a little propellant in a side chamber to make turbine gas, then dump it overboard → wasted propellant, lower efficiency.
Staged combustion: burn a fuel-rich preburner → very high performance but brutal turbine temperatures and pressures → complex, expensive.
Expander cycle (this note): don't burn anything extra. The nozzle walls are already blazing hot and must be cooled anyway. So route the fuel through cooling channels; the heat it steals for free boils it and pressurizes it enough to run the turbine.

HOW the loop works (trace the hydrogen)

Liquid H₂ leaves the tank → enters the fuel pump (pressure rises).
It flows into regenerative cooling channels wrapped around the nozzle/chamber. Here it absorbs heat $Q$ from the combustion, turning into hot high-pressure gas. This cools the nozzle so it doesn't melt.
The hot H₂ gas expands through the turbine, dropping in pressure and giving up work $W_{turb}$ .
That work spins the shaft → drives the fuel pump and oxidizer pump.
The now-lower-pressure H₂ (still gas) is injected into the combustion chamber and burned with oxygen.

Figure — Expander cycle — hydrogen-cooled nozzle drives turbine

Deriving the power balance from first principles

Step 1 — Heat picked up in the cooling jacket

The hydrogen enters the channels at temperature $T_{in}$ and leaves at $T_{out}$ . From the definition of specific heat (energy to raise temperature), $Q = \dot m_{H_2}\, c_p\,(T_{out}-T_{in})$

Why this step? $c_p$ is defined as heat per unit mass per kelvin at constant pressure. Multiply by mass flow $\dot m$ and temperature rise → heat rate absorbed. This $Q$ is exactly the heat that would otherwise melt the wall — we're recycling it.

Step 2 — Available enthalpy = turbine's fuel

That absorbed heat raises the gas enthalpy. The turbine converts enthalpy drop into shaft work. For an ideal (isentropic) turbine expanding from pressure $p_1$ to $p_2$ : $W_{turb} = \dot m_{H_2}\, c_p\, T_1\left[1-\left(\frac{p_2}{p_1}\right)^{\frac{\gamma-1}{\gamma}}\right]\eta_t$

Why this step? For an ideal gas, enthalpy $h = c_p T$ . Isentropic expansion links temperature ratio to pressure ratio via $T_2/T_1 = (p_2/p_1)^{(\gamma-1)/\gamma}$ . The work is the enthalpy drop $\dot m c_p (T_1 - T_2)$ ; factor out $T_1$ and multiply by turbine efficiency $\eta_t$ .

Step 3 — Pump power required

A pump raises pressure by $\Delta p$ on a volume flow $\dot m/\rho$ : $W_{pump}=\frac{\dot m\,\Delta p}{\rho\,\eta_p}$

Why this step? Work per unit volume to push fluid up a pressure step is $\Delta p$ (from $W=\int p\,dV$ turned around). Volume flow is $\dot m/\rho$ . Divide by pump efficiency $\eta_p$ because real pumps waste some input.

Step 4 — The self-sustaining condition

$\boxed{\,W_{turb}\;\ge\;W_{pump,fuel}+W_{pump,ox}\,}$

The square-cube limit (WHY expanders can't be huge)

Worked Examples

Worked example Example 2 — Ideal turbine work

The 3 kg/s hot H₂ at $T_1=400\ \text{K}$ expands from $p_1=12\ \text{MPa}$ to $p_2=8\ \text{MPa}$ . Take $\gamma=1.4$ , $\eta_t=0.75$ . Find turbine power.

Pressure ratio term: $(8/12)^{(0.4/1.4)} = (0.667)^{0.286}$ . $\ln(0.667)=-0.405$ , times $0.286 = -0.116$ , $e^{-0.116}=0.890$ . $W = 3\times 14000\times 400\times(1-0.890)\times0.75 \approx 1.39\times10^{6}\ \text{W}=1.39\ \text{MW}$

Why this step? We used $h=c_pT$ and the isentropic $T$ – $p$ relation. Note the turbine uses only a modest pressure drop: expanders keep pressure high because the gas must still enter the chamber afterward.

Worked example Example 3 — Does it close?

Pumps need to raise 3 kg/s H₂ ( $\rho=71\ \text{kg/m}^3$ ) by $\Delta p=13\ \text{MPa}$ and 18 kg/s O₂ ( $\rho=1140\ \text{kg/m}^3$ ) by $\Delta p=13\ \text{MPa}$ , $\eta_p=0.7$ .

Fuel pump: $\dfrac{3\times13\times10^6}{71\times0.7}=7.8\times10^{5}\ \text{W}$ Ox pump: $\dfrac{18\times13\times10^6}{1140\times0.7}=2.9\times10^{5}\ \text{W}$ Total pump demand $\approx 1.07\ \text{MW}$ .

Turbine gave 1.39 MW $>$ 1.07 MW ✅ → the cycle sustains itself with margin to spare (extra used for startup transients / throttle).

Why this step? This is Step 4 in action — the entire cycle "works" only when this inequality holds. Note the ox pump is cheaper per kg because O₂ is dense (low volume flow).

Common Mistakes (steel-manned)

Flashcards

What working fluid drives the turbine in a hydrogen expander cycle, and where does its energy come from?

Gaseous hydrogen; energy comes from heat absorbed through the nozzle/chamber walls (regenerative cooling), not from any combustion.

Difference between closed and open (bleed) expander cycle?

Closed sends all heated H₂ through the turbine then into the chamber (nothing dumped); open dumps a fraction of turbine flow overboard for more power but some loss.

Why can't expander cycles scale to very large thrust?

Heat available scales with wall area (

\propto L^2

) but pump power demand scales with propellant flow (

\propto L^3

); demand outruns supply — the square-cube law.

Formula for heat absorbed in the cooling jacket?

Q=\dot m\,c_p\,(T_{out}-T_{in})

Why hydrogen and not kerosene as coolant/turbine gas?

Hydrogen has very high

c_p

(~14 kJ/kg·K), boils cleanly, and won't coke; kerosene carbonizes in channels and stores too little heat.

Ideal isentropic turbine work formula?

W=\dot m c_p T_1[1-(p_2/p_1)^{(\gamma-1)/\gamma}]\eta_t

Self-sustaining condition for the cycle?

W_{turb}\ge W_{pump,fuel}+W_{pump,ox}

Why can't the turbine expand hydrogen to near-zero pressure for max work?

The exhaust must still be injected into the chamber at ≥ chamber pressure, so downstream pressure

p_2

is bounded from below.

Pump power formula?

W_{pump}=\dot m\,\Delta p/(\rho\,\eta_p)

Name a real closed expander engine.

RL10 (Centaur/DCSS upper stage).

Recall Feynman: explain to a 12-year-old

Imagine a rocket where the engine's bottom part (the bell) gets super hot. Instead of letting it melt, you run cold liquid hydrogen — the fuel — through tiny pipes wrapped around it, like water in the cooling coils of a fridge. The hydrogen soaks up all that heat and turns into a hot, fast gas. That gas then blows through a little pinwheel (a turbine) and spins it. The spinning pinwheel runs the pumps that push more fuel in. So the engine cools itself AND powers its own pumps using the same heat — heat that would otherwise be wasted. After spinning the pinwheel, the hydrogen goes into the fire and gets burned for real. Clever! But it only works for small-ish engines, because a giant engine needs way more pumping than its hot walls can pay for.

Connections

Regenerative cooling — the heat-transfer mechanism that feeds this cycle.
Gas generator cycle — contrast: burns extra propellant, dumps turbine exhaust.
Staged combustion cycle — contrast: fuel-rich preburner, highest performance.
Turbopump design — the load the turbine must drive.
Specific impulse — closed expander gains Isp by wasting no propellant.
Isentropic flow relations — source of the $T$ – $p$ turbine equation.
Square-cube law — the geometric scaling that caps expander thrust.
Liquid hydrogen properties — why $c_p$ and boiling point matter.

Concept Map

Hinglish (regional understanding)

Intuition Hinglish mein samjho

Dekho, expander cycle ka core idea bahut elegant hai. Rocket ka nozzle jab jal raha hota hai to uski walls itni garam ho jaati hain ki melt ho sakti hain. Toh hum kya karte hain? Cold liquid hydrogen (fuel) ko nozzle ki walls ke around chhote channels mein daalte hain — isse wall thanda ho jaati hai (regenerative cooling), aur hydrogen us heat ko soak karke ek hot high-pressure gas ban jaata hai. Yeh free heat hai bhai — waise bhi wall ko cool karna hi tha!

Ab yeh garam hydrogen gas turbine ko spin karta hai. Turbine shaft se fuel pump aur oxidizer pump chalte hain. Matlab engine apni hi waste heat use karke apne pumps chala raha hai — koi extra fuel jalaye bina. Yahi expander cycle ki khoobsurti hai: gas generator cycle mein thoda propellant side mein jalाना padta hai aur exhaust bahar phenkna padta hai (waste), lekin closed expander mein kuch waste nahi hota, isliye specific impulse zyada milta hai. Turbine se nikalne ke baad hydrogen chamber mein jaakar oxygen ke saath jal jaata hai.

Hydrogen hi kyun? Kyunki uska $c_p$ (specific heat) bahut bada hai (~14 kJ/kg·K), thoda sa mass bhi bahut heat store kar leta hai, aur woh cleanly boil hota hai — kerosene channels mein coke (kaala carbon) bana ke jam जाता hai.

Ek important limitation yaad rakhna: expander cycle bade engines ke liye kaam nahi karta. Heat toh nozzle ki surface area se milti hai ( $L^2$ ), lekin pumps ko jitna propellant push karna padta hai woh volume/flow ke saath badhta hai ( $L^3$ ). Toh engine bada karo to demand supply se aage nikal jaati hai — yeh square-cube law hai. Isiliye RL10 jaise expander engines chhote-medium upper stage engines hote hain. Self-sustain condition simple hai: turbine ka power $\ge$ dono pumps

Expander cycle — hydrogen-cooled nozzle drives turbine

WHY does this cycle exist?

HOW the loop works (trace the hydrogen)

Deriving the power balance from first principles

Step 1 — Heat picked up in the cooling jacket

Step 2 — Available enthalpy = turbine's fuel

Step 3 — Pump power required

Step 4 — The self-sustaining condition

The square-cube limit (WHY expanders can't be huge)

Worked Examples

Common Mistakes (steel-manned)

Flashcards

Connections

Concept Map

Hinglish (regional understanding)

Go deeper — visual, from zero

Test yourself — Rocket Propulsion

Connections