Transpiration cooling: Foundations — Physics deep dive

This page assumes you have seen nothing. We build every letter, ratio, and picture the parent Transpiration cooling note uses, in an order where each block rests on the one before it.

0. The scene before any symbols

Picture a flat slice of rocket wall. On one side: monstrously hot gas rushing past. On the other side: cool fluid we pump in. The wall is riddled with microscopic pores, so the coolant seeps through and mixes into the hot flow.

Figure 1 — The scene. Amber arrows (left) are the hot gas at $T_{g} \approx 3000 K$ streaming past the porous wall. Cyan arrows (right) are the coolant at $T_{c} \approx 300 K$ being pushed through the pores. The short cyan arrows escaping the left face are the "sweat" blanket. The wall itself sits at some middle temperature $T_{w}$ (white label). Read the figure left-to-right: hot in, coolant out, wall in the middle.

Everything else is a name for a quantity in this picture. Let's earn each one.

1. Temperature — the "how hot" number

The parent page uses three different temperatures. Keep them straight — this is the #1 mistake:

Symbol	Plain words	In the picture
$T_{g}$	temperature of the hot gas	the fire on the left, $\sim 3000 K$
$T_{c}$	temperature of the coolant as it enters	the cold fluid on the right, $\sim 300 K$
$T_{w}$	temperature of the wall itself	the metal in the middle

2. Heat flux $q$ — "how much heat, how fast, through how much wall"

Heat is a form of energy, measured in joules ( $J$ ). But we rarely care about total joules; we care about the rate and the area.

Why do we need a flux and not just total heat? Because the throat of a nozzle is small but gets hammered per unit area — the flux is what melts metal, not the total. (This is the same quantity studied in Nozzle Throat Heat Flux.)

The parent uses two versions:

$q_{0}$ — the flux without any coolant blowing (the bare, worst case).
$q$ — the actual flux with coolant blowing (smaller).

3. Convection and the heat-transfer coefficient $h$

How does heat get from the gas into the wall? Not by the gas touching and giving up its jiggle once — the gas keeps flowing, constantly delivering fresh hot particles to the surface. This flowing-fluid heat delivery is called convection (see Convective Heat Transfer).

That proportionality — Newton's law of cooling — comes in two flavours. The bare wall (no coolant blowing) uses its own coefficient $h_{0}$ : $q_{0} = h_{0} (T_{g} - T_{w})$ and the blown wall (coolant sweating) uses a smaller coefficient $h$ for the very same temperature gap: $q = h (T_{g} - T_{w})$

All the messy fluid mechanics of the flow near the wall — speed, turbulence, thickness of the slow layer — get packed into this one number. That layer is the subject of Boundary Layer Theory.

Figure 2 — Why blowing shrinks $h$ . The horizontal axis is distance $y$ out from the wall; the vertical axis is gas temperature. The amber curve (no blowing) rises steeply right at the wall — a steep temperature gradient means heat pours in fast, i.e. a large $h_{0}$ . The cyan curve (with blowing) rises gently, because the coolant blanket has pushed the hot gas away; the shallow wall-gradient means a small $h$ . Look at the two arrows at the wall: the steeper amber one delivers more heat per second. Same $T_{g} - T_{w}$ , smaller coefficient — that is the whole point.

4. The boundary layer — the thin slow film that does the insulating

Why does it matter for heat? Heat has to cross this sluggish layer to reach the metal, and a thick, slow layer is a good insulator. The trick of transpiration cooling is:

5. Mass flow $\overset{m}{˙}_{c}$ and mass flux $G$ — "how much coolant"

Now the coolant side. We need to count how much coolant we push through.

But just like heat, we care per unit area, because a big wall naturally carries more coolant. So we divide by the area:

6. Specific heat $c_{p}$ — "how much heat one kilogram can soak up"

The coolant works by warming up: it enters cold and leaves hot, and the heat it absorbs on the way is what it stole from the wall. How much heat does warming take?

So the heat one kilogram carries away from $T_{c}$ up to $T_{w}$ is $c_{p} (T_{w} - T_{c})$ , and the heat all the coolant carries per second per square metre is: $kg per m^{2} per s G J per kg per K c_{p} K (T_{w} - T_{c}) = W m^{- 2} = a heat flux. ✓$

Notice the units multiply out to a flux — that is your guarantee the equation is honest.

7. The blowing reduction factor $η$ — "how much the blanket helps"

Why a ratio and not a new coefficient from scratch? Because it isolates one clean idea: "how much did blowing thin the heat delivery?" Small $η$ = strong shielding. And crucially, $η$ shrinks as you blow more ( $G$ up) — more coolant, thicker blanket, weaker heat delivery.

8. Two "conductances", the balance, and $T_{w}$

The whole parent derivation is one sentence: in steady state, the heat the gas delivers equals the heat the coolant carries off. Using the blown flux $q = η h_{0} (T_{g} - T_{w})$ on the left (since $h = η h_{0}$ ) and the coolant pickup $G c_{p} (T_{w} - T_{c})$ on the right:

$gas delivers, per m^{2} η h_{0} (T_{g} - T_{w}) = coolant carries off, per m^{2} G c_{p} (T_{w} - T_{c})$

Now solve for $T_{w}$ step by step (nothing but algebra):

$η h_{0} T_{g} - η h_{0} T_{w} = G c_{p} T_{w} - G c_{p} T_{c}$

Move every $T_{w}$ to one side, everything else to the other:

$η h_{0} T_{g} + G c_{p} T_{c} = G c_{p} T_{w} + η h_{0} T_{w} = (η h_{0} + G c_{p}) T_{w}$

Divide by the bracket:

$T_{w} = \frac{η h _{0} T _{g} + G c _{p} T _{c}}{η h _{0} + G c _{p}}$

The two quantities doing the pulling both have units $W m^{- 2} K^{- 1}$ — they are "conductances":

Quantity	Name	Side it belongs to
$η h_{0}$	gas-side conductance	how strongly the gas pushes heat in
$G c_{p}$	coolant-side conductance	how strongly the coolant carries heat out

Figure 3 — The tug-of-war. The horizontal white bar is a temperature scale from $T_{c} = 300 K$ (cyan, left) to $T_{g} = 3000 K$ (amber, right). The white marker is the resulting $T_{w} = 1527 K$ from Example 1 of the parent. The cyan arrow shows the coolant conductance $G c_{p}$ pulling the marker left (cooler); the amber arrow shows the gas conductance $η h_{0}$ pulling it right (hotter). Where they balance is exactly the weighted average from the boxed formula.

Edge cases — pushing the tug-of-war to its limits

Plug extreme values into the boxed formula and watch it behave:

These four corners bracket every real case: $T_{w}$ always lands between $T_{c}$ and $T_{g}$ , never outside.

The cooling effectiveness $ϕ$

Finally, a clean scorecard from 0 to 1:

9. Specific impulse $I_{s p}$ — why coolant is not free

Every kilogram of coolant you sweat out is a kilogram you could have used for thrust. The efficiency of turning propellant into thrust is measured by Specific Impulse ( $I_{s p}$ ). This is why the parent warns against blindly maximizing $G$ : cooling costs performance.

The prerequisite map

The full picture, assembled

The single equation the parent derives just says: the heat the gas delivers (left) equals the heat the coolant carries away (right), and solving for $T_{w}$ (done in §8) gives the tug-of-war average. $gas delivers η h_{0} (T_{g} - T_{w}) = coolant carries off G c_{p} (T_{w} - T_{c})$

Related cooling cousins you can now compare: Film Cooling, Regenerative Cooling, Ablative Cooling.

Quick recall

Recall What are the three temperatures and their order?

$T_{c} \leq T_{w} \leq T_{g}$ : coolant cold, wall in between, gas hottest.

Recall Which direction is positive

$q$ , and why does it stay positive here? Positive = heat flowing from gas into wall; it stays positive because $T_{g} > T_{w}$ always.

Recall Why divide by area to get flux

$q$ and mass flux $G$ ? Because melting and cooling happen per unit area — a small hot throat can have a huge flux even with modest total heat.

Recall What does

$η = 0.4$ physically mean, and when does $η = q / q_{0}$ ? The blown coefficient is $40%$ of the bare one ( $h = 0.4 h_{0}$ ); $η = q / q_{0}$ only holds if $Δ T$ is identical in both cases.

Recall What does

$T_{w}$ do as $G \to 0$ and $G \to \infty$ ? $G \to 0$ : $T_{w} \to T_{g}$ (no cooling). $G \to \infty$ : $T_{w} \to T_{c}$ (perfect cooling).

Recall What does

$ϕ = 1$ mean and why is it impossible? Wall as cold as the entering coolant — perfect cooling, which needs infinite coolant.

Equipment checklist

Cover the right side and answer aloud. If you can, you are ready for the parent derivation.

I can convert 3000 K into "how hot is that vs a melting metal?"

Yes — kelvin is a straight scale from absolute zero; 3000 K is far above any alloy's melting point (~1900 K), so no material survives unaided.

I can state the units of heat flux

q

and its sign convention

W m^{- 2}

; positive means heat flowing from gas into wall.

I can write both convection laws

q_{0} = h_{0} (T_{g} - T_{w})

and

q = h (T_{g} - T_{w})

Same temperature gap, different coefficient;

h < h_{0}

after blowing.

I can explain the boundary layer and how blowing changes it

The thin slowed layer of gas at the wall; coolant thickens and cools it, so the wall gradient (and

h

) drops.

I can compute mass flux

G

from

\overset{m}{˙}_{c}

and

A

G = \overset{m}{˙}_{c} / A

, in

kg m^{- 2} s^{- 1}

I can explain what

c_{p}

measures

Joules to raise one kilogram of coolant by one kelvin — its heat-sponge capacity.

I can define

η

and say when

η = q / q_{0}

holds

η = h / h_{0}

; equals

q / q_{0}

only when

Δ T

is the same in blown and un-blown cases.

I can derive

T_{w}

from the energy balance

Set

η h_{0} (T_{g} - T_{w}) = G c_{p} (T_{w} - T_{c})

, collect

T_{w}

, get the weighted-average box.

I can state the two limiting cases of

T_{w}

G \to 0

gives

T_{g}

;

G \to \infty

gives

T_{c}

I can say why more coolant is not free

Coolant is propellant mass lost, lowering specific impulse

I_{s p}