Regenerative cooling — heat flux, coolant flow, pressure drop: Visual walkthrough

Step 1 — Draw the battlefield: three materials in a row

WHAT. Before any maths, we draw the physical setup. On the left is the roaring combustion gas. In the middle is a thin slab of metal wall. On the right is the cold coolant flowing past. Heat marches left-to-right, from hot to cold.

WHY. You cannot write an equation for heat flow until you know what heat crosses, and in what order. Heat cannot teleport from gas to coolant — it must pass through the wall. That single fact ("in a row, one after another") is the seed of everything.

PICTURE. Look at the figure. Four temperatures are marked, and they only ever drop as you move right (heat always flows downhill in temperature):

$T_{a w}$ — the hot-gas driving temperature (leftmost, hottest).
$T_{w g}$ — the wall temperature on the gas side.
$T_{w c}$ — the wall temperature on the coolant side.
$T_{co}$ — the coolant bulk temperature (rightmost, coldest).

Figure — Regenerative cooling — heat flux, coolant flow, pressure drop

Step 2 — Steady state: the same $q$ everywhere

WHAT. We claim one single number $q$ describes the flow through all three layers — gas film, metal, coolant film — at once.

WHY. Suppose more heat entered the metal on the gas side than left on the coolant side. Then heat would be piling up inside the metal, and the metal would keep getting hotter forever. In a running engine the wall temperature settles down and stops changing — that is steady state. No pile-up means whatever flows in, flows out: the flux is the same through every layer.

PICTURE. Three identical amber arrows, one per layer, all the same thickness — the "thickness" of the arrow is $q$ . If any arrow were fatter or thinner than its neighbour, heat would accumulate at the join. It doesn't, so they match.

Step 3 — Layer 1: heat leaping the gas film (Newton's cooling)

WHAT. We write how much flux crosses the thin gas layer touching the wall.

WHY this tool — Newton's Law of Cooling. A moving fluid does not conduct heat lazily; it sweeps heat onto the surface by convection. The experimental rule for that is Newton's Law of Cooling: flux is proportional to the temperature gap across the film. We use it (not conduction) because the gas is a flowing fluid, not a solid.

PICTURE. A steep temperature cliff sits right at the wall face. The gap from $T_{a w}$ down to $T_{w g}$ is the "voltage" driving heat across the film.

Step 4 — Layer 2: heat crawling through the metal (Fourier's law)

WHAT. Now the flux crosses the solid wall itself.

WHY this tool — Fourier's Law. Inside a solid there is no flow to sweep heat; heat diffuses down the temperature gradient. The rule for that is Fourier's Law of Conduction: flux equals conductivity times the temperature slope. Different physics, different law — that's why we switch tools here.

PICTURE. Inside the metal the temperature falls in a straight line (constant slope) from $T_{w g}$ to $T_{w c}$ across thickness $t_{w}$ . A gentle ramp, not a cliff — metal conducts easily.

Step 5 — Layer 3: heat handed off to the coolant (Newton again)

WHAT. The last leg: flux leaves the cold face of the wall into the flowing coolant.

WHY. The coolant is again a moving fluid, so we use Newton's Law of Cooling once more, with the coolant-side coefficient $h_{c}$ . The value of $h_{c}$ comes from how fast the coolant flows — that's the Dittus-Boelter Correlation — but for this derivation we just need the form.

PICTURE. A second cliff, on the coolant side, dropping from $T_{w c}$ to the coolant bulk $T_{co}$ .

Step 6 — Stack the drops: the telescoping sum

WHAT. We now have three tidy statements, each a temperature drop equal to $q$ times a resistance: $T_{a w} - T_{w g} = q \frac{1}{h _{g}}, T_{w g} - T_{w c} = q \frac{t _{w}}{k _{w}}, T_{w c} - T_{co} = q \frac{1}{h _{c}}$ Add all three left sides and all three right sides.

WHY add them. Watch the middle temperatures cancel — this is called telescoping. On the left, $T_{w g}$ appears once with a $+$ and once with a $-$ ; same for $T_{w c}$ . They vanish, leaving only the outermost temperatures. The unknown wall faces we couldn't measure directly disappear on their own.

$(T_{a w} - T_{w g}) + (T_{w g} - T_{w c}) + (T_{w c} - T_{co}) = q (\frac{1}{h _{g}} + \frac{t _{w}}{k _{w}} + \frac{1}{h _{c}})$

PICTURE. The three drops stacked as a staircase from $T_{a w}$ at the top to $T_{co}$ at the bottom. The total height of the staircase is $T_{a w} - T_{co}$ ; each step's height is $q \times$ that layer's resistance.

Step 7 — The electrical picture: why they add

WHAT. We reveal the analogy that makes the whole thing obvious.

WHY. Heat flux $q$ behaves like electric current; temperature difference behaves like voltage; each layer is a resistor. Resistors in series add, and current through a chain is $I = V / (R_{1} + R_{2} + R_{3})$ — the exact shape of our boxed formula. This is not a coincidence; both obey "flux = (driving difference) / (resistance)".

PICTURE. Three resistors in a line, current $q$ flowing through, voltage source $T_{a w} - T_{co}$ across the ends. The biggest resistor controls the flow.

Step 8 — Degenerate & limiting cases (never leave a gap)

WHAT. We test the formula at its extremes to make sure it never lies.

WHY. A trustworthy formula must give sensible answers when a term goes to zero or infinity. If it blows up nonsensically, we've made an error. Check every corner.

PICTURE. Four dials, each pushed to an extreme, with the resulting behaviour of $q$ .

The one-picture summary

Everything above compressed into a single diagram: the temperature staircase on the left, its identical-twin resistor ladder on the right, and the boxed formula reading straight off both.

Recall Feynman: retell the whole walkthrough in plain words

Heat starts in the fire and wants to reach the cold coolant, but it has to walk through three rooms in a row: a hot-gas "air gap", the metal wall, and a cold "air gap" on the coolant side. Because the engine is running steadily, nothing piles up — the same amount of heat walks through every room per second (that's $q$ ). Each room slows the heat down by a certain amount; we call that slowing a resistance. The hot-gas room and coolant room resist by convection (Newton's cooling, $1/ h$ ); the metal room resists by conduction (Fourier, $t_{w} / k_{w}$ ). The temperature drops a little in each room, and if you add the three little drops the wall-face temperatures cancel out, leaving just "top temperature minus bottom temperature". That total push, divided by the three resistances added up, gives the heat flux — exactly like electric current equals voltage divided by resistors in series. Test it at the edges: no temperature gap → no heat; thick wall → big resistance → hot wall (that's why real walls are thin copper). One formula, three rooms, and it never lies.

Recall Active recall — cover the answers

Why is $q$ the same in all three layers? ::: Steady state — heat can't pile up, so flux in = flux out for every layer. Which law governs the gas and coolant films? ::: Newton's Law of Cooling, $q = h Δ T$ . Which law governs the metal wall? ::: Fourier's Law of Conduction, $q = (k_{w} / t_{w}) Δ T$ . What happens to the wall-face temperatures when you add the three drops? ::: They telescope (cancel), leaving only $T_{a w} - T_{co}$ . What is $U$ ? ::: The overall heat-transfer coefficient, $U = (1/ h_{g} + t_{w} / k_{w} + 1/ h_{c})^{- 1}$ . Why does making the wall thicker make it hotter, not safer? ::: Thicker $t_{w}$ raises $R_{w} = t_{w} / k_{w}$ , increasing the wall's temperature drop and pushing $T_{w g}$ up.