3D Popcorn Popping: Heat, Steam Pressure and the Bursting Hull

Kernels in a pot on a hot coil: heat transfer, steam pressure, hull rupture and the jump, in a Blender-built kitchen.

Switch to 2D

Presets

Playback

Parameters

Camera

Layers

Quality

Model check

The science

What is modelled

Each kernel is a small sphere of starch with about 14 % water, sealed inside a hard hull. The pan heats it; the water inside turns into superheated steam and the pressure rises along the saturation curve. When the pressure reaches what the hull can hold (about 0.9 MPa, around 175-180 C) the hull ruptures, the starch foam flashes out and sets as the white puff, and the kernel jumps. Every kernel in the pot has its own size, moisture and hull strength, drawn from normal distributions with a fixed seed, so the pops spread out over time exactly as in a real pan. The kitchen, pot, lid, burner coil, kernels and popped corn are Blender-built models; the pan temperature drives the glow of the coil.

Equations

  • Heating of a kernel (sphere of radius R, density ρ = 1300 kg/m³, specific heat c = (1 − w) 1500 + w 4180 J/kg K, conductivity k = 0.2 W/m K): τ dT/dt = Tpan(t) − T, with the Biot-corrected lumped time constant τ = ρ c R/(3h) (1 + Bi/5) and Bi = hR/k.
  • Pan warm-up: Tpan(t) = T∞ − (T∞ − 20) e−t/τpan. For a constant pan temperature the pop time of a kernel is t = τ ln[(T∞ − 20)/(T∞ − Tpop)].
  • Steam pressure (Antoine, T in C, P in mmHg): log10 P = 8.14019 − 1810.94/(T + 244.485). The pop temperature of a kernel is the T at which P equals its burst pressure.
  • Old maids: if the moisture is below about 11.5 % the hull leaks steam and the pressure never builds up; with moisture w ~ N(μ, σ) the expected share is Φ((11.5 − μ)/σ).
  • Flight of a popped kernel: dv/dt = −g − (½ ρair Cd A / m) |v| v with Cd = 1.2; without drag it would rise v²/2g.

Checks run in the page

  • Antoine pressure against steam-table values at 100, 120, 150, 180 and 200 C.
  • Stepped kernel heating against the closed-form pop time.
  • Monte Carlo share of dry kernels against the normal-distribution formula.

Honest simplifications

The kernel is treated as a sphere with one temperature; real kernels have a temperature gradient (Bi is shown, and above about 2 the lumped model is rough) and heat arrives through the contact patch. The burst pressure is a distribution, not derived from hull mechanics. The dry-kernel threshold of 11.5 % is empirical. The jump (an observed hull-leg mechanism) is modelled only as a launch speed, popped kernels do not collide with each other, and the pile height in the pot is estimated from the popped volume. The heating clock can be sped up; the flights always run in real time.

What to try

  • Lower the heater to 165 C: the steam pressure never reaches the hull strength and nothing pops.
  • Choose old, dry kernels and watch how many stay in the pot.
  • Raise the heat transfer coefficient (oil in the pan) to shorten the whole run.
  • Take off the lid and let the popcorn jump into the bowl.
Loading model...
Loading Blender model...
Fraction of kernels popped
Pop rate
Steam pressure vs temperature
Made with 🟠 Blender · Three.js