From split atom to spinning turbine: a nuclear power station simulator for NCEA Physics
A nuclear power station is, at heart, a very elaborate way of boiling water. That sounds like a joke, but it's the honest truth and it's a genuinely useful thing for students to realise: strip away the reactor and the radiation, and the back half of the plant is the same steam-and-turbine machinery as a coal station. What's remarkable isn't the electricity generation — it's the source of the heat. This simulator lets students follow the energy all the way through, from a splitting uranium nucleus to power flowing onto the grid, and then break the plant in instructive ways.
It's a single self-contained web page: open it in a browser, put it on the projector, and it runs — the plant starts up already producing power, and you drive it from a control room. Below I'll walk through the physics it shows, then flag some ways to use it in class.
Following the energy
The clearest way to understand a power station is to trace the energy through it, and the simulator is laid out as one continuous chain you can watch flowing:
Nuclear → thermal. In the reactor core, uranium-235 nuclei undergo fission, and the energy released heats water in the primary loop. This water is kept under enormous pressure — around 155 bar — so that even at over 300°C it stays liquid rather than boiling. You can watch the core glow with fission power and the hot water flow out of it, its colour tracking its temperature.
Thermal transfer. The hot primary water flows to a heat exchanger, also called a steam generator. Here it passes its heat to a separate body of water in the secondary loop, boiling it into steam. In the simulator you can see the bubbles rising as this second water boils.
Thermal → kinetic. The steam rushes to a turbine and pushes its blades, spinning them — you can see the rotors turning at a rate that tracks the actual steam flow.
Kinetic → electrical. The spinning turbine turns a generator, which produces the electricity that flows out along the transmission line to the grid.
The used steam is then cooled back into water in the condenser and pumped around again, while waste heat is carried off through the cooling tower as that characteristic plume. So the full energy story is: nuclear → thermal → kinetic → electrical, and the simulator labels each transformation as you watch it happen.
Why two separate loops?
The heat exchanger is the design feature students most often find puzzling — why not just send the reactor's water straight to the turbine? The answer is a nice piece of applied reasoning, and it's about safety, not efficiency.
Water passing through the reactor core becomes radioactive. If that water went on to the turbine hall, the entire turbine and generator building would be radioactive too. So the primary loop is kept sealed inside the containment building, and only its heat is allowed to cross into the clean secondary loop. The steam that reaches the turbine has never been near the core — the two loops never mix. It's a case where a whole piece of expensive engineering exists purely to keep radioactivity contained, and that's worth making explicit rather than letting students assume it's about performance.
The control room: three dials, three different physics
The plant is driven by three controls, and the reason it's worth having all three is that they do genuinely different things — this isn't one "power" knob wearing three hats.
Control rods. These are made of a neutron-absorbing material. Push them into the core and they soak up neutrons, so fewer are left to cause further fissions and the chain reaction slows. Pull them out and the reaction builds. They control the rate of fission, and so the reactor's power. In the simulator, one important detail is built in: the rods move at a realistic finite speed rather than snapping instantly, because a reactor that could change reactivity instantaneously would be uncontrollable.
Coolant pump. This drives the primary water around the loop. Slow it down and heat isn't carried away from the core fast enough — and something instructive happens, which I'll come to.
Grid load demand. This is how much power the grid is asking for. Turn it down and less steam is drawn off, so less energy leaves the plant, and the reactor settles to a lower power to match. It's a small illustration of how a real plant follows demand.
At normal settings the plant produces around 3,000 megawatts of thermal power and about 1,000 megawatts of electricity — the real numbers for a large pressurised water reactor.
The two lessons hiding in the failures
The controls are most valuable when you misuse them, because two of the most important ideas in reactor physics only show up when things go wrong.
Cut the coolant pump, and the reactor calms itself down. You'd expect losing cooling to send the power soaring. Instead the core temperature spikes, but the reactor power falls on its own. This is the negative temperature coefficient, and it's the single most important safety feature of a reactor design: as the core heats up, the water expands and moderates neutrons less effectively, so the reaction becomes less reactive automatically. A reactor of this kind is inherently self-stabilising — a hotter core is a less reactive core. Students can watch this happen rather than take it on trust, and it's a lovely example of negative feedback in a physical system.
Hit the emergency shutdown, and the power doesn't reach zero. The SCRAM button slams all the control rods in and fission stops within seconds. But watch the thermal power: it drops sharply, then levels off well above zero and only slowly declines. That residue is decay heat — the radioactive fission products already sitting in the fuel keep decaying and releasing energy for a long time after the chain reaction has stopped. It's a genuinely counterintuitive result and one of the most important in the topic: shutting a reactor down does not make it safe to walk away from. It still has to be cooled, for a long time. If you cut the pump and SCRAM, the simulator's warning names this as the exact failure mode behind the Fukushima accident — reactors that had shut down correctly but then lost the power needed to keep pumping coolant, and overheated on decay heat alone.
Where the energy actually goes
There's an energy bar along the bottom of the simulator that quietly makes an important point: only about a third of the nuclear energy released ends up as electricity. The rest — roughly two-thirds — leaves as waste heat through the cooling tower.
Students often assume that's just bad engineering, something a cleverer design could fix. It isn't. It's a fundamental limit set by thermodynamics. Any heat engine's maximum possible efficiency depends only on the temperatures it works between:
η_max = 1 − T_cold / T_hot (in kelvin)
With steam around 290°C and a condenser near 30°C, that ceiling is about 46%, and real friction and turbine losses bring it down to around a third. This is also why the condenser and cooling tower aren't wasteful extras — a heat engine can only do work if heat flows from somewhere hot to somewhere cold, and the condenser provides that cold end. Rejecting heat there is what allows the turbine to work at all. That's a subtle idea, and being able to point at the cooling tower while explaining it helps it stick.
A note on the physics
Underneath the diagram is a proper coupled model — reactor power, two coolant loops, and the turbine all linked together and responding to your controls, not a scripted animation. I tuned it numerically so that the steady state lands on real PWR figures (about 3,000 MW thermal to 1,000 MW electrical at roughly 33% efficiency), so that a SCRAM drops fission power about 90% within a few seconds while decay heat lingers, and so that losing coolant makes the reactor self-limit through its temperature feedback. Testing along the way caught genuine errors — an early version let a fully shut-down reactor keep generating, and another spiked past its power ceiling on startup — which is a reminder that "looks about right" and "is right" aren't the same thing.
One honest caveat: it's a lumped-parameter model, so it won't show coolant boiling away or pressure transients, and the timescales are compressed — roughly twelve times real speed — so a SCRAM or a heat-up plays out in seconds rather than minutes. The behaviour is faithful; the clock is sped up so it fits a lesson.
Using it in class
- Trace the energy chain first. Before touching a control, just follow the flow from core to grid and name each transformation: nuclear → thermal → kinetic → electrical. The plant runs on its own, so this can be pure observation.
- Ask why there are two loops. Let them puzzle over it before you explain. "Safety, not efficiency" is a satisfying answer once they've felt the question.
- Cut the pump and watch the power fall. Ask them to predict what happens to reactor power when cooling is lost. The result contradicts the obvious guess, and the negative feedback is the payoff.
- SCRAM at full power and watch for zero. It never arrives. Decay heat is the lesson, and the Fukushima connection makes it concrete rather than abstract.
- Point at the energy bar. Two-thirds of the energy is "wasted" — ask whether better engineering could recover it. The answer is no, and η = 1 − T_cold/T_hot is why.
It sits alongside the rest of the AS91172 suite, so the fission heating the water here is the same reaction that splits the nucleus in the fission simulator and runs the weapon — the same physics, harnessed instead of unleashed. Start by boiling water, and let students discover that the interesting part was never the turbine; it was where the heat came from.