Imagine building a car from scratch, using no mass production techniques at all.
Imagine having to engineer each part individually, from scratch: every bit of the engine, every nut and bolt of the transmission, the drive train, the interior, down to the doors and latches and the steering wheel and the seats.
Now imagine both the entire design and each separate part of it had to be certified for safety by an arms’-length regulator. And imagine that if you made any change at all — no matter how trivial — the entire safety certification process had to start again from scratch.
You’d end up with a five million dollar car. It probably wouldn’t run very well. It’d be slow and clunky and uncomfortable. And it’d be stuck in a technological time warp: continuous improvement couldn’t happen, so you’d be stuck with 1950s automotive technology forever.
That’s how we build nuclear power plants.
One bespoke design after another, every piece custom built, under a regulatory system signed by Kafka.
This is the thing to keep in mind when you hear people complain that nuclear is way too expensive. They’re not wrong, exactly. If you build it in the very stupidest way possible — the way we build it now — nuclear certainly is way too expensive.
That’s not a feature of nuclear physics, though: it’s an artifact of messed up human systems.
If we built cars the way we build nuclear, we’d think of cars as this shitty, clunky, insanely expensive tech that clearly won’t scale. The car, we’d be told, is a technological dead end: an idea that once seemed promising, but didn’t work in the real world.
But what if we built nuclear reactors the way we build cars?
On an assembly line? At scale? As part of a continuous flow process where engineers continuously tweak routines to squeeze a 0.1% efficiency gain here, and a 0.07% efficiency gain there, day after day, week after week, for decades?
In Japan they call this “kaizen” — routine incremental improvement.
It sounds banal — it is banal. But it compounds over time. Kaizen is how we went from the crappy, dangerous, gas-guzzling, slow, uncomfortable cars of the 1950s to the sleek, safe, fuel-efficient, fast, comfortable cars of the 2020s.
The road to building nuclear power plants the way we build cars is not as fanciful as you might think. The key is to make more, smaller reactors designed to be physically incapable of melting down (what the industry calls passive safety). Reactors designed to be deployed in the same sites where old coal or natural gas power plants were, places already connected to the grid. Or designed to go on barges that can be redeployed as needed. Reactors that aren’t cooled with water, and so don’t need an expensive, failure-prone pressurized chamber. Reactors that produce high-temperature outputs and so can be used for industrial applications and district heat as well as to produce electricity. Reactors that rely on cheap, abundant fuel that can’t be used to build a weapon, and generates hardly any nuclear waste at all?
Reactors that are as far removed, technologically, from the reactors we have today as this is from this:
Right now, today, more than 60 companies across 15 countries are racing towards this vision. The race is mostly invisible to normal people, because the hugely dysfunctional legacy regulatory framework has kept them from reaching production so far.
In the next few years, that will change. And that change can’t come too soon.
Because nuclear fission is the only technology proven to be able to take a developed-world energy grid to carbon neutrality in real-world conditions: France, Finland, and Ontario all proved this back in the 1970s.
That those places have solid, reliable, inexpensive zero-emission power in the 2020s, even though they’re working on technology from the 1960s built in the 1970s and 1980s, tells you all you need to know about nuclear’s potential.
Because just think: even nuclear built in the dumbest, most dysfunctional way imaginable “works,” just about, even now.
Imagine what nuclear built intelligently could do.
A car where every part needs government certification- you just described an airplane. That is the reason all airplanes have had the same basic shape and technology since the 1960s.
The same process described would result in small nuclear reactors economical enough to power transport ships, replacing all the bunker crude currently used as fuel. Instead, ship powering nuclear reactors are so expensive that only the military uses them. The US Navy has 6,200 reactor years of operation, and counting, without a single radiological accident. No one on these ships has ever exceeded any Federal radiation limit or been harmed by radiation from the reactors. Transport ships are responsible for about 3% of global carbon emissions.