Hacker News Reader

Comment by DennisP | Hacker News Reader

Gigawatt-scale nuclear reactors take a long time. Ycombinator is investing in factory-produced reactors that fit in a truck. There are quite a few other companies working on small factory-produced reactors.

Solar with batteries is a lot more expensive than solar alone. Solar is doing well right now with natural gas backup, but a lot of us would like to avoid fossil entirely. Solar with nuclear backup might be a great combination, assuming demand correlates reasonably well with daylight hours.

View on HN

> nuclear backup

That wouldn't work with current nuclear reactors, whose startup / shutdown sequences take hours. Typically, the grid operator runs nuclear plants at nearly 100% 24/7 (so called "baseload") due to this.

I'd be curious how truck-sized reactors work and whether or not you could operate them as load-followers rather than baseload. That would make them extremely attractive for replacing natural gas peakers, especially as more wind / solar get onto the grid.

My point mainly is that if there's more demand when solar produces more energy, then you can pretty much run the nuclear plants all the time and get a decent balance.

But I suspect the small reactors will be more flexible. Molten salt reactors are supposed to load-follow automatically with a lag of thirty seconds or so. (And of course with small fusion reactors it wouldn't be a problem at all.)

Solar generation is less predictable, and the grid needs an energy source that is capable of quickly adjusting to the current fluctuation in demand of consumers.

That's where the batteries come in. To continue the software metaphors, batteries provide caching, and the cache is tunable for cost and performance. This wasn't viable a decade ago, but with the advances in solar/wind cost/performance, and the advances in battery cost/performance, it's really becoming a powerful option.

You're right. It is also a common misconception that nuclear reactors cannot load follow, they can:

http://ansnuclearcafe.org/2013/02/14/responding-to-system-de...

One of the main limitations is the stress it puts on the fuel.

Many advanced reactors overcome these limits, and if financially incentivized, they will definitely load follow. On top of that, the UPower reactor has a thermal transport time constant nearly 10 times that of other reactors, and its fuel is immune to the shocks that bother LWRs. In fact, the same type of fuel was used in a research reactor and would be ramped in power from 5 watts to 150 billion watts in less than 50 millionths of a second. That puts a lot of stress on fuel, yet this fuel kept its stride without breaking a sweat.

This reactor is built like a tank, and is designed to be quite resilient and flexible. It can definitely load follow to support a renewable heavy grid system. In fact it's been considered for use as a grid stabilizer at substations because of these abilities.

What kind of fuel and coolant does it use? Googling only turns up a couple fairly uninformative articles.

So how small a reactor? And what sort of cost, what sort of setup time, what sort of safety constraints (ie, current reactors are generally water-cooled and need access to large bodies of water)? It's possible to trim the cost of reactors from the billions to the millions, maybe, but how many millions?

Certainly, solar with batteries is a lot more expensive - but to my point, startup cost is low. Latency, or bandwidth? The long-term cost/kwh may be higher for solar, but the short-term startup expense will be much lower, unless you can get the factory-built reactors down into the thousands rather than millions of dollars.

If nuclear is cheaper than solar in a decade or two, but costs ten times as much in the short run, there's a tremendous advantage to solar, battery cost or not. Opportunity cost matters tremendously.

Startup cost for solar is not all that amazingly cheap per watt, and if you include batteries it's probably more than nuclear.

UPower's design doesn't require a large body of water: http://spectrum.ieee.org/energywise/energy/nuclear/startup-d...

For setup you deliver it with a truck and bury it. Other modular nuclear designs are a bit bigger but still in the range of small natural gas plants, which are competing quite well.

If Helion works out then fission and solar will both be mostly obsolete. They'd be 50MW plants, retailing power at four cents per kWh, with no significant safety concerns.

Wouldn't trust a Silicon Valley startup to care about our properly execute for "safety." This is a pump and dump, making a quick buck.

You mean "release early, release often"?

Safe = reliable = high capacity factor = revenue. Look at the INPO and NRC ratings of plant safety and economic performance. They are directly correlated. So it's in every nuclear startup's best interest to be very safe.

Solar scales MUCH more fine-grained. There isn't a convenient source for 10kw nuclear power, and there probably never will be, but it's easy to build solar at that scale.

You're exactly right, and that's what we designed for with UPower. I would have written the exact same thing when we began talking about doing something in nuclear 5 years ago.

Too many reactors are designed without the market or financing in mind.

We decided on the simplest possible reactor optimized to a size useful to a market in dire need- just MW scale.

It has no pumps, no water in the reactor, and builds upon a legacy of data so that there will be minimal fuel and materials qualification, which adds up very quickly in both time and money.

Why hasn't it been done before? The key, as you bring up, is in manufacturing, simplicity, a relatively new and hugely growing microgrid market that didn't exist much before, and a business model that doesn't require the customer to buy the unit as opposed to power purchase.

The problem with anything nuclear is insurance. The only way reactors got insured was because Uncle Sam underwrote them.

Indeed. What insurer could have covered the estimated 137 billion dollar hit [1] from the Fukushima disaster? The Japanese government has had to step in, to the very real financial detriment of its people and their nation.

[1] http://www.bloomberg.com/news/articles/2012-11-07/fukushima-...

Also, Chernobyl was one of the reasons Soviet Union collapsed.

Well, it's cheaper than Greece.

(Nuclear reactors have the same "too big to fail" problem as the banking system: the expected value is positive, but the worst-case is a huge open ended liability and toxic asset)

It's exciting work, to be sure. Before now, the only small-scale nuclear work I'd seen were plutonium batteries (like for powering satellites), which are horrendously expensive and not something you ever want in the hands of Bad Actors.

It looks like UPower is currently targeting environments where traditional power is impractical and lots of power is needed, and plenty of budget is available - remote mines, military installations and such. Do you see a market for urban/residential power grid in the future, too? Or would that be too difficult a squeeze between distributed solar and traditional power plants?

The short answer is yes. We see this as our Tesla roadster (well designed niche product for a market willing to pay, in this case however, desperately in need for a solution that doesn't involve constant shipments of expensive and polluting diesel for loud generators) from which we will streamline and optimize to make our "model 3" so we can produce something to meet and even beat grid prices. It actually isn't a big jump between the two, we have good indication now that it will be possible without much iteration to beat grid prices in all but the cheapest markets. And as you pointed out, the financing at that stage will play a significant part. :)

Confused: if you have nuclear, skip the solar and save money?

Sure, if nuclear ends up cheaper per kWh than solar without backup.

Depends if nuclear includes decommissioning and clean-up costs.

Do you really want lakes full of residual waste for the next few tens of millennia?

The UK currently has no idea what to do with a lot of its waste. So it's simply left to rust and ferment in water - not a good outcome.

Sustainable intermittency turns out to be something of a myth anyway. Intermittency effects across Europe turn out to be negligible.

So instead of building nukes, you can spend the money on mixed-mode sustainables and a hugely improved distribution grid and get a cleaner and more reliable outcome overall.

You can also make sustainables distributed, and run them on a domestic scale as well as an industrial one.

PV roof installations have worked well in Germany and are starting to work well in the UK, even though neither location is known for being sunny.

You'd get much better results in the sunnier parts of the US.

Long-term waste isn't a factor for UPower or Helion.

How?

UPower is a fast reactor. 99% of nuclear waste, and almost all long-term waste, is transuranics, which fast reactors use as fuel. What remains is the fission products, which go back to the radioactivity of the original ore in a couple centuries. Encase them in glass and bury them, and you're good.

Helion is a fusion reactor. Its "waste" is helium, and it uses a reaction that produces only 6% of its energy as neutron radiation.

In both cases the reactor itself may become somewhat radioactive, but that's another short-term problem.

Reply on news.ycombinator.com