Hacker News new | past | comments | ask | show | jobs | submit login

The whole "what about at night?!" line has been trotted incessantly out here, slashdot, hackaday, reddit, etc like people in the utility industry aren't aware. A huge amount of electrical load happens during the day that doesn't happen in the dead of the night, which is why many utilities offer nighttime cheaper rates.

That hasn't stopped utilities from deploying wind and solar for 40% of their new generation over the last year or two, while shutting down coal plants left and right. For half a decade it's been considered settled fact that solar and wind are cheapest and the way forward. So, maybe this isn't as big a deal as you think it is.

Attention long ago shifted toward planning the grid to accommodate that, and also on grid-scale storage projects. Pumped hydro has been used extensively in the UK for decades because every time the BBC finished a program, everyone watching TV would fire up their electric kettle at the same time. There are several pumped hydro installations in Canada and the US as well.

Most EVs can schedule charging cycles and numerous utilities provide special rates for customers who agree to some level of utility control on their charging similar to how utilities provide discounts to people who let them tweak their AC thermostat; you can still override it, in most cases. In Australia, going back several decades, there was a grid load reduction signal system that worked by generating a harmonic that load reduction switches would look for.

An increasing number of EVs are set up to be able to backfeed power into the grid, and the idea is that they could do so on request from the utility.

Lots of utilities offer off-peak billing incentives. Imagine a world where your fridge does most of its cooling during the day, and "rides" through the night, etc.

EV charging stations are starting to come with their own local battery systems because DC fast charging, particularly at 800V, is enough power that it can be difficult to get enough electrical service and in some locations the station sits unused for a fair period of time.

Add in the fact that lithium ion battery prices continue to plunge following expected rules about production costs, as well as other technologies maturing; there's an iron salt based flow battery that looks really promising, made of very cheap and extremely prevalent materials (salt, iron, and water mostly) and scales fairly well.




>That hasn't stopped utilities from deploying wind and solar for 40% of their new generation over the last year or two, while shutting down coal plants left and right. For half a decade it's been considered settled fact that solar and wind are cheapest and the way forward. So, maybe this isn't as big a deal as you think it is.

Conspicuously absent in this incomplete story is the fact that so much of the coal has been replaced by natural gas, which doesn't have the giant unreliability downside of solar and wind. You can get away with >=~40% of your energy generated by solar/wind on every single day of the year with no interruption. Bump that up to 80% and it'll take a long time and a lot of money to accomplish that without sacrificing reliability.

But why? Just use nuclear for the other half. The goal is not "renewable energy ASAP". The goal is "minimize fossil fuel usage ASAP".


The problem with nuclear is the long lead time to having a plant up. With delays you can expect a nuclear plant to be online in 15 years.

According to Wikipedia The US has also shutdown a couple nuclear plants because they were cost ineffective.

So the fastest path off fossil fuel are renewables.


Smaller series-built plants will solve the long lead time problem. Once installed these are to be run at max capacity all the time, providing the base load - actually running a nuclear power plant is cheap compared to all other energy sources, the costs are mostly made in the beginning - planning and construction - and end - dismantlement and conservation - of the plant. Those costs - both head and tail - will go down radically with series-built plants. Once a reliable and cheap nuclear or hydro base load is in place the rest can be filled in with a mixture of renewables with storage (which is not available for now except for regions with a large established hydro-power infrastructure), renewables with fossil-based backup or more nuclear capacity. Renewables can not be used on their own as long as the storage problem is not solved, insisting that they can will and does lead to extreme price hikes and brown/blackouts.

I'd be wary of using Wikipedia as a source on this politically contentious issue since it has a known and fairly extreme slant to the "left" on most of these issues, in part due to the Wikipedia "reliable/perennial sources" policy [1] which promotes the use of left-biased sources while demoting centre- or right-biased sources.

[1] https://en.wikipedia.org/wiki/Wikipedia:Reliable_sources/Per...


How many years before those new models are available at scale? Less than 15?


That mostly depends on the will of the regulator, technically these systems are mostly ready. The NuScale 50MW modular reactor has now gotten its safety approval [1,2 - interesting to see how Popular Mechanics claims to "like nuclear technology" while the Scientific American article shows the more negative attitude "green" proponents usually espouse ] so it is a step closer to reality, what remains is mostly more rounds of approval and certification. In the UK Rolls Royce is working on a larger (440MW) [3] type of modular reactor with a stated goal of having them up and running in 10 years, building at a pace of 2 per year. Russia and China are also working on this type of reactor, so is Argentina. The IAEA has a technology roadmap for the deployment of small modular reactors [4] for those who want to read more on the subject.

To answer your "less than 15" question succinctly: certainly, as long as the activists are kept at bay. Possibly, if they are allowed to roam free to wreak havoc. Probably not if they are put in control.

[1] https://www.popularmechanics.com/science/a33896110/tiny-nucl...

[2] https://www.scientificamerican.com/article/first-u-s-small-n...

[3] https://www.bbc.com/news/science-environment-54703204

[4] https://www-pub.iaea.org/MTCD/Publications/PDF/PUB1944_web.p...


It's doubtful the utilities are going to leave money on the table if there is a significant uptick in nighttime electricity usage due to to charging.

I'm not sure I'd want the lifetime of the most expensive component in my car being used to supplement the grid. But not only that, the incentives for feeding power back into the grid are under attack from the utilities:

https://www.theverge.com/2022/1/8/22872236/california-clean-...


“I'm not sure I'd want the lifetime of the most expensive component in my car being used to supplement the grid.”

I’d imagine that a service like that would be paid for, indeed. Tesla has talked about setting up a system to orchestrate grid support from car batteries and powerwalls.

As I understand it there is more money to be earnt to maintain grid frequency, than to shift load. I guess with more battery storage on the grid the variability in price at higher frequency will calls being made for seconds and minutes will dip.

There’s storage over the day night cycle but earning once a day will take longer to generate income.

But seasonal variability will be generating income a lot more slowly unless the price goes very high, but there would probably be other problems at that point.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: