Hacker News Reader

Comment by rmetzler | Hacker News Reader

Something I miss from reading those articles: what are you doing when the wind doesn't blow for a few days or even weeks? And in case of solar farms, the sun doesn't shine (like at night)? How do you store the energy efficiently?

If you think batteries, how long does the battery of your smartphone or laptop works? 5 years max? Maybe 10? Is this the timespan you plan your reliable infrastructure for? And how "green" is it to build these batteries?

If you think pumped-storage hydroelectricity, how much places do you have where you can have two pools (one uphill, one downhill) to store the water? How much energy can store in there?

The thing is, you'll build all these solar and wind farms and then still build and run the fossil plants, because you aren't able to store the energy to make it reliable enough.

View on HN

At the moment you just maintain a gas backup infrastructure. Depending on how large your grid is and the intermittency and variation of its sources of electricity, you can assume a certain level of output, then the gap between that and maximum demand needs to be maintained as backup. That’s not necessarily a bad thing, gas plants have a low up front capital cost, with a high marginal fuel cost, so having plants available but used less frequently is not a serious economic barrier.

In the long run energy storage techniques will come in which will displace some use cases for gas, how quickly that happens depends of technological development. As you say long term storage isn’t likely to be achieved by Lithium ion batteries, it’s more likely to be Flow batteries or Hydrogen production. Although Lithium batteries can shift storage throughout a day, and also balance the grid on near instantaneous timescales, which is useful.

Another possibility is overbuilding renewable generation, if solar is very cheap you could produce twice as much as you need in Summer in order to produce enough to meet demand in Winter. That can drastically reduce the need for long term storage, but depends on the properties and cost of the renewable technology in a particular location.

What do you mean by producing twice as much in Summer to meet demand in Winter? Because much of our energy consumption is immediate, and can't be easily stored (e.g. heating, cooking, browsing the internet etc).

I think the GP meant having a peak capacity twice the size of peak demand.

>> it’s more likely to be Flow batteries or Hydrogen production.

From an engineering point of view, this statement probably should be correct. Hydrogen in particular has such wonderful potential.

But, we have to consider future economies of scale in production. We don't know how long it's cost-performance may continue to improve, but of the three technologies mentioned, unfortunately only LiIon is currently on an improvement curve that makes it feasible for large scale storage in the short-to-medium term.

At the moment Lithium batteries are cost effective for about 2-4 hours of storage, if you halve costs that would go down to 4-8 hours, but the point is that it’s very unlikely that you will ever get reductions large enough to store energy over weeks or months. Even 2 days would require costs to fall to 10% of where they are now.

There seems to be an error in those assumptions.

Halving costs need not be equivalent to doubling of economic capacity. It might just as well cause an order of magnitude or higher of increased capacity.

I would not be able to estimate those factors, but it's almost certainly incorrect to assume a linear relationship between cost and productive capacity.

Each halving in cost doubles the amount of storage you can apply to any one project for the same cost. 2 hour storage on a solar or wind farm costs around about £5 per MWh, levelized over the batteries lifetime. There is no way to get around that, as far as I can see.

"The thing is, you'll build all these solar and wind farms and then still build and run the fossil plants, because you aren't able to store the energy to make it reliable enough."

So? If we only halve fossil fuel use does that not mean its worth doing?

The article states that wind power is now cheaper than just the fuel for NG power plants, so its financially advantageous to move to wind whilst maintaining NG plants. And its environmentally advantageous even though we may not have all the answers right now.

I'm not even sure if there is a 'the' answer. The solution will be probably be a combination of all of the above, plus HVDC, plus smart grid type features.

BTW laptops and phones are the worst environments for batteries, they're hot and enclosed. Look at EV battery warranties, the Bolt is 8 years, the powerwall warranty is 10 years. The expectation, and experience is they'll last much longer.

One, also inefficient, way to get to storage is by using surplus renewables (wind, solar) to generate bio gas (CNG or even LPG) for later use. Not the most efficient solution, but does that really matter when marginal costs for wind and solar are basically zero?

Depends on the cost relative to other options.

I would guess that if costs keep dropping you'd get a certain amount of overbuilding for the low energy demand season so you'd have capacity there for some kind of clean gas technology.

I wish the boffins would start crunching the numbers though. If my gas boiler breaks today, should I replace it or fit a heat pump? If I replace my kitchen should I keep a gas hob or move to electric?? These are less than once a decade purchasing decisions and theres absolutely no visibility on whether gas to the home is going to be a thing in the medium term.

It's as much seasonal as it is spread over time. Solar is strong around noon when demand is, as of now, rather low.

Intraday, batteries would make more sense. Higher efficiency, quicker to react, 365 cycles a year to recoup costs.

You are probably never going to get to a point where you're charging a battery now for use sometime in January. That's where some kind of bio gas would come in, because it is feasible to store it long term.

Exactly that!

Why is this comment being downvoted? These are all valid concerns, and even though the tone is a bit negative, they need to be addressed. We can't just ignore the viewpoints that challenge our views. I'd be much more interested in comments explaining how these problems with consistent energy supply can be solved, than downvoting this question. In fact, I'd like this question to be encouraged if it helps inform me of solutions to the storage side of the grid...

https://www.google.com/amp/s/www.forbes.com/sites/peterdetwi...

I see you got downvoted.

Your comment touches on topics outside the current article and its discussion. The reason is that Electrical grid balancing has been commented elsewhere; the economic operation of plants has also been commented in other discussions; if wind is blowing - there is energy left on the table without it being used; if operating a plant on fossil fuels, those fuels still have to be delivered. Wind energy is variable input, but review data on its variability over span of days.

Overall, having variable input of energy in an electrical grid is not the blocker. It is also an extensive - and ‘done enough for now’ area of research.

The variability is not what has slowed down wind energy deployments.

> The variability is not what has slowed down wind energy deployments.

It has been subsidized in a lot places. And yes it creates a lot of variability which is currently mitigated by powering fossil plants up and down.

People think they buy solar and wind power and it works on paper, but in reality it only works because there are fossil fuel plants as backup.

Nuclear power has been subsidized, most still is, coal is still subsidized. So what's the problem here?

Politics and ignorance, the usual culprits

It's not possible to discuss the cost of generating a kWh without also discussing availability aspects. Comparisons must be made like for like.

For example, certainly for solar the cost and practicality of storage must be taken into account.

There are similar constraints with wind.

Overall, renewables can be used a lot these days but still have limitations.

As someone said in a different thread: Fossil fuel will compete with storage, not with primary energy production. Wouldn't it make sense to move money and research into finding better energy storage?

Pumped-storage hydroelectricity does need special places to be most effective, but we already have a lot of one-way dams littering our rivers, of which some might be upgraded. Deep mines that can be used for the lower reservoir, maybe something to be give depleted coal mines a new purpose?

Energy production will change a lot over the next years, there's no way stopping it.

There is a YC startup looking at converting CO2 and H2O into "Fossil Fuel" called Prometheus (namesake of the monitoring tool I guess!) https://news.ycombinator.com/item?id=19842240.

> but we already have a lot of one-way dams littering our rivers, of which some might be upgraded

It would be better to locate pumped hydro off rivers. Dams are terrible for the ecology of rivers.

Pumped storage, but anywhere:

https://www.forbes.com/sites/peterdetwiler/2019/08/14/tower-...

There was an interview with the founder on Bloomberg TV last week as well.

Doesn't this not scale? Such that you would need millions of enormous towers to store enough energy to power the US total electricity usage?

Between wind, solar, and storage, you would be hard pressed to “run out” of energy in a well designed system. Stating otherwise is nonsense.

I don't understand how to design such a system.

The sun doesn't shine in a lot of places for less than 12 hours a day. In the winter, a lot of places have much less sunlight.

When men needed wind to sail the sea, there were situations when the wind didn't blow for weeks.

How much energy would your "well designed system" need to store and what is possible?

Yes, I read about that Tesla battery in Australia. Then I calculated how many Tesla walls a city like Munich would need to be able to survive for 1 week. I don't believe it is possible.

People rely on electric power. If the grid goes down in places like Germany like once every month, there would be uproar.

You don't need to survive for a week but max. a day or so. Anything longer and electricity will come from somewhere else in the grid, either a different region or a back-up conventional plant being gas or nuclear or even coal.

At least in Germany even large industrial energy consumers are for years now an active part of grid balancing. Either they can stop and resume production as needed or continuous processes serve a similar function as your base power plants. The silver bullet to get them there was money, it became financially viable and profitable and all of a sudden businesses jumped at the opportunity.

Disclaimer: Worked at two of these power hungry places and know of of another one making quite some money on the electricity exchanges by just timing his production runs properly.

Yes but the real aim should be to produce totally clean, renewable zero carbon power cuts.

You're betting against human ingenuity here.

Solar still works in cloudy weather, and you can compensate by building more than you typically need. You can transmit increasingly long distances. Here's a line working at 2300+km.

https://www.power-technology.com/features/featurethe-worlds-...

Most weather doesn't span 2300km in all directions so I think it's possible to handle most situations and fill in the gaps over time.

You can use a Tesla as home battery. Charge at work or home, or at a supercharger, power your house at night.

Sure there are edge cases like far north where you need coal or nuclear. But I'd bet the bulk of humankind can be supplied well given another 10/20 years of innovation.

Have you made simulations or calculations to this effect? (or alternately, point me to the analysis of those who have). I'm not sure hand-waving it away as it'll wash out on average should be a valid response to something as critical to modern life as the power grid.

Some questions that might be interesting: What is the current downtime of electricity. What would be the desired goal for the new mixed / renewable grid? Given existing patterns of wind and solar generation, how much storage needs to be installed to reach this goal? How much would this storage be expected to cost at today's prices, and then with projected future savings from scaling (this could be used as a higher bound)?

The reason "running out" might be a valid concern is that the power grid is currently reasonably robust. Turbines are massive, with a great deal of inertia, meaning that even if something drastic were to happen, they can often cope with spikes in load long enough for extra production to ramp up. Wind and solar less so. Batteries presumably would be pretty great for ramping up, assuming we get enough of them on the grid. But then the economics needs to take into account the price of not just the renewables, but also generation.

Michael Moore is about to release a documentary essentially arguing that what you stated is largely a hoax. I don't have an opinion because I don't have the knowledge. But Moore is pretty credible to people on the left, I wonder what the impact of this will be.

Realistically it's between wind, solar and natural gas. Unless you're lucky in a hydro area.

Storage will happen, but it's a long way off.

Yeah where I live the vast majority of our power is hydro.

Even if renewable plants are supplemented by fossil-fuel plants, that still reduces fossil fuel consumption.

Storage isn't all that feasible (yet) but they also don't build wind farms in places where the wind doesn't blow, and they don't build solar plants where the sun is inconsistent.

>Something I miss from reading those articles: what are you doing when the wind doesn't blow for a few days or even weeks? And in case of solar farms, the sun doesn't shine (like at night)? How do you store the energy efficiently?

As the renewable industry, you still have people depend on the conventional grid for the vast majority of power, while taking advantage of generous subsidies, tax breaks and other deals to convince people renewables are anywhere near a match for it at the moment...

I don't see how it is even thermodynamically possible for there to be no wind period. It is produced by temperature differentials and it would take some very extreme geoengineering to not have those - like a planetary superconductor grid.

Not to mention geographic distribution of grids levels it out.

At that point you may as well ask what if positive and negative charges repel each other?

A good grid (because it's always windy somewhere), batteries for a day or two of low production (e.g. the batteries in electric cars, one of those can power the average home for several days), power-to-methane for cold winters without wind.

Though I have only a 24kWh battery (2015 LEAF), that’s nowhere close to powering my home for several days, even if I were able to park the car entirely during that time period.

Looks like the average US electricity consumption is close to 30 kWh/day.

Storage is very important, but there are many places in the world where wind and weather are pretty constant.

And near the equator there's much less seasonal variation in solar.

I wonder if energy-intensive heavy industry is going to migrate to low latitudes in a solar powered world. Sorry, previously industrial Europe.

I read a few times about a proposal to build super-high voltage power lines between the Sahara and Europe to transport solar power...

Reply on news.ycombinator.com