Hacker News Reader

Comment by aardvarkr | Hacker News Reader

Is no one pointing out that this isn’t the part that makes replacing fossil fuels hard? This is focusing on a completely irrelevant metric. It’s a great motivator for why we need to shift but it has no impact on how hard it will be to replace oil. There’s 100 practical reason for why it’s hard, namely that clean energy sources are not under our direct control to scale up and down as we need more/less. We can’t tell the sun to be strongest at 6pm when everyone gets off work so we’ll need more energy. You can force the wind to blow. The best way to solve this is improving battery tech but that’s a HUGE task. What we have currently is a focus on quickly growing increase in clean energy sources as a baseline energy source with a diversified mix of other sources along with natural gas peaking plants that are far cleaner than coal and provide near immediate response times to surges on energy demand which is incredible.

View on HN

South Australia already has days when solar alone is providing most of it's power [1] [2]. It's not that hard. You scale up renewables, and build curtailing (throwing away excess clean energy) into your cost model (until storage gets cheaper, which it will; Tesla is shipping almost 800 MW of utility scale battery storage per quarter). Australia's entirely grid will be clean by 2030 at their current rate of renewables deployment. Yes, yes, not everywhere is Australia, but the sun and wind potential is significant in most geographies where people exist. Let us not use edge cases to guide our path forward.

[1] https://opennem.org.au/energy/sa1/?range=7d&interval=30m (OpenNEM scoped to South Australia, trailing 7 days 30 min granularity)

[2] https://www.abc.net.au/news/2020-10-25/all-sa-power-from-sol... (All of South Australia's power comes from solar panels in world first for major jurisdiction)

> Let us not use edge cases to guide our path forward.

Funny you should say that after presenting an edge case.

In North America during the winter the sun is setting during peak demand in the Northern states. That means we need to overbuild wind so much that we can run entirely on wind on non-windy days. The economics do not support that.

Storage is critical to make the US case work.

Edge cases are the Arctic Circle. Naysayers keep naysaying while it's being done [1]. Enough sunlight falls on the Earth in an hour to power the world for a year, and there's roughly 200GW of solar PV manufacturing capacity globally.

[1] https://www.eia.gov/electricity/monthly/#tabs_unit-4 (Maps, right hand side showing retirements and commissioning; note wind is green, solar is yellow)

> Enough sunlight falls on the Earth in an hour

This isn’t relevant because we don’t have transmission from where the sun is at a given point in time to where the electricity is needed.

It’s like having enough nuclear plants in Fiji to power the world and calling this a “solved problem”.

We would need a global grid and solar more than 2X built for demand to handle the fact that the earth rotates and that there will be clouds on average.

Yes, we're going to need more renewables to support peak demand, I thought I mentioned that above (ie curtailment). A global grid is not necessary, although a more integrated grid (likely long haul HVDC transmission lines [1]), storage (utility and distributed), and demand response is [2].

[1] https://publications.jrc.ec.europa.eu/repository/bitstream/J... (warning: pdf, "HVDC Submarine Power Cables in the World")

[2] https://www.energy.gov/oe/activities/technology-development/...

Storage or better transmission infrastructure...

South Australia was famous recently [0] for having not only the most expensive electricity in Australia, but competing for the most expensive energy in the world. I would encourage the world to be cautious about copying whatever it is they are doing.

And if the South Australia Big Battery is an example of utility scale storage, store's about 1-2 hours worth of energy from a small plant. It isn't capable or expected to help power on through a night or anything, it smooths the grid when there is instability. Ie, it doesn't mitigate the main problem with renewables (intermittent generation). It mitigates secondary problems with renewables (short term grid instability). Or it mitigates whatever problems the South Australians have with their grid, I don't know. Point is, if they were relying 100% on renewables they'd have developed a culture of early bed times.

[0] https://www.news.com.au/finance/business/south-australia-has...

Not to mention that it has destabilized the national Australian grid to the point where a 35C day means that you get rolling blackouts across multiple states.

Currently the only way you can be sure you have power on a hot summer day in Australia is to share a substation with a hospital designated to provide life support to patients.

When has Australia ever experienced these rolling blackouts across states due to solar? The most recent substantial blackouts (2019 in victoria) were due to extreme heat causing the power stations to be unable to meet demand, plus some stations being offline [0]. This is totally unrelated to the uptake of solar which ameliorates peak demand for aircon power. [0] https://www.abc.net.au/news/2019-01-26/victorian-blackouts-w...

If you refer to your citation (from 2017), it states that electricity prices were high because of natural gas shortages and coal generator closures, and that the payback period on storage and solar is down to 7 years (storage systems have 10 year warranties, rooftop solar and their inverters typically have 25 year warranties, so after your payback period, your power produced is free to you). Renewables drive down the cost of power [1], being the cheapest form of electrical generation.

The "South Australia Big Battery" at the Hornsdale Power Reserve, recently expanded, has saved electrical consumers $150 AUD million [2], cannibalizing frequency response revenue from thermal fossil generators (natural gas). It is not intended for long duration discharge.

[1] https://www.theguardian.com/australia-news/2019/dec/24/south... (2019: South Australia’s clean-energy shift brings lowest power prices on national grid, audit finds)

[2] https://www.cefc.com.au/media/media-release/neoen-completes-... (According to an independent review conducted by Aurecon, the initial 100 MW battery has delivered over $150 million AUD in savings to South Australian energy consumers)

> were high because of natural gas shortages and coal generator closures, and that the payback period on storage and solar is down to 7 years

That isn't as compelling as you might think. If they screwed up their energy grid by closing down gas and coal generators then (1) solar would look better than nothing and (2) would have a short payback period.

That fact could just as easily be cited in an argument as "these clowns really need more gas and coal". The lack of fossil fuels is linked to very high power prices.

> Renewables drive down the cost of power [1]

Your link doesn't actually support that. It says they drive down the cost of wholesale power. That is how it typically plays out when countries go heavy on renewables (same thing happened in Germany). Wholesale prices drop, retail prices rise because of there is too much energy when it isn't wanted.

> The "South Australia Big Battery" at the Hornsdale Power Reserve, recently expanded, has saved electrical consumers $150 AUD million

Again, proabbly evidence of a mismanaged grid. Given that it was built in SA rather than somewhere important, it seems likely Big Battery is fixing up losses from installing an unstable amount of wind and solar.

>That is how it typically plays out when countries go heavy on renewables (same thing happened in Germany). Wholesale prices drop, retail prices rise because of there is too much energy when it isn't wanted.

There is too much energy because obsolete base load plants can't keep up with a modern grid. The primary source of excess energy isn't renewables because we want to use every single renewable kWh. Excess energy is energy that isn't wanted not just because we don't need it but also because we don't like how it was created. Coal plants not only force flexible power generators to turn off prematurely, they also generate CO2. So any insistence on protecting coal plants is completely misguided considering their fatal flaws as a technology.

>it seems likely Big Battery is fixing up losses from installing an unstable amount of wind and solar.

Is this supposed to be some kind of joke? Flexible power generators cannot increase grid instability because their response time is measured in minutes or seconds. You can turn solar panel on/off immediately. You can turn on/off turbines at any time with a slight delay. The Tesla battery stores an insignificant amount of energy so it can't even do what you are talking about. It's primary purpose is frequency stabilization which was usually done with peaker plants and unconventional energy storage like flywheels.

I worked as a quant for a power company in Australia. Solar and wind are a great way for power companies to make money and a terrible way to run an industrial society.

Running anything but an aluminium smelter off renewables is impossible. Because people have the nasty habit of wanting elevators to work even when it is raining.

>I worked as a quant for a power company in Australia

This type of work sounds very interesting to me, do you mind sending me an email (see my bio)? As a university student, I'd really like to know what I can be doing on the side to land a position like this.

Good thing most of us are transitioning to a knowledge society ;)

Hey, not to be too harsh but you do realize that was only for 1 hour in the middle of the day when usage is near its lowest? It’s impossible for solar to provide 100% of the power used in a day because the sun doesn’t shine for 24 hours. It also doesn’t reduce the amount provided by baseline power sources because they can’t scale down quickly (they take days to start up and slow down) so they ended up generating a wild excess of energy those days. The amount of power generated != power used and if there’s no one to use it then it goes unused. On top of that, that first link was great and showed just how much solar fluctuates week over week and day over day making it hard to predict.

Renewables are fantastic, don’t get me wrong, and I’m a huge proponent of them. Problem is they’re just one piece of the bigger picture and making wildly infeasible claims like 100% renewable energy just makes it that much harder to implement a reasonable and highly effective strategy that relies on growing renewables while maintaining a robust grid that can provide power cheaply. Southern Australia is one of the most expensive power markets in the world so it’s not really a good example and the world shouldn’t be following their model.

You’re not being harsh. I do realize it was only for an hour. How else would you expect an energy transition to occur? First an hour. Concept proven. Then multiple hours at a time. Finally, those remaining fossil generators sit idle so long they’re decommissioned (or converted to burn hydrogen or ammonia generated from renewables). The South Australian state itself has set its 100% renewables target before 2030. You set ambitious targets, and then you work to meet them. This isn’t any different then the UK pulling their new petrol vehicle sales ban forward to 2030 from 2035 because EVs have been proven out.

Cheap, dirty power isn’t good energy policy, and it’s likely others follow this lead in order to meet their zero emissions targets.

https://www.cleanenergycouncil.org.au/news/south-australia-s... (“The South Australian Government has recognised the incredible success of renewable energy to date, and has now set a firm plan for getting to 100 per cent renewable energy. This is the future, and the Clean Energy Council looks forward to working with the South Australian Government to make this a reality.”)

Following up, as the Australian state of Tasmania just declared that they're able to run 100% on renewables as of November 27th 2020 [1], with the local government targeting 200% renewable capacity by 2040.

[1] https://reneweconomy.com.au/tasmania-declares-itself-100-per...

[2] https://reneweconomy.com.au/tasmania-liberal-government-tabl...

This problem is way overstated, as most fossil fuel and nuclear power can't be easily or efficiently ramped down either. We also have energy storage technologies other than batteries: the most common being pumped-storage hydroelectricity (basically a reversible hydroelectric dam, where we pump water up during excess power usage and use the generators during high demand).

> being pumped-storage hydroelectricity

Which isn't great as far as the "Power Generated" : "Land Used" ratio is concerned.. the same is true of most hydro-electric solutions.

Neither are solar or wind farms.

What's wrong with wind farms? They literally take no land, you can graze cattle / farm right under them.

They decimate bird populations.

Natural gas absolutely can be ramped up and down and that’s critical to the stability of our grid. Just because there are other fossil fuel plants that can’t doesn’t somehow downplay the critical nature of peaker plants.

> We also have energy storage technologies other than batteries: the most common being pumped-storage hydroelectricity

These take tons of land and aren’t really anything like reverse hydro electric dams because if you tried to reverse a dam during high production you will be running a river dry and run out of water quickly.

> as most fossil fuel and nuclear power can't be easily or efficiently ramped down either.

Ramping it down is irrelevant, you can always use the extra to process aluminium at cheaper rates, pump water back in reservoir, etc. The ramping up is critical, as you expect the light to turn on when you flick the switch, not 3h later.

Every time I see hysterical comments (not from you) about how terrible it would be if excess solar and wing power regularly crashed the price of electricity I see and opportunity instead. Seriously regular periods where electric power is 'free' is a serious opportunity. You could smelt aluminum, electrowin iron. Run electric furnaces in mini-mills.

I don't think it's that simple. What does your aluminum smelting plant do when the power isn't free, sit there unproductively? Or purchase electricity when it's expensive? Either will cost you money. Maybe you can still come out ahead in either of these cases, but that depends on quite a few factors.

Even if you are running an aluminum smelter 24/7 having periods when the price of electricity goes to zero helps your bottom line a lot. Aluminum is difficult case because the electrolytic cells have to be kept hot. I've saw paper describing electrowinning iron. The process used aqueous sulfate electrolyte. One assume the capital cost of the cells themselves would be minimal.

Yes but if the average price of the clean energy is sufficiently high, brief periods of free clean energy may not be enough to offset that.

It's the opposite. Low whole sale energy prices in Germany cause policy failures that increase the EEG surcharge.

Here are some made up numbers. If the wholesale price is 0.10€ per kWh but the subsidized price is 0.20€ per kWh then the EEG budget has to make up the 0.10€ difference. That difference is added to the retail price so you end up paying 0.10€ for the electricity itself and 0.10€ per kWh for the EEG surcharge.

Industrial users are exempt from the surcharge so consumers end up paying electricity bills for the biggest energy consumers.

You actually end up paying 0.10€ for the electricity, 0.10€ for the EEG surcharge and 0.05€ for the EEG portion that industrial companies didn't pay for. Total: 0.25€ which is higher than even the subsidized price. Meanwhile industrial users only pay the wholesale price of 0.10€.

Of course I have simplified grid costs and taxes but this is the general problem with the EEG surcharge.

You have variable pricing on electricity, and heavy industrial users build up business plans that take advantage of the market. Prices will adjust until the inefficiencies are eaten up by market forces.

It’s not “maybe you can come out ahead” the occasional abundance of very cheap electricity creates an opportunity that will be filled by whoever can make the most of it.

> most [...] nuclear power can't be easily or efficiently ramped down either.

As you say, most, but naval reactors are very responsive. Responsive reactors can be designed when that property is desired.

Sure you can reduce power production, just insert control rods, but that comes at a severe cost to efficiency, costs that at large scale become far more of a concern, especially when nuclear is already nearly prohibitively expensive.

It's not just about control rods. Naval reactors also ramp up fast, which allows you to ramp them down more often without incurring long ramp-up periods. They're very good reactors, and the USN has a better safety record for running reactors than any other organization on the planet.

FWIW long distance transmission lines work too. Solar peaks a few hours earlier than demand. So ship that power east two timezones.

And which solar panels power California?

Batteries and hopefully offshore wind. And people will time shift energy consumption when incentivized. If we were able to charge cars at work, for example, that would shift massive amounts of load to peak solar hours, when electricity will be essentially free on the generation side, and transmission/congestion costs will be the major factor.

We are well on our way to making enough batteries to replace our car fleet. If we do that, we will have already built the industrial manufacturing capacity to store several days worth of energy.

I forget the numbers, but a huge proportion of wind and solar installs are planning foe good chunks of battery right now. The contracts I've seen indicate a storage cost of $80-$110/MWh, which is cheaper than new nuclear, and pretty much in coal territory.

The future is one of super abundant renewable energy, with the huge fractions of it thrown away, unless somebody can figure out used for it or ways to transmit/store it.

If California is the only market powered by batteries, that's a huge reduction in battery-scope. Washington has a lot of hydro, etc.

It is relevant to the improved battery tech problem however and makes the task less huge. Wind is also fairly reliable in that the day night cycle alone creates it since it is created by temperature differentials. Across a broad area it is impossible for there to not be wind short of being messed with by hyperadvanced aliens scenarios like ensuring all areas of the earth are being heated equally.

Reply on news.ycombinator.com