Hacker News Reader

Comment by radicalbyte | Hacker News Reader

I was skeptical until I saw how good my sun panels work. We only have installed about 50% of capacity and they generate enough to cover 120% of our annual electric usage.

Everything here is electric except the heating and I work from home all day (with an energy-drinking PC set-up).

I live in The Netherlands - the same panels will generate about 2x as much energy a year if they were in Spain or North Africa.

I wouldn't be surprised if we could run our houses and light industry for 6 months a year purely on solar panels if we put them on every roof top in the country.

Then you look at wind: we have a huge potential for wind energy especially off-shore (and in our big lakes).

Like most of the EU the main problem is heating in winter - that requires a reduction in energy usage through insulation and heat-pumps combined with energy import from countries which have high potential for solar and wind (North Africa has huge potential here).

View on HN

I think the new question then becomes, can we make enough solar panels for everyone without destroying the Earth?

A more pertinent question is possibly can we make enough storage to smooth the supply sufficiently to allow adoption of solar as the principal electricity supply technology.

Seems like we're going to need to become more diurnal and more seasonal in our approach to industry? Make hay while the sun shines, as the idiom says.

We're getting to the point where wind > hydrogen > combustion will be cheap enough to cover that kind of swing. We're going to be connecting electric systems over a huge landmass which will resolve much of the non-seasonal swings.

Sodium-sulfur batteries are made of two very abundant elements, where the other active elements are mostly aluminum oxides and the structural elements are iron and carbon.

Hydrogen is hard to store, but there is sufficient space for satisfying most of our needs by itself. Hydrogen generators are also mostly made of iron, carbon, and aluminum, or alternatively a lot of organic chemicals and aluminum.

Pumped hydro has some problems to compete on the open market, but it also can satisfy most of our storage requirements by itself. Pumped hydro systems are made mostly of steel.

What kinds of resources do you expect we to run out of?

Hydrogen is extremely easy to store underground. It is expensive to make (i.e. inefficient and capital intensive) and to move around with compressors but actually storing it if you have access to salt domes and depleted gas fields is not hard at all.

If the thesis is that electrical energy will be hyper abundant 80% of the time and deeply scarce the remaining 20% (which is the case if you use only solar and wind) and that periods of scarcity are correlated - so a "bad" day is likely to be preceded and followed by other "bad" days, then large scale hydrogen storage is quite a good match to get you through your worst week.

"Extremely" is really not a good way to describe it.

Yes, salt domes are quite viable. They are more the "expensive" kind of storage, and require a lot of scale. They are probably not economically viable for electricity storage as it's usually understood, but they are cheap enough that it's perfectly viable to dig them at a loss for societal assurance. They are certainly a lesser drag on the economy than nuclear power.

AFAIK the vast majority of depleted gas fields can't be used for hydrogen storage. The ones that can are obviously a major gain, but those aren't enough to sustain a society.

Storing hydrogen is perfectly viable, and won't have any catastrophic impact on electricity prices. But the "extremely easy" is a bit too optimistic.

I think industries that are overwhelmingly dependent on energy often are like that now. Energy prices vary throughout the day and some industries (aluminum?) are very price sensitive.

Yes. Quite easily, the materials are common. They require energy to process and ship, but that does not have to come from fossil fuels.

I'm confident we'll get there. There will be lots of mitigating the damage from climate change, because that train wreck is still in motion. But we will mostly decarbonize our economies by the end of this century.

> But we will mostly decarbonize our economies by the end of this century.

Unfortunately unless we completely decarbonize by the middle of the century, we'll reach catastrophic levels of global warming (>3 degrees warming, likely leading to the polar ice caps melting and significant ocean level rises, especially at the equator).

Once we have mostly decarbonized there are many techniques for capturing carbon if you have enough energy

Perhaps, though we would need to do that long before the turn of the century. People just don't want to believe how little time there is left to stop catastrophic global warming.

All the trends I've read recently put us to going to no higher than 3 degrees of warming. The trends that put us toward over 3 degrees of warming were made around the time period of ~2010 when the trends were much worse than they turned out to be over the 2010s.

3 degrees is very bad, but it's quite survivable. We can of course always do better. But right now I'm quite happy with the direction things are going and the current trends but still push for trying to do more where we can.

Most economies are aiming for net zero by 2050.

A goal, which it is now clear, that would have been absolutely trivial to meet while making a solid profit on the deal, if we'd just started earlier, like when the problem became widely known four decades ago.

But we didn't. So we can probably still do it, and we can probably still come out richer, but some people are still trying to slow it down so they can profit from the short term chaos and impoverishing and risking our lives to do so.

The only possible solution four decades ago was nuclear, and even then it was only a partial solution, and which people oppose for illogical reasons. The story now is much more hopeful. I also hope nuclear can be a part of it, but I think battery advances are more likely to beat it to the starting line.

Climate change is not just about electricity production.

There was a wide array of solutions available 40 years ago, including, but not limited to nuclear.

Notable mentions: carbon fees, preventing deforestation, not subsidizing fossil fuels, not actively encouraging methane release from oilfields, more efficient cars, trains, insulation, heat pumps and solar water heaters but really the list is very, very long.

> Climate change is not just about electricity production.

Agreed, but that was the part that could be addressed with nuclear.

> There was a wide array of solutions available 40 years ago, including, but not limited to nuclear.

I think we're in agreement, except for the word "solution". The things you mention are not solutions, but rather mitigations, bandaids, things that slow the pace of climate change. Which still would have been extremely useful to buy us more time to decarbonize.

> Most economies are aiming for net zero by 2050.

We emitted what the eartg can absorb (3Gteq) back in 1990. Since then, emissions trippled.

Net zero is not enough, there is still a lot of CO2 in the atmosphere, so the temperature will reach +3°C.

Net Zero by 2050 and peaking global emissions before 2030 should be enough to meet the 1.5C target according to the IPCC.

Assuming we follow that plan and don't hit any tipping points, we should stay under 3C, but yeah it's possible.

I think three degrees of warming is unavoidable at this point. The massive coordination and effort it would take to avoid that seems like wishful thinking at best. I still hope I'm pleasantly surprised here, but I'm skeptical.

It's going to come down to mitigating, and that might mean different things in different places.

I think there will be some amount of geo-engineering to cool the poles and avoid some of the sea level rise and permafrost melting. There will also be some active carbon removal, but we also will not reach zero emissions either. That may mostly balance out, or reach "net zero".

There will also likely be a slow migration away from low lying coastal areas, as well as seawall construction.

The biggest risk to this is that efforts to decarbonize will lead to cost of living increases that disproportionately impact the poor and working class, leading to a populist uprising against these measures and an enthusiastic return to burning coal. (Coal remains the cheapest way to generate energy in most places.)

We have already seen a little bit of this. It's only been a preview of what a more serious revolt would look like.

It's important to realize that anything that increases energy costs is regressive. The wealthy can afford these cost increases and their lifestyle is barely impacted. For the poor and working class increases in these costs can be absolutely devastating.

The upper classes can also afford EVs, solar panels, batteries, and can very often telework at least some of the time. Working classes are stuck with older technology that is more expensive to fuel and generally have to commute to work jobs in person with no option for remote work.

You do a CO2 dividend and the problem is solved. There is no easier problem than that. It is literally one person one vote.

From my perspective people hate democracy and want people to be unequal by birth.

How often do people with power make decisions for the greater good?

Maybe I'm cynical, but the observer who said that the problem with political power is that the worst people will do anything they can to take it (as a result, good people won't have it)

I don’t see how decarbonization will lead to a cost of living increase for the poor and working classes. Do you have any sources for this?

Solar and batteries are still rapidly dropping in price. At the same time, the fossil fuel industry is engaging in rampant hostage taking (Putin) and price gouging (oil is under $100/barrel. Why is gasoline 2x what it was the first time oil was at this price?)

If anything, I’d expect decentralized generation of renewable power to stabilize energy costs, eliminate wars for oil, and create well paying jobs for unskilled labor.

As for EV affordability: All EVs in the US have a 10 year battery warranty (or lose subsidies), which should create gluts in the used car market moving forward.

All of those things help the poor and working class: lower, more stable energy prices, better jobs, fewer wars to fight and cheaper used cars are each individually huge wins.

We have to. The alternative is destroying the earth even more.

And overall, the lifetime environmental impact of solar panels is much lower than producing the same energy using fossile fuels.

There is also the option of winding down energy consumption - we produce so much useless crap for no particular reason anyway.

I think that would need to go hand in hand with social change to allow people greater quality of life outside of consumerism (e.g. greater economic stability, autonomy, home ownership, access to education, etc.).

Right now it seems that the social contract is "give up your autonomy and stability to capitalism and in return you get to consume lots of stuff". If we remove the "consume lots of stuff" part of the equation, then I doubt people would be happy with the "give up your autonomy and stability" part.

Energy efficiency is skyrocketing though. My house uses 25% the energy it would with if we installed 20 year old appliances (even ignoring improvements to insulation).

Yes you're absolutely right

You might be generating 120% but what happens if everybody installs?

Do you have storage installed to smooth out the dead hours?

Unfortunately the answer is that you need days of storage for solar and wind to be anywhere near 100% of the energy mix.

The cost of that is astronomical.

That's true, and it's why other sources of energy are useful still, such as wind and nuclear and hydro.

But I also think we'll never get to the point that everyone has solar installed. There are plenty of situations where it doesn't make sense, such as properties with lots of taller trees around short buildings.

Also, generating 120% generally doesn't make sense because the payout from the electric company for the other 20% is pitiful.

A battery makes sense for a lot of people for a few reasons, too, so the power company doesn't necessarily have to be the one to do all of that, either. We did it mainly for relief from power outages, but it also works well for places with prices that change by time of day.

Solar is so damn cheap it almost makes sense to set up supplemental systems even if you only get like 10% of the power rated on the panel.

Panels are like $1 per nominal watt

A comment like this comes up in every discussion of renewables and needs to die. We don't need separate stationery storage, enormous storage capacity will be free as car batteries. EVs will have 17 times the battery capacity of stationary storage systems by 2030 [1].

By the time we completely transition to EVs, it will be 100s of times of storage. All this storage capacity is available at zero cost, people buy cars anyways. It is decentralized, no single point of failure. This is the sweetest kind of energy demand, can charge whenever electricity prices are negative or zero. Cars are parked nearly all the time, grid adaptations will evolve demand management. Cars will supply power back to the grid at its highest prices, providing a source of income for EV owners.

[1] https://assets.bbhub.io/professional/sites/24/BloombergNEF-Z...

Power prices have been going negative and spiking up 100x in the UK for a decade or more.

Yet the vast majority of EV's don't have the ability to discharge into the grid to earn money... Most can't even charge when power is negatively priced unless you pay for a third party charger and service.

Basically, Car owners, charger owners, Utility companies, and EV manufacturers would all like to earn the money by charging/discharging your EV at times to earn the most money. But all of those people don't want someone else being able to take that sweet sweet revenue. So they put roadblocks in the way.

EV's usually have the capability to charge into the grid, but firmware that won't allow it.

Chargers (AC and DC) won't allow power to flow back into the grid, and will disconnect any car who tries.

Utilities refuse to let home users take part in the minute by minute pricing - instead the best available is dynamic pricing set half hourly a day in advance.

And users want to do this, but have three parties who've assembled roadblocks in the way. And there is no indication of that impasse ending anytime soon.

>Utilities refuse to let home users take part in the minute by minute pricing - instead the best available is dynamic pricing set half hourly a day in advance.

That seems reasonable considering what happened with griddy in texas.

There are better approaches. For example a supplier could sell you 5 kWh per day at a fixed price, and anything extra you use is at instant market price, with some kind of pay-as-you-use system.

Then if the prices go sky high, users who chose that plan at least get to use 5kWh each day. 5kWh is plenty to not freeze and to keep your phone charged and fridge going. If you want more, you can pay more.

The UK already has a system of electricity meters that have a 'balance', and when the balance reaches zero it cuts off till more credit is added. Such a scheme integrates well with that, and means a user can never rack up a massive unexpected bill if the price goes crazy high or low.

> For example a supplier could sell you 5 kWh per day at a fixed price, and anything extra you use is at instant market price, with some kind of pay-as-you-use system.

I suspect 5 kWh/day is way to little to keep the house warm and still be able to cook food. The exact number doesn't really matter here. What's more problematic is how the 5 kWh of guaranteed/insured/subsidized power would be allocated.

If it's your first 5 kWh/day, it will be you'll blow past the quota by noon, and if there was a price spike in the evening you would pay the full price of it, or shut off your breaker until prices go down.

Having it as a some sort of pool you can draw from anytime you want doesn't suffer from the above problem, but would make such a service very expensive to run for the counterparty due to adverse selection. Anyone using such a service would opt to use the market price when electricity is plentiful/cheap, and switch to fixed price when electricity is scarce/expensive. The net result is that the 5 kWh ends up always being the most expensive electricity for the day, and a service providing such a guarantee will be priced accordingly. In a competitive market, you would expect the price of a fixed price plan to be the same as the "5 kWh per day at fixed price and anything extra at instant market price" plan that you propose. You might be tempted to argue that utilities/the government can subsidize the latter plan as some sort of green energy initiative, but that would be be bad. The most expensive energy also tends to be the dirtiest, because peaker plants (power plants that can quickly turn on/off) tend to run on fossil fuels. Subsidizing such a scheme would also be subsidizing electricity consumption during peak hours, and therefore subsidizing fossil fuel plants.

5/24 kWh per rolling hour, with an audible beep if you'll run out of credit before the end of the hour would be a suitable way to do it.

Then you can hear the beep and either decide to turn off a few things, top up your balance, hit the 'emergency 1 kWh' button, or just sit in the dark for 15 minutes.

Most of the benefit can be achieved in one step by putting an 'eco mode' button on home chargers that makes it dispense power if and only if there is curtailment (which can be looked up over network) or the car battery is critically low. Make this an overridable default with some messaging about owners doing their part for a greener future to make it palateable and you don't need the battery to feed back out for the EV to be acting as storage.

This benefits the grid operator because they get to massively overcharge for curtailed energy. It benefits the EV manufacturer because they get to greenwash. And the consumer doesn't get a say because they don't actually own their DRM'd remote controlled car and charger.

Hooking up the car to the roof solar is still very expensive. It needs to be done for every house. What if the car is at work and someone needs the battery power at home?

Would you be drawing down the battery on your car every night and leaving in the morning on an empty tank?

Car batteries are not going to be a substantial source of grid energy. You need those for driving and additionally the battery chemistries used in cars isn't optimized for the kind of loading that a grid would put on it.

>On a road trip

>Stop for lunch, plug in car to recharge

>Cloud passes in front of sun

>Grid empties car

People won't want to put their plans at the mercy of a fluctuating grid. There will never be a time when people are happy to plug in their car and come back to find it empty, electricity spot price be damned. It's my car, not your grid storage system.

I keep hearing that a problem with solar is that it doesn't work at night. So I can see that as a (solvable) problem for industry, but for personal life?

This idea that you need to have your lights on all night is rather recent. For a very very long time people survived by following the natural rythms of daylight. Some people even think this is healthier, and suggest special apps to adjust screen brightness to follow natural light patterns.

A good fridge/freezer should hold its cold overnight, especially if you aren't snacking at midnight.

So while it might be true that your current lifestyle is based on 24/7 electricity, but what makes that lifestyle sacrosanct? You'll probably change it over the next few years anyway, and odds are it is different today than it was 5 years ago.

> This idea that you need to have your lights on all night is rather recent. For a very very long time people survived by following the natural rythms of daylight.

People have used fire, candles for a very very long time to provide light at night.

Lighting is a very small fraction of our energy usage (about 5%): https://www.eia.gov/tools/faqs/faq.php?id=99&t=3

If we built our houses right, we could drastically shift energy usage requirements for heating and cooling to daylight hours (utilizing thermal mass storage like https://www.steffes.com/ets/ )

Also you can't underestimate moon light either in premodern times. Deep in the trees or an urban landscape with tall buildings all around it is less reliable, but even a half moon on a clear night without washing your vision out with bright artificial lights leaves you with a ton of light. And even when overcast you aren't completely blinded at night as long as you have a decent amount of sky above you.

I think the idea of people sleeping as soon as the sun is down and not just doing low-light activities even 3000 years ago as ridiculous despite being a pervasive idea.

Would be especially inconvenient in places where you have daylight from 9AM till 4PM and where people rely on electric heating.

The answer actually isn’t that you need days of storage. A researcher I follow on Twitter here in Australia called David Osmond has been running a simulation for over a year that scales up the wind and solar we have to the levels we would need to achieve around 100% renewable energy, and using real time data simulates storage requirements.

He’s found that for the NEM grid in Australia at least, just six hours of storage can bring us to over 99% of demand being provided by renewables and storage over a month, almost every month. On a weekly basis, some weeks it’s 100%, but worst case will be something like two or three percent being provided by something like gas turbines over the week.

We have more pumped hydro coming online and will have more in the future, so not all those six or so hours worth will be chemical batteries.

Part of what makes this work is because on the eastern seaboard, often it might be, for example, windy in northern Queensland while it’s calm in South Australia, so being interconnected helps a lot. At this point we really don’t have (and it’s not simulating) any kind of serious demand management (apart from in emergencies, but I’m talking everyday kind of management), which could potentially remove some of the need for gas peaking. We also don’t have any offshore wind yet (legislation only legalised it last year and there are some early projects). As some of that offshore generation comes online with better capacity factors than onshore, it will likely make things more consistent.

Is there any code for that simulation available publicly? I'd like to plug in numbers for other locations.

I don't think so unfortunately, he just posts the results, e.g. https://twitter.com/DavidOsmond8/status/1590215438323322880

How many gigawatt hours is 6 hours of storage though? Also is this just to cover current electricity demand or total primary energy?

In Australia, based on current electricity demand it's 24 GW / 120 GWh.

Primary energy replacement obviously will require more than that, but it won't be so bad as people make out, since electrification can drive massive efficiency increases - electric cars are 3-4x as energy efficient as ICE cars, and there are very few places in the parts of Australia that get decently cold that are too cold for air source heat pumps, so that's a 3x efficiency increase over gas space and hot water heating (even as much as 5x for CO2 hot water heat pumps!).

You need those interconnects though.

Here in Norway the electricity prices this summer and fall were up to 100x higher in the south compared to the north, due to lack of interconnect capacity between the north and south.

Building hundreds of miles of high-capacity interconnects ain't done in a weekend either.

People misunderstand the impact of interconnects.

It's like saying, this remote town in the hills has cheap housing because it takes so long to drive to it so lets build a high speed express train right to the city. Then we can have cheap housing and access to the city.

Except you can't, because as soon as you have access to the market demand, the price goes up to match the market price.

Interconnects are useful, but people wanting to build interconnects so they can access the cheap power from places that only had cheap power due to a lack of interconnects is upside down thinking.

Oh we've noticed that as well. Our interconnects to Germany and UK certainly contributed to the high prices in the south this year.

But better internal interconnects between south and north would certainly have contributed to averaging out the prices across the country, avoiding the extremes we saw this year.

After all, there were hydro plants in the north which halted production due to the low prices at the same time as we had prices >10x normal in the south.

What do you mean by astronomical? My back-of-the-napkin estimate is that you'd pay about $8k a year in battery costs to supply a household ($20k 18kWh battery system lasting 5 years, then double it).

Obviously that's missing commercial power, utility-scale storage, heavy industry, electric vehicle charging, etc., but it seems like factoring those in would give you a big yet feasible cost, especially if storage gets cheaper over the next decades.

$20K buys closer 30kWh these days, and they last more like 10 years instead of 5, so $8k/year is a wild overestimate.

That strengths your point.

However, 30kWh isn’t quite enough for heating and cooling an all-electric off-grid house, unless you carefully monitor battery capacity, etc.

The batteries alone have actually gotten quite affordable. https://signaturesolar.com/eg4-lifepower4-lithium-batteries-...

Here is 30kWh of LFP batteries for $9,000. Now that doesn't include the inverters or installation or anything but LFP batteries have excellent longevity, in this case 7000 cycles to 80% depth-of-discharge. That means charging and discharging the battery every day for 19 years and at that point, you still have 80% of the remaining capacity.

> except heating

How much of an impact did this actually make, though, since heating was probably the majority of your emissions to begin with?

Solar may be sexy, but depending on your heat source you possibly could have done better by installing a heat pump on grid power.

Really depends on the insulation if you're in Western Europe close to the sea (making the winters less harsh). I live in a flat and the whole complex has been upgraded with modern insulation and modern windows about five years ago. I turn on the heat on a few days when I feel exceptionally cold, otherwise I don't and it still maintains 18-19° (being chubby helps, that's my comfort temperature). It's not because my neighbors are doing the heating for me -- I've talked to them, and they barely ever turn it up, either.

Granted, that would change somewhat if we get a super harsh winter here, but otherwise: I consume much more electrical power than heating.

Reply on news.ycombinator.com