Hacker News new | past | comments | ask | show | jobs | submit login
Something funny is about to happen to some prices (ifost.org.au)
105 points by solresol 29 days ago | hide | past | favorite | 98 comments



This is built in to wholesale energy markets and well understood. Energy prices go negative with some degree of regularity and it is built into the system.

If you already have a grid interconnection and a handy energy sink, you can definitely make money just buying excess power for a negative price at certain times. So far, though, it's not quite frequent or extreme enough to justify a bespoke infrastructure investment.

And if you're going to make some kind of investment, better to spend money on some kind of energy storage tech (i.e a chemical or physical battery.) Then you not only take excess power off the system when it isn't wanted, but also help power the evening peak when usage ramps up just as solar is ramping down.


Iceland developed a aluminium-smelting industry to make use of the country’s plentiful (cheap) geothermal energy supply.


Except that aluminum smelting requires near constant power levels or the smelting pots will freeze and need to be replaced. So it's not really relevant here.


The smelteries use hydropower, and the power wasn't just laying around, each smeltery requires building a massive dedicated hydropower plant.


Fun fact: this is why the US aerospace industry grew up in the pacific northwest.

At the beginning of the 20th century we went on a bit of a dam building spree up there (and all over, but heavily up there). Then when WWII broke out and we got a huge demand for airplanes (which require light aluminum), the industry congregated around where there was an excess of cheap hydro. Hence Boeing, Lockheed, and friends all being out of Seattle.



How does one enter the market to buy negative priced energy? How much money could I make if I boiled off my swimming pool?


In the UK there is a provider than bills the "live" electricity price using smart meters, (in reality I think it's 30 min chunks but close enough). Sometimes at low demand and high wind and solar output prices go negative.

https://www.energy-stats.uk/octopus-agile-london/. I think the all time low was -10p/kWh. Most of the time it's around -3p though when it does go negative. Assuming you had a 20kW (massive) heater you'd only make maybe £1-5/day running it when the prices are negative, which isn't hugely often ATM. So probably not economical, yet.


I think some people with a pool would be totally fine, esp in colder months, using it as an energy sink, make it possibly more pleasant / usable.


The author mentioned just such an energy storage tech: water.


To be fair, they didn't really mention water as a battery. Yes, they used the phrase "bottled electricity", but that was in reference to the large energy cost of desalination. You can't then get the energy back out like you could with an elevated lake.


but you can schedule large swathes of energy intensive industries that don't rely on human fingers so that they happen at cheap power and no longer happen at expensive power.

humans don't desalinate water with their bare hands...


Water (pumped hydro) storage is really great, but only works well at scale where the geographic features support it.


Existing hydro (not pumped) also acts as storage: a decrease in the load on a normal hydro dam means more stored power... and all the infrastructure exists including grid connection.


It a lot of cases that doesn't happen. The water levels in the headwaters and the amount of flow on the downstream river are controlled with goals other than power production. Like not causing floods, and having enough water depth for navigation. In Ontario, Canada, if the plant doesn't run at capacity, generally the water is sent through the spillway without generating power. I'd assume that to be common in many hydro-electric installations.


the article mentioned desalinating water as a way to use negatively priced energy and storing the value in a commodity for sale or consumption at a later date, not pumped hydro storage


Stored water for hydro power generation is also a commodity for later sale or consumption.


The problem with water is that to make it cost-effective requires appropriate geologic configuration and massive infrastructure investment. In California for example, all the appropriate locations are used, there are no more viable locations (according to the USGS)


What about electric water heaters? I am unfamiliar with the usage patterns and losses over time, but having my home water heater having some variability in temperature would be ok with me — I’d simply adjust the handle in the shower.


Colleagues of mine are working on a system using these kinds of water heaters. Specifically, they are working for a standardized protocol to allow these devices to negotiate prices / indicate to the grid how much they could take.

From what I heard they said something like a half-time for the heat in these devices of about 1 to 2 days.


Could do the same with freezers. And maybe a few degrees with fridges.

And boost those with phase-change materials at the right thaw points.


Interesting idea. It'd have to be done carefully because there are problems with going too cold or too hot:

* Too cold: Legionnaire's disease. Presumably not right away, so maybe there's some wiggle room if you're careful about timing.

* Too hot: scalding, particularly if you have kids. Maybe a new design could have a thermostatic valve on the output to give more flexibility.


Thermostatic valves are common in institutional water heating installations.

There are also quite a few people who use thermal storage with wood fired boilers. The boilers burn the most efficient and cleanest when they are near full output, then hot water is stored in an insulated tank to intermittently supply the house with heat.

Description of one such system: https://www.newhorizoncorp.com/PDF/ekomanual.pdf


With a properly installed thermostatic mixing valve, you don't even need that. The water will always exit at a safe, constant-ish temperature regardless of the water temperature in the heater, by mixing in the cold supply.


You can't build pumped hydro storage. You can only "discover" suitable sites.


Yeah, I was surprised by the OP’s assertion that energy prices never go negative. I think spot prices have been negative in eg Texas on windy (and maybe also sunny) days.


They have gone negative briefly in Germany, as well. Both still rare occurrences at this point, though.


Briefly? It's very common. Even UK _retail_ energy prices go negative a fair bit. https://www.energy-stats.uk/octopus-agile-london/


Due to issues with transferring energy, German producers will often pay Netherlands ones to shutdown their windfarms and export electricity there.


An interesting thing is happening because of Colorado's new anti-flaring law. They are filling a shipping container with GPUs and dropping it off at well-sites to mine bitcoins. Seems silly to me but apparently they think it will be profitable.

https://www.crusoeenergy.com/news-and-media/2020/1/10/denver...


There is a company (whose link I can’t find) who will drop a natural gas compressor and storage unit for capture and storage. Not sure if the economics work, but they’re making the attempt.

Edit: Found it, Cryobox. Seems like a great way to monetize abandoned wells that are emitting methane.

https://www.verdek.com/Nano-LNG-Station.htm


Before the natural gas enters a gas turbine it has to be pressurised. This takes some energy and it can be done beforehand. Power station generates electricity at times when it is expensive and compresses the natural gas when power is cheap. Compression usually takes around .8% of turbine power output.


The device my comment mentions is for compression and capture on site at a well where there is no pipeline or electric transmission infrastructure. The compressed natural gas is hauled by truck for export at a pipeline endpoint.


Which means that someone, somewhere is going to get paid to create a dummy electrical load. We've never (as far as I can tell) seen "grid stability" payments for using electricity before.

Kinda, sorta. The whole conceit behind wholesale energy markets is balancing congestion on the system by changing the locational marginal prices (LMPs). If you have a very congested line, a generator with a negative shift factor (their generation will reduce flow on the line relative to the constraint) will see high prices to turn up, while generators on the other side with positive shift factors (will put more flow on the line and exacerbate congestion), will see low-to-negative prices to turn off.

There is a lot of development in these markets (in terms of energy, capacity, and ancillary services) with respect to energy storage resources, and co-located (solar+storage) facilities. If you want to know more about that, reading up on FERC Order 841[1], and the various RTO/ISO implementations to date is a good place to start.

[1] https://www.energy-storage.news/news/ferc-order-841-us-about...


Can’t excess power be used to remove carbon dioxide from the air?


Yes, if the equipment isn't too expensive.


See also (in the UK): https://octopus.energy/agile/ which can go (and has gone) negative, to the retail consumer, which seems to be a step further than TFA.


I mostly dislike/distrust smart meters in the U.K. but this seems a pretty good use of the technology.


Ditto, has an API too. I'd sign up in an instant if only it wasn't (typically, y'know, when not paying me to consume!) much more expensive than my provider which just tracks wholesale price (monthly, plus margin of course).


In no time someone will come up with a box that will switch to your neighbor’s power. One will have this and the other one the classic one. You will both use the cheaper one at a given time. That will give you the access to the highly volatile power prices with a cap in more expensive ones. People usually have to pay O(N) money for things like this on the exchanges. You’ll have it for O(1) money. Or maybe just resell the power to the neighbor with a small discount compared to a regular utility price and pack the difference. The box should pay for itself in no time.


Installation is regulated and, from what I understand, these boxes' tamper resistance is more sophisticated than guerrilla IEDs. (That is to say, not necessarily 'sophisticated', but a non-trivial amount of 'you're doing something wrong if you're trying to defeat me' counter-measures.)


What if the two neighbours share an electric vehicle or two, and swap which garage they are in based on where it is cheapest to charge? The car that isn't charging can then supply electricity to the other house.


'Supply electricity to the other house' is the bit that you can't do.

I mean, by all means power everything from your car and stop using the mains supply to your house - but that sounds pretty inconvenient to me - you can't interfere with the supply as a DIYer, and I doubt there's any way to do it in a regulator-approved way short of completely disconnecting from the grid, not off and on based on price.


The great thing about electricity (like fresh water) is that it's the ultimate fungible good and there's a huge amount of useful work we can do with it. The problem is getting it from point A from point B without losing it.


There's some articles about getting Australia to dump it's spare energy into making Ammonia that can then be relatively easily exported.


There have been negative electricity (spot) prices already - https://www.montelnews.com/en/story/sweden-finland-see-negat....


Negative prices in summer are going to hamper investment in generation plant needed for winter. PV is great while the sun shines, and for a few hours afterwards, not so great when there's hardly any sun for months.

Extending out the time to profitability (equivalently, reducing ROI; because the plants are idle in summer) means many proposals for generation in winter won't rise above the hurdle rate of ROI and won't get built.

Some regulation might be necessary. Maybe it should be that an energy provider must commit to a guaranteed minimum power output that applies year-round (of course excepting well notified planned maintenance and small outages for equipment failure).

IOW, a provider uses PV (perhaps) as one of a portfolio of generation methods, whatever works cheapest for reliable supply year round in the particular circumstances of the provider and market.

Allowing point of use (rooftop) installs of PV will make seasonal demand on the grid peakier too.

People will say, all those EVs are a distributed energy source. For seasonal variations, they're not.


very quickly we're seeing battery grid tied which eliminates most if not all of your concerns


I'm not sure I buy the possibility of negative water prices predicted in this article. Water is pretty easy to store until the market is willing to pay for it.


It also doesn't even make sense. If you had to pay to get rid of fresh water, why wouldn't you just put it back in the ocean?


The idea that the inputs to an industrial process could go so negative in price that the outputs would also go negative in price is a fantasy.

Let's say A is running a desalinated water giveaway and B is storing the excess electricity as potential energy, which can be sold as electricity when the price goes positive.

B can afford to take the energy for a less-negative price, so outbids A by accepting less money with the juice.

Sure, if things get ridiculous enough, it might be a race to see who can afford to give away the most water / aluminium / nighttime power / Funko Pops.

But, it won't.


it happens at weird critical points like when the price of crude was negative in june/july when no one would take physical oil in oklahoma


Sure, but what the article was proposing is analogous to a situation where the price of crude becomes so negative that gas stations start giving away gasoline at the pump.


Prices are negative because nobody expected such a scenario. Otherwise someone would have bought it off for at least $0.


Or you could just turn off some solar panels for a while.


I don’t know much about solar tech, it sounds simple but how easy is this to do on a big scale? - eg does having an open circuit damage the panels if they are illuminated?


No, you just turn them off. It's a big photodiode.

However, what's quite common is that renewable energy is not sold at the spot price but instead at a "strike price" agreed at time of construction, which gives the stability to finance the project. So asking a plant to turn off their grid input will invoke a penalty clause for "curtailment". This is more common with wind turbines.


You can short or open them to achieve zero power. Neither is detrimental.


And I assume you can have open circuits on generators (from nuclear or other) -- there'd be somewhat less inductive drag on the generator, but I assume that's not a requirement for operation, esp. at low power. (Though I assume you don't want to short/shunt all your power unless you really want to heat something up...) Seems like you could at least be doing inertial storage by pumping water uphill or lifting huge cement blocks so you can harvest that potential energy later.


Opening the circuits on a steam turbine without reducing the incoming flow of steam is a bad idea: the 'inductive drag' on the generator is where a large fraction of energy is leaving the system! If you don't do anything the turbine will destroy itself fairly quickly. The next thing you can do is route the steam past the turbine, which will effectively waste some of the thermal energy through the condensors. However in most nuclear power plants these condensers aren't designed to take the full energy output of the reactor, so if you remove all the load from the system you'll need to reduce the reactor's power output, which is not a fast operation to reverse (fossil fuel based power stations will generally be able to react faster in this case).

But long and short is if you interrupt the power going out from a nuclear power station it will not be producing useful power for a while! (it won't damage anything but it'll take some time to get it back up to normal operation).


Generating electricity creates mechanical load on the generators that in turns allows to discharge the energy in generated high pressure steam.

At some point you hit a case where you can't go lower without shutting down completely, and that involves all parts of the system - just spinning up or down a steam turbine can be a harrowing moment as depending on the whole setup there could be an RPM range where the harmonic oscillation might destroy the whole thing (as my grandfather, who was working as engineer on steam turbines for electricity generation, described it: "we would slowly spin the turbine to the critical RPM range, then open the throttle to cross the critical range in as short time as possible, then do slow and steady to the rated RPM")


Well you can't just magically do nothing with all that energy, it's not something I know a lot about, but I assume the risks (i.e. reasons you cannot open circuit) include arcing across contacts, and excess heat build-up.


And since there is no load on a disconnected generator it will start to speed up until it destroys itself.


Generators act as a flywheel and store energy. If you abruptly disconnect them that energy still has to go somewhere.


"Solar panels are getting cheaper. The learning rate for solar energy is (probably) the steepest of any energy technology yet deployed. So we can be pretty much assured that within the next decade we'll have enough solar power to power anything we want"

Again, predicting the future is notoriously hard.

Current rate of progress cannot be extrapolated to infinity.

Most exponentials are S-curves in disguise.

Take a look at Moore's Law.


Moore's Law is alive and well. See the progress of GPUs.


Energy storage into hydrogen always made sense to me. Would make vehicles more attractive. Could be converted back to energy later onto the grid when it was cost effective. Certainly not efficient, but it is still useful work.


Hydrogen usually makes a lot of sense on a general theoretical level until you try it out. Consider that to get any useful work out of hydrogen you have to put it under high pressure. And if you put it under any pressure it leaks and seeps through any container, seeing how it is the smallest molecule known to man (or theoretically possible).

There are other ways to store electricity. You can store it into electrical heavy processes, such as making aluminum or steel from scrap. (Something that half researched blog post did not even mention).


Why does it need to be stored under high pressure? Compressing any gas is a waste of energy unless you have a need for those higher pressures, or space is an issue. If you're simply storing the hydrogen temporarily as a reserve, to later use for electricity generation, then storing it in large mylar bags, or silos at 8psi (close to isothermal expansion) or less seems the ideal way to go.


> then storing it in large mylar bags

May I suggest “Hindenburg” as a brand name?


Funny. Would you call an ice maker, Iceberg? Or name a generator after any of the fiery car or plane crashes?

I can assure you that hydrogen can be stored just as safely as other fuel sources. In many ways it is actually safer because it's escape velocity and the fact that it doesn't pool. Not compressing hydrogen would add to it's safety as there wouldn't be any stored forces to counteract, small leaks can be detected and don't cause turbulent ruptures.

Lithium ion batteries == spicy pillows


What few people note, is that during electrolysis of water, besides hydrogen you also produce oxygen. Oxygen is useful in gasification of organic waste products (corn stover, bigasse, and other woody biomass) to produce syngas that is not diluted by atmospheric nitrogen (a problem with using oxygen from atmospheric air). So not only are you producing hydrogen, which can later be converted into electricity or motive force (in a combustion engine), but also another form of fuel which can help to even out the energy needs when solar is not available. Up-cycling and recycling materials is a great use of excess electricity in my opinion.


Why not take this one step further? Surplus electricity produces green hydrogen produces conventional fuel using Fischer-Tropsch [0].

Not very efficient, but that would be an incredibly flexible green energy storage system.

[0] https://en.wikipedia.org/wiki/Fischer%E2%80%93Tropsch_proces...


there are relatively simple options for energy storage: lifting things up with the excess energy, then letting them back down if the energy is needed (at night). traditionally this is hydro, but it's also possible to do it with pulleys and concrete blocks.

edit: e.g. https://gravitricity.com/


I remember a similar concept that would work by stacking barrels of concrete using automated cranes. They had a press release like 2 years ago, and since then I never heard anything from them anymore.


The fixed costs of this are a problem, as the usual approach requires platinum electrodes.


The author points out that periodic negative prices for energy make inefficient processes a good use of capital, so no platinum required, just plain old electrolysis will be fine.


What is “plain old electrolysis”? If you do it carelessly, you risk producing the wrong outputs, and you also risk destroying your electrodes.


Well, someone will during the next decade have to make an absolutely ridiculous amount of batteries. Imagine what kind of profit margin those folks would get on free electricity!


Get paid to charge your batteries, then get paid later to discharge them! Elon is smiling. https://www.tesmanian.com/blogs/tesmanian-blog/neoens-revenu...


Do not forget that batteries are expensive and that somebody has to actually pay the negative prices. If prices were to go negative too much, the production would become unprofitable, it would then be cheaper to just give it up.


Use cheap (above, but close to zero) to liquify air, store liquids in one set of tanks, store the heat in salt solution in another.... there was a recent story about a company that was getting 60% round trip efficiency.... they could store energy at the rate the compressor worked.... I imagine getting energy back could happen quicker.

With large enough tanks, it's not unrealistic to imagine storing 5MW-10MW for several days worth.... so about 50-100 MW-hour.


It’s not at all obvious why the energy load during periods where you are paid to consume energy would be inefficient as opposed to shifting load from other times.


This happens regularly in Texas, and you can occasionally reduce your bill (very slightly) by turning on your electric devices if you use Griddy as a provider.


What about getting paid to winch a heavy object up a tower during the day, and then getting paid again to lower it again (generating energy) at night?


Idea is not new and it was actually implemented here https://en.m.wikipedia.org/wiki/Kruonis_Pumped_Storage_Plant :-)


What's this about a timeframe of a few minutes to 0% power for load-following coal or nuclear generation? Quite a strawman.


> A possible product that I think is likely to play out this way is water desalination. Desalinated water is very expensive because of the embedded energy cost: it's often described as "bottled electricity". Currently we tend to use reverse osmosis because it's more energy efficient; but if you don't have to worry about energy costs, you might choose instead to boil water and distil it: since the capital costs are cheaper.

Here's another possibility: Bitcoin mining.

It's been discussed a lot - mostly hypothetically, but has a fraction of the distribution cost. There is a non-negligible capital cost that will turn over periodically. But there's nothing to ship at the end of it all. And if you're getting paid to use power, then it might be possible to come out ahead.

And then there's the non-negligible cost of locating a desalination plant next to the ocean, where the effluent pollutes the environment.


Bitcoin mining is dominated by capex, not opex. It makes much more sense to buy your rigs and pay whatever you have to pay to keep them on 24/7.


Perhaps some limited utility for remote locations where methane is being flared off, because both capturing the methane and transmitting electrical power are too expensive.

Putting a generator in place to convert the methane to electricity, then crunching the value into bits on-site, might add up. I agree that it isn't going to be useful for periodic cheap energy, however; there's no scenario where intermittently running your ASICs is going to add up.


Don't believe it. If anything it'd allow for new businesses to flourish. I.e. Sea water filtering. It will gladly take all the extra electricity to create more fresh water.


“Dummy loads” are unnecessary - we could divert excess energy into desalination, carbon capture, or energy storage efforts (molten salt, for instance).


"someone, somewhere is going to get paid to create a dummy electrical load".

Did anyone else think of Crypto mining, where there #1 cost is electricity?


This isn't going to be funny because we need to eliminate all CO2 emissions before 2030 and these negative prices aren't encouraging.


I would say that negative prices are very _discouraging_ to continued use of fossil fuels, since it's hard to get investment for additional oil drilling capacity when the prices are negative.


That's true. Lots of oil companies have gone bankrupt.


This assumes we can't lay massive power cables across many latitudes so that one region's morning would power someones night.


Germany can't do it within it's own borders. SuedLink is basically a failure and if it ends up happening it's more of a miracle.




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

Search: