Comment by sephamorr
5 years ago
Power markets are more complicated than most people realize. One thing to note is that solar power can be cheaper than gas, but still not be economic. The fact of the matter is that an intermittent kWh is not as valuable as an on-demand reliable kWh to a utility who's number 1 priority is reliability. Even as solar is acquired at lower and lower prices, if evening power is generated by expensive and inefficient gas turbines, the customer might not see costs go down (and emissions might not go down either!). Solar is clearly economic in many markets, but we'll never get grid emissions in a place like California down much more without storage.
As mentioned in the article, solar power is insanely cheap. So while you are right that a kWh of solar is not, on average, as valuable as a kWh of semidispatchable coal, the insane cheapness means that at some point, it is economical to massively overbuild renewable and decommission fossil fuel plants. You provide no evidence for the implausible claim that solar causes emissions to go up: your argument is only that the emissions reducing effect is muted, which may be true.
Also, while storage would be helpful, it is not the only way to enable a renewable transition. Additional transmission is enormously helpful: as the sun goes down on California, the wind is picking up in the Midwest. And don’t forget demand response: if smart thermostats received price signals (maybe we should precool this house...) that would alleviate the evening ramp-up issue.
So I claim we’ll need less storage than “a whole day’s usage”. But the learning curve applies to batteries as well! This storage won’t cost as much anyway.
The whole issue of intermittency is overrated. While a single solar panel might generate intermittently, the solar fleet across a whole state generates more predictably.
I am predicting that grid emissions will come down a lot over the coming decades. Partially I’m predicting the past: they’ve already come down, a lot!
Exactly, the only thing limiting solar at this point is the rate of investment. Wind and hydro etc have a huge role, but for some back of the envelope estimates.
On the storage issue, at ~100$/kWh batteries that do a conservative 1,000 cycles are ~10c/kWh stored + generation costs + conversion inefficiency. Take current unsubsidized grid solar prices of 2c/kWh solar and double that for 4c/kWh as a conservative redundant safety margin. Tracking solar for example has much better morning and evening generation though at slightly higher prices.
If 2/3 of your electricity is at 4c/kWh and 1/3 is at 15c/kWh that’s 7.7c/kWh for pure solar 24/7 including peaking power needs. Obviously a specific mix of generation determines storage needs, but those are also really pessimistic estimates.
PS: Hydro power is 6.1% of the total U.S. electricity generation. If 80% of that is released at night that’s a huge reduction in storage needed. Similarly transmitting power east or west makes a large difference in storage needs.
Hydro isn't quite so fungible, I think a couple areas account for most the hydro generation.
It doesn't matter that if solar/wind is cheap (or even zero) because you still have to have an alternate power source for when your intermittent power has gone AWOL. You need cheap grid-scale storage, which doesn't exist at the moment.
Or some other dispatchable low-carbon energy source, which also doesn't really exist at the moment. Load-following nuclear reactors (hooked to thermal storage) could do this physically but a lot has to happen to get the cost down. Meanwhile, various CCS options could play a role if there's a solution for safely storing vast quantities of carbon out of the atmosphere.
I don't know about its efficacy, but my favorite means of mass energy storage is a gravity battery: literally stacking heavy blocks in a huge tower, which are raised and lowered based on energy demands by automated cranes.
https://en.wikipedia.org/wiki/Energy_Vault
6 replies →
Storage would be some combination of diurnal storage (batteries, say) and long term storage (hydrogen). The former benefits from high efficiency; the latter from lower capital cost. There is also thermal storage (an order of magnitude cheaper than batteries) for any application involving heating or cooling, including industrial users of heat.
Hydrogen... which can then be put back through a fuel cell to produce electrickery. The germans are looking at this. I believe.
> if smart thermostats received price signals (maybe we should precool this house...) that would alleviate the evening ramp-up issue.
Is there an existing model for retail intraday rates? Would intraday rates be desirable for all market participants?
"Add area for curtailment data?" https://github.com/tmrowco/electricitymap-contrib/issues/236...
Or just smarter appliances. Plenty of opportunities for thermal banking/battery banking.
Most of my loads, on average, don't need to run that exact second.
I don't mind if my hot water tank super-heats water in the middle of the day for the rest of the day. If electricity is really cheap, my freezer can jump into overdrive.
I don't need my clothes to dry in the next hour, just over the next 8.
I don't care if my fridge/freezer takes a break while I run the microwave or pre-heat the oven.
I don't care if my car charges ASAP as soon as I park, as long as it's charged by 8AM. And let it run as a grid-power bank for a fee.
Then you could have A/C systems that make ice or compress refrigerant in a tank.
I think the point is that in any case, price isn't a finish line for solar to replace everything. This process will be much longer and require lots of infrastructure changes in multiple places.
> I don't need my clothes to dry in the next hour, just over the next 8.
Then why not just hang them up for drying? Zero power consumption and they will dry in eight hours, maybe not under all conditions but under many.
I'm usually with you on that.
I'd even go a step ahead: dryers ruin clothes.
A bit diff in a humid area, or where the A/C would need to condense the added humidity (if dried indoors), or the furnace would have to counter-act the cooling from evaporation, but HVAC is usually more efficient than the dryer.
>I don't mind if my hot water tank super-heats water in the middle of the day for the rest of the day.
Water does not have the capability to store much energy.
>If electricity is really cheap, my freezer can jump into overdrive.
Your freezer can't if it's not a ammonia refrigerant. You will actually be wasting energy.
>I don't care if my fridge/freezer takes a break
Your fridge actually does not use electricity constantly. It detects the temperature and run the motor, stops it when it reaches the desired temp. It's already having a break.
It's common here to have 150L+ water tanks. Can take it from 50C to 75C and then let the anti-scald valve temper the output when drawing.
OK on the freezer. Could still take the fridge down to 2C.
The concept of the fridge taking a break while running other loads is to reduce peak draw current. If everyone did that, it would make the grid more stable.
8 replies →
> Water does not have the capability to store much energy.
Systems that make ice at night and use it for A/C during the day date back to the 1950s.
The general class of these facilities is phase-change materials. Water is pretty amazing, both for ice and for steam.
You may want to look into phase change materials for home heat and ac.
The block of ice or refrigerant liquified in a tank serve as PCMs unless we want to get really fancy :)
Intermittency and the unsuitability of grid storage batteries to compensate is heavily played up by the carbon industry.
The future of low emissions energy production will be largely driven by overproduction and demand shifting, not banks of grid level batteries.
This likely won't happen as soon in America, however, the economy is too tied in to fossil fuels and the appetite to upgrade the electric grid by utility companies heavily invested in gas isnt really there.
It would be nice if you can have your appliances charging per day, and using a built in battery suited to run at night.
But the solution isn't batteries, it stored hydroelectric. Only works where you have mountains though.
Batteries are a bit uneconomical now but may well fall in price like solar has. See for example:
>A startup run by a Tesla veteran and backed by Bill Gates is promising to build a long-duration battery that's 50 to 100 times cheaper than lithium-ion https://www.businessinsider.com/form-energy-battery-startup-...
Europe is already using something like 90% of our hydro electricity potential. So far, countries either: - added renewable energies on top of existing fossil fuel based power generation (gas, coal) - added renewable energies and replaced coal with gas (US for instance)
How is that a solution when it's difficult to build more of it?
1 reply →
then by your own logic it would happen other places, it doesent.
It is "played up" because that is how the physics/science works right now. Grid-scale storage is required to smooth out grid-scale intermittent generation.
Solar all day, NG all night would be a major improvement over our current system. Especially as batteries get better.
Depending on where you are, that's pretty close to what's happened. Right now (4pm), the California ISO is generating 73% from renewables+hydro, 7% nuclear, and the balance gas. Naturally, the night replaces solar with gas. This is largely how California has reduced electricity emissions to their current levels.
Always gets me that the above is a no brainer. But people married to the argument that we can't have renewables until we solve the storage problem just refuse to even consider it.
Disconcerting.
And that nuclear power is just from a single plant!
NG is still extraordinarily high carbon. "A little better than coal (if you ignore the pipeline and wellhead losses)" is not something we can settle for given the seriousness of climate change. We need 24/7 low-carbon power. Fracked natural gas for entire nights (recall in winter nights are quite long) is not an option.
> "Solar is clearly economic in many markets, but we'll never get grid emissions in a place like California down much more without storage."
even before reliable utility-scale storage, there is a lot of low hanging fruit from covering the southwest US in solar and wind. but yes, the costs of intermittent power like wind and solar doesn't by itself end with installation.
> The fact of the matter is that an intermittent kWh is not as valuable as an on-demand reliable kWh
Don't worry -- every single time renewables are mentioned on HN, this aspect is at the top of the posting.
Every HN post has something negative at the top.
And as long as it's true, may it long keep getting mentioned.
I wish more places had hourly pricing of power, with good integrations. Would be really cool to be able to set up your appliances/AC/electric car to consume more power when it is cheap. And then you could get battery banks for your house to store power when it is cheap and use it when it is expensive, absorbing the peaks in a decentralized way.
Battery prices are also dropping, although as not as fast as solar, but its at the point where in some regions solar + battery is cheaper than coal
Modern Gas turbines can also act as a good backup for solar in times where there is unusual demand because they can start up on under a minute
These days most contracts are being bid solar+battery at a ratio of about 5:1, if memory serves.
All they need to do is build mechanical/hydraulic systems for storing solar power during the day and unleashing it at night.
For example, use solar power during the day to pump fluid from a lower reservoir into a higher reservoir, and then harness the energy of the water flowing from the higher reservoir into the lower reservoir through turbines.
As long as your output energy is always coming from the turbines, and as long as your solar powered pumps running during the day can keep up with double the rate of drainage flow, then you should have a constant loop with a steady supply of power.
This type of system could be retrofitted onto virtually any dam, giving you a way to create a closed loop with constant power and without the water loss typical from a dam (other than evaporation).
For areas where water is scarce and a dam isn't feasible, there are also other ideas, such as gravitational potential energy systems that use solar powered energy to lift weights on pulleys, which then power a generator as the weights are lowered by gravity.
Other ideas: Thermal storage including molten salts which can efficiently store and release very large quantities of heat energy, compressed air energy storage, flywheels, cryogenic systems, etc...
You're wildly underestimating how complicated pumped hydro is.
For one thing, existing dams absolutely cannot be converted to pumped hydro. Dams do not store water below them. Water flows downstream because a dam is in a river. There is no water to pump uphill. Unless, of course, you also build a second dam very close downstream to create another lower reservoir. This is usually a bad idea, and better to just find better geography that will support a new pumped hydro dam.
Some dams, such as Lake Oroville in California, do use an afterbay to allow for off-peak water pumping back into the lake (https://norcalwater.org/efficient-water-management/efficient...).
Sadly, dams tend to destroy healthy fishery ecosystems —- a side effect of not “wasting” the water.
>...For example, use solar power during the day to pump fluid from a lower reservoir into a higher reservoir, and then harness the energy of the water flowing from the higher reservoir into the lower reservoir through turbines.
Trying to rely only on intermittent power sources has huge storage requirements due to weather along with daily/seasonal variation. If grid energy storage was a simple problem it would have been done decades ago.
For example, one estimate is that for Germany to rely on solar and wind would require about 6,000 pumped storage plants which is literally 183 times their current capacity:
>...Based on German hourly feed-in and consumption data for electric power, this paper studies the storage and buffering needs resulting from the volatility of wind and solar energy. It shows that joint buffers for wind and solar energy require less storage capacity than would be necessary to buffer wind or solar energy alone. The storage requirement of over 6,000 pumped storage plants, which is 183 times Germany’s current capacity, would nevertheless be huge.
https://www.econstor.eu/bitstream/10419/144985/1/cesifo1_wp5...
If you want to make renewable storage seem ridiculous, you just model storage as if every electron generated is precious and can't be wasted and you get an answer like the above.
Overprovisioning is so simple and widely accepted a concept that anyone ignoring it is likely trying to intentionally mislead.
> If grid energy storage was a simple problem it would have been done decades ago.
Except storage is much less useful in the old paradigm, so the motivation wasn't there. Going forward, prices will swing wildly, so storage will be more valuable.
1 reply →
Good ideas but not currently feasible at grid-scale and so can't be really considered today as viable components of a power grid.
Yup, stacked concrete: https://qz.com/1355672/stacking-concrete-blocks-is-a-surpris...
Which all costs a lot of money.
There's nothing more handwavy than electricity storage
Does cheaper prices open up some less efficient ways of storage?
I'm still hoping for this one, just because it's so epic: the giant rock-piston gravity storage: https://www.heindl-energy.com/
It's impressive how systemic predictability matters. When nothing happens at the right time, you basically die.
also, solar might not be solar.
"Utility scale solar" is ambiguous and can mean solar thermal or pv solar.
I think solar thermal might be a completely different animal that is not long-lived and not clean.
It might be a similar situation to where "renewable power" turns out to me mostly burning garbage or burning trees.
This documentary might be a little biased but also have some interesting insights: https://youtu.be/Zk11vI-7czE
Concentrated solar thermal is basically dead. PV just slaughtered it.