Something this article doesn't address is the potential impact batteries everywhere could have. This quote in particular tells me the author, and the speaker, don't see it this way:
“The battery-in-every-home idea—not only do I think it doesn’t make economic sense, I don’t think it’s necessary,” said Brian Warshay, an analyst at Bloomberg New Energy Finance. “Having a centralized grid is incredibly useful and incredibly efficient.”
The difficulty with our current power infrastructure is that there is no buffering at the consumption side. Because there is no buffering at the consumption side, the overall grid must be able to handle peak demand. That is, we must over-provision. With batteries everywhere, this isn't necessarily the case. In theory, we could all have generators, but we don't because they're noisy, smelly, and a maintenance hassle. But sticking a giant battery on the wall? I can see everyone having one, and big institutions having large arrays of them. I think that has the potential to fundamentally change our power grid.
Do you have any idea how many Li-Ion batteries would be needed to match the 30 GIGAWATT-hours of storage potential in the Bath County pumped storage plant?
The problem is that California didn't build enough of these 20 years ago. California has a couple of pumped-storage plants coming in... and as soon as they're built then net-metering will be even better for consumers.
In general, centralized storage at the utility-level will benefit from scale. Redox batteries will probably be cheaper and more efficient than Li-Ion batteries that Tesla is putting out.
And Pumped-Hydro is already much much cheaper, and has existed since the 60s.
I ran some back-of-the-envelope calculations based on Tom "Do the Math" Murphy's nation-sized battery, 336 TWh, sufficient for two weeks' reserve power.
Tesla's Superfactory could produce that. In about 12,000 years.
Even if the capacity's excessive by a few orders of magnitude, that's a hell of a lot of battery.
Assuming 56 kWh Tesla batteries. Half a million battery packs is 28 GWh of total storage capacity. That's the energy equivalent of 16,000 barrels of oil. Assuming a modest 30% Carnot thermal generation efficiency, you'd need 55,000 bbl of oil to actually generate that electricity.
Yeah, batteries are a hard sell.
If each battery is 2m x 3m or 6m in area, the Nation Sized Tesla Battery would cover 36,000 km^2, or an area 190 km on a side. Maybe we'll stack them.
(The actual measurements are smaller, around 1.5 x 2.5 m or so, so that's an overstatement.)
A buddy of mine working at ORNL claims there's interesting stuff on the storage front that's competitive with pumped storage, but can't say just what yet. I'm highly curious myself.
A buddy of mine working at ORNL claims there's interesting stuff on the storage front that's competitive with pumped storage, but can't say just what yet. I'm highly curious myself.
"Table 4" is probably what you want to look at. As you can see, only Compressed Air is competitive with Pumped Hydro as far as energy storage (w-hrs).
It looks like Compressed Air is going to be the golden standard a few years from now. There was a relatively cheap compressed-air plant created out of abandoned mine tunnels that was hundreds-of-Megawatts in scale. I don't quite remember the name of it however.
Also, Compressed Air doesn't require you to empty a lake (and refill it) every day. So California's low-water situation may prefer compressed air storage systems.
Redox Flow batteries look to be the most promising technology at utility-scale chemical. Chemical solutions are needed for their versatility... but "long-term" energy storage (such as day-night load balancing) will likely be handled by Compressed Air and/or Pumped Hydro.
Currently, over 95% of the US's energy storage capacity is in pumped hydro.
Thanks for that, could be, and you're likely right, though my understanding was that the work he's discussing hasn't been published. Hard to see it _not_ being CAES though.
Another avenue I've looked at is ocean-based pumped-hydro. Same as the usual suspects, but the ocean is the lower (and hence inexhaustible) basin. That means creating an elevated high-fluctuation saltwater estuary, and you've got corrosion issues associated with salt water, plus marine life fouling. Japan's built a test facility. Tidal barrier systems can also be considered as such (with or without active pumping).
Advances are being made at _all_ energy storage solutions. Frankly, CAES is the most promising, now that they've figured out how to store excess heat during compression.
Centralized storage doesn't offer some of the benefits in-home batteries do, though, like the ability to store up power from home solar, or the ability to keep your furnace running in a power outage.
like the ability to store up power from home solar
Why would you want to do this when net-metering exists? Let the grid store your excess power, and buy it back from them. Run the math, you'll see which one is cheaper.
or the ability to keep your furnace running in a power outage.
A $3000 gas generator will get you 10KW of power. In contrast, you need $10,000 of Tesla's "cheap" batteries to get to 10KW (not counting installation costs... or the inverter... or any additional stuff)
In any case, a "total" solution for a 10kw gas backup power can be had for less than $7,000, including installation costs. While the 10kw battery solution is closer to $15,000+.
A Gas Generator can be wired up to either a propane tank, offering perpetual energy. (Gas lines rarely rupture). If you don't want to be hooked up to gas lines, then propane tanks are also an option.
They kill people on a regular basis (due to misuse, but misuse happens). I don't need 10KW, either, I just need enough to power the furnace so I don't freeze to death.
Why would you want to do this when net-metering exists? Let the grid store your excess power, and buy it back from them. Run the math, you'll see which one is cheaper.
The math will change if everyone does it, I'd imagine.
Ignoring the fact that furnaces are typically 10kw or more... then lets say you "only" need 3kw of power (so you can run off of a single Tesla battery). (Maybe you're running a smaller room-heater or something)
A natural gas furnace, which needs a small amount of power to start and continue running.
Then why not get a smaller $500 generator?
As a non-mechanical person, I'd rather not have to maintain one. The lawnmower is enough for me. I also like the idea of being able to have some additional powered items plugged in in an extended outage.
A natural gas furnace, which needs a small amount of power to start and continue running.
Hmm, at which point I wonder if your standard UPS actually does the job. When I think "PowerWall", I'm thinking of things that power the entire house.
But UPSes are useful for the small-scale "important" appliances. Like the sump pump or I guess in this case... a gas furnace.
APC UPSes are only about $500. My parents had flooding issues with their Sump Pump as power went out a few years ago, and they just installed some UPS on it. Haven't had a problem since then.
Why would you want to do this when net-metering exists? Let the grid store your excess power, and buy it back from them. Run the math, you'll see which one is cheaper.
That's great in places where the grid is required to take your excess power at rates similar to what you will end up paying at the times you may want to buy it back.
How many places does that apply to? (EDIT: I'm genuinely curious; it's not a rhetorical question, just to be clear)
That's great in places where the grid is required to take your excess power at rates similar to what you will end up paying at the times you may want to buy it back.
On the contrary. Any form of net-metering makes batteries a hopeless situation.
Lets say you use net-metering, and it returns power at less than HALF of "off peak" levels. Lets say you're paying absurd California prices, of 40-cents kw-h at peek, and 10-cents at non-peek. And then you only manage to sell power back at 4-cents.
IIRC, these numbers are somewhat close to California numbers. Mind you, Virginia has 9-cents peek and 1-cent off peek. So... California prices are about as good as it gets for batteries.
Well, lets see here... first off all, you don't pay peek anymore. Because your solar panels are active between 11am and 4pm. Lets say you're selling it back at 4-cents. So you're paying in essence... 6-cents for nighttime power through the net-metering system (for every kw-h stored in excess by your solar panels)
You need to use 58333 kw-h before you break even with Tesla's $3500 battery (not counting inverter costs or installation costs). Considering that the battery has a capacity of 7kw-h, that is 8333 cycles or 22-years.
Unfortunately, the Tesla Battery Pack only has a 10-year lifespan. So you've lost money in the long run.
Not counting inefficiencies in the battery, inefficiencies in the inverter or installation costs.
Using the 7 KW-h power wall (the daily cycle one), about 4.3 million. Tesla vehicles use 80-90 cells per kilowatt hour of capacity, so about 365 million lithium ion batteries would get you to 30 gigawatts (also, the claims about the Tesla Gigafactory have it producing a bit more than that per year).
Congrats, that's $1.277 Trillion in batteries to match the $5 billion Bath County Pumped Storage station.
Its not about whether or not someone can produce the batteries. Its about whether or not its cheaper to build centralized storage. Frankly, Pumped Hydro is extremely cheap in contrast to batteries.
Do you want to spend $5 Billion on a pumped hydro station? Or do you want to spend $1,200+ Billion mining a volatile metal out of the ground and building chemical batteries?
I think it is clear which solution is better for America in the long run.
The 365 million is 18650 cells, they retail for ~$8-10.
So hopefully less than $3.6 billion in cells (factory direct, new factory aimed at reducing costs). There is a bunch more of stuff in the integrated units, at the quoted $3000 price, 4.3 million of them would be ~$14 billion.
I think pumped storage is great (I've posted enthusiastically about it here in the past), but the sneaky factor is that they did a good job siting the ones they built decades ago, many of the great spots to build them are in use already.
(I edited in the $14 billion after I realized I was talking about a product with a proposed price, I had originally said something about less than $10 billion)
Technically, the Bath County station was built with $1.6 Billion. I adjusted up to $5 Billion for an off-the cuff "inflation" estimation, also because other pumped hydro solutions are coming around $5 Billion as well.
Something to note is that the Bath County station has been running since the 1970s. That's a useful lifespan well of 40 years at this point. I don't think Lithium Ion (especially when they are cycled daily) will achieve that lifespan.
It's hard to do a facile comparison. The fancy batteries from Tesla obviously lose right now against places with an existing grid and convenient site for a storage plant, but Bath County serves something like 60 million people, which is a lot of grid, with a lot of ancillary equipment in between the station and those users.
And there are reasons to expect the benefits of new pumped storage to decline over time (the best, most cost effective sites get built first, that sort of construction is well developed technology), and conversely, some reasons to expect the costs of batteries to decline over time (chemistry is still inching forward, as costs come down the number of uses goes up).
Ultimately, I can see that I'm not running out to buy solar panels and house batteries, so my real opinion on what is best right now is clear enough.
The centralized grid is going to stay regardless of how many homeowners buy powerwalls; only so much can be buffered on the consumption side. Because another thing this article doesn't address is the fact residential power use occupies a mere 18% [1] of power consumption. Commercial is only 12%.[1]
Maybe you could store enough energy to run your datacenter overnight. I'm speculating, but powering a chip fab or even a machine shop on batteries seems infeasible.
But, I agree that there is a major imbalance and generation/storage on the consumption side would help correct that. Reduced need for peak generation --> greater net efficiency.
Batteries are also a form of over-provisioning, just a distributed one.
The real market potential for batteries is not in those parts of the West where there is cheap, reliable mains. It's those places where the grid is unreliable enough that people with money would buy generators, or places where the grid is a tiny island (Hawaii, Orkney).
"But sticking a giant battery on the wall? I can see everyone having one..."
High-rise office and apartment buildings probably don't have space to store all those batteries. If you have a suburban home with a garage you can hang a battery on the garage wall, but if you live in a 20-story apartment building, where would you put it? I doubt that fire codes would allow putting huge lithium batteries inside every apartment. A large fraction of the U.S. population lives in this kind of densely-packed urban environment.
Comments
Something this article doesn't address is the potential impact batteries everywhere could have. This quote in particular tells me the author, and the speaker, don't see it this way:
“The battery-in-every-home idea—not only do I think it doesn’t make economic sense, I don’t think it’s necessary,” said Brian Warshay, an analyst at Bloomberg New Energy Finance. “Having a centralized grid is incredibly useful and incredibly efficient.”
The difficulty with our current power infrastructure is that there is no buffering at the consumption side. Because there is no buffering at the consumption side, the overall grid must be able to handle peak demand. That is, we must over-provision. With batteries everywhere, this isn't necessarily the case. In theory, we could all have generators, but we don't because they're noisy, smelly, and a maintenance hassle. But sticking a giant battery on the wall? I can see everyone having one, and big institutions having large arrays of them. I think that has the potential to fundamentally change our power grid.
Erm... or centralized storage mechanisms could be built.
https://en.wikipedia.org/wiki/Bath_County_Pumped_Storage_Sta...
Do you have any idea how many Li-Ion batteries would be needed to match the 30 GIGAWATT-hours of storage potential in the Bath County pumped storage plant?
The problem is that California didn't build enough of these 20 years ago. California has a couple of pumped-storage plants coming in... and as soon as they're built then net-metering will be even better for consumers.
In general, centralized storage at the utility-level will benefit from scale. Redox batteries will probably be cheaper and more efficient than Li-Ion batteries that Tesla is putting out.
And Pumped-Hydro is already much much cheaper, and has existed since the 60s.
I ran some back-of-the-envelope calculations based on Tom "Do the Math" Murphy's nation-sized battery, 336 TWh, sufficient for two weeks' reserve power.
Tesla's Superfactory could produce that. In about 12,000 years.
Even if the capacity's excessive by a few orders of magnitude, that's a hell of a lot of battery.
http://physics.ucsd.edu/do-the-math/2011/08/nation-sized-bat...
Assuming 56 kWh Tesla batteries. Half a million battery packs is 28 GWh of total storage capacity. That's the energy equivalent of 16,000 barrels of oil. Assuming a modest 30% Carnot thermal generation efficiency, you'd need 55,000 bbl of oil to actually generate that electricity.
Yeah, batteries are a hard sell.
If each battery is 2m x 3m or 6m in area, the Nation Sized Tesla Battery would cover 36,000 km^2, or an area 190 km on a side. Maybe we'll stack them.
(The actual measurements are smaller, around 1.5 x 2.5 m or so, so that's an overstatement.)
A buddy of mine working at ORNL claims there's interesting stuff on the storage front that's competitive with pumped storage, but can't say just what yet. I'm highly curious myself.
http://prod.sandia.gov/techlib/access-control.cgi/2011/11273...
"Table 4" is probably what you want to look at. As you can see, only Compressed Air is competitive with Pumped Hydro as far as energy storage (w-hrs).
It looks like Compressed Air is going to be the golden standard a few years from now. There was a relatively cheap compressed-air plant created out of abandoned mine tunnels that was hundreds-of-Megawatts in scale. I don't quite remember the name of it however.
Also, Compressed Air doesn't require you to empty a lake (and refill it) every day. So California's low-water situation may prefer compressed air storage systems.
Redox Flow batteries look to be the most promising technology at utility-scale chemical. Chemical solutions are needed for their versatility... but "long-term" energy storage (such as day-night load balancing) will likely be handled by Compressed Air and/or Pumped Hydro.
Currently, over 95% of the US's energy storage capacity is in pumped hydro.
And now with more bandwidth...
Thanks for that, could be, and you're likely right, though my understanding was that the work he's discussing hasn't been published. Hard to see it _not_ being CAES though.
Another avenue I've looked at is ocean-based pumped-hydro. Same as the usual suspects, but the ocean is the lower (and hence inexhaustible) basin. That means creating an elevated high-fluctuation saltwater estuary, and you've got corrosion issues associated with salt water, plus marine life fouling. Japan's built a test facility. Tidal barrier systems can also be considered as such (with or without active pumping).
https://www.reddit.com/r/RenewableTech/comments/29t6t7/seawa...
http://www.hitachi.com/rev/1998/revoct98/r4_108.pdf
http://www.inference.phy.cam.ac.uk/sustainable/book/tex/Lago...
http://www.sciencedirect.com/science/article/pii/S1876610214...
Flow, liquid metal, and molten salt batteries also have my eye.
Doesn't CAES require gas or other fuel for reheat on expansion?
Early designs did. But CAES has begun to store the excess heat that is made during compression.
For example: http://sigmaenergystorage.com/?page_id=447&lang=en
Advances are being made at _all_ energy storage solutions. Frankly, CAES is the most promising, now that they've figured out how to store excess heat during compression.
Centralized storage doesn't offer some of the benefits in-home batteries do, though, like the ability to store up power from home solar, or the ability to keep your furnace running in a power outage.
In any case, a "total" solution for a 10kw gas backup power can be had for less than $7,000, including installation costs. While the 10kw battery solution is closer to $15,000+.
A Gas Generator can be wired up to either a propane tank, offering perpetual energy. (Gas lines rarely rupture). If you don't want to be hooked up to gas lines, then propane tanks are also an option.
They kill people on a regular basis (due to misuse, but misuse happens). I don't need 10KW, either, I just need enough to power the furnace so I don't freeze to death.
The math will change if everyone does it, I'd imagine.
And tons of people have died as they fallen off of roofs as they installed solar panels.
Erm... what kind of furnace is in your house?
http://www.shop.thefurnaceoutlet.com/Electric-Furnaces_c15.h...
Ignoring the fact that furnaces are typically 10kw or more... then lets say you "only" need 3kw of power (so you can run off of a single Tesla battery). (Maybe you're running a smaller room-heater or something)
Then why not get a smaller $500 generator?
http://www.amazon.com/DuroStar-DS4000S-4-Cycle-Portable-Gene...
Or I dunno... a fireplace at that point. Since we're talking about small scale if you're using something smaller than a 10kw furnace...
A natural gas furnace, which needs a small amount of power to start and continue running.
As a non-mechanical person, I'd rather not have to maintain one. The lawnmower is enough for me. I also like the idea of being able to have some additional powered items plugged in in an extended outage.
Hmm, at which point I wonder if your standard UPS actually does the job. When I think "PowerWall", I'm thinking of things that power the entire house.
But UPSes are useful for the small-scale "important" appliances. Like the sump pump or I guess in this case... a gas furnace.
APC UPSes are only about $500. My parents had flooding issues with their Sump Pump as power went out a few years ago, and they just installed some UPS on it. Haven't had a problem since then.
http://www.tripplite.com/shared/literature/white-paper/how-t...
I'll have to go to their basement to figure out which brand they bought. But the above whitepaper gives you an idea of how it works with a sump pump.
Sounds a hell-of-a-lot cheaper than the Tesla Powerwall still.
EDIT: Seems like some guy in the UK has done the calcs already. http://kissurvival.com/running-your-gas-furnace-from-a-gener...
That's great in places where the grid is required to take your excess power at rates similar to what you will end up paying at the times you may want to buy it back.
How many places does that apply to? (EDIT: I'm genuinely curious; it's not a rhetorical question, just to be clear)
On the contrary. Any form of net-metering makes batteries a hopeless situation.
Lets say you use net-metering, and it returns power at less than HALF of "off peak" levels. Lets say you're paying absurd California prices, of 40-cents kw-h at peek, and 10-cents at non-peek. And then you only manage to sell power back at 4-cents.
IIRC, these numbers are somewhat close to California numbers. Mind you, Virginia has 9-cents peek and 1-cent off peek. So... California prices are about as good as it gets for batteries.
Well, lets see here... first off all, you don't pay peek anymore. Because your solar panels are active between 11am and 4pm. Lets say you're selling it back at 4-cents. So you're paying in essence... 6-cents for nighttime power through the net-metering system (for every kw-h stored in excess by your solar panels)
You need to use 58333 kw-h before you break even with Tesla's $3500 battery (not counting inverter costs or installation costs). Considering that the battery has a capacity of 7kw-h, that is 8333 cycles or 22-years.
Unfortunately, the Tesla Battery Pack only has a 10-year lifespan. So you've lost money in the long run.
Not counting inefficiencies in the battery, inefficiencies in the inverter or installation costs.
Using the 7 KW-h power wall (the daily cycle one), about 4.3 million. Tesla vehicles use 80-90 cells per kilowatt hour of capacity, so about 365 million lithium ion batteries would get you to 30 gigawatts (also, the claims about the Tesla Gigafactory have it producing a bit more than that per year).
Congrats, that's $1.277 Trillion in batteries to match the $5 billion Bath County Pumped Storage station.
Its not about whether or not someone can produce the batteries. Its about whether or not its cheaper to build centralized storage. Frankly, Pumped Hydro is extremely cheap in contrast to batteries.
Do you want to spend $5 Billion on a pumped hydro station? Or do you want to spend $1,200+ Billion mining a volatile metal out of the ground and building chemical batteries?
I think it is clear which solution is better for America in the long run.
The 365 million is 18650 cells, they retail for ~$8-10.
So hopefully less than $3.6 billion in cells (factory direct, new factory aimed at reducing costs). There is a bunch more of stuff in the integrated units, at the quoted $3000 price, 4.3 million of them would be ~$14 billion.
I think pumped storage is great (I've posted enthusiastically about it here in the past), but the sneaky factor is that they did a good job siting the ones they built decades ago, many of the great spots to build them are in use already.
(I edited in the $14 billion after I realized I was talking about a product with a proposed price, I had originally said something about less than $10 billion)
Ah, that at least makes them reasonable.
Technically, the Bath County station was built with $1.6 Billion. I adjusted up to $5 Billion for an off-the cuff "inflation" estimation, also because other pumped hydro solutions are coming around $5 Billion as well.
Something to note is that the Bath County station has been running since the 1970s. That's a useful lifespan well of 40 years at this point. I don't think Lithium Ion (especially when they are cycled daily) will achieve that lifespan.
It's hard to do a facile comparison. The fancy batteries from Tesla obviously lose right now against places with an existing grid and convenient site for a storage plant, but Bath County serves something like 60 million people, which is a lot of grid, with a lot of ancillary equipment in between the station and those users.
And there are reasons to expect the benefits of new pumped storage to decline over time (the best, most cost effective sites get built first, that sort of construction is well developed technology), and conversely, some reasons to expect the costs of batteries to decline over time (chemistry is still inching forward, as costs come down the number of uses goes up).
Ultimately, I can see that I'm not running out to buy solar panels and house batteries, so my real opinion on what is best right now is clear enough.
The centralized grid is going to stay regardless of how many homeowners buy powerwalls; only so much can be buffered on the consumption side. Because another thing this article doesn't address is the fact residential power use occupies a mere 18% [1] of power consumption. Commercial is only 12%.[1]
Maybe you could store enough energy to run your datacenter overnight. I'm speculating, but powering a chip fab or even a machine shop on batteries seems infeasible.
But, I agree that there is a major imbalance and generation/storage on the consumption side would help correct that. Reduced need for peak generation --> greater net efficiency.
[1]: http://www.eia.gov/tools/faqs/faq.cfm?id=447&t=1
Not really. Nowadays here in NY the delivery portion of my electric bill exceeds the supply cost.
Batteries are also a form of over-provisioning, just a distributed one.
The real market potential for batteries is not in those parts of the West where there is cheap, reliable mains. It's those places where the grid is unreliable enough that people with money would buy generators, or places where the grid is a tiny island (Hawaii, Orkney).
"But sticking a giant battery on the wall? I can see everyone having one..."
High-rise office and apartment buildings probably don't have space to store all those batteries. If you have a suburban home with a garage you can hang a battery on the garage wall, but if you live in a 20-story apartment building, where would you put it? I doubt that fire codes would allow putting huge lithium batteries inside every apartment. A large fraction of the U.S. population lives in this kind of densely-packed urban environment.