5 ms·

Iron-Chromium Flow Battery Aims to Replace Gas Plants

17 points by pwg 12y ago

reitanqild 12y ago
Interesting, especially for its benefits for solar power. (One sentence sticks out though: ". The company projects it can deliver energy for utilities and other users at a cost of less than $250 per kilowatt hour." This must be a mistake?)
- danbruc 12y ago
  This are (probably) the installation costs.
  - stcredzero 12y ago
    Amortized over 10 or 20 years? That sounds comfortably low to me.
- bencoder 12y ago
  There's discussion of that in the comments [1]. It looks like it's referring to the storage capacity. [1] http://spectrum.ieee.org/energywise/energy/the-smarter-grid/new-flow-battery-aims-to-replace-gas-plants#comment-1400912704 http://spectrum.ieee.org/energywise/energy/the-smarter-grid/...
- ndonnellan 12y ago
  Storage energy, not cost per unit of production. For example, to store 1 MWh, it would cost $250,000 to build the storage. That cost would be amortized as suggested below. Since solar and wind are still very small parts of the energy production industry, it's not clear what utilities will pay for pure storage (at least, not the last time I checked). Places like Hawaii, where the PV concentration is very high, might be interested.
  - mikeash 12y ago
    Even absent solar and wind, aren't there some fairly substantial costs associated with fluctuating demand? I'm not particularly knowledgeable about this stuff, but I seem to recall that even among traditional methods of generating power, the cheaper forms tend to be less able to respond to quick changes in demand. Beyond that, there's also the fundamental conflict between average and peak demand, meaning that a lot of generating capacity (and thus capital expenditures) sits idle a lot of the time, and just exists to be able to handle those few days when the temperature hits 105F and everyone runs their air conditioners at full power all day long. I have no idea how those costs stack up against $250/kWh for storage systems, but it's at least conceivable that this stuff could be worthwhile even absent any significant solar/wind deployment.
XorNot 12y ago
Be interesting to see how they're solving the cross-contamination problem. Iron-Chromium batteries fell out of favor because once the electrolytes mix (and they always do) the cell is done. The vanadium-redox battery was an advance because with the same electrolyte on both sides you don't have that problem (but vanadium is expensive in this application). Last I heard there'd been some significant progress on house-scale iron-air flow cell type systems, which is the one I'm really excited about - no cross-contamination to worry about their either.
ginko 12y ago
How does this compare to pumped-storage hydroelectric plants? Those usually have a conversion efficiency of 70-85% and can deliver serious capacity/power. I guess good places for solar and mountains aren't always close together.
- twic 12y ago
  Presumably you can also implement a pumped storage system with a massive hole in the ground, using that and the surface as the two levels. A use for old mines, maybe? Or maybe even for those persistent mine fires that were posted today. Two birds with one stone. Prior art, of a sort: http://en.wikipedia.org/wiki/Lake_Peigneur http://en.wikipedia.org/wiki/Lake_Peigneur
  - maxerickson 12y ago
    The pumped storage plant in Ludington, Michigan doesn't have a whole lot of head, 363 feet. It's basically just an artificial reservoir at the top of a hill by Lake Michigan. http://en.wikipedia.org/wiki/Ludington_Pumped_Storage_Power_Plant http://en.wikipedia.org/wiki/Ludington_Pumped_Storage_Power_...
- afarrell 12y ago
  the problem with a pumped-storage plant is that it is not nearly as responsive to changes in demand as a natural gas turbine. context: On the bulk electricity market, each topological region has a set amount of megawatt demand and suppliers bid in supply curves. A nuclear plant and a hydro plant (including pumped-storage) thats already running will bid negative in order to ensure that they get scheduled and don't have to shut down. A coal plant will bid slightly positive, and a natural gas plant will bid high. The linear programming problem that schedules these plants will start with the low-cost megawatts and work its way up in price until it fills its demand. The price of the last megawatt to get scheduled is applied to the whole market so the nuclear plant gets paid the price of a natural gas plant. Natural gas plants also provide another product called "regulation". That is, providing power at a certain level but being able to ramp up or down quickly in real time to keep the whole system operating close enough to 60Hz. It looks like these batteries are going to be at least comparable to natural gas turbines, so they'll be able to be take over some of the role of the marginal watt and provide some regulation.

17 points by pwg 12y ago

Interesting, especially for its benefits for solar power. (One sentence sticks out though: ". The company projects it can deliver energy for utilities and other users at a cost of less than $250 per kilowatt hour." This must be a mistake?)

danbruc 12y ago
This are (probably) the installation costs.
- stcredzero 12y ago
  Amortized over 10 or 20 years? That sounds comfortably low to me.
bencoder 12y ago
There's discussion of that in the comments [1]. It looks like it's referring to the storage capacity. [1] http://spectrum.ieee.org/energywise/energy/the-smarter-grid/new-flow-battery-aims-to-replace-gas-plants#comment-1400912704 http://spectrum.ieee.org/energywise/energy/the-smarter-grid/...
ndonnellan 12y ago
Storage energy, not cost per unit of production. For example, to store 1 MWh, it would cost $250,000 to build the storage. That cost would be amortized as suggested below. Since solar and wind are still very small parts of the energy production industry, it's not clear what utilities will pay for pure storage (at least, not the last time I checked). Places like Hawaii, where the PV concentration is very high, might be interested.
- mikeash 12y ago
  Even absent solar and wind, aren't there some fairly substantial costs associated with fluctuating demand? I'm not particularly knowledgeable about this stuff, but I seem to recall that even among traditional methods of generating power, the cheaper forms tend to be less able to respond to quick changes in demand. Beyond that, there's also the fundamental conflict between average and peak demand, meaning that a lot of generating capacity (and thus capital expenditures) sits idle a lot of the time, and just exists to be able to handle those few days when the temperature hits 105F and everyone runs their air conditioners at full power all day long. I have no idea how those costs stack up against $250/kWh for storage systems, but it's at least conceivable that this stuff could be worthwhile even absent any significant solar/wind deployment.

XorNot 12y ago

Be interesting to see how they're solving the cross-contamination problem. Iron-Chromium batteries fell out of favor because once the electrolytes mix (and they always do) the cell is done. The vanadium-redox battery was an advance because with the same electrolyte on both sides you don't have that problem (but vanadium is expensive in this application). Last I heard there'd been some significant progress on house-scale iron-air flow cell type systems, which is the one I'm really excited about - no cross-contamination to worry about their either.

ginko 12y ago

How does this compare to pumped-storage hydroelectric plants? Those usually have a conversion efficiency of 70-85% and can deliver serious capacity/power. I guess good places for solar and mountains aren't always close together.

twic 12y ago
Presumably you can also implement a pumped storage system with a massive hole in the ground, using that and the surface as the two levels. A use for old mines, maybe? Or maybe even for those persistent mine fires that were posted today. Two birds with one stone. Prior art, of a sort: http://en.wikipedia.org/wiki/Lake_Peigneur http://en.wikipedia.org/wiki/Lake_Peigneur
- maxerickson 12y ago
  The pumped storage plant in Ludington, Michigan doesn't have a whole lot of head, 363 feet. It's basically just an artificial reservoir at the top of a hill by Lake Michigan. http://en.wikipedia.org/wiki/Ludington_Pumped_Storage_Power_Plant http://en.wikipedia.org/wiki/Ludington_Pumped_Storage_Power_...
afarrell 12y ago
the problem with a pumped-storage plant is that it is not nearly as responsive to changes in demand as a natural gas turbine. context: On the bulk electricity market, each topological region has a set amount of megawatt demand and suppliers bid in supply curves. A nuclear plant and a hydro plant (including pumped-storage) thats already running will bid negative in order to ensure that they get scheduled and don't have to shut down. A coal plant will bid slightly positive, and a natural gas plant will bid high. The linear programming problem that schedules these plants will start with the low-cost megawatts and work its way up in price until it fills its demand. The price of the last megawatt to get scheduled is applied to the whole market so the nuclear plant gets paid the price of a natural gas plant. Natural gas plants also provide another product called "regulation". That is, providing power at a certain level but being able to ramp up or down quickly in real time to keep the whole system operating close enough to 60Hz. It looks like these batteries are going to be at least comparable to natural gas turbines, so they'll be able to be take over some of the role of the marginal watt and provide some regulation.