4 ms·

Shipping containers of Li-ion batteries in the power grid

30 points by davi 16y ago

bradleyland 16y ago
I was rather surprised that the energy transfer loss is only 10% in batteries, but 15% in flywheels. It's pretty amazing that the chemical process in a li-ion battery (developed in the last 50 years) is more efficient than our best engineering efforts in the simple wheel and axle design (over 5500 years old). On a side note, an engineer at our local power plant once told me that the inertial energy available on "the grid" was sufficient to cause one of the steam turbines at our local power plant to "jump out of the floor" should someone fully engage it out of phase.
- stcredzero 16y ago
  The US might benefit from some investment in Nickel-Iron batteries. These were invented by Edison in 1901. They have a basic service life of 20 years (as opposed to 3-5 for Li-ion) and can be stretched to 50 years with electrolyte replacement. The battery tech is worked out, but investments in manufacture and economies of scale would enable market access to a technology with lower cost than Li-ion. (Also, the materials used in NiFe batteries are cheap and available domestically. They are a bit challenged with regards to high currents, though.) Wind and solar plants “introduce some additional variability that you don’t have with traditional thermal units,’’ Mr. Zaharuncik said. With a system that has a lot of renewable energy generators, he explained, “you need some other kind of resource that complements it. ‘’ I'm surprised that flywheels aren't a bigger deal because of wind and solar power. Why doesn't someone make a flywheel unit the size of a major appliance, incorporating inverters and regulating hardware, for home wind and solar?
  - khafra 16y ago
    Perhaps safety issues[1]? [1]http://news.ycombinator.com/item?id=309413 http://news.ycombinator.com/item?id=309413
    - stcredzero 16y ago
      Flywheel safety isn't such a big deal if you can bury the flywheel chamber in your backyard. Decades ago, there were fiberglass flywheels whose fibers simply unraveled when they failed, becoming a mass of very hot fuzz.
  - yardie 16y ago
    Flywheel technology were an area of research over a decade ago when battery research was still in its infancy. IIRC, GM had an electric car concept that had a huge fly wheel behind the cabin. But your options were slow, heavy (thus inefficient) flywheels or light, fast flywheels. And by fast I mean 100,000 RPM. At those speeds they were using Kevlar to keep it from flying apart, making it impractical because of the price of kevlar.
  - AaronM 16y ago
    At the last company I worked for, which was a manufacturing facility. They installed a giant flywheel as a line conditioning device. (We manufactured NAS, and it is terribly costly when a power outage interrupts 5 days worth of testing forcing you to start over). I think it gave us about 17-18 seconds of power, in the event of any brief interruptions in power (Most of the outages we received were 1-2 seconds in length)
    - aristus 16y ago
      Some datacenters (eg the Miami IPX/NAP) use flywheels the size of large trucks to provide failover until the generators kick in. They are seriously scary machines up close.
  - Symmetry 16y ago
    If you're doing these on an industrial scale a liquid salt battery might be worth while.
    - jpitz 16y ago
      Are you referring to the sodium sulfur batteries that NGK makes?
      - Symmetry 16y ago
        Sodium Sulfur, Sodium Aluminumchloride, Potassium, whatever. This actually seems like an "almost there" technology which would blossom if there was a big demand for it.
  - coryrc 16y ago
    Nickel is not a cheap metal and NiFe batteries are not efficient[1]. Nickel is much more damaging to the environment to extract. LiFePO4 batteries have lifetimes exceeding 15 years (no one really knows how long yet -- they were invented in 1996). Their efficiency is well above 90%. Its raw material costs are less than NiFe. [1] http://en.wikipedia.org/wiki/Nickel-iron_battery http://en.wikipedia.org/wiki/Nickel-iron_battery
- lutorm 16y ago
  I don't think it's the wheel and axle design that limits the flywheel efficiency, it's more likely the electric<->mechanical energy conversion.
  - sliverstorm 16y ago
    What about the bleed-off? Flywheels loose energy due to friction, and batteries slowly self-discharge.
- zefhous 16y ago
  Indeed. I have done some research on flywheels and I believe that they can be much more efficient than batteries. Another huge benefit to flywheels is that they are consistent and do not suffer loss of capacity over their extremely long lifespan. Flywheels really should be used for this type of thing, not batteries. Regarding efficiency, the following is from a 2003 report: "Various losses in the system, including bearing losses, and electrical losses (stator losses, power converter losses), contribute to the overall conversion efficiency. The power electronics switching losses usually dominate the total losses, and the overall efficiency is a function of power, usually > 80% for power in the range 10% to 100% rated. The maximum efficiency quoted by some manufacturers is 96%. This gives rise to a typical maximum in-out efficiency of 92%." - http://www.itpower.co.uk/investire/pdfs/flywheelrep.pdf http://www.itpower.co.uk/investire/pdfs/flywheelrep.pdf
- jrockway 16y ago
  I was rather surprised that the energy transfer loss is only 10% in batteries, but 15% in flywheels. Unrelated, but this led to an interesting thought for me. When flywheels fail, they shatter into explosive shrapnel fragments, killing everyone nearby. When shipping containers full of LiIon batteries fail, they blow up and leave a crater the size of ... well, big, probably. Why such energetic failures? Because that's what they do, concentrate a lot of energy all in one space. Electricity and explosions are both energy. I know this is obvious but it made for an interesting few seconds of mental dialogue so I thought I'd share :)
cagenut 16y ago
AES did a pilot project of the same type with Altairnano batteries back in 08, you can see some of the numbers and details here (warning pdf): http://www.b2i.cc/Document/546/KEMA_Carina_validation_report_public_final.pdf http://www.b2i.cc/Document/546/KEMA_Carina_validation_report...
chopsueyar 16y ago
What about some sort of hydraulic "battery"? I have seen water pumps that can be used as generators and vice/versa, especially during offpeak periods, water is pumped uphill, and during peak demand, water is sent downhill to generate electricity. Would a hydraulic turbine/pump/generator be feasible?
- cryptorchidism 16y ago
  Well yes, and that's what the system you mentioned does. In addition to pumped hydro, there are schemes for pumping air into underground caverns. You need too much volume to make artificial tanks cost-effective. That's why natural formations are used.
  - chopsueyar 16y ago
    That pneumatic system sounds interesting. Do you have any links, por favor?
- yrral 16y ago
  I don't think so, since the article states that the system needs to vary its output in 5 second intervals, the generators that use water will take too long to start spinning up to keep up with a sudden strong current draw.
- asmithmd1 16y ago
  Pump power storage facilities have been online for decades. They are often near nuclear facilities which can not easily throttle up and down: http://www.nypa.gov/facilities/blengil.htm http://www.nypa.gov/facilities/blengil.htm I think your analogy to a battery is correct for these pump storage operations while a container of batteries is, by analogy, closer to a capacitor: able to deliver high pulses of current quickly but only for a very short time.

30 points by davi 16y ago

I was rather surprised that the energy transfer loss is only 10% in batteries, but 15% in flywheels. It's pretty amazing that the chemical process in a li-ion battery (developed in the last 50 years) is more efficient than our best engineering efforts in the simple wheel and axle design (over 5500 years old). On a side note, an engineer at our local power plant once told me that the inertial energy available on "the grid" was sufficient to cause one of the steam turbines at our local power plant to "jump out of the floor" should someone fully engage it out of phase.

stcredzero 16y ago
The US might benefit from some investment in Nickel-Iron batteries. These were invented by Edison in 1901. They have a basic service life of 20 years (as opposed to 3-5 for Li-ion) and can be stretched to 50 years with electrolyte replacement. The battery tech is worked out, but investments in manufacture and economies of scale would enable market access to a technology with lower cost than Li-ion. (Also, the materials used in NiFe batteries are cheap and available domestically. They are a bit challenged with regards to high currents, though.) Wind and solar plants “introduce some additional variability that you don’t have with traditional thermal units,’’ Mr. Zaharuncik said. With a system that has a lot of renewable energy generators, he explained, “you need some other kind of resource that complements it. ‘’ I'm surprised that flywheels aren't a bigger deal because of wind and solar power. Why doesn't someone make a flywheel unit the size of a major appliance, incorporating inverters and regulating hardware, for home wind and solar?
- khafra 16y ago
  Perhaps safety issues[1]? [1]http://news.ycombinator.com/item?id=309413 http://news.ycombinator.com/item?id=309413
  - stcredzero 16y ago
    Flywheel safety isn't such a big deal if you can bury the flywheel chamber in your backyard. Decades ago, there were fiberglass flywheels whose fibers simply unraveled when they failed, becoming a mass of very hot fuzz.
- yardie 16y ago
  Flywheel technology were an area of research over a decade ago when battery research was still in its infancy. IIRC, GM had an electric car concept that had a huge fly wheel behind the cabin. But your options were slow, heavy (thus inefficient) flywheels or light, fast flywheels. And by fast I mean 100,000 RPM. At those speeds they were using Kevlar to keep it from flying apart, making it impractical because of the price of kevlar.
- AaronM 16y ago
  At the last company I worked for, which was a manufacturing facility. They installed a giant flywheel as a line conditioning device. (We manufactured NAS, and it is terribly costly when a power outage interrupts 5 days worth of testing forcing you to start over). I think it gave us about 17-18 seconds of power, in the event of any brief interruptions in power (Most of the outages we received were 1-2 seconds in length)
  - aristus 16y ago
    Some datacenters (eg the Miami IPX/NAP) use flywheels the size of large trucks to provide failover until the generators kick in. They are seriously scary machines up close.
- Symmetry 16y ago
  If you're doing these on an industrial scale a liquid salt battery might be worth while.
  - jpitz 16y ago
    Are you referring to the sodium sulfur batteries that NGK makes?
    - Symmetry 16y ago
      Sodium Sulfur, Sodium Aluminumchloride, Potassium, whatever. This actually seems like an "almost there" technology which would blossom if there was a big demand for it.
- coryrc 16y ago
  Nickel is not a cheap metal and NiFe batteries are not efficient[1]. Nickel is much more damaging to the environment to extract. LiFePO4 batteries have lifetimes exceeding 15 years (no one really knows how long yet -- they were invented in 1996). Their efficiency is well above 90%. Its raw material costs are less than NiFe. [1] http://en.wikipedia.org/wiki/Nickel-iron_battery http://en.wikipedia.org/wiki/Nickel-iron_battery
lutorm 16y ago
I don't think it's the wheel and axle design that limits the flywheel efficiency, it's more likely the electric<->mechanical energy conversion.
- sliverstorm 16y ago
  What about the bleed-off? Flywheels loose energy due to friction, and batteries slowly self-discharge.
zefhous 16y ago
Indeed. I have done some research on flywheels and I believe that they can be much more efficient than batteries. Another huge benefit to flywheels is that they are consistent and do not suffer loss of capacity over their extremely long lifespan. Flywheels really should be used for this type of thing, not batteries. Regarding efficiency, the following is from a 2003 report: "Various losses in the system, including bearing losses, and electrical losses (stator losses, power converter losses), contribute to the overall conversion efficiency. The power electronics switching losses usually dominate the total losses, and the overall efficiency is a function of power, usually > 80% for power in the range 10% to 100% rated. The maximum efficiency quoted by some manufacturers is 96%. This gives rise to a typical maximum in-out efficiency of 92%." - http://www.itpower.co.uk/investire/pdfs/flywheelrep.pdf http://www.itpower.co.uk/investire/pdfs/flywheelrep.pdf
jrockway 16y ago
I was rather surprised that the energy transfer loss is only 10% in batteries, but 15% in flywheels. Unrelated, but this led to an interesting thought for me. When flywheels fail, they shatter into explosive shrapnel fragments, killing everyone nearby. When shipping containers full of LiIon batteries fail, they blow up and leave a crater the size of ... well, big, probably. Why such energetic failures? Because that's what they do, concentrate a lot of energy all in one space. Electricity and explosions are both energy. I know this is obvious but it made for an interesting few seconds of mental dialogue so I thought I'd share :)

cagenut 16y ago

AES did a pilot project of the same type with Altairnano batteries back in 08, you can see some of the numbers and details here (warning pdf): http://www.b2i.cc/Document/546/KEMA_Carina_validation_report_public_final.pdf http://www.b2i.cc/Document/546/KEMA_Carina_validation_report...

chopsueyar 16y ago

What about some sort of hydraulic "battery"? I have seen water pumps that can be used as generators and vice/versa, especially during offpeak periods, water is pumped uphill, and during peak demand, water is sent downhill to generate electricity. Would a hydraulic turbine/pump/generator be feasible?

cryptorchidism 16y ago
Well yes, and that's what the system you mentioned does. In addition to pumped hydro, there are schemes for pumping air into underground caverns. You need too much volume to make artificial tanks cost-effective. That's why natural formations are used.
- chopsueyar 16y ago
  That pneumatic system sounds interesting. Do you have any links, por favor?
yrral 16y ago
I don't think so, since the article states that the system needs to vary its output in 5 second intervals, the generators that use water will take too long to start spinning up to keep up with a sudden strong current draw.
asmithmd1 16y ago
Pump power storage facilities have been online for decades. They are often near nuclear facilities which can not easily throttle up and down: http://www.nypa.gov/facilities/blengil.htm http://www.nypa.gov/facilities/blengil.htm I think your analogy to a battery is correct for these pump storage operations while a container of batteries is, by analogy, closer to a capacitor: able to deliver high pulses of current quickly but only for a very short time.