5 ms·

Isn’t 4 hours of battery storage much too low? How is that considered “reliable”? What am I missing?

cptcobalt 5y ago
Indeed, if we're assuming that they'll charge during the day to discharge at full power at night, it won't cover all of overnight use, but with the current composition of grids typically it's what you need to cover after sundown until most go to sleep (and energy use reduces). A project like this doesn't need to reduce base load, but there surely needs to be renewable (or nuclear) energy projects to replace base load. (This is an over simplification, but base load supply is built to be more steady state, less real-time reactive to the needs of grid.) Also, the "time" of the battery is just a function of storage size and inverter sizing at peak discharge. It's often selected based on how the operator expects it to be used. (For example, Tesla sells megapacks in 2hr/4hr variants [1], and I think a lot of other grid-scale storage works this way. A 2hr megapack replaces some battery bays with more inverters.) While the battery is 4 hour capable, the operator could really intend for this to provide a constant overnight supply, and it could certainly provide that. [1]: https://www.tesla.com/megapack/design https://www.tesla.com/megapack/design
- throwaway894345 5y ago
  Interesting, but does this assume Morocco won’t have a cloudy week or something (I understand that Morocco’s climate is almost always sunny, but even still)? I always assumed the threshold for “reliable renewable energy” began at a week’s worth of storage.
  - fauigerzigerk 5y ago
    Not that I know anything about this, but isn't it the job of "the grid" to make sure all sources of energy combined can provide the necessary power at all times? No individual power source is ever completely reliable. Please forgive my naive question. I have exactly zero relevant expertise.
    - cptcobalt 5y ago
      Yes! The grid is one of the world's largest real-time reactive machines. Power is generated on the fly as you need it. You could draw a direct path from your home A/C switching on all the way back up the chain to some source needing to push power to meet your demand. The supply of the grid is a big mix of diverse sources: renewables, natural gas, hydro, nuclear, coal—whatever. Each power source has different pros and cons, and you cannot really over-index on one to have a reliable grid. For example, nuclear is a fantastic and relatively clean source for base load (obviously, when handled responsibly), but not great to provide power that reacts to the shifts of demand through the day. Solar, obviously, doesn't work at night. The grid operator has to build a supply mix that fits the needs of their region. (And that's typically by facilitating energy market with different energy products where generators bid and participate.)
  - cptcobalt 5y ago
    This project also includes wind, which doesn't require sun. "A week's worth of storage" feels somewhat like an armchair simplification. It roughly sounds on target if we're moving toward the march of nines for real-time zero-carbon power supply, but...that honestly doesn't need to be the target expectation for today. Long term goal, yes. Today's targets really should be to get to 90-95% renewable, since chasing the last 5% will be the hardest and most expensive—and that's probably where we can acceptably use some dirtier backup sources. With the state of the grid today, there's lots and lots of power sources that can be replaced for cheaper and cleaner before we need to chase the long tail of 100% renewable at all hours during every season. The reality though, is that all of these requirements are modeled, and the operator surely has their install min/maxed for typical use. At the risk of being off topic, I personally have what energy-wonks could consider a "three hour" system (10kW solar, 15kW/43.5kWh battery, my typical daily energy use is between 25-50kWh, higher if I need to charge the car) and have done two weeks fully off-grid last summer, which included some cloudy weather. (True off-grid: the utility breaker was open to hard disconnect from the grid). While I could probably live full-time off grid in the summer, I definitely feel its better to have my system help clean up the grid where possible. So there's really a lot of flexibility to be had depending on the system's requirements.
    - throwaway894345 5y ago
      > "A week's worth of storage" feels somewhat like an armchair simplification. It is. The way I've heard it explained by energy experts (or people who were introduced by the media as such) was that storage needs to be on the order of days/weeks, not hours. I'm sure that's something of a crude oversimplification for a lay audience and certainly not a rugged analysis for this particular project. > With the state of the grid today, there's lots and lots of power sources that can be replaced for cheaper and cleaner before we need to chase the long tail of 100% renewable at all hours during every season. I'm of the impression that "clean carbon" peaker-plants are fairly inefficient because there's a fair amount of overhead in starting them and consequently they don't actually end up being that much cleaner. :/ Unfortunately, I wish I had more to offer than "this is what I've heard an alleged energy expert say". :(
      - 5y ago
Retric 5y ago
Capacity factor, across 24 hours a 30% capacity factor works out to 7.2 hours at max output. Being able to 4 hours of max output is therefore 55% of the total energy produced in the average day. Grid demand is also significantly higher in the daytime so that 4h of battery allows you to very closely match users demands over 24 hours or ramp up during peak demand and fall back on other sources at night.
- ncmncm 5y ago
  Just now, money is overwhelmingly better spent on top-line renewable generating capacity. There is no value in storage you are not generating enough power to keep charged up. Furthermore, cost of storage is falling even faster than wind or solar ever did. Wait until next year, and you get lots more storage for your money. Finally, batteries are good for very short-term storage -- their round-trip efficiency and fast response are unbeatable -- but they cost more than alternatives. For storage that you don't need to draw down every night, something you can fill cheap tanks with is better even if round-trip efficiency is low. Something you can also sell when your local tanks are full, and buy if they seem likely to go dry, is better yet. So, expect to see a lot of anhydrous ammonia long-term storage. Also, hydrogen, and liquid nitrogen. And, lots of tropical sites synthesizing for export to higher latitudes in winter, and lots of higher-latitude cities importing it, via ship, in place of LNG and oil. It is kind of surprising to see the batteries set up in Morocco, not in UK. That might be politics. Typically, storage is best sited near point-of-use. Maybe part of the deal is Morocco gets to use a share of the energy.
  - tuatoru 5y ago
    > expect to see a lot of anhydrous ammonia long-term storage. Also, hydrogen, and liquid nitrogen. This is wishful thinking. These are very much research projects at present. > Also, hydrogen, and liquid nitrogen. Hydrogen makes no sense as energy storage. The energy costs of compression (and liquefaction, if you do that), and the risks, are just silly. Liquifying and regasifiying nitrogen are energy-expensive too. It'd be better to combine hydrogen with air-captured carbon to make medium-chain hydrocarbons. We can store those at room temperature and pressure safely for season-long periods of time, as demonstrated by hundreds of millions of motor vehicles and tens of thousands of fuel depots. Leaks of liquid hydrocarbons are much less likely to kill people than leaks of ammonia.
MrPatan 5y ago
It's very reliable compared to 0 hours of battery storage
imtringued 5y ago
Those 4 hours don't mean wall clock time, they mean it takes 4 hours to discharge the battery at its maximum discharge rate. In other words, the battery would bridge exactly 4 hours of 0% power generation. The moment you have more than 0% power generation but still less than 100%, the battery will last more than 4 hours of wall clock time.
- ncmncm 5y ago
  The battery will mostly be drawn down at a time of much less than peak demand, so will provide power for longer than 4 hours. But it will be available too for when demand peaks above generating capacity. It is usually a mistake to overbuild batteries. The money is better spent on PV panels, wind turbines, and grid tie-ins and, eventually, other storage media.

Indeed, if we're assuming that they'll charge during the day to discharge at full power at night, it won't cover all of overnight use, but with the current composition of grids typically it's what you need to cover after sundown until most go to sleep (and energy use reduces). A project like this doesn't need to reduce base load, but there surely needs to be renewable (or nuclear) energy projects to replace base load. (This is an over simplification, but base load supply is built to be more steady state, less real-time reactive to the needs of grid.) Also, the "time" of the battery is just a function of storage size and inverter sizing at peak discharge. It's often selected based on how the operator expects it to be used. (For example, Tesla sells megapacks in 2hr/4hr variants [1], and I think a lot of other grid-scale storage works this way. A 2hr megapack replaces some battery bays with more inverters.) While the battery is 4 hour capable, the operator could really intend for this to provide a constant overnight supply, and it could certainly provide that. [1]: https://www.tesla.com/megapack/design https://www.tesla.com/megapack/design

throwaway894345 5y ago
Interesting, but does this assume Morocco won’t have a cloudy week or something (I understand that Morocco’s climate is almost always sunny, but even still)? I always assumed the threshold for “reliable renewable energy” began at a week’s worth of storage.
- fauigerzigerk 5y ago
  Not that I know anything about this, but isn't it the job of "the grid" to make sure all sources of energy combined can provide the necessary power at all times? No individual power source is ever completely reliable. Please forgive my naive question. I have exactly zero relevant expertise.
  - cptcobalt 5y ago
    Yes! The grid is one of the world's largest real-time reactive machines. Power is generated on the fly as you need it. You could draw a direct path from your home A/C switching on all the way back up the chain to some source needing to push power to meet your demand. The supply of the grid is a big mix of diverse sources: renewables, natural gas, hydro, nuclear, coal—whatever. Each power source has different pros and cons, and you cannot really over-index on one to have a reliable grid. For example, nuclear is a fantastic and relatively clean source for base load (obviously, when handled responsibly), but not great to provide power that reacts to the shifts of demand through the day. Solar, obviously, doesn't work at night. The grid operator has to build a supply mix that fits the needs of their region. (And that's typically by facilitating energy market with different energy products where generators bid and participate.)
- cptcobalt 5y ago
  This project also includes wind, which doesn't require sun. "A week's worth of storage" feels somewhat like an armchair simplification. It roughly sounds on target if we're moving toward the march of nines for real-time zero-carbon power supply, but...that honestly doesn't need to be the target expectation for today. Long term goal, yes. Today's targets really should be to get to 90-95% renewable, since chasing the last 5% will be the hardest and most expensive—and that's probably where we can acceptably use some dirtier backup sources. With the state of the grid today, there's lots and lots of power sources that can be replaced for cheaper and cleaner before we need to chase the long tail of 100% renewable at all hours during every season. The reality though, is that all of these requirements are modeled, and the operator surely has their install min/maxed for typical use. At the risk of being off topic, I personally have what energy-wonks could consider a "three hour" system (10kW solar, 15kW/43.5kWh battery, my typical daily energy use is between 25-50kWh, higher if I need to charge the car) and have done two weeks fully off-grid last summer, which included some cloudy weather. (True off-grid: the utility breaker was open to hard disconnect from the grid). While I could probably live full-time off grid in the summer, I definitely feel its better to have my system help clean up the grid where possible. So there's really a lot of flexibility to be had depending on the system's requirements.
  - throwaway894345 5y ago
    > "A week's worth of storage" feels somewhat like an armchair simplification. It is. The way I've heard it explained by energy experts (or people who were introduced by the media as such) was that storage needs to be on the order of days/weeks, not hours. I'm sure that's something of a crude oversimplification for a lay audience and certainly not a rugged analysis for this particular project. > With the state of the grid today, there's lots and lots of power sources that can be replaced for cheaper and cleaner before we need to chase the long tail of 100% renewable at all hours during every season. I'm of the impression that "clean carbon" peaker-plants are fairly inefficient because there's a fair amount of overhead in starting them and consequently they don't actually end up being that much cleaner. :/ Unfortunately, I wish I had more to offer than "this is what I've heard an alleged energy expert say". :(
    - 5y ago

Retric 5y ago

Capacity factor, across 24 hours a 30% capacity factor works out to 7.2 hours at max output. Being able to 4 hours of max output is therefore 55% of the total energy produced in the average day. Grid demand is also significantly higher in the daytime so that 4h of battery allows you to very closely match users demands over 24 hours or ramp up during peak demand and fall back on other sources at night.

ncmncm 5y ago
Just now, money is overwhelmingly better spent on top-line renewable generating capacity. There is no value in storage you are not generating enough power to keep charged up. Furthermore, cost of storage is falling even faster than wind or solar ever did. Wait until next year, and you get lots more storage for your money. Finally, batteries are good for very short-term storage -- their round-trip efficiency and fast response are unbeatable -- but they cost more than alternatives. For storage that you don't need to draw down every night, something you can fill cheap tanks with is better even if round-trip efficiency is low. Something you can also sell when your local tanks are full, and buy if they seem likely to go dry, is better yet. So, expect to see a lot of anhydrous ammonia long-term storage. Also, hydrogen, and liquid nitrogen. And, lots of tropical sites synthesizing for export to higher latitudes in winter, and lots of higher-latitude cities importing it, via ship, in place of LNG and oil. It is kind of surprising to see the batteries set up in Morocco, not in UK. That might be politics. Typically, storage is best sited near point-of-use. Maybe part of the deal is Morocco gets to use a share of the energy.
- tuatoru 5y ago
  > expect to see a lot of anhydrous ammonia long-term storage. Also, hydrogen, and liquid nitrogen. This is wishful thinking. These are very much research projects at present. > Also, hydrogen, and liquid nitrogen. Hydrogen makes no sense as energy storage. The energy costs of compression (and liquefaction, if you do that), and the risks, are just silly. Liquifying and regasifiying nitrogen are energy-expensive too. It'd be better to combine hydrogen with air-captured carbon to make medium-chain hydrocarbons. We can store those at room temperature and pressure safely for season-long periods of time, as demonstrated by hundreds of millions of motor vehicles and tens of thousands of fuel depots. Leaks of liquid hydrocarbons are much less likely to kill people than leaks of ammonia.

MrPatan 5y ago

It's very reliable compared to 0 hours of battery storage

imtringued 5y ago

Those 4 hours don't mean wall clock time, they mean it takes 4 hours to discharge the battery at its maximum discharge rate. In other words, the battery would bridge exactly 4 hours of 0% power generation. The moment you have more than 0% power generation but still less than 100%, the battery will last more than 4 hours of wall clock time.

ncmncm 5y ago
The battery will mostly be drawn down at a time of much less than peak demand, so will provide power for longer than 4 hours. But it will be available too for when demand peaks above generating capacity. It is usually a mistake to overbuild batteries. The money is better spent on PV panels, wind turbines, and grid tie-ins and, eventually, other storage media.