3 ms·

However you slice it, you have to deliver say 40KWh to that many cars overnight, or they fail to charge and thus are useless. So the grid has to handle that and

by lstodd 6y ago

However you slice it, you have to deliver say 40KWh to that many cars overnight, or they fail to charge and thus are useless. So the grid has to handle that and even if you can get rid of extreme peaks with "smartness", you can't lower the cables and switches ratings much.

Reason077 6y ago
First of all, the average car does not need 40 kWh every day. The average driver in the USA drives 13,500 miles per year (far more than the average driver in the UK/Europe!) That's about 37 miles per day, so about 12.5 kWh per night on average at ~3 miles per kWh. And remember, those who are driving long distance are probably going to be using fast chargers on-route, so their domestic charging demand will be lower. Secondly, even if they did need 40 kWh every night, that only equates to 4 kW over a 10 hour charging cycle. Easily manageable by avoiding the typical late afternoon / early evening peaks. Most grids have a lot of spare capacity between, say, 9PM and 7AM.
- lstodd 6y ago
  You fail to take into account inefficiencies in transmission and charging. That would multiply required input by a factor of two or three, depending where you measure it. So no. Switch most commute to electric - then at first grid goes down, then power plants struggle. "Most grids have a lot of spare capacity between, say, 9PM and 7AM" - that's not the grids we're talking about. HV circuits might not feel it, but something laid down for a suburb was scrupulously designed to carry only just enough, because it's basically burying refined metal, be it copper or aluminium. There is no spare even 4KW for every house and never has been because that would have cost more in about everything - meters, cables, multiple stages of step-down transformers, transmission lines, generation. Electicity is not free, never has been, and distibution costs are surprisingly a very important consideration.
  - Reason077 6y ago
    > "There is no spare even 4KW for every house and never has been" Yes, there is. It's just the timing that matters. Domestic electricity demand during off-peak overnight hours is as little as 1/3 as it is at peak time. If all those houses can run 4 kW of air conditioning, or heating, or electric ovens at 6PM, then there is 4 kW to spare at 2AM. This issue has actually been studied in some detail by the UK's National Grid. The conclusion was that additional demand from millions of electric vehicles would cause issues if they all were charged at peak times. But, provided demand can be managed, existing grid infrastructure can cope with an entirely electric vehicle fleet with modest investments in the coming years/decades. Fun fact: UK grid demand used to be significantly higher than it is today. It peaked at 62 GW in 2002, but it now rarely gets above 45 GW, despite significant population growth! Much of this reduction in demand is due to improved energy efficiency. If we could handle 62 GW in 2002, then there's no reason that we can't handle 62 GW again in, say, 2042.
    - lstodd 6y ago
      > If we could handle 62 GW in 2002, then there's no reason that we can't handle 62 GW again in, say, 2042. There is - supply and grid maintenance was scaled down with demand. The grid certainly can be rebuilt to 2002 state in UK, but this will not be cheap. The other point that I was making is that any deviation from current patterns, like night demand significantly growing will require significant changes in generation.
      - Reason077 6y ago
        > "The grid certainly can be rebuilt to 2002 state in UK, but this will not be cheap." It doesn't need "rebuilding". The grid, broadly speaking, hasn't changed since 2002. Generation has changed dramatically, but generally the transmission lines that served old power plants do not get removed when the plants close. They're still there, still active, and are often reused (for example, off-shore wind farm built off the coast from closed coal plant). > "The other point that I was making is that any deviation from current patterns, like night demand significantly growing will require significant changes in generation." First you were talking about lines, now you are talking about generation. But it's the same story. Power plants get turned off and curtailed at night because there is less demand. If demand increases at night? Plenty of spare capacity available, they just need to run for longer.
  - Reason077 6y ago
    > "You fail to take into account inefficiencies in transmission and charging. That would multiply required input by a factor of two or three, depending where you measure it." I do not account for transmission inefficiencies, as this is highly variable depending on your location and grid infrastructure. But it would be extremely unusual for transmission losses to be as high as 66% like you suggest! As for charging losses, that's something like 10%. About 90% of the energy as measured at your domestic electricity meter will make it into the battery. I did account for this, by giving a deliberately low miles per kWh figure. In reality, most EVs will get significantly better efficiency than the 3 miles per kWh that I quoted.

by lstodd 6y ago

Reason077 6y ago

First of all, the average car does not need 40 kWh every day. The average driver in the USA drives 13,500 miles per year (far more than the average driver in the UK/Europe!) That's about 37 miles per day, so about 12.5 kWh per night on average at ~3 miles per kWh. And remember, those who are driving long distance are probably going to be using fast chargers on-route, so their domestic charging demand will be lower. Secondly, even if they did need 40 kWh every night, that only equates to 4 kW over a 10 hour charging cycle. Easily manageable by avoiding the typical late afternoon / early evening peaks. Most grids have a lot of spare capacity between, say, 9PM and 7AM.

lstodd 6y ago
You fail to take into account inefficiencies in transmission and charging. That would multiply required input by a factor of two or three, depending where you measure it. So no. Switch most commute to electric - then at first grid goes down, then power plants struggle. "Most grids have a lot of spare capacity between, say, 9PM and 7AM" - that's not the grids we're talking about. HV circuits might not feel it, but something laid down for a suburb was scrupulously designed to carry only just enough, because it's basically burying refined metal, be it copper or aluminium. There is no spare even 4KW for every house and never has been because that would have cost more in about everything - meters, cables, multiple stages of step-down transformers, transmission lines, generation. Electicity is not free, never has been, and distibution costs are surprisingly a very important consideration.
- Reason077 6y ago
  > "There is no spare even 4KW for every house and never has been" Yes, there is. It's just the timing that matters. Domestic electricity demand during off-peak overnight hours is as little as 1/3 as it is at peak time. If all those houses can run 4 kW of air conditioning, or heating, or electric ovens at 6PM, then there is 4 kW to spare at 2AM. This issue has actually been studied in some detail by the UK's National Grid. The conclusion was that additional demand from millions of electric vehicles would cause issues if they all were charged at peak times. But, provided demand can be managed, existing grid infrastructure can cope with an entirely electric vehicle fleet with modest investments in the coming years/decades. Fun fact: UK grid demand used to be significantly higher than it is today. It peaked at 62 GW in 2002, but it now rarely gets above 45 GW, despite significant population growth! Much of this reduction in demand is due to improved energy efficiency. If we could handle 62 GW in 2002, then there's no reason that we can't handle 62 GW again in, say, 2042.
  - lstodd 6y ago
    > If we could handle 62 GW in 2002, then there's no reason that we can't handle 62 GW again in, say, 2042. There is - supply and grid maintenance was scaled down with demand. The grid certainly can be rebuilt to 2002 state in UK, but this will not be cheap. The other point that I was making is that any deviation from current patterns, like night demand significantly growing will require significant changes in generation.
    - Reason077 6y ago
      > "The grid certainly can be rebuilt to 2002 state in UK, but this will not be cheap." It doesn't need "rebuilding". The grid, broadly speaking, hasn't changed since 2002. Generation has changed dramatically, but generally the transmission lines that served old power plants do not get removed when the plants close. They're still there, still active, and are often reused (for example, off-shore wind farm built off the coast from closed coal plant). > "The other point that I was making is that any deviation from current patterns, like night demand significantly growing will require significant changes in generation." First you were talking about lines, now you are talking about generation. But it's the same story. Power plants get turned off and curtailed at night because there is less demand. If demand increases at night? Plenty of spare capacity available, they just need to run for longer.
- Reason077 6y ago
  > "You fail to take into account inefficiencies in transmission and charging. That would multiply required input by a factor of two or three, depending where you measure it." I do not account for transmission inefficiencies, as this is highly variable depending on your location and grid infrastructure. But it would be extremely unusual for transmission losses to be as high as 66% like you suggest! As for charging losses, that's something like 10%. About 90% of the energy as measured at your domestic electricity meter will make it into the battery. I did account for this, by giving a deliberately low miles per kWh figure. In reality, most EVs will get significantly better efficiency than the 3 miles per kWh that I quoted.