4 ms·

Not my area of expertise, but people that claim to be experts suggest that global electricity generation is enough: https://www.barrons.com/articles/theres-enou

Not my area of expertise, but people that claim to be experts suggest that global electricity generation is enough: https://www.barrons.com/articles/theres-enough-electricity-in-the-world-for-electric-vehicles-heres-who-will-charge-them-51605368406 https://www.barrons.com/articles/theres-enough-electricity-i... My intuitive guess is your simple model is wrong, but I'm not an expert.

robomartin 5y ago
Thanks for the link. Sadly, no, the "analysis" in the article --and I am being kind-- is overly simplistic. This is the kind of hand-wavy thing I've been running into when I look for prior work on the subject. Here's an analogy that might convey the nature of the hand-waving: A typical home swimming pool requires 20,000 gallons of water. Let's say the average flow rate of a good quality garden hose is 20 gallons per minute. That means you can fill that pool in 1,000 minutes, or, about 17 hours. Can you fill the pool in 8.5 hours if you use two hoses? No. Why? Because your plumbing will establish a limit on just how much water you can flow per unit time. Anyone who has ever dealt with plumbing understands this. In some installations, if you flush the toilet and someone is taking a shower they get burned because less cold water flows to mix with hot water. In my home, when the sprinkler system is on there's a marked difference in flow rate everywhere in the house. What if a thousand of your neighbors want to fill their 20,000 gallon swimming pools simultaneously? Could they? Very likely not. Well, not in 17 hours. The large pipes supplying water to a neighborhood have a flow rate limit as well. Under normal usage they are adequate for the usage profile of a certain number of homes in a neighborhood. Are these pipes designed and specified to manage twice that design rate? Likely not. And certainly not five or ten times. Most neighborhoods in the US are supplied with drinking water from large water tanks strategically located higher than the homes they service. This isn't universal across the entire nation but it is very common. Here comes the next problem. Filling a thousand 20K gallon swimming pools requires two million gallons of water. This, over and above the water required for all the other things people do with water, from drinking to flushing the toilet. A typical neighborhood water tank might hold one or two days of water, as defined by the typical usage for the area. It certainly does not hold an extra two million gallons to dispense over 17 hours. As the water level decreases due to either normal or abnormal usage, pumps are engaged to maintain water level. Once again, these pumps are sized for typical usage as it might pertain to the specific installation. They are not sized to provide a massive step change in water pumping from the main aqueduct into the neighborhood tank. As you can imagine, this analysis can extend beyond neighborhood tanks and explore issues with aqueducts, water treatment plants, etc. What these articles do is they say something like "the world capacity is N GWh. We need 1/2N GWh for electric cars. We only use 1/2N for everything else. We have what we need." Well, no. We don't. There's also a confusion between power and energy. Power is energy per unit time. Put a different way: There's a huge difference between walking a mile and running a mile. What these articles ignore is that the electric car charging problem is like the neighborhood swimming pool issue. When everyone in the neighborhood wants to charge their cars in eight hours or less, what you need isn't just the energy (the water level in the pool), you also need power, you need to be able to deliver a lot more gallons per minute at every home, street, neighborhood, town, city and region. Exiting power plants cannot do this. Furthermore, by definition, if we want coal power plants to deliver twice their average power to charge cars, you have to burn at least two times the coal you were burning before. Hence me reaching for nuclear power plants...because any other option would be somewhere between unrealizable (the plant just can't produce twice the power) to unthinkable (pollution). Put a different way: Let's say your city is fed by a 1 GW nuclear power plant today. Everyone is driving gasoline cars. Median load factor for nuclear power plants is in the order of 80%. That means you only have a 20% excess capacity. Now everyone buys electric cars and, of course, everyone wants to charge them as quickly as possible, with some willing to wait overnight. Well, your power plant will not be able to supply this extra power. We don't design power plants to be twice or three times larger than what we need. Which means you'll need to build one or two additional plants as soon as possible. I have yet to see a study of model of electric car charging energy and power requirements that passes the physics test. Lots of hand-waving and "trust me, I am a <insert appeal to authority>" but no numbers. Science is about numbers. My attempt isn't complete at all. However, like I said, even if I am off by a factor of ten, we have a problem. Heck, we likely have a problem if I am off by a factor of 20. I think what everyone is counting on is everyone covering neighborhoods with solar power. As the owner of a 40 panel array I built myself to, among other things, power my CNC shop and, after expansion, support electric vehicles, I can tell you without a shadow of a doubt that the idea of a clean solar future for electric cars is a fantasy. My energy logs say so. I would need somewhere in the order of three to five times the array to truly be "green" as far as electric vehicle charging would be concerned. Any way you look at it, that is impossible.
- fossuser 5y ago
  I don't think everyone wants to charge their car in eight hours or less or at the same time in most cases. Seems like load balancing would probably work out in the short term if necessary. You've obviously thought about this more than I have so I don't have much to say - I'm just skeptical of answers like this since I think they often turn out to be wrong. Not very satisfying, but I guess we'll see what ends up happening. When there’s demand, capitalism and markets are usually pretty good at solving for it.
  - robomartin 5y ago
    The reason I modeled an 8 hour charge is because this would be what most people would have available after they get home from work. I modeled a rolling scenario where 1/8 of the fleet starts charging every two hours. It isn’t a perfect model, of course. The intent was to simulate both time zones and people coming home from work at different times. If we accept the above and add the idea that people will charge at work for an additional eight hours, for a total of 16, then power requirement is cut in half. This would still require a massive increase in power generation and delivery capacity. And then, of course, there’s a percentage of the population that will need to charge fast. That means half an hour. The problem is power. Energy can be delivered over time, power is instantaneous. We don’t have enough power. While I agree that entrepreneurs often find a way, we still haven’t come up with a way to violate the rules of physics. That’s a hard limit.

robomartin 5y ago

Thanks for the link. Sadly, no, the "analysis" in the article --and I am being kind-- is overly simplistic. This is the kind of hand-wavy thing I've been running into when I look for prior work on the subject. Here's an analogy that might convey the nature of the hand-waving: A typical home swimming pool requires 20,000 gallons of water. Let's say the average flow rate of a good quality garden hose is 20 gallons per minute. That means you can fill that pool in 1,000 minutes, or, about 17 hours. Can you fill the pool in 8.5 hours if you use two hoses? No. Why? Because your plumbing will establish a limit on just how much water you can flow per unit time. Anyone who has ever dealt with plumbing understands this. In some installations, if you flush the toilet and someone is taking a shower they get burned because less cold water flows to mix with hot water. In my home, when the sprinkler system is on there's a marked difference in flow rate everywhere in the house. What if a thousand of your neighbors want to fill their 20,000 gallon swimming pools simultaneously? Could they? Very likely not. Well, not in 17 hours. The large pipes supplying water to a neighborhood have a flow rate limit as well. Under normal usage they are adequate for the usage profile of a certain number of homes in a neighborhood. Are these pipes designed and specified to manage twice that design rate? Likely not. And certainly not five or ten times. Most neighborhoods in the US are supplied with drinking water from large water tanks strategically located higher than the homes they service. This isn't universal across the entire nation but it is very common. Here comes the next problem. Filling a thousand 20K gallon swimming pools requires two million gallons of water. This, over and above the water required for all the other things people do with water, from drinking to flushing the toilet. A typical neighborhood water tank might hold one or two days of water, as defined by the typical usage for the area. It certainly does not hold an extra two million gallons to dispense over 17 hours. As the water level decreases due to either normal or abnormal usage, pumps are engaged to maintain water level. Once again, these pumps are sized for typical usage as it might pertain to the specific installation. They are not sized to provide a massive step change in water pumping from the main aqueduct into the neighborhood tank. As you can imagine, this analysis can extend beyond neighborhood tanks and explore issues with aqueducts, water treatment plants, etc. What these articles do is they say something like "the world capacity is N GWh. We need 1/2N GWh for electric cars. We only use 1/2N for everything else. We have what we need." Well, no. We don't. There's also a confusion between power and energy. Power is energy per unit time. Put a different way: There's a huge difference between walking a mile and running a mile. What these articles ignore is that the electric car charging problem is like the neighborhood swimming pool issue. When everyone in the neighborhood wants to charge their cars in eight hours or less, what you need isn't just the energy (the water level in the pool), you also need power, you need to be able to deliver a lot more gallons per minute at every home, street, neighborhood, town, city and region. Exiting power plants cannot do this. Furthermore, by definition, if we want coal power plants to deliver twice their average power to charge cars, you have to burn at least two times the coal you were burning before. Hence me reaching for nuclear power plants...because any other option would be somewhere between unrealizable (the plant just can't produce twice the power) to unthinkable (pollution). Put a different way: Let's say your city is fed by a 1 GW nuclear power plant today. Everyone is driving gasoline cars. Median load factor for nuclear power plants is in the order of 80%. That means you only have a 20% excess capacity. Now everyone buys electric cars and, of course, everyone wants to charge them as quickly as possible, with some willing to wait overnight. Well, your power plant will not be able to supply this extra power. We don't design power plants to be twice or three times larger than what we need. Which means you'll need to build one or two additional plants as soon as possible. I have yet to see a study of model of electric car charging energy and power requirements that passes the physics test. Lots of hand-waving and "trust me, I am a <insert appeal to authority>" but no numbers. Science is about numbers. My attempt isn't complete at all. However, like I said, even if I am off by a factor of ten, we have a problem. Heck, we likely have a problem if I am off by a factor of 20. I think what everyone is counting on is everyone covering neighborhoods with solar power. As the owner of a 40 panel array I built myself to, among other things, power my CNC shop and, after expansion, support electric vehicles, I can tell you without a shadow of a doubt that the idea of a clean solar future for electric cars is a fantasy. My energy logs say so. I would need somewhere in the order of three to five times the array to truly be "green" as far as electric vehicle charging would be concerned. Any way you look at it, that is impossible.

fossuser 5y ago
I don't think everyone wants to charge their car in eight hours or less or at the same time in most cases. Seems like load balancing would probably work out in the short term if necessary. You've obviously thought about this more than I have so I don't have much to say - I'm just skeptical of answers like this since I think they often turn out to be wrong. Not very satisfying, but I guess we'll see what ends up happening. When there’s demand, capitalism and markets are usually pretty good at solving for it.
- robomartin 5y ago
  The reason I modeled an 8 hour charge is because this would be what most people would have available after they get home from work. I modeled a rolling scenario where 1/8 of the fleet starts charging every two hours. It isn’t a perfect model, of course. The intent was to simulate both time zones and people coming home from work at different times. If we accept the above and add the idea that people will charge at work for an additional eight hours, for a total of 16, then power requirement is cut in half. This would still require a massive increase in power generation and delivery capacity. And then, of course, there’s a percentage of the population that will need to charge fast. That means half an hour. The problem is power. Energy can be delivered over time, power is instantaneous. We don’t have enough power. While I agree that entrepreneurs often find a way, we still haven’t come up with a way to violate the rules of physics. That’s a hard limit.