Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

How may EVs does it take before demands become an issue for the Grid?

This site may earn commission on affiliate links.
Cottonwood's points are exactly right. I'll add that solar doesn't generate at night which is when most EVs will charge. Imagine a world (maybe 50 years from now) where everyone has solar + EVs. We generate a ton of distributed electricity during the day but then use a ton of electricity at night to charge our EVs. Local storage is the only good solution I can see for this whether it's at the customer location or at the neighborhood or substation level. Fortunately, someone is stepping up in a big way to facilitate exactly that solution (thanks Elon!). Battery storage will keep solar's continued growth viable, stabilizing the grid and defending against the naysayers who can't see past the end of their nose into our inevitable renewable energy future.

- - - Updated - - -



Energy is energy. That "most" that is not electricity could be used to make electricity instead or do lots of other things, no?

Well, the obvious answer to the scenario you describe would be to shift most of the charging to happen in the day at work instead. Done properly, it wouldn't be any more inconvenient.

Yes, that energy could be used to make electricity - but you'd only get ~40% of it as electricity due to losses on the conversion by way of steam.
 
Well, the obvious answer to the scenario you describe would be to shift most of the charging to happen in the day at work instead. Done properly, it wouldn't be any more inconvenient.

That's almost certain to happen. Take a look at a midsummer grid demand curve for South Australia, which has a very high penetration of distributed behind-the-meter solar:

SA-Pitty-sherry.png


Their peak demand is at midnight when everyone's programmable water heater turns on to take advantage of "super off-peak" rate! Oops! An adjustment to time-of-use pricing seems likely to occur there soon.
 
Well, the obvious answer to the scenario you describe would be to shift most of the charging to happen in the day at work instead. Done properly, it wouldn't be any more inconvenient.

Yes, that energy could be used to make electricity - but you'd only get ~40% of it as electricity due to losses on the conversion by way of steam.

Charging at work doesn't seem too obvious to me. That requires a lot of charging infrastructure to be built in concentrated areas (work parking lots) that doesn't exist today. Every spot would have a charger of some kind. A spot with a broken charger becomes an unused spot. Then there's payment (who pays, how, etc.). I'm not sure any of that is practical. That also shifts the distributed charging load (one or two cars per house) to concentrated charging areas (office buildings). You're generating distributed power to push through the grid to concentrated parking lot daytime charging. Another down side is that people would plug in to start their charge when they arrive at work, typically 8 or 9 am, when solar is still not producing much. They're not going to care enough to schedule the charge for later or they'll just want to have the car charged asap anyhow. The obvious advantage of charging at night is that I'm asleep and there's far less chance I'll need to run out at 2am than at 11am.

I think the only way to ensure long term grid stability and convenience is some type of storage. For now it's batteries. In the future maybe it will be something even better.

- - - Updated - - -

That's almost certain to happen. Take a look at a midsummer grid demand curve for South Australia, which has a very high penetration of distributed behind-the-meter solar:

Any idea what that graph looked like before all that solar?
 
Well, the obvious answer to the scenario you describe would be to shift most of the charging to happen in the day at work instead. Done properly, it wouldn't be any more inconvenient.

Well, maybe not more inconvenient for the EV driver, but it could be problematic for other reasons. Rather than charging stations distributed out among owners' homes, you would now require concentrations of them at, for example, workplace or commuter parking lots. This localized high load concentration could be taxing on the local electricity distribution system.
 
You can use the growth of the Internet speeds as a rough model for organic grid growth over time, making the assumption that generation will keep up. I believe you'll see an organic growth across the entire distribution grid as more demand gets placed on the system. T1's led to T3's, led to OC3's, led to OC12's and OC48's, which led to OC-192's. At several steps in that trajectory, you replaced the physical infrastructure - copper lines with optical multiplexers, then bigger multiplexers, then direct fiber attach, then WDM. That same physical infrastructure build-out will occur in the grid over time.
 
You can use the growth of the Internet speeds as a rough model for organic grid growth over time, making the assumption that generation will keep up. I believe you'll see an organic growth across the entire distribution grid as more demand gets placed on the system. T1's led to T3's, led to OC3's, led to OC12's and OC48's, which led to OC-192's. At several steps in that trajectory, you replaced the physical infrastructure - copper lines with optical multiplexers, then bigger multiplexers, then direct fiber attach, then WDM. That same physical infrastructure build-out will occur in the grid over time.

The first optic fibers had nothing like the theoretical maximum of BPS traversing them - there was a lot of development in a short period of time and it was cost-effective to bury a big bundle of fiber rather than put in individual strands. Despite this after two years of effort I still can't convince te only ISP I can access to light-up a second strand to reduce their backhaul congestion in my area. Am considering bringing the towns they serve together with a muni fiber proposal. %$(&% rent-seeking monopolies...

Barring new physics (very high-temp superconductors?) power lines and substations are already running near their theoretical maximums at peak times, and will have only linear additive growth in capacity.

That said -- there's still a LOT of room for end-use efficiency to reduce peak electric demand in commercial and industrial processes, and this can free up a whole lot of spare capacity for vehicle charging.
 
The first optic fibers had nothing like the theoretical maximum of BPS traversing them - there was a lot of development in a short period of time and it was cost-effective to bury a big bundle of fiber rather than put in individual strands. Despite this after two years of effort I still can't convince te only ISP I can access to light-up a second strand to reduce their backhaul congestion in my area. Am considering bringing the towns they serve together with a muni fiber proposal. %$(&% rent-seeking monopolies...

Barring new physics (very high-temp superconductors?) power lines and substations are already running near their theoretical maximums at peak times, and will have only linear additive growth in capacity.

That said -- there's still a LOT of room for end-use efficiency to reduce peak electric demand in commercial and industrial processes, and this can free up a whole lot of spare capacity for vehicle charging.

I recognize you can't do a 1/1 comparison, but at each step, infrastructure had to be touched / upgraded. Whether fiber was fully utilized, it still was replaced along the way with better types that could handle different dispersion characteristics needed for WDM.

At the same time, though, we're talking about the typical distribution grid. At each step of the build-out, some additional infrastructure work was required. Going from T1's, provided over the good ole' copper bundle going into a building led to T3's (whether coax-provided or from an FMT150) on old SMF, which led to fiber and equipment replacement when OC12 and OC48 SONET mux equipment came into play. That, in turn led to new fiber being pulled for direct fiber and WDM applications.

Organically, the grid will grow as customers place more demand upon it. It'll grow in the residential areas (we just had all of our poles replaced and new distribution line run within the past 5 years in my rural neighborhood), it'll grow at the distribution, it'll grow in the highline cores. It will happen naturally... there will be hot spots, but they'll each be solved over time.

And I agree with you that, like data communications leveraged compression and other interesting mechanisms to delay infrastructure upgrades, we'll see things like Tesla's energy storage and peak management help to defer massive upgrades.
 
If you're going to be putting in mass charging at "work" locations, it might as well be grid-aware so that the utility can shape the demand and perhaps paired with local storage. Also all of those businesses should be putting in solar as well, both on their roofs and over all those charging EVs, so there should be significant solar capacity that is local.
 
Cottonwood's points are exactly right. I'll add that solar doesn't generate at night which is when most EVs will charge. Imagine a world (maybe 50 years from now) where everyone has solar + EVs. We generate a ton of distributed electricity during the day but then use a ton of electricity at night to charge our EVs. Local storage is the only good solution I can see for this whether it's at the customer location or at the neighborhood or substation level. Fortunately, someone is stepping up in a big way to facilitate exactly that solution (thanks Elon!). Battery storage will keep solar's continued growth viable, stabilizing the grid and defending against the naysayers who can't see past the end of their nose into our inevitable renewable energy future.

- - - Updated - - -



Energy is energy. That "most" that is not electricity could be used to make electricity instead or do lots of other things, no?

Destination solar charging would solve this. Every surface of the destination structures, buildings (between windows, entire roofs), canopies, parking pavement, and street pavement would be covered with solar. Meters would charge car users for energy used. Automatic charing snakes would plug into vehicles without interaction by the users according to user preprogramming settings, and would unplug as soon as the vehicle was unlocked when the customer comes back, and the car would not shift into gear until the snake was safely away. Special lensing on the rough-surface tire-gripping roadways would direct the energy from the roads into the collectors beneath them, and the lenses would be physically built and supported to handle heavy vehicular traffic, vacuumed daily and rinsed weekly (with collection of the rinse water for recycling). Destination charging would definitely solve daytime charging questions. When destination charging fails, home charging with batteries to shift time would work as you mentioned.

Regarding refinery energy from oil refining to gas, the energy gained from burning that oil would no longer be done if the oil wasn't in the refinery to burn, and the process-particular energy transferences which used that oil not only wouldn't happen, but the oil wouldn't be in position for that process to exist in the first place, and the particular processes wouldn't be done (obviously). Plus we don't want to burn oil for energy anyway. Energy not gained from burning the oil during refining would not be otherwise gained by not refining. That's what I'm guessing and I'm almost certainly correct.

- - - Updated - - -

Charging at work doesn't seem too obvious to me. That requires a lot of charging infrastructure to be built in concentrated areas (work parking lots) that doesn't exist today. Every spot would have a charger of some kind. A spot with a broken charger becomes an unused spot. Then there's payment (who pays, how, etc.). I'm not sure any of that is practical.

I find it fascinating that you mention all the same issues I do, and I say "therefore it's possible and easy" and you come to the opposite conclusion.

We have extra people on the planet. Finding more work for them to do and more purposes for them to live to do that work is a good thing, not a bad thing. Solutions that save our planet (pollution mostly) are a good thing, not a nuisance. The only entities that think solutions are a nuisance are old-fashioned corporations, governments, etc., that are vested in broken ways, and hate solutions. Everyone else wants the solutions, is willing to do the work, and can definitely afford the more affordable future options than the less affordable legacy options that are growing cancerous, outdated and expensive.

- - - Updated - - -

Another down side is that people would plug in to start their charge when they arrive at work, typically 8 or 9 am, when solar is still not producing much. They're not going to care enough to schedule the charge for later or they'll just want to have the car charged asap anyhow. The obvious advantage of charging at night is that I'm asleep and there's far less chance I'll need to run out at 2am than at 11am.

The computers can be programmed appropriately. For instance, I arrive at my parking garage at 5:05 AM. It would plug in. At around 10AM, it would start charging it. When I get in to leave at 2:30PM, it stops charging. Then I drive home, arriving at approximately 3:55PM give or take (due to traffic). It has about 1 hour left of charging at home, then no more sunlight. That's a solid 4 hours of direct sun to car right there (possibly more, like 6, if you count non-peak sun).
 
Last edited:
That's almost certain to happen. Take a look at a midsummer grid demand curve for South Australia, which has a very high penetration of distributed behind-the-meter solar:

SA-Pitty-sherry.png


Their peak demand is at midnight when everyone's programmable water heater turns on to take advantage of "super off-peak" rate! Oops! An adjustment to time-of-use pricing seems likely to occur there soon.

If this isn't proof that government and monopoly policy have major effect, I don't know what is. That's why real smart people need to be involved in that policy, such as diffusing the systems to natural selection (marketplace competition), or something equivalent. Otherwise you have entirely broken policies like the joke above. All the use at a particular hour? Dumb to begin with! Duh! Anybody who has used multiprocess multiuser computer farms when cron jobs start would know this!