Electric cars--can the power grid handle the load?

Factual Questions
1

A couple years ago, there was a lot of discussion here at the Dope about the viability of electric cars.
And one of the big issues was the incapability of the nationwide electric grid to produce enough power for 50 million vehicles to recharge sumultaneously. This was expected to be a critical problem, not just in the US, but also in many other countries.

So what happened?

We are now realistically approaching the day when most new cars will be electric. But it seems that nobody is worried any more.

2

Seems like with the discovery of the AI Data center, we’ve suddenly figured out how to add a shitload of power to the grid. They’re building them everywhere.

Each one uses enough power to charge around 300k EVs per day.

3

I can’t say I’ve seen any expressions of concern in the UK over capacity to cope with electric cars. Indeed, where I live, one energy company’s been allowed to install charging terminals on the streetlamp posts.

No doubt the planners in the supply system are building all these considerations in - AFAIK planning applications for data centres usually include plans to generate their own supply, but there is concern that much of the “planning” for data centres has been speculative hype.

4

There’s never been any real concern. There has been lots and lots of FUD spread by the fossil fuel and gas car industries and the RW political haters that there should be concern. About 99% of which are raw lies.

The cars mostly charge when the electrical supply is large and the electrical demand is low. It’ll be pure profit for the power companies at little marginal cost of new upstream infrastructure.

Instead, the infrastructure expense is mostly building out the charger networks and installing chargers in individual homes. Exactly the kind of entreprenureal and consumer stuff we usually succeed at.

5

Asking if there is enough capacity for 50 million vehicles to recharge simultaneously is like asking if there is enough petrol to refill 50 million cars simultaneously. The situation simply doesn’t come up.

6

Our power company encourages EV ownership. Thus, I have concluded that they believe they can handle it. They know more about the grid than I do.

7

It’s not really an issue at the grid level, but might be an issue at the level of the neighborhood gas station, rural county homes, etc. It depends on the lines and substations present in the area.

It’s one thing for a homeowner to plug in a level 1 charger (it’s just like plugging in a space heater; don’t overdo it and you’re fine) or a level 2 one (might require some rewiring and a new panel but only in that home).

But for a random gas station to be able to add a few DC fast chargers, it’s a whole lot more expensive. They might need to add more transformers and get 3 phase power from somewhere. If they want to support more than just a couple, like a whole supercharger section, it gets even more expensive. And those things are huge and take up a lot of parking space for typically 30 to 60 minutes at a time.

In other words, we’re not really faced with an inability to produce enough power, but distributing it across this vast country at EV scale is a challenge that is difficult to solve. In a way, data centers are easier to deal with because of the scale and concentration involved; a large steady draw at one place is a very different problem than a bunch of peaky bursts scattered all over the place.

Anyhow, we’re at no risk of EVs taking over in the US anyway. They are a tiny niche here. The rest of the world is quickly moving on but we’re going backwards and doubling down on fossil fuels. Many car companies are giving up on their US EVs and the federal government is actively trying to sabotage efforts.

8

One difference; the location of data centers can be chosen based on where there is sufficient excess power. That’s unlike domestic electric car recharging, which is going on everywhere, though much less concentrated.

9

Theoretically could be. Reality not. Data centers are rapidly building out their own dedicated power plants, mostly natural gas.

EVs OTOH are generally charged at times that are less than peak …

10

Not even close in the U.S., according to Edmunds What Is the Percentage of Electric Cars in the U.S.?, and right wing governmental pushback is getting stronger.

11

And even if a majority of new cars were EV it would be a good long time until the majority of cars were. Albeit with some local lumpiness.

12

Something that’s not discussed often enough, IMHO, is this “wires vs roads” contrast in infrastructure.

To add a new gas station anywhere, all you really need is a road. The fuel gets delivered by tanker trucks and gets stored in underground tanks for the pumps to use.

With electricity, you can’t really do that (in a economically practical way, at least, like moving giant batteries around would be way too impractical and expensive). So you have to drag physical wiring, sometimes really thick and heavy physical wiring, for dozens or maybe even hundreds of miles, to be able to support the draw of DC fast charging. You can’t just say “I want a new fast charging station here” and be done with it. The entire local electrical infrastructure up to the nearest transformer substation (and sometimes even past that) might need to be upgraded.

By analogy, look at our fiber internet rollout. 20-30 years underway now, with various federal and state subsidies over the years, and still the overwhelming majority of US households do not have access to fiber internet. There is dark fiber everywhere, and a lot of regional telecom hubs have migrated over to it, but the “last mile” problem is still with us.

Same with EV charging infra. It’s easy enough to put up a few fast charging stations alongside major highways (like California has done), because then you’re running the wiring mostly along a single corridor. There the costs could be recouped in a matter of months or years from the sheer amount of traffic going through them.

It’s much harder to do a hub-and-spoke distribution system into the heartland rural areas that might see only a dozen or so users a day (if even that); the massive investment it would need to upgrade all of those areas to support DC fast charging would likely never pay themselves back.

Onsite generation isn’t really fast enough (solar panels don’t produce THAT much power; you’d need a hundred 500W panels to support a SINGLE 50 kW fast charger, and that’s on the slow end (they’re up to like 300+ kW now). And even if you had giant onsite batteries, well… each battery is basically the size of an EV, and would take just as long to charge (12 hours or so without DC fast charging), so even if you slowly charged a bank of them up through the week, you’d only be able to discharge a few of them before running out.

Fossil fuels are incredibly energy-dense and easy to transport in liquid form. Electricity isn’t as easy to move around. In a big country like the US, there will probably always be that urban-rural divide. As far as I know, other geographically big countries deal with that not by magically adding EV chargers everywhere but by having better alternative transportation (like higher-speed passenger rail networks).

To my dismay, the civic discourse around this topic rarely ever touches upon the nitty-gritty. The right puts out FUD, the left goes “just do it already, it’s not that hard, climate change can’t wait”, but rarely does a bureaucrat on any side actually sit down and talk about the difficult logistics in an engineering sense. California was leading the way with their state incentives and buildouts, but even that (sensibly) focused on major corridors, intersections, and hubs first, with the rural areas lagging far behind. Most states are not California (in terms of either values or the sheer amount of money they can throw around), and the federal attempts at this were half-hearted at best and usually paused every time the administration turns red and dismantled whenever it stays red.

It’s not something we can just handwave away either, lest the chicken-and-egg situation (between chargers and EV adoption) remain an issue forever.

It’s the sort of problem that IMHO only a sustained and coordinated federal effort could really tackle — and I mean on the scale of the Manhattan Project, or at least major dams & Interstates. The far left kept trying to push such projects through in Green New Deal-type proposals, only to be gutted by the mainstream Democrats and then altogether abandoned by the Republicans who end up inheriting them. As a country, we’re just not able to work together to build decades-long infra projects anymore.

13

On the other hand, once the power lines have been upgraded, delivery of the additional power can be accomplished easily. But a gas station requires regular delivery of tanker trucks.

14

I suspect interest in EVs will accelerate in the US as long as gas prices are high. Chart from Paul Krugman: yes, some countries are more forward-looking than others.

15

I don’t think upgrading the power lines are a one-and-done deal, though (unless we massively overprovision them to start with). It’s one thing to be able to support a couple of fast chargers, another thing if we were actually facing EV critical mass (which, to repeat, we are nowhere close to) where the majority of “pumps” at the majority of gas stations needed to be replaced with EV chargers.

Like any infra, it’d have to be maintained and upgraded over time to keep up with changing usage needs. We’re notoriously bad at being able to do that with any infra, even for things like bridges and roads, much less a values-driven thing that only a small part of the population actually wants and that more than half the country actively demonizes in their heads.

16

And that chart doesn’t capture much of the nuanced differences between those countries.

China can pull off a national infrastructure project every decade or so without any meaningful resistance; Great Leader just pushes a few buttons like a game of Civliization and it gets built. Need high-speed rail? Let’s go!! Renewables? Give us a decade and we’ll make it happen. New factories and ghost cities? Yes, sir! Meanwhile we’d be lucky to pull something like that off once every half-century or so… and honestly, probably never again? China is just beginning their technological leapfrog, while we’re reaching the end of our line.

The EU is much more population dense (in a hub and spoke sense) and also much more left-leaning compared to the US, with a population that cares more about climate change. And they have excellent existing rail infra, so cars aren’t the only way to go on long road trips. They also, importantly, have access to the Chinese EV market without all the roadblocks and tariffs that we put up, and many of their automakers are partnered with Chinese firms to codevelop their EV techs. They also have cross-border policymaking via a relatively functional meta-government in the form of the EU that the US federal government, with its massive corruption and destruction today, cannot really compete with any more.

That chart, IMHO, more depicts policy-side incentivization than organic consumer demand (that uptick in the early 2020s can be overlapped with the incentives that managed to survive the Green New Deal compromises). EV sales have to be spurred on in an environment supportive of them or they will die out. We’re not yet at a point where they can be a stable, self-supporting system. We’re already seeing a sales decline this year, but it’s not a fair comparison because there was a huge uptick in sales late 2025 due to the end of the federal subsidy; we’ll get better information by the end of this year into next.

Gas is cyclically expensive again, but that in and of itself won’t solve EV adoption any more than it has in the past. Besides, rising gas prices also means rising energy prices, which also means rising EV production costs and charging prices.