Updating California's grid for EVs may cost up to $20B | Hacker News Reader

Updating California's grid for EVs may cost up to $20B

arstechnica.com

18 pointsBender71 comments

That's a big scary number to your average California taxpayer.

What's frustrating about these articles is there's no talk of what's on the other side of the scale. What's the cost of doing nothing (and presumably stifling the rise of EV's so we can do nothing)?

Continued dependency on fossil fuels (and the associated oil infrastructure, health, and the you know, global existential stuff), and the occasional transmission line hook failing from severe wear, dropping the line, and burning down a good chunk of the state.

On the other hand, presumably if you're upgrading the transmission lines, you're going to replace those 80-year-old worn down hooks at the same time. I don't like using grid update subsidies to bail out corporate malfeasance on negligently-deferred maintenance, but pragmatically, how many Paradise's is 6-20B worth?

There are alternatives beyond "do nothing". For example, reducing vehicle-mile demand through better development patterns, or widespread adoption of the dramatically more efficient electric bicycle.

Simply spending $20B is like at least two orders of magnitude easier than any of your proposed alternatives in this country.

why don't we just build giant Arcologies and move everyone into them. easy. /s

A terrestrial version of Ark B??

Converting the state to EVs also costs 100x more than $20 billion because you are ignoring the trillions it will cost to replace the ICE fleet.

That's money you're spending anyways over a 20 year period. Approximately 0% of daily driven cars are > 20 years old.

If by 0% you meant 23%, then yes.

https://hedgescompany.com/blog/2024/02/average-age-of-cars-t....

And you only need to spend that 1-2 trillion dollars if you insist on perpetuating car dependence. You are not "spending it anyway". It is an opportunity cost that you could spend on a better world instead.

I said "daily drivers". A vintage car parked in a garage and used twice a year isn't relevant.

A 2002 Corolla is not a vintage car. The amount of going outside and looking around you are admitting to avoiding is pretty remarkable.

Nobody is putting 50,000 miles a year on a 2002 Corolla.

Nobody is putting that many miles on any car.

https://www.fhwa.dot.gov/ohim/onh00/bar8.htm

All for orange-pilling the water supply, but grid modernization isn't just for EVs.

Lots of other benefits as well like replacing fossil fuel heating with more heat pump penetration, and making things handle the variable energy flows from putting more storage on the grid, which to be frank, is needed to continue adding solar at CA's stage of solar development.

A grid designed for the centralized coal/gas plant era doesn't work as well in the era of distributed resources. EV's are just sexiest way this shows up.

$20B is about 10,000 homes at typical CA prices, including infrastructure, or less than one small-sized city. It's significantly cheaper to spend that to upgrade the grid for EVs than to re-do how our cities are laid out. California Forever, the real estate developer that's planning a new city in Solano County, has already dropped $900M on the land plus another $700M or so on incentives for the county to greenlight them, and they're still facing local political opposition and haven't even broken ground.

Electric bikes are good options for single commuters in their 20s, but have a lot of practical problems as primary transportation for families. I can't really pick up my kids from daycare on an e-Bike, or schlep a car full of groceries from Costco.

My day care has basically devolved into a contest to see who can carry the most kids on an e-bike. Your town is just behind the curve.

As for the costs of development, you have made a common error. 20 billion in public infrastructure can't be compared with 20 billion in private construction activity. 99%+ of construction is privately funded. In my county, which is only the 7th-largest in California, private parties undertook $28 billion in construction in 2023, and this was considered a very slow year for construction.

Is it that scary? California has 40 million people, so it's about $500 per person. And the electric car mandate is 10 years out (2035), which means it's about $50 per year per person, and the median household income in CA is about $90K.

It shouldn't be that scary of a number.

Former Californian; sky high taxes including 10%+ sales taxes and even just paving asphalt properly is a fool's errand. No, we don't speak about high speed rail.

So yes, it's a scary number. Scary in the sense that the abyss of fiscal laundering and waste is terrifying.

Yeah it's terrible optics. Actually $20B over a decade is reasonably cheap for such a massive energy usage transition. (Consider the estimated $88 billion and $128 billion for the SFO <--> LA high speed rail.) And as a costal state, California will feel some the most intense financial impact from climate change. I wish it was easier to put a number on that cost.

That seems like a really small number to me. Grid updates by 2045 may cost $20B, so for $2-4/mo/customer, California's grid can be upgraded for EVs.

Bad headline.

Not only are the costs spread over decades, the extra sales of electricity (replacing fuel sales) cover the cost of the upgrade, while lowering the per unit cost of electricity.

They mention this in the article so there's no real excuse other than clickbaiting.

In addition to that, spreading the costs evenly across the 40 million residents of California, it’s a whopping $500 per person. Not exactly cheap, but also not a political nightmare to find the funding.

Keep in mind, many areas of California already have absurd electrical pricing and rising ($.50+/kwh) and everyone was promised a long time already that there would not need to be grid upgrades to support EV’s.

Which was transparent BS.

So here we are again, likely going to get more out of proportion increases due to things that were promised to not be needed.

There was a short time where people were making an argument (and it wasn't entirely impossible, either) that on-home solar would "balance out" EV charging so that the overall increased load on the grid would be minimal.

That didn't pan out for various reasons.

I remember many of these pronouncements recently being during rolling brownout/blackouts due to high loads in summer AND after PG&E had stopped providing decent pricing for infeed from home solar systems.

I agree that some of the discussion was before the BS was apparent, but a lot of it was well after it was obviously not how this was going to work.

Still possible (if in a setup with good insolation!) to provision your own home solar setup and charge off that - and high income folks, it would likely pay off pretty quickly due to the current insane rates - but then that’s why the utility companies seem dead set on trying to stop it too.

What I've seen is that where it used to be quite worth connecting to the grid and "feeding back" because you'd get paid the current retail price, now it is complicated enough and paid little enough that many people just don't bother feeding back at all.

Yeah, it was part of the short term incentives to get people to roll out solar.

Now they pay wholesale prices when buying it back (often zero or even negative during peak insolation) and charge retail when selling it - any by retail I mean ‘crazy high usage based numbers which make no sense’.

I know a bunch of folks which installed solar themselves to cover their AC costs because the utility still charges crazy high rates during peak insolation - even while they are having to load shed - and their electric bill dropped to almost nothing afterwards. They didn’t even need batteries.

For starters - many of the utilities charge more for daytime and "peak" usage hours in the TOU plans.

Which is also when they're overproducing solar and having to shed load.

But hey, no sense looking for logic here.

Plenty of logic if you assume their goal is to be paid as much $$ as possible, while throwing up the maximum smokescreen possible.

Near as I can tell, the CPUC is in on it for many reasons. Some, like PG&E being paying the tab on their numerous past misdeeds without formally declaring bankruptcy/putting the state on the hook.

But also if for no other reason than it's the only way the various green mandates can actually be hit, since costs to actually deliver a kwh when needed are far, far higher and require much larger capital outlays than the public would be willing to accept if it was obvious to them.

It's frog boiling/gaslighting the public - 'oh it won't cost anything extra - it'll probably save us money!', 'oh, we might need to add some restrictions', 'oh it won't cost that much more really, it will be worth it', 'oh, you've already paid for most of it, it will just be a bit extra', 'oh, yeah it is a bit expensive but what are you going to do about it', etc.

For some reason it seems to work REALLY well in California.

"for various reasons"... they killed home solar with net metering charges!

The PV is about ~$300 per person (when spread out over decades).

You're fighting a losing battle living in California if you're going to complain about spending ~$0.03 per day to have electric cars...

That’s the claim, right now. Originally it was $0.

I left California awhile ago.

My current (peak) electrical rates are less than $.15/kwh, and i’m not far from the border though.

Who ever claimed $0? I get why you left California.

Up here in Seattle we pay 10-13 cents? But I probably wouldn’t notice much if we paid 50 cents.

Pre ~ 2023 the party line was 'we don't need to do anything in generation/distribution infrastructure, we just need to roll out more chargers'. With hints or even promises of REDUCED electric rates, not increased rates.

Search seems to be prioritizing more recent articles, so harder to find references, but there are plenty out there. I remember reading dozens of articles at the time which all insisted that costs would stay the same or go down, and we didn't need to make more generation capacity because EV use would never be a meaningful portion of electrical demand.

"In addition, by including new rate structures in the pilot programs, California may be able to shift load in ways that defers expensive new generation or transmissions assets. This remains to be seen, but could result in lower rates overall." [https://www.csis.org/analysis/utility-involvement-electric-v...]

Here is the California Legislative bill passed on the topic in 2022 - and the focus is on building more chargers. [https://leginfo.legislature.ca.gov/faces/billNavClient.xhtml...]

Here is also the mandate to increase more charging infrastructure too, and assess effectiveness. [https://www.energy.ca.gov/data-reports/reports/electric-vehi...]

By 2022, they were trying to get folks to switch to grid controlled charging [https://www.canarymedia.com/articles/ev-charging/new-califor...] which is a bit crazy for the end consumer, and could even be a large scale safety issue. Essentially centralized rationing.

In 2023 is when supply/demand issues got so obvious that people started to call BS [https://calmatters.org/environment/2023/01/california-electr...].

Notably, around then is also when the Diablo Canyon nuclear plant decommissioning got canned (despite it being past it's design life, with known and expensive to remediate radiation leaks).

Seattle doesn't need AC, but a lot of housing does have electric baseboard heating. Those folks would definitely notice if electricity ~quadrupled in cost!

This very article is claiming that the per unit electricity prices will decrease because EV charging will gravitate to "fill in" low demand regions, spreading fixed costs across more units.

You appear to be using this new research as proof of the opposite, and calling out previous times the exact same claim was made.

Ah, of course - $/kwh will drop, despite spending an additional $20-50bln we previously didn't expect on generation. /s

Did costs drop for AC, despite load being during peak insolation? Or unit kwh costs overall? Because it just keeps going up, faster than inflation.

The 'we'll only charge during high supply/low demand' times is fantasy, near as I can tell.

> we don't need to do anything in generation/distribution infrastructure, we just need to roll out more chargers

I just never heard that argument, if it was common, or was it a Californian thing?

Grid controlled charging is common in Europe where electricity rates are higher (well, outside of Norway who has plenty of cheap hydro). Americans have always seemed to take their car freedoms for granted (cheap gas! free parking! cheap electricity!). Having lived in Europe for a couple of years and then in Asia, it boggles my mind how many free things we get over here.

Diablo Canyon has a warm water problem (that kills fish when the warm water is put back into the ocean), not a radiation problem.

> Seattle doesn't need AC, but a lot of housing does have electric baseboard heating. Those folks would definitely notice if electricity ~quadrupled in cost!

We are basically Norway: we have lots of hydro and a relatively mild climate. Heat pumps are pretty efficient.

This discussion was specific to California, with some side discussions.

Diablo Canyon has been leaking tritium. [https://www.nrc.gov/docs/ML1116/ML11166A174.pdf]

And numerous fault lines were discovered in and around the plant after construction that lead to serious concerns of structural failures and radiation leaks in the event of earthquakes. The most recent was found in 2008.

Water temp hasn’t been the issue, and I have no idea where you got that from.

https://www.sanluisobispo.com/news/local/environment/article...

https://www.hcn.org/articles/nuclear-energy-the-divide-over-...

Seems less paywalled.

> Each day, Diablo Canyon sucks up about 2.5 billion gallons of ocean water to generate steam and cool the reactors, before spitting it — 20 degrees warmer — back into the Pacific. The water intake kills thousands of fish each year and sometimes other sea animals. The warm water discharges are also changing the marine ecosystem in Diablo Cove, driving out the bull kelp and having a detrimental effect on abalone.

None of that impacted permitted or regulatory oversight for Diablo Canyon that I saw.

Regulators explicitly mentioned the issues I raised when discussing shutting it down, and even the environmentalists were saying their concerns were due to radiation/meltdown.

Also, the state wanted to only run the plant at half power in favor of renewables, because reasons, making it lose money like crazy.

Here is a contemporary article from when the shutdown was approved originally.

[https://www.sfgate.com/business/article/End-of-an-atomic-era...]

I’m sure some folks are worried about the hot water discharge - but that has been going on for forever now, and the situation isn’t getting any worse. Someone will always find some reason to complain.

I live in California and only pay 0.10-0.12 a kWh.

What area are you in?

Average rate when I was in the Bay Area was > $.55/kwh.

While this may very well be true, the vast majority of California is serviced by 3 main providers and their rates are nowhere near that price.

It's also more of a range.

> Li and Jenn came up with a range of anywhere between $6 billion and $20 billion

That's the latest trick in news reporting these days. Sometimes they report the total cost, sometimes the yearly cost, sometimes over ten years. I guess that's part of the effort to make numbers have an emotional effect while making them hard to compare. School buses, football fields, width of a hair, elephants and so on, 2/3rds vs 66%. Most people have no idea how to compare them.

Although the headline isn't great, the article does do what I've been taught to do by the BBC's More-or-Less programme, and ask 'is this a big number'.

It's very easy to not comprehend big figures, so there are a few techniques to make them relatable - bring it down to the cost of an individual and/or per year, and to compare to other costs.

Here they have chosen to compare it to the existing cost of the grid, and the existing maintenance and operation costs, but which seems like a good comparison (if you haven't read the article, it's about one third of the cost of the existing grid, or about three times the annual operation/maintenance costs)

An alternative would be to say it's a cost of about 500 USD per Californian, or about 50 USD/year which isn't nothing, but doesn't feel like an outrageous figure.

That’s not that bad when you consider the cost of oil and gas infrastructure

Current lack-of-supply causing increased cost and delays in just about every single major grid interconnect/hookup/switching component would indicate that the present day cost of the upgrade is not likely to be fully correct. I'm a little too lazy to grab the link, but for context listen to an episode of Odd Lots from Bloomberg from a week or two ago about issues facing property developers all over the sunbelt getting switch gears and transformers for new commercial multi-tenant buildings. The combination of EV charging stations and AI data centers is making grid upgrades more than double their previous decade growth trends, which is throwing power companies and electrical infrastructure component manufacturers for a loop and they're not bringing supply online nearly fast enough, particularly in light of complex grid rollouts like all the renewable that CA is requiring by law. Combine that with a general lack-of-supply of the skilled labor necessary (linesmen and electricians with high voltage/commercial training and experience) and you won't see it happen on time or in budget. Also, for comparison, grid upgrades require a certain amount of new infrastructure installation in places without current right-of-way, which has gone particularly poorly for the high speed rail out there, so I would be willing to take a bet that this follows a similar growth curve of at least tripling the high end estimate.

Details:

> "Paying for the necessary upgrades will be pricey, but there's a lot of uncertainty here. Li and Jenn came up with a range of anywhere between $6 billion and $20 billion. They put this in context in two ways. The total capital invested in the existing grid is estimated to be $51 billion, so the cost of updating it could be well over a third of its total value. At the same time, the costs will be spread out over decades and only total up to (at most) three times the grid's annual operation and maintenance costs. So in any one year, the costs shouldn't be crippling."

There's no doubt that switching from a primarily fossil fuel-based transportation system to an electricity-based one is going to be expensive, but good transportation systems have positive knock-on effects on economic growth, and the side effects - cleaner air, independence from fossil fuel price swings, etc - are more than worth the investment.

In contrast, giving $100 billion away to finance endless war in Europe and the Middle East - after dumping well over $1 trillion on Iraq and Afghanistan (pre-debt-interest) - is just a recipe for long-term economic decline. This is why China is an economic success story - they built a national high-speed rail network, the world's best, while the US was playing at imperial neocolonialism.

Actual paper (edit, nope previous paper on similar topic by same authors): https://evs36.com/wp-content/uploads/finalpapers/FinalPaper_...

edit 2: the actual paper

https://www.pnas.org/doi/10.1073/pnas.2317599121

Believe it or not, there are a lot of opportunity to solve a lot of these challenges with software.

https://smartcar.com/blog/electric-vehicles-power-grid

https://smartcar.com/blog/what-increasing-power-outages-tell...

Disclosure: I work for Smartcar and my company has a small role in the greater efforts solving these challenges with APIs.

And the cost of updating California's grid to support the increased use of air conditioning is also $20B.

There's no comparison of how much grid improvements would cost without EVs so I don't know if $20B is alot. California has 5th highest GDP in the world at $3.9T so .5% if all that money was spent in one year but actually over a decade it's .05%.

Is this actually a cost?

At some time in the future oil prices will eventually tend towards infinity, at some point before then it'll be 200, 500, 1000 a barrel.

At that point the grid upgrade and EV transition will be a huge cost saving.

It's just investing for later, surely?

Same thing in Europe. Governments are getting involved providing bridge financing to grid operators so the upgrades are done on time - or at least, less late. Bit of a failure of planning though

$20B? That’s it?

13 million households in CA. $1500/household. $9/month electric utility surcharge for 30 years at 6%. Hardly noticeable.

Sounds big by itself. However, this is spread over time, and the *annual* budget for transportation in California is $33B

If you think that is a lot, you're never going to believe how much 30 million EVs would cost.

Cool - just roll that into the next monthly 600% PG&E rate increase.

Twenty billion dollars is 20,000 clean hydrogen refueling stations.

Mhm then how much to transport fuel to all those stations?

can still do that after electric doesn't pan out

Hyrdogen cars are DOA.

Too late.

https://electrek.co/2024/04/19/dc-fast-charging-station-for-... ("The US now has 1 DC fast charging station for every 15 gas stations")

https://afdc.energy.gov/fuels/electricity-locations

https://supercharge.info/map

https://web.archive.org/web/20240425142724/https://www.white...

I feel like this thought pattern just assumes all electrical generation will be done the same way as always, from a central source. Two big things are already happening to change this paradigm though.

1) Batteries are allowing us to transmit energy asynchronously

2) Distributed generation with solar

“The grid” needs updates for sure, but it doesn’t need 10x capacity increase. That’s just fud.

What does it mean to "transmit energy asynchronously"?

Transmit the energy when it's cheap, instead of at the moment it's going to be used.

async is a strange choice but perhaps the closest equivalent from a software vernacular. time-shifted energy transmission/generation.

Batteries managed by individual users will never replace big infra. They are simply not resilient enough.

This kind of logic is akin to saying mesh networks are all we need to deliver reliable internet to everyone. No, CTTH, FTTH or satellite is what we need to deliver reliable internet to everyone, and those require big infra.

Furthermore, to a point in the article that folks on Ars are spouting, if everyone is charging during off-peak hours, then those aren't off-peak any more. They're now on-peak hours.

Grids are required to deliver steady reliable power even under peak usage. 10x is likely an underestimation of how much extra capacity we will need if everyone switches to EVs.

Under the current rules, yeah, but the rules and how we think about this need to change too. For example in my state the process for selling energy back to the grid from a residential account is extremely limited and had a capped number of users.

Batteries aren’t resilient enough today but they have grown leaps and bounds over the last few years. New chemistries and ways of controlling them are making all kind of new possibilities.

As much as I don’t like some of the rules in CA the requirement for solar on new builds is brilliant. The hardware has become so cheap and rolling the labor and design into the initial construction means a short pay back period.

Selling energy back isn't as efficient as it sounds. Transmission losses are a major factor for delivering power. In a star topology, any power you generate must go back to a local base station before being delivered to your neighbour, at minimum doubling the transit distance.

So, of course, the next step is (to my point): Let's rely on a mesh network to deliver power locally and directly. Sure, until one of your batteries has an issue or its a rainy day and the batteries are depleted. Suddenly your neighbour needs to source power from somewhere else. That somewhere else could be other neighbours, who are, due to local weather conditions, experiencing the same issues. Okay, let's go back to the grid. Now this weather is affecting multiple locations at the same time, and the grid itself is under-provisioned since it was reliant on these local sources of power. If the grid has sufficient capacity to deal with this failure mode, then it has sufficient capacity to be the regular, default power source.

Power generation and delivery are hard problems that require big thinking. Wishing it were simple (oh just do this!) doesn't make it so.

PG&E CEO pay was $9.3 million in 2018, $18.5 million in 2019, $6.2 million in 2020, $51.2 million in 2021, $14.1 million in 2022 and $17 million in 2023.

Just sayin.

Reply on news.ycombinator.com