The EV Transition Explained: Can the Grid Cope?
spectrum.ieee.org
spectrum.ieee.org
> For instance, Marshall stated, it is not unusual for a 37.5 kilovolt-ampere transformer to support 15 households, as the distribution system was originally designed for each household to draw 2 kilowatts of power.
It wasn't that long ago that 10 light bulbs would consume about half of that. So, lets assume 60 watt bulbs with 4 in the living room, 4 in the kitchen, and 4 in a bedroom all on. Now someone turns on a couple of floodlights outside. We are at ~1KW with no TVs, computers or air conditioners running.
At that point, do we just hope noone turns on a hairdryer? :) Someone has messed up something fundamental in this article.
Remember the early days of the internet when "experts" predicted it'd collapse as the user base grew? Much of this article reads like those.
The article gets this right and it's a completely normal part of utility design and distribution, which is unsurprising because the person quoted is a PE-licensed electrical engineer with the title "Director of Utilities".
Someone picked the wrong numbers. Even older homes tend to have 100A panels, in newer homes 150A and even 200A are more common.
It sounds like he's presenting a real edge case but not making it clear that its an edge case.
But to me the kicker is that he went into no further detail. Transformer setups like that are done with knowledge that overloading can happen. If your defending him, then you'd surely know all about how much and for how long, right?
Instead, he just said 2kw per house, which is literally absurd.
A clear saving grace of EVs is that there is a lot of hours during the evening to charge a car (and most aren’t going from zero). Coordination on when to start and trickle charging across the whole evening seems like a much cheaper way to stretch the grid’s capacity than mass upgrades.
Coordination is an in interesting aspect. The nice thing is that these cars tend to have internet connections and the ability to remotely control charging. In my area, we are already seeing utilities experiment with offering incentives to allow remote control to improve load management.
The technical solutions in many cases no longer require additional hardware.
1) can handle peak air conditioning load AND 2) has time of use pricing to encourage EV's to charge at times other than peak air conditioning load.
Can handle EV charging.
Palo Alto satisfies neither of those two conditions. That's the problem, not the EV's.
That's why only few weeks ago we have seen an initial standard for biderectional DC support. That's why there is still NO PV inverter who actually output 400V DC for stationary batteries who happen to be the same in EVs avoiding wasting 25-30% of energy in useless multiple transformation.
That's the issue of private led initiatives: public research propose visionary solution, implement them, give them to the public with some standard concepts. The private sector then pick and implement.
Like we do now, anyone try to offload research costs to customers, push their own standards and nothing really work.
And that's a VERY good example of the fact EVs designers have no clue on what they are doing: modern domestic storage is 400V like most EVs batteries, there is no damn reason to avoid a direct MPPT (DC) link between PV inverters and cars nor no damn reason to mandate charging at constant amperage. It's simply the lack of PUBLIC research who also produce PUBLIC standards...
Imagine if in the 20th century we stopped at "Automobiles are doomed. Who's going to build all these highways and gas stations!?" or "The lightbulb is a nice idea, but we have no power plants, an they're super expensive!"
No, we frequently build things.
Imagine if in the 20th century we stopped at "Automobiles are doomed. Who's going to build all these highways and gas stations!?" or "The lightbulb is a nice idea, but we have no power plants, an they're super expensive!"
This happened literally all the time.
1. That ↑, but also…
2. … the spread of lightbulbs or automobiles happened organically, simply because people perceived them as better than what was there beforehand and wanted them. There was no big centralised push like "We must switch from horse carts to 100 % automobiles within two years in order to avoid the big horse dung apocalypse", so the spread of those new technologies simply happened at whatever speed was feasible and sensible. If it turned out that building the required new infrastructure for lightbulbs or automobiles was a little harder than initially estimated, the buildout simply happened a little slower and that's that.
Whereas now there are external goals attached to the process (must have X % electric cars by 20xx in order to avoid the apocalypse), so if it turns out that reaching that goal is harder then expected, there's immediately much hand-wringing.
We would need 43,000 GWh/y of new electricity generation for CA alone (15m cars * .24 KWh/m * 12,000 annual miles). For reference CA has a yearly electricity generation capacity of just under 200k GWh/y, and imports another 100k GWh/y.
So it's not just rapid charging load concerns, but overall generation capacity which will need to be addressed.
ref. https://www.energy.ca.gov/data-reports/energy-almanac/califo...
I don't think the "change all of our infrastructure" option is even realistic, to be honest. We can barely maintain our existing infrastructure in too many cases, let alone rebuild it all.
EDIT: Corrected units from kWh/y to GWh/y.
I get about 3.5mi/kWh in my EV. If we cut the average yearly mileage about in half to ~6,000mi, that would be ~1,700kWh at my driving efficiency, plus a bit for charging losses. Even at 4.5mi/kWh it is still ~1,333kWh.
43,000 GWh/y for all CA EVs at the average miles per year per vehicle.
Total CA generation is 200,000 GWh/y.
It’s an interesting problem to me, once you fold in people’s behaviors.
I think this makes sense, generally speaking, but the problem is that this isn't a "market" in the overall sense, only in choosing between vehicles. Gas and oil infrastructure for example has a lot of government involvement, so will charging infrastructure. So you can't really let the market decide here because the competition isn't really there.
Do you mean the total population and the total land area? Doing the division that's an absurdly low number, but probably an irrelevant one. Most trips don't go that far.
The real problematic "low density" is suburban neighbourhoods.
Suburbanites will certainly oppose this as well, but eventually economic physics will force them to change their habits regardless of what they want.
> We can barely maintain our existing infrastructure in too many cases, let alone rebuild it all.
Right, which is why doubling down on cars is bananas. We already can't afford to maintain all of these roads, let alone build new ones or expand. Fortunately, there is a lot of low-hanging fruit. Repurposing existing highways to initially allow for bus rapid transit alongside personal automobile transit, for example, can help. We probably need a nationwide moratorium on new road construction for 20-50 years to really get things back in a good state.
and water treatment, and sewage system
I can't help linking a NJB video when this topic comes up.
https://www.youtube.com/watch?v=7Nw6qyyrTeI
"the replacement cost of all infrastructure in Lafayette, something that you would expect to happen over about one generation, was about $32 billion. ... Yet the existing tax base was much less than that."
---
I don't think it can change though, given that majority of the US lives in the suburbs. They can vote to stop you from changing their precious lifestyle.
For me again it's just a question of whether we will manage the change or whether it'll just be a big disaster.
Switching to solar AND switching to EV's is trickier, but since switching to solar adds electricity generation, it's part of the solution rather than part of the problem.
This means that the increased generation requirements would still be a concern. Even if all it takes is for the existing power plants to switch to running all day around, it still means there will be an increase in non-power-plant materials needed: additional maintenance, fuel, personnel, storage, etc.
TANSTAAFL, in other words.
Electricity is far from free, especially in California. That should be enough to pay for the marginal costs.
But if I was being cynical, I'd point out that the prices paid for electricity are not even sufficient to maintain the electrical infrastructure in CA - certainly not enough to stop the equipment from regularly starting fires.
So something else (perhaps regulation on PG&E) is likely to be required.
I’m worried I missed a division and I’m off by a factor of 100x somewhere :)
California is aiming to get there by 2045, and 90% clean by 2035 and 95% by 2040 so in the next ten years it already needs to build enough renewable production to displace the 100k GWh/y that gas (and a smattering of other fossil sources) provides.
So, it's not really a big deal. It's only currently planning to ban ICE sales in 2035 (though I think the EV transition will accelerate to the point that becomes moot).
Even assuming your other numbers are correct on total energy needed, your conclusion doesn’t follow since more imports can be found (or more import capacity built). The western interconnect is regional but your analysis assumes state-level constraints.
“Installing, wiring, and connecting the pieces of a traditional post-and-frame switchgear takes about 50 hours of on-site electrical labor,” Seitz said. “And that doesn’t take into consideration the time, costs, and difficulty of sourcing components from various manufacturers and supply houses.” Traditional switchgear is composed of electrical panels that are tasked with receiving, distributing, and protecting the site’s power. The components include a power disconnect panel, a current transformer (CT) cabinet, utility meter, distribution panels, and a transformer.
“Just measuring, cutting and assembling the typical post-and-frame structure for these panels takes two days,” Seitz said. “That’s before the electrician even begins handling the electrical components.” Next steps include:
Removing all dead fronts from tubs panels Mounting the panels, cabinets and enclosures Removing all the conduit knockouts Installing conduit fittings and conduit Mounting all breakers and internal equipment to bus bars Running wire through conduit Terminating all wires Making the external field wiring conduit connections
source: https://www.utilitydive.com/spons/ev-charger-deployments-mus...
So EV's shouldn't be a problem.
We're now a net importer of electricity, cuts are already planned for January.
Most people (myself included) that want to reduce their energy bill go with something else than electric.
The power companies should've planned the maintenance more prudently.
especially when they *(the transformer)* are toward the end of their lives, which many already are.*
Also, a fair bit of the infrastructure is maintained by the individual states, making any country-wide coordination of investment or fixes neigh impossible.
sounds like a capacity issue, not a maintenance issue. Who would invest in more capacity for which there was no demand?
This article has an issue with numbers here clearly. A heat pump isn’t really going to consume more electricity than the AC it replaces (and hopefully it is more efficient than the cool-only-AC it replaces as well so electricity uses decreases in the summer).
For electric cars, no-one needs 48A (11.5kw) charging (unless you bought a hummer EV and then you’re an idiot). The city can start denying permits for charging at more than 24-32A unless you have enough battery storage to cover it up.
Palo Alto also does not have TOU plans, and no demand charges.
While this wouldn't be a problem for most folks, there are definitely many who could easily outstrip the power provided by a Level 1 charger (in my area, construction workers who are driving around 100 miles per day to their worksite is an example).
Also, you need a minimum of level 2 charging to support long distance travel by road, where a full battery can be consumed in about 3 hours of highway driving.
> or electric cars, no-one needs 48A (11.5kw) charging
refers to L2 DC fast charging. The next level down (which is implied by "nobody needs L2") is L1 charging, which is done on a household 120V or 220V circuit.
In other words, if you don't supply L2 (and L3) DC fast charging for customers who need it, EVs will never replace ICE.
> The city can start denying permits for charging at more than 24-32A unless you have enough battery storage to cover it up.
https://www.jdpower.com/cars/shopping-guides/what-is-dc-fast...
And 11.5kW is in the mid-L2 range per that. It could be out of date, for sure.
L1 = single phase 120V power. People typically think of common household outlets, which will be limited to 12A or 16A, but it could also be a TT-30 at 24A.
L2 = Split phase 240V. Typical amperages are 32 and up, but lower is possible with some cars and higher is common. Eg, the F-150 Lightning can charge at 80A.
L3 (very uncommon) = 3 Phase AC at up to 400A - Could be >300KW. I've never seen this.
Most home chargers will likely be 240V at ~24 Amps, that's about 6 kW and plenty for the vast majority of users. People who regularly need more can easily install more.
DC Fast Charging, (L3) is done at 400-800V and really only needed for road trips so few places outside of highways will likely have it
I live in a weird house that has 3 gas furnaces but only one of of the furnaces has AC attached at the moment. I'm going to swap out two of the three for ducted heat pumps, so my peak electrical load in both seasons will increase by roughly 5kw. Replacing the third furnace (with the AC) would surely reduce the cooling electrical load since it's old and outdated, but would add another few kw to my winter heating load if I were to do so.
You're wrong about this. We frequently take day trips that are about an hour each way from home, and without the L2 we would get dangerously low after another day of commuter driving. As a co-poster noted, standard chargers will take days to recharge. Ours does about 6 miles of charge / hour, whereas the L2 does about 30 miles of charge / hour.
Naively you'd think that cars with a larger battery would be more likely to need a level 2 charger since it takes so much longer to charge a large battery on L1 than it does a small battery. But in fact it's often the opposite -- a big battery means you only need enough charger to handle an average day rather than a peak day.
Maybe charging after your daily driving only charges at 3 kW, but charging after a road trip does the full 11.5 kW.
But yeah, I've certainly come home from a road trip with 10% battery, plugged into my 11 kW home charger, and had it say it would take 5 hours to recharge to my 70% limit. If I was only charging at 2 kW, an overnight charge wouldn't get me very far the next day.
I don't know about you, but I mostly do long trips on the weekend. As long as I have enough to get to work Monday morning and am full by the weekend, that's almost always good enough for me.
It's definitely something Tesla has solved for to achieve their cross nation road trip planning with 0 home charging.
The answer isn’t to ban high AMP L2 chargers but instruct cars to only use high amp when lower amp charging isn’t sufficient.
For most users who do home charging, they leave them plugged in for 10+ hours at night and drive nowhere near their battery’s capacity, so trickle charge them each night, ensuring car is ready at their scheduled departure time.
People are no longer spending money on gas, it’ll be going to the electric companies. They’ll have more revenue for investment in capacity.
BEVs was designed with a specific target, even if no one formally state that in such crude terms: to keep a grid with significant amount of intermittent sources (like p.v. and eolic) we need something fast enough to compensate their peaks and drops, witch happen to be batteries. They are too expensive respect of their limited service life so better offload such costs to private Citizens to keep company profit. Private cars are actually batteries on wheel, cars are already not that cheap stuff, people do not use them for more than 8-10 years anyway, due to limited range they got plugged in frequently like when you stop traveling anywhere.
The above paragraph was initial theory. But while people have bought a bit p.v. because it's already interesting in vast areas of the world far less interest happen for EVs and since many people even if casually do travel for longer range they need quick charging witch means a hyper big peak loads for the grid.
EVs adoption was forced with laws (emission limits, travel limits in cities etc) and got pushed by soon to happen frequent blackouts, for some reasons people decide to buy them en masse recently but there is still NO SOLUTION for the grid: actual p.v./eolic volume is grown, but far less than the skyrocketed energy demand, quick charge need to be more widespread, fast (3-phase 7.3kW / 22kW) chargers need to be FAR more widespread etc so grid operators, aka private scammers are still in trouble...
Although critical to safety, much of the clean energy transition is in the hands of utilities to move quicker. With a 50+ year track record of not doing so I'm curious what government and/or private sector can do to remove the utilities involvement or red tape.
Level 2 chargers have been easier to install but the DC Fast Chargers backed by big IRA bucks will be a tough slog. Utility transmission line drops (12 months), battery storage, installation/engineering (3 months) trenching, concrete pouring, permitting (3-6 months), union labor, DOT involvement (6 months). The day is soon approaching where we have far more EVs than charge stations ready to install.
Meanwhile, as we adopt electric cars... and practically every manufacturer is moving this direction and is far behind or is there already. Tesla, Ford, GM, VW, Hyundai/Kia all have fantastic electric cars.
But its not just this. Another major consumer of power is greenhouses; also coming online and expanding now. They tend to be going very high power HPS. These lamps are so bright not even sunglasses are good enough. Your skin burns far more than any beach has ever done.
So we have several new high energy consumers. We dont have any new nuclear plants built. We are shutting off coal plants. The grid cant cope because it's bipolar and dumb.
Why is nobody building gridscale batteries? Why does nobody have huge hydro pumped setups?
our grids will start to grow batteries at pinch points, already are, but it will be an exceptions/corner-cases thing, the economics don't work for the broad-average.
https://en.wikipedia.org/wiki/Grid_energy_storage
NREL FAQ https://www.nrel.gov/docs/fy19osti/74426.pdf
Tesla's approach https://www.tesla.com/megapack
Form Energy, one of the best funded startups working on novel grid storage batteries https://formenergy.com/technology/battery-technology/
> Vehicle-to-grid (V2G), also known as Vehicle-to-home (V2H) or Vehicle-to-load (V2L) describes a system in which plug-in electric vehicles (PEV) sell demand response services to the grid. Demand services are either delivering electricity or by reducing their charging rate. Demand services reduce pressure on the grid, which might otherwise experience disruption from load variations.
for example, it would be technically trivial to implement a kind of "lock acquisition" handshake where all the meters downstream of a bottleneck (transformer in this case) request their incremental kW and the utility does basic fair- queuing. even if only new meters and cars did it, it would still help immensely.
EVs already let you specify when you want to leave in morning, so it’s a small extension to that.
First, smaller cars, smaller batteries. Still can have ICE vehicles for range. Smaller cars help density, 2 Wallaby EV cars are designed to fit in one US garage space (other models may be larger).
Second, the UTV class will employ upgradeable solar panels for replacement as technology mandates - no OTA stuff, just a few hex screws probably and some plugs. Like modifying an RC car.
Third, we can't get away from energy density problems at scale with EV even trying to supplement with solar, and I'm thinking about making EV UTVs that are off-grid self-supporting as long as the generator system (sold separately, also useful in other applications) is brought along.
Fourth, the Wallaby EV system is not aiming to penetrate high-dollar households with the discretionary income to make changes of this nature - the transition clients need to save money on gas, car insurance, and then they can afford electricity and heat and maybe some school supplies. This is the real world and these folks need help and are #1 to benefit from getting away from petroleum. Cost fluctuations, being in a ghetto where the gas is a dollar more than 2 zip codes away, food desert issues, urban density...all these are the natural habitat of the Wallaby EV.
Fifth, the grid in Texas recently had trouble coping with a giant storm which fortunately my hospital grid location saved me from freeze issues, and it's still an issue. Until mentally incompetent grand-sitters get out of the way, it's not even worth talking about potential fixes. This has to be grassroots, not top down.
Tesla already tried top down. The build quality is embarrassing. Also on the top end a Focus ST will catch a Model S easily at anything starting over 90 mph. Smoked by a $15,000 used car. The Wallaby EV racer on the other hand will likely require a fighter pilot G-suit because of the active aero and tricks I have up my sleeve (all thanks to EV plug and play type concepts).
I am so excited about the future I actually spent real money for a change to setup the structure to house the pages and pages of business plans, roles, materials, and logistics to make it work. Great USA jobs for USA employees making something start-to-finish with USA materials. My ultimate middle finger to the legacy of Steve Jobs.