Power use is hitting record highs – what summer outage risks tell us about grids | Hacker News Reader

Power use is hitting record highs – what summer outage risks tell us about grids

smartcar.com

36 pointssmartcar_98 comments

I'd be surprised if power use weren't hitting record highs every single year. More people, more EVs, more more industrial automation, more data centers, more new construction with climate control, and on and on. Efficiency gains can't possibly be enough to offset overall growth.

The switch from incandescent to CFL and LED bulbs was a big jump in efficiency, and that one was pretty easy to implement at the consumer level. There should have been a modest reduction in power usage with the switch from CRTs to flat panel displays as well. However, that was low hanging fruit and there aren't too many other places where we are going to see that kind of jump.

Maybe some kind of powerful heat pump based kitchen oven? The heater coils they currently use are basically unchanged since the invention of appliances, but it's hard to imagine what a more efficient replacement would look like with foreseeable technology. Some sort of induction oven where you always have to stick your food in an insulated metal box before sticking it in there?

Maybe a hot water heater that uses a heat pump? Maybe it would have two tanks and you pump heat from one tank into the other, making your cold tap super cold at the same time?

There just aren't a lot of places left for your typical household to save a significant amount of power. I'd love it if industrial users put more thought into power savings but that requires buy in from rich people and that's a lot harder to do.

What percentage of electricity usage was incandescent light bulbs?

It's hard to say, but the changeover was noticeable:

https://www.nytimes.com/interactive/2019/03/08/climate/light...

When you have a technology in very nearly every home in the country and used daily that you can replace with something a whopping 85% more efficient it adds up.

Ground to air heat pumps could increase efficiency for air conditioning (which in hot and humid states is probably the largest line item for electric usage). Part of the problem here is the install cost means you likely want to share the ground source piping with 4 or more homes, but we don’t really do small scale self governance (build or share things amongst a dozen or fewer households) in the US anymore.

There are heat pump water heaters already. I think people usually have them setup so that the excess cold air is either used to keep the building cool in hot weather or it's routed outside in cold weather.

The next low-hanging fruit improvement is community solar installations, or barring that, individual home solar installations.

Technically, things like better insulation are low hanging fruit in the sense that the technology exists and the materials for it are cheap, but it's incredibly labor-intensive, especially when retrofitting old housing stock.

Solar is much easier to retrofit onto a home than making a 50+ year old house well-insulated.

Maybe steam convection ovens are more efficient? I haven’t looked at power consumption, but it might be less since steam requires a lower temperature to work with. But people use their ovens so little that it might not matter much in the large scale.

Even if we were using substantially less energy overall, transitioning away from fossil fuels for heating and transportation is likely to result in using a lot more electricity.

I expect most modern countries are going to need to be constantly upgrading their power grids and power generation capabilities over the next few decades. Hopefully most of the new energy generation will be renewables.

Where do you think the electricity comes from?

The same place hamburger, bacon, and keebler cookies.

elves

It depends. Fossil fuels are still a major source, which needs to be replaced as soon as possible with something else (where "something else" probably means some combination of solar, wind, and nuclear).

What I'm saying is that in our long term planning, we shouldn't be thinking "we have X terawatts of fossil fuel production capacity so we need to build out X terawatts of non-fossil-fuel production capacity to replace it (including storage, as needed)" We should be thinking "We probably need ~2X terawatts of additional production because we're replacing fossil fuels for electricity generation in addition to transitioning to electricity-based heating and transportation."

EV's mostly charge at night when demand is low, so they should have an undersized impact.

We'll need more generation, but only about half as much more as a naive analysis would indicate.

Interestingly, energy use actually declined during the first year of the pandemic (I know for sure in the US, and probably globally). You can see it here for Texas, https://findenergy.com/tx/#overview , in the annual production graph, which is the state with the most production in the US. On a national level IIRC, it was approximately -4% from 2020 to 2021.

It's one thing for power use to hit record highs - I would expect increase to be more or less constant with economic development and/or population growth. The sources from this article state that the mid-continent ISO is actually expecting to _lose_ 3,200MW capacity 'as result of retirements, and the decreased accredited capacity of new resources.'

https://cdn.misoenergy.org/2022%20PRA%20Results624053.pdf

Yeah, it's nuts that power companies either didn't notice their capacity was decreasing, or they didn't care. Might be time for them to lose their public utility status, and have that passed to someone who can add.

The PDFs below are a little dry, but they have noticed and cared since at least their 2018 forecast. As for the remedy, what exactly does it mean to lose public utility status?

[1] https://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20...

[2] https://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20...

[3] https://cdn.misoenergy.org//Aligning%20Resource%20Availabili...

They're a state-sponsored monopoly. If they're not competent to maintain enough capacity for the easily-projected demand, go find someone who can and make them the new state-sponsored monopoly.

It'd be ugly, but if no one loses their job nothing will change.

We already have carrots and sticks to ensure reliability. That's what regulation is for. Changing those is a lot easier than changing companies.

I would hope that the expansion of renewables would offset the loss of fossil fuel based sources due to retirement. That would be the ideal scenario.

They are not directly interchangeable. Fossil fuels work at night and when it's not windy out.

We use vastly less power at night especially when you remove people specifically using power at night do to lower costs. So for now that’s a non issue.

I should have said "when not sunny", not "at night". Point being - without vastly improved and expanded energy storage, you cannot have a reliable grid powered by renewables. But with nuclear, you can ditch fossil fuels.

Nuclear becomes exponentially more expensive the more you add to the grid and thus the lower it’s capacity factor becomes. France was exporting a lot of nuclear power and importing non nuclear and still fell down to 70% cf when the US was at 90%. One way of looking at that is a .9/.7 - 1 = 30% increase in cost per kWh overall. However compared to 90% for 35% the next 35% cost 60% more per kWh. And aiming for 90% would be even more expensive.

If we want a zero carbon grid nuclear simply isn’t viable on it’s own, it needs massive energy storage or peaking power just like wind and solar.

Yeah, I was thinking more in terms of nuclear base capacity (humming along at ~constant output) and renewable for peaks. Still need vastly more and better storage for that, but not as much as exclusively renewable.

Also, not being in this field, I may have misunderstood you or confused the terminology.

Yeah, but if your extra load is coming from climate change that makes the planet more windy and is at its worst when the sun is out then it matches up nicely with the extra demand.

This should bode well for electric vehicles if an electrical outage were to occur.

One of the inherent weaknesses of external combustion engine-powered vehicles.

Unless you routinely run your electric car down to a low battery, it's not a problem. 300+ miles per charge is normal now. Most people top up every night. It would take a power outage that lasted longer than a week for my car to run out of juice with my normal driving habits.

By the way, if the power is out, gas station pumps don't normally work either.

They could have on-site generators for the gas pumps for such an emergency.

It's not like they don't have a tank of gas nearby .....

If we all had electric cars, those batteries would BE the grid.

This isn't really true without massive distribution system redesign. It could be true for individual homes with transfer switches that disconnect them from the grid if that's what you mean, but a peer-to-peer setup where your house is powering your neighbor down the street doesn't work without some substantial investment.

Even the first part (powerwall setup basically) could be useful.

But it’s best when coupled with reduced usage where possible - if we get to a point where the majority of residential electrical usage is cars we’ll be doing pretty well.

What substantial investment is required, at the hardware level?

A big inverter would be the first thing you need.

Right, I guess my question is if anything that needs to be done to the grid itself (outside of individual homes). You're going to have a big inverter anyway to power just your house from your solar panels, batteries, whatever.

You need transmission lines capable of handling the current. Too much current through local transmission lines can cause drooping lines, which touch trees (i.e. grounding that line), which shuts those transmission lines down automatically.

It's something like this that helped cause that big power outage in Chicago. Power couldn't flow through the big transmission lines (they got shut down for a couple of reasons), so it flowed through the local lines towards the demand, and then those were shut down because they hit trees.

Ironically, this electrical current issue is also a problem for charging electric cars at stations, since one fast charger at 20% utilization can draw 2-3x more power than the store whose power grid it's attached to.

This is the original claim I'm asking about: "a peer-to-peer setup where your house is powering your neighbor down the street doesn't work without some substantial investment."

Wouldn't it be less current, in general? If your neighborhood gets 20% of its power from solar panels on your neighbor's houses, it seems clear that power lines inbound to the neighborhood would need to carry 20% less load, and lines inside your neighborhood would also carry somewhat less on average (your power consumption has not changed, and presumably your neighbor is powering his house off his own power before exporting the rest to you).

I understand there are probably some topological hot spots, like if one dude has a 1MW solar installation in his backyard, but as a general principle I don't understand what big upgrades need to be done to the grid to support P2P.

Honestly, I'm not qualified to answer this specific usecase. But here's a few numbers to chew on:

An average home consumes just a little over 1kW on average, and its connection to the grid will be built to that tolerance.

Solar generation is generally peaks around 5kW for an average house installation; individual panels measure between 250 and 400 watts.

An electric car's battery pack can, pretty easily, provide in excess of 100kW (based on their charging rates, and the fact that batteries charge more slowly than they can discharge - about 4.6x slower for Tesla battery packs).

Huh, this is interesting to me. My house has 400A service (or at least that's the main panel breaker size). 400A at 240V is somewhere in the ballpark of 80kW possible draw.

Does something bad happen if I actually use my home's full service for an extended period? I understand bad things would happen if everyone on my block did.

As a thought experiment, would anything bad happen if a magic wand was waved over a neighborhood, and each individual house now supplied power to the grid in the same amount it would usually be consuming it? A whole neighborhood with normal electric flow, just reversed. Are there components of the grid that are unable to deal with reverse power in the exact same quantity and load distribution as forward power?

You would have to spread that load across multiple circuits in your house to avoid tripping breakers/blowing fuses. You might not have enough capacity to absorb all 400A if the original installer oversized the main breaker to allow for additional capacity later.

The bad thing is you would get a huge power bill at the end of the month.

I think the thing I'd be worried about first is the transformers. Would they handle that much power flowing in reverse without overheating/catching on fire? They're already one of the weaker parts of our system; they don't handle overcurrents well (they're largely what folks are concerned would fail due to a solar flare).

The second thing would be all the sensors attached to the grid. Would they identify a sudden and dramatic change in flow as a fault, and throw their switches?

Also, where would that supply of energy flow? If there's not enough of a sink for that electricity, what happens to the generators? If the sink is outside the local area, what would happen to those transmission lines?

Distribution systems are designed for unidirectional power flow from the transmission system at the top to individual users at the bottom. If you start doing things like reversing power flow and exporting power through already very heavily loaded distribution transformers you create bottlenecks, stability issues, and safety issues that were not originally designed around. If you expect users to be importing and exporting power from their residence you need to have automated protection that can disengage those residences from the grid at a minimum, and probably a lot more that I am not an expert on. My day job is designing control systems for renewable generation and as a result I have some exposure to transmission level and generator bus level protection schemes and they can be quite complex!

This is what I'm looking for, thanks! Any pointers on good places to read about these issues?

> but a peer-to-peer setup where your house is powering your neighbor down the street doesn't work without some substantial investment.

We already have grid-tie solar, and beyond that, grid-tie batteries at homes and them doing work to support frequency and voltage of the grid in many locations. Having an EV do it isn't any harder.

The big challenges are figuring out how to compensate/bill for all this.

Rather then some strange comment about electric vehicles and your apparent distaste because they depend on a electric grid (as opposed to gas, which is also a commodity and has the same issues, transportation being one of them). Why not look at how do we fix this? Maybe we need to offer more incentives for solar panels or fix the grids by providing money to the states?

Electrical outages take out gas stations, too.

We have an electric car, and we survived a 3 day outage caused by a wind storm a few weeks ago just fine. That's because we started with a full battery, just like we do every morning.

OTOH, many people with gas cars had empty tanks and had to endure 4+ hour lines at the few gas stations that had electricity and hadn't yet run out of gas.

I suppose you'd have been fine if your gas car had a full tank too?

It's harder to ensure a full tank of gas every day than to ensure a full EV charge every day.

Yes it takes more effort to ensure you stop by a gas station on the way home than it is to plug in at night, but still fairly minimal.

My father instilled in to me to never let my tank get below half, in case there are emergencies and I still follow that practice.

Two points:

1. When I switched to an EV as a daily driver, I realized I had underestimated how much time and hassle I was spending at gas stations. When it's been engrained in your psyche as being just part of life, we don't really realize how nasty gas stations are compared to just plugging something in.

2. Another good reason to never let your tank get below half: If you get "bad" gas, you'll only have half a tank of it, diluted with presumably "good" gas. Rare occurrence nowadays, but a benefit nonetheless.

A buddy of mine from the Air Force (I wasn't in AF, he was) told me they never fill above half-full unless you are going long distances. Apparently something to do with MPGs and the weight of fuel.

I never listed to him, but to this day he still only fills his car up half way for better gas mileage. I'm sure it matters if you have a 200 gallon tank, but 60 pounds of fuel will not make a difference in a vehicle that weighs 3000 pounds. Adding a single passenger is going to be more than double that.

I can see it for something with a very large fuel capacity, that doesn't often get driven around motorpool/base unless deployed.

Agreed - it was dumb of him to adopt that in his civvy life, but I couldn't convince him otherwise.

The EPA says that for every 100 pounds of weight carried around by a car, fuel efficiency drops by around 1%.

So if you avoid carrying around 8 gallons of fuel everywhere, you could increase fuel efficiency by around half a percent.

Not a lot of savings.

https://www.fueleconomy.gov/feg/driveHabits.jsp

This is where EVs don’t work so well: you are carrying the entire weight of the battery no matter how much it is charged. Not so important for cars, but for electric planes it is significant.

Regenerative braking negates much of the weight penalty in cars.

You don’t have to. I have 15 gallons of ethanol-free gas I get from the airfield (unleaded) in the garage for filling up the mower, generators, the snowblower, etc. It usually takes about six months to get through it all. I fill up a jug when it goes empty, etc., etc. If it really becomes a problem, I’ve got plenty of gas to get to where someplace has working gas pumps.

Sure, but I think the point is that EV owners are topping off everyday vs ICE owners don't fill up until they need to (maybe weekly or if it runs low, certainly not daily).

I am sure many EV owners also woke to mostly empty batteries because they didn't charge them that night and other ICE car owners who just happened to fill up the night before. Not sure what advantage you see for electric here. Both need to be "fueled" somehow.

In fact you can fill an ICE car in a few minutes so you can have many cars per single pump. EVs require long periods of time at a dedicated plug that many apartment owners, renters, and others don't have access to.

Could you imagine the lines at a charging station when it takes 30 minutes best case per vehicle?

As long as the sun rises, my house has power and my EVs charge. The sun is more reliable than petroleum supply infrastructure. It’s Mr Fusion 8 light minutes away.

I don’t have batteries yet because I have favorable net metering and my utility is natural gas fired (which pairs better with distributed renewable generation vs coal or nuclear), but if/when their power quality declines, my system is future proof to accept battery storage.

Most people have grid tied systems that disconnect when the grid is down :/

Most US grids are reliable, and more rooftop solar would bolster generation when it’s needed: hot days when the sun is shining and cooling loads are straining generation.

"The sun is more reliable [in my locale]"

Fixed it for you.

Germany has 60GW of solar generation capacity (with plans to expand that to 200GW by 2030) at its relatively high latitude. When it’s hot, the sun is shining. When the sun is shining, you use it for cooling. You even can use it for heating in the winters with heat pumps.

There are very few inhabited places on Earth where the sun is unreliable enough that today's solar is uneconomical. My parents live in upstate NY—hardly a place known for its high rate of sunny days—and since they installed their rooftop solar, they've said it generally provides enough power for the house unless it's actually covered in snow.

Would you spell it out for us? What exactly happens when the grid fails at your site? Overnight, obviously you have no power. Then "the sun rises" but surely if there's any load, it must get above a certain point in the sky before it can run your house and charge your EVs, especially in the winter. And same thing in the afternoon, there must be a point where the sun is shining but too low in the sky and the load cannot be sustained- what exactly happens then?

(also, no clouds where you are?)

Assume there is no utility power. The sun must rise high enough on the horizon to energize the system (I haven’t done the math to know what the altitude/sky position is yet, I’ll report back). Once energized, the system will operate as a microgrid and attempt to maintain frequency and voltage if there are clouds until the sun sets and falls below the horizon. Some loads are more tolerant than others. The EVs of course will charge whenever there is power, and the hot water heater and pool pump will operate just fine only when power is present. Only the AC unit is fickle about power quality.

[removed strawman comment for being unnecessarily combative] When my utility is operational, I’m not just offsetting my consumption but also pushing back clean power into the grid for others locally to consume (which reduces the natural gas generation required of my utility) and while the failure mode isn’t perfect when the grid is down, it’s better than having no power at all (until battery storage costs decline).

Edit: https://pv-magazine-usa.com/2021/10/25/enphase-launches-micr...

Not a strawman (I don't think? not sure what you meant by this..), just very curious. You have a setup that is in between useless when grid is down, and fully self-sufficient. Seems possibly very practical to me, but I haven't heard of it before.

It is a very uncommon setup, I've met installers that don't know about it. The new generation of microinverters enable this setup more easily, so it should become more common.

My EV is plugged in whenever it is in the garage. I'm sure most other EV owners have the same habit.

Pretty much every EV owner that can plugs in every night.

People just do not get how much different an EV is vs gas. Never having to go to a gas station is just SO nice once you never have to do it.

If you arrive home by 6, it normally takes 4 hours to top off. It should be done by 10pm. What time did the grid go down again?

The car charges at a rate of 48 km/hr on a 32 amp garage circuit. So on an average day it recharges in less than an hour. However, it's programmed to charge at the cheap overnight rates, so it doesn't start charging until midnight.

Therefore we were out ~30km of range when the power went out. 370km rather than 400km is still "full" in my opinion, though.

> That's because we started with a full battery, just like we do every morning.

Which - if we're being honest - is a bit of a privilege. Most folks with cars do not have a dedicated garage, nor a parking space with an outlet.

> That's because we started with a full battery, just like we do every morning.

You're assuming the electric outage happened after you filled up your car with electricity.

It could also be possible that the electric outage happened after you filled up at the gas station and the gas station ran out of power.

Actually, it did happen later in the day, so it was down 30km, an average days usage. So we had 370km of range rather than 400km. That's still full, essentially.

My point is that the only time the car drops below 300km of battery life remaining is when we're on a road trip, so maybe once a month.

Whereas our gasoline car will have somewhere between 1/8 and full on a random day.

Yes, it should. EVs make great standby power sources.

Should be fine if the vehicle is kept charged. Ford F-150s can power most homes for a few days on a single charge [0]

[0] https://www.thedrive.com/tech/40695/the-electric-ford-f-150-...

First note: the so called smart grid with EV so far is a myth, we do not have not a common and fast enough and trustable enough signaling network nor fast enough spread inverters. Even if we ever arrive to agree a common standard for signalling who can trust it at a grid scale? So far we have just frequency shifting witch is reasonably trustable since it's so dumb that can't be really sabotaged but it's far from being usable at grid scale, so IMVHO we can count on EV batteries where we are at home for our own home backup. Something VERY local, for a very local microgrid. No more.

II note: with the premise that I do not know how electricity grid in USA was made, in EU the grids was made by States years ago, then get privatized and so far almost no real upgrades have happened, witch is a classic plot for all privatization: profit for the company, no investments, cry for State money and help "it's an emergency!!!!". This model MUST be ended ASAP if we want to have something still working at certain scale in the future.

How is backfeeding the grid to power it with EV batteries any different than backfeeding the grid with home solar generated power, this already happens with grid-tied solar installations (which is almost all of them)?

Lithium batteries can push a lot more current than solar panels. Orders of magnitude more current.

For a point of reference, charging lithium batteries has to happen slower (lower wattage) than discharging, and a car battery pack with a fast charger can pull over 350kw.

In comparison, a residential solar system for a house will provide somewhere around 5kw, peak.

Yet somehow my car doesn't take down the grid every time I plug it in and the car tries to draw 150KW to charge the batteries at max rate.

Why would grid storage configure the cars to supply power at the max battery rate?

Even if my car and the grid-tie system malfunctioned and tried to send unlimited power to the grid, the 30A breaker its plugged into would trip before it hits 7500W.

It's not like it needs to handle a lot of power because EV's don't have that much power to spare, if I have a 75KWh battery and am willing to give up 30% of my capacity to the grid, that's 25KWh, or around 2.5KW for 10 hours.

The grid, even if its your connection between your mains and your car, needs to be set up to limit the output of the car's batteries. Or the car (powerwall, etc) does.

> if I have a 75KWh battery and am willing to give up 30% of my capacity to the grid, that's 25KWh, or around 2.5KW for 10 hours.

That depends on how much current you can push. The battery is quite capable of giving up that 25kWh in 10-20 minutes too.

The grid, even if its your connection between your mains and your car, needs to be set up to limit the output of the car's batteries. Or the car (powerwall, etc) does.

Yes, and it does have those limits in place for charging, some in software (by the EV charge controller telling the car how much power it can use), some hardware (through at least 2 circuit breakers). Why wouldn't the same limits be enforced if using the car for grid storage?

Why does the max power output of a car battery have any bearing on whether it can be used for grid storage?

Because batteries can pump out magnitudes more power in a very short amount of time than solar. The two are not comparable.

But ultimately, what speaks more than anything I could ever say is the requirement for backup power generation (including household battery arrays) requires mains power be cut off before engaging - to prevent damage to the grid.

Because batteries can pump out magnitudes more power in a very short amount of time than solar. The two are not comparable.

The batteries aren't connected to the grid directly, the grid can't absorb DC power, it has to go through an inverter.

Even if your batteries can supply 500KW of instantaneous power, if they are connected to a 1KW inverter, the most the grid will ever see is 1KW - there's no failure mode that would let a 1KW inverter pump out 500KW of AC power, even for an instant.

A lot different: backfeeding the grid is not about stabilizing the grid but instead stabilizing the local MICROgrid of our homes, essentially a solar inverter take time to ramp up production when a load appear, take time to step down production when load drop. The grid compensate, with strict limitation, for instance in France you can't push more than 3kW if mono-phase, in Italy 6kW, even if you are allowed to have a monophase 10+kW for instance to being able to recharge your EV at 7kW, the needed common minimum to keep cells in balanced state (3kW slow charge is simply too low for most vehicles, they recharge but the battery might suffer).

A simple practical example: my dishwasher start heating water is +2kW in an instance, my solar inverter take 2-3" to ramp up to +2kW. The home microgrid need or the national grid or a battery inverter to avoid collapsing due to a too big frequency drop. The dishwasher have finished heating water, again a sudden -2kW, again I need to divert extra energy to the grid or a battery to avoid a frequency peak big enough to make the inverter disconnect.

AC grids are like a tense rope with various "needles" attached to it, some push the rope at a certain frequency, let's say 50 or 60Hz, some are dragged by that frequency. Those who push are generator, like a solar inverter, a NPP etc, those dragged are loads. Those who drag need to keep the frequency so when a new load arrive they need to push stronger at the same rate, when a load detach itself they need to push less at the same rate. It's NOT easy. We have made public grid of a certain size, not too little, not too big, to average the load, we have observed that at a certain size demand is "constant enough" to made easy keep the frequency. Below or above is hard, the resulting grid is unstable.

Now, since served people became bigger and grid infra was not evolved with them we start to make "grids interconnections" to help compensate, that's worked a bit, but also sometimes that have made catastrophic cascading effects. We are at a point we can't compensate anymore. Witch means grid instability on grid complex enough that restoring might take time, non in seconds or minutes but in hours or days.

Some hope that grid-tied EVs so batteries and powerful, fast inverters, can help absorbing excess energy when grid loads drop, pushing energy to the grid when grid load ramp up. I'm not in the field but given the very small example of my home system (solar + lithium storage) I really doubt such idea can work in practice. Also being an IT guy I fear about it's potential crapload of vulnerabilities. The sole solution I see is re-do the public grid, at State level, for current and foreseeable future needs, with enough NPP, gas PP and distribution networks to have again a stable national grid, no smartness needed. Since I doubt that's happen that's why I invest in a system with lithium storage witch here (France) is definitively NOT convenient economically speaking, preparing to more and more frequent blackouts, longer and longer...

That doesn't really answer my question.

If I have a home solar installation that can send up to 10KW into the grid, how is that any different than an EV battery that can send the same amount of power into the grid?

I could see the problem if people were sending 100's of KW into the grid with EV batteries, but that's not the case, they're sending a similar amount of power that a home would use.

3kW slow charge is simply too low for most vehicles, they recharge but the battery might suffer

I charge my EV at 1KW (9 amps at 120V), it's enough to recharge my 30 mile commute overnight. I'm not aware of any cars (in the USA) that have a minimum charge rate.

> If I have a home solar installation that can send up to 10KW into the grid, how is that any different than an EV battery that can send the same amount of power into the grid?

Your system have a purpose: give you energy NOT stabilizing the grid. Your system (like mine) actually disturb the grid and that's why countries have regulations on them. The actual grid have issues ALSO due to the augmented presence of p.v.

If you want to stabilize the grid with cars you do the very opposite, your system need not to give you energy using the grid as a kind of "energy buffer" but to be itself an energy buffer. Let's say we have such systems: at a certain point in time locally our domestic power plants see a sudden frequency drop, around 1Hz. As fast as they can they start pushing energy to the grid. All connected vehicles/plants. As a result the frequency peak in 2-3". Again as fast as they can all connected systems try to get energy from the grid to help the frequency going down. Again a network effect. The frequency drop. They reverse. In a minute you have generated enough oscillation that no one is able to keep the frequency, local grid segments got disconnected, all generators in the segment shut down.

Few big enough power plants can coordinate themselves to monitor and decide, too many small and spread can't. Even with frequency shifting witch is the fastest and simplest way possible we can't.

Trying proving the above: if you have a monitoring system fast enough, like an MQTT frequency sensors or a multimeter with frequency measure and graphing ability try to see a minute at night, with pv disconnected, no battery etc and a minute when it work. On grid ONLY you'll see a far more stable frequency. On p.v./battery you'll see a schizophrenic variation. Try to interpolate at a grid scale the resulting mess.

If we just inject to the grid our inverters simply try to push energy watching frequency, they push less as the frequency goes up but nothing more. It's up to the big power plant regulating. Being one-direction only they generate disturbances but still something manageable at least at a certain scale. A bi-direction system I can't really say if it's sustainable nor I know if a far bigger presence of p.v. can be allowed...

Edit: sorry, forgot a part, yes you can recharge and EV even at 100W but the battery do not balance cells, as a result it last less time. Witch means for instance 8 years instead of 10 of usable life. 7kW is generally the "needed mean" to balance for most EVs, normally you have such issue mentioned in the manual.

Can you show me an example, I looked at the owners manual for a few cars and none mentioned an issue with slow Level1 charging.

Few big enough power plants can coordinate themselves to monitor and decide, too many small and spread can't. Even with frequency shifting witch is the fastest and simplest way possible we can't.

My utility can already turn off my air conditioner during peak demand, why is it harder to turn off a 4000W load when they need more power than to turn on a 4000W battery when they need more power?

I have no example at hand, but if you look at "lithium battery cell balancing" you find many sources, two at random

A very generic article https://www.saftbatteries.com/energizing-iot/lithium-ion-bat...

A thesis who cite many balancing technique https://tel.archives-ouvertes.fr/tel-03584252/document

Another normal advise is not to charge at 100%, not because it's bad in general, the contrary it's NEEDED for a proper cell balance, but on cars have a potential issue, the very same who is an issue for the grid stabilization: when you ask for more power your battery and inverters do their best to give it, but when the demand drop, for instance because you have hit the accelerator then release it, there is an excess of power and on a battery-only system like a car you have no grid to push excess energy to nor you can shut down. As a result you divert excess back to the battery crating high DC ripple effects overheating and stressing the battery itself.

> My utility can already turn off my air conditioner during peak demand, why is it harder to turn off a 4000W load when they need more power than to turn on a 4000W battery when they need more power?

It's not an on-and-off problem but a how quick such on-and-off can happen. As I suggest try seeing your p.v. system when it works. Try see how the frequency spike constantly with p.v. inverter active and how stable is (normally) on the grid, try to se what happen when you start a big load and when it drop. To see that better, if you have lithium storage try to run exclusively on it+solar and see what happen.

AC is veeeery nice for many aspects, but it's hard for others. Try to keep a grid at a certain frequency is NOT an easy business, trying to do so with a gazillion of small generators is next to impossible.

Smartness is a modern hope we add in both regulatory capacity for generators and avoid sudden loads spikes because smarter loads can potentially advise generators before starting or stopping and eventually ramp up or down slowly. However so far we are in an uncharted territory with just very little experiments and nothing on scale.

One really important improvement that we can make for free* is increased memberships in regional automatic power markets and balancing authorities like the California Independent System Operator. Organizations like these make the exchange of power between neighboring entities smarter, faster, and smoother than they ever have in the past. It's not magic and it won't make an overloaded transmission system magically not overloaded but it's a huge step in the right direction.

As a small fish in that particular pond it concerns me a bit that the government of California has some degree of control over the organization, though. I'd say a positive change would be to massively expand the size of the various balancing authorities and put them all under the purview of the federal government.

*not actually free, but free on the scale of other necessary changes.

Can you be more specific? Naively, it seems like California state government having some level of control over the California Independent System Operator is entirely reasonable. Even under federal control I’d still expect California to exert some control over the process.

No, you specifically don't want state government control because they will tend to favor regional interests over national ones. California happens to be the ones who organized this particular ISO (which was a valuable service) but once it leaves the borders of California someone else needs to be in the driver's seat.

If there is one given application for solar power I'd say it is to avoid these silly power peaks when the sun shows its power. Given that it is the presence of the sun which creates the (often imaginary) need for air conditioning it is a no-brainer to run that air conditioning off a bunch of PV panels. Now that these have gone down in price so much that their purchase price is but a fragment of that of an HVAC installation I see no reason to forego on this component. With electricity prices going up during periods of high use - if not now then in the not too distant future, where I live (Sweden) the price varies by the hour - this is an investment which will pay back for itself in less than a decade while PV panels last for at least 25 years.

Electricity demand must rise dramatically if we want to decarbonize the economy. Everything that uses fossil fuels today needs to use electricity in the near future. Of course that doesn't mean that primary energy demand needs to rise, on the contrary, heat pumps and EVs are much more efficient than their fossil counterparts.

Reply on news.ycombinator.com