Hacker News Reader

Comment by goodSteveramos | Hacker News Reader

It will be several decades before a combination of renewable overcapacity and energy storage can meet any increase in demand for electricity. In the mean time, anything that causes an increase in demand, such as electric cars, will be met with an equal amount of additional coal or natural gas supply. This means electric cars in the US are 100% fossil fuel powered, and the Model 3 only has a ~20% smaller carbon footprints than average gas powered cars and a slightly larger carbon footprint than the gas powered Toyota Prius Hybrid.

EV subsidies have mostly lined the pockets of wealthy luxury car buyers who would have bought an EV anyway. Those subsidies would have reduced emissions far more if the subsidies had gone to making economy hybrids the same price as non-hybrids.

View on HN

A modern combined cycle power plant is significantly more efficient than a typical internal combustion engine, even when taking into account transmission losses. So, an EV is going to use less fossil fuel per mile, even if the power plant is fully fossil fuel based.

On top of that, per megawatt, building out new solar is cheaper than keeping existing coal plants running. The solar transition is largely constrained by the duck curve, which can be mitigated using v2g technology, so as time passes, the power sources get greener out of purely economic considerations.

As long as you're not throwing out brand new ICE vehicles and just replacing worn out ones, both the TCO and the total carbon footprint goes down when switching to EVs.

> It will be several decades before a combination of renewable overcapacity and energy storage can meet any increase in demand for electricity. In the mean time, anything that causes an increase in demand, such as electric cars, will be met with an equal amount of additional coal or natural gas supply.

There's a lot wrong with this comment, but this bit especially is just made up nonsense.

According to [1] 39.7% of USA energy generation on average is already either renewables or nuclear, and as other commenters have pointed out even if it was 100% coal including transmission losses it would still be more efficient than petrol in an ICE.

And saying that all additional power production built in the US grid will be 100% coal or natural gas for the next two decades is a bizarre claim.

[1] https://www.eia.gov/energyexplained/electricity/electricity-...

>And saying that all additional power production built in the US grid will be 100% coal or natural gas for the next two decades is a bizarre claim.

The electricity grid has something called a Capacity Factor. Capacity is the ability to produce power measured in Watts. The Capactiy Factor is the fraction of capacity which is actually producing that power at any instant in time. Power demand and supply must be almost exactly in sync at all times to maintain grid stability. This is achieved by turning on and off coal and natural gas power stations, which can be done relatively quickly, to ramp up and ramp down production as demand fluctuates. Capacity must always greatly exceed typical demand so that unusually high demand can be met. Nuclear, wind, solar and hydroelectric power plants always produce whatever power they can independent of demand. This is for a diverse set of safety and environmental reasons. Only fossil fuel plants can "load follow" to meet changes in demand. Like I stated in my original comment, all load following in the US is done with fossil fuels, and that will be the case for many years or several decades to come. Increases in capacity will mostly come from carbon free sources, but until those sources can load follow, either by dumping unused energy in off-peak hours or by storing that energy until it is needed, all load following will be done with fossil fuels. This means additional demand is going to be met entirely with fossil fuels for a long time to come. I hope that helps you understand the concept of marginal demand better.

I'm confused - you posted on a thread about EVs, the problem here is a long term ramp up of demand over the span of years or decades as more and more people swap their ICE vehicles for EVs. This is a problem to be solved by the building out of more power production capacity over the long term. Dealing with short term load spikes with existing infrastructure is a different problem that isn't really relevant in this context.

Power demand from EV charging is pretty predictable too - the bulk of it happens overnight when demand is usually lowest from everything else.

>Dealing with short term load spikes with existing infrastructure is a different problem that isn't really relevant in this context.

EVs sold today are going to be charged on the grid we have today. if that makes their carbon footprint barely better than gas and slightly worse than a hybrid it means our subsidies would be better used to increase carbon free electricity generation and increase hybrid sales. Once EVs have smaller carbon footprints then it will make sense to subsidize them. If you don't care about the short term then yes. The long term solution is full carbon free electricity and EVs

EVs are actually a nearly ideal load for load-scheduling to match net grid draw to times of overcapacity. If anything, they have the potential to vastly improve grid stability, which would otherwise become a concern as an increasing amount of power is supplied by fluctuating renewables.

Adding the charging demand, some of which will inevitably occur at peak-demand, is not going to improve the stability of the grid compared to a gas car that never places any load on the grid at all. EVs will only improve grid stability if owners chose to feed the grid during peak demand.

> Some of which will inevitably occur at peak-demand

Some, but nowhere close to most. If 90% charge their cars at night, or even just at the grocery store after work, that is off peak.

It's not that simple. Feeding power back to the grid certainly helps (and this will likely become ubiquitous; even just tapping into the top 10% of capacity would be a massive benefit). But EVs connected to the grid can even help improve grid stability, even without feeding power back to the grid, through simple frequency-based dynamic demand methods. At peak demand, the grid is most sensitive to marginal loads. EVs that are charging at this time can modulate their power draw, even without returning power to the grid. In effect, if you turn on your kettle, a charging EV can simultaneously lower its draw rate by almost the exact same amount, so that the grid load gets smoothed out. This can be done without any data or networking. It is equivalent to adding additional rotating inertia to the grid, and very helpful for grid stability. It's a technology that is not widely adopted today, but very cheap to implement and likely to spread quickly.

the effect of increasing peak demand is going to outweight that advantage massively

It actually does not.

The fact that EV charging will mostly happen at night means the fraction of EV energy coming from renewables will lag far behind other electricity uses even after some of it is able to be met with renewables. Fossil plants pick up the slack from renewables at night and that effect will increase as solar replaces nuclear.

Why don't you mention "big batteries" ? These make wonderful peakers. Victorian Big Battery (VBB) in Australia is 300MW. Comparable in size with a nuclear reactor.

Thanks to the EV success, lithium batteries costs are falling exponentially. Coincidentally, photovoltaics costs are also falling exponentially.

Conclusion: regarding electricity production, fossil fuels are soon not going to be competitive, PV + batteries would be so much cheaper...

When? With the actual costs curves, in less than a decade.

No way they are "100% fossil fuel powered" when, for example, 20% of US power is nuclear. If you're just going to throw out something like that so casually, why should I believe any of it?

Nuclear power is used to provide baseline grid capacity. You can't just burn more or less uranium/plutonium as demand changes with the reactors that are in service. On the margins the surge demand largely comes from hydrocarbons, because you can just burn more or less coal, oil, or gas. Thus electric vehicles that tax the grid are going to result in new power being added as needed, and it will be nearly all from hydrocarbons.

Most EVs charge at night, which is typically the time of lowest grid demand, and used to set the operating point of base-load plants like nuclear and coal.

Raising the demand at night will disproportionately come from nuclear, and if anything by raising the baseline, it'll mean those nuclear plants are running more around the clock _decreasing_ the amount of gas peakers needed during the day.

This is totally wrong. Nuclear has a very low marginal operating cost and to maximize return on investment Nuclear plants run at 100% of their allowed output irrespective of demand. Increases in the price of baseload power are unlikely as intermittent sources like wind and solar contribute a higher fraction of power. In particular, midday solar output will lower the baseload far below the current low point at night.

You can, however, charge your EV at night. Natural gas and hydro work well with nuclear, you can fire them up at peak and wind them down at night. Bonus if you can do pumped storage using extra power at night from the nuclear plant, but I don't hear that happening much in practice.

You should beleive it because its true. Please read my comment again carefully. Marginal demand is not an intuitive concept, but it is quite real I assure you. Please don't be upset. My figures are from the paper "Conventional, hybrid, plug-in hybrid or electric vehicles? State-based comparative carbon and energy footprint analysis in the United States". Also see my reply to virtue3 for what I hope is a clearer explanation of the marginal demand concept.

https://www.sciencedirect.com/science/article/abs/pii/S03062...

Another point to consider here.

Even if EVs are powered by fossil fuels. A powerplan producing electricity and transmitting it to your house is orders of magnitude more efficient than generating energy from gasoline in a motorvehicle using an internal combustion engine.

Also we can curb emissions at the power plant and track them dramatically easier.

And more EVs does not mean more power plants turning on per se. Most EVs are charged at night when power usage drops dramatically (and when you get cheaper rates on your electricity).

The real answer here is using more green house gas free power options like nuclear and solar and wind. I doubt we'll see much nuclear in the near future because it is so cheap to create solar and wind farms at this point and there's dramatically less risk despite it being dramatically less productive.

EVs generate a bunch more carbon during their manufacturing process.

> As of 2022, Woodley's research showed that aggregate use of EVs below 55,749 miles may — in the U.S., at least — fail to generate any emissions benefit over gas-powered vehicles.

https://news.harvard.edu/gazette/story/2023/08/when-buying-a...

Also, If we end up using coal to charge an electric vehicle, its carbon emissions are actually higher than that of can you an ICE vehicle.

Then if you have to replace the battery on your EV, that will reset some of that carbon offset.

A Tesla Model 3 has a lower carbon footprint than an Toyota Corolla after 13,500 miles per Argonne National Labs (vs a Harvard puff piece).

Only 20% of electricity in the US is generated with coal, and phaseout continues at speed. There is over 1TW of renewables in US interconnect queues.

https://news.ycombinator.com/item?id=38243912 (citations)

https://cleantechnica.com/2023/12/01/us-coal-electricity-dro...

https://www.eia.gov/todayinenergy/detail.php?id=55439

https://www.pv-tech.org/nearly-1tw-of-renewables-in-us-inter...

Correction:

“The United States has at least 1,350 GW of wind and solar capacity and 680 GW of storage is waiting to be connected, which is enough to double the country’s electricity supply.”

https://www.pv-magazine.com/2023/12/08/solar-lcoe-now-29-low...

This is so boring stop.

Expect the FUD to get more and more unhinged as it becomes harder and harder to for anyone not either malicious or stupid to believe it.

BRB I’m gonna go build a new coal plant because my neighbor got a polestar.

>And more EVs does not mean more power plants turning on per se. Most EVs are charged at night when power usage drops dramatically

Consider:

Scenario A: 1,000,000 EVs are put on the road. Scenario B: 1,000,000 gas cars are put on the road.

Scenario A: 100% of zero carbon energy is consumed by existing demand. Remaining demand is met by fossil fuels including the energy needed to power the 1,000,000 EVs. Scenario B: 100% of zero carbon energy is consumed by existing demand. Remaining demand is met by fossil fuels, not including the energy to power additional EVs.

In Scenario A, more energy is generated by fossil fuels by exactly the amount of energy used to charge the EVs. In the US, at no point in time, day or night, summer or winter does any zero carbon energy go to waste. So increasing demand means more fossil fuels being burned.

> In the US, at no point in time, day or night, summer or winter does any zero carbon energy go to waste. So increasing demand means more fossil fuels being burned.

Are you just forgetting the part where there's less gasoline demand or any efficiency differences?

Yes. Those two factors are completely irrelevant for calculating the source of electric power used to charge new EVs. Gasoline demand and efficiency differences are instead captured in the carbon footprint estimates.

But you yourself estimate 20% savings and then hand wave that away under "fossil fuels."

I have no idea what you are trying to say.

In Scenario A, more clean energy deployment immediately reduces transportation CO2 emissions. In fact, every single bit of roof top solar being deployed contributes, in some small way, to reducing CO2 emissions from transportation.

In Scenario B, no amount of clean energy deployment changes the emissions of transportation. In fact, a good chunk of that production is likely wasted due to the mismatch of production and demand, without the benefit of storage.

Why are you only concerned with emissions of transportation and not total emissions? Replacing 10 GWh of fossil fuels with 10 GWh of renewables will reduce total emissions in both scenarios because that 10GWh of fossil fuel energy was being consumed fully in both scenarios. Eventually, when almost all electricity is carbon free, adding more carbon-free capacity will stop reducing emissions because your capacity greatly exceeds demand. Once that happens, then yes, scenario A will result in greater reduction of emissions for every unit of new carbon free electrical production because it is offsetting gasoline combustion. But my point is that we are decades away from electricity being nearly 100% carbon free.

Because the scenarios in question are only concerned with emissions from transportation, but more importantly disregard storage.

Multiple countries at present, at times, exceed their energy demand by renewable production...but the excess is pretty much just lost, their isn't adequate storage to bank that energy.

EVs do bank that energy, and more importantly they're likely to do it during the parts of the day demand isn't as high (i.e. peak sunshine).

Moreover, centralized emissions control or reduction is much more addressable then 1,000,000 individual combustion engines.

> orders of magnitude more efficient

This is simply not true. It's more efficient, but a typical gasoline car gets around 20% efficiency (large diesel trucks can be significantly greater).

A modern, state of the art combined cycle natural gas fired power plant is only up to around 60% efficient. There's some additional loss for distribution, charging the EV battery, and in the electric motor, but it's not large and you're probably about 50% efficient overall.

Another win is that natural gas emits around 25% less CO2 than gasoline per unit of energy. And of course if you're powering your car with solar, then that's much better.

Anyway, thermal efficiency is on the order of 2.5x better, not "orders of magnitude".

Reply on news.ycombinator.com