Electric car power generation

Factual Questions
61

Well, that’s dumb. You don’t build HVDC links for short distances, because the cost is in the endpoints and the savings is in the distance. I believe you, but really that’s the stupidest infrastructure project I’ve heard of (since the Alaskan bridge to nowhere) unless it’s only to serve as a testbed (which is still dumb since the tech is proven).

By comparison, the Pacific DC Intertie cost, probably, a few billion to build (it’s hard for me to find figures, but they’ve more than doubled capacity with upgrades that cost hundreds of millions). It’s 846 miles long and efficiently transmits 3 GW of power from hydro resources in Oregon/Washington down to LA. There’s no reason the same can’t be done elsewhere (and more cheaply due to modern tech).

It should be possible to do a lot, though I guess it depends on the details. A normal hydro plant is going to be sized for typical inflow. That might be a gigawatt or so. But there’s no reason you can’t install (say) 10 GW worth of turbines to handle much higher peak loads. Use batteries to smooth out the short-term peaks.

Still, adapting older stations should be a significant benefit. I like hydroelectric, but I also like nature. In California, we had two valleys of similar beauty: Hetch Hetchy and Yosemite. We kept Yosemite and converted Hetch Hetchy to a reservoir. That was an ok tradeoff as far as I’m concerned but I sure as hell don’t want to convert Yosemite into a reservoir.

62

BTW, Biden’s plan allocated $100B for electric grid improvements. I can’t tell if this involves long-range HVDC interties, but regardless, that should be enough for tens of gigawatt-kilomiles if the Pacific DC Intertie is any indication. Maybe we can get a decent connection to Texas at some point to use their wind resources.

I claim that grid improvements can happen much more rapidly than nuclear installations. They can be built in parallel, so long like don’t actually take longer than short ones. And they don’t have nearly the NIMBY problems of nuclear. Farmers, ranchers, etc. will be happy to lease their land for a few towers. We must construct additional pylons.

63

The map looks pretty good to me. The top consumers are the US and China, and both have significant solar resources (and plenty of land). The populations may be biased a little to the north, but not a lot, and it’s straightforward to move the power production that short distance.

Europe and Russia are not in a great position. But Europe has resources to the south, and Spain already imports electricity from Morocco. They can build that out further. And there are significant wind resources in the area, so they’re less dependent on solar.

Russia, I dunno. Most likely the rest of the world will have to make up for them. They’ve got serious problems even ignoring any internal switch to renewables.

Ideally, we’d move some of the highest power industries to places where it makes the most sense. In particular, aluminum and steel processing. Put 'em right next to the cheap solar locations. Doesn’t make sense to keep them where the only cheap power is from coal.

64

Quite a lot of aluminum is smelted in Iceland, taking advantage of its electricity generating capacity.

65

The early aluminum industry in the US was located in the Northwest and Tennessee Valley areas to take advantage of cheap hydro power. There’s even a town in Tennessee named Alcoa, a former company town. The cheap hydro power was a result of dams built under the New Deal. Most of the plants have closed here in the Northwest (don’t know about Tenn), I expect because of high labor costs.

66

On the face of it, this makes sense. However, please consider this : Past Aluminum manufacturing in the US, signed long term power offtake agreements with hydro-electric projects. All was hunky dory when the demand for Aluminum was good, but when the demand tanked, these Aluminum companies started having massive inventories, but kept the plants running because of the offtake agreements / local subsidies to keep jobs. This caused the Aluminum crisis of circa 2013.

So cheap operating costs of Solar Plants may not be the only reason to move metals smelting near them,

67

The biggest cost is a tunnel through the Pyrenees. A second cable planned to run underwater along the west side of the Pyrenees will lift French-Spanish interconnection capacity yet further to 5 000 MW.

68

Because @Sam_Stone’s map is total consumption by country, it’s showing us population size almost as much or more than it’s showing us energy consumption per capita or per unit land area. Which are the issues for generation and transport to the end users. It’s not wrong, but it’s not apples-to-apples comparable to the insolation maps.

His point remains legit however, that much of the best areas for insolation are economically poor, under-inhabited, or both while many of the densest concentrations of consumers (both people and industry) are in areas relatively poor in insolation.

All of which says planet-scale solar electricity will have to come with a long extension cord if it comes at all. Which will increase costs over some theoretical ideal where supply and demand are conveniently co-located everywhere.

Will this be greater than the true cost to civilization of emitting monster tonnages of GHGs as we do today? Probably not. Adult decisions take the true costs into consideration, not just the obvious or traditionally recognized ones. Humanity needs to become an adult pretty darn quick.

69

Seconded. However, seeing how things are going; CO2 emissions reductions are/will be too little and too late.

And I have a nagging feeling that we will be forced to look at global dimming / geoengineering. That will be sad and non-adult thing to do, but it looks like we are heading that way.

70

Just to create a side trail to this conversation - my carbon footprint - myself - if Google serves me right, is about 2000kg carbon from my non-electric car and 1000kg carbon from my natural gas furnace each year. Since one car is a Tesla, I could cut my personal emissions by two-thirds by making my other car a Tesla also. OTOH, I seriously don’t expect most of Canada to switch from natural gas furnaces to electric without significant financial incentives, and I doubt the generating capacity or grid infrastructure is up to the task yet. I figure issues like how cleanly the electricity is generated, or details like how the products I consume reach stores, etc. are issue I can do very little about.

For example, I suspect that moving goods by rail rather than by individual semi-trailer over intercity distances is probably most efficient - but my impression of the Canadian rail system is that it is stuck in the 19th century and not interested in making intermodal transport rapid and efficient to compete with trucks. If it could, how much traffic would be off our highways? Canada is characterized by long distances and sparse population, so electrifying the railways is probably not as efficient as it might be, for example, in the eastern USA.

My electrical consumption went from about 1.5MWh/yr to 2MWh/yr with an electric vehicle. But this is off-peak, so not adding to the peak grid capacity demand.

I gather they’ve begun constructing the demonstrator large fusion reactor to try to make Tokomak tech commercially feasible, so actual fusion power is only 20 years away… just as it has been 20 years away since I started reading about it in the 1960’s. :slight_smile:

Canada at one time had a design for a sealed small nuclear reactor that could produce enough heat to, for example, provide heat for a large apartment building. This was of course before anyone considered “do we want every building superintendent to be a nuclear power engineer too?” (Doh!) and also “Do we want highly radioactive materials to be widely available on every city block to anyone who might want to build a dirty bomb?”

71

It sure seems that way. If pandemic lockdowns barely registered a blip in reducing carbon emissions, I don’t see what legislation could possibly do it.

72

I did notice a lot fewer cars on the road - maybe for 2 months at most. My house and office were heated/cooled the same whether someone was there 24 hours or never - same I assume with every other building. I still consumed the same amount of groceries, so those trucks and ships were still running; for every restaurant that was closed, there were probably 100 people using their ovens and stovetops. Fed-Ex also probably doubled their shipments thanks to shopping online.

What really impressed me was the level of pollution drop once people stopped doing many commercial/travel activities - clean skies over major cities, or the stories about the canals of Venice so clear you could see the bottoms.

73

That’s marketing woo from your supplier. As stated above, electricity is fungible. There is no true storage of electricity. So if you select power from a “green” supplier, that just means that power across the grid shifts to more fossil fuel sourced. Your choice may motivate producers over the long term to think about more green energy production, but it doesn’t change how much energy across the grid is produced fossil fuels. That’s what I mean when I say you can’t cherry pick your power source.

74

When did power transmission by HV DC become a think? Were people being Westinghoused to death left and right?

75

Yeah, I looked into it a bit more and it’s not quite as bad as what I was thinking. Even more important than the tunnel, though, is that the link serves to connect two independent grids. That’s a useful feature of HVDC (AC is hard to synchronize). But it does mean that bringing it up as an example of the expense of HVDC is not useful. Distance wasn’t really the point.

I also hadn’t realized how much progress China had made just in the last couple of years. They now have a 1100 kV, 12 GW, 3300 km link operational. And a whole mess of links in the 5-8 GW range. That’s the kind of thing we need to be building in North America to connect the wind-heavy and solar-heavy regions with the populated areas. We should have an easier time at it than China, really. We just need to do it.

76

Cross-country transmission via DC is more efficient, even considering the inefficiencies of converting from/to AC at either end.

77

The main reason AC was ever an advantage was because transformers are an easy and reliable way to increase voltages to the levels needed for efficient long-distance transmission, and transformers require AC. But it’s possible to convert that to DC and then back again through more complex means (first, with mercury-arc rectifiers, and then with semiconductors).

Due to the expense of the conversion, it only makes sense for long distances or for independent grids that can’t be synchronized.

What makes it attractive for long distances is that the system (insulators, etc.) has to be designed for peak voltages. DC always runs at the peak voltage, whereas AC spends much of the time at less than that. DC links can transmit 1.4x the power compared to a similar AC link.

78

The Pacific Intertie (mentioned upthread), connecting the Columbia River to LA, was built in the 1960s and went operational in 1970. It was the first really long one, although the Chinese built a longer one a few years ago.

79

An AC transformer is just a couple of coils of wire around a hunk of iron. There’s no simple DC transformer. Changing high voltage DC to anything else (including a lower voltage DC) requires more complex machinery at either end of the line, which is very expensive.

On the plus side, AC wires, insulation, standoffs, etc. all have to be designed for the peak voltage, but the actual power you get through the wires is RMS. High voltage DC on the other hand always runs at peak, so for a given set of wires, etc. you can always push more power through the line using DC. Or to put it a slightly different way, for the same amount of power, you need more expensive wires, insulation, etc. for AC than DC.

So you have always had a tradeoff. The “transformers” on either end of the line are significantly more expensive for DC, but the wires are cheaper for DC. So if the line is long enough, DC is actually cheaper. Exactly how long is “long enough” has varied over the years, with practical DC systems needing shorter and shorter wire distances to break even with AC.

There are also inductive and capacitive losses (aka reactive losses) in wires, which are proportional to the rate of change of the voltage and current. For AC, these losses are significant. For DC, since the voltage and current are basically constant, you basically eliminate the reactive losses. Again, this makes DC more attractive, though you still have the added expense of the transformers and switchgear at both ends.

Reactive losses are even more significant under water, so undersea and other underwater cables have a big incentive to use DC over AC.

Speaking of switchgear, DC switchgear is much more expensive. At high voltages, when you open a switch, the electricity will arc across the open switch contacts. AC arcing naturally extinguishes itself since the AC voltage drops to zero twice during the AC sine wave cycle. DC isn’t naturally extinguishing, and will draw an arc over a very impressive distance at high voltages. This makes it much more difficult to design a high voltage DC switch, so again, much more expense.

Improvements in DC swtichgear and semiconductor-based DC transformers, rectifiers, and inverters have all reduced the equipment cost at either end of the line, but DC lines still require very expensive gear on either end, so don’t expect to see DC lines used for electricity distribution any time soon. But for point to point transmission over long distances, DC lines work well, and have for decades.

As for when high voltage DC first became a thing, that was way way way back in the late 1800s. Fancy shmancy semiconductors hadn’t been invented yet, so they used huge clunky mechanical motor-generator sets to do the AC to DC and DC to AC conversions. Expensive and inefficient, but it worked.

Thyristor-based semiconductors for high voltage DC were invented in the 1960s and started to be used in practical systems in the early 1970s.

80

Saving summer energy for winter is quite possible, to a limited degree (but then, we don’t need to save all of the summer energy). Hydroelectric dams store energy on timescales of decades. What we need to do is, first, establish regulations that dams can’t draw down reservoirs more than their time-averaged replenishment rate, over the course of a year. That is to say, however many million cubic meters of water flow in each year, that’s how many you can draw out in a year. But you don’t have to draw it out uniformly, as long as it averages out: You can run the hydro generators at full capacity during the part of the year when other energy supplies are low and demand is high, and not run them at all during the part of the year when supplies are high and demand is low. Which is what would naturally happen, with such regulation, because the operators of those dams would naturally want to sell their quota of energy at the highest price they can.

As for solar, don’t forget that in the best parts of the world for solar power, the major consumption of electricity is for air conditioning. And there’s a good correlation between when it’s sunny and when you need more AC. So supply and demand are automatically largely balanced, there.

You just couldn’t resist, could you?

As for AC vs. DC transmission, doesn’t AC also start seeing some problems when the length of the lines grows comparable to c divided by the frequency? At the few-tens-of-hertz frequencies used in most of the world, this is basically cross-continental distances.