How the gas turbine conquered the electric power industry
construction-physics.com
construction-physics.com
2. A gas turbine can be up and running in days
3. Green washing - it's "clean" energy for some definition of clean, certainly compared to coal
I'm exaggerating a bit on points 1 and 2, but the fuel costs are cheap and you can setup a gas turbine plant comparatively quickly compared to a coal or a nuclear power plant. Most people don't care about a gas turbine plant in their backyard, but a coal or nuclear power plant? You're facing a multi-year legal battle that's going to cost you millions of dollars. You have to do environmental impact studies that are going to cost you tens of millions of dollars. There's a lot of complications for how to handle the waste (not just nuclear: coal has SOX, NOX, fly ash and pot ash wastes that you must manage) that cost tens of millions of dollars. A gas turbine is quick, cheap, and easy. Guess which form of electrical power generation is proliferating?
Fast construction speed was an advantage even 30 years ago, and so is the lower number of full time employees needed to operate a CCGT plant vs. equivalent capacity from a nuclear or coal plant. But note that the turbine revolution arrived back when the US was still building coal and nuclear plants. They weren't built just for "green washing" reasons or for avoiding legal battles.
[1] https://www.eia.gov/dnav/ng/hist/res_epg0_r5302_nus_bcfa.htm
And if it's an illegal leak, maybe it needs some help to flare.
Its arguable what the correct timescale to consider it is. Personally I think 100 years is too long given methane's potential feedback effects. IE, warming causes methane to be released from Arctic tundra melt causing more warming, etc.
And there's also the issue that the higher the concentration of methane in the atmosphere the longer it takes to decay.
Now that it is closed, and there is no smoke - they can see not just the towers, they now see a long way behind them as well.
Best guess: you're confused by the word "smokestack". Smokestack is a synonym for chimney. The power station is closed. The chimneys are still there, but there's no smoke coming out of them.
It's very clean when compared to coal, 50-60% less CO2.
Gas emits a relatively low amount of CO2, but a much higher amount of methane, which is a potent greenhouse gas (easily 10x of CO2, 50x as much depending on how exactly you measure it (e.g. methane has a shorter atmospheric lifetime than CO2)). Measuring only CO2 here is pointless.
Err... no.
Natural gas is methane, to a good first approximation, but if your plant is emitting methane, it means your plant is not operating efficiently. Natural gas power plants burn methane, producing CO2 and water.
Edit: if you're talking about production leaks, coal mines emit methane, too, and there's not even a token amount of effort to stop it, nor any incentive to do so. With gas production, every cubic meter that leaks is a cubic meter you don't get to sell.
Edit #2:
https://www.epa.gov/sites/default/files/2021-04/documents/fa...
Note that coal mines are currently the 5th largest methane emitters. However, that would certainly go up a lot if natural gas were replaced with coal.
The number 1 methane emitter, by a fair amount, is still "enteric fermentation" (i.e., farts and burps).
Not too that this table does not include all the naturally-produced methane resulting from the normal rotting of vegetation and non-domestic animal manure. I would almost bet that exceeds even the enteric fermentation.
Peaking gas turbines are great compared to coal. They also can fail on the supply commitments. South Australia suffered disconnection from a national grid which forced shut-downs and then found the gas generators contracted to do black start couldn't do it for various logistical reasons. The contracted parties got sued into the ground later.
I love that Rolls Royce and Pratt&Whitney supply both this market and others from fundamentally the same technology as flight engineering. Obvious differences to mesh a power take off instead of a turbo fan but it's an RJ11 under the hood (for instance)
> By comparison, France’s much-vaunted nuclear power construction spree accounted for roughly 60 gigawatts of capacity between 1975 and 1990, and China built about 400 gigawatts of solar PV capacity between 2007 and 2022, though today China is building about 100 gigawatts of solar a year.
This does not capture the sheer size+speed of the PV revolution. We're at 800GW production of PV panels EACH YEAR. This will increase to 1TW by the end of 2024. For comparison, the whole world fossil fuel electricity capacity is around 5TW. Accounting for a lower capacity factor, solar could provide almost all of the electricity in just 15 years. This is shorter than the life expectancy of solar panels, so prepare yourself for some major disruption.
Panels will be cheap as dirt, we'll install them as fences, roofs, sidings, you name it. Electricity during daytime will be free. We'll see some changes to the structure of electrical companies: they will charge a higher subscription to connect to the grid, and much less per KWh. Net metering is obviously going away.
> Panels will be cheap as dirt
If you don’t consider the capital of a panel then sure free / cheap as dirt.
Then the sun goes down and the wind doesn’t happen to blow, so you fire up your gas turbine.
That's what I meant, the capital cost is already so low that you could use them as fence panels. They will only get lower with the looming supply glut.
Look at this monocrystaline (higher eff) panel in china: 15 cents per watt. There are cheaper ones too: https://www.alibaba.com/product-detail/Haitai-Solar-Panel-Pr...
A 400W panel will cost 60 dollars and has a size of 1.8mx1.8m. Per square meter that is $18.52
A generic plastic fence panel from lowe's 6ft by 8ft costs $120 https://www.lowes.com/pd/Freedom-Actual-6-ft-x-7-82-ft-Ready...
price per square meter: $26.87
You're glossing over the inherent danger of solar panels in that they are relatively high voltage devices that can't be turned off easily. This means they get saddled with expensive safety regulations. The glorious utopia where you buy a pallet of the things from Wal*Mart with your pocket change and stick them everywhere the sun shines runs into some logistical hurdles.
> they are relatively high voltage devices
On the contrary, they are low voltage (12V), which makes them more dangerous (because you need higher amps to carry the same amount of energy). To make them safe, you can install inverters closer to the panel, to raise the voltage to 110 or 220, making them much much safer.
> The glorious utopia where you buy a pallet of the things from Wal*Mart with your pocket change and stick them everywhere the sun shines runs into some logistical hurdles.
That can work if you install them as fences, or just on the ground in general. If you install them on rooftops then yeah, you need some serious consideration. The installations are still grossly overpriced, if both EU and Australia can do it under 1USD per Watt, so can the US.
Lets check a random common panel.
https://www.solar-electric.com/rec-solar-rec400aa-pure-alpha...
Voc: 48.8 Volts
And the original post was talking about cheap installs, so you would string several together to put it on an inverter. You hit high voltages very quickly. Even just 4 panels and you're up to almost 200V at full sun. It's more typical to put 8-12 panels on a string. The voltages hit dangerous levels very quickly.
You want to tie them in series first, to increase the voltage and keep the amps as low as possible. 110V or 220V are not dangerous voltages, that's why houses are wired at these higher volts. People all over the world handle cables carrying 220V with very few issues. If you wired the house at 12V-48V, some wire in the wall would catch on fire every-time you wanted to toast a slice of bread, lol.
They are typically first wired in series to increase the voltage as early as possible, to make them safer. If you were to wire them in parallel you would keep a low voltage and increase the amperage, which would make their wires glow like Christmas lights.
That's the whole reason why they started selling micro-inverters that go on each panel: to increase the voltage and decrease amps before even leaving the panel area, to make them safer.
Look at a typical micro-inverter: input is up to 50 volts, 10 amps. Output is 240 volts, 1 amp. https://solartown.com/solar-products/enphase-iq8-microinvert...
For panels, the voltage is not dangerous part, the intensity is.
If they were actually 12v they would be safer. You could handle them with dry skin. It's the same reason people aren't electrocuted while working on cars. The danger would be starting fires as you noted.
Anyway, in the real world when you have panels wired in series you typically need to shield the connection wires in conduit because both the voltage and amperage reach dangerous levels and you can start fires or kill people if they try to handle the wire, and you can't easily turn them off. This is part of the expense of installing solar panels that you can't really avoid. You will always need someone who knows what they are doing to handle the electrical part or you will kill people.
https://news.ycombinator.com/item?id=38304405
Add in the large number of storage options that are coming to market besides lithium ion, from sodium to iron to straight up thermal storage, and we may still have gas turbines on the grid in 20 years but it's unlikely they will generate more than a few percent of a year's electricity.
We are seeing the first disruption of the electricity industry in a loooong time and it's going to upset all other energy processes too, as electrical sources of process heat become cheaper than burning gas or anything else.
Looking at where prices are today misses the picture, look at where they will be in five years or ten years in order to understand our future.
Some of this is people (usually required to by the installers) choosing expensive solutions like microinverters or optimizers that add a high fixed cost to each panel, even though both technologies offer very little utility to the customer. The old talks about how one shaded panel would take an entire array out are out of date and just plain wrong these days. The other is that some of the costs like permitting and interconnection are just as hard or even harder as they always were, especially as power companies get nervous about the number of people installing solar panels.
Enphase IQ7-60 is $85 on Amazon right now.
I have to disagree with you there... they're not that expensive, and they eliminate the most common single point of failure in the system.
When I installed mine last year, I got quotes from every major supplier: I paid a roughly 10% premium for panels with microinverters over a traditional system.
We're talking about extending the time it takes for panels that will work for 20+ years to pay for themselves by a few months, in exchange for a massive increase in redundancy: that was an obvious decision from my point of view.
Power companies already buy much of their electricity from independent power generators. However, for the economics to work, producer prices need to be lower than retail prices. These prices change throughout the day, based on demand.
Net metering results in people getting paid the retail price instead of the producer price.
A decentralized power system is going to have a lot of battery backup and that might even be more important than the solar panels?
The article explains why many years have passed from the development of efficient steam turbines until the development of efficient gas turbines, due to the differences between the Rankine Cycle and the Brayton cycle.
Even if the Americans like to name the Joule cycle as the Brayton cycle, the American name does not have any justification.
While George B. Brayton has patented his engine in 1872, James Prescott Joule has published already in 1851, 21 years earlier, a scientific paper (“On the Air-Engine”) describing what is named now as the Joule cycle, a.k.a. the Brayton cycle.
While the Brayton patent contained very little information, the paper published by Joule was very important and it described in great detail how to design an engine using this thermodynamic cycle and why this is useful.
Moreover, in 1859, 13 years before Brayton, William John Macquorn Rankine has published a very influential manual (“A Manual of the Steam Engine and other Prime Movers”), where all the thermodynamic cycles used in engines known at that time were classified, and he already referred to this cycle as Joule’s cycle.
Therefore there is no doubt about the priority of the term "Joule cycle" over "Brayton cycle".
An interesting fact that is usually not mentioned in most manuals is that at equal maximum temperature and maximum pressure (which are typically limited by the materials used for the engine), the Joule cycle is more efficient than either the Atkinson cycle or the Otto cycle, so the fact that it is the easiest thermodynamic cycle to approximate in a gas turbine is favorable for its efficiency.
I found the preceding argument more compelling.
Brayton needed to solve actual hard core engineering problems to make the device practical, thus it wasn’t the same idea only similar. There’s little comparison as Brayton was selling commercially viable engines vs writing a paper on a much older idea.
Europeans have the biggest ongoing case of butthurt I've ever experienced. Is it some kind of inferiority complex because the US is a superpower?
However I care deeply about historical truth and about remembering to whom we are indebted for the things we use every day.
In this case there is an indisputable temporal priority, as the term "Joule cycle" had already been in use for a long time before Brayton invented his engine.
Even if the priority would not have been so evident, between two people, one who has made some money by building and selling a thing and a second one, who has taught everybody how to build such things and why they should build them, I believe that the reasons to remember the second are much more important, even if it is not unusual for the American spirit to honor more the first.
Language and history is messy. If you don't accept this you'll spend your life tilting at windmills in the most pointless way.
Another example, Von Neumann did not invent the architecture named after him. It's named after him because he wrote a very influential report on ENIAC.
His papers are based on observations and measurements of heat, force, cooling, etc based on comparing multiple engine design types .. at scale and with models.
> FWIU, heat engines are useful with all thermal gradients: pipes, engines, probably solar panels and attics; "MIT’s new heat engine beats a steam turbine in efficiency" (2022) https://www.freethink.com/environment/heat-engine
"Thermophotovoltaic efficiency of 40%" (2022) https://www.nature.com/articles/s41586-022-04473-y https://scholar.google.com/scholar?cites=1419736444024563175...
"Capturing Light From Heat at 40% Efficiency, NREL Makes Big Strides in Thermophotovoltaics" (2022) https://www.nrel.gov/news/program/2022/capturing-light-from-.... :
> The 41%-efficient TPV device is a tandem cell—a photovoltaic device built out of two light-absorbing layers stacked on top of each other and each optimized to absorb slightly different wavelengths of light. The team achieved this record efficiency through the usage of high-performance cells optimized to absorb higher-energy infrared light when compared to past TPV designs. This design builds on previous work from the NREL team.
> Another crucial design feature leading to the high efficiency is a highly reflective gold mirror at the back of the cell. Much of the emitted infrared light has a longer (less energetic) wavelength than what the cell's active layers can absorb. This back surface reflector bounces 93% of that unabsorbed light back to the emitter, where it is reabsorbed and reemitted, improving the overall efficiency of the system. Further improvements to the reflectance of the back reflector could drive future TPV efficiencies close to or above 50%.
Thermoelectric effect: https://en.wikipedia.org/wiki/Thermoelectric_effect
Thermophotovoltaic energy conversion *: https://en.wikipedia.org/wiki/Thermophotovoltaic_energy_conv...
Thermophotonics: https://en.wikipedia.org/wiki/Thermophotonics
Gas turbine: https://en.wikipedia.org/wiki/Gas_turbine :
> gross thermal efficiency exceeds 60%. [100] (2011)
GE-7HA https://www.ge.com/news/press-releases/ha-technology-now-ava... (2017) :
> that its largest and most efficient gas turbine, the HA, is now available at more than 64 percent efficiency in combined cycle power plants, higher than any other competing technology today.
How do TPV operating and lifecycle costs differ from gas turbine's costs?
An electric motor's efficiency is not necessarily the same as its generator efficiency in reverse.
Gravitational Potential Energy
CAES Compressed Air Energy Storage:
Solar thermal energy > https://en.wikipedia.org/wiki/Solar_thermal_energy :
> Electrical conversion efficiency: Of all of these technologies the solar dish/Stirling engine has the highest energy efficiency. A single solar dish-Stirling engine installed at Sandia National Laboratories National Solar Thermal Test Facility (NSTTF) produces as much as 25 kW of electricity, with a conversion efficiency of 31.25%. [66]
Szilard-Chalmers MOST process: https://news.ycombinator.com/item?id=34027647 ...18 years at what conversion efficiency?
Combined storage + solar/wind I believe is currently at LCOE parity with natural gas.
With Sodium Ion batteries coming online for storage (no lithium, no cobalt, no nickel), the writing is on the wall for fossil fuels.