I wrote previously that coal is just not that hot of a fuel. Compared to natural gas, coal isn’t logical for electric generation, and getting less and less logical with time. Meanwhile, solar power is getting cheaper every year, with grid parity on the way in a few short years (maybe one). In some places, parity has already been reached. Coal pushers still try to tell us that they’re the reliable, American energy. The claim is that solar (like wind or other renewables) varies, and cannot keep power grids stable and consistent in the face of demand. What they conveniently forget to tell you is: Coal isn’t too hot at that, either.
As solar prices keep crossing parity, along with the rest of the portfolio (gas, nuclear, hydro, wind, etc.), there will soon be no credible pitch for coal. My advice: get out. If you’re exposed to coal financially, sell. If you’re actually looking at a coal seam, get out via the nearest exit ASAP, and find a new career. Of course, nothing ever really goes away, as I’ve stated multiple times. There’ll be a transition period, and even after that the grid will cling to legacy plants. However, there is no coal growth on the horizon; it’s a matter of coal staying level or contracting, and at what contraction rate.
Electricity use varies through the day, and today’s grid does “load following” to match supply with demand. In a few situations, usage ramps up strongly, and utilities call on “peaker plants” to fulfill demand. It is thus desirable for generation to be throttleable, or in grid terms “dispatchable.” This holds when you can’t find ways to smooth out usage (“peak shaving”), or store electricity on these large scales. Coal lobbyists then claim they are dispatchable, but no, coal really isn’t.
Coal plants operate on the “Rankine Cycle,” named after William Rankine. In this power cycle, a burner is used to heat water; the hot water then turns turbines. After driving turbines to make electricity, the water is run in a radiator of some sort. The now cool water goes back to the burners for another loop. In other words, these plants make water heat up, then cool down. The more hot water, the more power.
Natural-gas plants, however, mostly operate on the Brayton Cycle (after George Brayton). In this cycle, the gas burns with outside air, of course, just like coal burns. But then, the exhaust itself turns a generator. No liquid water is involved (except maybe some incidental steam in the exhaust gases).
Since this never has to boil any water, it can start up and yield power far faster. Just picture boiling a pot of water on a stove, or turning on a hair dryer- which one’s faster? A _lot_ faster? The hair dryer, by a huge margin. Similarly, a natural gas Brayton plant can be dispatched a lot faster than an equivalent Rankine plant, burning coal, oil, or even natural gas. Sure, a gas turbine has to warm up its own metal. But so does any Rankine plant- a Rankine has metal burners, water pipes, etc. that also soak up startup heat.
Natural gas “plants” are also smaller than coal ones, letting them heat up even faster. To cut costs, coal plants got bigger and bigger with time. The same number of parts, making more power, make a big coal plant more cost-efficient than a small one. But this hurts startup even more: fewer, bigger coal plants take longer to warm up and start producing. In contrast, natural gas sites use several, much smaller turbines. This is partly since turbines are developed with the jet people, and jets hardly fly epic-sized turbines. Small gas turbines then start far faster than coal plants warm up. Individual gas turbines can also be dispatched singly, versus one big coal plant, which must commit to firing up or staying cold.
Neither cycle can match hydroelectricity, of course. Dams need no warmup since they don’t use a heat cycle at all. Throttling dams up or down is just throttling water flow; this athermal process is as fast as the valve is. When it comes to dispatchability, coal isn’t even close.
It’s true that a Brayton-cycle turbine has to spin up, taking time. A gas turbine may spin at hundreds of thousands of RPMs, far faster than a piston engine or the turbine in a Rankine plant. But you know what? It doesn’t have to be this way, either.
We are now seeing the deployment of natural gas power “plants” with no water loops or turbines- in fact, arguably no moving parts at all. Direct natural gas fuel cells take fuel and combine it with air directly to make electricity. There are no turbines to spin up; there is no water to heat up. The power cycle is, technically, athermal. (In practice, switching off a fuel cell for too long causes its temperature to fall too low.) For that matter, there’s also less metal in a well-built fuel cell. It too gets to operating temperature much faster than Rankine plants.
Fuel cells, like gas turbines, are smaller. Natural-gas fuel cells, the size of diesel backup generators, are sited by server buildings, hospitals, or other key infrastructure. In addition to fast starts and dispatchability, such sources put power directly in places that use it, cutting grid costs and losses. Without long-distance lines, onsite assets save grid costs and boost reliability. Try that with coal; you’ll get soot and stench complaints before the first bill arrives.
Natural gas power is not only growing, it’s been growing for over twenty-five years now. As I’ve stated before, Wall Street issues favor natural gas, since individual, small turbines finance and install gradually- not one, big coal project. The interest costs favor many, small upgrades, not one mega-debt. In the worst case, you simply order an emergency gas turbine, which comes on a trailer. You haven’t made an investment at all, you’ve contracted “peaker plant services” from generation firms. It is then a big coincidence that solar power has been installing as the grid has been getting more and more dispatchable. A big, fortunate coincidence: the one-two punch of solar plus natural gas will be a knockout for coal growth.
Of course, this is assuming intermittent solar is a problem in the first place. One solar panel or array doesn’t simply power one site, and only that site. Homes, shops, offices, etc. connect to a grid, and a grid connects generation assets in its entire service area. An interconnection then spans many power utilities, in several states. A weather front, meanwhile, is about the size of a state. As a front passes by, solar assets on either side send power through the interconnection, including across the front. To a lesser extent, neighboring interconnections (the greater grid) sell power to each other. In cases of high demand, your electricity comes from generation capacity in your state, in neighboring ones, and further down the line by one or two more states. One weather front is almost not an issue; one cloud is not worth discussing.
Notice I said “generation,” not “solar arrays.” Grids connect dirty and clean generation alike. On the grid are solar arrays in different places, any one of which may be clouded at any given time. Also on that grid: wind turbines, which will be affected differently. Some weather patterns mean low sun, but high wind; in some areas, winds blow strongly at dawn and dusk. Other areas have stronger winds at night; there are also local variations (“microclimates”). This is before we even bring up other sources: nuclear, hydropower, and oil or biomass. As a troublesome weather situation is forecast (36-48 hours in advance), utilities simply call in natural gas turbine trailers.
Overall, solar clouding is simply not an issue at present levels. It won’t be, until solar passes the level of, at the very least, hydro (several percent of our national power portfolio), and more likely the level of nuclear (~20% of national generation). By that time, we will have options.
Even this is assuming that dispatchability is the one, shining goal. It isn’t, not by a long shot. There are two ways to make supply and demand for electricity meet: manage supply, and manage demand. I’ll talk about demand shaping. We can also manage supply through energy storage. Although grid-scale storage of electricity isn’t a big factor now, it will be…