By Matthew L. Wald
Forget what you’ve heard. Small isn’t beautiful, at least when it comes to energy. Small is ugly, dirty, and expensive. Clean, affordable and reliable energy has to be a collective enterprise, no matter how much that conflicts with the mythic American-cum-environmentalist ideals of self-reliance and going it alone, ideas that may have fit on the frontier but don’t translate well to urban and suburban life.
I am, inadvertently, running a real-world confirmation of this proposition.
Last summer I bought a cabin in the woods in upstate New York. The previous owner had installed an 11 kilowatt generator. My wife and I picked the place for a variety of reasons, the generator not among them. But now I’m in a neighborhood where many, perhaps most, of the houses have emergency generators, because our supplier has a lot of “service interruptions.” (I’ll identify the company only as the Friendly Local Utility, FLU for short; names have been changed to protect the guilty.)
The generator kicks in at least once a week, but in stormy weather, several times a day. We could handle an interruption of a few minutes or a few hours, but the region is prone to storms and FLU may be out for days, which is another matter. Especially since the well and the septic system are essential, and they both have electric pumps.
And herein is an electric equivalent of the Tragedy of the Commons. When everybody has access to a public resource, people tend not to care for it. In another way, it resembles the plight of public education; people with money can insulate themselves from its troubles, and those who remain suffer. In the electricity case, people who can afford a private back-up will pay for that, and the impetus to strengthen the common system declines.
It’s a tragedy for everybody because small is ugly and the back-up is worse.
The Trend to Go It Alone
My decision to bail (or, more properly, the decision by the previous owner of the house) is hardly unusual. In August, the Washington Post reported that in 2002 only 0.63 percent of houses had emergency generators, but now that number is 5.77 percent, and rising. This counts installed systems. Uncounted are the portable units on a contraption like a wheelbarrow, that require more gasoline every eight hours or so, that require the homeowner to go out and start it up, often in the dark, rain or snow, and then string extension cords through the house windows, to the refrigerator or other critical equipment.
On the higher end are permanently installed units with an automatic cut-over switch. These start up on their own within a few seconds after the utility stops supplying electricity, and can shut themselves down when the power comes back on. They often run on natural gas, which should be permanently available, or liquid propane gas, which rural houses usually keep in tanks that store hundreds of gallons, meaning days and days of supply.
The permanent models are a little bigger than a whole-house air conditioner, and a lot noisier. The portable models are louder still, whine like a gasoline mower, and change pitch when the load changes.
Either solution can make life a lot more comfortable, at least for the family that has one, but there are drawbacks for the world as a whole.
How does my little power plant stack up? How does it compare in cost, and cleanliness, to the electricity that FLU supplies to run the cabin for 99 percent of the time?
Precise comparisons are difficult, but the short answer is, not well. We are better off relying on the power grid whenever we can, and the world would be a cleaner, more prosperous place if the high voltage grid and the distribution system were reliable enough that people with a little extra cash didn’t invest it in private generators like mine.
The benchmark is the carbon footprint of a “marginal kilowatt-hour” on the New York State grid. Marginal because if my generator is running, and producing, say, 5 kilowatt-hours over the course of an outage, then the state grid is producing 5 kilowatt-hours less. The New York State Energy Research and Development Administration says the carbon footprint is 0.55 tons per marginal megawatt-hour. In household terms, that’s 1.1 pounds per kilowatt-hour. The margin means more than the average, because the average is pulled down by all the zero-carbon resources — the state’s hydroelectric plants and the nuclear plants — that former governor Andrew Cuomo didn’t kill. Those are running close to 24/7, and are not on the margin. Ditto, the state’s limited solar and wind generation are running whenever they can, so anything above that is met with natural gas.
Burning a gallon of propane releases about 12.6 pounds of carbon dioxide. According to the owner’s manual, my generator, a Generac, consumes 1.97 gallons per hour at full load, and 1.22 gallons per hour at half load. It’s significant that efficiency is lower at part-load; it’s another reason that small is ugly. In a system that teams together multiple generators, most can be run at full load, which is their highest efficiency, and only a few have to run at part load, to vary their output every time somebody turns a vacuum cleaner on or off.
My cabin’s load is certain to be variable, not full, because the well pump and the refrigerator cycle on and off over the course of the day. The generator’s production is unmetered, and I don’t track how many hours it runs, so I can’t say with certainty how heavily I’m loading the generator on average. But for a back-of-the-envelope calculation, I’ll assume it averages half load. That makes it about 6.7 pounds of carbon dioxide per kilowatt-hour, or roughly six times the carbon as the marginal kilowatt-hour from the grid.
This doesn’t count the diesel burned by the big tanker truck that delivers my fuel a few times a year. Or the losses in the transmission system between the utility generator and my house. (The emergency generator sends its energy about four feet to get to my house.) But these are minor factors.
The efficiency is low because the generator is an internal-combustion engine, which is inherently less efficient than the standard on the grid, called a combined-cycle generator. These generators are a combination of a turbine engine, resembling the jet on an airplane, and a second system that takes heat out of the exhaust to boil water and uses it to spin a steam turbine, and make more electricity.
In fact, I could compare my generator to a propeller plane of 60 years ago, and the grid’s marginal generators to a jet engine. My generator has two cylinders, with spark plugs, and is about as powerful as the engine on a modest motorcycle. Utility-sized turbines (which were originally derived from jet airplane engines) squeeze the fuel-air mixture tighter, and burn it at higher temperatures, to get more work out of the same unit of fuel.
Propane has 91,500 BTUs a gallon, about 80 percent as much as in a gallon of regular gasoline. If my generator is giving me 11 kilowatt-hours per 1.97 gallons of propane, then I’m putting in 180,255 BTUs per hour, to get 37,543 BTUs worth of electricity, for an efficiency of around 21 percent. The utility plant has an efficiency closer to 60 percent. And carbon dioxide production is directly proportional to fuel consumption.
What does it cost?
Utilities figure their cost as the total of capital plus maintenance plus fuel, plus a little bit for transmission and distribution. I pay FLU about 11.8 cents per kilowatt-hour, plus some odds & ends of fixed charges.
Here’s how those costs show up in my private solution:
To run my generator, buying the 1.22 gallons of propane costs me about $2.75. That means the fuel alone costs me about 26 cents per kilowatt-hour.
Calculating the other factors depends in part on how many hours I run the generator.
According to the website of a contractor that installs these generators, 100 hours a year is a good rule of thumb. The two-cylinder, 530 cc engine needs an oil change and new spark plugs every year, and an occasional air filter. An electrician offered to do this work for $125.
Assuming an average of half load, it’s producing 550 kilowatt-hours a year. That would put the maintenance cost at about 23 cents per kilowatt-hour.
The contractor’s website says that the brand I have has an average lifetime of 3,000 hours. That’s only a little over 4 months (there are 8,760 hours in a year, but mine won’t run continuously; most often it runs for tens of minutes or a few hours at a time). If I get 100 hours a year out of the machine each year for 30 years, that’s 3,000 hours and 16,500 kilowatt-hours. The generator cost $2,850 new, at a local big-box hardware store, in 2016. So the capital cost is 17.3 cents per kilowatt-hour. If I had to buy a replacement today, I’d spend closer to $4,000, and the capital cost would run about 24 cents.
The total is about 66 cents per kilowatt-hour, or about 71 cents if I were installing the system today. If I use the generator more frequently, the maintenance cost per kwh declines, and if I load it more heavily, the fuel cost per kwh declines. But in round numbers, it’s five or six times more costly per kilowatt-hour than energy from FLU, and a lot dirtier.
On the other hand, my kitchen refrigerator-freezer uses about a kilowatt-hour a day, and 66 cents is a small price to keep the milk sweet and the beer cold. And how much is it worth to run the thermostat and the blower motor for the propane heating system? What’s it worth to be able to flush a toilet?
Residential customers are voting with their feet, driven by extreme weather, or fear of extreme weather. According to the Wall Street Journal, Generac reported late last year that demand was so strong that it couldn’t find enough installers or dealers to keep up. More recently, demand has receded.
Other Approaches
There are other approaches to emergency power. I have a friend who owned a plug-in Toyota Prius, which uses the car’s engine to charge the battery. He modified the car to be able to plug his refrigerator and home computer into the Prius. He figured it beat buying a generator, because the Prius always had gasoline in it, to recharge the car’s battery, and it came with a catalytic converter and a good muffler, two refinements that a portable generator wouldn’t have. His wife was afraid not only that he’d void the warranty on their car, but that Toyota would come and take the car away.
I liked his solution, but I’m not that handy.
Another possibility is solar panels and batteries. That would work fine for the short “service interruptions.” But when you really need the system, in winter, the sun is high enough in the sky here to wake up the solar panels only about 8 hours a day, presuming it’s not cloudy. My consumption runs about 25 kilowatt-hours a day. In a pinch I could get by with less, but it’s easy to go three days here without much sun. That would be four or five Tesla Power Walls, in the neighborhood of $9,000 each. Propane would have to get rather scarce and expensive before the batteries looked like a good deal to most consumers.
Most houses with solar panels don’t have batteries, and in a blackout, their solar panels would be useless to them and to everybody else; for a variety of reasons, including the safety of utility workers who are servicing distribution lines that are fed by rooftop solar panels, the panels can’t run when the utility service is down.
As my house has the energy equivalent of a private lifeboat, there is a deeper problem to consider here, the problem of sharing capacity. Electricity is like the dial tone on a landline or the availability of a cell phone connection: demand varies. But if FLU’s customers are left short at a peak time, they will be very unhappy. In some cases, notably in Texas, they could be dead. But building enough capacity for each utility to meet its peak demand is like buying a dining room table to accommodate everybody for the once every five years when it’s your turn to host Thanksgiving dinner. Borrowing some folding tables and chairs is a more realistic strategy.
And that is a big chunk of what the grid does. It allows sharing capacity, so that on a really hot day in New York, demand can be met because a generator fired up in Manitoba.
In almost all of the United States (but not Texas, Alaska, and Hawaii) system reliability is vastly improved by connections with neighbors. Alaska and Hawaii have a geographical excuse. Texas has blackouts.
This conflicts with an American principle (or at least, American myth) of personal independence and self-sufficiency. Hence the embrace of rooftop solar by both left and right. But when ideology and self delusion meet engineering reality, engineering generally wins.
Or Blame Your Sinful Self
There is a second myth about shrinking the impact on our planet: that it’s a matter of personal virtue. There are, broadly speaking, two different approaches to limiting the carbon-loading of the atmosphere and the oceans, and the climate changes that will follow. One is to treat each kilowatt-hour coming from the socket on the wall, each drop of water from the well or the municipal water system, and each gallon of gasoline from the neighborhood gas station, as precious, something to be used carefully, prudently and sparingly. And everything brought back from the supermarket, from food to laundry soap, should be chosen with an eye to its environmental footprint.
The other is to focus on the source, to worry about where the megawatt-hours of electricity are coming from, what method was used to raise the meat and vegetables, and to improve the sustainability of everything in the supermarket bag, including toilet paper to laundry detergent.
In other words, replacing an incandescent lamp with a light-emitting diode is a good step, but somebody has to look at the upstream, and replace the coal plant with a nuclear reactor. Hooray for eating less beef, but a more realistic approach is to improve the carbon efficiency of the food industry. Low-flow shower heads are a good step, but if we deplete the aquifers and get to the point where a lot of water has to run through a desalination plant, we’ll need one that operates without emitting a lot of carbon dioxide.
Personal virtue, in other words, is nice, but a lot of the critical decisions are made collectively. (You can’t decide on your own to build a reactor.)
Energy and Equity
And, as the current cliche goes, this is a first-world problem. There is an equity issue; I can afford the 66 cents per kilowatt-hour, and a lot of other people can’t.
So people like me are opting out, or at least setting themselves up in a way where they’re not dependent on the common system. An analogue is sending your kids to private school and de-funding the public ones.
The real solution is more reliance on the grid. The grid, although much maligned as “creaky” or “outdated,” is the mechanism for moving clean energy around. Yes, solar can go on my roof, with a lot of expensive batteries in the basement, but solar “farms” are far less costly. Wind is only practical on a large scale, and so is hydro. Nuclear energy, an essential ingredient of a low-carbon system, only comes in large packages. (A small modular reactor means up to 300 megawatts, which is 2,700 times larger than my 11 kilowatt generator. Even a micro reactor is in the range of thousands of kilowatts, and the most effective way to use one of those is to connect it to the grid, so that it can meet the needs of lots of users, and thus run at full power for as many hours as possible.)
By necessity, the decision to build a reactor rests with a lot of people: investors, regulators and local and state governments. But that understates the communal nature of the decision, because a reactor can’t exist in isolation. It’s part of an industry. It needs a supply chain of parts from certified suppliers, and fuel from enrichment plants and fabrication plants that serve a lot of reactors. It relies on university engineering programs at a lot of different campuses, because it needs intellectual infrastructure as well. Eventually it needs a communal solution to spent fuel disposal, either before or after reprocessing.
Just as the car industry relies on the fuel industry and the road industry, or a passenger jet requires a network of airports, air traffic controllers and trained pilots and mechanics, a nuclear reactor is part of an interconnected whole. Artisanal energy projects won’t cut it.
The same is true of electricity in general, it works better as a system.
The lesson here is that what I pay my Friendly Local Utility for a kilowatt-hour is cheap relative to its value to me, and a lot cheaper than the go-it-alone strategy. The clean and equitable solution may be to pay a bit more for grid power, to improve reliability and reduce the number of defectors, like me, and to preserve the system, which is the backbone of modern life.