David Foster: “but much of the actual generation will wind up being done by natural gas”

According to Wellinghoff: “Natural gas is going to be there for a while, because it’s going to be there to get us through this transition that’s going to take 30 or more years.”

After that, you will be on your own. I am suggesting diesel generators will be the new fashionable accessories, like sub-zero refrigerators.

Jay Manifold: Let’s compromise on a one syllable word like nuts.

I disagree. I think it’s Brobdingnagian. Or perhaps Cyclopean.

I think what is really likely to happen is that a lot of capital is going to be deployed to wind and solar…but much of the actual generation will wind up being done by natural gas. This will, of course, increase electricity prices sharply. (It may also create investment opportunities in nat gas.)

Thanks for the link to Jevon’s paradox. I’ve been trying to explain to people why increasing efficiency of technology doesn’t work.

“the United States can reduce energy usage by 50 percent.”

This is the Lovins fallacy. Greater efficiency in using energy does not cause the total amount of energy used by an economy to decrease. The truth is Jevon’s Paradox: “It is a confusion of ideas to suppose that the economical use of fuel is equivalent to diminished consumption. The very contrary is the truth.”

Problems with unsteady power generation from wind will be overcome, he said. “That’s exactly what all the load response will do, the load response will provide that leveling ability, number one,”

Load response is his way of telling us that if the wind isn’t blowing, the digital grid will turn off your stuff to keep from collapsing. Better have a UPS and a diesel generator.

he said. “Number two, if you have wide interconnections across the entire interconnect, you’re going to have a lot of diversity with that wind. Not all the wind is going to stop at once. You’ll have some of it stop, some of it start, and all of that diversity is going to help you, as well.”

This is pure hand-waving. It sort of makes sense, but you must remember, the weather is the product of systems that span the entire continent. The likelihood is that the winds are correlated.

It is interesting to noodle through his prophecy. But it is best to start with some facts.

GE, it should not surprise you, makes wind turbines. Their on shore 2.5 MW model is the apple of their eye. It has 100 meter long rotors. It’s a big boy.

4 of them can generate 10 MW if the wind is between 28 and 56 mph. And 40 can generate 100 MW, and 400, 1 GW. 100 GW would require 40,000 of the pups.

How far apart do they need to be? This document from Australia says 3 to 5 rotor diameters across the wind (N-S in the Great Plains) and that the rows need to be 5 to 7 rotor diameters apart. Trying to work this and be generous to the case for wind energy, I take the rotor diameter of the GE 2.5 MW turbines as 200 meters. I would guess that they need to be spaced about 0.5 km N-S and 1 km E-W.

In the Great Plains it is about 2000 km from Mexico to Canada. 2000 km will host a N-S line of 4000 machines. Its nameplate capacity would be 10 GW. To get 1 TW there would have to be 100 rows spaced over 100 km. 200,000 km^2 is about the size of Nebraska. This is not a big project, it is pharaonic.

And how much would it cost? I don’t know precisely, and I don’t think anybody does. My research tended to indicate that the generators themselves could be had for $1 to $2 per Watt, installed on dry land. (Anything in the water will cost a lot more). But that does not include things like land, service roads, and auxiliary buildings. Nor does it include any type of energy storage.

As you say, Wellinghoff is either postulating magical fairy wind generators that would have a combined peak power output approaching that of every coal, natural gas, and nuclear generator in the entire US (900 GW), or else he’s got his units and his math so messed up that you can’t tell what he’s claiming.

So I repent of my suggestion that your math might be worse than his. Such an accusation was beyond the bounds of reasonable discourse. :-)

I did in fact mean 4.1 trillion kilowatt hours (4.1 petrawatt hours) and all my linked sources are in watt hours. I dropped the hours because I had previously been doing some calculations in Quads (quadrillions of btu’s) and I dropped the hours of the terawatt hours by a kind of unit sympathy. That used to bedevil me in the lab all the time in college.

The problem here is that the kilowatt i.e. one thousand (10^3) watts is the basic unit of measurement of electricity generations and consumption. So the link to the Energy Information Agency shows that American consumes 4,157 billion kilowatts hours. So, tidying that up that comes to (4.1*10^3) * (1.0*10^9) * (1.0*10^3)=4.1*10^15 watt hours.

For simplicity sakes it would be best to use kilowatt hours of kwh as the base unit. In that case, the U.S. consume 4.1 trillion kwh per year which we could write 4.1 Tkwh.

I assumed that Wellinghoff was speaking in watt hours because that is the standard measurement. If so then his maximum claims for wind generation would be 1,000 billion kilowatt hours or (1.0*10^3)*(1.0*10^9) * (1.0*10^3)= 1.0*10^15 watt hours (1.0 Tkwh). That translate to 1 trillion kilowatt hours per year of roughly 1/4 of present consumption. So my original argument remains.

Wellinghoff might have been talking about Declared Net Capacity or Nameplate capacity both of which are based on the theoretical output of a generator system. The U.S. currently has 1,087,791 megawatts of nameplate capacity which translates in 1,089 gigawatts of name plate capacity. So perhaps that is what Wellinghof is claiming that windpower could produce up to 1,000 gigawatts of name plate capacity which would be almost enough to provide our current power needs.

However, name plate capacity tells us nothing about how much power you actually get out of wind generator. With fossil fuels, hydroelectric or nuclear the operator can determine how much power comes out of the generator. With wind power, the weather determined out put which cam be anywhere from zero to the rated power. Most people don’t even try to give a name plane or DNC for windpower because of it unreliability.

For example, windpower currently has 1.08% of the U.S. nameplate capacity but it produces only 0.07% of current U.S. electricity production. Nuclear power by contrast has 9.8% of U.S. nameplate capacity but produces 19.4% of our power. Windpower underperforms due to it unreliability and nuclear power over performs because of its rock solid reliability.

So, in short, Wellinghof is smoking crack either way.

Math is hard.

Setbit…it’s hard to believe that peak (total nameplate) capacity is only 2X average 24-hour-a-day use:

Yeah, that seems counterintuitive to me as well, but both the links I gave and the one Shannon gave say basically the same thing: a little over 4000 GW hours total energy generation and around 1000 GW total power capacity. Not knowing exactly how those two numbers are arrived at, I can’t say if it’s an apples-to-oranges comparison, but the orders of magnitude are consistent.

The per-source subtotals line up pretty well, too. For example, nuclear and natural gas constitute similar percentages of total energy production, but natural gas has about 4 times the peak power capacity of nuclear. That makes sense when you consider that nuclear plants run at or near capacity as much as possible, while natural gas plants spin up and down as needed.

I suggest that there is already one metric we can look at: the price of electricity where we live. If it’s cheap, brownouts are unlikely. If it’s expensive, things may be a lot closer to going over the edge.

I also suggest that we vote with our feet and be prepared to move to places not yet put in jeopardy by the techno-illiterates. Notwithstanding my suggested criterion above, these may not always be easy to identify. I think the authorship and readership of this blog is more aware than most that there is no truly pro-science political party in the US, and that “intellectual,” for the politically active, rarely includes a significant mathematical component.

From AER (linked above) table 8.2a, Total Electricity Net Generation for the US in 2007 was:
4,159.5 Billion Kilowatt hours = 4.1595 * (10e3) * (10e9) * (10e3) Wh = 4.1595 * 10e15 Wh
So, Setbit is correct call it 4.2 PWh.

Dividing the 4.2 * 10e15 Wh by the 8760 hours in a year we get 0.47 10e12 W = 475 GW of generation. The peak summer capacity for generation at ~1 TW is given by Table 8.11b.

http://en.wikipedia.org/wiki/List_of_countries_by_electricity_consumption

1)diurnal curve–the 24hr number includes midnight to 8 AM when power demand is much lower
2)seasonality–heavy seasonal influences, especially for air conditioning in summer
3)regional factors–nameplate capacity in Georgia doesn’t do any good for demand in California
4)mainenance–units have to be taken out of service occasionally for maintenance, and sometimes just plain break…

http://www.eia.doe.gov/emeu/aer/pdf/pages/sec1_5.pdf

shows 101.6 X 10^15 BTU consumed in 2007. Convert to TWH and divide by the number of hours in a year: 3.399 TW.

Right now, the U.S. consumes roughly 4.1 terawatts of electricity and heat from electricity generation.

Check your units, here. I think you biffed it.

Renminibi and Robert are right, I think. Per the lower right-hand number on this page (http://www.eia.doe.gov/emeu/aer/txt/ptb0802a.html) the US generated some 4,160 billion kilowatt hours (i.e. 4,100 terrawatt hours, i.e. 4.1 petawatt hours) of electric energy in 2007. Dividing by 8,760 hours in that year gives overall average power generation of a little under 500 gigawatts.

Obviously peak capacity has to be higher than that 500 GW average. This page (http://www.eia.doe.gov/cneaf/electricity/epa/epat2p2.html) says that nameplate power capacity was a little over 1 terawatt in that same year.

I still think you’ve got an article here; I expect Wellinghoff’s numbers are delusional. But unless you’ve got some completely different data source that’s better than the DOE I think your numbers are junk as well.

Could it be that your math’s worse than the Obama administration’s? Man, that’s just sad.

Annual Energy Review 2007
Report No. DOE/EIA-0384(2007)
Posted: June 23, 2008
Next Update: June 2009
Table 8.2a Electricity Net Generation: Total (All Sectors), 1949-2007

If you had plants running 24*365, you would need about 500 GW of generating capacity. That doesn’t happen, and the total generating capacity is almost a TW.

I cannot tell from the article, but when knowledgeable people discuss wind energy installations, they are usually referring to the “nameplate capacity”, i.e. the productive capacity of the machine, if it run under optimal conditions. Of course optimal conditions occur less than full time, and most wind turbines yield about a third of their capacity.

The inconstancy of the wind is a byword. The sun sets every evening. Any of these systems needs back-up by fossil fuel generators, mostly natural gas. They also will need systems to store and redistribute energy. All in all we are talking about an incredible expense. Nuclear is far cheaper, and uses far fewer resources.