December 28, 2006
Standby Generator Costs Limit Wind Power

Writing for the New York Times Matthew Wald examines the economics of wind power.

He said that in one of the states the company serves, Colorado, planners calculate that if wind machines reach 20 percent of total generating capacity, the cost of standby generators will reach $8 a megawatt-hour of wind. That is on top of a generating cost of $50 or $60 a megawatt-hour, after including a federal tax credit of $18 a megawatt-hour.

Note that a tax credit on one party is a tax on another party. So that wind tax credit is not free and causes market distortions. Though other energy sources have their own external costs that cause market distortions.

By contrast, electricity from a new coal plant currently costs in the range of $33 to $41 a megawatt-hour, according to experts. That price, however, would rise if the carbon dioxide produced in burning coal were taxed, a distinct possibility over the life of a new coal plant. (A megawatt-hour is the amount of power that a large hospital or a Super Wal-Mart would use in an hour.)

A few things to note here. Take the $18 per megawatt hour US government tax credit away from wind and it costs from $68 to $78 per megawatt-hour plus another $8 per megawatt-hour for standby capacity coming from other electric power sources. That puts it at double or more the cost of coal electric.

But the economics above understate the problem with wind. Suppose we shift to dynamic pricing of electricity (which we should btw) so that the price of electricity varies as a function of demand and supply. Electricity would cost more at 2 PM on a hot summer day than at 2 AM on a cool fall day. Well, wind tends to blow when electric demand is lowest!

In many places, wind tends to blow best on winter nights, when demand is low. When it is available, power from wind always displaces the most expensive power plant in use at that moment. If wind blew in summer, it would displace expensive natural gas. But in periods of low demand, it is displacing cheap coal.

The wind power industry wants a far more sophisticated electric power distribution grid so that wind electric can get carried from wherever the wind is blowing to wherever it is not blowing. Some industry analysts are skeptical about the feasibility of such an undertaking and whether it would even work since we could have weak wind days over a very large area. I wonder what it would cost.

Curiously, wider usage of wind power would favor coal over nuclear. Why? Coal has a larger variable cost than nuclear because coal as fuel is a larger fraction of total coal electric cost than uranium or plutonium is as a cost for nuclear electricity. In a nutshell, nuclear plants have the highest capital cost but the lowest fuel cost. Next comes coal and then finally natural gas. Natural gas electric plants cost the least to build but have the highest fuel cost. So they are used for peak power. Wind is so unreliable that natural gas plants probably would cost too much as back-ups to wind and therefore coal would be the best back-up for wind.

Nuclear, by contrast, works best as baseload power. Nuclear plants cost so much to build and save so little in operational cost when idled that once a nuclear plant gets built it makes sense to run a nuclear plant continuously 24x7.

Photovoltaics (if only they didn't cost so much) have far more favorable supply characteristics as compared to wind. They produce the most electricity during summer days when demand is highest. Though they are far from perfect. First off, in the northern hemisphere (and a similar problem occurs in the southern hemisphere just 6 months out of phase) the hottest days are in July and August and yet the longest day of the year (when the most suns shines to generate the most electricity) is in late June. Also, electric power demand does not peak at high noon. As the day heats up people turn on more air conditioners into the afternoon as the sun is past its peak and into the evening when people go home and turn on air conditioners, TVs, computers, and assorted home appliances. Solar's output peak does not match the market's demand peak for electricity.

Wind (and solar and nuclear) economics would improve if a carbon tax was levied on coal and natural gas burned to generate electricity. But coal would still retain a large cost advantage even with a hefty carbon tax.

The economics of wind would change radically if the carbon dioxide emitted by coal were assigned a cash value, but in the United States it has none. Coal plants produce about a ton of carbon dioxide each megawatt hour, on average, so a price of $10 a ton would have a major impact on utility economics.

I've read estimates of the cost of full carbon dioxide sequestration of about 2 cents per kwh or $20 per megawatt-hour. That'd still leave coal cheaper than wind. Though full carbon sequestration would probably make nuclear cheaper than coal (see Phil Sargent's links in the comments).

When comparing between wind and coal the wind tax credit is economically similar to forcing coal burning utilities to do full carbon sequestration on coal in the sense that the wind tax credit narrows the gap between wind and coal by about the same amount as the cost of carbon sequestration. However, the wind tax credit does not cause a big shift in demand away from coal because wind costs too much. An elimination of the wind tax credit combined with a requirement for full carbon sequestration would cause a partial shift away from coal toward nuclear and would eliminate the economic argument in favor of wind.

The wind tax credit currently causes a small reduction in demand for nuclear power. How? To the extent that wind farms get installed the effect is to increase the demand for back-up power sources which are cheaper when not used all the time. The back-up power is needed for when the wind does not blow. Since coal plants cost less than nuclear plants they are cheaper as back-up power for wind.

Note that the relative cost of nuclear, coal, wind, natural gas, and other electric power sources varies within the United States and even more globally. For example, in the Middle East natural gas is far cheaper than in the United States and coal is far more expensive. Similarly, the amount and reliability of wind varies. In some regions (e.g. the southeastern part of the United States) winds are pretty weak. Whereas in other regions (e.g. the Aleutian Islands of Alaska) winds are very strong.

Note as well that the relative costs of electric power sources will change with technological advances. Photovoltaics strike me as having the greatest potential for big cost declines. But being the most expensive photovoltaics most need big cost declines. Nuclear, wind, and cleaner coal costs will decline as well. But how much and how soon?

Two more wild cards: dynamic pricing and better electric energy storage technologies. Big declines in battery costs would greatly help wind and photovoltaics. Electronic switches could charge batteries when electricity is cheapest.

Share |      Randall Parker, 2006 December 28 04:49 PM  Energy Wind

aa2 said at December 29, 2006 6:02 AM:

And if a carbon tax goes on coal, it makes nuclear better. Nuclear is keeping on declining in cost, and in addition capital is cheap right now. The only time I can see going for something like natural gas is when capital is so expensive. I bet the big boys like GE will even finance a nuclear plant for an area.

I notice most people don't want to believe that with technology we won't have to 'cut back', 'conserve'. If you go further and say not only will we not conserve, we will dramatically expand our per capita use over the next century.. then they really emotionally disagree.

Jake said at December 29, 2006 9:13 AM:

Germany found that backup plants burned up more energy than the wind farms produced. Thus they have given up on wind and are moving towards nuclear power.

It is very important that we not subsidize wind farms as that would cause us to create too many backup power plants that waste energy.

Ned said at December 29, 2006 9:40 AM:


Your assertion that wind power is too expensive (i.e., just another eco-fad) is correct. But if one believes, as I do, that human activity does not play a major role in global warming, then there is no need for a carbon tax. The cost of electricity for industry in the US in 2004 averaged 5.3 cents per kilowatt-hour, which is quite low by international standards, lower than Germany, Japan or even Mexico ( This is in part due to the vast US coal deposits. A carbon tax will destroy this advantage, cost the US jobs (especially in fields such as heavy industry, which use lots of electricity), worsen the trade defecit and not do anything about global warming. Even if one believes in human-induced global warming, reducing US levels slightly would not make a dent in global CO2 levels when countries such as China are massively increasing their emissions and not doing anything about it. Some coal plant emissions, such as mercury, are potentially dangerous and do need to be curtailed, but CO2 isn't one of them.

Randall Parker said at December 29, 2006 9:56 AM:


What I'd like to know: Which futuristic nuclear power plant designs will have lower capital costs?


How much can wind power costs come down? Will bigger and bigger propellers bring us to a point of much lower costs? I'd like to see a graph of historical costs of wind power to see what the trends have been recently.


I really dislike coal due to the particulates, mercury, and other crap coal plants release. As for CO2: Seems to me there's a real risk higher atmospheric CO2 will cause big climate changes. Whether those changes will be a net plus or minus depends on where you are sitting and what your values are. I'd say that if you are Bangladeshi then global warming is bad news. If you are a Russian or Canadian or Alaskan (unless you are an enviro-weenie) it seems like a net benefit.

I also think if we required CO2 sequestration from power plants industry would find cheaper ways to do it and that the rate of advance in nuclear power technologies would accelerate. We might be better off in the long run in terms of energy costs if we sequestered.

Paul Dietz said at December 29, 2006 12:03 PM:

But if one believes, as I do, that human activity does not play a major role in global warming, then there is no need for a carbon tax.

I used to be a GW skeptic, but I'm not one now. I believe the preponderance of the evidence is against your opinion now, and that the case will be proved beyond a reasonable doubt (as has happened with anthropogenic halocarbons and the ozone hole) before too very much longer.

I see a great deal of shoddy thinking on the contrarian position on GW, repetition of well-debunked talking points, and just general wishful thinking. You should ponder the intellectual company you're keeping and ask just how sure you are in the position you've put yourself in.

Randall Parker said at December 29, 2006 12:16 PM:

Paul Dietz,

I take the position of prudence. Is it plausible that the CO2 increase will lead to global warming? Yes. Is it absolutely proven that the result of high CO2 will be harmful to a substantial chunk of humanity? No. But it is imprudent to wait for absolute proof before we do anything at all.

Still, even if you accept that higher CO2 will lead to global warming there are still arguments to be made against doing anything now. One economic argument against doing anything is that reduced fossil fuels consumption will slow global growth and therefore leave more people in poverty and cause more deaths than might be caused by global warming.

Also, another economic argument against action is that if we wait it will be cheaper to fix the problem later and we'll have more wealth with which to fix the problem.

I favor the position that we ought to accelerate energy technology development so that the cost of doing something will drop more rapidly. Also, accelerated energy technology development will eventually make non-fossil fuels cheaper than fossil fuels and that will be very pro-growth.

However, maybe carbon taxes are the most effective way to accelerate the development of new energy technologies.

Paul Dietz said at December 29, 2006 12:31 PM:

However, maybe carbon taxes are the most effective way to accelerate the development of new energy technologies.

How do you feel about carbon taxes coupled with corresponding income tax reductions? The general principle is to tax things you want less of. We don't want less income!

Randall Parker said at December 29, 2006 1:15 PM:


I question whether it is politically possible to cut other taxes while increasing carbon taxes. In Europe VAT came in addition to income taxes and VAT basically is what allowed European governments to take a larger percentage of GDP than the US government and state governments can manage. I oppose VAT for this reason and am really unthrilled with the prospect of carbon taxes. I'd rather we just banned the emission of CO2 by electric power plants. Such a ban would create less government mischief.

Carbon taxes, because they might increase the size of government, might even slow economic growth more than optimists claim.

Again, I have serious concerns about CO2 emissions. But there are sound economic arguments for why we might want to wait before requiring a reduction in CO2 emissions. I'm not set on my position. I'm quite open to smart arguments on this issue.

The best sequence of events would be for a big push to advance non-fossil fuels energy technologies that would lead to fossil fuels becoming more expensive than alternatives. Then the market would displace fossil fuels with no carbon taxes involved.

Also, since I'm leery of taxes as a mechanism for accomplishing goals I'm more open to regulatory changes. Not that I'm thrilled with regulations either. But building code regulations and certainly emissions restrictions regulations could accomplish a partial reduction in CO2 emissions. For example, if coal plants were not allowed to emit any regular pollutants (e.g. mercury, particulates) that'd make the environment cleaner and, by driving up the cost of coal electric, cause more nukes and wind to be used. It'd also encourage conservation without increasing government tax revenue.

Gerald Hibbs said at December 29, 2006 10:01 PM:

We don't need no stinkin' energy conservation Mr. Carter. The answer is streamlining regulations and placing incentives -- whatever needs to be done -- so we can build hundreds of pebble bed reactor nuclear plants ASAP. Also, we must heavily subsidize electric vehicle purchases (while penalizing pure gas powered vehicle purchases with a multi thousand dollar tax -- excluding vehicles for industrial usage) for a period of years. A 3-5 year subsidy would probably be enough for the price per unit to decrease sufficiently due to economy of scale. Give the car companies a year or two notice of the policy change and watch them go. If this policy were set in place can you imagine the line up of vehicles we would see when the starting gun went off? In ten years the new cars would trickle down so that almost all consumer cars were electric.

I always read, "When the technology is ready. . ." I think the Tesla Roadster's acceleration with fast charging batteries that have >90 mile range meet the needs of >%90 of consumers. And that's pure battery cars. A hybrid design that allows for gas operation as well will take care of the rest of the consumer market including rental vehicles for the all battery people who need to travel cross country. How much more ready does the technology have to be?

Seriously, what exactly is the standard the technology needs to meet? To me this feels like constantly waiting for the next level of television or computer before purchasing. There is always going to be a next level until the technology is completely replaced by a new technology (ie. cassette -> CD.) The main blockage to the average consumer buying an electric vehicle at this point seems to be cost and perception. Show me a well styled, good acceleration, mid size hybrid sedan selling for an IC comparable price that has a 5 minute charge time (with gas stations adding charging stations -- yes, subsidizing capital costs) and I'll show you a popular vehicle.

Perhaps I'm wrong on where the battery technology is right now. If I'm wrong it will be there in a year or two. But I think I'm right and we just need to kick the industry/populace in the butt. Most of the time I say let the market work itself out. However in this case national security forces me to conclude we need a major change to our operations yesterday. Consumer acceptance isn't going to kick in enough until there are charging stations at every gas station and the cost is comparable to pure IC engines. I don't see those things happening for a long, long time without intervention.

As to the nuclear plant issue? Don't even get me started. For once I look at France with approval and envy. Japan, Germany, France. . .what the heck is our problem?

Gerald Hibbs said at December 29, 2006 10:36 PM:

OK, I went off on electric cars in the wrong thread. It's just that the nuclear topic is to me inextricably linked to electric vehicles. So, oops, and I apologize. I would remove it and repost another time, but that's not an option.

Since I'm doing another post I wanted to point out that should my plan be implemented, which it won't, we would see a ton of money go into battery technology immediately. I urged giving the car companies a year or two notice of the policy change. That would also be a year of two notice to venture capitalists and existing companies that there was a money storm on the horizon so they better do their best to position themselves under the cash cascade. I predict we would see a new car lineup with capabilities significantly advanced of our current tech.

For the people who would object to putting so many billions into a project that will eventually happen anyway (moving away from carbon based fuels) I would note that the benefits of an electric/nuclear based economy far outstrip the obvious and I think once in place cheap electricity from nuclear plants and really cheap per mile transportation costs would be a boon to our economy. Anytime you lower the cost of doing business it is a good thing.

Further, we would no longer be funding people who want to kill us. As it is we are propping up hostile regimes who maintain power, in part, by spending our petro dollars to pacify the populace. Add into that the, to me, second order benefits to our environment and you've got a policy with lots of good (and essential as regards security) benefits for a cost that is a mere miniscule percentage of our annual federal budget.

Finally, we would once again be subsidizing positive change in the rest of the world. A major problem is the economy of scale. It's great when it is in your favor. But when you are on the uphill portion it's a real bitch. Our taking the lead on this would make it easier for the rest of the world to follow suit. Not only would we see changes in our own country sooner but in the rest of the world as well.

/Gore - Kaczynski screed mode

Randall Parker said at December 29, 2006 11:01 PM:


The Tesla is going to cost about $100k. That's partly because of their target market and small production runs. But it is also partly because the cost of batteries is too high. Also, they are building a sports car. Their idea works probably because it is a cramped vehicle. Does it even have a trunk?

I'm all for cost effective battery cars. Ditto cost effective nuclear power. But France is able to build large numbers of nuclear reactors in part because of a market controlled by government. Coal is cheaper unless coal burners are forced to capture all emissions.

I'd really like to see all electrics displace most gasoline cars and for nuclear power to displace coal. But while nuclear is cheaper than wind it is still too expensive.

aa2 said at December 30, 2006 6:28 AM:

Randall said "what nuclear plant future ideas have cheaper capital costs?"

I'm actually nto one who believes we'll go pebble bed reactors or anything too exotic. Just more refined, better engineered, more mass produced evolutionary improvements on the current models. I generally agree with you the problem with nuclear has been its upfront cost versus coal. Especially when the mass introduction of nuclear stopped in the late 70's and early 80's... its not pure coincidence to me that it was a time of very high interest rates. There were other factors but that is a big one.

The last Canadian nuclear plant built was projected to cost 4 billion.. and ended up costing an idiotic 14 billion for a 700mw plant. But things are far different now.. the Koreans just built a number of plants for under 2 billion for a 1 gigawatt plant. I believe the cost was 1.6 billion each and they built 4 plants. And that wasn't an estimate that was the final cost of the completed now operational plant.

If you look at what Toshiba, GE and the others are saying the cost is going to decline further, especially as they are selling more. For example you don't need to engineer the whole thing again, you don't have to construct the auxilary factories again.. And like any industrial process you find better cheaper ways of doing things. We're looking at a handful of reactor designs for the whole world.. and even that could come down.

A handbuilt car is fabulously more expensive then a mass produced one.. and much less reliable. So all and all I believe its safe to say over the next two decades the price of a 1 gigawatt reactor will come down under 1 billion US 2006 dollars.. and its already at or under 2 billion.

Don said at December 30, 2006 3:42 PM:

Electric cars might fit wind power well (assuming they are plugged in at night). Run the climate control for a couple of hours before the car leaves in the morning to pre-cool or pre-warm on shore power and you stretch the battery range a bit more.

Randall Parker said at December 30, 2006 5:36 PM:


But how does that Korean nuclear plant construction cost translate into dollars per megawatt-hour produced?

To put it another way: What would a 1 gigawatt nuclear plant have to cost in order to achieve $33 per megawatt-hour of produced electricity?

Gerald Hibbs said at December 30, 2006 7:55 PM:

Randall, both of your objections were met within the (far too long) thread hijacking posts. If you can have that kind of acceleration in a sport model, surely a sedan model (with trunk) can be made with reasonable acceleration. You don't need 4 seconds like the Tesla. Yes, the Tesla cost $100,000. Put it in a plant making >80,000 and you'll see a huge price drop. Similarly, has the economy of scale kicked in for next gen electric car batteries? Again, I'm talking out my behind, but I doubt it. Even then the cost will currently be higher than similar model IC cars, I admit. I think, due to security reasons, subsidizing (hell, we subsidized Hummers!) them is a worthwhile investment.

This is not an instant fix, it's the beginning of a decade long switchover as IC cars lifespans naturally end and high cost new cars filter down into the used car market for lower income people. Over that decade the technology will get better and better spurred on by (government subsidized) market forces. The computer chip makers have no problems updating their factories every year/six months why will the battery makers? This technology will spread around the world into developing countries. In a decade hostile governments will be paupers without the funds to continue weapons R&D, pay soldiers/terrorists, fund extremist schools/mosques, retain power by paying off the populace, etc.

What level of capability do electric vehicles need to reach before it is considered a viable technology?

As to nuclear, my proposal is for heavy capital subsidization -- essentially adding money to the subsidy until the market reacts and builds almost exclusively nuclear. According to this chart nuclear beats out all competitors, including coal, in operating costs. As the capital costs decrease so does the subsidy. I'm less adamant about this proposal than the car proposal as we do have US sources for our coal. But the switchover is going to happen, gaining the benefits of abundant/cheap/clean energy now for a relatively small governmental investment seems a fair trade off. Frankly, I don't see nuclear plants evere being cheaper than coal without governmental intervention of one sort or another. If the government is going to force a mandate, I would rather it not be unfunded. The longer we wait the more coal plants that are going to have to be decommissioned at a cost to owners when the government eventually does change the rules.

TTT said at December 30, 2006 10:00 PM:

Is it true that battery technologies improve at a rate of 5-8% a year?

If so, then the chances for electricity replacing gasoline within cars increases with time, and could be easy by 2017..

aa2 said at December 31, 2006 1:53 AM:

ttt I've heard the rate is 9% a year since the introduction of lithium ion for consumer electronics. That was looking at the increase in watt hours per kilogram, for the same cost. And you are right even if they are only half what they need to be today.. that means in a decade they will be there. And in two decades there will be no comparison.

aa2 said at December 31, 2006 2:07 AM:

Randall if we assume Korea has the asme interest rates as the US of 5% for bonds.. being generous. That would mean a capital interest cost of 80 million dollars a year for one plant if it was totally financed. They are one gigawatt plants and they are the baseload power.. but lets assume only 75% of the time can it sell power.. say in the dead of night all the nuclear isn't used. So in a year there is 365*24 hours.. or 8760 hours. So 8760*.75= 6500 hours. So lets cut it down further .. 6000 gigawatts or 6,000,000 megawatts of power a year sold for 33 dollars a megawatt hour. Which works out to 180 million dollars a year. 80 million for capital cost.. and 100 million left for running the place which is imo way more then you need.

33 dollars a megawatt hour is also 3.3 cents a kilowatt hour to compare to Gerald Hibbs numbers.

aa2 said at December 31, 2006 2:16 AM:

btw I agree Gerald that because the enemies of America are mainly financed by oil it is a national security issue. To replace as much oil as possible to cut demand and 'starve' the enemy out. Also if carbon in the air is as bad as some people think that should be a major issue if not the pre-eminent issue. And then the question shouldn't be do we build the next generation of plants nuclear? It should be how rapidly can we replace all the coal plants in America with nuclear plants? China should then be asking the same questions too.. and the world probably would follow America's lead. If it offered a realistic solution, and started taking real action.

Gerald Hibbs said at December 31, 2006 7:20 AM:

Right aa2, one of the major foundations to my arguments is that the U.S. will make these switchovers feasible. I talked in an earlier post about how all the cabs in China are mandated to run on natural gas. China does indeed care about its environment and takes these issues quite seriously, but at the same time the reality is they gotta do what they gotta do. If that means dangerous and dirty coal mines? So be it. If US money can lead the way to an affordable EC? Expect to see Chinese cars switch over soon after.

As you noted aa2, there is a big drop in building costs waiting to happen. I love economy of scale, but the uphill side is a major bitch. If the US gets into the nuclear power plant business in a major way we help other countries over the hump. If we can get China over the hump, then when they start building for their 1.4 billion people that helps other countries over the hump as well.

I found a recent article, from November, stating that Duke Energy was seeking to build two coal powered plants in SC for a cost of 2 billion, but an assortment of changes had raised the cost to 3 billion. The article doesn't state what the power output is, but from these numbers it looks like we are moving into the ballpark of nuclear power plants. Frankly, I think at this point we are simply seeing a lack of political will to embrace nuclear and streamline some of the endless regulatory roadblocks to nuclear plant creation. Fortunately, it looks as though we are seeing movement in this regard with even some supportive noises from the Greens.

Philip Sargent said at December 31, 2006 12:45 PM:

Off topic perhaps, but some intersting news recently, the new Westinghouse nuclear design may actually get built. The AP1000 is an attractive idea; highly simplified, lots of passive safety features, modular, cheap etc.etc. - the only problem was that it's most direct competitor design was being built (or will be built) in two sites (Finland and France) and the AP1000 was just paper.

On Dec.16th China did a deal with Toshiba (parent company of Westinghouse) to build APs at two sites in China. Since otherwise the AP design was looking increasingly unlikely to fly, the buyers must have got a very good price.

Randall Parker said at December 31, 2006 1:27 PM:


What do you think that AP1000 price might be? Any guesses?

Also, what do you think the prospects are for getting the price of nuclear power down to the price of coal? Do you think the Finnish and French reactors will generate electric for more or less than the cost of cheap coal electric in the US?

It seems to me that cheap nuclear electric power might be our best prospect for slowing and stopping the growth of CO2 emissions. But I have a hard time judging whether nuclear reactor construction costs can fall all that much.

boonton said at December 31, 2006 3:44 PM:

But the economics above understate the problem with wind. Suppose we shift to dynamic pricing of electricity (which we should btw) so that the price of electricity varies as a function of demand and supply. Electricity would cost more at 2 PM on a hot summer day than at 2 AM on a cool fall day. Well, wind tends to blow when electric demand is lowest!

Indeed but it would seem to be an increase in electric supply. There would be a little bit of supply increase during periods of peak demand (those rare summer days with a good breeze) and a big supply increase during a period of low demand. If cheap electric gets even cheaper then it becomes more economical to find uses of electricity when it is cheap such as storage or night.

In other words let's say today the cost of daytime electric is $0.10 kwh and nighttime electric is $0.05 kwh. Let's say wind supply causes that to become $0.09 during the day and $0.02 during the night. We are better off because we get cheaper electric either way...even though the impact at day time is tiny compared to night.

What is interesting is that the difference increases dramatically. In the above example, if you could find a way to store night electric energy and sell it during the day the markup is $0.05. But with wind the markup becomes $0.07. Perhaps the real winner then would be the fuel cell industry and other forms of energy storage as well as those industries that can shift energy consumption away from the daytime and towards night.

Randall Parker said at December 31, 2006 3:55 PM:


But wind costs more than the market price. So there's little incentive to build wind generator sites without either a tax subsidy or a regulatory requirement.

If wind generator farms start lowering the price of electricity that'll make wind power even more uneconomic. If most wind electric has to sell for 2 cents per kwh then there'd be no point in building any wind farms at all.

boonton said at December 31, 2006 5:07 PM:

Whether it's economical to build them or not isn't relevant to the fact that they will increase supply. Consider commercial real estate. If there's a boom then a lot of office buildings get built but cannot cover their costs with rental income. That's not good for the builders nor for the efficient use of resources but the fact would remain that supply will increase lowering rental costs.

If they do increase the differential in costs then they might increase the utilization of energy at night. In the long run that would be good for big baseline producers like coal and nuclear because it's more efficient to run such plants at full capacity non-stop 24-7.

aa2 said at January 1, 2007 12:47 AM:

Randall and Phillip the Toshiba/Westinghouse deal with the Chinese was 4 plants of 1 gigawatt each.. to be built in the same giant complex.. The deal, estimated in the past at about $8-billion (U.S.) Check NEI nuclear notes the blog.. Then look around dec 16.

The Chinese are going with what I've been arguing for a few years.. big nuclear complexes with extra room on them, near major load centers. They seem to be focused on the idea of 6 plants of 1 gigawatt each built in one complex. This facility will have room for more to be built.

It will be interesting to see if the Chinese keep expanding their nuclear ambitions.. each year I look it up and their plans are always progressed further, including more and bigger reactors.. and a more well thought out plan.

Engineer-Poet said at January 1, 2007 6:47 AM:

Getting back to wind power, the cost of backup goes down to zero if demand-side control is sufficient to cover the short-term variation in output.  The cheap coal-fired plants can only ramp their power slowly.  This is a problem if the plants have to track rapid variations in either demand or generation, but advanced DSM systems such as ice-storage A/C and PHEVs can balance that from the other side.  This frees the grid manager, because firm demand for X megawatts right now become something closer to Y GWH plus or minus Z% between 9 AM and 5 PM, and the minute-by-minute variation can be managed by trading off vehicles and A/C against other loads.  This will also eliminate the need for other spinning reserve, reducing capital and operating costs as well as fuel costs.

Nuclear may be cheaper than wind, but the soonest we're going to get new nuke plants in the US is about 2016.  The rate of wind installations is doubling about every 2 years, with about 2.5 GW installed in 2005; at that rate, 2011 will see about 20 GW installed and account for ~60% of new electric generation.  Wind can fill the gap for carbon-free electricity.

If PHEV's come along at the same time, annual production of ~15 million PHEV's per year driving 8000 electric miles per year at 250 Wh/mile will add an average of 3.4 GW/year to demand.  If these vehicles are connected to the grid with 220 V 30 A cables (440 V 20 A is also possible), the annual addition will supply 99 GW of controllable load; even if they only connect to standard outlets, that would still be 25 GW/year.  It would only take a few years of production to provide enough DSM capability to eliminate the gas-fired backup plants for wind and every other bit of spinning reserve on the grid.

Randall Parker said at January 1, 2007 8:09 AM:


Most of the need for backup power for wind is for baseload demand. That need does not go away. In fact, since wind blows stronger at night when demand is lowest it requires a parallel baseload capacity for running only during the day! How can wind compete?

Currently the rev-up of wind in the United States that you speak of is caused mostly by two factors:

- tax subsidies. $18 per megawatt-hour.

- state level mandates for renewables as a percentage of total power. California is the biggest driver here.

The rise in wind power is therefore not sustainable. Its supply characteristics do not match the demand peaks. It costs more. Its growth is driven by government mandates.

Engineer-Poet said at January 1, 2007 10:04 AM:
In fact, since wind blows stronger at night when demand is lowest it requires a parallel baseload capacity for running only during the day! How can wind compete?
That's where DSM comes in.  A large part of the afternoon demand peak is A/C.  If you use the wind to charge PHEV's and put ice in storage whenever it's available, you have many hours or even a day to get your backup powerplant going.  The off-peak availability of wind reduces the peak load on lines and transformers, extending their life and reducing maintenance.  You can also forecast the availability of wind and use the forecast to start ramping the fossil backups up or down; the difference between instantaneous generation and uncontrolled demand is met by DSM and/or V2G.

If we get the 15 million PHEV's/year and charge them 8 hours/day, each year's production will consume about 10 GW average when charging.  If we produce flat-out for 5 years, that's about 50 GW.  Even my 8x increase in wind installations by 2011 (20 GW/year) would only yield 6 GW average at 30% capacity factor; the growth in DSM potential would outpace the variable supply by a considerable margin.  We'd need another doubling to 40 GW/year (~12 GW average) before wind would start catching up.

If the full social cost of CO2 is figured into the price of coal power (about 7.5¢/kWh per the Stern review's figure of $85/ton) wind is cheaper than anything we can bring on-line between now and 2016.  The PHEV's and ice-storage A/C will actually save money by replacing imported oil and flattening the demand curve, eliminating the need for gas-fired peaking plants and even a lot of spinning reserve.

Randall Parker said at January 1, 2007 12:45 PM:


People only run A/C in the summer. What about the other 3 seasons? What about the fact that wind peaks in the winter? Plus, the freezer system will cost money and take up space. How long for the pay-back?

Keep in mind, again, that wind costs more. The night time when wind blows harder is going to be when a variable pricing electric market will pay the least for electricity. The summer A/C peak is when wind blows less hard. Also, the southeast is the biggest user of A/C in the United States and the southeast has the weakest winds in the US.

As for social costs: Noone now wants to pay those guesstimated social costs. But full CO2 sequstration for coal electric will only cost 2 cents a kwh according to a US DOE EIA estimate I read. So there's no way that wind makes sense. Better to sequester coal CO2.

Of course, the problem is that the public doesn't even want to pay 2 cents more per kwh to sequester or to get wind power.

PHEVs: We need to get them first. Plus, we need to get dynamic electric pricing. But once we get dynamic electric pricing and people have an incentive to charge up at night that strikes as an argument for sequestered coal or nuclear, not wind.

Boonton said at January 1, 2007 2:32 PM:

I'm not sure what PHEV's are. It sounds like they are giant ice boxes? Freeze water at night then during the day use it to asorb heat out of your building thereby saving on air conditioning. But doesn't this require either a big building or lots of small house sized ice boxes? Sounds like it would be easier to use off peak electric costs to make fuel cells which you could then discharge during high peak periods. If wind brings the off peak cost of energy down then it could still be of use. Charge up the fuel cells during those cold winter nights and discharge them during heat waves in the summer when power prices spike.

Bob Badour said at January 1, 2007 4:28 PM:


PHEV=pluggable hybrid electric vehicle

The problem isn't so much that wind will reduce off-peak prices so much as dynamic pricing means the price for wind generated electricity will drop to the point where it is unattractive to produce.

PHEV's will mean stable demand for electricity because one will charge the batteries in them when demand drops and potentially discharge the batteries in them during high demand periods--both for household use as well as transportation.

Engineer-Poet said at January 1, 2007 9:29 PM:

Residents of dwellings may only run their A/C in summer, but a lot of industrial and commercial operations run it year-round.  There are also domestic and commercial refrigerators (suitable for ice storage?), freezers (freeze a salt mixture instead), hot water tanks (obvious)...

A lot of the electric DHW systems could and should be converted to solar with electric resistance or heat-pump backup.  To the extent that demand can be forecast, hot water makes a good storage medium and a great opportunity for DSM.  If you expect to shower in the morning, run a load of laundry in the afternoon and the dishwasher the next day, you can take your time heating the water so long as it's ready when needed.  You can do this by over-heating a tank during an early power surplus (and tempering the water at the outlet) and coasting for hours or days, or maintaining just enough hot water to squeak by until power becomes available.  (The complementary nature of wind power from winter storms and space-heating loads is also good fit.)

PHEVs provide the DSM capability which is the natural complement to wind (and solar), as you yourself noted:

Dynamic pricing combined with pluggable hybrids that can easily respond to pricing changes will do something else too: They will create more growth potential for energy sources that are not reliable. Wind and solar photovoltaics will both become more useful if a large portion of the demand for electricity was highly responsive to pricing changes.

Randall Parker said at January 1, 2007 9:46 PM:


Scraping the bottom of the barrel to use a reference to an argument I made! ;>

I agree that dynamic pricing and PHEV both will make intermittent power sources more useful. I also agree that we can find ways to store energy such as by storing heat and cold. We can develop cheaper ways to do that too.

My point is that, geez, wind is still expensive. The ability to store expensive energy is not all that exciting. Of course, what is true of prices today is not necessarily true tomorrow. Technologies advance. Now, can wind electric costs drop by, say, half? You have any insights into that? I'd really like to know.

I think solar has prospects for much larger price drops than wind. Of course solar needs bigger price drops. But I expect solar to eventually become cheaper than wind. Propellers seem like a heavily developed technology. Though GE keeps making the propellers bigger to lower costs, right? Maybe nanotech materials can some day drop the price of props and towers by a factor of 2 or more. I've read the electronic components for converting and switching wind electric have dropped in price by some large amount. Probably the switching circuitry will get cheaper still.

Boonton said at January 2, 2007 10:06 AM:

Wouldn't it be much more economical for a coal or other large plant to simply build a massive battery to asorb energy during off peak periods as opposed to having everyone plugging in what are essentially smaller batteries in their cars?

K said at January 2, 2007 2:38 PM:

boonton: wouldn't it be.....

no, the batteries, inverters, and maintenance are more expensive.

The cheapest way of bringing coal or natural gas power online is simply to fire another generator when needed. Added generation is amazingly cheap once everything else exists. And total capacity can be increased in small increments.

I suppose opening the gates at a hydro generator or increasing the burn at a nuclear plant is even cheaper. Nuclear and hydro can increase output at little cost until capacity is reached. At that point adding another reactor or dam is an enormous cost.

Hydro also has the problem of fuel supply, nature must send water down to the dam, you can't order more by phone.

Randall Parker said at January 2, 2007 5:44 PM:


Batteries in cars would be cost-justified by their ability to propel cars down the road.

Huge stationary batteries cost too much just to store electric. As K argues it is cheaper to just build more generating plants and turn them on when you need them.


Nuclear can't get turned up to increase capacity because all nuclear plants run at full capacity. Nuclear plants are always used for baseload. Their fuel costs are low. Their capital costs are high. So it pays to run them all the time.

K said at January 2, 2007 6:26 PM:

Randall: I try not to argue against the facts. If 'all nuclear plants run at full capacity' is facts I bow to em.

Since reactors produce heat - hence power - variably depending upon the position of the control rods why would they always be run at capacity?

That question may date me, I have little idea what control mechanisms are used in modern high-power reactors. Certainly all-on or all-off makes some sense for safety - the more variable a control is, the more chance of malfunction.

Your comment about using the nuke power first makes sense, the marginal cost/watt is probably so low as to not exist.

Sione said at January 3, 2007 12:02 AM:

But as soon as another nuclear accident occurs (and it will), one which releases radioactive poisons into the atmosphere and/or the water and/or the food chain, then there will be some serious reservations expressed by the public. Let's see how many people will suddenyl choose to remember all those who got sick and died after Chernobyl. I wonder how ya'll feel when it's 'merican babies and children getting chronic sicknesses or born with cruel deformaties after getting irradiated by a good ol' boy nuklear whoops-a-daisy-sorry. How much more ya'll going to relate to your own getting in the neck rather than the time when some slavic commie furriners got irradiated. How will it go when large areas (whole neighbourhoods or regions) are rendered dangerous due to a wee accident in your home country? Goodbye nuclear option.

I like nuclear power. I admire the technology. I just don't want any more of it in the Pacific. I'd sure hate to see it within a few thousand miles of where I live. There has to be a better way than to use something with such a huge downside. And please don't ramble on about how safe it will all be. Accidents occur. Always have, always will. More certain that taxes. As certain as death.

BTW how long do you think it'll be before the first Airbus A380 crashes with the loss of all on board? If that possibility can't be excluded, then a nuclear stuff-up can't be denied either.


As far as battery cars are concerned-

Gerald, even if they get better slightly performance and range you are still left with the problems of battery pack lifetime and charging cycle issues (hours and hours to charge- remember you need to move Megawatts to fill in minutes, where is the infrastructure for that going to come from?). Then you have the problem of all those nasty chemicals getting into the local environment. And as for accidents, big volts, big amps, big troubles... Non-trivial issues. Tough stuff to sort out (even assuming it's possible to do so).


Boonton said at January 3, 2007 8:39 AM:

My point is that if car batteries can make good use of off peak cheap electric then one centralized battery at the powerplant should be even better. Coal generators are massive creatures who are not just switched on at a moment's whim. Gas generators are more often used for this purpose but even then each 'on-off' switching increases maintaince costs.

A massive battery at the plant itself (or located off plant but connected to the grid) could be tapped within minutes I would think and could be switched off just as fast. Unlike thousands of relatively small car batteries it could make use of economies of scale. Why would electric companies not want to make such good use of off peak capacity? I suspect the reason is that today battery storage remains too expensive to essentially perform electricity arbitrage (buying cheap watts at night, selling dear watts during the day) profitably in most cases.

This may change, though, if the price differential between peak and offpeak increases (which it would if off peak supply was increased...say by wind) or fuel cells become more efficient.

K said at January 3, 2007 11:37 AM:

Boonton: your last. If huge batteries make sense then fine. I just haven't seen any indication that they do.

I don't follow this. You wrote:'This may change, though, if the price differential between peak and offpeak increases (which it would if off peak supply was increased...say by wind) or fuel cells become more efficient.'

I have seen figures that winds are stronger at night. So more power would be available then. That may be your argument. But what do fuel cells have to do with this?

The problem I see is to get power to homes and store it there during off-peak. It can then be used for AC, etc. loads during the peak. To just keep adding peak power capacity at the generators means you must also beef up the transmission structure. The pure electric or plug hybrid vehicle is a great match for off-peak power too - they will charge at night.

One way to store off-peak power at home might be to chill water overnight - down to say 34 F - engineers could figure the optimum temperature - and use it as a heat sink during peak hours. Simple technology for residences, complex in offices, apartments.

Boonton said at January 3, 2007 1:14 PM:


What we are talking about here is arbitrage...prices are low at night but high in the day....if we were talking about apples we'd go out and buy them at night and then sell them during the day for an automatic profit. With electric power you have to store it and that ends up costing you both in terms of money and lost power. Since that isn't free then before you embark on a scheme like this you have to compare your storage cost to the arbitrage profit.

If the difference in prices was large enough then it can overcome that cost. Right now it doesn't appear to be because I dont' see why large power generators wouldn't be the first to exploit the situation since firing their huge plants up and down is a major source of trouble for them. If they could exercise arbitrage then they would run the plants at an even level and use storage to make up for the fact that demand goes up and down.

I'm not sure why the emphasis is on getting power to homes where it will be stored. It would be horribly inefficient if we expected every home to have its own coal or gas powered generator why would it be any better for each home to have some type of large battery inside of it (either in the form of a hybred car or other battery)? I suspect the first place to try a battery scheme would be at the power generators themselves. They would minimize the energy lost in sending their generatred energy over long distances and since they already have large plants and lots of money they can build and manage huge batteries or fuel cell that would never be proper to put in a home.

K said at January 3, 2007 2:45 PM:


First. I said nothing about having large batteries or generator in the home.

Second. You don't explain what fuel cells have to do with this. Maybe you mean keep fuel cells at home and use off-peak power to extract H2 to fuel them. Then convert the output to alternating and use it during peak hours. Well that will work. Honda is currently estimating the cost may fall to around $20K by 2018. (If you have to ask what it costs now you can't afford it.)

I would just as soon have a large battery at home as a tank of H2 and a compressor, if it comes to that.

Third, you suspect the place to try a battery scheme is at the power generators - but since you say it is probably too expensive, why?

Fourth, I indicated keeping grid and generation loads lower seems more desirable than just boosting or modifying the amount of power the utility can deliver at peak; and one way to do that is to get cooling power to homes at night. And those hybrid/electric cars charging at night also level the load at the utility and optimize their performance.

Fifth: The purpose of batteries in electric/hybrid vehicles isn't to store power at home; it is to power the vehicle. The fact that night is when it can best be done is a happy coincidence.

Sixth: Utilities already exercise arbitrage. They buy/sell electricity depending upon each days conditions and time-of-day. The big impediments are regulation and the lack to a true national grid. Better long-distance transmission lines would help too.

You seem interested in arbitrage and profit making for either power traders or utilities. I think they know how to make money already. Someday that way may involve huge batteries but that seems unlikely today.

Engineer-Poet said at January 3, 2007 7:51 PM:

I note that Boonton doesn't see the error in assuming that transmission losses will not apply if batteries are located at the powerplant (the point of use is still just as far away), and I further note that he doesn't realize that it would take a smaller and cheaper battery to store energy at the usage end of a transmission line (after all the losses were taken) instead of at the production end.  Last, with the battery at the usage end, the transmission line can be run closer to the average demand, cutting losses from resistance (which scale as current squared).

Neither does Boonton see the utility of batteries aboard vehicles.  Of course those batteries have value; the price per kWh of electricity is vastly smaller than gasoline even before pollution and greenhouse gases are considered.  Electricity is about 75¢/gallon equivalent.  The fact that the battery-powered vehicle can exercise arbitrage in the electric market (as well as providing regulation and perhaps other services such as VARs) just adds more and more benefits.

Getting back to the cost of wind power:

The claim was made up at the top that wind power's generation cost is 5¢ to 6¢/kWh.  The cost of backup generation would add about 0.8¢/kWh to this, if there was no other measure such as DSM to deal with that requirement.  PHEV's appear more than capable of doing this.

Note that those figures are for the current cost of wind power.  Today's state-of-the-art land-based wind turbines are rated at about 1.5 MW/unit (similar units designed for offshore use are rated at about 3.6 MW).  A couple decades ago, the typical plant was about 30 kW and produced power for 25-30¢/kWh (this is the territory solar PV is breaking through now).  5 MW turbines were in testing last year, and the optimum size is purported to be about 10 MW.  The historic 50-fold size increase cut the cost by about a factor of 5; if the next factor of 6 increase cuts cost by a factor of 2, we're down in the 2.5-3¢/kWh range.

This is not to say that we shouldn't exploit PV, or solar thermal electric, or solar heating, or anything else.  IMHO we should just slap a $85/ton fee on the carbon in fossil fuels and let the other technologies shake themselves out.

Engineer-Poet said at January 3, 2007 8:01 PM:

Pardon, that would be 3.4-3.9¢/kWh, before tax credits.  Cheaper than even un-sequestered coal.

I'm more than happy to pay a few billion dollars a year to help make the cost of RE cheaper than fossil fuels, because it will cost far less in the long run.

Randall Parker said at January 3, 2007 9:08 PM:


Boonton doesn't appear to want to understand that battery costs make electricity more expensive. The expense will be justified sooner if the batteries are used to propel cars. Their capital cost will be justified by savings in gasoline costs. Then their storage capacity for the grid will come as an additional benefit.

As for wind: The cost is 5 cents to 6 cents with a tax subsidy. Also, that cost is for producing an unreliable power source and at times other than when you most need it.

Your information about increased size and cost is very interesting. Maybe wind will be able to compete and without subsidy. But I'd like to see some wind industry estimates for how scaling up will cut costs. Since I have connections to a local wind power system design and manufacturing company I'm going to try to find out that way.

Tax on carbon: Not going to happen. The public wants cheap fuel prices.

Boonton said at January 4, 2007 10:17 AM:

Second. You don't explain what fuel cells have to do with this.

sorry, my science here is weak. From what I understand fuel cells are essentially batteries that use H2. Like a traditional battery, a breakthrough would make it more viable to stash surplus power in such a battery.

Third, you suspect the place to try a battery scheme is at the power generators - but since you say it is probably too expensive, why?

1. No one is seriously doing it. Since generating companies are profit seeking institutions with both money and plenty of engineering brainpower why wouldn't they do it?

2. As I did point out there is economies of scale at work here. A huge industrial battery system right next to the generator would probably be more efficient at storing energy and delievering it at peak moments than lots of scattered smaller batteries.

Fifth: The purpose of batteries in electric/hybrid vehicles isn't to store power at home; it is to power the vehicle. The fact that night is when it can best be done is a happy coincidence.

True but I suspect to make this really economical on a vaste scale the price of traditional fuel would need to go up much more. I think the jury is still out on electric cars and hybreds... There's still a good argument IMO for simply making cars smaller and designing the most efficient fuel based engine possible rather than toting around a heavy battery AND fuel based engine.

Sixth: Utilities already exercise arbitrage. They buy/sell electricity depending upon each days conditions and time-of-day. The big impediments are regulation and the lack to a true national grid. Better long-distance transmission lines would help too.

Arbitrage means buying in one market and selling in another. For example, if you buy cheap cigerattes in West Virgina for $20 a carton and sell them in New Jersey for $40 (in NJ they easily cost $50-$70) that is arbitrage. What we are talking about is not a market seperated by geography but by time...nighttime versus daytime. To exercise arbitrage you would need to buy power at night and sell it during the day but to do that you need to have a way to store it.

I note that Boonton doesn't see the error in assuming that transmission losses will not apply if batteries are located at the powerplant (the point of use is still just as far away), and I further note that he doesn't realize that it would take a smaller and cheaper battery to store energy at the usage end of a transmission line (after all the losses were taken) instead of at the production end. Last, with the battery at the usage end, the transmission line can be run closer to the average demand, cutting losses from resistance (which scale as current squared).

I think you're double counting the benefit. Let's just pretend that it requires 100 units of power to put 50 units into a home. To put an extra 50 units into a home at peak time the plant must have a battery capable of putting out 100 units from the plant. A home battery need only put out 50 unites but in order for the utility to put 50 units into that battery at off peak hours it needs to send 100 units out its door. (This is, of course, assuming the battery suffers no additional loss as it stores the 50 units coming into the house).

Either way the transmission cost is the same so the question comes down to the loss of putting the power into the battery. I suspect there the plant will have the edge because of simple economies of scale AND the fact that they would have the expertise to optimize the battery charging process.

Needless to say there are some old fashioned environmental concerns as well. Batteries contain lots of unpleasant chemicals. While it isn't yet a major issue what's going to happen if you have 50% or more vehicles driving around with huge chemical batteries and many homes contain similiar batteries in their basements? A centralized plant based battery could be monitored for pollution a lot easier.

Neither does Boonton see the utility of batteries aboard vehicles. Of course those batteries have value; the price per kWh of electricity is vastly smaller than gasoline even before pollution and greenhouse gases are considered. Electricity is about 75Ę/gallon equivalent. The fact that the battery-powered vehicle can exercise arbitrage in the electric market (as well as providing regulation and perhaps other services such as VARs) just adds more and more benefits.

I'm skeptical. I don't think batteries as they work today make sense. While they may only be "$0.75 per gal" to fuel their weight and other costs make them unwieldy to use in cars which need a light fuel and engine. Think about an airplane. Sure its fuel is more expensive than gas or diesal but since they require so much more energy they need to get every bit of power per ounce of fuel. Let's not even imagine doing a battery powered plane. The fact remains while it may cost less to fuel the battery you are toting around several hundred extra pounds. If you're putting a battery in your basement or at a plant you don't have to worry about moving it around so go ahead and let it weigh several tons or more!

Maybe fuel cells could deliever the bang for the buck which is why I'm holding out for breakthrus there.

Boonton doesn't appear to want to understand that battery costs make electricity more expensive. The expense will be justified sooner if the batteries are used to propel cars. Their capital cost will be justified by savings in gasoline costs. Then their storage capacity for the grid will come as an additional benefit.

On the contrary, I keep emphasizing that for arbitrage to make sense the differential in prices has to be great enough to overcome the costs of storage. If that wasn't the case, if battery storage was free then there would be no difference between peak and offpeak prices. Power plants would simply fuel their batteries off peak and discharge them during peak hours.

Boonton said at January 4, 2007 11:06 AM:

There is a plus for home based batteries. If you have a personal windmill or solar cells that may put out extra power when it isn't needed then it can be channelled into the battery with much less transmission loss. But most people do not have the open land needed to accomodate enough wind/solar energy generators to make a surplus...even late at night when only a few light bulbs and a fridge are pulling juice.

Engineer-Poet said at January 4, 2007 7:52 PM:

Randall, I figured the subsidy into it.  6¢/kWh + 1.8¢/kWh subsidy = 7.8¢/kWh generation cost.  7.8/2=3.9.

Boonton, since you're not doing so well with the explanations I'll go over them more slowly this time:

  • Generating and storing 100 kWh at the powerplant requires a battery capable of holding 100 kWh.  Do that at the far end of a line with 50% losses, and you only have to store 50 kWh.  You can either cut the size of the battery in half, or you can store the equivalent of 200 kWh from the generator.  (Real line losses are about 9%, not 50%.)

  • Arbitrage means buying cheap and selling dear; it doesn't have to be in different places, it can be at different times.  Buying cheap off-peak power and using (or selling) during peak times is arbitrage.

  • Transmission lines are sized by their peak carrying capacity, not average.  If you had a generator making 800 MW feeding a line with a capacity of 1 GW, you could only add an extra 200 MW from other sources (or storage) before the line would be at capacity; this limits the load at the far end to 1 GW.  But if you had the 800 MW feeding through the line all the time and storage at the far end, the line is no longer the limiting factor; you could store 400 MW for 8 hours overnight, and then pour all of that back in 2 hours in the afternoon (1.6 GW) PLUS the 800 MW from the generator.  2.4 GW > 1.0 GW.

  • Sure, batteries contain unpleasant chemicals.  You want MORE of them so you can take your line losses afterwards?

  • "I don't think batteries as they work today make sense.
  • "  Until you learn enough of the science to at least calculate some numbers, you don't know what makes sense and what doesn't.  Not a slur, just fact.

Everyone who's making a useful contribution to this has studied enough to master the facts.  Now it's your turn.

Boonton said at January 5, 2007 9:38 AM:


I'll be happy to agree that your command of the facts is better than mine, if you wish you can take my objections to be more like questions. Such as:

1. If batteries as they work today make sense why are utilities not using them? Certainly the lines are not typically runing at peak capacity all the time so a plant could keep 200 MW in storage and use it on the 1 GW line during a spike in demand (when they would normally only be sending 800 MW down). That would save them the need to bring standby generators online. The only reason that comes to my mind is that the cost of using batteries is too much relative to the potential revenue from arbitrage.

2. I'm not sure I understand this:

Generating and storing 100 kWh at the powerplant requires a battery capable of holding 100 kWh. Do that at the far end of a line with 50% losses, and you only have to store 50 kWh. You can either cut the size of the battery in half, or you can store the equivalent of 200 kWh from the generator. (Real line losses are about 9%, not 50%.)

Ok, so if you wanted to have 50KWh stored in home based batteries the plant would have to send 54.945 KWh down the line. If you wanted to store the power at the plant to be sent down the line when needed you would need a battery with 54.945 KWh. It still seems like a wash, battery at the plant or in the home.

Yes I see your point about the lines. If the line can only accomodate another 200 MW but demand spikes to 250 MW then the power plant doesn't have much it can do but home based batteries could draw as needed without having to care about the capacity of the line.

But line owners have a market incentive to build enough capacity in their lines so that it is rare that they are overloaded. From my experience it is rare in summer to lose power for more than a half hour due to overloaded lines (storms and other events are a different story). Even if that's atypical I don't see a reason why it would be difficult to build enough excess capacity to handle demand spikes 99.9% of the time.

So again it still seems to come down to where does it make the most sense to put a battery. In every home or at the plant. The plant still seems to have these advantages:

1. A centralized location to make environmental monitoring easy.
2. Economies of can correct me on this but isn't a big battery more efficient at storing energy than a small one?
3. Plenty of engineers and experts on hand to monitor and maintain the battery(ies) to ensure they are operating at full efficiency.

If home based batteries were a good idea why don't we see them? Home Depot will sell you a fuel based generator to cover you for a while during a blackout but I've never seen a battery based system for sale. Obviously it's cheaper to pull power from a line than it is to make it yourself burning fuel oil. Why not a big home battery that would charge at night and discharge during a blackout? (even if you don't get variable pricing you're still buying power cheaper from the line than from the gas station).

This isn't just homeowners who may not be experts. Institutions like hospitals also use fuel based generators for backup power instead of massive battery systems. The only explanation that seems plausible is that when you factor in the costs battery storage is simply not economical either at the plant or at the end user at this time.

Which is why I said a breakthrough in batteries is needed.

Engineer-Poet said at January 5, 2007 8:02 PM:

Today's batteries are more expensive than generation.  They are roughly the same cost as gasoline (even today's short-lived lead-acid batteries; Firefly Energy's carbon-matrix cells will last a lot longer and be a lot cheaper per kWh).  That's why (a) nobody's using batteries for grid backup, and (b) all the talk is about PHEV's.

Nick said at January 7, 2007 1:18 PM:

Randall, E-P,

Do we know that the 5-6 cents cost for wind (after subsidies) is accurate? It sounds high to me: I've seen estimates of 3-8 cents before subsidies.

The .8 cents for intermittency management is for wind at 20% of market, which we're quite far from - we're about .9% in the US now, and may be at 1.8% by the end of 2007.

I wonder at the 2 cent estimate for CO2 sequestration: is this actually being done anywhere at any kind of scale? Surely this is affected by geography: what if you don't have a handy place to store it?

The value of the wind tax credit is much less than 1.8 cents. 1) Keep in mind that the credit only lasts 40% of the useful life of the wind turbine, so even with adjustment for present value the credit is only worth about 1 cent, and 2) many wind developers must partner with for-profits to capture the value of the tax credit, which reduces it's value further.

I think the negative correlation between wind supply & overall demand is overstated. It exists in some places and not others. In fact, in some places there is positive correlation. I'd like to see a real analysis (rather than hand waving by someone who appears to have been fed their facts by Duke, a utility that likes coal & nuclear - although I have to say I was unable to read the article).

Cutler Cleveland, a professor who has involved with these kinds of things, especially E-ROI analysis, feels that nuclear is the most expensive alternative due to Price-Anderson's unfunded liabilities. Personally I feel that proliferation is a pretty expensive externality. Consider that the Iraq war was started in the name of WMD, principally nuclear, and it's likely to cost a trillion dollars. Sure, that wasn't the real reason, but it was considered good enough to start a war....


natural gas generation has, until just the last couple years, been a very cheap way of handling peak generation needs. Water pumped storage is likely the cheapest way to handle utility-scale power storage. It's been done for many years, and can cost less than a cent per kwhr, but they are very large projects, and require cost certainty for some years. A good example is Ludington, MI, which was built about 30 years ago to smooth out power demand for nuclear power plants.

V2G is interesting, but is complicated and involves battery wear & tear. Demand management of battery charging is much easier and lower cost, and will (as E-P notes) greatly facilitate wind & solar.

Randall Parker said at January 7, 2007 6:14 PM:


I think the NY Times has the reporting expertise in science and technology and the contacts to get accurate answers from reliable sources. So I tend to give their numbers a lot of credence. That's why I wrote this post in the first place.

Here's a US Department of Energy Energy Information Administaration estimate

CO2 is currently recovered from combustion exhaust by using amine absorbers and cryogenic coolers. The cost of CO2 capture using current technology, however, is on the order of $150 per ton of carbon - much too high for carbon emissions reduction applications. Analysis performed by SFA Pacific, Inc. indicates that adding existing technologies for CO2 capture to an electricity generation process could increase the cost of electricity by 2.5 cents to 4 cents/kWh depending on the type of process.

Note that is the cost for retrofit. For a new electric generation plant using gasified coal the cost of CO2 capture would be substantially less.

Here are some coal CO2 sequestration estimates from May 2006:

The World Coal Institute notes that at present the high cost of carbon capture and storage (US$ 150-220 per tonne of carbon, $40-60/t CO2 - 3.5 to 5.5 c/kWh relative to coal burned at 35% thermal efficiency) renders the option uneconomic. But a lot of work is being done to improve the economic viability of it, and the US Dept of Energy (DOE) is funding R&D with a view to reducing the cost of carbon sequestered to US$ 10/tC (equivalent to 0.25 c/kWh) or less by 2008, and by 2012 to reduce the cost of carbon capture and sequestration to a 10% increment on electricity generation costs.

More recently the DOE has announced the $1.3 billion FutureGen project to design, build and operate a nearly emission-free coal-based electricity and hydrogen production plant. The FutureGen initiative will comprise a coal gasification plant with additional water-shift reactor, to produce hydrogen and carbon dioxide. About one million tonnes of CO2 (at least 90% of throughput) will then be separated by membrane technology and sequestered geologically. The hydrogen will be burned in a 275 MWe generating plant and in fuel cells.

Construction of FutureGen is due to start in 2009, for operation in 2012. The project is designed to validate the technical feasibility and economic viability of near-zero emission coal-based generation. In particular it aims to produce electricity with only a 10% cost premium and show that hydrogen can be produced at $3.80 per GJ, equivalent to petrol at 12.7 cents per litre.

In Denmark a pilot project at the 420 MWe Elsam power plant is capturing CO2 from post-combustion flue gases under the auspices of CASTOR (CO2 from Capture to Storage). Flue gases are passed through an absorber, where a solvent captures about 90% of the CO2. The pregnant solution is then heated to 120įC to release pure CO2 at the rate of about one tonne per hour for geological sequestration. Cost is expected to be EUR 20-30 per tonne.

A 2000 US study put the cost of CO2 capture for IGCC plants at 1.7 c/kWh, with an energy penalty 14.6% and a cost of avoided CO2 of $26/t ($96/t C). By 2010 this is expected to improve to 1.0 c/kWh, 9% energy penalty and avoided CO2 cost of $18/t ($66/t C).

Figures from IPCC Mitigation working group in 2005 for IGCC put capture and sequestration cost at 1.0-3.2 c/kWh, thus increasing electricity cost for IGCC by 21-78% to 5.5 to 9.1 c/kWh. The energy penalty in that was 14-25% and the mitigation cost $14-53/t CO2 ($51-200/tC) avoided. These figures included up to $5 per tonne CO2 for transport and up to $8.30 /t CO2 for geological sequestration.

So 2 cents an hour is a rough estimates and the real cost depends on whether it is a new or existing plant and whether we use today's technology or technology we'll have within 2 to 5 years.

Nick said at January 8, 2007 9:45 AM:

Well, thatís encouraging. That seems affordable to me, even if itís probably outside the pale of politically doable things right now.

You asked about wind turbine improvement possibilities. Several things to keep in mind: 1) power is the square of the length of the rotor arms, as it is proportional to the area swept by the arms. So, double the length of the arms, quadruple the power. 2) wind speeds increase with height, and 3) power is the cube of wind speed. So, as you make wind turbines taller and larger power increases much faster than cost, probably roughly twice as fast as cost.

I see two major possibilities: offshore floating windturbine structures , which are based on proven oilrig technology but still in development. They'd reduce costs by eliminating the sea-floor based structure; allow bigger turbines; allow greater distance from shore which would take care of visibility problems, and take advantage of better winds. The 2nd is kite-based windturbines, which I believe will be faster to install, can take advantage of stronger & more reliable high winds, and have reduced structural costs - they're still in R&D, though, and more speculative.

I think wind will eventually get to 3 cents per kwhr on average in the US, without subsidy.

My estimate is that if we were to take ALL the external costs into account, including pollution, GW, occupational health, esthetics (including the esthetics of chopping off mountaintops for coal), strategic/security concerns, nuclear proliferation, economic competitiveness in new industries and tech export possibilities, oportunity cost of long investment times for coal and especially nuclear, etc that we would find that renewables (wind and eventually solar, wave, etc) are the cheapest energy source and the best place to invest our money.

Randall Parker said at January 8, 2007 7:01 PM:


I wonder if wind turbines could be put on existing oil rigs.

Nick said at January 9, 2007 8:36 AM:

Well, hereís a project to put turbines on oil platforms as an alternative to decommissioning. I canít find recent articles, so it may have failed. I think there are clear synergies. I suspect thereís a chicken & egg problem which will prevent widespread adoption until someone invests the time & money in developing a standard module for addition to platforms.

and hereís Shell using wind & solar to power an unmanned gas platform, with an interesting discussion of the complementarity of wind & solar. It looks like a subcontractor called Tss4U (an acronym for 'The solar solutions for you') may have developed the kind of standard module that's needed.

Another synergy: I would expect to see offshore wind farms combined with wave energy farms. The advantages just seem blindingly obvious. Thereís an interesting side benefit to wave farms: many coastal communities spend enormous amounts of money protecting their coasts from erosion, and wave farms would soak up a significant % of wave energy, thus prolonging the life of beaches.

Dr. Warren Reynolds said at January 23, 2007 11:19 AM:

The article has missed a vital point. Windpower to hydrogen ! North Dakota Co-op Electric is constructing a sind-to-hydrogen power system. Hydrogen storage and use levelizes the windpower output so it is controllable. Control is what the electric utilities want.
Denmark has over 1,000 wind mills generating MWs of electricity. One windy day, they "dumped" many MWs onto the grid which caused problems in Germany, France, etc. Why did not Denmark just convert all that excess electricity into hydrogen for later conversion to electricity ?
Germany, by the way, has agreed to drop all nuclear power projects and phase out all nuclear by 2020. The European Commission on Sustainable Power has the goal of 100 cities on renewable energy by 2015.

Randall Parker said at January 23, 2007 7:03 PM:

Warren Reynolds,

The German government is reexamining the decision to phase out nukes and may reverse it. They are having problems scaling wind to the level needed.

As for converting to hydrogen: It is costly to store. Also, the conversion back to electricity costs more and wind electric already costs more than nuclear and coal without going thru a hydrogen stage.

All sorts of solutions are possible if higher prices are acceptable. But since people generally oppose higher prices all sorts of theoretical solutions do not get implemented.

prathap said at July 1, 2010 12:30 PM:

if the 200mw generator has a weight of 3tonne in standby condition,,,,,
then what will be the weight of the generator in running condition

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©