US electrical generation - Where we are now

Last week, Hannes Kunz and Nate Hagens put up one post talking about electricity, and they are planning another post in the near future. I thought now might be a good time to put up a few graphs regarding US electricity generation, showing where we are now. Most of the graphs are summaries of data from the Energy Information Administration website, either from the renewables section, or the electricity section.


Figure 1 - Amounts of US Electricity Generated by Source

Figure 1 shows the largest source of US electricity is from coal. Following behind these are natural gas and nuclear. The amount of electricity generated using natural gas and nuclear have both grown in recent years--natural gas because more natural gas generation was built; nuclear because ways were found to increase the amount of electricity obtained from existing plants, by upgrading their capacity and by operating them more hours. "Other renewables" have grown from 2.2% of the total in 1996 to 3.57% of the total in 2009. Pretty much all of this increase has taken place in the last four years.


Figure 2

Figure 2 just breaks out the "other renewables" into their component parts during the last five years. Wind is responsible for nearly all of the increase. Solar is not large enough in quantity to show up on the graph. Geothermal, biomass waste (which includes biogas), and wood have all been close to flat, at least in terms of the electricity actually generated in the past few years. The amount of electricity generated by wood has been declining.


Figure 3

Figure 3 is similar to Figure 2, but shows the "other renewables" as a percentage of total electricity generated. Wind grew from 0.4 % of US electricity generated in 2005 to 1.8% of US electricity generated in 2009. Wood (including wood chips) represents about 0.9% of electricity production; waste has grown from a bit under 0.4% to a bit over 0.4%% duriing this period. Geothermal is a little under 0.4% of US electricity. In total, as mentioned above, these other renewables amounted to 3.57% of US electricity in 2009.


Figure 4

US electrical generating capacity has grown greatly in recent years. The biggest share of the increase has been in natural gas generation, particularly in the 1999 to 2003 period. Prior to about 2000, gas prices were about $2 per thousand cubic feet at the wellhead, and this inexpensive price no doubt played a role in the ramp-up. Natural gas plants were also relatively quick and inexpensive to construct, and could act as "peaking plants" (plants operated for short periods of time, but at high prices, when demand was high), and thus could take advantage of high wholesale electricity prices, for short periods of time.

Other renewable generation capacity also increased, reaching 3.8% of total generation capacity by the end of 2008.


Figure 5

Figure 5 compares the utilization of the various types of power plants, and shows how this has been changing. In this calculation and in Figure 6, the calculation is made by dividing the actual electricity produced by the theoretical capacity, where the theoretical capacity is the average of the year-end capacity for the year indicated and the year prior. This averaging process with respect to capacity is used to get an estimate of how much capacity was, on average, available during the year.

Nuclear plants now operate at a little over 90% of their theoretical capacity. There is little room for improvement.

Coal plants operate on average a little over 70% of the time. Many of them are taken off line at night.

Renewables ex hydro have been trending down in their utilization of capacity, most likely because of a changing mix away from wood and geothermal (with high capacity utilization) to more wind. More detail will be shown in Figure 6.

Natural gas utilization has declined significantly, with the building of all the new natural gas plants. If the average utilization of capacity is only about 25%, and some plants intended to operate most of the time (50% to 70% or capacity), then some plants must be used very little.


Figure 6

Figure 6 shows the percentage of capacity used by other renewables. Geothermal is highest, using over 70% of capacity.

Wood utilization of capacity has dropped from a shade over 70% to about 60%. (We noted above that total electricity generated from wood was declining slightly.) Some of the wood is waste product from industrial processes (paper manufacturing, construction). To the extent that this source of fuel is depressed because of the economy, it may be affecting the percentage of wood electricity generating capacity being used.

Wind utilization of capacity does not seem to be trending upward. The average for the four years is 29%, which is better than Europe, but not as good as folks manufacturing new wind turbines are advertising.


Figure 7

Figure 7 is graph of FERC estimates of the cost of new capacity. Of course, some of these sources are operated more hours than others.


Figure 8

Figure 8 is a little calculation I made, trying to figure out the cost of capacity on more of a comparable basis, adjusting for utilization.

In Figure 8, the nominal costs per 1,000 watts of capacity are my estimates regarding what the FERC amounts shown in Figure 7 are, underlying the graph amounts. The percentage of capacity is the average of the last four years of the actual percentage of summer capacity, from Figure 5 or 6. The adjusted cost of comparable capacity shown in the last two columns is simply the nominal capacity divided by the utilization.

One can argue whether these numbers are comparable, even after adjustment. For one thing, there are differences in the quality of electricity produced--whether it is available on demand, or not. There are also differences in how many years plants can be expected to operate, operating costs, and the cost of fuel.

On this basis, wind comes out comparable in cost to nuclear, but without the cost of nuclear fuel. Coal and natural gas seem to be quite a bit cheaper (neglecting the considerable cost of fuel). Geothermal seems to be especially cost effective when one considers the lack of fuel costs.

I have added a line for Solar PV, based on the indications of this study from Berkeley regarding the installed cost of solar PV. The numbers are so high that one wonders whether something is wrong with the calculation.


Figure 9

This is a graph from the Berkeley report showing the trend in installed solar PV costs. There are a lot of costs in the installation besides the PV itself--including the inverter, and the cost of installation. I also don't know how accurate the percentage of capacity that seems to come out of the EIA data is. It seems to average only about 16% - reflecting a combination of factors including how many hours sun is available, how much shading there is on the panels, how well the panels are installed, and how clean the panels are kept.

The indicated solar PV cost is extremely high compared to the other costs. I can delete the solar PV indications, if someone can explain how they are incorrect. But they are somewhat concerning.

TVA was once one of the largest coal fired generators but they are clearly moving from a coal + Nuclear + hydro utility to a Nuclear + natural gas + hydro & pumped storage utility.

No new coal fired plants at TVA since 1970, they recently announced the closure of another 1,000 MW of coal capacity and a 600 MW NG plant is near completion.

In nuclear, Brown's Ferry 1 (1,065 MW originally, now 1,155 MW) was restarted in 2007 after a 1985 shut-down of BF 1, 2 and 3. Brown's Ferry 1, 2 and 3 are all scheduled for uprates to 1,280 MW.

Watts Bar 1 (1,121 MW) was started in 1973 and completed in 1996. Watts Bar 2 (1,180 MW) is scheduled to be completed in 2013 and partially completed Bellefonte 1 (1,256 MW) is about to restart construction after decades of delay.

This year, a once 80+% coal fired utility should be less than half coal with substantial future drops in coal clearly ahead.

Best Hopes,

Alan

PS: There are very limited wind resources in the TVA area.

Nuclear is a 'false fireman', Alan, as is more hydro.
Dixie needs to go heavily into PV and biomass.

Louisiana has lots of CCS sites so clean coal is an option. And don't forget geothermal and geopressured gas.

Nuclear is going to run out fuel long before ugly coal unless fusion makes a break thru. The US is more dependent on foreign sources of uranium--90% than oil--70%.

http://www.asianage.com/india/%E2%80%98uranium-reserves-be-over-2050%E2%...

Also drop the dorkey electric trams.

Try the transmilenio(but in the future run on biofuels).

http://www.youtube.com/watch?v=SRGoketbIZE

Majorian -your Asianage ref is simply quoting two people who work for "Clean Energy Research Centre at the University of Florida" who are engaging in boilerplate antinuclear rhetoric. Their one sided position references nothing and amounts to their opinion only.

Yeah, look how freakin' expensive nuclear is per kwh compared to everything else.

For every dollar invested, you get about can get about two kwhs of solar for every nuclear kwh, and solar is getting cheaper all the time.

And wind is about 3-6 times cheaper.

Talk about EROI!

So in a financially constrained world, which should we moving faster on. We can't do it all folks. Every dollar spent on nukes is depriving us of two to six times the energy we could have gotten out of solar and wind. Even if for some reason you don't understand how dangerous nuclear is, it's just too expensive of a technology and too slow to come on line for this very late stage of the game.

There is a place for nukes, after we have saturated the grid with as much wind, solar, geothermal, hydro, biomass as either the grid can take or we can produce, even with expanded pumped storage.

If we are constrained to 10% FF, Florida, for example, will need to get at least 60% of their GWh from nukes.

Alan

Florida has pretty decent off-shore wind, and pretty decent insolation. Further, there's really no reason every area in the country has to produce their power locally.

I imagine we'll probably use nuclear power, it's probably cost-effective, and it may be a good idea overall (I'm still very concerned about weapons proliferation), but it's not necessary.

There is only a small area off the Florida Keys with good off-shore wind (no geographic diversity and not NEARLY enough MWh) and hurricanes make wind too risky on the coast and off-shore anyway.

The massive number of HV DC lines required to power Florida from, say, a combination of wind from Western Oklahoma, South Dakota, Manitoba and Labrador (some geographic diversity) is "beyond raional analysis". Last I checked the max/line was 3 GW, but plans in China for 4+ GW.

Clouds reduce insolation. Florida is cloudy. I was surprised to find out that April was the highest insolation month (although June 21st has the highest "cloud top" insolation).

http://www.thermomax.com/TampaFL.php
http://www.thermomax.com/MiamiFL.php

And some days pretty much all of the state is cloudy.

Good for hot water and a supplement for electricity.

The only viable non-carbon source for the bulk of Florida's electricity is nuclear. Not true for every state.

Alan

Nonsense.

Miami has an annual average of 5.26 kwh/m2 per day versus
5.38 kwh/m3 per day for Phoenix, Az.
NOLA gets 4.76 kwh/m2 per day (89% of Phoenix).

http://www.solarpanelsplus.com/solar-insolation-levels/

Florida uses ~300 Twh/ year of electricity or ~1 Twh/day
so 1 Twh/day / 5 kwh/m2 x 10% eff = 2 billion m2 or

There are 4 million single family homes in Florida.
If 500 m2 of solar PV were put on each roof(100'x50') you'd
make up that much power.

The fact is that solar and wind doesn't work with electric trams but would work fine with something like the 12 kwh/d Nissan Leaf EV; 12kwh/d / .5 kwh/m2d = 24 m2.

You also ignore that Florida's renewable biomass potential which could provide 20% of Florida's electricity.

http://www.fl-dof.com/forest_management/fm_pdfs/woody_biomass_economic_s...

Nuclear is not renewable and looks to run out of fuel after 2050.

Those are some eyebrow rasing numbers..

5000 sq ft of solar on each roof - just how big is the average Florida house? And how much of it is not at the optimum orientation?
And to power the cars, many of which are away from their house during the day, you'll need a separate battery storage at night, so your costs and losses are going up pretty quickly.

I think the solar would work better with electric trams which run mostly during the day and need much less electricity per passenger mile

And as for the study, basing it on a sustained yield of 35 green tons (about 18 dry) per acre per year, is based on the best results obtained in field trials of eucalyptus, but there is no guarantee of it being sustainable (eucalytpus is notorious for depleting soils of phosphorous).

Not saying they can;t have a biomass industry - I am a supporter of that and am actively researching prospects for such here in British Columbia, but their figures are best case figures.

One thing not mentioned there though, if you did produce 50m green tons of eucalyptus (specifically blue gum), you could extract about 0.5-1m tons of eucalytpus oil from it, as part of a gasification-burning process. It makes a great biofuel and can be used directly in diesel engines, and is a great co solvent for ethanol and gasoline (prevents water separation). That is equivalent to about 20,000bpd of oil. Not huge in the scheme of things, but that oil will probably worth more then than it is now. Makes for nice smelling exhaust too.

to power the cars, many of which are away from their house during the day, you'll need a separate battery storage at night, so your costs and losses are going up pretty quickly.

Better to put PV on the car roof, and on top of whatever commercial garage they'll be in.

electric trams which...need much less electricity per passenger mile

Surprisingly, they use more than EVs.

Better to put PV on the car roof

I would be surprised if the average PV panel on a car roof would have an EROEI greater than one. Orientation and lack of shading are extremely important for PV; two things for which average car roof will not be advantageous.

The commercial garage idea is better.

Roof based solar can be very good, for powering Air Circulation, and that can slash the start-up AirCon requirements.

So whilst the Watt-Hours/$ might not stack up wonderfully, the ability to budget for a smaller AirCon, plus the better initial comfort, are what you compare against.
(not just what those watts could have cost from a wall plug).

On the one hand, many cars won't have a high % of time in the sun. The best: EREV/EV taxis, which are out 100% of the time, and need range.

OTOH, the comparison is power from ICEs: 10kWhs from a $3 gallon of gas gives us $.30 per kWh, which factory installed PV ought to be able to beat handily.

10kWhs from a $3 gallon of gas gives us $.30 per kWh, which factory installed PV ought to be able to beat handily.

Nick, my point is that you're not going to get $.30 per kWh for PV on a car, factory installed or no. Grid tied PV systems in ideal locations struggle to meet that price at a capacity factor of 15 percent. A PV panel on a car is probably going to have a capacity factor of one or two percent, even if the driver makes an effort to park in the sun when that's convenient.

I agree that cars will have low capacity factors - that's what I was saying with my 1st sentence.

OTOH, some cars are outside all of the time.

Grid tied systems have to deal with BOS costs, but a panel in a car should be able to eliminate most costs: it's manufacturing, which is far more efficient than grid-tied systems that require field installation; redundant support structures; and dedicated power electronics. If a car can add a panel for $1 per Wp, and get just 2.5% capacity factor, it could get $.30/kWh.

Of course, new-car PV options are going to be very expensive: such options carry very high markups and generate large profits for the manufacturer. But, that's price, not cost.

nick,

So then we need covered garages for all the cars, not street parking or basement parking?

C'mon, I have nothing against rooftop solar, the fact is, it produces electricity that coincides fairly closely with the daily peak consumption, which is great. i find it hard to believe that there would not be enough nightime cheap electricity to charge ev's .

You have to agree his numbers are a bit over the top.

As for the trams/trains using more power than EV's I'd be very interested to see your data for that.

So then we need covered garages for all the cars, not street parking or basement parking?

Well, I figure just put PV either on the car, or on whatever structure the car is in.

You have to agree his numbers are a bit over the top.

I certainly wouldn't suggest a 100% solar solution. Heck, I wouldn't suggest 100% anything. OTOH, his assumptions as to numbers of SFHs, % efficiency, and type of structure were very conservative. IIRC, the total roof space (industrial/commercial, residential, garage, etc) in the US is adequate to provide all of our current electrical generation, should we want to do such a thing. IOW, the roof space isn't the barrier.

As for the trams/trains using more power than EV's I'd be very interested to see your data for that.

Keep in mind that transit systems generally have pretty low utilization: rush hour may be packed, but only in one direction. Night and weekend service, and service to lower-density areas, is necessary to provide a complete solution, but greatly lowers utilization.

I had a source for US transit rail using about .35 KWH/pax-mile, but I can't put my hands on them right this second.

I seem to remember you had some stats for electric rail trips in Calgary. Here's their average annual power costs: $4.8M (2006) http://www.calgarytransit.com/html/technical_information.html . What would you calculate?

Calgary's train system is the 2nd highest utilised one in N. America (apparently the Monterrey system is the first).
They do 260,000 passenger trips per day, and the average commute in Calgary is 7 miles, each way. (that figure came from somewhere else, form the City, posted by Rocky Mtn Guy some time ago).
So using those numbers, we can get a rough estimate of 7*260,000=1.82m passenger miles/day, and we'll round that down to 1.5m.
That power bill works out to $13k/day, and if we use an optimistic price of 8c/kWh (they actually buy wind power, exclusively, so it is likely higher) that works out to 160,000kWh/day, or 0.1kWh/passenger mile.
Now, an EV can beat that, if it is has more than 2 people, but they suffer from an even worse "capacity factor" than trains, since the average car is 1.1 or 1.2 people.

The other part of the equation is the investment required. The 130,000 people will need 100,000 EV's , and at $33k for the Leaf, that is $3.3bn, for vehicles that have a 20yr life. The train system has about 70 vehicles that cost$3m each and have a 40yr or longer life, for 1/30th the investment - you could use the difference to build a lot of wind turbines (about 2200MW) The cost of building track and road is comparable. Added to that, the cost of ownership (insurance etc) and maintenance is much less for the train than for 100,000 ev's.

There is actually a really go paper on the Calgary system here, I think it's a great example of the 80/20 rule, they didn;t go for a cadillac system, and got 80% of the result for 20c on the dollar.
http://www.calgarytransit.com/pdf/Calgary_CTrain_Effective_Capital_Utili...

I think what's important, when making these comparisons, is to look at the efficiency, and load factors, of systems being built today. Some of the older ones, with very heavy trains (where they run on freight lines and have to meet their collision requirements) etc are less energy efficient. So, it's a bit like comparing to an old electric car instead of the state of the art .

Properly done, both are very energy efficient. Transit has the effect of gradually densifiying the areas they serve, which is also a good thing. In Vancouver this is already being seen along the corridor of the 1 yo old Canada Line - developers are preferring to develop around the train line than away from it, and homebuyers are preferring to live closer too. This means more people are being housed without having to build any new roads, water/sewer lines, etc etc, it makes better use of a lot of existing resources.

And doesn't mean spending $bn on (imported) lithium batteries, a lot of the train stuff can be built locally - all the $ spent on EV's will leave that city and much of it will leave the country.

Well, I like rail. I think we should expand it. I just hate to see it discussed unrealistically.

Now, we were talking originally about trams, and the Calgary system is a central rail system. A hub, in effect, not the spokes that a tram system would provide. The spokes, whether tram or bus, would have far lower utilization.

2nd, as you note, the electricity isn't really that important - it's cheap and plentiful. The dominant cost is operations: if you look at the overall cost of running the Calgary system, I suspect you'll find it close to $1 per pax-mile. That's 3x the cost of personal vehicles.

The key: personal vehicles are self-service, while buses, trams and rail require operators. Rail, of course, can have one driver for a number of cars, while buses and trams require a high ratio of driver to passengers.

$33k for the Leaf, that is $3.3bn, for vehicles that have a 20yr life.

The life is as long as you want. The current life of personal vehicles is entirely based on planned, fashion-based obsolescence, not on actual physical life. EVs especially are extremely low maintenance.

Now, we were talking originally about trams, and the Calgary system is a central rail system. A hub, in effect, not the spokes that a tram system would provide. The spokes, whether tram or bus, would have far lower utilization.

and for more flexibility, those spokes could be hybrid buses ?

Which I see are hitting commercial critical-mass quite quickly.

Yes, a very good thing.

Or electric trolley buses, Vancouver has hundreds of those.

But the hybrid buses are appearing too, which is good. these ones from Design line need just a 30kW microturbine to power them. (they could do it much cheaper with 30kW diesel, but that's not as sexy.

http://www.designlinecorporation.com/EcoSaver%20IV.%20pdf.pdf

Nick, I will agree that trams, running in the street or in unprotected row's are no more efficient than ev's, though they do add a certain character.

For the Calgary train system, from that report;

Although this paper has focused on capital costs, it is noteworthy that the effectiveness of
Calgary’s LRT has resulted in a low operating cost per passenger. For 2005, the average hourly
operating cost of LRT is approximately $163.00 ($139.40 USD). This figure includes operating,
maintenance and utility costs. With an average of 600 boarding passengers per operating hour
the average cost per LRT passenger is only $0.27 ($0.23 USD). In comparison, the average cost
for bus passenger boardings is approximately $1.50 ($1.28 USD) or almost 6 times the cost of
carrying an LRT passenger. Of course buses have considerably lower capital cost and have
different capabilities.

Note that $163 is per train, per hour. That is 23c per passenger, and with an average trip of 7 miles, you are looking at 3-4c per pax-mile. It really is a testament to what a well planned and utilised train system can achieve.

Trains don't actually require operators - the Vancouver Skytrain is completely automated.

And , of course, you don;t need to provide parking for trains. In most cities, there are three parking spaces per vehicle - that is a lot of land that is being used up, which further spreads things out etc.

I think 20yrs is quite a reasonable upper limit for the average car. Some will go on much longer than that, of course, but even though EV;'s are low maintenance, I don;t think it's fair to assume any significant part of the fleet will last for longer than 20 yrs. Especially given that the technology will improve these first Ev's will become obsolete faster.

EVs are significantly LESS energy efficient than trams because EVs help preserve Suburban Sprawl and all the waste & energy consumption that goes with that, while trams support Transit Orientated Development.

Trams save more energy indirectly, via TOD and less sprawl, than they do directly.

The ideal "gathering system" for trams is shoe leather and bicycles. Both work well with trams, neither works with EVs.

Quite frankly, I think a 20 year average life span for EVs is excessive. Accidents take their toll every year, as does salt further north. "Auxiliary systems" from window openers to seats to plastics wear out. And then there is battery life. Replacing EVs is another energy cost.

Alan

EVs are significantly LESS energy efficient than trams because EVs help preserve Suburban Sprawl and all the waste & energy consumption that goes with that, while trams support Transit Orientated Development.

As you know, I disagree: housing costs are far larger, and more important as a factor for people choice of location. If I agreed, I'd have to point out that relocating half the population from the suburbs to TOD would be a very, very long and expensive proposition.

As far as EV live goes: the average car these days is kept effectively for 12-14 years: their half-life is 6-7 years. But....they could be kept going for 100 years, if we wanted to. Proper preventive maintenance and inspection is far cheaper than replacement every 12 years. If, as a society, we wanted to make that even more feasible, we'd make them out of stainlesss steel, as we do rail cars. But, as a society, we choose to throw away personal vehicles far before their functional life is over.

Commercial fleets keep their rolling stock much longer than consumer fleets: trains, planes, trucks, etc. For all of them, 30 year old vehicles are not unusual. Why? Because they concentrate on function, not fashion. The fact is, personal vehicles are just as reliable as commercial vehicles. Heck, both ICE engines and electric motors last almost forever these days.

Accidents take their toll every year

Only a very small % of vehicles are completely smashed. Most have relatively less damage, but are "totalled" because their market value has fallen below the cost of repairs. Market value falls because of fashion, not function: a 15 year old Corolla has a market value of what, $1,500? But, it's perfectly functional.

"Auxiliary systems" from window openers to seats to plastics wear out.

That's a minor cost.

then there is battery life. Replacing EVs is another energy cost.

Battery costs fall reliable by 7-10% per year, and li-ioni costs are falling faster. EVs don't require significantly more energy to manufacture than ICE vehicles.

I bought an 1982 Isuzu I-Mark diesel new (could not afford the M-B 240D) and drove it till structural problems killed it. Seat spot weld came loose (rewelded), plastic fell apart, drivers seat torn, got rid of it when bumper support failed (repaired damage that corner a decade before) shortly after electrical short and other electrical problems (age related). Engine still good. About 2000 from memory. $650 to someone that was going to export it to Mexico.

Ended up buying a 1982 M-B 240D (manual transmission). Better late than never.

I can assure you that even 2010 M-B quality would not hold up as well as this old tank.

The Leaf is a Versa with a different drive train. By 2031, most will be in salvage yards for one reason or another.

Best Hopes for old time Mercedes quality,

Alan

I bought an 1982 Isuzu I-Mark diesel new (could not afford the M-B 240D) and drove it till structural problems killed it. Seat spot weld came loose (rewelded), plastic fell apart, drivers seat torn, got rid of it when bumper support failed (repaired damage that corner a decade before) shortly after electrical short and other electrical problems (age related). Engine still good. About 2000 from memory. $650 to someone that was going to export it to Mexico.

So, it's market value was only $650, but someone was going to keep maintaining it in Mexico. What would the annual cost of maintenance have been for the next 5 years? More than $1,000 per year? Far, far less than the cost of depreciation on a new vehicle.

I can assure you that even 2010 M-B quality would not hold up as well as this old tank.

M-B quality has fallen quite a bit relative to other makers, sadly.

The Leaf is a Versa with a different drive train.

Could you show me backup on that? AFAIK, that't not true at all.

By 2031, most will be in salvage yards for one reason or another.

Very likely, but not because of necessity.

The Nissan Leaf, for example, is similar to the Nissan Versa. A converted electric Versa, in fact, was used as a stand-in for the Leaf as a test-drive vehicle during Nissan's recent North American tour to introduce the Leaf.

http://sunpluggers.com/2010/03/how-much-will-nissan-leaf-or-chevrolet-vo...

Nissan saved a couple of years and quite a bit of development money by starting with a production car and then modifying as needed.

In 2000, I saw a rising cost of ownership ahead of me. I think I got the lowest annual cost of ownership by selling when I did. Absent the diesel engine, this car would have been crushed.

Best Hopes for my 240D outlasting me :-)

Alan

Nissan saved a couple of years and quite a bit of development money by starting with a production car and then modifying as needed.

Nissan says that the Leaf is a unique platform - they didn't start with a Versa and modify it.

I see where the writer got this idea: In car-maker jargon, the Versa body was used as a "mule" during the testing and demonstration phase.

In 2000, I saw a rising cost of ownership ahead of me. I think I got the lowest annual cost of ownership by selling when I did.

Sure, but how much would it have risen? The cost of maintenance would have been far below the cost of a new vehicle. The fact that you could find another used vehicle, for very little money, that has stood you in good stead, says something about how much value car owners abandon every year in the rush for fashionable new vehicles.

Best Hopes for my 240D outlasting me

M-B quality has fallen relative to other makers, but they've all gotten much better over the last -several decades. Did you know that small plane engines are now much less reliable than car engines? Most new cars have the same potential to last forever as your 240D.

We can relocate a third of the population from Suburbs to well built, energy efficient TOD in twenty years.

A better investment (and life cycle cost) than EVs to keep them in the Suburbs.

Alan

We can relocate a third of the population from Suburbs to well built, energy efficient TOD in twenty years.

Have you looked at new housing construction stats lately?

A better investment (and life cycle cost) than EVs to keep them in the Suburbs.

Have you looked at housing prices in dense urban areas (not NOLA), vs suburbs??

It's nice to see that Calgary's LRT is so well designed. But, of course, it handles less than 5% of pax-miles.

Again, I like trains, and I think we should expand them. But...very few people in Calgary have given up their cars because the rail system, and that's not likely to change soon.

As far as EV live goes: the average car these days is kept effectively for 12-14 years: their half-life is 6-7 years. But....they could be kept going for 100 years, if we wanted to. Proper preventive maintenance and inspection is far cheaper than replacement every 12 years. If, as a society, we wanted to make that even more feasible, we'd make them out of stainlesss steel, as we do rail cars. But, as a society, we choose to throw away personal vehicles far before their functional life is over.

Commercial fleets keep their rolling stock much longer than consumer fleets: trains, planes, trucks, etc. For all of them, 30 year old vehicles are not unusual. Why? Because they concentrate on function, not fashion. The fact is, personal vehicles are just as reliable as commercial vehicles. Heck, both ICE engines and electric motors last almost forever these days.

" Nuclear is not renewable and looks to run out of fuel after 2050. "

Right, like we ran out of 2 cent/gallon gasoline in 1908. We pay about 10 cents per kWh on average, but uranium cost is less than 1/3 cent per kWh with our 1960 reactor technology. A simple uranium burning MSR reduces uranium cost to 1/15 cent per kWh and with breeders it will be 1/300 cent per kWh.

The sun will run out of fuel before earth runs out of uranium and thorium. Nuclear is more renewable than wind and solar.

A simple uranium burning MSR reduces uranium cost to 1/15 cent per kWh and with breeders it will be 1/300 cent per kWh.

Please contact the Indians, Chinese or Iranians about your excellent idea!

I heartly approve your suggestion of a molten salt reactor than makes electricity for 1/15th of a cent per kwh( but I draw the line at 1/10th of a cent per kwh).

Let us all know when you convince somebody to actually build one of these miracles.

I guess we will have to start using the thorium cycle and the hundreds of years of fuel we have in storage which is called spent fuel from the pressurized water reactors. Which many want to bury rather then use in the unspent fuel that is in them in another type of reactor such a liquid fluoride thorium reactor(LFTR).

Thorium is much much harder(not impossible) to make a bomb out of which helps. A LFTR also transmutates the plutonium and other nasties out of existence which is a good thing.

There is also heavy water reactors which use natural uranium( no enrichment needed).

I question the veracity of your source. A commercial source, they do not appear to account for the haze that high humidity locations have.

Solar is simply too erratic to be much more than a useful supplement. The storage required for a dominant source is tremendous !

Biomass is a mirage. The available land and soil quality in Florida not already taken up by other viable crops (including pulp wood) is small.

Alan

Solar is simply too erratic to be much more than a useful supplement.

Too wide-sweeping of a statement. In FL, it could be combined with nuclear baseload and nat gas turbines to be the basis of the entire electrical supply. It could supply 95+% of hot water needs.

In FL, it could be combined with nuclear baseload and nat gas turbines to be the basis of the entire electrical supply

I agree.

With enough HV DC lines, western Oklahoma could supply, realistically, 6%, 8% or even 10% of the MWh from wind.

Connect to new pumped storage @ Chattanooga and they could be cycled twice/day. At night (11 PM to 5 AM say) excess nuke could fill them, Draw them down in early morning (minimal solar PV at 7:30 AM), fill them up again with surplus solar PV in late morning till just past solar noon (solar PV in Florida will likely peak an hour before solar noon because of clouds & haze). Draw down from mid-afternoon till late evening (say 10 PM).

Imported OK wind fills in as available, either meeting demand instead of pumped storage/FF or filling pumped storage.

Properly proportioned, this will reduce NG demand considerably.

On cloudy days and cold nights, when the nuke-solar-pumped storage + wind as available fails to meet demand, burn FF.

Alan

There is only a small area off the Florida Keys with good off-shore wind (no geographic diversity and not NEARLY enough MWh).

The US East Coast has more than enough off-shore resource to provide 100% of electrical consumption. I'll see if I can find the study.

hurricanes make wind too risky on the coast and off-shore anyway.

I thought building wind turbines rated for cat4 hurricanes was pretty straightforward. Certainly some existing turbines are.

Florida is cloudy.

FL has less sun than SW US, but it has above average for the US.

Many of the most populated areas in Florida will be under water in a few decades. How much should we spend on infrastructure there?

People won't leave Florida - they'll just creep back, following the waterfront.

The trouble is "back" will be flooded long before the Miami area coast is.

Play around with http://flood.firetree.net/ for fun.

By the way, how exactly do hundreds of high-rise apartment buildings "creep back"??

If those now residing in high rises are no longer in them, they will take up much, much more land, but land is exactly the thing that will be vanishing.

That doesn't look good...

Yes, I'm fascinated by the question of how cities like NYC and Miami will deal with rising sea levels. A lot of people have settled in low-lying areas, and moving them will be a big project.

Some people wonder what we'll do when automation and rising labor productivity eliminates most jobs. I should think that relocation will keep us plenty busy...

http://www.nrel.gov/analysis/lcoe/lcoe.html

Nuclear power works out to cost about 8.5 cents/kwh when, amortized over 20 years, so for the remaining 40 years or so of the plants life you would be down to around $0.0186kwh.

A nuclear plant is online 90% of the time and lasts for 60-80 years and PV solar maybe 25 % and last perhaps 20-30 years. So be careful when you quote how cheap solar is when 70% of the time the sun is not shining.

Solar PV lasts longer than 30 years, likely longer than nukes (and see Trojan and Ft. St. Vrain for short lived US nukes), Over a century seems reasonable for a well built silicon unit.

And there will likely be scrap value in the high purity silicon & frame when it is finally scrapped. Scrapping nukes has taken as much money, inflation adjusted, as building them. So add costs at the end to nuke for scrap.

90% CF comes after a decade of good management & operations. LOTS of examples of 60%, 70% and 80% CF. 90% CF is *NOT* a given !

PS: Check out the capacity factor of Brown's Ferry 1 since it first went critical, slightly over a third.

Alan

For every dollar invested, you get about can get about two kwhs of solar for every nuclear kwh, and solar is getting cheaper all the time.

And wind is about 3-6 times cheaper.

Dobhoi, are you kidding us or are you being dishonest? I know you know better!

The article showed wind and nuclear at the same construction cost (but nuclear lasts three times longer). Solar PV was shown to be ten times more expensive, and the cost curve seems to have flattened already.

[Double post]

"nuclear lasts three times longer"

See Alan's rebuttal of this above.

tHEY generate electricity for a nickle and sell it to us for twenty cents. I'd like to see a breakdown of where my .20 goes. How much to the billing deprtment, the lobbyists, the line crew, the lawyers whatever. It costs me fifteen cents for my PV power but I cut out the middle men.

PV capacity in Tucson is 25%. If they're talking Northern Europe or Canada I can believe 16%.

Robert a Tucson

I just did the calculation. The low percentage surprised me.

One thought is that if there is off-grid electricity, perhaps it goes into the total capacity, but not into the electricity produced. I suppose if I wrote to the EIA, I might be able to find this out. It doesn't seem like the off-grid piece would be more than 10% of the total though--but that is something else I don't have a way of checking.

There may also be an issue with the how well the PV is situated, and how well it is maintained. Wind utilization seems to always be below the stated capacity, for a variety of reasons. If it is necessary to wait a few days for a repair, this reduces output, for example. It may be that PV has some of the same issues.

What values did you use to calculate PV actual capacity and where did they come from? Don't forget, the vast majority of residential PV systems in the US are net-metered, so their kwhr output wouldn't show up on any utility summary.

Gail, your figures (graph) for PV is through 2008. Prices have droped dramatically since then. Regarding down time and maintenance, living on PV for nearly 15 years, these have not been an issue. The reliability of my mixed 3.5kw system has been remarkable. Zero downtime for PV. I've had small issues with tracking but the systems default to solar south (making the arrays fixed). Output has exceded projections over time in an area that averages over 65 inches of rainfall annually. I question the numbers. Properly installed, PV is about as reliable as windows. They just sit there and make power when the sun shines. Controllers and inverters have survived lightning strikes and have all lasted well beyond their warranties. I clean panels about twice per year (rain does the rest).

I thought there could be a dirt issue. Snow also.

Snow is a definite issue, though it melts quickly off of my panels. Dust may be an issue is some areas though PVs are great at shedding dirt (witness the Martian rovers unexpected longevity).

All forms of electrical generation require maintenance. My experience with PV is that it is very low in the maint. dept. I spend more time washing windows for passive solar gain. Perhaps we need to compare EROMD (energy returned on maintenance dollar) over 20 years. Or ask how many jobs a coal fired plant provides over its lifetime vs. several large PV installations. Jobs per watt might be a good topic for future discussion ;-)

I spend a lot of time maintaining/repairing things. Fossil fuel generator, log splitter and saws for firewood, our (great) water system, farm equipment, washer/dryer, etc, etc. PV and related equipment? Almost never yet it provides the majority of our electricity. PV has been the best technical investment I've ever made and well worth the cost. As for the big, complicated global energy situation, time will tell, but I expect more individuals and communities will realize that it is worth investing in and adapting to.

Storage, storage, storage......the really big hurdle.

In the ten years I've had my PV panels in operation, I've never had to clean them. The rain carries away any wind-blown dirt.

Snow is really rather easy; all I do if there is significant snowfall is clear the bottom portion of the array. Then the morning sun warms the panels and the rest of the snow slides right off, also providing a cleaning 'swipe'.

Maintenance is a breeze...

no down time? you have 50% down time when the sun does not shine at night.

for those who wish to do their own calculations and compare.
http://www.nrel.gov/analysis/lcoe/lcoe.html

He means no downtime due to maintenance or system malfunction.

Looks like Ghung's solarpanels beat every power plant in the 'availability' factor. Pitty it's the capacity factor that's a bit on the low side ;-)

PS. 'availability' is the amount of time the plant is ready to produce power when wind blows, the sun shines, coal is in stock or the gas pipeline is pressured.

I just did the calculation. The low percentage surprised me.

I'm guessing the PV is being attracted to those places with outsized subsidies. Doesn't New Jersey have a very high renewable energy credit? IIRC it was something like $.69 per KWhour! At that rate, even in a cloudy state, it is too good an investment to pass up. So my guess, New Jersey (and perhaps a couple of other fairly cloudy states) are soaking up the lions share of installations, via massive subsideies. Thats what happened worldwide, with Germany, and Japan outbidding sunnier locations.

80 cents in Ontario

There was an essay here on TOD a few months ago about electricity in Germany, and the capacity factor of solar quoted there was all of 11%.

Which is not bad for their latitude and weather. Much of the US can do much better.

My own system in southern california 16% empirical capacity. I produce 4200 kwh/year from a 3.06 kw system. Given that there are 8760 hours per year this works out to be 16.0%. (The true number might be slightly higher as there are a tiny handful of hours that my 2500W inverter limits the 3 kW system).

-dr

One problem is how they rate panels. Rated power is for somewhat intense insolation 1000watts/meter**2, but the kicker is it is measured at a cell temperature of 25C. Because panels are encased in protective glass, there operating temperature is much higher than 25C, except in very cold windy weather. They lsoe about a half percent per degree C, and typical operating temps are well over 50C. I calculated that on a totally sunny day (including the effects of atmospheric scattering), my panels recieve well ove twice as much insolation on June 21st as december 21st, but a good summer day is only about 1.5times a good winter day. I guess that is one of the selling points of thinfilm, the thermal sensitivity is more like .2 to .25%, so you get more of the rated capacity.

On a partly cloudy day, just when the sun clears the clouds, power output is boosted by two effects. Firstly, the cells are still cool. Secondly you get scattered light from the rest of the cloud. But my 2500watt inverter has never read more than 2498, so I think it is programmed not to exceed that.

Note that 16% capcity factor in the article, was a summer capacity factor, which should be much higher than an annual one.

I assume these graphs don't include the Hydro imports from Canada?

Newfoundland is looking for a market for over 3 GW, Manitoba 4 GW and HydroQuebec has potential of 25 GW (1.5 GW in 4 dams on Romaine River under construction). BC has potential as well.

Alan

There is lots of potential from from Canada, but net exports are surprisingly low, at present. Lots of it is hydro plants running at peak, and then shutting down and buying back at night. BC is the master at this -we are actually net importers by about 15%, but make an annual net profit on electricity exports-imports (assuming California actually pays the bill). BC has a target to be electricity self sufficient by 2020, and should get there, but hydro projects here are controversial - lots of pristine rivers that are being eyed up.
Eastern Canada is a bit more pragmatic (BC has all the treehuggers) and their rivers aren't as beautiful.

As I understand it, these are just US amounts.

I have also left out Hydro pumped storage, except from the overall totals.

Gail. Nice piece of work. A rough calc is showing you are at about 13.7 MWh per capita (4100 TWh / 300 million) in 2007; the EU equivalent will be about half that. As ever, you can judge the state of a modern economy by its electricity consumption. Hopefully, that chart (fig 1), will start turning up again soon.

Your fig 8, I am not sure of. Your column "% of capacity", I am assuming is percentage utilisation. That is, your Nukes are averaging 90.5 percent of what they say on the nameplate. For instance, Wind is showing 29.2 percent, which is about the world average for the good sites.

It is becoming fairly standard to compare different technologies on there "Levelised Cost" basis. The following is a UK example. Don't think you have to factor in Carbon price yet; US should avoid it if possible. Solar PV is not on the map for large scale generation yet. The big EU solar farms have a utilisation factor of around 16 percent.
http://webarchive.nationalarchives.gov.uk/+/http://www.berr.gov.uk/files...
http://en.wikipedia.org/wiki/Moura_photovoltaic_power_station

Regarding Figure 1 turning up again soon, I am not certain this will happen. It seems to me that oil problems will affect the economy as a whole enough that we should expect debt defaults and recession going forward. This will tend to cause a decline in electricity demand. If electricity costs are rising faster than the CPI, this will also tend to depress usage--Jevons' Paradox in reverse.

THe US does use a lot more electricity that Europe. I am not sure that reducing electricity use to Europe's level will be as easy as it might look, because US residences and offices are bigger, and because there is more need for both more heat and cooling because of climate differences.

My percentage of capacity is really percentage of utilization, with one minor difference. I am using the amounts underlying the average of the last four years shown in Figures 5 and 6. These amounts have "Summer Capacity" as their denominator, and "Summer Capacity" seems to be lower than nameplate capacity. I don't understand why this is, and the why the percentage varies from year to year. (I used summer capacity figures because these were more readily available in the EIA data.) For 2008, the ratio of summer capacity to nameplate capacity is as follows:

Coal .929
Natural gas .874
Nuclear .949
Wind .987
Solar .994
Wood .888
Geothermal .688
Waste .862

I wonder if nameplate capacity represents what was originally installed, and summer capacity represents what is available, in summer 2008. Geothermal especially seems to have a huge difference--perhaps part of the installed capacity is permanently (or semi-permantly) off line. The ones that have been installed more recently seem to have smaller differences between summer capacities and nameplate capacities.

It is interesting that the big EU solar farms have a utilization factor of about 16% also.

Growth in generation should be modest.

In the IT world there is a lot of emphasis on replacing workstations with notebook computers, eliminating power usage in networking, consolidating data centers, replacing old power-hungry servers.

Retail and commercial square footage should be going down as on-line shopping and mobile workforce concepts reduce demand for these spaces with their high HVAC and lighting requirements.

There is an emphasis on using energy efficiently in many areas.

The production from thermal power plants rely on a temperature difference which might explain the lower summer capacity. This would be more significant for smaller units that operate with air cooled radiators or air cooled condensers. Plants with cooling towers would face the same problem.

I don't understand this:

As ever, you can judge the state of a modern economy by its electricity consumption. Hopefully, that chart (fig 1), will start turning up again soon.

I didn't know the fact that the US doesn't tax electricity as much as European states do is the discerning factor for the state of the economy? Is the US economy really doing twice as good as the European one? For decades on end? I find that hard to believe...

What I do believe is that the much higher EU tax is one, if not THE, factor in the much lower EU electricity usage per capita (or household), while prosperity is roughly equal. The much lower EU usage just shows me that the US has much to gain in electricity usage efficiency. In other words: the US is unnecessarily wasting a lot of energy, depleting energy sources faster then they need to. On another note, isn't inefficient energy usage somehow linked to a sorry state of development? ;-)

PV calculations:

1. In another Berkeley Study, we find;

(Mis)valuing Solar Photovoltaic Power
It has long been recognized that the timing and location of power production greatly affects the value of the electricity. Combustion turbine “peaker” plants have a much higher cost per kWh produced than baseload coal, nuclear or combined-cycle gas turbines, but they are still worth building for use only at peak electricity use times. Likewise, a high-cost plant located in a transmission-constrained area can make economic sense when compared to the cost and feasibility of transporting the power to that location. While few dispute that the direct cost of electricity from the currently available solar photovoltaic technology is relatively high, proponents point out that the value of the power is also high because of its favorable timing and location.

Conclusion
Using actual real-time prices, the change in value is between 0% and 20%, but using
prices from a simulation model, which assures that peaking gas capacity covers its fixed
costs through high energy prices, the increased value from real-time valuation of solar
power could be in the 30%-50% range
.

2. What is your source for the 16% actual PV capacity? I didn't see it in your reference.

Wind Calculations:

You have 29% as the actual capacity, though more recent installations are refining micro-siting and other techniques to improve capacity, so the actual capacity numbers for recent installs are approximately 36%.

Recent Texas wind installations are averaging 39% capacity factor.

Alan

It is strange that this is not showing up in the calculated utilization ratios.

The utilization ratios by year are

1999 - 35.3%
2000 - 32.4%
2001 - 24.7%
2002 - 30.2%
2003 - 24.7%
2004 - 27.8%
2005 - 31.8%
2006 - 33.9%
2007 - 28.4%
2008 - 31.0%

Maybe manufacturers are overstating real world expected results?

It is strange that this is not showing up in the calculated utilization ratios.
..
Maybe manufacturers are overstating real world expected results?

Both numbers can be right, as they have Different coverage : one is NEW installations, (using better site choice technology), and the other is a industry-wide-average, which likely includes Maint. downtimes, whilst someone quoting 'new capacity factors' will consider an operating turbine, (and likely for less than 12 months )

I am wondering if over time, the quality of site selection goes down. Part of the problem is that the best (windiest) sites are taken first. But part of it is that with renewable mandates, and everyone wanting to get on the bandwagon, wind turbines are being located in places where, with less advanced turbines, the economics didn't seem to make sense.

I imagine that will occur, but not for a while. There are a plenty of good sites.

As prices drop, some may decide to build more on existing windfarms, and that could involve a trade off of a small reduction in capacity factor, to get more MW.

Also, countering this, will be improved blade and operating-area, as well as generator technology, which can all increase the area under the curve.

Superconductor generators may yet make it over the horizon....

I am wondering if over time, the quality of site selection goes down.

No, the quality of micro-siting improves every year, and turbine installations also get higher, both of which show why the capacity of newer sites are higher. Also, turbine designs are increasingly benefiting from field experience, which improves overall reliability. Turbines from the 80s and 90s should be expected to be having lower capacity values as they age, though averaging them into future expected capacity unfairly gives those future installations a 'black eye' they don't deserve.

Part of the problem is that the best (windiest) sites are taken first.

Actually, the lowest hanging fruit are taken first, with respect to states that have the highest incentives (and lowest barriers), those sites closest to existing powerlines, land acquisitions that are the easiest, etc.

All I did is divide the numbers that are shown in EIA reports, based on what is reported to them. I tend to believe this more than wishful thinking on the part of the manufacturers.

Electricity generation comes from Table 3 for Renewables. Electricity capacity comes from Table 4.

For example, for wind for 2009, electrical generation is 70,760,934 thousand kilowatt hours for wind, according to Table 3. Wind capacity at he end of 2008 was 24,651 megawatts at the end of 2008, and 33,542 megawatts at the end of 2009, according to Table 4, making an average capacity during 2009 of approximately 29,096,5 megawatts. Multiplying by 365 days a year and 24 hours a day gives a total capacity of 254,885,340 thousand kilowatt hours. Dividing 70,760,934 by 254,885,340 gives 27.8%.

The calculation for solar is similar.

Detail is available at the state level through 2008, if you go into some of the backup files and do calculations.

I understand your source of PV electrical generation, though these must be from utilities. As mentioned above, most residental (and quite a bit of commercial) PV installations are net-metered, which means the utilities do not measure and report their output.

As we've discussed before, older wind installations aren't as well sited or as high, so using them as part of your data averaging does not capture the current wind capacity values being attained.

I put together some state wind indications for a longer period a while back. This shows that some states are clearly better than others, and there is an improving trend over time. The highest state seem to be Montana, at 36%, with new wind turbines and ideal siting.

In Texas, the average capacity factor of wind farms installed in 2004 through 2005 is 39 percent, compared to 32 percent for projects installed between 2000 and 2001 and 19.6 percent for those installed before 1998. The West Texas wind farms that generate power for the city of Austin’s utility company, Austin Energy, have capacity factors ranging from 35 percent to 40 percent.

This is from an official State of Texas website.

http://www.window.state.tx.us/specialrpt/energy/renewable/wind.php

Both tables cannot be correct.

May I suggest that you took the total year generation and divided by the 12-31 nameplate. Since quite a few wind turbines are erected mid-year, this artificially lowered the capacity factor.

Erect a WT at a 40% CF site on July 1st and it will report about 20% CP that first year.

Alan

Gail, as you see, installations are improving their micro-siting, turbine height, turbine efficiency, etc, so mixing in old wind farm data gives a distorted view of what future wind farm capacity would be.

Austin Energy is a municipal utility and keeps quite good tabs on all the various factors of production. In graduate school I learned a lot from them about how fuel efficiency (heat rate) varies by load, warm-up, uncertainties in load projections for the next hours and economical dispatch. Also transmission losses, etc.

Due to the University of Texas and the desire of many graduates to stay in Austin, Austin Energy has an extremely numeric & analytical workforce @ HQ, more than any other utility I have seen.

If Austin Energy says that the various windfarms they have contracts with have capacity factors between 35% to 40%, I believe those numbers are calculated to at least 5 decimal places and exhaustive statistical analysis has been made (actual vs. predicted, seasonal by time off day, etc.)

Alan

PV installations are net-metered, which means the utilities do not measure and report their output.

Just what the heck is a "net meter" anyway? PG&E replaced my smartmeter with one. But the smart meter, and even the decades old analog meters could run backwards just fine. So any old meter has the needed capabilities. I was assuming that somehow they could get output estimates from my system (to help me renewable portfolio standards), but I don't know how they get it. Otherwise they were just wasting a lot of money replacing meters! My guess is they know the PV capacity of their customers and just use a model feed by weather data to estimate it.

Net metering is simply the legal act of backfeeding power onto the grid from a home or business, letting a meter turn backward. It used to be illegal in most states.

From http://netmetering.com.ar/what-is-net-metering-definition/ ;

Net metering is a program offered by a utility company for customers that install renewable energy systems to generate their own electricity that can be used to offset a portion of the electric energy provided by the utility. Any excess energy generated by the customer during the monthly billing cycle would be sold to the utility company and credited to the customer.

In order to utilize net metering, the customer’s generation must be interconnected to the utility grid with a meter that can register the amount of electric energy that is used and produced during the billing cycle.

What is interconnection?

Interconnection is the physical connection of the customer’s generation source to a utility’s distribution line.

The interconnection required with net metering allows the electric utility to provide the extra electricity or back-up power that self-generation may need. It also assures that safety protocols are adhered to for the protection of the utility linemen.

Interconnection standards vary from utility to utility. Check with your local utility company for the standards required to execute a net metering agreement.

Who can use net metering?

Any customer – residential, commercial or industrial – that is able to generate their own electricity through renewable energy sources or microturbines that can be interconnected to the utility grid.

Net Metering Benefits

* Generate your own electricity and reduce your electric bill

* Protect against future rate increases

* Produce electricity from free sunlight and watch your meter spin backwards

* Reduce the burning of fossil fuels contributing to a healthier environment

* Increase the value of your home

* Save money on your system purchase with state rebates and incentives

I believe that wind installation timing is such that many installations are finished quite close to the end of the year. That means that the beginning figure needs to be weighted by more than 50% in your average capacity figure.

in order to provide peak power at a premium the sun would have to shine 100% of the time. Soar Pv power will increase peak power plants because they are cyclic.

in order to provide peak power at a premium the sun would have to shine 100% of the time.

I don't think you understand what peak power is. For one thing, it usually occurs precisely when the sun is shining.

It looks like the Berkeley figures are indeed very high, probably too high. Let's have a look at average retail prices for solar panels:
http://www.solarbuzz.com/Photos/moduleprices10-9.jpg
The price for the panels has been on average between $4 and $5 per Wp for the last decade and dropping fast the last few years. Someone looking for a good deal could have gone considerably cheaper than that. Add $1 for inverter and materials and the range is $5 to $6 per Wp instead of the $8 to $11 in the Berkeley figures.

The Berkeley study is really a study by Lawrence Berkeley National Laboratories. It is a very impressive 50 page report, with information by state, as well as on a national basis, and many other breakdowns. The study is updated each year, using the same assumptions. According to the study,

This document was prepared as an account of work sponsored by the United States Government.

I think the difference is that the Berkeley study includes considerably more than the solar panels themselves. The cost would include inverter, solar panels, shipping costs, and cost of installation.

Gail I appreciate the response, but I also included the inverter and other materials into the calculation. But I'll take up the glove ;-)

I have experience with a 2000Wp system in Europe, so I'll go with that.
* 2000 Wp needs an inverter costing about 1500 Euro, lets assume Euro = $, so that means $0.75 per Wp. Add misc materials (including rails) and it's $1 per Wp.
* Transporting a pallet of panels over a medium distance assuming a pallet contains about 4000 Wp? Would $0.10 per Wp be in the ballpark?
* Installing 2000 Wp on a roof takes one day by two persons, so that'll be around $1000 or $0.50 per Wp.

So everything included and assuming current average panel price: $4.20 + $1 + $0.10 + $0.50 = $5.80 per Wp. Going bigger would be cheaper per Wp. I wouldn't buy my new system through Berkeley if I were you...

Also I'm sure there are TOD members who bought their system (including labour) for much less then that, say $5 per Wp?

The report makes some comments that may explain some of the differences:

Average Installed Costs for Residential Systems Are Lower in Germany and Japan than in the U.S.

Installed Costs are Generally Lower for Residential Systems than for Similarly Sized Commercial and Public-Sector Systems

The New Construction Market Offers Cost Advantages for Residential PV, Despite the Higher Cost of BIPV Relative to Rack-Mounted Systems

Systems >10 kW with Thin-Film Modules Had Lower Installed Costs than Those with Crystalline Modules

Tracking Systems Had Higher Installed Costs than Fixed-Axis Systems

Module Costs Were Lower for Large Systems than for Small Systems in 2008, While Non-Module Costs Were Relatively Constant Across System Sizes

The calculations seem to be based on what individuals and businesses reported as costs in paperwork associated with different state incentive programs. I presume that those making claims for benefits also had to include copies of receipts.

"Average Installed Costs for Residential Systems Are Lower in Germany and Japan than in the U.S."
Do they say why? Perhaps because the market in the US is still small and therefore there's less experience and competition, hence higher prices?

"Installed Costs are Generally Lower for Residential Systems than for Similarly Sized Commercial and Public-Sector Systems"
Ok, but why would a factory, office building or shopping mall etc install only 1 - 5 kWp systems? They usually have much more room and higher power requirements...

"The New Construction Market Offers Cost Advantages for Residential PV, Despite the Higher Cost of BIPV Relative to Rack-Mounted Systems"
This amazes me. Germany proves that field based ram-foundation installations are considerable cheaper then roof mounted systems. But again, this might be because of the lesser developed market in the US.

"Systems >10 kW with Thin-Film Modules Had Lower Installed Costs than Those with Crystalline Modules"
Oh I would surely hope so, because almost the whole point of TF is that it's promising to be cheaper...

"Tracking Systems Had Higher Installed Costs than Fixed-Axis Systems"
Sure, that's a no-brainer. But does this mean that tracking systems are more used (preferred) in the US because of their higher yields? I ask because tracking systems are more expensive in the EU too, so this can't be the cause of the price diffence. Tracking systems would increase the capacity factor though, which is a nice bonus.

"Module Costs Were Lower for Large Systems than for Small Systems in 2008, While Non-Module Costs Were Relatively Constant Across System Sizes"
Again a no-brainer. Larger quantities of a mass produced product are usually cheaper then smaller ones. Anyway, this figures for the EU and Japan too, so it cannot be the cause of the higher price in the US.

So, to summarize, the only real argument for higher US prices I'm seeing is that the US market isn't as much developed as in the EU/Japan.

Most of these are section headings in the report, and are followed by discussion and charts. Read the report.

* Installing 2000 Wp on a roof takes one day by two persons

My roof has the fake tiles (concrete). For my 2450 system, it took two guys more than three full days. Fortunately for me, they didn't appreciate how hard it would be when they quoted the price, they thought they would get it done in half the time. They had to cut a lot of concrete shingles, but did a very professional job. And I haven't had any leaks!

Ok, that's different then most EU countries where roofs are usually covered with concrete or ceramic tiles (not shingles) hanging from a wooden support (batten).

For that kind of installation you'd just push a row of tiles under the tile row above, place the rail hooks (they hook around the batten secured with a single screw) and pull the tiles down again. There are rails construction systems like click-fit who make the job even easier. It's really quick and easy. I have seen multiple systems of around 2000 to 3000 kWp going up a roof in a single day using only two people.

For that kind of installation you'd just push a row of tiles under the tile row above, place the rail hooks (they hook around the batten secured with a single screw) and pull the tiles down again.

Maybe what you use for battens in Europe is a lot stronger that what we use here in California. Otherwise that sounds really amateur to me.

We put two 5/16" lags into a rafter to secure the bracket to the roof. Then you have to grid away concrete tile material to make room for the bracket so that the tiles lay flat instead of breaking or otherwise letting in water.

And all that grinding usually adds exactly a day to a two day job.

Still, I agree that a 2kW system should take two days.
I've done the roof part of a 3kW system on a shingle roof in half a day as part of a three man crew.

Maybe what you use for battens in Europe is a lot stronger that what we use here in California. Otherwise that sounds really amateur to me.

I think the EU solar installation market is more developed than in the US, it certainly isn't amateurish. Don't forget that concrete roof tiles way much more than solar panels. The additional carrying load is not meaningful. The rails are used to support the people working on the roof too and I've never heard of rails being sheered of the roof by storm or snow. It's a non-issue.

What about these? $2.32/Wp

Add $1/Wp for an inverter. Interesting!

But is EconoSolar a reliable brand? Can one get state rebates with these? (i.e. have they been certified? The implication is that they have been.)

If you add in balance of system, and use a commercial installer the prices come in in that range. My system, adding in subsidies etc, came in just below $8/watt. A lot of larger PV installations seem to be coming in the $6-$8 range. The just approved Abengoa 250MW solar-thermal plant got a $1.4billion loan guarantee, so if I assume that is the cost, thats just under $6/watt.

My previous career we had a similar expectaions problems selling supercomputers. People would look up chipprices and figure they could buy raw microprocessor chips with capability for $X, how dare we charge more for a complete system. We called it the bag of chips problem. A real system isn't just a bag of components.

Please look at my second comment in this thread. I did use a commercial installer and other system components. But I assumed roughly equal circumstances for the EU and US which they are not, apparently. I'm suspecting that the US market isn't as developed as in the EU...

The installed costs comparison is interesting. Nuclear is the most expensive. But...

Wind currently is almost an off the shelf item, standardized products with spec, and product numbers. This is now stage of automotive production: Plenty of models, but fair volumes available, with economies of scale kicking in. And even then it is quite expensive.

In contrast, nuclear is at the level of handmade prototypes. There are 400-500 installed commercial reactors, of who knows how many types. Situation would change if it were possible to make "Honda Civic" of nuclear reactors, a several hundred MW standard reactor, using something less primitive than water moderation.

Most current reactors use 60 year old technology, it's like driving a carburated car with diagonal tires.

On top of that, there is a problem with renewables is that they just do not scale, never mind EROEI, grid issues, $$$ etc.

Which is more feasible, 3 million 1MW wind turbines, or one thousand 1000 MW nuclear plants, to cover 1TW of currently installed power. This argument is only about scalability; nothing to do with grid and actual mix of energy sources.

Thorium/salt cooled reactor to the rescue?

On top of that, there is a problem with renewables is that they just do not scale

This is an unsupported pronouncement.

Perhaps you missed this NREL article that determined that renewables could provide 35% of US western grid electricity.

Not to mention that the NREL study doesn't include DSM, overbuilding, or hydrocarbon synthesis using excess power (from overbuilding).

It is the grid that can take 30% renewables. The issue I raise is that it is nor practical to actually generate those 30% using renewables. For example a Roscoe farm, wind faceplate capacity of 781 MW. USA need roughly 1 TW of installed power, so if the grid takes up to 30%, then we could get 300 GW from wind. But because of capacity factors, we would need to install 1 TW of wind power, so another 1200 Roscoe wind farms.

But even now we covered only 30% of electricity.

And when we start thinking about using renewable electricity to replace non-electrical (heat, industrial processes, transportation) use of gas and oil. ouch.

I believe that in a few years we will finally hear about necessity of going nuclear and we will start building nuclear plants.

Help me understand what you are trying to say.

The study showed that the US Western grid could sustain 35% penetration of renewable electricity, which included 30% wind and 5% solar. The rest would be coal, nuclear, and dispatchable hydro and gas turbine. So yes, if we built wind farms that generated 30% of the electricity, we could realize that level of penetration.

Why would renewable electricity not be available to be used by industrial processes, heat, transportation? How is one electron different from another?

First there is a problem with generating those 30%. It takes HUNDREDS OF THOUSANDS of towers for the West and nationwide millions of towers. Let's say it's done, but it takes 1200 extra Roscoe farms to do it...Then we would want to replace diesel trains with electrical, all the house heat (natural gas now), all industrial heat (natural gas now) etc etc. with renewable electricity. This is even more electricity and we would need several more thousand of these Roscoe farms. Nuclear is so much more dense in a sense of having a large power capacity in literally one spot, taking a few hundred acres at most. Nuclear is coming back, but quietly:

http://www.chinadaily.com.cn/business/2010-03/23/content_9629907.htm

Again, the issue is that nuclear is added in chunks of 1GW at 90% capacity, wind in 1-3MW at 30%. I bet China will use one design for all these reactors. They seem to look cheap at $2.1B per GW, compared to bigger numbers summarized by Gail.

A piece of trivia about nuclear vs coal (not wind): Coal is contaminated with radioactive Thorium and Uranium. The amount of nuclear energy contained in these contaminants is comparable with thermal in carbon. I look for details and post later as my lithium battery is dying :-<

I don't see this an an either/or question; I support both wind and nuclear (and solar, hydro, geothermal, wave, etc).

90% capacity factor for nuclear is typically reached after a decade of debugging and with skilled operations & management. Not likely for new Chinese reactors IMHO.

China is building several different types of reactors now. EPR, AP-1000, CNP-600 & -1000 (local designs) and another pair of Russian VVERs are in the works. Plans for more types.

There is a risk of a common design flaw if all the eggs are in one basket.

Alan

Once those 90% are achieved, the reactor can run for another 50 years, with major refurbishing half way down life span? I understand the reasons for multiple reactors. But isn't it a good reason to design a simple(r) reactor to achieve some economies of scale. They must have good reasons for doing that (multiple designs), and all eggs in one basket can't be enough of a reason. What are the other reasons for that variety of designs?

Some old reactors perform poorly, so 90% CF is not assured and the experience with 50 y/o reactors is very limited. But mid-life, WITH GOOD MANAGEMENT, seems reasonable.

The CNP-600 (a smaller 600 MW reactor is for special cases) like Hainan island that is getting two of them. Limited grid connections seems the likely reason elsewhere.

I think China wants alternatives, not too many eggs in one basket. Build a couple of EPRs and compare notes with AP-1000. Perhaps after building 18 AP-1000s and problems appear, they can switch to EPRs.

The VVERs are going next to two operating VVERs. Always best to keep commonality on-site. Expanding an existing site is always easier than starting a new site. A back-up choice to the EPRs as well. Good politics also.

The CNP-1000 is to create an all Chinese unit. Higher risk of problems, so build AP-1000s in the meantime along with CNP-1000s. My guess is that after 2020, CNPs will dominate new starts almost exclusively.

Alan

Nuclear is coming back, but quietly:

http://www.chinadaily.com.cn/business/2010-03/23/content_9629907.htm

I'm trying to track all the numbers in that link.

It reads like they now plan 28 @ ¥14B, which comes to ¥392B, which is the same budget as the 2005 plan to spend ¥400B. (but on 40)
'Construction of 20 of the 28 reactors has already begun..'

What seems to have changed, is the 40 has declined to 28, due to the higher price ?

You misread. The plan is revised, but not to 28 reactors, but to 70 GW nuclear by 2020. The 28 reactors' construction permits is just PART of the plan, not THE plan.

In 2005, they had some 5 GW of nuclear. They plan to add 65 GW to that number, and as they originally planned for 40 reactors of around 1 GW each, the revised plan is actually quite a bit more aggressive than the previous plan. That's quite impressive - after a third of the time frame has passed, they increase the target by 60%!

Part of the reason for the higher cost estimates is simply inflation. The yuan inflation rate peaked at 6% year-on-year in 2008. But $2B a piece is not too bad.

It takes HUNDREDS OF THOUSANDS of towers for the West and nationwide millions of towers.

Not really. The US has average generation of about 450GW. We'd need 450GW nameplate of wind power to generate 30% of that, at 30% capacity factor. Divide 450GW by 3MW, and you get 150K. That's not millions.

Well a 1MW wind system is small these days. Farms with several larger turbines can add up. And the wind turbines can be put up relatively quickly & easily as long as there is grid transmission support nearby. Those nuke plants a massive multi-billion dollar systems that take 10 years to build.

But in the end, there is no reason to choose . . . do both. And yes, someone needs to start cranking out those Thorium reactors.

On top of that, there is a problem with renewables is that they just do not scale, never mind EROEI, grid issues, $$$ etc.

There is around 120000 TW of solar energy reaching the earths surface.
There is around 2000 TW available in windpower.
There is around 5.5 TW of wave energy available for exploitation
There is around 0.8 TW of tidal energy available for exploitation
Then there's also geothermal, biomass, and other renewables.

Total human electricity consumption is around 16 TW... Renewables have more scale then we will ever need TM

An example: EROEI of kite windpower is possibly up to 375 with no grid or capacity issues and possibly cheaper then any real cost of current production ensemble. As reported by TheOilDrum earlier.

Edit: Never mind. Solar is so small that it doesn't show up. Wow, I'd think it would at least show up as a line. It is starting to get reasonable in price.

And electricity is generated with wood? Where?

Solar shows up just barely (between wind and geothermal) but yeah, it is not much of a line.

And electricity is generated with wood? Where?

Yes, largely from waste from the lumber industry. Among other things, tops, branches, sawdust, bark, and just generally crappy trees that get harvested in clear-cut lumbering are waste that can be burned. Most often in a cogeneration arrangement, where some of the heat is used as process heat and some for generating electricity. The most common generation technology is boiling water and steam turbines, but gasifiers are starting to appear.

Burlington, Vermont, is often used as an example. New Hampshire is also operating at least one plant. Tri-State Generation has at least experimented with mixing wood waste with coal as the fuel in a coal-fired generating plant in Colorado. IIRC, Tri-State is using a fluidized bed combustion system that can burn a wide array of solid wastes, as well as coal. Southern Power is building a wood-fired generating plant in East Texas that will use waste from the lumber industry there.

Austin Energy has a contract for (memory) 100 MW of new wood power near Lufkin Texas (lots of timber in that part of East Texas).

Alan

And electricity is generated with wood? Where?

PG&E claims 4% from biomass. Most of it is wood waste. In Minneapolis when available they add sawdust to the coal. Of course the supply of wood waste, like the supply of used cooking oil is rather limited.

Go and find any pulp mill - they almost always have a steam electricity system for burning their black liquor (the unuseable lignin portion of the wood, typically 25-305 of mass), and these plants can be 10's of MW.

This press release, from yesterday, is for the mill down the road from me. Their electrical system now totals 60MWe. I can see them doing more electricity and less pulp and paper as time goes on!

There are also numerous 20-50MWe wood plants throughout the Pacific NW. Some are attached to pulp mills, but numerous ones have been built as stand alones.

Here is a report from NREL in 2000 that looks at the major systems
http://www.nrel.gov/docs/fy00osti/26946.pdf

The study picked 20 plants, though there are many more in operation. These 20 added up to 700MW total, equivalent to a decent sized coal plant or one unit at a nuke.

My favourite line from this study, as their last conclusion;

Subsidy Programs Do Not Last
As a final note, the Shasta general manager’s list of lessons learned includes this one:
“Beware of entering a regulatory system in which the utility commission or legislature has
determined that it is acceptable for ratepayers to pay the full cost of your technology. Such
things do not last.”

"Wind utilization of capacity does not seem to be trending upward. The average for the four years is 29%, which is better than Europe, but not as good as folks manufacturing new wind turbines are advertising."

I live near an enormous wind farm located in Fowler Indiana. There is usually pretty good wind most days of the year. One thing that I have noticed while driving through this wind farm on the 65 is that usually only 1/3 of the turbines are positioned correctly to turn. I have wondered for the longest why they don't fully utilize what they have. It seems very strange for a company to spend so much on building these and then only using 1/3 of what they could be generating. Any ideas on what may be happening here?

WIND. Are there days when they are all turning; and, the wind is in a particular direction? It is unusual for wind generators to be "constrained off" a distribution or transmission network for system control reasons.

Some wind farms do have down wind wake problems, when the wind is in a particular direction. The old rules were that you needed ten blade diameters space downwind when positioning turbines. And, about five diameters cross wind. Recent farms use a more statistical approach to planting turbines.

If you can get a wind farm up to 40% utilisation you are on a winner. The higher the Hub height and preferably out a sea, the better your chances. Turbines are getting better; they have lower start up wind speeds and higher max wind trip speeds nowadays. And they are getting bigger, up to 7 MW currently. Though they still are constrained by physics; Betz limit and the infuriating fact that if you halve the wind speed, you only get one eighth of the output.

Nope, can't say that I have ever seen more then 1/3 of them turning. The ones that aren't turning are pointed incorrectly.

The reason why summer capacity is used to rate generation facilities is that condenser cooling is least effective in the summer, hence actual generating capacity is lowest in the summer just when it is needed most for air conditioning loads. Capacity payments to producers are often based on summer capacity.

Gail, your solar PV installed cost is a little higher than current costs, but not much. Please remember that the installed cost of capacity is not a good indicator of the relative cost of electricity, especially when comparing fuel burning plants with those that use no fuel. A better comparison would be with the levelized cost of electricity produced, preferably adjusted to negate the effects of subsidies.

Why the explosion in costs for nuclear generation?

Capital costs? In a capital starved economy (e.g recession) getting 5 billion Dollar for a 1 to 2 GW nuclear plant will not be cheap because locked-up capital means a lot of interest needs to be paid while there's no power production due to long build time. And interest rates will be especially high because there are a lot of uncertainties due to the uniqueness of each plant, but also because the risks of a possible accident are practically unknown.

Higher oil costs mean higher costs for everything mined is one contributing factor.

Many mines use almost no oil today, and higher oil prices will reduce the mining oil use significantly. Electric conveyor belts instead of giant trucks as one example of where economics may dictate oil use today but not tomorrow.

Alan

A friend bought a PV system a couple of years ago. The unsubsidized purchase cost divided by the actually delivered watts averaged over a year, was 61 dollars per delivered watt, not peak rated watt! That's the horrid truth.
\
Or putting it another way, that's 61,000 dollars per kW purchase price. a honda genset costs about 250 dollars per kW capability.

"Nother thing that bothers me. People mix up talk about solar PV and CSP something awful. "Now we get solar power by two different ways. PV is one way, and steam or some other kind of heat engine run by mirrors focussing on a receiver, called CSP, is another. CSP is cheaper than PV. ( Period, end of any further reference to CSP). Now --PV is gonna get cheaper and cheaper because--(quantum dots etc etc etc) and pretty soon it will be too cheap to meter" (or some other nonsense conclusion). No comment on the obvious fact that solar concentrators/heat engines- start off cheaper than PV right now, and can be improved a long long way better, so there is no reason to think that CSP will not STAY cheaper in the future.

Example from ancient history. I listened with fascination to the last lecture of the year in gas turbine design. My prof, a renowned expert in the field, told us gullible grad students- " Watch, in only a few years gas turbines will improve to the point that they drive all heavy truck diesels off the road." Ha! here it is 58 years later, and how many gas turbine trucks have blown your fenders off recently as they zip by you on the superhighways??

I call it the fallacy of the last move. The chess novice has a hard time keeping in mind that his is not the last move, that the other guy is gonna have his move next.

A friend bought a PV system a couple of years ago. The unsubsidized purchase cost divided by the actually delivered watts averaged over a year, was 61 dollars per delivered watt, not peak rated watt! That's the horrid truth.

$61/Watt? Are you sure he wasn't building a house or running a drug lab on the side? Solar panels tend to run about $2.50/Watt to $8/Watt installed these days, depending on what panels/mounts/inverter someone goes with and whether or not they install the system themselves. Panel prices have been as low as $1-$2/Watt depending on type (thin-film Si versus monocrystalline) for small orders and seem to have stabilized around $1.5-$2.5/Watt. W/o incentives, the break even price for a solar install is ~$4/W installed provided the owner doesn't have to finance it. W/ incentives it's a lot lower than this, and w/ financing it's naturally higher.

We are looking at the cost, relative to the kilowatt hours actually delivered. The problem was that the solar PV didn't deliver much electricity.

That depends on the install, but at $6/W installed there would be an upper bound of ~12+c/kWh assuming no finance charges. That's assuming the person pays someone to install a system w/ no rebates out of pocket, lets it generate electricity in conditions halfway between Maine and Arizona for 25 years, then chucks off a cliff or something. Realistically most installs will last longer than 25 years, albeit at reduced output, so given a $6/W install price they would probably see less than 12c/kWh. W/ rebates/tax credits that could be less than 6c/kWh, and a DIY installer could see costs of less than 3c/kWh after 25 years.

Around here, according to the notoriously honest local PV dealer, a friend of mine, to get the real purchase $/W of a PV setup, you take the peak watt system cost, like $8/watt, and multiply it by 8 to get the delivered purchase $ per averaged watt over the year. So, 8x8 is 61, right? Right. I got the same number by using my friend's actually measured numbers over the year.

My friend isn't stupid. He bought the PV because he saw a huge subsidy on a thing he wanted for non-economic reasons. The numbers above are for the non-subsidy cost, not the one he paid.

BTW. I really like PV, for stuff like wrist watches and little calculators. I even have some for somewhat bigger jobs like pumping solar hot water. I just happen to like solar thermal better -because that's where I grew up.

Installed cost for PV is just the price of the hardware (panels, the inverter, the mounting, cabling/cutoffs) plus the price of the install. You don't arbitrary multiply the price by another number. If it's $8/W installed then it's $8/W installed. If you're referring to the price of electricity then it's generally about twice the installed cost in cents over most systems warrantied lifespan.

Sorry, but I don't understand the "times 8" bit of your story. The complete system costs $8 per Wp, that's it. Now actually you aren't really interested in $ per Wp but in the $ per kWh over the lifespan of the system.

A simple calculation:
So let's assume an average sunny place then your PV panels would produce 1.5 kWh per installed Wp in a year. You can assume that good quality panels function well over 30 years, which would mean 45 kWh per Wp over it's lifespan. The cost per kWh in that case is $0.17 kWh (I know an iverter replacement every 10 to 15 years should be included, but this is just a simple calculation).

In the EU 17ct per kWh is cheap. Bear in mind that while this 17ct/kWh is a fixed price for the entire system lifetime gas, oil and coal are very likely to become more expensive in the future. Add-in inflation and you'd laugh about the 17ct in 10 to 20 years from now.

If your insolation is 2 kWh per Wp the price goes further down to $0.13/kWh and if your system cost is $5 per Wp then you pay $0.08/kWh.

I think what wimbi is saying is that the capacity factor of solar PV is 1/8th (12.5%).

For a constant 1 watt, 24/7, you need to buy 8 watts of panels. Buy & sell from the grid.

Alan

Thanks, Alan, I can always trust you to say something sane, and many times something highly informative and loaded with good numbers. Your simple explanation is unarguable. It allows me to quit this quibble and go back to having fun making little tiny solar powered stirling engines.

The one I am working on is cheap, cheap, cheap. It will only last a year (constant, 24hr/day) but after that you put it in its box, send it back, and get a rebuilt one nearly free. Or maybe not. For a few bucks more, we could jack up the life by factor of 10. Might make more sense.

For a few bucks more, we could jack up the life by factor of 10. Might make more sense.

Ya think?

Ya think?

Well, maybe. It's not all that simple, ol' boy. Some markets, big ones, won't pay a nickel more for more than 10Khrs. Example, my rusty old 21 year old car's engine still runs like new. Would I have paid a nickel more for a longer lived engine? No. I already knew the engine would outlast the rest of it.

Another example, we tried to make a stirling fridge cooler-- great performance, all kinds of good features, like no startup surge, no CFC's. No sale. To the fridge maker, 50c more was too much. What he had was plenty good enough- and cheaper, which is what mattered.

That's interesting, and too bad. Cooling seems to be an area where Stirlings are better suited than power generation. Given that you have to pay to have a fridge de-gassed now,, there would seem to be a cost advantage, but no one ever thinks of that when they buy a fridge...

Is your Stirling company still around?

I wonder if he is measuring average power over a year. Which would be peak times the capcity factor. If his capacity factor was low say 12% and the system cost $8/watt(peak) that would be about right. Lots of people throw out numbers with units attached to them without getting them right.

Averaged over a year...........as usual folks apply a standard to PV that they don't hold any other power source to.

"a honda genset costs about 250 dollars per kW capability."

Another ridiculous apples and oranges comparison. BTW, I have been through 5 gensets in 16 years (including a Honda). I have 40 PV panels doing different duties, the oldest will be 16 years old next month. Not one failure. Zilch! No oil changes. No noise. No fumes. No fuel to haul. I have installed over 30KW of PV for other folks. NOT A SINGLE PANEL FAILURE IN 13 YEARS. Try holding a genset to that standard. PV is solid state, all other viable generating methods are mechanical, moving parts, high maintenance (CSP included). PV BOS (inverters, controllers) have had a very high reliability factor. Gensets? The bane of my existence.

Finally, my math tells me either your friend payed too much or the install/location was poor. Likely both.....

.....and a lot of people seem to disagree with some of the opinions expressed here regarding PV:

World solar photovoltaic (PV) market** installations reached a record high of 7.3 gigawatt (GW)* in 2009, representing growth of 20% over the previous year.

http://www.solarbuzz.com/Marketbuzz2010-intro.htm

I have 40 PV panels doing different duties, the oldest will be 16 years old next month. Not one failure. Zilch!

Do you have any numbers, for the change in efficiency over that time ?
Some people quote a slow degrade in capacity over time.\

Any comparisons in old panels W/m2 vs newest panels W/m2 ?

Solarworld says to expect .27% loss per year. I think they are
a bit conservative, The Old Arco 33's are very close to nameplate
after 30 years of exposure.

" A PV Module is the closest thing we have to a perpetual
motion (and) is the most reliable electric generator in the know
universe ", Joel Davidson, SOLutions in Solar Electricity

Most warranties are 80-85% output at 20-25 years. Measuring output over time is problematic due to environmental variables and best done in a lab. My experience has been excellent. I have 3 Siemans 80W panels that are putting out full rated power after 16 years.

PV warranties typically allow for 20 percent output degradation over the module’s 20- to 25-year warranty life. But measurements of many modules put into service in the 1980s show that it’s unusual to see even half that much degradation. Many of those earliest modules still perform to their original specifications. It is safe to say that modules carrying warranties of 20 years or more have a high probability of working well 30 years from now.

http://homepower.com/article/?file=HP118_pg12_AskTheExperts_1

Hi Ghung,

I am truly impressed with trhe durability and dependability shown by your pv modules, especially considering you have indicated they are a mixed lot from different manufacturers.

But I have a very hard time believing that new panels purchased today or in the future will prove to be as reliable.

There is no doubt in my mind that the bean counters will find ways to cut corners on materials and assembly to the point that problems become very common , perhaps even with new panels, in a lot of cases.

And so far as collecting on a twenty year warranty is concerned, nobody has yet posted any details about the terms;a prorate returning say ten percent of the purchase price eighteen years from now wouldn't be worth filling out the paperwork, considering inflation.

And anybody who thinks the company you are buying from today is necessarily going to be around and in financial good health even ten years from now is poorly informed;wars, mergers,bankruptcies, bank panics, and plain old thievery, not to mention honest problems such as the death of your small businessman dealer, are facts of life.

I believe the only warranty I would REALLY count on would have to be gauranteed by a third party such as an insurance company.

Being a cynical old farmer, right now today I would probably have MUCH more faith in used modules from a well known manufacturer than brand new ones from a recent startup.Especially a third world start up, or an American startup.

What do you think?Anybody else?

Yeah, Mac, there have been issues with warranties. Since I haven't seen any failures I haven't had the chance to test the process. My panels are Siemens, BP and Kyrocea. I've installed and some other major brands. I have always purchased from a couple of well established suppliers who have been great at handling warranty claims on other equipment and I expect that they would make sure that any PV they sell is covered. These guys would likely eat the loss on anything they sell even if the manufacturer doesn't.

I'm trying to bring up a post I made a couple of months ago about Shell Solar after they aquired Siemans Solar. They created a huge network of seller/installers in Asia (mostly) and pushed millions of modules on the market. Many have failed and the warranties are not being honored. There are actions going on in several countries.

COLOMBO, Sri Lanka, Jan. 5 (UPI) -- The World Bank and green energy companies have accused oil giant Shell of not honoring warranties on solar power systems sold to the developing world, the Guardian reports.

http://www.upi.com/Science_News/Resource-Wars/2010/01/05/Shell-fails-to-...
The gist of one article was that this has dealt a blow to the renewables market in places where the World Bank was promoting renewables. Despite the setbacks, folks in that part of the world are still adopting PV in a big way due to their unreliable grids.

Shell Solar has morphed into SolarWorld, IIRC.

Shell Solar has morphed into SolarWorld, IIRC.

If by 'morphed' you mean "was acquired by", then yes.

Hey Ghung. What you say is right! I got no problem with it whatsoever. I made the comparison with a honda to get the response you gave. After all, I'm trying to sell small heat engines based on the fact that they beat the hell lout of that bane of your existence.

I am all in favor of solar, I just happen to think that we can make a solar thermal gadget that will have the SAME reliability as a PV panel and be cheaper. Also puts out 60 Hz AC directly, if you wish.

Don't believe me? Why should you. But fortunately these are not faith-based engines, they work regardless.

Here are some examples of things with moving parts that last a long time.

good wrist watches
mediocre refrigerators
dirty old industrial electric motors.
water turbines
human hearts (if lucky)
NASA- small stirlings for isotope power to outer planets-continuous operation for 15 years, over 35% thermal efficiency.
lots and lots of others.

People think moving parts = short life. Not true. Hey, how 'bout the solar system?

Don't be offended, but I think you are making a mistake which is common for people that did not go through thorough physics training: thermal power needs large temperature differences to be able to do real work (work=energy) and that very high temp differences are needed to achieve meaningful efficiencies. Without sufficient efficiency your EROEI is very likely to be very low and therefore the economics will be bad too. I'm referring to the Carnot cycle here (there are other methods that are more efficient but are also more difficult).

Wrist watch: very low power, hence no stresses
Refrigerator: scroll pumps don't last 10 years at e.g. 50% duty cycle (they will become less efficient over time). Otherwise it's just a heat pump like many aircon suppliers (mitsubishi, daikin etc). You'd have a hard time competing with these industrial giants.
Industrial e-motors: but what do you want to do with them? Attach them to a stirling engine? E-motors are durable but by themselves they don't produce anything.
Water turbines: need a large vertical drop and lots of water to produce meaningful amounts of energy (force = mass times velocity)
Human heart: constantly rebuilt and maintained, doesn't work without it for more then a few minutes.
Space based Stirlings: have a very large temperature difference (temperature of space which is a few degrees above abs zero) but need to radiate the energy to get a heat flow (convection doesn't work in space). Stirlings can be made reliable but hardly as reliable as solid state PV if a lot of money is spent on expensive and durable materials.

Solar system? How about no friction, no wind, dust, rain or kids fooling around?

Please accept this humble word of advice: Please, please, consult a trained physicist first if you plan on investing a lot of your hard earned money on a new wonder machine that should efficiently convert solar energy to electricity by means of mechanical power. Also, have a very close look at the current commercial Stirling and solid state PV systems first.

Styno, I think you are being a bit harsh on some of these technologies. Many refrigerators run for decades and decades without failure - they get less efficient, but so will most moving things over 30+ years, and these have had zero maint.

One of the big refrig companies, Carrier, has used their off the shelf components to make an Organic Rankine Cycle system, that can run off a temperature gradient as low as 50C.
http://www.pw.utc.com/vgn-ext-templating/v/index.jsp?vgnextoid=7400b9246...
Granted, the Carnot efficiency is terrible but this is a hot springs resort in Alaska that has unlimited warm water. They put in two of the 250kW units and turned of their diesels, saving huge $$ . Lots of scope for waste heat and low grade geothermal or concentrating solar when you only need a temp difference of 100-150C. Their systems cost about $1.50/Wp, so if you can engineer a cheap solar collector (pipes under black metal roof, etc) you can have a cheap solar thermal system. It;s just the concentrating part that gets expensive

Hydro plants don;t need high head - they are more cost effective, for sure, but you can buy, today small or large hydro systems that work on as little as 1m head. There are many, many opportunities for small hydro.

I will agree that he should be talking to an experienced physicist or engineer first, but there are lots of ways to do cheap systems. The commercial ones almost always and up being the expensive ways.

I will also agree that he should be very careful of any Stirling system. This has been the perpetual also ran since forever. To get high efficiency needs high temp and pressure, and at low temps, and ORC system will be better. I did see a project done using car a/c components to make an ORC system with parabolic dish - far cheaper than any stirling could be.

I actually like this approach, from a former NASA fuel cell engineer, though he hasn't had commercial success yet either, but it sure is a very simple concept;
www.proepowersystems.com
Neatly sidesteps most of the problems of steam, ORC and Stirling. Engines will have just as little torque, and most efficient operating pressures at at 3-5 atm, so cheap to build.

But, just another of hundreds of miracle engines out there - choose wisely, if at all.

Styno, I think you are being a bit harsh on some of these technologies. Many refrigerators run for decades and decades without failure - they get less efficient, but so will most moving things over 30+ years, and these have had zero maint.

You might very well be right, but I've measured a few older refrigerators (~15 yr) and they consumed 800 - 900 kWh/yr where they were advertised as ~300 kWh/yr. So low maintenance they were indeed, but the efficiency dropped by 2/3 in that time period. The Carnot cycle isn't that efficient to begin with, let alone when wear and tear multiply that efficiency by 0.33.

One of the big refrig companies, Carrier, has used their off the shelf components to make an Organic Rankine Cycle system, that can run off a temperature gradient as low as 50C.
http://www.pw.utc.com/vgn-ext-templating/v/index.jsp?vgnextoid=7400b9246...
Granted, the Carnot efficiency is terrible but this is a hot springs resort in Alaska that has unlimited warm water. They put in two of the 250kW units and turned of their diesels, saving huge $$ . Lots of scope for waste heat and low grade geothermal or concentrating solar when you only need a temp difference of 100-150C. Their systems cost about $1.50/Wp, so if you can engineer a cheap solar collector (pipes under black metal roof, etc) you can have a cheap solar thermal system. It;s just the concentrating part that gets expensive

All very well, but temperature differences are one thing. You will have a hard time making 100 degrees deltaT with a black roof and some piping. Carnot efficiency at 100 deltaT is still very poor, and Stirling engines are Carnot cycle engines no? Perhaps vacuum tubes would do better in that regard. The other thing is the amount of solar energy that hits the surface. Solar panels are already at 20% commercial efficiency. I would be very very impressed if there is a mechanical alternative that beats that with the same low maintenance and installed cost. Call me sceptical. That's why I wrote my earlier message, solar physics on earth has very real limitations that prohibit cheap simple, low maint. mechanical power.

Hydro plants don;t need high head - they are more cost effective, for sure, but you can buy, today small or large hydro systems that work on as little as 1m head. There are many, many opportunities for small hydro.

Physics limits the amount of power you get from an amount of water give a certain hight. 1m lead needs LOTS of water to get meaningful power. I know a boat prop is about 30% efficient max. Add a tube casing (i.e. pump) and you'll get perhaps 60% efficiency max.

I will agree that he should be talking to an experienced physicist or engineer first, but there are lots of ways to do cheap systems. The commercial ones almost always and up being the expensive ways.

Why? You should ask yourself. Why has noone made a cheap reliable and efficient mechanical way in harnessing the sun's power? Because it's darn difficult. So, if someone claims to have the simple answer that noone else has come up with (but it just needs some engineering and sometimes a little investment), then anyone should be sceptical.

I will also agree that he should be very careful of any Stirling system. This has been the perpetual also ran since forever. To get high efficiency needs high temp and pressure, and at low temps, and ORC system will be better. I did see a project done using car a/c components to make an ORC system with parabolic dish - far cheaper than any stirling could be.

The O-part in ORC is just a marketing concept, it's just a Rankine cycle machine which is in essence also a Carnot cycle. There are no miracle efficiencies to be gained. If you use it on waste-heat then anything is better then nothing, but using the Carnot cycle as primary power on a low temp difference (i.e. roof mounted solar) is just a headache.

Please think hard about the basic physics: solar insolation (W/m2) and Carnot efficiency before attempting to spend a lot of money on the miracle solution.

You will have a hard time making 100 degrees deltaT with a black roof and some piping. Carnot efficiency at 100 deltaT is still very poor, and Stirling engines are Carnot cycle engines no? Perhaps vacuum tubes would do better in that regard. The other thing is the amount of solar energy that hits the surface. Solar panels are already at 20% commercial efficiency. I would be very very impressed if there is a mechanical alternative that beats that with the same low maintenance and installed cost.

Well, 100C may be pushing it for a simple system, but you can get easily over 70. This is an example of the way I am thinking, a clever way to use the roof as the concentrator.
http://www.builditsolar.com/Experimental/RidgeVentHeater.htm
A black metal roof (e.g. of a large farm or industrial shed etc would be a good candidate.

On our farmhouse back in Australia, we heated the pool by pumping water onto the roof of the house with a garden soaker hose, so the water ran onto our black tile roof, and we tapped into the downspout by the pool. Didn't believe my Dad when he said not to touch the water because it would be too hot, but it scalded my hand. The point being, you may be able to take advantage of existing roof area, very cheaply, compared to using flat plate or tube collectors. Or, to get the best of both worlds, use the expensive panels/tubes to do the last part of heating from 70 to 90C.
I did not say anything about using Stirling engines, either, I am talking about the ORC system. With a capital cost of about $1.30/W, all you need to do is get enough hot water to it, and you are making electricity. So then it becomes a question of what is the most cost efficient way to get the hot water. A drop in output temperature, in exchange for a large drop in cost, will be worth it in some situations.
This is not to say it's a panacea, but the ability to make electricity from hot water, not steam, does open up new options.

1m lead needs LOTS of water to get meaningful power. I know a boat prop is about 30% efficient max. Add a tube casing (i.e. pump) and you'll get perhaps 60% efficiency max.

sounds like you need to get a better prop for your boat. 30% is at the low end, properly sized and pitched props can get up to 70%
(http://www.psychosnail.com/InfoBoatPropeller.aspx) Shrouded props on slow speed tugboats can do a bit better.
Anyway for hydro turbines, you are about right with 60% for low head operations. You can use centrifugal pumps running backwards and get up to 80% with carefully selected pumps (I have worked with a company that does just that for their business). Other turbine types like the crossflow can get 60% out of as little as 2m of head - they are sometimes used at drop structures in irrigation canals.
These guys can sell you a small, off the shelf low head system, they get from 45 to 52% eff at 1 to 3m http://www.microhydropower.com/Low%20Head.htm
Of course, larger units are more efficient still. There are many places with large flows and low head, like irrigation canal drop structures, outlets from sewage treatment plants, and many rural creeks and rivers and small cascades. lots of opportunities there (I have built micro hydro systems so I can speak from some experience here)

So, if someone claims to have the simple answer that noone else has come up with (but it just needs some engineering and sometimes a little investment), then anyone should be sceptical.

Quite so, but that doesn't mean the person is wrong. I used to work with an electrician at a ski resort (where I managed the energy and water systems) who was amazing for coming up with ways to do things, cheap and simple, that the electrical engineers never thought of. Their solution to a situation was $60k and he would come up with a $5k way to achieve 80% of the result. this happened enough times that the engineers would ask him first before venturing their own ideas. There are many times (not all) when the 80/20 rule works.

With solar, thermal efficiency is only one part of it cost efficiency normally more important, and less efficient ways can sometime be simpler and more profitable, especially if they avoid "custom engineering"

The O-part in ORC is just a marketing concept, it's just a Rankine cycle machine which is in essence also a Carnot cycle. There are no miracle efficiencies to be gained. If you use it on waste-heat then anything is better then nothing, but using the Carnot cycle as primary power on a low temp difference (i.e. roof mounted solar) is just a headache.

Please think hard about the basic physics: solar insolation (W/m2) and Carnot efficiency before attempting to spend a lot of money on the miracle solution.

I don;t know if I'd call it just marketing, it is a useful way to distinguish these from steam. And Rankine cycle is not really the Carnot cycle - that one involves isothermal expansion and compression, neither of which occur in the Rankine Cycle. The Rankine cycle is, of course, subject to the Carnot cycle limits. No one has ever claimed miracle efficiencies, what the ORc is able to do is create useful work out of lower temp differentials than steam systems can, and that is significant. It is allowing many low grade geothermal and waste heat sources to be productive.
As for solar power, I'm not saying it's a miracle solution, but there will be situations when it is the best one, especially where hot water is a desired output of the system, this allows you to squeeze out some work along the way. The Sydney Olympic pool has a huge rooftop hot water system, gets up to 70c on a good day. The off the shelf ORC was not available when they built it in 1998, but they have lots of m2, so lots of W, a hot water system that already exists, and a use for all the waste heat - that would appear to be a good application for ORC, as everything else is already there.

You can;t change the laws of physics, but you can sometimes game the economics, especially when some parts of the system are already in place or available for free- those are the sorts of projects I look for for a living.

Well, 100C may be pushing it for a simple system, but you can get easily over 70. This is an example of the way I am thinking, a clever way to use the roof as the concentrator.
http://www.builditsolar.com/Experimental/RidgeVentHeater.htm
A black metal roof (e.g. of a large farm or industrial shed etc would be a good candidate.

70 degrees is what's called: very low quality heat. Combined with Rankine or Carnot cycle this has difficulty to do a lot of power as you also need to substract outside air temperature (say 20 -30 degrees, which leaves only 50 degrees max to work with). Vacuum tubes with special liquids (like the cooling liquid used in cars) can do up to 140 degrees, but then volume is probably low and therefore the amount of heat too). Really, simple collectors do best in combination with e.g. underfloor heating (deltaT is biggest and Carnot efficiency isn't at play).

These guys can sell you a small, off the shelf low head system, they get from 45 to 52% eff at 1 to 3m http://www.microhydropower.com/Low%20Head.htm
Of course, larger units are more efficient still. There are many places with large flows and low head, like irrigation canal drop structures, outlets from sewage treatment plants, and many rural creeks and rivers and small cascades. lots of opportunities there (I have built micro hydro systems so I can speak from some experience here)

Sure, it looks like a cool machine, but I have some reservations:
- Wear and tear. How long does that prop work efficient?
- The cheapest variant is $2300 per 1kWp, Solar PV will last at least 30 years for $4000 per 1kWp
- How much water is needed to produce 1kWp with a 3 meter water drop? I cannot find the number, but it looks like something in the order of 10l/s which is a significant flow.

But like I said, it's a cool piece of hardware. I would like one (if just only to measure it) if only there was a stream with sufficient gradient nearby...

With solar, thermal efficiency is only one part of it cost efficiency normally more important, and less efficient ways can sometime be simpler and more profitable, especially if they avoid "custom engineering"

Sure thing, but does it work reliably for many years without maintenance in harsh conditions (blistering hot, freezing cold, hail, wind and snow)?

The hot water collectors placed on the White House under Nixon's tour are still working fine (albeit at some other place because Reagan had them removed). Some of the very first PV panels recovered from satellites launched in the 70s still work fine...

I don;t know if I'd call it just marketing, it is a useful way to distinguish these from steam.

Well, Organic means there is an organic medium used instead of a non-organic medium. So, apart from the boiling points the rest is exactly the same, hence I called it a marketing term: it's just a Rankine Cycle system.

And Rankine cycle is not really the Carnot cycle - that one involves isothermal expansion and compression, neither of which occur in the Rankine Cycle

Sure, Rankine phase change of the working fluid while Carnot is without phase change. While Rankine cycle machines are used to process low quality heat like waste- or solar heat, they also follow the Carnot cycle closely albeit not efficient. For example an ideal Rankine cycle machine can get 63% efficient at 515 C deltaT, but actual machines only reach about 43% efficiency for that temperature. Lower temperatures result in much lower efficiencies.

So if you live in a very hot (i.e. very sunny place at low latitudes) environment then ORC can make electricity from solar waste heat at low efficiencies. But I still think that PV will have higher efficiencies, lower maintenance and longer lifetime and a wider application area (i.e. work at high latitudes as well). The ORC solution should be really really cheap to compete. And quick to market, because thin-film PV will be very cheap too within the foreseeable future.

CSP is cheaper than PV. ( Period, end of any further reference to CSP). Now --PV is gonna get cheaper and cheaper because--(quantum dots etc etc etc) and pretty soon it will be too cheap to meter" (or some other nonsense conclusion). No comment on the obvious fact that solar concentrators/heat engines- start off cheaper than PV right now, and can be improved a long long way better, so there is no reason to think that CSP will not STAY cheaper in the future.

Firstly, PV and CSP target quite different usages :

* Most new CSP now includes some storage, which PV lacks.
* CSP usually needs water rights, and some volumes for cooling.
* PV can install in places CSP cannot touch, such as on roofs
* PV panel prices are falling so quickly, tracking is less mandatory.

If you look at a large multi MW site, there are many common elements:

Both CSP and PV need land area, and tracking, and both need the same sub-station transformers & switchgear.

So, the [PV Cell cost], trades off against [Mirrors+Boiler+Turbine+Cooling]
- and one is coming from high volume fabs....

Seems the saving grace for CSP looks to be the easier storage (and it can also Gas-Augment, to better use the Turbine+Cooling+sub-station transformers & switchgear investments, provided of course, their site has a Gas supply ....

I have not seen a design yet, but falling PV Cell prices, could open up a Dual PV+CSP approach, where you use the infrared for CSP and the visible for PV.

jg. what you say is good, but I happen to know that very very small CSP is here and now. I can easily pick up a 1kW heat engine that will convert sunlight into any kind of electricity you want at a thermal efficiency of over 25% and at a cost per watt of about half a dollar manufacturers price based on typical multiplier of commodity cost in large numbers. This thing can easily compete with roof mounted PV in places with bright sun that can be concentrated.

I can visualize, for example, a solar thermal system mounted on a big box store roof in LA generating more than the building uses from a concentrating thermal converter, coupled to a gas booster for 24 hr service, and including a pumped hydro storage.

Pure fantasy! So, show me where that little daydream violates any law of physics or any experience in whittling out steel.

And I like your suggestion of a combo of PV and CSP, sounds natural, and sure to happen. And I agree about the advantage of gas boost and storage for CSP.

My gripe in all this is that people simply don't seem to want to factor in the obvious- that EVERYTHING gets better with time and experience and effort, not just PV. But that effort has gotta be there, and it ain't.

But I have a saving hope. We here in the good ol' USA may indeed be incurably stupid, but the Chinese are not. So my grandkids might still have a moderately good life as long as they can sell anything to China that the Chinese would care to buy-- like movie scripts on Armageddon.

Can you provide a link to the 1kW heat engine? A detailed datasheet would be most welcome!

http://www.microgen-engine.com/

That one is over 20 years old in design. The new ones are very similar in basic layout, but way lighter, more efficient, and cheaper, but not yet on the market.

BTW. Thanks for the kind advice about getting some tutoring from a competent physicist on basic thermo before mouthing off on TOD. It is obvious that my somewhat flippant remarks above are easily misinterpreted as was in fact done in several comments. I should have been more explicit. Of course I know a honda is a piece of crap for long term use. That was what I was getting at- people tend to buy on first cost and to hell with life cycle cost. I know that one all too well from years of fruitless effort in trying to sell on the basis of life cycle costs.

PS. I have advanced degrees in thermal machines from two of the most rooty tooty engineering schools you can name, and taught engineering thermo for 15 years. Now I am told I know from nothing. Dam, all that academic stuff-- a waste of time! Why didn't somebody tell me before this??

PPS. I remember a fascinating lecture by Edward Teller, a physicist fairly well known at the time, on the nature of negative temperatures. Teller's leg squeaked and he needed a shave. He could play the piano pretty good.

PPPS On life of things with moving parts. I was talking about the entire class of things with moving parts that last a long time, heat engines can be in a subset of that class.

Thanks wimbi, much appreciated.

I happen to live in a country with a rich source of nat gas and accompanying developments. The micro-gen is a similar development as we've seen here where nat-gas is burned and the heat is used to run a Stirling engine. Now, the heat to electric production of these machines is about 8 to 1. In other words, you generate 8 parts of true waste heat to 1 part of generated efficiency (12%). That is how efficient the Stirling engine is: not very. And this concept has the advantage that it runs on nat-gas and therefore at much better quality heat (higher temperature) then collected solar heat (vacuum tubes or hot-plate).

The whole idea of the micro-gen is: you use the low quality waste heat for warming a building and there is some electricity production as a bonus. It is NOT an efficient electricity power plant by itself. But you apparently know more about the future developments, it would be great if they'd become a viable alternative. Your qualifications are much better than mine. I'm sure my general physics degree hasn't been as thorough on thermal machines so I'll be looking forward to hearing good news of your new enterprise.

I happen to live in a country with a rich source of nat gas and accompanying developments.

I see their road-map claims :
September 2010 : Scheduled commercial availability.

So, they should be installing in numbers, right now ?

It seems to target Dual-heat of Central heating, and Water heating, with some bonus electricity.

A bit sparse on info, on how independently those three can interact ?

So is best suited to colder climates, and those with costly, non hydro grid power.
- and a reliable gas supply.

Yes, a field trial period is already running since 2008. Don't know if they're selling in significant numbers now. Haven't heard much about it lately actually. It doesn't surprise me as the boiler uses more nat-gas for the same amount of heat and feed-in is still troublesome due to bureaucracy at various levels (government as well as the commercial monopolies). The whole concept seems like a good way to ensure more nat-gas use by civilians who don't know what gas prices will to compared to electricity feed-in income. I wouldn't touch it.

Yes, it is a Dual-heat (building warming and hot water) with bonus electricity (at a cost of more gas usage).

Normally these systems provide either building warming or hot water. If it's warming the building and hot water is needed then it will switch quickly until hot water demand is gone and go back heating the building. Electricity is produced in both situations.

I agree with your assertions that they would be best suited to colder climates where gas is cheap and abundant compared to electricity.

These folks are combining PV and solar thermal, in what seems to me a completely natural way:
http://www.pvtsolar.com/home.html

See the Flash demo here:
http://www.pvtsolar.com/how_demo.html

I posted this link in another thread and an employee of a competitor in the UK pointed out what he considered to be design flaws in this system. A quick search should turn up the discussion if someone is interested. It was in the last week.

Well, from a solar physics point of view it's not completely natural... Solar panel efficiency degrades with higher temperatures while on the other hand you want the boiler to store as much heat as possible (warmer water). These contradict each other. It's more a compromise solution for people with little roof surface.

Edit: Oops I assumed it was a combined solar hot-water PV panel system. PVSolar uses a fan to suck in colder air under the PV-cells where it's heated by the solar panels and then fed to a heat-pump for hot water production. Sorry. Yes, this seems a natural combination that benefits PV electricity production. I only worry a bit about the temp differences between the lowest and highest mounted panels, but overall, it could be beneficial.

On second thought, I wonder how optimal this system is: PV thermal degrading is mostly a problem for the summer months while warm water needs are mostly required in the winter.

Here is the sub-thread link:
http://www.theoildrum.com/node/6933/716406

His point is valid, about "heat quickly becomes superfluous in areas where irradiance is optimum."

I was actually thinking of CSP dual use, but these examples are great to see, and
I like the cooling concept, to increase the PV output.

Missing is a Watt budget, of the cooling energy needed, and the gain in PV that results.

I don't think people are going to go with big satellite dish looking reflectors w/ Stirling engines in 'em instead of PV panels, but they might be cheaper for large scale installs. Large scale thin film solar hit ~$3+/W installed cost in 2008, so it'll be tough for SES to beat that. We'll know in a few years once the two SES projects are finished.

http://www.greentechmedia.com/articles/read/first-solar-reaches-grid-par...

http://www.renewableenergyworld.com/rea/news/article/2010/09/top-6-utili...

Or putting it another way, that's 61,000 dollars per kW purchase price. a honda genset costs about 250 dollars per kW capability.

Try to avoid comparing apples to appleseeds. With the honda genset you also have to pay for FUEL.

Assuming (generously, I think) that 1kW of honda genset can run for ten hours on a gallon of gasoline, and that gas costs 3$, then it will cost $.30 to have 1kW for an hour.

A PV system is usually expected to last for 25 years. Since there are (count 'em) 219150 hours in 25 years, your honda genset will cost you $65745 per kW to run for 25 years. Plus $250. If the motor lasts that long and gas prices never rise.

I agree that your friend is not stupid.

I think this data needs both expanding and adjusting.

Figure 7 Estimated Cost of New Generation gives capital cost per kilowatt and Figure 8 gives capacity utilisation. By dividing the number in each technology category we get an idea of how much overbuilding is required for the nameplate output to be produced. For nuclear we get 1/.91 = 1.10 so adjusted capital cost increases by 10%. For solar PV 1/.16 = 6.25 so overbuilding increases costs 525%.

I don't see a table for Levelised Costs of Generation. I think that needs to be included both for current costs and also factoring in possible future carbon costs. Thus if pulverised bituminous coal without carbon capture worked out at say 6c per kwh I would add 2c for a future carbon tax so the levelised cost goes to 8c. That 2c is similar to what the Australian Greens Party want. They might get it and who knows it could spread to the US if Obama gets re-elected.

Just want to add that I, for one, appreciate electricity getting some added attention. Thanks!

It seems like quite a few people understand oil (or think they understand oil) and don't want to bother figuring out electricity.

When comparing prices of different sources we also need to compare what those sources are competing with as opposed to other sources that in general are all or part baseload/large scale dispatchables. When all is said and done we need to end up with a weighted cost of electricity comparing a specific wind or solar install to whatever they are competing against, not to baseload or large scale dispatchables, unless that's what they're competing with.

Solar for instance is cost effective in most cases because it competes w/ natural gas peaker plants that are only operated during the summer months. It may not be competitive with load following natural gas in the winter, but because of the large differential between it and peaker natural gas in the summer, as a whole it's cost effective. Wind power installations tend to follow the same trend although they compete w/ more sources. That's why capacity factor tends to be lower than the manufacturers advertise. Most developers will gladly reduce capacity factor by 25% if they can compete with a producer that's twice as expensive.

The indicated solar PV cost is extremely high compared to the other costs. I can delete the solar PV indications, if someone can explain how they are incorrect. But they are somewhat concerning.

My take would be they have tried to include the capacity factor, in their W numbers.

If you do that, then the average watt price of course elevates significantly.
- but everyone else talks about nameplate watts, and includes capacity factor on GWh and pay back calculations, so it is a muddled thing to do.

You can reality check component prices easily enough, even in smallish volumes

http://www.ecobusinesslinks.com/solar_panels.htm
http://www.ecobusinesslinks.com/inverters_sma.htm

Shows ~$1.55/watt for panels, and ~50c/watt for inverters.
So those are 2010 spot component prices.

Then you can do a reported industry revenue work back, which gives

2008: $37.1B/5.95GW = 6.23 $/W
2009: $38.5B/7.50GW = 5.13 $/W

which shows there are good margins in the built-outs, and grid connects, but many of those costs are common to all power projects.

and factory-costs are under $1/W, but need a margin added,

First Solar claims ~76c/W, and there is this too...

http://www.oerlikon.com/ecomaXL/get_blob.php?name=100907_PressRelease_Th...

["...“ThinFab” for manufacturing of thin film silicon modules, which will achieve record breaking production costs of € 0.50 per Watt peak ....
Oerlikon Solar developed a new champion Micromorph® lab cell in cooperation with Corning Incorporated with 11.9 percent stabilized efficiency "]

fairly close to FirstSolar, but using Silicon.

Figure 5 & Figure 1 seem to conflict ?

In Fig 1 it is clear the recent decline was mostly shouldered by COAL, so one would expect the Coal capacity factor to decline in Fig 5, but it seems to hold steady ?
Did most of that reduction come from closed plants, or is there some other reporting reason to miss the coal decline in Fig 5 ?

There is one more year of data in Figure 1 (ends 2009) than Figure 5 (ends 2008). Data regarding capacity was only available for renewables for 2009 for some reason.

Related to, but not included in the post (very good BTW Gail) is the relative cost of electricity as compared to foundation energy sources. Another poster on TOD produced a very good synopsis of the cost of a kWh from typical FF sources. The bottom line, including efficiencies, etc. is oil electrical generation is 0.12/kWh, or $120/MWh.

This is very interesting because we are working with $120/MWh for a typical - if not acceptable upper end price - for new electrical generation. I can't think of a clearer indication or empirical fact that supports FF's as the foundation energy source of our civilization as we know it. We know this somewhat theoretically, viscerally, but here is a hard and fast physical number that stands up to scientific or engineering scrutiny.

The message to society and industry is all boats rise on a tide and electricity is included. This is the new bar to clear (that's kind of personal because I was a high jumper back in the day). If you are going to develop an electrical generation project, make sure the delivered price to the retail utility (wholesale) is 0.12/kWh or less. If oil rises in price, then adjust accordingly; but keep in mind we are approaching a societal, entropy related Reservation Price on FF's.

I must asdmit that my knowledge of electrical theory and practice beyond simple grid ac 120/240 and automotive type low voltage dc is VERY limited.

But I simply can't see why low voltage dc appliances are so expensive;surely a 48 volt dcmotor can't be that much more expensive than a 120 volt ac motor, and there are so many off the shelf mass produced cheap electrical gizmos available that I don't see a need for a single purpose built dedicated electrical part in a 48volt dc refrigerator.

Ditto a washing machine,small airconditioner, many other appliances that could run directly off a panel array when the sun is out and bright;this would lower costs, simplify the system, and increase efficiency considerably- at the expense of convenience of course.

Somebody fill me in, please.

But I simply can't see why low voltage dc appliances are so expensive;surely a 48 volt dcmotor can't be that much more expensive than a 120 volt ac motor...

Because they are made in such low volumes, and in a low-tech way.
It's not likely to change anytime soon.

Given the infrastructure we have, it is (much) cheaper to add inverters to Solar, than try to morph appliances and household wiring to match solar.

Plus a Grid-Tie Inverter allows you to SELL excess power.

OFM, this is a larger question than you may appreciate. I've given it the same considerable consideration and what the modern house requires is an AC and DC system. A majority of our functional purposes are DC devices. Only the high wattage devices are AC and it is questionable whether they remain or not. Air conditioner units and washing machines, probably and maybe, but the rest??

Here's some EE 101 edification. The voltage you get at the outlet is 120 Volts RMS. That's Root Mean Square because AC average is zero. RMS is the DC equivalent in energy. So in some respects we haven't strayed too far from Edison.

I challenge/ask anyone to take an inventory of their home on what runs on DC and what doesn't. Guaranteed the DC list will be longer than the AC.

At the present power levels of most household devices the current interrupting technology has been long overcome. If we are to look to real leadership in energy conversation I'm afraid it won't be technically exciting or sexy. Simply put, DC end use systems should be installed in all new homes and small business; and, there should be incentives for retrofitting.

Furthermore, these systems will readily integrate with home based generation and distributed generation systems. Perhaps the real resistance is the paradigm from centralized generation to distributed, but I have no opinion on the merits of either. We need both. As Gandhi said, the middle path is best.

AC is only there for the convenience of transmission. 9/10's of the devices in the home and small business will work equally well, or better, on direct current, DC. That's a monumental change at the local level, but as said before, not sexy.

U.S., Canada, N. America and Europe want to cut back 5-10%? Convert over to DC at the local level.

Now wasn't that easy...

Further EE edification...

One of the major drawbacks of DC electrical systems is interrupting current and especially fault currents (short current). Because a direct current does not return to zero, or a very low point like AC, DC is very hard on electrical contacts and equipment. Just try drawing off the jumper cables while starting a car and you'll get a very unpleasant demonstration of what interrupting DC is like.

(And, you can't sue me for trying this at home because I'm in Canada. The lawyers in our Yellow Pages is only 1/4 of the U.S.)

We now have power electronics and common household devices that interrupt DC all the time. The fact is we live in a lower energy dense world at the unit level, or per unit production level. Does this mean our total energy consumption is less, no. Only, we are getting more out of a kWh than before. That my friends is efficiency.

BC. We don't draw arcs switching DC anymore. HVDC Transmission is the future.
http://www.renewableenergyfocus.com/view/3567/hvdc-transmission-from-ene...

GAIL. Have a look at http://www.renewableenergyfocus.com/view/12299/eu-pvsec-solar-thinfilm-r... . Note the price and efficiency they are claiming.

Thanks BC,

My technical education extends far enough to be familiar with the points you have made;I understand ohm's law, and stepping up and down voltage , which is necessary for long distance transmission of electricity with ordinary (read existing ) copper or aluminum cable.

Trying to install a dual ac/dc electrical system in a new or existing house would be a major headache due to lack of a standardized design and building code;and I don't know if existing dc loads are large enough to justify the seperate circuits and the cost of an ac to dc inverter;of course such devices have one built in at the factory.

Now a handyman such as yours truly would simply run some extra circuits fitted with a receptacle plugs manufactured or modified in some fashion that it would not accept a cord for an ac appliance.

Hence such a circuit cpould exist side by side with my ac circuits, but isolated from them, and power my interior lights,run my spare computer, charge my power tools,and even a small air conditioner when it is most needed.

Dc switches are still a knotty problem for sure but if the load is a small one, a simple switch such as is used currently can be beefed up at small cost.I am using some ordinary heavy duty toggle switchesof the type used to control residential lights to switch some 12 volt dc at about two amps for homemade floodlights with no problems.

Here is a bit of backwoods hands on for those interested:

Sealed beam headlights,the kind used on older cars burned out on either low or high, can be had frre from the trash at a service station, but it is better to rob the plug with a few inches of the wiring and "bucket" which holds the light from a junked car.

Mount it in a small box or frame made from wood or cardboard in such a way that you can point it in any direction, and attach an old extension cord up to a hundred feet long to the light with electrical tape.

Plug up your automotive battery charger, 12 volt 4 amp minimum spec, and clip the other end of the old cord in the alligator clips.

The result is a fine and safe semiportable worklight for next to nothing;

I use one like this to work in wet or damp locations such as crawl spaces, as it is much brighter and yet much more diffuse than a flashlight, and with the charger outside in a nice dry spot, I am exposed only to 12 volts in the event something goes wrong.

Some local guys have a few lights mounted in a similar fashion in barns and spots where lights are occasionally needed, but out of reach of the farm grid.In this case , they put a set of alligator clips on the extension cord end that attaches to the battery, and take along a spare tractor, car or fishing boat(deep cycle)battery ;this rig needs a seperate switch-a couple of bucksat the hardware store- as connecting or disconnecting a battery from a load directly at the terminals can concievably result in an exploded battery with possible severe injury.

I can work for up to four hours by means of such a light powering it with an old car battery that wouldn't start the car-one with a big v8 engine-on a very cold day.

It is easy for a handyman to drive an automotive alternator with a small horizontal shaft gasoline engine such as the ones used on garden tillers and power as many as six to eight headlights silmantaneously, or to use such a rig up to charge batteries in the field on a piece of heavy equipment such as a bulldozer.

This is of course a solution ideally suited to OCCASIONAL use only.

Actually, the cost differential is primarily volume sales; DC household appliances are a very tiny market, which AC household appliances are a monster market. Non-recurring engineering, factory design and equipment costs, advertising, etc all are significant overheads for DC appliances, not so for AC.

Re: DC power lines - For a simple implementation, the biggest draw will be the refrigerator with perhaps a couple of lighting fixtures between the kitchen and living room. Dual wiring the whole house would be excessive. Of course, going DC-only is another option.

OFM, If you like tinkering around with stuff, and it certainly seems you do, then check out this site
http://www.ecoinnovation.co.nz/c-60-kits.aspx

They have mastered the art reconfiguring a special dc motor from washing machines, called a Smart Drive, into 12/24/48V motors and generators, and in their information section there is a free pdf manual on how to do it.
They have used these things for wind turbines, motor driven generators, hydro driven generators , motors for go karts/trolleys and on and on.
All I want for Xmas is their "inventors kit" and I think I'd be playing with it for weeks!

Interesting to hear BC-EE's comments on DC, and I would agree, most everything either converts the AC, or could run directly on DC.
Even the refrigerators, if you want to see an efficient refrigerator, check out these marine ones (made in Vancouver) using 12 or 24V compressors.
http://www.novakool.com/index.htm
At maximum power they draw 9 amps at 12VDC, 108 watts, for up to 9cu.ft fridge.
For a domestic sized one you would user a larger version of these compressors (by Danfoss), but your power/cu.f should go even lower for large units from the square-cube law between surface area and volume.

The SunFrost are the gold standard for home DC fridges - they use less than 1 kWh/day, on DC. The secret is the 3" thick styrofoam insulation, that's all! Good article about the house built by the owner of Sun Frost here.
http://www.sunfrost.com/extreme_efficiency.html

He runs a net-zero house with a 1.6kW array and worked out the capital cost for his grid connected system at $8.50/w (2005), to get 3.5kWh/day, yearly average, per watt.
This gives a capital cost per kWh/day of $2444, and I think this is the most meaningful way to compare sources, as you can easily add on a fuel cost per kWh if appropriate.
Based on that, his theory is that if you can get an more efficient appliance, that saves you 1kWh/day, for less than a $2444 increase, then it is worth doing - conservation is cheaper than expanding the supply - as is almost always the case.

There is some fantastic info and advice on these links, thanks Paul Nash. I am convinced that this is the future of housing for a large number of people.

" I challenge/ask anyone to take an inventory of their home on what runs on DC and what doesn't. Guaranteed the DC list will be longer than the AC. "

One 5 watt cell phone charger does not equal one 5,000 watt cloths dryer. Most of the power is consumed by AC devices.

" DC end use systems should be installed in all new homes and small business; and, there should be incentives for retrofitting. "

The DC loads in a home do not all operate on the same voltage level. More likely they all operate on different voltages. In fact many devices like computers, TV’s and printers often contain power supplies that produce multiple voltages, both AC and DC to run various components in the device.

Semiconductor switches combined with inductive/capacitive circuits can efficiently produce the required voltages without the use of transformers. This is true regardless of whether the source power is AC or DC.

" U.S., Canada, N. America and Europe want to cut back 5-10%? Convert over to DC at the local level.
Now wasn't that easy... "

Home electricity consumption represents only 1/3 of the electricity that supports our lives. Home consumption would have to be reduced 30% to create an overall savings of 10%. There is no evidence to support the claim that home DC would save any energy.

OFF Topic

I tried to respond to your post on the other thread several days ago, but it was closed.

I appear to have confused you with Charles Barton and my strong statements should have been directed towards him. You have no eMail link in your profile so I could not contact you that way.

ATM, I am working on a significant article and that is taking most of my time.

Alan

Bill, sorry to respond late, but I believe you missed the point. "The functional devices", that is, the number of devices and not the total load. How many devices do we have around the house that convert power from AC to DC; or how many could work equally well on either system.

Sure, there are numerous devices with different voltages. The N. American AC system was that way initially. The tone of your counter argument is the stuff that drives me nuts because the sub-text states we cannot make change or show leadership because of they way things are. That is driving by looking in the rear view mirror.

Engineer-types take great comfort in these types of arguments because it shows a command of the facts, but does not show a command of the future. We could standardize the DC voltages and loads in a day if there was the will. They don't need to be all over the place anymore than industrial DC devices. Industrial DC devices are either 24 V or 48 V (usually -48 VDC for telecom), and they all operate off the same distribution system.

No, we shouldn't rewire houses only for this purpose; but we did rewire houses to current code with knock-and-tube wiring when it became obvious it was inefficient and dangerous. Sooner or later we are going to wake up and ask why are we inefficiently converting AC to DC all over the place with a multitude of little transformers? Maybe it would hit home if we had to generate our own power by bicycle like Ed Begley Jr.

A toaster works equally well on either AC or DC BTW. Resistive loads don't know the difference and don't care.

Hi BC-EE, good to see you posting - I haven;t been on TOD much lately. Tried to email you (from profile address) but it bounced back. Time to talk to you about some renewable elec projects on coast and around Sun Peaks. Can you send me an email to paulrnash101 at gmail ? Thx,

Paul.

The new Apple Mac Mini (just 10 watts for CPU) boasts over 90% efficient AC > DC conversion.

Spending more on high efficiency conversion seems a better use of resources than rewiring houses & businesses.

Alan

A toaster works equally well on either AC or DC BTW. Resistive loads don't know the difference and don't care.

If you were trying to make a point, heating is a very poor example to choose.
Power arrives at the house as AC, and likely always will.
It is clearly foolish to convert that AC to DC, for heating!.

Even claims of DC reticulation, need careful scrutiny.

* That adds a LOT of copper to a house
* You STILL have AC to DC conversion

In fact, you now have non-optimised AC-DC, as now you have to GUESS what size you might need, for future DC loads.

Perhaps you guess 1KW - Now, that 1KW DC inverter, could be powering a single Cellphone charger - and wasting FAR more watts, that that cellphones own charger would have!.

Small chargers ARE becoming standardised, and they DO have stricter standby numbers, which is the correct way to manage this.

The tone of your counter argument is the stuff that drives me nuts because the sub-text states we cannot make change or show leadership because of they way things are. That is driving by looking in the rear view mirror…. Engineer-types take great comfort in these types of arguments because it shows a command of the facts, but does not show a command of the future.

Sooner or later we are going to wake up and ask why are we inefficiently converting AC to DC all over the place with a multitude of little transformers?

A lot of irony there BC. The old plug in power supplies that were always warm from copper and iron losses in the transformer, even when no power was being consumed by the downstream device, have been history for years. My solid state cell phone charger gets slightly warm when charging, but it is room temperature otherwise.

Someday the transformer on the pole behind your house may be replaced with an efficient solid state model.

More likely, a "metallic glass" transformer on the pole. The core was cooled so fast, metal crystals could not form. Significantly lower losses, especially at low VA levels.

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

up to 80% lower core loss than conventional transformers

http://www.metglas.com/products/page5_1_5.htm

Alan

BC. Would not disagree with your numbers but if you are building a commercial power plant you would be looking at "Spark Spread" numbers and comparing them to site specific factors.
http://en.wikipedia.org/wiki/Spark_spread

"Don't forget, the vast majority of residential PV systems in the US are net-metered, so their kwhr output wouldn't show up on any utility summary." Yes it does. The utility installed a digital meter on our main service panel when we installed solar PV. It reads out kwhd (kwh delivered from the grid), and kwhr (kwh received by the grid from the panels), and reports them on our energy bill. I agree with low maintenance. My panels are about 2 years old, have only a very fine coating of dust, and have never been washed - just cleaned by brushing snow off of them. And I live in high dusty desert. A fine coating of dust has little effect on the panels, and my reflective telescope mirrors.

While your meter is counting the amount imported/exported from the grid, it does not know how much electricity was directly consumed by your house without ever going on to the grid.

For example: Let's say that your PV system is generating 2 kW, your big screen plasma is on and drawing 1 kW and nothing else in the house is using power.

The meter will see 1 kW being exported to the grid, not the 2 kW your system is generating.

You would need two separate meters to measure this - one dedicated for your PV system and one dedicated for your house.

You would need two separate meters to measure this - one dedicated for your PV system and one dedicated for your house.

Which, incidentally, is what Sun Run systems have.

Pretty good analysis of installation costs.

But it seems to completely ignore operational costs.

For example, while PV has a very high installation cost - it has the benefit of very low maintenance and no fuel costs. It also has the benefit of producing electricity during daylight which is when demand for electricity (and thus the price) is highest.

Hi, just to tell I used your useful info here, in France :

http://www.infowargulfblog.ch/?p=717

I deeply thank you. War on COREXIT MUST BE WON.

My blog (90% in English) has this only purpose : to inform French SCIEnTISTS and Media. results are hard to come quick but some comes, slowly, as it seems that I am THE FRENCH who has an overview of this catastrophe.

Plus the media black_out exposed on my site...

This task is a part of a strategy : http://www.infowargulfblog.ch/?p=306

http://www.infowargulfblog.ch/

See you.

Some renewables like sun and wind-turbines have a low capacity factor. This is well established. Capacity factor can be increased at a cost, but is capacity factor increase necessary?

In the old days -where wind was the only non-human power source besides horses- this was not a problem because work was done when energy was available (even by night if necessary).

These days, due to demand side management -which was needed because of rigid slow response times of earlier coal and nuclear and their big turbines/boilers- nighttime electricity usage has increased a lot. This has lead to the concept of "base load" which in turn is now often used against adding more renewable power like sun and wind despite the inherent advantages:
- less CO2
- no air pollution
- less dependency on foreign countries
- less funding of terrorists due to oil money flowing to unstable/hostile regimes
- less mining accidents
- less pollution due to spilling of mining/drilling waste
- energy democracy instead of energy monopolies
etc.

Can we go back to a situation where energy usage is more governed by energy availability instead of the other way around? What are the pro's and con's?

Good comment.

Yes, we can, though we can also use dispatchable power and storage systems to fill in the 'gaps' when they occur.

For example, gas turbines and hydropower can respond quickly, which can help solve the supply side of the issue. Realtime pricing can help reduce the demand during those time periods.

Wei Zhang; Feliachi, A.; Residential load control through real-time pricing signals, Dept. of Comput. Sci. & Electr. Eng., West Virginia Univ., IEEE System Theory, 2003

The concept of shaping residential loads can be an effective way of controlling the load profile of a distribution company. Flat energy rates do not provide incentives to customers to use power as would be optimal from a utility point of view. With the restructuring of the power industry it is expected that prices will fluctuate in any given day. Actually real-time pricing (RTP) and demand side management have been discussed for quite some time since the 1970s. However, RTP is still a concept, and often misused with time of use. This paper describes how to design an RTP system for residential customers, by looking into some of the current problems in the electricity market, and the potential impact of demand response from residential customers.

In order to achieve the load control by modifying the electricity consumption patterns of residential customers, in addition to creating incentive by pricing schemes, it is critical to design the RTP system friendly to conquer the year long consumption habit due to the historical flat rate charge. This paper gave a design that retrofits the common houses facility. The system itself processes the real time price information with automatically managing the power usage for the
householders at the most preferable and economical way. Without dramatic lifestyle changing, residential customers can avoid high electricity price charge at peak time, and the power grid can benefit from load control. Eventually the RTF' system used by residential customers can help
to prevent and reduce price spike and energy shortage in the electricity market.

Yes ofcourse, dispatchable power remains needed. But storage and dispatchable power is expensive, that's why I wondered if it's an option to reverse some of the demand management towards energy usage when it's available instead of trying to iron the day/night usage changes out.

Many residential buildings here already have two counters for high/low cost electricity and most of them use the cheap hours of the day to turn on the dish-washer and washing machine. In France they use them to pre-heat the buildings at night (afaik). There's also talk here about smart meters which can turn a special group on/off when power is cheap.

Virtualization in datacenters can be used to move servers to parts of the world where energy is cheapest at that moment. Cold-stores can cool a bit more when preferred and let temperature rise a bit after that.

So it seams that at least some of the mechanisms are in place.

DSM works great when the customers can reasonably predict when it is best to do and not do stuff, like the day/night rate differentials, or even split the day into three of four time periods. But when you try to bring it in line with the real time spot pricing, you run into the problem that sooner or later, you are going to have to run the fridge, water heater or whatever - you can;t go without if for days.
While it certainly is possible now to manage loads according to spot prices (I'm sure there'll be an iphone app for that), unless you program it once, it will be just too much effort, for too little rewards for most residential customers.

Flattening the load curve as much possible is absolutley a great thing (I spent two years doing just this at a ski resort). Itmakes the patterns more predictable, generators are able to run at higher efficiencies, and, often overlooked, it relieves peak loads on the transmission system, which results in lower line losses, and avoids costly expansions.

When we look at "predictably unpredictable" sources like wind, we then need to have loads that can be turned off for days at a time,if need be, and there are not many homes or businesses that have those. IF we look at the old style example of milling flour, you can predict, fairly accurately the average annual output of a windmill, possibly even the monthly, but the weekly and daily production are impossible to predict more than a week or day ahead. So the miller has to have a higher peak capacity (-=larger milling machines) and then be able to store flour to meet suppliers needs (=cost of storage facilities + standing inventory). He may be able to convince his customers to store the product, which merely shifts the cost to another party.
In any case, it is clear that there is a cost to taking advantage of this intermittent energy, which may be worth it for some customers, if the intermittent energy is cheap enough. Unfortunately, wind is usually subsidised today, so it's chances of being viable (i.e. unsubsidised) when selling surplus power (=lowest cost) in the spot market are slim to none.

Larger commercial and industrial customers have more loads they can move around, but only for so long. here in British Columbia, there is a time of day rate available to industrial customers (i used to be the energy manager for one of them) and only two out of 500 qualifying customers have chosen it. Admittedly, the rate differential here is not the great, but you can see that the other 498 have decided it is not worth the trouble to game the system.
Most people will be happy to schedule things to different times of the day or week to game the system, but few will be willing to manage according to daily or hourly fluctuations in price. When your average electricity bill is $3/day, it's not worth spending half an hour each day to get it to $2.

But it certainly is worth making one time or programmed changes to get it to $2.50, and in doing so, you will probably find ways to get to to $2, or even less.

When we look at "predictably unpredictable" sources like wind, we then need to have loads that can be turned off for days at a time,if need be, and there are not many homes or businesses that have those.

We need to make the distinction between load shedding (turning somethin off) and load shaping (adjusting parameters of the load). For example, A customer could set A/C thresholds to have 75 as the default, 76F when the realtime price goes up over $0.xx/kWh, 77F when it goes over $0.yy/kWhr, and 78F when it goes over 0.zz/kWh. Likewise with electric hot water.

Electric car charging would be more focused on hourly changes during the night, similar to laundry.

Yes, most DSM will be automated. EV/EREV charge settings will be set in the dealer's showroom, and the settings may be changed a few times by aficionados, never by others.

Extended low periods over very wide areas will be rare, and so they can be solved by cheap backup power. For instance, there's a very large supply of backup power in hospitals, jails, etc which needed to be tested and exercised 2x per year - the first and 2nd useage of DSM per year from these sources is therefore free.

In the medium term we'll have plentry of excess peak capacity. In the very long-term, extended low periods would be easily handled by biomass, and moderate overbuilding and conversion of excess power (even at low efficiencies) to synthetic hydrocarbons.

EV/EREV charge settings will be set in the dealer's showroom,

I think this is a misreading of human nature, likely driven by wishful thinking.

Not on the Leaf or the Volt BTW.

I think mass use of EVs will result in much higher 6 PM peaks.

Alan

I know you feel this way. We've had many discussions.

Not on the Leaf or the Volt BTW.

What are you thinking of? Both these vehicles will allow automated charging.

Keep in mind that the Volt eliminates "range anxiety", so that people won't be frantically recharging whenever they get the chance.

And, 6 PM peaks are an artifact of flat pricing. People respond to prices. If people are charging too much at 6 PM, then prices at that time are too low. Set the prices high enough, and the peak will go away.

Anybody with a cell phone knows that people respond to time-of-day pricing.

I pay absolutely no attention to time of day pricing for cell phones. The convenience of calling when I want to/need to is simply too great.

If there was a 3% chance that I would go out again that night, I would likely recharge a Leaf as soon as I got home.

Alan

I pay absolutely no attention to time of day pricing for cell phones.

I suspect that's because you use it very little.

If there was a 3% chance that I would go out again that night, I would likely recharge a Leaf as soon as I got home.

Not if the price of power was high enough. OTOH, if you were one of those people with much more income than others, or a particular need for mobility, then probably prices wouldn't rise to that level, because other people would do the shifting.

Alan, you've suggested that increased cost of commuting would force people out of the suburbs, so you clearly believe that people respond to prices, right?

Finally, I agree that many people will be anxious about range - they're the logical candidate for a Volt, instead.

Yes, excellent comment.

People tend to assume that current patterns of demand were handed down by god. In fact, US industrial/commercial consumption has been shifted to night time by very primitive time of day pricing ("demand charges"), while residential pricing is absolutely flat.

We could move much more consumption to the night, to take advantage of wind and nuclear, or we could move existing night consumption back to the daytime, to take advantage of solar.

People respond to prices.

The indicated solar PV cost is extremely high compared to the other costs. I can delete the solar PV indications, if someone can explain how they are incorrect. But they are somewhat concerning.

I think the reason it's not worrisome is that the installation cost of capacity is a relatively meaningless number. It's the levelized cost of electricity that matters. For homeowners, the levelized cost of PV is already below retail electricity rates for peak hours here in California. And that's before rebates which bring it down to the price of non-peak electricity.

I'm 90% sure that the FERC cost numbers are merely for installation and do not include operating costs, although it would be nice to have a proper footnote to follow up on the source. Since PV has near zero operating costs, it doesn't look good in this comparison. But this comparison is misleading.

Again, compare levelized costs of electricity, or you are comparing apples to appleseeds.

http://www.nrel.gov/analysis/tech_lcoe.html

Have some fun with the Simple Levelized Cost of Energy Calculator.