Thanks for posting this, Kate. I'd like to ask people to use the main Sagacious Iconoclast link to access the model, as I can't change the keepandshare file, so when I issue any new releases to account for corrections or suggestions made below, the keepandshare link will change, which I will then note in the Sagacious Iconoclast page.

As mentioned, I'm interested in corrections, extensions, or enhancements to the model, which since Kate has provided this space, we can discuss here rather than via the email link provided in the model. Once again, that's:

sagaciousiconoclast.blogspot.com/2008/03/solar-power-v-oil.html

(Oops, just noticed, the main link you have up, Kate, is missing
the "l" after ".htm". The link I just posted above is correct.)

Jeez, I can't even cut and paste right today...

Jeez, I can't even cut and paste right today...

then I'll just pass on any "dinner" offers:-)))))

But it is the way of the future! just think,when we are all back living in the mud huts, then we'll have all the solar power we need! If it can get through snow/ice to us. Almost -10 trillion bucks. that should buy the Goracle/Fruitfly some more homes/yachts,etc.. Run, of course, on the black market gas they will have.

despite the goracle and dr. mengele suzuki economics always, always, always prevail. it may take a generation or two (aka China) but you cannot fight value or cost with rhetoric, theory or bull$hit.

Yes, Justthinkin, but note that if oil doubles in price (Cb), and solar panels become five times more efficient (Pp), then solar saves trillions of dollars. That may well happen over the next 20 years.

Whoa. And just when I'm struggling with Excel spreadsheets.

Oops, I should not have said more efficient,
I should have said more cost-effective: Cp ÷ Pp.

Vitruvius:

I recently was examining solar just for something to do one evening. I happened upon the following. There is a company called .... Solar Systems ... (how original) located in Australia. Apparently they are going to build a 154 MW power station for $420 million. I put those numbers into your model and came up with a factor of only 4.22 times as much. Guess what? It's done with mirrors. :)

http://www.solarsystems.com.au/the_technology.html

Consider also that the company thinks the solar part may be able to be replaced for about 20% of the cost of the unit so the replacement cost will not be equal to the original cost.

I see you built your model for cars so this may not work in your model but it may be a direction for long-term large-scale solar.

The trouble with the model is that the cost of manufacture of the solar panel is ignored. In fact the solar panel costs a lot energy to produce from the mining, smelting and refining of its materials, to the manufacturing and marketing of the solar panel.

It is also assumed that the solar panel will last forever which it will not. Materials of which the solar panel are made (such as gallium) will corrode over time.

Solar panels are also assumed to work everywhere with the same efficiencies. This is very untrue. What works in the Sahara close to the Equator works poorly or sporadically at higher latitudes (sun at low angles, filtered by greater thickness of atmosphere with lower number of sunny days).

Dalton McGuinty's billion dollar investment in a solar farm near Windsor, Ontario will produce electricity at $0.42 per kilowatt hour versus $0.055 cents for nuclear or hydro which is a very poor investment despite its green appeal.

actually with the small area of solar panels given with my spreadsheet example 60km square ,it might be a hint that solar cycles have far more to do with the temp. or the climate than any human contribution. a hint , just maybe.

it an area small enough to drive a bandwagon around twice in a few hours.

And in sunny England, land of spring:

http://news.bbc.co.uk/2/hi/in_pictures/7310286.stm

Yikes....too scary for me... ;)

No, Earl, the cost of manufacture is not ignored, it is a component of the selling price, Cp. It is not assumed they last forever, their life expectancy is kx. The average effective duty cycle is represented by kd, feel free to lower it if you think it's too high.

Also, Brent, I shouldn't have said vehicles, I should have said oil-based stuff that needs to be replaced because of a switch to electricity. Moreover, the Cv and Xo parameters are in fact general purpose fudge factors ~ if you think I missed some component, you can plug its annuallized cost into one of those to see the effect.

(I have to pop out for Easter Sunday dinner ~ back in a couple hours.)

Does it account for all the subsides put into oil production? An interesting model....

sean S,

yes it does.

calculate out the oil subsidies and post them.
remember depletion is a tax effect not a subsidy. royalty is a tax not a subsidy . income tax is a tax not a subsidy . road tax is a tax not a subsidy, PST is a tax, GST is a tax. shipping is a cost , pipelining is a cost . so keep your ledger tight , we are watching.

according to Car and Driver (i know a great source) the cost of solar panels is rising due to increased demand and a limited supply.

the installed/produced $/kwhr will also rise.

residential panels produce about 100W/panel at about $800 each at CanuckistanTire

McGuinty didn't really blow a $Billion on solar -- did he !!?? With WK's advice ?

Interesting exercise Vitruvius. Not to add much, but a question/complication. Have you considered the nature of electrical vehicles vs internal combustion vehicles i.e. the power plants are not equivalent wrt weight, therefore mass of the vehicle, mass of the brakes, frame etc? I've read articles suggesting that electrical vehicles may be able to realize some economies through independent motors in each wheel for instance -- offset of course by the weight of batteries for storage. Might simple electrical generation for a power grid make for a simpler calculus?

What if the US dollar recovers some ground and oil settles in at 85$ per BOE? Or is that not possible? Also, as solar becomes more desirable and starts replacing oil, won't the demand for oil lower and thus decrease the the price per barrel making solar even more expensive or is it recognized that the cost of solar will decrease as it becomes even more popular, like vcr's?

Regardless, your intellect and approach to life's mysteries is refreshing and much appreciated by this SDA reader. I look forward to your posts, comments and insights.

Also, as solar becomes more desirable and starts replacing oil, won't the demand for oil lower and thus decrease the the price per barrel making solar even more expensive or is it recognized that the cost of solar will decrease as it becomes even more popular, like vcr's?

I think this is the Xo variable.

The bottom line is this: the cost of solar panels will only decrease and the cost of a barrel of oil will only increase. There is no logic that says that oil is the better long term investment.

Well, I got away from this for a while and am back at it. The numbers are actually more interesting than I first thought for the large scale Australian system. Here are the numbers I used.

I took the AU$420 million to build a 154 MW station. I took today's close from futuresource.com on the $AU to be 0.89 $US and arrive at $373.8 million for the 154 MW station. This works out to be about $2.43 per Watt. Solar Systems says that each dish generates about 35 kW so I use that for the Pp variable.

35000 * ~$2.43 = $84954. Now, Solar Systems says their PV part is about 20% of the cost of the dish so I enter (84954 * 0.2) $16991 for Cp and 500% for k(i). Ap get 0.23 m**2 (from the Solar Systems website) and I take your default 6 hours per day for k(d). k(e) is simply set to 100% [because I have no knowledge to choose any other value :) ] and I leave k(x) at 20 years.

I set C(v) to be 0. I also set Xo to be 0; however, over time this number must be > 0. I already mentioned k(i) at 500%. The last variable is k(m). I am having trouble understanding your default of 100%. I know that the value should not be 0 if I take it the way you have labelled it; however, I am setting it to be 0 since I reason that the value should actually be a factor of the increase over the oil technologies and not a factor of the absolute cost of the solar technologies. I.E. both technologies will have maintenance costs - or do I miss something here?

Setting k(m) to 0 yields a cost advantage of oil of only 11%. Even setting km to 50% (for arguments sake) yields solar being only 66% more expensive. Maybe this is why the Australians chose it? Hmmm.

There is, of course, the initial AU$420 investment. However, if one needs to build a plant anyway (coal, PNG, nuclear, hydro, etc) solar is not quite competitive, but it looks like it might be getting there.

A few comments:

1) You cannot simply "replace" a barrel of oil with the same amount of electricity comparing BTU's. Most gasoline engines are 30% efficient, diesel around 35-40%, batteries are much more efficient then that. Ke would have to change drastically to affect that.

2) The numbers are too large, and deal with too large of a cross section. It is fun to play with these numbers, but not really meaningful. It would be better to be specific to what the energy would be used for(for example, automobiles, or house heating)

3) Speaking of home heating, what if you went from heating oil to a geothermal setup, powered by solar? This would be probably the most drastic example. BTU "equivalent" of the second example(KW-hr converted to BTU's) would be about 1/5 that of heating oil used.

It would also be interesting to "throw in" nuclear and hydro power. We all know solar(and wind) would be blown out of the water.

Just one more thought to add to the sunny outlook for solar. I suppose one could also run numbers for wind systems and geothermal using this setup. I might do that later. However, my wife wants me to enjoy a film with her tonight :)

Thanks, Vitruvius. Look for your response later. Hope you enjoyed your meal with your friends/family.

In Carpentry, we do not use only one of, screws, nails or glue to join everything.

In energy use we do not use only solar cells to power *everything*.

Solar cells are baing manufactured in continuous roll sheets now and will become cheap as dirt. = TG

(Ok, three hours. Mom and dad are 82, and there's blessed food involved, so mom's desires trump all hands ~ but I did leave a bit early after explaining the modeling we're working on here ~ they said to say hi ;-)

For the relative efficiency of electric vehicles to petroleum vehicles, DrD and Allan, the model has the ke parameter. I set it to 100%, with a note. But yes, that's why I included that factor. If you set it to 200% (electric vehicles getting twice as many meters per Watt-equivalent of input power), then solar does, as we would expect, much better. But you certainly can convert volumes of oil to BTUs to kWh, independent of the efficiency factors (which is why they are included later).

On the demand curve, Brent, I think we have to assume that the hydrocarbons we harvest for chemical feed stocks, currently at 40% of the global total, are going to rise as the rest of the world modernizes, so the total demand for hydrocarbons will probably not be abated. Nevertheless, I will admit that betting on the long-term prices of commodities is not something I care to undertake.

The ki and km factors, Brent, (one of which is arguably redundant) are provided to convert from the cost of solar panels to the cost of energy delivered to the end user. That is to say, a dollar's work of oil includes everything, but a dollar's worth of solar panels doesn't include infrastructure, operations, maintenance, and profit.

No, wait a minute, that's wrong. There is an omission in the model! It's a dollar's worth of gasoline at the pumps that includes everything. However, a barrel of gasoline, which is 158 liters, costs 158 bucks(ish), so there should be an explicit factor for the infrastructure, operations, and maintenance on the hydrocarbon side of the ledger. For now, we can just set Cb to 158, giving us only 490% for Sg ÷ Og.

I like what you did with the model, Brent, that's what I intended it for. When Kate originally asked me for my opinion of Butler's essay, I noticed its lack of amortization of long-term expenditures, and so started to do my own calculations.

The model I have made so far is the result. When I started, I had no preconceived notion as to what the final numbers would be. Yet it does look like the near-term difference is less than an order of magnitude, and longer term solar may become less expensive for those cases in which it is appropriate.

There are lots of problems left to be worked out, like storage of off-peak generation for peak usage, intelligent metering to allocate optional usage to times when the portion of the grid getting good photons is high, et cetera. And, of course, we still have to figure out where to get the 42 billion square meters of land to put the panels on (though, on a global scale, 206 × 206 km isn't that much).

All in all though, to the degree it is reasonably pragmatically possible, I am in favour of reducing the use of hydrocarbons for transportation, heating, cooling and lighting, because they are so valuable as feed stocks for making toxic chemicals. That's what I tell all the environmentalists.

Oh yeah, and wait a second, a dollar's worth of gas at the pumps includes taxes. We need a tax factor for solar; logically that's part of km (so maybe, Brent, it's a good idea I set km so high ;-)

Right, now watch this. If we start with the sample model, increase oil's Cb to $200 to account for the omission described above and a little growth, increase the price- performance of solar to $1/W by changing Pp to 1000 W/panel, and change ke to say that an electric vehicle gets twice as many meters per watt of input, then solar only costs 38.6% of oil, a saving of nearly 2 trillion dollars per annum.

I'm starting to think that the numbers are saying that the medium term economic prospects for solar are entirely feasible. So, naturally, whenever I start to think something, I ask myself: what am I missing?

Vitruvius: "note that if oil doubles in price (Cb), and solar panels become five times more efficient (Pp), then solar saves trillions of dollars. That may well happen over the next 20 years."

Actually that's rather unlikely with respect to solar cells. Remember, you can do nothing about the input energy source (the sun), nor can you affect the MeV of the incoming sunlight. That being the case, solar panels are essentially a materials science problem. These have shown themselves to be extraordinarily intractable over time. Yes, solar panels have been developed that are more efficient than silicon wafers. They are nowhere close as economically effective because of their material content and manufacture methods required.

Second, as your later post pointed out, the space requirements are prohibitive.

Third, what none of the solar industry sources have acknowledged is how extremely sensitive these devices are to scratching. Even light scratching, barely detectible to the human eye can result in the loss of as much as 3/4 of the energy from reflection/refraction effects.

One little thing you might consider Vitruvius is the energy payback period. For solar cells, it was negative 20 years ago, meaning you get less energy over a lifetime than it cost to make it in the first place.

No Real Conservative, the cost of solar cells will not necessarily come down over the next few decades. The cost of physical objects, does not tend to decline over time very much in real terms, particularly when it is energy intensive to produce in the first place, as solar cells are. Price declines are mostly the result of using cheaper materials. In the case of solar cells, the trend for more efficiency is the reverse, more expensive, not less expensive, materials. Battery technology has gone the same way over the past 20 years as well, which is why we still don't have an electric car remotely comparable in performance and endurance to an IC engine.

In short, solar may well turn out to be like fusion; it's always 20 years away.

Good points, CGH. Fortunately for me, the model can handle those parameters. I may have the kd full-power equivalent duty cycle too high, considering the scratch factor (though they'll probably find a way to mitigate that problem, remember we're talking decades -- see below). I don't think the comparison to fusion is fair. The unknowns there are far more suspect. It would perhaps be more fair to compare to the advances that are being made in fission, such as the pebble-bed stuff that is just starting to come on-line now -- see below.

The incoming MeV is certainly a hard limit, CGH, but I think the current average is somewhere around 15%, so a maximum increase of a factor of five is possible. Moreover, I think there are some non-bogus plans in place that don't, as far as I can tell, rely on too much magic to get from $5/W to $1/W, which would be critical. And the energy payback in the cost of manufacturing is simply a part of the price, unless you're selling at a loss, though I suppose you can then always rely on volume. And the nano-technology capacitor work doesn't look too bad ~ when you think about it it's a natural ~ so that may help alleviate the store-and-forward problem.

Clearly I am a technophile. I think that all the great advances in the human condition have come from technology. From ancient agricultural technology, through manufacturing technology, to the fantastic results we have seen over the last hundred years. The reason we have the Magna Carta, Women's Liberation, and treasures beyond the dreams of even not-long-ago kings is because technology has made us so wealthy. Which is also partly why people spend so much time b1tching: they can afford to.

Let's not forget three things though. Even if solar does become a substantive part of our energy infrastructure, (1) as TG alluded to, there are things you can't do with electricity that you can do with hydrocarbons, (2) there will be a huge build-out period, from the time it takes to advance the technology, to the time it takes to assemble the infrastructure, we're talking decades, and (3) there will be the usual interminable parade of fraudsters, snake-oil salesmen, and all-round general-purpose shysters trying to screw us on the evolution of technology for unearned personal gain.

Thanks for the feedback, everyone. I'll make the oil markup correction described above and generate a new release, but what you've got me thinking about is a multi-dimensional analysis of this sort. As Allan alluded to, we're talking about hydrocarbons replacement, but the parameters involved are dependent on principal usage categories, namely, transportation, heating & cooling, and already electric things, like lighting and appliances (what you're reading this on).

Also, as Brent mentioned, it would be interesting to parameterize parallel models like this for hydrocarbons (oil, gas, tar, shale, coal), nuclear, solar, wind, et cetera, and then assign contribution coefficients in the cross product with usage categories, to come up with overall utility measures.

I doubt I'll do that, my customers would prefer I spend my time elsewhere, but if I do, I'll let Kate know. And let's not forget, though utility is the closest measure of utopia, utopia is not possible, so as John Stuart Mill drew the line in the sand, pace his father's colleague Bentham: utility is not everything.

(At which point H. L. Mencken would probably jump in and say some-
thing like: of course utility is not everything, just look at government.)

By the way, the Xo parameter is commonly called the Hennesy variable ;-)

20 years away?

Solar is used in vast numbers of remote industrial applications, Marine, boat,camper and home roof tops today. Most Marine application solar panels can be walked on. Tough stuff.

42 Billion acres for panels? Silly idea. Solar panels free us from grid liability and line losses. They feed energy storage in both sunny and cloudy weather.

Check the Canadian Tire catalogue. = TG

Understood, TG, yet it remains the case that you can't generate and distribute tens of peta-watt-hours per year of energy using either the catalogue or the warehouses of Canadian Tire. And I'm not saying the real-estate problem is a show-stopper. There's lots of places you can put the things. I appreciate your enthusiasm, TG, yet pace Allan's comment, the numbers aren't too large, they're something very close to the relevant scale at hand, of which global dollars per annum is a decent first-cut approximation.

That's the single biggest problem with the cheerleaders, on both sides, who don't understand the magnitude of the global scale of the problem. Canadian Tire doesn't enter into it. You're talking AB&B, Fluor, Bechtel, and/or whatever adds to or supplants them.

That's one of the reasons we had to take Saddam out. It was partly about oil, and a half-dozen other things. He was about to privateer the gulf oil flow and hold the world ransom, up with which is something we cannot put. That's something lymbic people don't always understand: the social costs of disrupting the global hydrocarbon system are far higher than a simple major war somewhere.

Meanwhile, I remain unconvinced that pursuing the potential for deriving some non-trivial portion of our eventual future energy supplies from evolving solar power technology should not be pursued.

Jeepers, I went to post today's Late Nite Radio show, and I spent two hours listening to Vladimir Horowitz. I see it's been quiet on the Sunshine Front though ;-)

Since this is a discussion in which quantative values will tend to be appreciated, I will note that in the eight hours since Kate published my model, the Sagacious Iconoclast site has had about 800 visits and 1,200 page views, and the model has been downloaded from the keepandshare site over 700 times. That's one every ten seconds.

Thanks again, Kate, and good night everyone, I wish you all the best. And remember the words of W. S. Krabill, who said, "Those who are mentally and emotionally healthy are those who have learned when to say yes, when to say no, and when to say whoopee".

Nice work on the model Vitruvius.

Just to be annoying, I thought of another variable. Dual-use materials.

Its been my considered opinion over the years that dedicated solar power "stations" are an idiotic idea except possibly in the equatorial deserts. Just the cleaning bill alone for dusting all those acres of flat surfaces is enough to give one pause.

However. Its a nice sunny day today, what if my roofing material was busy running my hydro meter backwards? What if my driveway was storing up kilowatts to be used to cook dinner?

Solar collectors don't have to be hyper efficient to work, they just have to be CHEAP. One way to make things cheap is to do two things at once with them. If a roofing material can be made that will keep the water out AND collect electricity, plus be installed by your standard roofer-type guy, plus cost roughly the same as a standard roof, that's a lot of potential acreage for solar collectors. Driveways are another available surface area.

If there were some type of film, sprayable emulsion, rigid panel or whatever that was both building material and collector, any power you get is "free" because you have to have a roof anyway. More capitalistically, you can reduce the cost of your roofing by selling the electricity it makes to the grid. At a certain cost/generation efficiency, the roof starts making you money.

What does that do to the model?

An insteresting variable, not yet modeled, I suspect, is the need of the 'grid' to enlist the aid of its supplicants~

http://masticshirleychamber.com/photov/html/jerry.htm
http://masticshirleychamber.com/photov/html/jerry2.htm
http://masticshirleychamber.com/photov/html/Paul.htm

http://www.lipower.org/pdfs/company/pubs/brochures/solpioneer.pdf....current rebate sans (federal&state) tax rebate considerations is $3.50/watt,,, and though the panels' prices have remained high, the ancillary support items have come down in price. Do it yourself'ers would not require very much of an investment in exchange for 20..30 years of exceedingly cheap energy-if not income ~

Nice work on the model Vitruvius.

Like any prototype, it needs some sailing on a *breakdown* cruise. With refinement and tweaking it could become VERY useful.

LDK Solar in a falling market has bucked the trend. 150 Million$ added to a previous Multi-million$ deal with Mosel.

http://finance.aol.com/quotes/ldk-solar-co-ltd/ldk/nys

Having previous highs over $70, there is a lot of up room from today*s $24.07.

LDK is only one of dozens of solar cell makers and most of them are in the dumps with the DOWN market. New-Techs are always first to tank in a slide.

This is NOT a stock tip! What looks ok one day can die on a rumour. This is only and example of many people throwing Million$ into the future of Solar.

Then there is Canadian Solar - [ CSIQ ]

And. . GT Solar.com who supplies manufacturers, [ ground floor ].

Too risky for me though. They signed a 49 Million$ with the Russian company, Nitol Group. GT Solar’s reactors and silicon tetrachloride converters will be installed at Nitol’s production facility in Irkutsk, Siberia.

I don*t trust that Putin may decide to take it over for some pretense.

gtsolar.com/promos/showpromo.php?id=111

= TG

Thanks for posting this Vitruvius; what it tells me is that we're stuck with hydrocarbon fuels for the next few decades. Despite the disadvantages of solar cells, I'd love to put some up on my roof as I live in the interior of BC with abundant sunlight in the summer. Cost is a BIG issue and there was a good series of article by Steve Ciarcia in Circuit Cellar Ink detailing his $65K solar system. $65K just gets you an 11 Kw system and for $10K I can get a 10 Kw natural gas backup generator for my house.

What hasn't been considered is alternative means of using solar power. It is possible to get H2 from photosynthetic reactions although the H2 yields haven't been that great thus far. I'm confident that tinkering around with photosynthetic bacteria might produce industrial scale H2 production which solves the problem of storing the energy once one captures it. There are other photovoltaic technologies of which only silicon has been commercially developed. What would be nice would be some form of photovoltaic house siding which would have the added bonus of producing power. The Phantom alluded to utilizing his driveway as a solar collector, but think a lot bigger like making highways into solar power collectors. I think it's going to be a while before we see solar cells that one can drive a semitrailer over that still function, but there must be a way to make use of the temperature differential between blacktop and surrounding countryside on hot summer days. There are thousands of miles of solar collecting blacktop in N. America doing litle except burning the feet of anyone foolish enough to step on the road with bare feet on a sunny July day.