That cubic mile
Posted by Engineer-Poet on February 28, 2007 - 11:50am
Topic: Alternative energy
Tags: efficiency, heat, nuclear, photovoltaics, solar power [list all tags]
A lot's been said lately about how much energy is in a cubic mile of oil. This is roughly the amount the world uses in a year.
Assumptions: The Three Gorges Dam is rated at its full design capacity of 18 gigawatts. A nuclear power plant is postulated to be the equivalent of a 1.1-GW unit at the Diablo Canyon plant in California. A coal plant is one rated at 500 megawatts. A wind turbine is one with a 100‑meter blade span, and rated at 1.65 MW. A solar panel is a 2.1‑kilowatt system made for home roofs. In comparing categories, bear in mind that the average amount of time that power is produced varies among them, so that total energy obtained is not a simple function of power rating.
src: Joules, BTUs, Quads—Let's Call the Whole Thing Off, IEEE Spectrum, January 2007
Illustration: bryan christie design. Click to enlarge.
Leaving aside some errors (the coal and nuclear numbers are off by about 10% to each other, and the capacity factor of wind turbines should be closer to 30%) the most essential oversight in that equation is elephantine:
Compared to that, the rest is small potatoes.
- 4 Three Gorges dams, cranking for 50 years.
- 32850 1.65 megawatt wind turbines, cranking for 50 years (100% capacity factor).
- 91,250,000 2.1 kW solar PV installations, for 50 years.
- 104 500 megawatt coal-fired electric plants, for 50 years.
- 52 1.1 gigawatt nuclear electric plants, for 50 years.
- A barrel of oil has 6.1 gigajoules (GJ) of chemical energy.
- A cubic mile is 26.2 billion barrels (42 gallons/bbl). (The USA uses about 20.5 million bbl/day, or 7.5 billion barrels/year; this comes to less than a third of a cubic mile annually. World annual consumption is closer to 1.3 cubic miles.)
By this, a cubic mile of oil is even more impressive: 1.60*1020 joules. That's 5070 gigawatt-years of energy, nearly twice IEEE's estimate. But that's what we put in. What do we get out of it, and what would it take to replace it?
Ins and outs
Oil gets turned into a bunch of different things, and those uses vary widely in efficiency. If used for heat, oil can be very efficient. Bunker fuel burned in low-speed marine diesels can yield 50% efficiency. But our most important uses of oil are also the least efficient.
Take the average car or light truck. They don't run on crude oil; they require a highly refined fraction known as gasoline. Demands of octane rating, vapor pressure and sulfur and aromatic content increase the losses in the refining process. One source claims 82.9% efficiency from an oil well to a refinery's gasoline output. That cubic mile just got smaller.
But the losses are just starting! The average vehicle is very inefficient, turning just 14.9% of the energy that goes into the tank to work done against air and rolling resistance. The rest is lost in the engine, transmission and brakes. From well to wheels, the total is a pathetic 12.4%. That cubic mile just shrunk by half... in all three dimensions! Diesel is more efficient both at the refining end (87.9%) and the consumption end (35-40% engine efficiency in heavy trucks) but overall throughput is still around 1/3.
If the world followed US patterns (it doesn't, but it's not that far off), refineries would average perhaps 90% efficient. Gasoline would be about 43% of the total energy of the product supplied, distillate (diesel and heating oil) 22%, jet fuel 8.3%. All the rest comes to 27%. If we drop jet fuel as non-essential and add the rest up by efficiency (14.9% gasoline, 40% diesel), the total useful energy comes to 42.2% of the input. Best of all, only 15.2% of that is mechanical work; the rest is heat.
Adding up the end products
If we were going to supply a cubic-mile-of-oil equivalent of heat and work from nuclear plants at 33% thermal efficiency (3.3 GW thermal input, 2.2 GW thermal + 1.1 GW electric output) it would take a lot less. If you cranked them for 50 years, a mere 14 1.1 GW plants could supply 771 GW-years of electricity and another 1540 GW-years of low-grade heat, more than satisfying the requirement of 1370 GW-years of heat from oil. Coal would do about about the same, but it would take 31 500 MW plants to equal the 14 nukes. Wind has no waste heat stream and couldn't do as well (the energy would have to be all electric), but the possibilities for solar are amazing. Solar heat (for space heat) can be collected for very little, sometimes for free with careful design. Supplying 770 GW-years of electricity from solar PV at 25% capacity factor would require only about 40 million 2.1 kW installations; doing a year's worth per year would require about 2 billion 2.1 kW systems, or about 700 watts per capita.
700 watts is about 10 of today's PV panels. The industrial nations could almost afford to give 10 panels to every child at birth, and cost improvements in the pipeline could extend this to much of the world in the next decade or two.
Imagine clean, cheap energy as a birthright. Something to ponder.
Edit: Note that this analysis only considers energy obtained from oil. Weighing this against the total consumption of (especially first-world) humanity leads to inaccurate conclusions; coal, gas, nuclear and hydro sum to considerably more total energy than oil. Ponder and discuss accordingly.
[ED by PG] You may also want to check out Khebab's story called "Getting a Grasp on Oil Production Volumes" which also discusses this topic.
Hit reddit, hit digg, hit your favorite link farm! :) Send it to slashdot, metafilter, del.icio.us, stumbleupon, etc.
Let's get EP some eyes for this really good post.
Great post EP. I gotta call you on the Nuclear numbers, though. How do you deliver the low-grade heat to homes that currently use heating oil, etc? There is also the transmission loss, and then the loss in the presumed charging of the electric vehicles that would be required in an electric-only world. A quick guess would be that you'd need about double your 14 1.1GW plants/year. Still under the 52/year from IEEE.
Do what was done in London with Battersea Power Station - simply have big pipes conducting the hot water to the houses across the river. It was then like a giant central heating system.
(From the book London Under London):
Should we put our Nukes in the city centers? What is the chance of getting that done?
I'm all for harvesting the wasted heat. For instance, check out the Ecopower combined heat/power unit (goggle it, it is pretty good). That takes care of the 60%+ energy loss due to heat/transmission. Unfortunately, it runs on fossil fuels, though a hydrogen version should be possible. Still, it is no Mr Fusion, but then I'm not holding my breath for that one.
I initially read that is should we Nuke our city centers.
I thought it was a pretty good idea.
In fact, it's called combined heat and power (CHP) with district heating.
This is something widely used in Europe (especially Denmark and Germany). 30% of Berlin's (ca. 3.5 million people) space heating requirements are covered by CHP.
So, it's absolutely no rocket science and has been refined and optimized for decades already. Just not in the UK or the US.
CHP causes slightly lower electric conversion efficiency (like, 5% less), but makes use of this 5% loss to deliver another like 20%-30% of the otherwise wasted primary energy to the homes.
CHP with nuclear plants is not exactly a good idea. Consider a radioactive leak in the plant, contaminating the hot water loop. It would have to shut down immediately to avoid pumping radioactivity into the homes, leaving entire cities cold from one minute to the next.
Micro-CHP units in individual homes are half-nonsense, because
- the capital cost for a home owner to buy such a thing is much higher than for large-scale utility companies,
- maintenance needs to be organised in much more decentralized and therefore more inefficient way,
- a change of fuel is almost impossible and
- propagation of such units is much slower than with a central solution which would serve whole cities or at least city quarters in one go.
Cheers,
Davidyson
Have to disagree here in USA circumstances. With most homes heated by natural gas (at least presently), the CHP unit does not necessarily mean more maintenance than present if current furnaces/boilers are swapped out with CHP. It also gives the homeowner electric power that is competitive with utility-based power (due to addition of taxes, etc), and also gives backup power during (the rare) outages.
Here in Connecticut, with some of the highest rates in the country (Long Island and Hawaii are the only higher areas), the CHP unit I'm installing will pay for itself in under 8 years. It helps that we have net-metering. If we could get hourly net-metering, the unit would pay for itself in probably 5 years or less.
Hi goinggreen,
thanks for the reply.
You are right - there are of course places and circumstances in which micro-CHP actually makes sense.
However, please note that what might make sense for you as an individual might not make sense for society. The capital cost point remains, as does the fuel switching point and the speed of propagation point.
Also, micro-CHP units are more complicated than simple gas boilers, so probably subject to more or more expensive maintenance.
Cheers,
Davidyson
As for the comparisons, I plead haste. (I noticed, too late, that I hadn't dealt with the amount of heat we get from natural gas and coal, directly and indirectly. That would have to figure somewhere, and I intend to go back and insert a note to that effect.)
Delivering heat to homes isn't necessarily impractical. If you are willing to put a nuclear plant in tunnels beneath a city (and what better place to put it to eliminate the threat of terrorist attacks?), you could transfer the heat as medium-pressure steam to neighborhood energy recovery turbines [1] and then the exhaust low-pressure steam or hot water for space heat. The water goes back down to the steam generators by gravity. [2] I did a writeup on this almost two years ago.
When heat is not required for space heat or to drive absorption A/C, it could be vented through cooling towers. These might be integrated with office towers or other buildings.
Electric transmission losses, battery losses etc. would probably be on the order of the efficiency gains from electric drivetrains. This looks close to a wash.
The one thing I didn't consider is higher-grade heat requirements for e.g. industrial process heat. This could also be supplied by nuclear (which was the original intent of what became the Midland Cogeneration Venture in Midland, MI) but there would be a greater impact on electric output.
[1] Medium-pressure steam is probably better than low-pressure, because the pipes will be much smaller, cheaper and have lower heat losses.
[2] If the reactor is deep enough, gravity could provide a large part of the pressure required for the boiler feedwater.
Electric transmission losses, battery losses etc. would probably be on the order of the efficiency gains from electric drivetrains
And if the batteries are replaced with an overhead wire ?
No battery cycle losses (out/in), no weight to haul around (include structure to support the Battery), no wasted time and distance refueling.
Best Hopes,
Alan
This illustrates the perennial problem of comparing apples to mangoes. I will stipulate the relative efficiencies in your calculations, but of course you're still begging the question of usefulness. Oil can't do the same amount of formal work as electricity, but electricity can't do the same amount of useful work as oil. And of course by "useful work" I mean "run cars and light trucks". Given that 60% to 70% of our oil consumption is for that purpose, unless we have a generally available mechanism for turning electricity into vehicular motion (and I don't think current BEVs need apply) the whole discussion is moot.
We use a lot of oil. Some of that usage could be supplanted at low cost by electricity or waste heat streams, some couldn't. Some could be supplanted at higher cost, but some still couldn't. For me, the only utility of the "One Cubic Mile" image is to help get across to laymen just how much oil we use. The implication that the various forms of energy are interchangeable is incorrect, unfortunate and distracting.
"unless we have a generally available mechanism for turning electricity into vehicular motion (and I don't think current BEVs need apply) the whole discussion is moot."
The latest generation of li-ion batteries really do fill the bill. A123systems and Altair have much greater cycle life than conventional li-ion, thus dramatically reducing lifecycle costs.
GM indicates they are satisfied with the specs for the 2 batteries they're considering for the Chevy Volt (A123systems and Saft), and are just waiting for the engineering of large battery packs suitable for vehicles.
The only limitation right now is cost - PHEV's/BEV's can't compete at the moment with very cheap gasoline (heck, BEV's preceded ICE's, but haven't been able to compete for the same reason). That will change in the next 3 years with the Volt, and other PHEV's. Of course, it could change even faster, if gas prices rise...
What is the penetration of A123systems-driven BEVs going to be like in Africa, Asia and South America? What's the global fleet change-over time? What about cost of retraining mechanics world-wide? How much will a 20% penetration of BEVs in these regions over the next 25 years help the global problem? This problem doesn't stop at the shores of the United States.
As I say in my Musings on Peak Oil Mitigation I look on BEVs strictly as a near-term technology that will promote a softer entry into the coming depletion phase. The way I see things unfolding, if we roll into a medium term characterized by localization, the technology requirements of such devices will be too great to maintain.
By all means we should develop them. Just don't be surprised if they have less impact on the situation than some are hoping.
Electrics are much simpler, and easier to maintain than ICE's. Their lifecycle cost is going to be less than ICE's.
If you think our ability to maintain complex systems will decline, electrics are just the ticket.
I'm thinking more about our ability to produce batteries that depend on nano-technology.
Nanotech in this case is a bit of a misnomer.
Originally it referred to very complex, very small scale systems, like tiny robots. Here it really refers to very tiny particles, or materials with very altered characteristics at a very small scale due to an understanding of material dynamics at that scale.
I don't think manufacturing it is all that much harder than conventional batteries.
Of course, if you think we won't be able to maintain central large manufacturing facilities due to catastrophic economic collapse, all bets are off. I think that depends on the assumption of catastrophically fast depletion combined with a lack of substitutes for oil. I don't see either being the case.
Oh, I have no question that oil consumption will look like that chart. I just see no reason why it has to cause economic collapse.
Peak oil does not equal peak fossil fuels - coal will be with us for a long time, and it will be used if it's needed to prevent economic collapse. Peak FF does not equal peak energy - renewables and nuclear will do just fine. For that matter, peak energy wouldn't equal peak economic growth - the US could easily function with 10% of our current energy. Heck, replacing heat engines (with renewables for electrical generation and electric motors for transportation) would reduce our energy consumption by 2/3, with no loss of functionality).
Given the sun pumps out 10^26 watts, peak energy is a good deal beyond the scope of reasonable conversation.
I agree that solar/wind can supply all the energy we need, i.e. it is technically possible. However, getting to that situation without accidents under the way is what most people here think the problem is, and it is caused not by technical difficulties but by the peculiarities of the human social behaviour.
Exactly. Can Be is different than Will Be.
As a test, go to Walmart parking(or any mall) lot for a half hour, Look at the people close and consider how easy/hard it would be for them to change radically their lifestyle. Go and convince them that Economies of growth may be a thing of the past for quite a while.
Tell them that their may not be NASCAR in 5-10 years.
Tell them to meditate on "Less is More" for a while.
Or (with GW thrown in) tell them to watch Grapes of Wrath.
Peace
John
this one is for SIX
In using the calculator on the numbers put forward in this ‘CUBIC MILE ‘- thread, you’ll see some ridiculous numbers/sizes coming up - after those 50 years of compensation (for dwindling oil , as I understand it all).
As for windmills the numbers comes to 1,8 millions of them – and at a rotor-diameter set to 100m and equator at 40.000 km – the line of wind turbines will, put ‘shoulder-by-shoulder’, circumference equator 4,5 times OR more or less cover all coastlines of this planet ……. Still possible??
AND as for the insane number of the 2,1kW solar arrays (demanding ca 30m2 pr unit) – you need a jam-packed area of more than half of that of the UK – or alternatively the area of the Czech + Slovakian Republics …. Still possible????
Both of these energy converting systems are subjects for replacements and maintenance – and PV’s are prone to be depleted through 20/25 years, rendered dead and gone ….
I reckon these systems will yield much less than what ever we can imagine at our worst, reality are some times brutal (!)
Fifty times 32850 wind units is a bit over 1.6 million. If you arranged them in lines 300 meters crosswind and 1 km downwind, the entire complement would only require an area of 493,000 square kilometers (190,000 square miles). This is less than 3 times the area of North Dakota, for the entire world. (The land between the turbines can still serve for agriculture or forestry.)
50 times 91.25 million 2.1 kW solar units (which would require about 15 m2 each, not 30) would require 68437 km2 of area. If I recall correctly, the USA alone already has about twice this much area beneath impervious surfaces like roofs and pavement. The world as a whole could put this much PV on existing rooftops.
Last, today's PV panels are warranted for 25 years. They will probably produce at upwards of 60% of their rated power for 50 years. If the cells on Pioneer 6 can operate in the high-radiation environment of space for 35 years, cells on the ground which aren't mechanically damaged should do just fine.
You are failing to see my point Engineer Poet.
And you are correct for the Wind turbine numbers – I used the number 35850 – for some erroneous reason, so my number came out 10% wrong.
But I googled an 2,1 kw array saying 1 kw needed 10x15 feet , giving my number 30m2 some truth.
BUT my overall assessments to these energy-systems are the shear scale of it all – and my claim is that the cubic-mile of oil will never be substituted by these systems on a MToes basis.
Surely the spaces are readily available – that’s not the issue. THE issue is to understand the ‘simple task of pumping oil/refine it’ in COMPARISON to the ‘complexities to manufacture, maintain and substitute PV/Wturbines as per needs’ – on this grandiose scale.
Both systems are dependent on their limitations –
a) it must blow – and nominal yield is roughly 15m/s …. Think again
b) it must be day and the sun must shine …. Think again
You will never be able (in the future) to depend on such systems – but they will constitute an add-on effect which we surely must go for ….
You are arguing against scale (argument from incredulity), when today's petroleum systems have an even greater scale — thus questioning your argument ab initio.
The USA installed about 2.5 GW of wind power in 2005, up from about 400 MW installed in 2002. This is more than doubling every 2 years. Potential of the continental 48 states is about 1.2 TW average, the continental shelves about 0.9 TW average; at 0.3 capacity factor, the USA could carry on installation at 50 GW/year for several decades without reaching limits of the resource. Production scales relatively well. What's the show-stopper?
15 m/sec is well above the average design speed for a typical wind turbine. The ones I've seen are generating considerable power in 7 m/sec winds.
There is nothing intrinsically expensive about PV. Silicon isn't especially energy-intensive (unless you try making it into single crystals), and one advance like a long-lived dye-sensitized TiO2 cell would slash costs radically. Once PV comes down to a small multiple of the cost of conventional roofing or glazing materials, it will replace them. There is an enormous installed base of structures in the world, and those structures require fairly regular roof repairs or replacements.
I can depend on wind and PV producing a certain amount of energy every year, if not every hour; PV is well-matched to one major load (air conditioning). Technology like thermal storage, grid-interactive vehicles and biomass-powered fuel cells can provide the buffering capability to manage a lot more. Keep 20% nuclear and 15% hydro in the electric mix, and you're probably there — I'd have to do the numbers to be sure.
Your last argument is equivalent to claiming that because individual electric plants go off-line for minutes or weeks, I can't depend on the grid. Your logic is faulty.
If your primary energy source peaks, I can't see peak FF being far behind, if it is behind at all. "A long time" for coal, won't be as long as you think, especially if it starts to substitute for oil, at least not at the required quantities. Natural gas is already peaking, or close to peaking, in some major regions. No doubt the US could operate on 10% of current energy consumption but there will be severe hardships getting there, unless it's done over many decades.
Renewables also take resources and would take a very long time to ramp up to what is required. No doubt they'll do fine but I don't expect them to be able to allow the party to continue. And growth will trump any "solution" eventually.
Just because you think something can be done doesn't mean that it will be done, or will be done in time, or that the transitions will be anywhere near painless.
Let me address things out of order.
"Just because you think something can be done doesn't mean that it will be done, or will be done in time, or that the transitions will be anywhere near painless."
There's no question in my mind that we will go to alternatives. The process has started: wind is 1% of US electricity and is growing at 25% per year at least: that's a doubling period of 3 years, so in 15 years we could be at 20% wind easily. If all of the wind projects planned for the US in 2007 actually get built, wind capacity will double to 2%, in just 1 year. That's not likely, but it tells you something about the demand for wind, which is mostly being held back by the speed with which turbine manufacturing can ramp up. Solar is doubling every 2 years, and in 10 years will be where wind is now. The needed batteries are here, and will be on the road in 3 years (whether it's GM or an asian manufacturer).
OTOH, I'm not suggesting that the transition will be painless. As I've noted elsewhere, if depletion happens relatively fast (or if war expands in the Persian Gulf to the point of greatly disrupting oil supplies), and we haven't prepared better than we have so far, then the transition will be much more painful than necessary. The question is, how painful will it be?
My hope is that the campaign against global warming will accelerate preparations.
"Renewables also take resources"
No question. OTOH, they don't take significantly more than conventional energy. Excess costs will arise if we have to retire infrastructure before the end of it's normal lifetime, as appears necessary. That's difficult, but doable.
"I don't expect them to be able to allow the party to continue"
Why not? And why do you phrase in a way that suggests that our current way of life is vaguely immoral, and that we should return to an ebstemious, pure life of ascetism?
"growth will trump any "solution" eventually."
Not really. This is an outdated notion. Growth levels off. Population growth is doing so, and manufacturing is doing so in OECD countries. The difficult question is how to raise developing countries to the standard of living of the developed.
"If your primary energy source peaks, I can't see peak FF being far behind, if it is behind at all. "
Oil isn't our primary energy source: it only provides 40% of our energy. Imported oil only provides 24% of US energy (and yes, US oil is declining, but very slowly).
"I can't see peak FF being far behind, if it is behind at all. A long time" for coal, won't be as long as you think, especially if it starts to substitute for oil, at least not at the required quantities."
A long time for coal is 30 years. No one thinks coal will be used up earlier than that, even with the highest estimates of growth in consumption. That's all we need it for - alternatives will be in place long before then.
"Natural gas is already peaking, or close to peaking, in some major regions. "
No question, NG is going to be painful. OTOH, we don't import much now, some imports will be available (as LNG, and probably in the form of fertilizer), and it will be around for a while, even declining.
"No doubt the US could operate on 10% of current energy consumption but there will be severe hardships getting there, unless it's done over many decades."
We could convert to PHEV's for 75% of vehicle miles driven in 20 years with relatively little pain (3 years to PHEV sales, 7 years to convert most vehicles to PHEV, 10 years of sales). That would accomplish a large chunk of the reduction. Actually, it could be really good for the domestic car industry to do it that fast or faster: it would keep them solvent, if done in the right way.
Nick - maybe - maybe not ...
As I see it Peak-Oil and the add-on of Peak-Coal will coincide with peak-Fossils somewhere down the line – which in turn will coincide with peak-everything so to speak.
Now, philosophically speaking or rather physically speaking: ARE we NOT in the progress of putting the physical parameters for the atmosphere back to the stages where FOSSIL-FUELS where produced at the first stage (?) some 90 – 170 million years ago…
As we now deplete the fossil-fuels and put them back into circulation again (CO2, methane, etc) – over a few hundred years –
… and eventually if we did succeed 100% in doing this – we would reach the same temperatures (greenhouse-effect) as way back than when the oceans went green from algae(oil/gas) and the peat-swamps(coal) blossomed …
ARE we about to close some kind a circle here ????
In the US we have a lot of under-used vehicles in our inventory. We can replace 50% of vehicle miles driven in 5-6 years, easily.
Poorer countries that tend to wait for our used cars will have a harder time, no question.
One paper I found shows the 50% crossover at about 9.5 years (link), but under pressure from high fuel prices and/or legal changes it might be quite a bit less.
hmmmm. I don't see the 9.5 years. The data on page 15 suggest that 8 years of sales accounts for 49.0% of total VMT (cumulative total of %'s in 4th data column).
The same data indicates that the median life of CA vehicles is 16.6 years (total vehicle population divided by last year sales), while the same figure for the country as a whole is 12.4 years (210M divided by 17M). So, apparently California is not representative.
The same ratio (8 to 16.6) applied to the national figure of 12.4 gives 6.0 years.
I agree, that could be accelerated.
A neat editorial on why car manufactures seem so reluctant to use all these wizbang new battery technologies we see popping up.
He basically goes through and explains the wide range of operating conditions cars are used in. Then points out that these new batteries have yet to be proven in real world conditions, and until that happens big car makers can't afford to use them.
----------------------------------------------------
Keep in mind that these new battery technologies don't even claim to be usable in the entire range of real world operating conditions.
http://www.technologyreview.com/Biztech/18086/page3/
--------------------------------------------------------
It seems to me we still have a long way to go before we can electrify transportation in any reasonable way. As one commenter in the first editorial said:
Necessity breeds invention. If we need battery powered vehicles because it simply is too expensive to fuel with gasoline, then we'll have them. They may have limitations compared to normal ICE vehicles and it simply will not matter. When the question changes from "which would you rather drive?", to "would you rather drive or not?" then the answer becomes very easy.
There are other solutions, of course, like increased public transportation. I think that is a very viable idea. But EVs will definitely play a part in the solution to our dwindling oil supplies, and all of the problems they'll face will fall by the wayside as time goes on.
betting that this will happen is like going out and buying a 500,000 dollar boat on the assumption just because you bought a lottery ticket you will win the lottery. it might happen but the chances are so remote and you only have one chance to win before the collectors start calling that it's just better not to take the chance.
more simply put modern(not counting bagdad batterys) battery's have been around for almost a hundred years and to expect a increase of efficiency larger then the it's history combined is asinine.
"to expect a increase of efficiency larger then the it's history combined is asinine."
Possibly, but the increase has already happened - check out Dewalt 36 volt tools.
Need a different kind of "efficiency" here: the high price of advanced batteries reflects (in part) embedded energy, and thus their price will RISE as energy prices rise. (They'll also fall as technology improves, those two effects will happen simultaneously.) Couple that with harder economic times, and the ability to "turn over the fleet" is going to be limited, IMHO.