You may only read this site if you've purchased Our Kampf from Amazon or Powell's or me
• • •
"Mike and Jon, Jon and Mike—I've known them both for years, and, clearly, one of them is very funny. As for the other: truly one of the great hangers-on of our time."—Steve Bodow, head writer, The Daily Show

"Who can really judge what's funny? If humor is a subjective medium, then can there be something that is really and truly hilarious? Me. This book."—Daniel Handler, author, Adverbs, and personal representative of Lemony Snicket

"The good news: I thought Our Kampf was consistently hilarious. The bad news: I’m the guy who wrote Monkeybone."—Sam Hamm, screenwriter, Batman, Batman Returns, and Homecoming

March 07, 2011

The Energy Talk, part 1

By: Aaron Datesman

Dr. Patricia Dehmer is the Deputy Director for Science Programs in the Office of Science in the Department of Energy (DOE-SC). DOE-SC is the single largest funding agency responsible for basic research in the physical sciences in the U.S. A talk she delivers frequently and refers to as “The Energy Talk” is archived on her web site here.

I attended one iteration of this talk in August of last year and learned a great deal. Even without any audio, it’s worth perusing, and I encourage you to give it a look. It’s full of little gems, of which I offer this one as a teaser: did you know that the overall energy efficiency of an incandescent light bulb, from coal car to bedroom lamp, is only 2%? Since lighting accounts for ~20% of our electricity usage, this inefficiency is a very big deal.

Light bulb.jpg
(Credit to Dr. Patricia Dehmer, DOE-SC)

I will post more slides from this talk along with some explanation of their contents if there is interest. It’s a great presentation and deserves a large public audience.

(Disclaimer: I work for DOE-SC. And I like it. A lot.)

— Aaron Datesman

Posted at March 7, 2011 10:23 PM

Mike of Angle, I am doing this for you.....

Posted by: Aaron Datesman at March 7, 2011 10:39 PM

You know, those 34 units of heat could be handy in the winter. Does the efficiency calculation change to account for heating degree days? Of course, its a massive inefficiency in the summer as it runs counter to the AC... hmmm.

Posted by: Bolo at March 7, 2011 11:31 PM

I'm not a scientist, but I think there are much more efficient ways to heat than incandescent light bulbs.

Posted by: knowdoubt at March 8, 2011 05:42 AM

I would like to cast a vote for this site to have much more of this and much less of the other guy who says the same boring thing about the media over and over and over again.

Posted by: Jonathan Schwarz at March 8, 2011 09:28 AM

I'm not a scientist, but I think there are much more efficient ways to heat than incandescent light bulbs.

I'm also not a scientist, which may explain why I heat my home with several thousand light bulbs.

It's very bright here!

Posted by: Jonathan Schwarz at March 8, 2011 09:34 AM

Asked in ignorance:

So we're working on getting that 34% loss at home down -- must be a good idea.

Can we do anything about the much-larger 62% loss at the power plant?

(And what part of an average American home's power consumption comes from incandescent bulbs, anyway?)

Posted by: Earth at March 8, 2011 10:11 AM

Thank you, Aaron! I wonder if this talk is online somewhere--I hope so, because I'd love to hear it. I'll look around.

Also: you know those YouTube videos with the Xmas displays timed to, say, Trans-Siberian Whatchamacallit? That's Jon's house YEAR-ROUND.

Also, the snow consumes electricity. Don't ask me how.

Posted by: Mike of Angle at March 8, 2011 11:38 AM


The incandescent light bulb in the slide has a 94% loss in energy. This is due to the physics of the light bulb. The process an incandescent bulb uses to produce light will invariably loose almost all of it's energy to heat and not to light. Several thousand light bulbs would be very hot for a very short period of time because, by now, Mr. Schwarz's house is once smoldering debris.

Florescent lighting is much more efficient in terms of lighting output to energy input as it uses a completely different process for making that conversion then the incandescent light. It produces a much different spectrum of light that some people claim to find annoying and it has other issues like damaging paintings.

LED's are even more efficient at converting energy to visible light but they are expensive to manufacture. They also produce only in narrow spectrum ranges making white light imposable (though a wide variety of different types of LED's in close proximity could approximate it) and the light is more focused making it more difficult to light a broad area.

There are physical (as opposed to engineering) limitations on the efficiency of any heat engine (which is what most electrical plants use). I have no idea how close this power plant is to that physical efficiency limit but contrary to popular belief it is not just an engineering problem.

There are trade offs between the size of a heat engine needed to produce a certain amount of power and the efficiency. While a larger engine can produce the same amount of power more efficiently this causes unavoidable engineering trade offs as large heat engines take more resources to build and maintain. The optimization of this total efficiency is far from trivial but 100% efficient or something close is unobtainable in a heat engine.

There are, of course, other methods of generating electricity. Most (all?) hydro-electric plants use a gravity engine to produce power and the physical efficiency of these engines are usually at-least 75% with a physical limit of 100% (but with engendering limits lower). The choice of which type of engine to use is effected by the fuel used. Fossil fuels and nuclear release energy in thermal processes and are rather limited to heat engines.

Even for processes like solar, on a large scale it might be more engineering efficient to use a heat engine even if another engine is more physically efficient. The production of traditional solar panels has costs in materials and energy that is not involved in the construction of a heat engine. The largest solar plants in operation (to my knowledge) use heat engines as opposed to photovoltaic engines and I believe this is to the lifetime efficiency of the engines (including costs in production and disposal).

I have no doubt that large scale energy manufacturers have optimized the economic efficiency of their power plants. This involves law and other elements of social nature (as opposed to physical or engineering natures) and thus different laws or other social restrictions might a result. There are, however, significant limitations on any large scale energy productions humans might endeavor in. The limitations are generally larger on small scale productions.

Posted by: Benjamin Arthur Schwab at March 8, 2011 12:42 PM


I forgot to mention. Other appliances in the home operate with different efficiencies. There are likely other devices that operate at close to 100% efficiency. From the presentation, residences currently have an aggregate energy efficiency of 80.0% (not counting the efficiency of the electricity production counting this the efficiency of source to consumption energy in residences in aggregate is no more then 57.6%)

Posted by: Benjamin Arthur Schwab at March 8, 2011 12:52 PM

Pedantically speaking, there is no energy crisis, but there is (or will be) a "Gibbs free energy crisis", in other words an "entropy crisis". Boltzmann has the last laugh, on earth as in heaven.

Posted by: Cloud at March 8, 2011 01:32 PM

The point made above regarding the heat input from the incandescent bulbs is not trivial. If energy is being expended to maintain a home's interior at a higher temperature than the outside air --which is often the case when incandescent lights are in use-- then virtually all of the 34 out of 36 units of lightbulb energy being 'lost' as heat will appear as a 'credit' on the energy ledger of the home heating system.

That is, incandescent light bulbs are extremely inefficient for their primary purpose (delivering visible light), but the actual fraction of 'wasted' energy is not as dramatic as the graphic indicates.

Posted by: SunMesa at March 8, 2011 02:20 PM

I vote that Jonathan should do what ever he wishes to do.

Posted by: Rob Payne at March 8, 2011 03:54 PM

Ditto the heat-credit comments above. Yes, oil and gas heat might be more efficient than electric heat, but an incandescent bulb is apparently a 94% efficient electric heat source.

I'm definitely interested in more slides and discussion. But I can't help but notice that the content of the slide is a bit confusing -- why talk about bulbs becoming "30% more efficient" when this figure could have used the same units as the rest of the slide? I assume that this means a change from 2% to 2.6% overall efficiency, but it's genuinely ambiguous. Also, frankly, it doesn't sound all that great, given the heat credit comments already mentioned and the other annoyances of flourescent lighting. Would I be happy with all flourescent lights in my apartment in exchange for a slight dip (about 7%) in the usage portion of my electric bill and a very slight increase in my heating bill? Given that units in my town get billed by the electric company at a flat rate plus a small usage portion, probably not.

Posted by: Picador at March 8, 2011 03:57 PM

I'm going to side with "not efficient" in the heating column. You're burning coal to produce electricity at 40% efficiency, which you're then converting back into heat; you'd do much better to just burn gas at 100% efficiency. That's 2.5X fewer greenhouse gas emissions, mercury tailings, etc.

Posted by: saurabh at March 8, 2011 04:03 PM

@ saurabh:

True, it is inefficient to convert coal-generated electrical energy back into heat. But the point is that, for a home that is already heated electrically, converting from incandescent to CFL will not alter the overall (lighting + heating) efficiency at all (to first order). And for homes that are heated directly by fuel combustion, the inefficient energy input from the lightbulbs is a small fraction of the total, so again the impact of CFL conversion on the overall efficiency would seem to be minimal.

Posted by: SunMesa at March 8, 2011 04:40 PM

when i was on the stage we burned coal in the lighting elements directly. i wonder if we could do that again.

Posted by: hapa at March 8, 2011 04:46 PM

A few comments to add to the thread, which is very good.

Sunmesa is correct for certain locations at certain times of year. It's that geographical and seasonal variation which makes the whole problem so hard to untangle.

Picador's critique of the slide is correct, but his conclusion is wrong. 2% is the overall system efficiency, but the efficiency of the light source is around ~5% (2/36). I agree that an additional 30% savings there still doesn't sound like much, but that 30% would amount to 7% of all of the electrical energy we consume.

How about that!

The principal story of this slide actually is not lighting technology, it's energy conservation - what Amory Lovins likes to call "negaWatts". Note that every 1Watt of power which makes it on to the transmission line costs us 2Watts of primary energy lost as waste heat. This is why the smartest decisions we can make are simply to use less energy.

This is a fundamental physical result which follows from the Carnot cycle. It applies to any combustion-based technology.

Posted by: Aaron Datesman at March 8, 2011 05:14 PM

That chart should be used to illuminate (bad pun intended) the folly of thinking that driving a "hybrid vehicle" (such as the Progressive Stalwart Toyota Prius) is a sound "green" move. In truth it's as ecologically stupid as driving an 8mpg Hummer... and just as vanity-laden.

The White Man's Fire is such a cowardism-fueled thing. How many here leave lights on at work, or at home, in rooms that are unoccupied? How many over-light a room that is occupied?

Never mind. Technology will save us. Science fiction tells us so!

Posted by: CF Oxtrot at March 8, 2011 05:27 PM

I vote for more energy info. Note: while no existing car is great ecologically (not even pure electrics) a Prius is better than a Hummer. In areas where coal if 50% or less of the Grid, and all-electric is better than a Prius. And yes I'm considering embedded energy and disposal. I do throw out life cycle studies created by public relations firms working for the auto industry who don't disclose their data or methodology - such as the notorious "dust to dust" series. Electric cars charge by a largely renewable grid would be fairly low on greenhouse gas emissions, especially if the cleanest rather than dirtiest manufacturing methods were used. (Those would be two huge changes from the present.) But there are important ecological consequences to auto manufacture and use aside from GHG issues.

Posted by: Gar Lipow at March 8, 2011 05:41 PM

charlie foxtrot, what are you talking about?

are you saying that plug-in vehicles are not as green as they seem because they draw power from the dirty electricity grid?

i hope not. because that would be shameful. the need to clean up our electricity supply as we electrify transport is about the most talked about thing in 'repowering america' that i can think of.

virtually every big plan estimates how much renewable power would be needed to push all the trains, trucks, cars & motorbikes.

btw here is a plan.

as you can see right in table 1 of part 1, they list big energy overhaul plans and whether those plans account for transportation footprint.

Posted by: hapa at March 8, 2011 05:56 PM

You throwbacks are still using incandescent lightbulbs? Jesus, next you'll be telling me you're still using the Imperial system of weights and measures.

Posted by: weaver at March 8, 2011 06:38 PM

hapa, I'm saying what I already said.

Apparently you want me to be saying something else.

I don't want to sully your or anyone else's Progressive Success at taking one type of emissions and substituting them for another, just to sell more cars.

How else will the auto mfrs stay afloat, if not for such shell games?

Alterna-fuel cars are just as dirty as gas or diesel cars. You can try to rationalize the "hybrids" as "green" if you like. I'm sure it makes for fun pseudo-intellectual sport. It's a great way to "win" non-existent "debates" on the Toobz.

Cash for Clunkers! Buy more! Throw more stuff away! Better 'cuz electric!

Madison Avenue sure saw a lot of Progressives coming down the pike, that's for sure. From here, however, it looks like calves being pushed onto the branding table. I'm stepping out before I get coated in scour and pee.

Posted by: CF Oxtrot at March 8, 2011 07:47 PM

cluster foxtrot, you wrote 7 paragraphs, and only 6 were crap! THAT IS NOT GOOD ENOUGH.

Posted by: hapa at March 8, 2011 08:52 PM
Sunmesa is correct for certain locations at certain times of year. It's that geographical and seasonal variation which makes the whole problem so hard to untangle.

My suggestion that conversion to CFL would have little impact on a home's overall (lighting + heating) energy efficiency would seem to be correct under any circumstances where the outdoor air temperature is lower than the set point of the indoor thermostat, excepting those homes (e.g., Jonathan's) in which a substantial fraction of the indoor heating is supplied by incandescent bulbs. Consider also the geographic distribution of the major lighting/heating energy consumers worldwide.

Just sayin', the incandescent bulbs in my own home are replaced, as they fail, with CFL. I support the principle, but have no illusions that it makes any appreciable dent in the broader predicament. Which is dire indeed.

Posted by: SunMesa at March 8, 2011 09:15 PM

LAUNCH DRYICE TO THE MOON, sequester CO2 safely for the long term.
Pump it in the ground, just not in MY backyard. Can't disolve it in the oceans. Magical thinking or not WE STIL will burn coal for the rest of this century. Get rid of that guzzler and buy an electric and THAT'S if it happened over night for EVERY vehicle on the road, there's aircraft and industry, bulk shipping, hell of an investment just to get off oil.
An electric rail launcher system, combined with distillation of coal to at least remove carbon dioxide and eventually pure carbon before use, SOME time may be bought to eventually bring non fossil fuel to 100% use.

Posted by: Mike Meyer at March 8, 2011 11:06 PM

when i was on the stage we burned coal in the lighting elements directly. i wonder if we could do that again.

That is very cool.

Posted by: Mike of Angle at March 8, 2011 11:11 PM

sunmesa in order to keep the costs of fuel switching EVERYTHING under control all electrical demand needs to be brought down. this means super efficient application of energy, in highly but not irritatingly energy conservative environments.

in other words the lightbulbs reduce demand for electricity from lightbulbs. if the old bulbs were seriously affecting how much electricity was demanded by the HVAC, something is very very wrong with that building's equipment & sealing.

Posted by: hapa at March 8, 2011 11:12 PM

I'm not sure I follow. An incandescent bulb produces 100 watts of power, and thus contributes 100 watts to the home heating, at ~ 40% efficiency in terms of GHGs compared to gas heating. A CFL bulb produces 30 watts of power at the same efficiency, requiring 70 watts of gas heating to make up the shortfall.

In the first scenario you produce 100/0.4 = 250 units of GHGs; in the second you produce 30/0.4 + 70 = 145 units of GHGs. If a typical home requires 3300 watts of heating and uses something like 10 lightbulbs, this seems like it would be a pretty significant difference.

Posted by: saurabh at March 9, 2011 04:12 PM

There are trade offs between the size of a heat engine needed to produce a certain amount of power and the efficiency. While a larger engine can produce the same amount of power more efficiently this causes unavoidable engineering trade offs as large heat engines take more resources to build and maintain.

The first is often true. The second is news to me. More efficient engines are more expensive, but not necessarily larger ones. Combined cycle turbines that convert natural gas to electricity with around 60% efficiency cost around $600 per KW of capacity in large sizes. (Been a long time since I priced t hem, so don't remember of the minimum size for this is 300 Meg or a Gig. Somewhere in that range anyway.) However the same sized turbine will cost $250 to $350 per KW in a 35% to 45% efficient single cycle turbine. A 25 KW turbine in contrast will cost a great deal MORE per KW. As far as I know larger engines are cheaper per KW, not more expensive.

In terms of maintenance: as far as I know MTBF (mean time between failures ) is smaller, not larger for smaller heat engines. In general maintenance costs per KW of capacity and per kWh produce goes down, not up as heat engine size increases. That is why, more and more, the current generation of solar farms are moving to models where you have many mirrors focusing on a single large heat engine, rather than modular farms with many tiny heat engines. One big heat engine not only increases efficiency, and lower initial costs. It lowers maintenance costs. If you want modular solar energy, that is where solar cells shine(heh) compared to heat engines.

One possible exception: some very big companies are about to deploy expermental systems where modular heat engines with no moving parts convert low temp heat (about the same temp as solar hot water) to electricity without about 15% efficiency. The inefficiency is made up for by combined heat and power where the waste heat is used to heat hot water, and in the winter to provide space heat. Obviously no moving parts means it is not a heat engine in the conventional sense, don't know whether it is thermoelectric or what. At any rate if it works the advantage will be lower initial cost that solar cells or CSP and low maintenance. We will see how it works in practice. Thermoelectric generation and other alternatives to conventional type of heat engines have their own problems with maintenance.

Posted by: Gar Lipow at March 9, 2011 10:37 PM


I know you're smarter than your comments here indicate, so I assume you don't understand how hybrid cars work.

Hybrids don't run off the electric grid. They recapture kinetic energy that is wasted in conventional cars (e.g. in braking) as electric energy, using this to power much of the vehicle's operations. They are simply more efficient versions of conventional automobiles, with these efficiencies manifesting especially in start-and-ctop driving environments like city streets.

You can legitimately ask whether they are more expensive (i.e. energy-intensive) to manufacture, and whether this outweighs the efficiency of their operation, but the data all point the other way.

Posted by: Picador at March 10, 2011 09:37 AM

Gar Lipov:

"As far as I know larger engines are cheaper per KW, not more expensive."

As a point of clarification, I was referring to absolute cost not per KW cost. While per KW cost over the entire life and all inputs of an engine are important, I have no knowledge of what these costs are. I see no reason not to believe what you have said and I accept it.

In your example (the 60% efficient vs. 45% efficient), the more efficient engine costs more per KW and I would expect this is due to a larger volume engine. In terms of lifetime pricing, building a larger engine (to produce more power with a larger volume) can reduce costs because of fuel consumption over the lifetime due to potential efficiency gains, one needs to fight the construction, disposal, and maintenance costs as well as potentially wasted excess capacity. These conflicting drivers lead to the best engine and the best efficiency not being necessarily straight forward. Why would a electric company spend x amount of money increasing efficiency by y amount if the savings in fuel are less than x?

I see no reason to doubt that electric companies are producing electricity in the most economically efficient fashion. This takes a lot of things like law into account. I think that if the full externalities were paid for fossil fuel usage and production (I have every reason to suspect that they aren't) then I think the way energy is produced and consumed in this country will change rapidly and drastically with minimal direct consumer involvement.

My heat engine discussion was in large part to point out that questions of efficiency are complex and a simple analysis would be inadequate. They power plant might have good reason not to produce electricity as physically efficiently as it can given the situation it is in. The efficiency of the power plant is economically optimal given the social situation. This might be very different from the socially optimal efficiency. Changing the power plants efficiency requires changing the social situation which is something I advocate for.

Posted by: Benjamin Arthur Schwab at March 10, 2011 11:06 AM