The Talk Turns to Energy at SXSW

South by South West is arguably the largest interactive festival in the world.  Certainly, it’s the most friendly.  When a bunch of tech geeks get together, the talk invariably turns to energy.

I moderated the panel, “Can the Creative Class Transform the Energy Business.” Garry Golden and I talk about the energy discussions I’ve been having at SXSW.

We talk about:

  • the Chevy Volt
  • takeaways from the energy SXSW panel.
  • upcoming podcasts, such as, lessons learned from the telecom industry that we can apply to electrical utilities.
  • when the grid becomes a software platform
  • are electric vehicles the new electricity reservoir?

Why talk about energy at SXSW?  Because this is a crowd of people who have grown up swimming in innovation.  The SXSW interactive audience has a lot to contribute.  Today began the conversation of how they can participate in energy.

Also, the energy industry feels a lot like the computer industry in 1982, or the Internet industry in 1992.  That is, we’re on the verge of new innovation.  You can feel it in the air.  No one was really having this conversation before at SXSW.

These were the people on my panel:

Mark Kapner
Austin Energy

Richard Donnelly

Gregory Kallenberg

The image is a Chevy Volt chasis that was on display outside the Austin Convention Center at SXSW.


  • Brent
    September 6, 2010 - 1:19 pm | Permalink

    This is less of a comment and more of a request for an article or podcast. The subject would be the so-called “long-tailpipe,” and more particularly, the amount of energy required to get a unit (gallon, liter) of gasoline to the pump. I have heard so many innuendos, allegations, and unsubstantiated assertions slung around on this subject that my head is spinning. Take these claims (I’m providing no sources because I can’t remember where I read them, but I’ll dig ‘em up if necessary):

    1) “It takes 1 barrel of oil (energy equivalent) to produce 10. It used to take 1 barrel to produce 100.”

    2) “Refining a gallon of oil uses 7.5 kWh of electricity. That’s enough to power a Nissan Leaf for 30 miles.”

    3) “Refineries are net producers of electricity and actually sell it to the grid.”

    4) “Refineries use between 1 kWh and 14 kWh to refine a gallon of gasoline.”

    5) “The delivery efficiency multiplier for oil is about .8 to the pump. It’s about .3 for electricity.”

    6) “If you include the carbon cost of food, bicycling at 20 mph emits about the same as my Tesla Roadster at 20 mph.”


    Actually, if the only number you could come up with is how many kWh to refine gasoline — and make a convincing, bulletproof case for it — I’d consider it a yeoman’s work.


  • Joel Greenberg
    September 7, 2010 - 2:29 am | Permalink


    Thanks for the comment. It’s an excellent question and one that I find surprisingly difficult to answer without turning to industry experts. You’d think it’d be easy to find how much energy it takes to make a gallon of gas. I mean, you’re boiling oil here. The technology’s well known. But there doesn’t seem to be any great resources for people who aren’t petroleum engineers without lots of digging.

    The EPA’s Energy Star program has a petroleum refining focus. In their 2005 Report Energy Efficiency Improvement and Cost Saving Opportunities For Petroleum Refineries: An ENERGY STAR® Guide for Energy and Plant Managers, the authors write “The petroleum refining industry is one of the largest energy consuming industries in the United States.The petroleum refining industry is an energy intensive industry spending over $7 billion on energy purchases in 2001

    Others write that 43% of a refinery’s operating cost is devoted to energy. So we know energy is major part of oil refining. No brainer, I guess.

    How do you get to your answer? Here’s the process I’d take.

    Crude oil’s specific heat is 0.8 BTU/degree F. That’s the amount of energy it takes to raise one pound of crude oil one degree. (A BTU is defined as the amount of energy it takes to raise one pound of water one degree, which it turns out, is about the amount of energy in one match.) Specific heat, then, is the amount of energy it takes to raise one pound of a substance one degree, in relation to the specific heat of water.

    This means crude oil takes less energy to raise one pound of crude oil one degree than water. (Now, crude is a complex mixture of all kinds of things, so I know I’m wrong already by thinking about it as a uniform substance, but for simplicity, let’s just go with this assumption.)

    According to How Stuff Works, high pressure steam is raised to 1112 degrees fahrenheit and that, in turn raises the temperature of the oil.

    So, to find the amount of energy needed to create that steam, assuming the water they’re using is at 75 degrees:
    1) Determine the BTU’s needed to raise 1lb of water 212-75=137 degrees. That’d be 137 BTU’s
    2) Determine the amount of energy needed to boil the water once it reached the boiling point (that has to do with the “latent heat of fusion”.)
    3) Determine the amount of energy needed to raise the steam to 1112 degrees F. To be thorough, add the energy to keep the pressure on as well as the energy that goes into the steam itself.

    4) Figure out the efficiency of the boiler unit as it transfers the heat from the steam to the oil to boil it. It will certainly be less than 100%, but how much?

    5) Multiply the energy efficiency of the boiler TIMES the energy needed to raise one pound of water to 1112 degrees TIMES the specific heat of crude oil (0.8, which means the energy needed to raise one pound of crude oil one degree, in relation to the energy needed to raise one pound of water one degree, measured in BTUs/Degree) TIMES the number of pounds of water needed to heat one pound of oil.

    THAT should give you the amount of energy needed to boil one pound of crude oil. That’s to start the process. That doesn’t take into consideration the energy of all the processes initiated once the crude oil is boiled.

    This is a start. Maybe it can be used as a double check on whatever numbers any one of us find.

    Now, let me give you some of my opinions on the statements you heard.

    1) “It takes 1 barrel of oil (energy equivalent) to produce 10. It used to take 1 barrel to produce 100.”

    According to the NEED Project, oil refineries are 25% more efficient today than they were in 1973, so the basic idea that we’ve gotten more energy efficient sounds right. Whether the numbers are right is another story.

    2) “Refining a gallon of oil uses 7.5 kWh of electricity. That’s enough to power a Nissan Leaf for 30 miles.”
    Don’t know. Although I’d imagine the refinery would more likely use natural gas to boil the water that creates steam, rather than electricity. Be that as it may, 1 KWh = 3412 BTU, so they’re talking about 25,590,000 BTU’s to refine one gallon of oil.

    3) “Refineries are net producers of electricity and actually sell it to the grid.”
    Putting my skeptical spectacles on, I’d say…um…no. Net producers, I’d think not, seeing as energy is 43% of a refineries operating expense.

    However, I think it’s totally conceivable that refineries sell energy back to the grid. They’re all about energy, so I bet they produce some of their own electricity and have excess to sell at times. Especially with combined heat and power, I could certainly see where a refinery would take the waste heat from boiling water and running it through a generator to create electricity instead of just wasting it. I can see where they have excess electricity to sell because they use fossil fuels for heat.

    4) “Refineries use between 1 kWh and 14 kWh to refine a gallon of gasoline.”
    Don’t know. Need to check that. But, it seems consistent with your previous question that mentioned 7.5 KWh to refine a gallon of oil. In doing research, just remember 1KWh=3412 BTU, so you can convert between one and the other.

    5) “The delivery efficiency multiplier for oil is about .8 to the pump. It’s about .3 for electricity.”

    Not sure what this is saying. Seems to me a “delivery efficiency multiplier” would be the ratio of energy in to energy out, with 1 being a frictionless, perfect system. The higher the number, the more energy efficient. If so, then these numbers don’t make sense. According to Amory Lovins in “Winning the Oil End Game” and in his talk on TED, he makes the point that electric motors are 80% efficient in moving passengers forward, as opposed to internal combustion engines, which are 7% efficient (if my memory serves). Meaning, 93% of the energy in the gasoline an ICE engine uses ends up as waste heat.

    Therefore, I’d say this statement is false. Maybe the author mixed up the two numbers?

    6) “If you include the carbon cost of food, bicycling at 20 mph emits about the same as my Tesla Roadster at 20 mph.”

    These comparisons are so imprecise. To clarify, we’re talking about the carbon cost of the fuel and the carbon cost of converting that fuel to energy. To get the answer, we’d need to:
    – decide where the food came from. Locally grown, organically grown food would have a lesser carbon footprint that farm factory produced, food I’d think.
    – understand how the electricity was generated to charge up the Tesla’s batteries.
    – know how much carbon a human being exhales when bicycling at that speed.

    I’m skeptical of this statement. Seems to me, bicycling would be one of the most efficient ways to cut the carbon on personal transportation. It’d also have increased health benefits. Maybe it’s a Tesla owner’s way to justify not exercising regularly?

    But to get to the heart of your post, I’m sure with a few phone calls, I can confirm the amount of energy needed to refine one gallon of gas. Then, we can convert to KWhrs if the answer comes back in BTUs.

    It’s an interesting question. Thanks for posing it as a direction to go.

  • Brent
    September 7, 2010 - 9:55 am | Permalink

    Thanks for the detailed response, Joel. I’m not so good with energy math, but I’ll try to see what I can do on a spreadsheet.

    I should clarify some of the numbered claims in hopes of pinning down some issues. I didn’t organize them very well for you, and I should rectify that.

    Claims 1 and 5 (concerning oil): These claims essentially say the same thing. If you assume a hypothetical gallon of gasoline in the ground, and then ask how much of it you’d have left after siphoning off all the energy demands of getting it to the pump, you’d have left either 80 or 90 percent of it. Claim 1 makes the additional point that in the old days (i.e., 19th & early 20th centuries) of oil extraction, you’d have left about 99 percent of it. Claim 5 makes the additional point that with electricity, you only have 30 percent of the energy left after all the losses associated with delivery.

    The claims matter when it comes to comparing electrics with gasoline cars. If you say that electrics are 90 percent efficient, and “gasolines” are 20 percent efficient, you might calculate total efficiency something like this:

    1) Electrics: 90 efficiency X 30 delivery = 2700
    2) Gasolines: 90 delivery X 20 efficiency = 1800

    And then you might argue that it’s not such a great difference, certainly nowhere near the “three times more efficient” claims you often hear about electrics. (Of course, my math is probably bad, but I seem to recall calculations almost exactly like this.)

    Claims 2, 3 4: These claims illustrate the range of answers on electricity used in oil refining. When Nissan has started using the 7.5 kWh number in its advertising, some people seized it as the most credible figure yet. However, there doesn’t seem to be any publicly released supporting evidence behind it.

    If the Nissan number is true, it would seem to eviscerate the long-tailpipe claim made on the gasoline side. However, if the 80-90 percent delivery efficiency number is true — which would seem to account for refinery losses — then the long-tailpipe people have a better argument.

    Claim 6: This claim comes from a Tesla Motors Club bulletin board posting. I wanted to argue against it as ridiculous (and even the author says it has real-world limitations), but when I looked at the assumptions and math, I found that I had few rebuttals. I think it’s obvious if you assume solar charging for the Roadster, fossil fuels for the bicyclist, and ignore carbon costs in constructing the car, the Roadster will come out ahead. There are a hundred objections to the proposal, but perhaps you might enjoy reading the post for yourself:


    Naturally, all these claims are usually predicated on the current mix of electrical generation. I’m usually an optimist about the future of electrical generation, and carry the hope that our power will only get cleaner from here.

    Thanks so much for taking an interest in the subject.

  • Derick
    September 7, 2010 - 4:41 pm | Permalink

    Several interesting questions raised…

    A few thoughts on a couple of them.

    When you ask about the energy cost of a gallon of gas at the pump, what type of crude are you starting with? Saudi sweet, Canadian shale? In the old days, crude could be scooped up from pools. The cost of getting and transporting can vary a lot. What costs do you want to include or exclude? Do environmental impacts or military support count? Refining is on piece, but so is extraction, transportation, etc.

    This link may not give you a complete answer on the bike question, but it has a few sources that may be of use:

  • Comments are closed.