Transition Culture

An Evolving Exploration into the Head, Heart and Hands of Energy Descent

Transition Culture has moved

I no longer blog on this site. You can now find me, my general blogs, and the work I am doing researching my forthcoming book on imagination, on my new blog.

15 Apr 2011

A chat with ‘The Nation’… and just how much energy is in a litre of petrol?

The other week I did a chat with Karen Rybold Chin for ‘The Nation’ as the final part of a thirteen-part series called “Peak Oil and a Changing Climate”.  Here it is….

Following it, Walter Haugen at Local Harvest wrote a post questioning my “1 litre of oil equals 35 days of human work” figure.  My source for that was from the end of a report done by FEASTA a few years ago which concluded that “a 40 litre fill-up at a petrol station is the equivalent of about four years of human manual work”. Walter’s calculations, which argue that my figure “overstates the energy value of a liter of petrol by almost a factor of four” are here.  Be great to hear your thoughts on this.  Not being a mathematician or an engineer myself I’d value your thoughts/analysis….

Comments are now closed on this site, please visit Rob Hopkins' blog at Transition Network to read new posts and take part in discussions.


Patrick Aucoin
15 Apr 8:12am

Transition Culture, everybody climb on board!

15 Apr 8:34am

Great to have an early morning Transition Top up !

Thanks Rob, you speak so well and your boundless enthusiam is infectious.

Hope you all have a good break at Easter !

Marcin Gerwin
15 Apr 8:38am

Rob, Walter Haugen seems to be right on this one – “a liter [of petrol] can power a car for 30 miles”.

This number is tricky, beacause it depends on a car. Small cars burn typically 5-6 liters of petrol per 100 km, and that’s quite efficient, because, for example, Volvo XC70 burns as much as 11,2 liters of petrol per 100 km. Let’s assume that the mean is 8 liters per 100 km. This gives us 12,5 km on 1 liter on petrol, which translates into 7,8 miles on 1 liter.

However, you could run even 62 miles on 1 liter of petrol, but your car would have to be a hybrid and look like this:

James Driscoll
15 Apr 8:48am

Hi Rob and Walter,
although I’m not an expert I suspect this is another energy returned on energy invested problem.
As walter points out 2500 calories is an often quoted as a useful intake for males. If I sit on my bum all day and do no useful work my body will still keep me breathing, digesting and my brain will still keep on pondering the lilly – all this activity will burn quite a lot of calories. )

When you stick petrol in a car pretty much all the energy of it goes into moving the car (if you don’t consider heat waste or what ends up in the exhaust)

So if a persons resting metabolic rate accounts for 75% is of daily calories burned it may be a Walter who is a factor of 4 out.

Walter may be calculating how many calories he needs over 9 days but if he’s not calculating how far he can push a car in 9 days (it might take him 35 days to get the same distance as a litre of petrol can)

kind regards

James Driscoll

James Driscoll
15 Apr 8:53am

apologies for all the typos in my last post – I’m still on my first cup of coffee – please feel free to edit 🙂

James Driscoll
15 Apr 9:41am

Further to my ealier post, if this site is correct I would spend just under 2000 calories doing nothing leaving about 500 to do some ‘useful’ work.

15 Apr 9:47am

Yes I’d agree his numbers look good. It also depends if you want to talk oil or petrol.

1 tonne of crude contains 41.9GJ of energy and there are 1,192 litres in a tonne. This gives 35.1MJ per litre.

1 food calorie represents 4.2kJ of energy so there are around 8,360 kilocalories (food calories) in a litre of crude.

Using his value of 125 calories per hour that’s 67 hours of human labour.

If you’re talking petrol then there’s 44.7GJ of energy and 1,362 litres in a tonne. But then there’s also an additional 14% embodied energy in the production of the litre of petrol giving you 37.4MJ per litre or 8,940 kcal. This works out at 71 hours of labour.

Call it an even 70 hours at 8 hours a day gives you about 9 days.

Energy content & density:
Embodied energy:

15 Apr 12:28pm

The comparison between oil and labour depends on whether we are looking at energy input or work output. I think Haugen’s numbers are roughly right, but for the wrong reasons (e.g. his work output assumption assumes that all 2000 calories of daytime energy consumption can be converted into useful work)

A litre of petrol contains roughly 10kWh of energy – run through a typical IC engine (~40% thermal efficiency), it could contribute 4kWh of useful physical work (if we’re talking about heating, you can use the whole 10kWh)

A human labourer can sustain an output of useful work of roughly 75W over an 8-hour day = 0.6kWh (this is half the 145W that Haugen’s 125cals/h consumption-based figure suggests)

4 / 0.6 = 6.7 days of physical work per litre of petrol (comparing output to output)

Dan Olner
15 Apr 12:56pm

Oh dear, I wrote a long comment, did it get rejected or not get through?

Dan Olner
15 Apr 12:59pm

Oh no, now I’ve made a real mess: I think there’s a moderation filter for long posts, not short ones. Please delete as appropriate and very sorry for messing up the thread…

Dan Olner
15 Apr 1:13pm

Sorry again: full workings out here, I get about 1 human day per litre of diesel, if we’re using people and diesel to generate electricity. It also turns out, at current UK fuel prices/minimum wage, direct people-power would be a hundred times more expensive than liquid fuel…

James Driscoll
15 Apr 2:28pm

I’ll try again…

Hi Rob and Walter,
although I’m not an expert I suspect this is another energy returned on energy invested problem.
As Walter points out 2500 calories is an often quoted as a useful intake for males. If I sit on my bum all day and do no ‘useful’ work my body will still keep me breathing, digesting and my brain will still keep on pondering the lilly – all this activity will burn quite a lot of calories. (e.g. see for what a person on ‘idle’ is likely to burn)

When you stick petrol in a car pretty much all the energy goes into moving the car (I know there are loses but hopefully the petrol is doing useful work rather than just idling the engine in a traffic jam!)

So if a person’s resting metabolic rate accounts for 75%ish of daily calories burned it may be a Walter who is a factor of 4 out.

Walter may be calculating how many calories he needs over 9 days but that’s very different from comparing how far he can push a car in 9 days compared to how far a litre of petrol could take it. (it might take him 35 days to push it the same distance because he can only use aruond 25% of his calorie intake to do ‘useful’ work)

To put it another way if we could all drink petrol 75% of the petrol would be used in keeping us alive and we would only be able to use around 25% of it to do useful work. This is clearly not the same as a car engine where much more of the energy in the petrol can be used to do something useful.

So, in a sense, you both may be correct but you are talking about slightly different things.

I hope this makes sense!

James Driscoll

15 Apr 3:51pm

This might make a useful experiment/fundraiser.

1- Get two high school football teams & coaches onboard to donate their services for two hours (or more).
In most of the world, this would be what the US calls a soccer team; in the US, it would be called a football team.

2- Find a track where you can push an SUV around and around in a circle. Calculate its MilesPerGallon [KilometersPerLiter] beforehand.

3- See how long various teammates can push the SUV around the track before they give up & which team can outdo the other.

4- Celebrate after. Age-appropriate beverages for the exhausted young athletes & applause for their hard work.

Not the same as a mathematical equation, but a dramatic human demonstration of the power of a tiny gallon of petrol.

Robert Wagner
15 Apr 9:43pm

About the same energy as in 100 CCs of food. After all, isn’t the ratio about 10 calories of oil to produce 1 calorie of food? Collapse is an amazing technological breakthrough, we get to produce our food directly and avoid the 1000% deadweight loss of having to convert to petrol first.

Gunnar Rundgren
16 Apr 9:03am

You can read my calculation about the relationship between human energy and energy in oil, on which is an extract from my book “Garden Earth, from hunter and gatherers to global capitalism and thereafter”. It is true what some other people commented that it depends a lot if you are speaking about input or output, both for the human and for the oil (or the petrol). Humans output is 20%-25% of the input, oil when used for various work has a somewhat higher ratio. My conclusion is that a barrel of oil corresponds to 14 years of work, so that is quite close to yours. But again it depends on how you count.

andrew ramponi
16 Apr 10:54pm

The Oil Drum had a discussion on this a couple of years ago for those really interested.

Also in response to James’ “pretty much all the energy goes into moving the car”.

“Despite 119 years of refinement, the modern car remains astonishingly inefficient. Only 13 percent of its fuel energy even reaches the wheels-the other 87 percent is either dissipated as heat and noise in the engine and drive train or lost to idling and accessories such as air conditioners. Of the energy delivered to the wheels, more than half heats the tires, road and air. Just 6 percent of the fuel energy actually accelerates the car (and all the energy converts to brake heating when you stop). And, because 95 percent of the accelerated mass is the car itself, less than 1 percent of the fuel ends up moving the driver.” – Amory B. Lovins from “More Profit with Less Carbon.”

I conclude: What a waste!

17 Apr 9:29am

The Feasta report reckons on an averagely fit person generating usable power of 70W. 70W for 8hours is 2 MJ of energy. A litre of gasoline is about 35 MJ (wikipedia) so that means 17.5 days.

I think the 70W is low. You need to consider someone used to physical work rather than an averagely fit person. David MacKay in Sustainable Energy Without the Hot Air reckons on 1 kWh/day/person which is equivalent to 125W. When I am on my exercise machine it says I generate around 200W but I couldn’t maintain that all day.

Walter Haugen
17 Apr 11:08pm

After doing some research on Rob’s source (FEASTAS) and several other websites, my conclusions remain the same. The problem is in the major theoretical flaw of not accounting for the energy “wasted” by the engine. FEASTAS used a generator powered by a human and lighting up a 70-watt bulb. This does not take into account the energy used by the operator to move his muscles and sweat and elevate his heart rate, etc. A lengthy treatise can be found on my Local Harvest blog here:

By the way, thanks for actually addressing this question Rob. I think it is very important. Not accounting for inputs is one of the reasons the economists have gotten away with dismissing energy use as an “external” for so many years.

Alex Hallatt
18 Apr 10:09am

Great interview – nice to be reminded of the positive outcomes of moving towards a more community-based economy.

I’ve just come back from visiting friends in Lyttelton (epicentre of February’s major NZ earthquake). Though not a transition town in name, it is in nature and the local business networks, timebank and farmers market will ensure that this town is back on its feet long before neighbouring Christchurch.

Peter Barber
18 Apr 2:55pm

I’ve done a simple comparison, which applies to manual labour, and I don’t agree with either Feasta or Walter Haugen. See what you think:

Energy content of petrol = 43.2 MJ/kg
Density of petrol = 0.720 kg/L (data from my chemistry textbook)
Therefore energy content of petrol
= 43.2 MJ/kg × 0.720 kg/L
= 31.1 MJ/L

Typical efficiency of conversion of chemical energy of petrol to _useful_ work by an internal combustion engine (ICE)
= 20% (this figure will vary with engine speed, compression ratio, etc.)

Therefore typical useful energy content of petrol
= 31.1 MJ/L × 20 / 100
= 6.22 MJ/L

Sustainable human mechanical power output (from David MacKay’s estimate)
= 125 W

Therefore daily output (assuming 8 h of work per day)
= 125 J/s × 3600 s/h × 8 h/day
= 3.6 MJ/day

Petrol equivalent of human labour
= (3.6 MJ/day) / (6.22 MJ/L)
= 0.58 L/day
(i.e. 0.58 L is equivalent to one person-day of labour)

Or, expressed as the reciprocal
= (6.22 MJ/L) / (3.6 MJ/day)
= 1.73 day/L
(i.e. one litre of petrol will perform the same amount of useful work as one person working at 125 W output for nearly two 8-hour days)

NB. This comparison applies to high-friction tasks (e.g. digging, sawing, ploughing) or stationary tasks (lifting, winding). It does not apply to transport, because muscle-powered and ICE-powered transport are not directly comparable. ICE-powered vehicles are heavy and travel at relatively high speeds, and therefore require very large amounts of energy to start and stop. They also expend large amounts of energy countering air drag (which is proportional to the square of the speed). Human- or animal-powered transport, on the other hand, is slow, requiring little energy to start and stop and making air drag irrelevant (except in headwinds!).

PS. I realise that this doesn’t factor in all the other issues such as EROEI, or land use, or petroleum use in agriculture. But then the point of contention was the equivalence of petrol to human power.

Dan Olner
18 Apr 5:12pm

Peter Barber: “(i.e. one litre of petrol will perform the same amount of useful work as one person working at 125 W output for nearly two 8-hour days)”

That’s kind of the ballpark I got for comparing diesel vs human powered electricity generation. Yours is 1 litre for 16 hours, I got 1 litre for 25 hours, so we’re on a 2/3 ratio to each other, pretty close.

The difference seems to be between those analyses figuring out the work that can be done (more a physicist approach? I’m not one, mind!) compared to thinking about total embodied energy.

I’d be interested to dig a little more into what assumptions people have about all this.

19 Apr 2:34am

While I like the physics discussion ( esp Peter Barber who I think is closer to the leverage issue in sound working metaphysics ) the anthopomorphic focus is a little disturbing- i am not competing against petrochemicals .
Thankfully in real life photosynthesis does all the work and …..we get to rest . Truly wonderful concept, especially when we find ourselves outside the garden from time to time

19 Apr 1:38pm

OK, let’s see if we can ground this discussion in the world in which we live, rather than the hypothetical world of mathematics…
I have a truck with 25 tonnes of bricks on board. I’m going to drive it from here to over there. Over there is ~5km away. I’m going to use 1L of diesel to get there (see for Mercedes and their Guinness book of records fuel consumption attempt). Oh, alright, I have a Kenworth, so I’m going to use 2L of diesel to get there (let’s be generous).
Now, you’re going to move another 25 tonnes of bricks from here to over there. I’m going to be really generous this time and give you a wheelbarrow, which will up your payload for each trip from six bricks, 12kg (my maximum load with no implements) to 30 bricks, 60kg (again, my maximum comfortable load with a wheelbarrow – and, unlike many geeks, I’m a long way from unfit).
For 25 tonnes, you’re going to do 417 round trips to move those bricks. With a full wheelbarrow, you’ll manage 4km/h maximum from here to there, probably less. From there to here with your empty barrow, you’ll probably manage 6km/h if you’re really fit. So, here to there will take 75 minutes, there to here will take 50 minutes, for a grand total of 125 minutes. Let’s be generous again and call it 2 hours…
So, for an eight hour day, four barrowloads of bricks from here to there, or 104 days to move the whole 25 tonnes. Since I used two litres of diesel to move my bricks, to get us back to a 1litre equivalent we divide by two to get 52 days of your labour to 1 litre of diesel.
I think, therefore, that Walter’s argument fits into that category that scientists call “not even wrong”

Dan Olner
19 Apr 3:15pm

Les: hey up, fine example! Mine was connected to real world examples… well, except that you wouldn’t actually ever try and power your house with cyclists. At least, not yet.

Anyway: fancy doing the sums for using a freight-bike rather than a wheelbarrow? E.g. –

‘freight bikes’ are commonly used globally (much less efficient ones):

I’ve seen recumbents in use at festivals: you can probably assume they won’t go much faster than walking with a full load, but the load can be much, much larger than a walker could deal with. (Compare distance a cyclist can go compared to a walker in a given time, using the same output. Bikes make much more efficient use of our legs.)

Walter Haugen
19 Apr 5:18pm

Here are a couple of points that seem to be missed by most of the posters.
1) It takes energy to do anything.
2) Most of that energy goes up the smokestack in the form of heat and gases.
3) Work output does not measure the energy that goes up the smokestack and is “wasted.”
4) The “wasted” energy has a more profound effect on the physical environment than the work output.
5) If you only measure work output, you are not measuring total energy.

Dan Olner
19 Apr 5:46pm

Walter: “If you only measure work output, you are not measuring total energy.”

True, but I don’t think people missed that point. It depends on the question you’re asking, doesn’t it? You have to draw a line around the problem, define it carefully. Otherwise, for example, do we have to discuss the total energy requirement to produce a healthy human being able to carry bricks? Answer, it depends what we’re discussing.

Going back to the title question: “how much energy is in a litre of petrol?” Depending on what exactly you want to know about that, it’s going to be important to find out what energy went into producing it and getting it to a point where it can be used. (But then, what about the machine using it? Do we include that too?)

Talking about the work than can be done given a certain technology set brackets the problem. It may only be one way of answering it, but just because people are focusing on ‘work done’ doesn’t mean they don’t see the bigger picture.

A good quote from Smil on this that underscores the need for setting boundaries: “net energy assessments encounter their most frustrating problems in the choice of boundaries and the treatment of mental labour. Decisions about here to stop the analysis has no acceptable universal solutions. Misleading results are specially likely if the goal of the exercise is the energy based evaluation of all cascading consequences. Energy flows may have a fractal structure, and hence there may be no finite and it energy cost. No less vexing is the necessity of converting to a common denominator.” [Smil, Energy in Nature & soceity, 345]

Walter Haugen
19 Apr 6:12pm

Dan – Excellent point. Yes, you can go to multiple levels of analysis, but I have been rigorous in restricting myself to fuel. Rob’s original statement was about fuel after all.

When I compare soil prep by horse or walk-behind tiller or tractor or a human with a shovel, I concentrate on fuel use. I do the same when I compare fuel use in transport by various means and mileage. Obviously there is a lot of emergy (embodied energy – Odum’s term) in the construction of the machine, etc., but that is difficult to calculate. There is only one account I have come across so far and there were many problems.

The fuel component is SO HUGE in energy use and abuse that it dwarfs the emergy. In terms of “reduce, reuse and recycle,” reduction in fuel use gives us more bang for our buck than any other efforts we can make.

A telling analogy you might find interesting is the reduction in atmospheric greenhouse gases in 2009 over the levels in 2008. This was due to the global recession. In other words, all the good works of us environmentalists were dwarfed by the simple downturn in global economics. Reducing fuel use right now by growing food with human fuel rather than fossil fuel provides big gains immediately.

mark forskitt
19 Apr 9:35pm

I know when I do a hard day’s work in the field using only hand tools I can easily get through more than 2500 (kilo)calories and still lose weight. I think olympic athletes can get through 5000 kilo calories per day – I’m certainly not in that league!

I also start better on cold mornings than most tractors.

Don’t forget to include the transport of the fuel to the place where is it combusted, and the embedded energy in the machinery you are running for a fairer comparison.

20 Apr 12:15pm

@Walter – Sorry Walter, when talking about how many days of human labour a litre of fuel contains, I don’t see that a comparison that is *not* about work output of the machines using the fuel vs. the human without the fuel would be in any way meaningful. Since we burn fuel to do work and we cannot build a machine that is anywhere even close to 100% efficient in converting oil to work, I think it is necessary to take into consideration those energy losses. I chose the diesel truck because it is the most efficient user of oil I could find real world figures for (yes, trains are way more efficient, but we hardly use them for moving general freight any more. We especially don’t use them to move bricks…).
So, in moving my 25 tonne load of bricks, my hypothetical truck is tossing maybe 50% of the energy in the fuel straight into the environment via the exhaust, radiator, noise, etc. So if we were to look at using 100% of the energy to move those bricks, our poor little human would be labouring for *twice* as long to move the bricks, making 1 litre of fuel worth more like 100 days of labour with the wheelbarrow!
In order to make our human more efficient, let’s give him/her one of Dan’s recumbent cargo tricycles from Sadly, the builders don’t identify which of the ~17 standard definitions of “ton” they are using when they say the trike has a capacity of 1/3 ton (I love standards, there are so many to choose from). I’ll assume they mean short ton at 907kg, for convenience. Let’s round the numbers, to make the sums easier, so 1/3 ton = 300kg.
@ Dan, while I have loads of experience with wheelbarrows (I’m the garden team leader for my local permaculture group), my sole experience with cargo trikes carrying bricks is watching other people pedal them incredibly slowly around the outskirts of Beijing, so this is now totally speculative hot air – the following should be consumed with a very large grain of salt – at least one long ton thereof…
So our recumbent tricyclist will carry 300kg of bricks per load for ~5km, as per the previous example and will need to make 83 round trips to move the 25 tonnes. We’ll assume that the recumbent nature of trike doubles the efficiency compared to the guy in Beijing, making us go only as fast as the person with the wheelbarrow (3km/h), but at least they have 5x the load. On the return trip, the trike might average 15km/h, which is a slower pace than a touring bicyclist, but I can’t imagine the trike going faster than that. (hot air warning: I can barely average 20km/h on my bicycle, but I live in a very hilly city). So, the outbound trip takes 75 minutes and the return takes 15 minutes, for a total of 90 minutes a trip – ~5 trips per 8 hour day, or 16 days labour compared to 1L of fuel in the German truck, 8 days compared to 1L of fuel in the American truck.
So, Walter, if we make the human as efficient as we can and make the truck as inefficient as we can, your calculations are bang on (even though I’m throwing away far in excess of 50% of the energy in the fuel in this example).
If we make the truck as efficient as we can, then the comparison lies mid-way between your calcs and Rob’s number.
If we assume 100% efficiency in the truck (*not* a real word possibility), Rob’s numbers look right.
My gut feel is that the calcs here probably make the human more efficient than is possible in the real world. I try not to think with my guts, however, but if this were the case, Rob’s numbers would be looking pretty good.
Then again, as a number of posters have alluded to, it all depends on your assumptions.
And Walter, I withdraw unreservedly my assertion that your calculations are “not even wrong”

Erik Buitenhuis
20 Apr 3:21pm

Walter Haugen said: “A telling analogy you might find interesting is the reduction in atmospheric greenhouse gases in 2009 over the levels in 2008.”

I know this isn’t what the discussion is about, but in the interest of not spreading misinformation: this is not true. The rate of increase in atmospheric CO2 in 2009 was (slightly) lower than in 2008, but still an increase.

Walter Haugen
20 Apr 3:51pm

Les – Thanks for your calculations and attention to this issue. I could do without the snarky comments however. Moving on, if you agree we have to take energy loss into account (which you do), it seems to me the proper place to measure that is on the input end. As you say, a human is never 100% efficient. Of course I never said any such thing. What I am measuring is energy used, NOT work produced. As I have said on this forum and in many other public places, 1) it takes energy to do anything, 2) many hands make light work, and 3) if we have a more efficient machine but it takes longer to do the job, we gain over the long term because we use less energy in the aggregate.

I am sure you are astute enough to realize that moving a load of bricks by a machine is done in a shorter period of time, but with more fuel than having a load of laborers do it by hand. As my wife likes to say about nuclear power, it’s like using a cannon to ring a doorbell.

Allow me to put on my biological anthropologist hat for a moment (and yes, I am a scientist). The most efficient machine we have at our disposal is the human organism. Because of bipedalism and the energy-capturing pendulum motion of our walking, we get a lot of work output for very little energy input. The fine discrimination of our opposable thumbs allows us to make precise motions using that energy and motion. Finally, our grotesquely-enlarged brains allow us to come up with new ideas (and even new machines) on a daily basis. However, we throw all this elegance away on a daily basis by subjecting ourselves to machines that use huge amounts of fossil fuels in a less efficient manner than either a horse (better) or a human (best).

As I say often, “The laws of physics are on my side.” When you don’t have the gas to move your ton of bricks, or the gas has gone up to $20 a gallon, you will still be able to get laborers to shift the bricks at a much lower energy cost. By that time, they will probably work for food, which you will still be able to get from your neighbor who farms with a shovel and hoe.

A fuel energy value of 125 kilocalories per hour is rather mild. One other poster remarked that he burns up more calories than that at his construction job. My calculation is based on doing tasks and being efficient. Bottom line: The human machine takes far less energy to run than any internal combustion engine. Beyond that, my personal human machine is more efficient than other human machines because of the way I work.

Consider the following: just 2 weeks ago I tilled up 15,000 square feet with an internal combustion engine on a tiller at a cost of ~2 kilocalories per square foot (1 gallon of gas at 31,000 kilocalories and 3 hours of my labor – 31,375 calories). The next day two of us dug up a 600 square-foot seed bed in a hoophouse in two hours (250 kilocalories) for a ratio of .42 kilocalories per square foot. The end result in both seed beds was the same. 2 kcal/sq ft vs. .42 kcal/sq ft. My tiller is the best I can buy in this country, a BCS and it is more efficient than a tractor. Yet the human engine is over 4 times as efficient in work output per kilocalorie. It just takes longer to do the same work.

You can devise all the theoretical scenarios you want. I grow food every day, year after year. The human engine wins, hands down. I see it right in front of me every day. The human engine will still be usable after we cannot use petrol either because of price or supply disruptions.

Dan Olner
20 Apr 4:36pm

We seem to have moved on from the original question, which was ‘how much energy is there in a litre of petrol?’ The question, perhaps naturally, has drifted into ‘which is better, petrol power or human power?’ Once we’re onto this question, things get a lot more complicated. There are lots of things going on in Walter’s post, but I’ll just pick up on one point: “if we have a more efficient machine but it takes longer to do the job, we gain over the long term because we use less energy in the aggregate.” That assumes using less energy is always better than saving time. At least as far as people’s choices go, that’s certainly not true: for instance, people fly to save time, when they could drive and use less energy, but take much longer. That is ‘better’ from that person’s point of view.

It is important to think about how we value time and what the opportunity costs of time are. A recent video by Hans Rosling looking at the impact of the washing machine sums up this point nicely: measuring the worth of washing machines based only on their energy use compared to hours of hand washing misses out everything Rosling discusses: all the things gained by spending that time on other non-drudge activities, including education.

Walter, you say “the human engine will still be usable after we cannot use petrol either because of price or supply disruptions.” That’s absolutely true, but I’m not sure that makes human power better than petrol power in some objective way, or electricity in the case of the washing machine. If we can find some way to keep washing machines running, I think that would be a good thing for humanity. I think people (mostly women as it happens) would prefer not to have to spend so much time scrubbing clothes, and use that time for other things.

But then we’re getting into people’s preferences. I’d be interested to know what people think about that. Personally, I think not accounting for the impact of preferences (how we value time being one facet of this) misses a massively important part of the problem. It cannot be reduced to a simple question of energy use. Again, though, we have to be careful what question it is we’re actually addressing…

Dan Olner
20 Apr 4:38pm

Wrong link for the Hans Rosling video, sorry! Here is the proper one. Do watch it, it’s jolly good.

Walter Haugen
20 Apr 5:19pm

Dan – Great post, except that I for one have not moved on from the original question of how much energy in a litre of petrol vs the amount of energy in human labor. Since I have been a community activist for over 40 years, I obviously have a bias towards renewables, sustainability, etc. and I usually take it for granted that people interested in the Transition Movement are already on board with peak oil, climate change and the need to reduce our human footprint on the planet.

Unfortunately, human preference has landed us in our present soul-destroying, planet-destroying mess. I always argue that our desires (or preferences if you like) must be curbed and damped down. This does not make me a Puritan – its more along the Buddhist lines.

Rob Hopkins is about in the middle of the “doomerosity scale.” (I love that term!) I am quite a bit higher on that scale – simply because I have been here before.

As one farmer I used to pick apples for liked to say, “Let’s plan for the worst and hope for the best.”

Kenrick Chin
22 Apr 1:32pm

When I give my peak-oil/sustainability/transition presentations I take Rob’s lead but say it differently to make the same point.

I bring along a 5-litre portable gas container (empty, of course) as a visual aid. I claim a small car can travel about 50Km on 5L of petrol. I use the local town and another 50Km away to reinforce the picture.

Then I ask the audience, what happens if you ran out of gas and had to push your car all the way back home?

I then add, how much would you be willing to pay to have your car towed home?

Everyone can easily get the point.

John Mason
25 Apr 9:24am

I had a go at similar calculations in a post on my blog last year, which I’ll paste in:

Energy is measured in joules, and energy output in joules per second. One joule per second is one watt, so a sixty watt lightbulb outputs 60 joules/second. OK so far?

Typical energy outputs of some familiar things are as follows:

A hummingbird in hover: ~1 joule/second

A human doing manual labour: ~100 joules/second

A horse at work: ~745 joules/second. This is equivalent to one horsepower, so a 50 horsepower outboard engine at peak performance is outputting 745 x 50 = 37,250 joules/second – or the work of 372.5 men.

Now let’s look at a barrel of regular oil. 1 barrel contains 34.97 UK gallons, with an available energy output of a whopping 6,100,000,000 – that’s six billion one hundred million joules of energy.

An 8-hour man-day for a non-slacker consists of 60 x 60 x 8 = 28,800 seconds @ 100 joules/second

The total output is therefore = 28,000 seconds x 100 joules/second = 2,880,000 Joules

Let’s now compare that to our barrel of oil:

6,100,000,000 joules (barrel of oil) divided by 2,880,000 (8-hour man-day) = 2118. Yes – a 34.97 gallon barrel of crude oil contains the energy output of 2118 man-days. That’s one man working every day of the year for 5.8 years.

Current global oil demand (purple in the above graph, with orange being supply) is about 86 million barrels per day. 86 million x 2118 man-days gives us the rather ungainly figure of 182,148,000,000 man-days of oil-based energy consumed by Mankind each and every day. This daily figure is the equivalent of some poor bastard having to work for 499,035,616 years without a day off. That’s from the Cambrian Period to now. Makes you feel tired just thinking about it!


(it was a post about energy and where it is produced!)

Cheers – John

Walter Haugen
25 Apr 6:02pm

John – Very nice post. However, you are still thinking in terms of output. This is like pouring out 5 beers at the local pub, drinking 1 and pouring out 4, and THEN only paying for 1 beer. This is what we have been doing with the 20% efficient internal combustion engine for the last 100+ years. [The other 80% of the energy input is going up the smokestack and into the environment as heat and gases.]

Another point is that if your manual labor output is 100 joules per second, that is 360,000 joules in an hour or 2.88 megajoules per day (as you rightly point out). Just taking the hour figure and converting to kilocalories (1 kcal = 4184 joules), we get the figure of 86 kcal per hour. This is much less than I consume every day on a 2500 calorie a day diet (500 kcal for 8 hours sleep and 2000 during the other 16 hours). Since I work manual labor and have done so for over 50 years, I know I burn up 125 calories per hour. Also, I don’t work any harder growing food than I do walking to the mailbox, thus the aggregate of 125 calories an hour all day long.

FEASTA (Rob’s source) said 60 kcal/hr; you say 86 kcal/hr. Both of you are measuring output. I say 125 kcal/hr of input. If your manual laborer is actually getting 86 kcal/hr of output, I suggest you triple his wages, as I suspect I am only getting 25 kcal of output per hour on 125 kcal per hour of input. [I am assuming I am at least as efficient as an internal combustion engine.]

Walter Haugen
25 Apr 9:51pm

Erik – The newspaper articles last year mentioned a reduction in gases, NOT a reduction in the rate of increase. There were reductions in Britain, Holland, Australia, etc. but not in some developing countries. The two articles I read spoke of net worldwide reduction. Here is a quote from The Independent, July 3, 2010:
“Britain’s greenhouse gas emissions fell by 8.6 percent in 2009, but this was largely because of the recession and levels will likely rise as economic growth returns, a study said Wednesday.”

If you can provide me with your references, I will certainly check them out.

John Mason
26 Apr 7:26am

Walter – thanks for the comments: your points on efficiency are well-made. The ITC is a remarkably inefficient use of liquid fuels really, isn’t it!

WRT to the links between emissions and the economy, I’m sure there are strong relationships and it would be interesting to see UK trends in the past 3 months as liquid fuel prices have escalated. However, it would also be necessary to look at to what extent increases in emissions in the East have cancelled out our reductions.

Cheers – John

Irv Mills
28 Apr 4:55am

Well, I did the calculations using:
1) 75 watts as a reasonable human output,
2) 131.76 Megajoules per gallon of gasoline (petrol), which works out to 38.42 Megajoules per litre
3) 20% for typical engine efficiency, though that may be a little high.

That works out to 25.8 hours of human labour being equivalent to 1 litre of petrol. 40 litres works out to just about half a year. Which is lower than Rob’s original estimate by a factor of 8.

But I am not about to quibble about this. I think most of us have been missing the point Rob was trying to make– that fossil fuels have provided us with “energy slaves” equivalent to a whole lot of people.

So let’s do that calculation. The car I drive uses about 8 l. to go 100 km. at 80 km/hr. That’s 6.4 l. per hour. Using my number of 25.8 hours of human labour per litre, that 165 hours of effort to do what my car does in 1 hour. Or 165 people working for 1 hour.

When the fossil fuels are done (or too expensive to use) very few of us will have 165 people seeing to our every need. So if our energy needs were to be met mainly by human muscle power, we’d be using a LOT less energy.

Walter Haugen
28 Apr 3:58pm

“So if our energy needs were to be met mainly by human muscle power, we’d be using a LOT less energy.”

Irv gets it.

8 May 12:56pm

Part of the problem here is that it’s comparing apples to oranges – an amount of fuel with an amount of time. There are several ways you could make the comparison:
(a) Calculate the amount of work a given machine can produce with 1 litre petrol (and this will depend on the machine) with the amount of work a person can do. Assuming the person is working flat out, 8 hours a day, you can calculate how long a single person would be working to produce the equivalent amount of work. This is the comparison of bike generators to diesel generators.
(b) Calculate how far you could move a given load with a given vehicle and 1 liter of petrol, and calculate how long it would take a person (with any mechanical help you choose) to move the same load. This is the truckload of bricks comparison, or the “bikers pulling a SUV” comparison.
(c) Calculate the calories contained by burning a litre of petrol, feed a human the same amount of calories, and estimate how long the human could be working flat out, 8 hours a day, with that amount of calories.

There may be other possible ways of answering the question that I’ve missed. Each of these calculations will give a different number, and it will depend on the assumptions. I suggest that Rob gives up on very quick soundbites and makes a comparison that is a bit more specific. It could be anything from “You can power a projector with a diesel generator with 40 people on bikes (Barbara Patkova does this in real life, so the number is reliable) or with X amount of diesel” to “You can move a ton of bricks with 1 litre of petrol for X miles with a truck, and it takes you X minutes, or with a wheelbarrow, and it takes you Y time.”

Walter Haugen
11 May 4:06pm

“You can power a projector with a diesel generator with 40 people on bikes (Barbara Patkova does this in real life, so the number is reliable) or with X amount of diesel” to “You can move a ton of bricks with 1 litre of petrol for X miles with a truck, and it takes you X minutes, or with a wheelbarrow, and it takes you Y time.”

Doly gets it too. As we transition into a world with far less petroleum energy available, we will need to substitute human labor and time for oil energy. We already have plenty of human labor and more time at our disposal than we realize.

Dan Olner
13 Jun 4:17pm

Just to open this one up again – back to preferences. A quote from Button’s transport textbook:

“Savings in walking and waiting times are valued at between two and three times savings in on-vehicle time – parameters that have proved to be remarkably robust over the years.” [p.104]

The upshot: people willing to commute for an hour each way would be much less likely to walk that distance. One can get into lengthy arguments about the methods used to make these findings (we can if anyone wants to!) but the point stands: once you’ve got past working out some abstract way of comparing energy use between people and fossil fuels, you then need to understand that we use and value our own-steam energy very differently. If we don’t get to grips with that problem, our solutions won’t work.

Perhaps it’s just a roundabout way of saying something the transition movement does very well: thinking in-depth about how to build human-scale societies. But I find myself often wanting to defend the remarkably simple and powerful use of utility as a way of helping do that – despite the fact that (is this true?) most people in the transition movement see such ideas as part of an economic outlook that got us into this mess in the first place.


Walter Haugen
13 Jun 4:31pm

Dan – Interesting comment that cuts right to a very important aspect – preference. When I was in my 20’s, I hated riding in cars and preferred to walk or ride my bike. I also dabbled in bike racing and did quite a bit of cycle-touring around the world. So I have a subjective preference for foot and bicycle transport.
Now that I am a farmer/researcher in my 60’s and work 16 hours a day, I don’t have much time for riding my bike. I only get to go to the pub one night a week too, which is another activity I enjoy. In other words the utility factor now takes precedence over my preferences.
The age factor is part and parcel of my approach and I am sure it looms large for nearly everyone. The young Transition types have a far different perspective than us older survivors of the 1960’s.
So, along with your point about preference vs. utilty, one other variable is the age factor.
I suspect that a tipping point will be reached about August, when the worldwide winter wheat crop comes in below expectation. At that point, I will still have a low-energy model that produces ample food.