Battery used Battery charging

Bike Powered Electricity Generators are Not Sustainable

The present approach to pedal power results in highly inefficient machines.

Image: K-TOR Power box.
Image: K-TOR Power box.
View original image View dithered image

Pedalling a modern stationary bicycle to produce electricity might be a great work-out, but in many cases, it is not sustainable. While humans are rather inefficient engines converting food into work, this is not the problem we want to address here; people have to move in order to stay healthy, so we might as well use that energy to operate machinery. The trouble is that the present approach to pedal power results in highly inefficient machines.

When operating a bicycle generator you are basically pedalling to produce the energy required to manufacture the battery.

There are two ways to power a device by pedalling. You can power it directly through a mechanical connection - as was the case with all pedal powered machines for sale at the turn of the 20th century. Or, you can pedal to generate electricity, which is then used to power the device. In the 1970s, most research was aimed at direct mechanical power transmission. Today, the interest in pedal powered machines is almost exclusively aimed at generating electricity, for instance for charging cell phones and laptops - products that did not even exist in the 1970s.

With one exception (the ‘Fender Blender’, a pedalled powered machine to make smoothies), the only pedal powered machinery that is now commercially available in the western world (offered by Windstream, Convergence Tech and Magnificent Revolution) are stands to fit your bike to, connected to an electric motor/generator and a battery. That is a combination that can quickly convert your regular road bicycle into an electricity generator. These are also the pedal powered machines which are used for educational and arts projects, like powering a music concert, a cinema projection or a supercomputer, or teaching kids the difference in energy use between, for instance, an incandescent light bulb and an energy saving lamp.

In an effort to raise awareness about energy use and global warming, the BBC even made a TV-programme in which an entire household was powered via these generators, with 80 cyclists generating up to 14 kW. These multi-person pedal power generators were pioneered in the 1970s by the Campus Center for Appropriate Technology (CCAT).

Generating electricity is very inefficient

There are several problems with the present-day approach to pedal power. First of all, it is important to know that generating electricity is far from the most efficient way to apply pedal power, due to the internal energy losses in the battery, the battery management system, other electronic parts, and the motor/generator.

These energy losses add up quickly: 10 to 35 percent in the battery, 10 to 20 percent in the motor/generator and 5 to 15 percent in the converter (which converts direct current to alternate current). (Sources: 1/2/3). The energy loss in the voltage regulator (or DC to DC converter, which prevents you from blowing up the battery) is about 25 percent (sources: 1/2).

This means that the total energy loss in a pedal powered generator will be 42 to 67.5 percent (calculation example for highest loss: 100 watt input = 80 watt after 20% loss in motor/generator = 57.5 watts after 25% energy loss in voltage regulator = 37.5 watts after 35% loss in battery = 32.5 watts after 15% loss in converter = 32.5 watts output = efficiency of 32.5% or energy loss of 67.5%).

You have to pedal 2 to 3 times as hard or as long if you choose to power a device via electricity compared to powering the same device mechanically

Furthermore, there will be an additional slight loss as the battery stands idle, and the charge efficiency (also known as “charge acceptance” or “coulombic efficiency”) of the battery will deteriorate over time. And to make the calculation complete, you should actually also include the energy loss in the electrical device that you are powering (we won’t do that here).

An energy loss of 42 to 67.5 percent of naturally means that it takes 42 to 67.5 percent more effort or time to power a device (say, a blender) via electricity compared to powering the same device mechanically. This can be considered an acceptable loss if you are using solar panels or a wind turbine connected to a battery as an energy source, but it becomes rather problematic when you have to deliver the energy yourself.

If you produce 100 watts of power and 42 to 67.5 percent is lost in the conversion, there is only 32.5 to 58 watts left to power the device. If you power the same device mechanically, you deliver 100 watts straight to it. You thus have to pedal 2 to 3 times as hard or as long if you choose to take the intermediate step of generating electricity and storing it in a battery.

Traditional bicycles were not made to generate stationary power

It does not stop here. The second problem with the present approach to pedal power is that it uses a traditional bicycle on a training stand instead of a pedal powered machine built from scratch - as was the case at the end of the 19th century. Of course, using a traditional bicycle has its advantages, but again it should be realized that this approach is considerably less efficient.

One reason is the use of a so-called friction drive - the rear bicycle wheel acts upon the small roller of the motor/generator. While chain and belt drives (used in late 19th century pedal powered machines) have an efficiency of up to 98 percent, a friction drive is only 80 to 90 percent efficient (and wears much faster). This energy loss should be added to the 42 to 67.5 percent efficiency loss calculated above, which rises to 48 to 73.5 percent. Low tyre pressure will further decrease efficiency.

It should be noted that there is also energy loss in the bicycle itself: your pedals are not attached to the rear wheel itself. You turn a sprocket, which turns a chain, which turns a sprocket, which turns the rear wheel. So, on top of the efficiency loss of the friction drive should be added the efficiency loss of a chain drive (plus the energy loss in the derailleur, if your bike has one).

Additional energy losses occur when using a racing bike or a mountain bike

Connecting a bike chain directly to the generator would prevent the energy loss of the friction drive, but it implies that you have to adapt the bicycle - destroying the whole concept of today’s commercially available pedal generators.

Racing bicycles

Additional energy losses can occur when using a road bicycle to generate electricity. For example, the picture accompanying the Windstream generator shows a racing bicycle. This is a very bad choice, because the position of a rider on a racing bike is aimed to reduce wind resistance. Tests on ergometers (stationary bikes used to measure the power output of cyclists) have shown that pedalling in such a position is only about 80 percent as effective compared to a normal upright position, again resulting in considerable energy loss.

On the road the rider position on a racing bicycle is beneficial because of the large importance of air resistance. However, on a stationary pedalling machine this position has no advantage whatsoever. The popular mountain bike is equally disadvantageous because of the corrugated tyres, which of course lower the efficiency of the friction drive. In short, while using a road bicycle to generate electricity has the advantage that you can use your own bike, this does not mean you can use just any bike.


Another important drawback of using a common road bicycle is the absence of a flywheel - a heavy disc made of concrete, wood or steel that continues to generate power after it has been put in motion. In a pedal powered machine built from scratch, like the ones used at the turn of the 20th century, the flywheel applies the function of the rear bicycle wheel in the training stand (although the flywheel is mostly placed at the front of the machine). The pedaller powers the flywheel, and the flywheel powers the machine (which can be a mechanical device or a motor/generator to produce electricity).

Why is a flywheel advantageous? Because there is an important difference between riding a bicycle on the road and pedalling a stationary machine. If we are pedalling, the power exerted by our feet on the pedals is inconsistent. It peaks every 180 degrees of crank rotation, and because the two cranks are placed 180 degrees out of phase this results in two power peaks per turn of the crank. Similarly, there are dead spots in between at the top and bottom position of the pedals (to be correct this minimum torque is not zero but about one third of the maximum).

On a stationary bicycle without a flywheel, the natural pedalling rhythm results in jerky motion, limiting the energy output of the rider

On a bicycle, this uneven exertion has little effect because of the inertia of both bike and rider. But on a stationary pedal powered machine, this natural pedalling rhythm results in jerky motion and additional stress on parts.

Because of its large mass and rotational speed, the flywheel evens out the difference between power peaks and dead spots. Evening out the power input means that the rider tires less quickly and can thus generate more power. The obvious disadvantage of a flywheel is that it is heavy - from 10 to 80 kg for stationary pedal powered machines - and thus not exactly mobile.

Generating electricity is not eco-friendly

Generating electricity is not only ineffiicient, it also makes pedal power less sustainable, less robust and more costly. To begin with, batteries have to be manufactured, and they have to be replaced regularly. This requires energy, which can completely negate the ecological advantage of pedal power.

According to this research paper (pdf), the embodied energy of a 150Wh lead-acid battery (like the one offered with the Windstream pedal power generator) is at least 37,500 Wh, which equals 250 full charges of the battery (more sources: 1/2). In other words: if you can deliver 75 watts of power to the battery, you have to pedal for 500 hours in order to generate the energy that was needed to manufacture the battery. Because the life expectancy of a lead-acid battery can be as low as 300 discharge/charge cycles (sources: 1/2), you are basically pedalling to produce the energy required to manufacture the battery. If you also factor in the embodied energy of other electronics and parts, the ecological advantage of a pedal powered generator connected to a battery becomes rather doubtful. It might costs more energy than it delivers.

A pedal powered generator might cost more energy than it delivers

Of course, it also takes energy to manufacture a pedal powered machine that does not take the intermediate step of generating electricity. This concern lies mainly with the production of steel, and quite a lot of it. The commercially available Fender Blender mentioned earlier weighs 25kg (55 pounds).

If made from recycled steel, and using these figures to calculate the embodied energy of steel, this comes down to an energy cost of at least 41,625 Wh, slightly more than the battery needed for the electricity generator. If freshly made steel is used, the embodied energy is at least 138,750 Wh (3.7 times the embodied energy of a single battery). However, these machines can last at least 100 years (pedal powered machines surviving from the late 19th century are still in use), while the battery of the electricity generator has to be replaced every few years.

If we ignore the embodied energy of other parts than the battery (both training stand and electronics), and take a life expectancy of 4 years for the battery (rather optimistic), a pedalled powered generator would require an embodied energy of 937,500 Wh over the course of 100 years - 6.7 to 22.5 more than a mechanical unit. Moreover, it is easy to make the frame for a mechanical pedal powered machine from scavenged materials, bringing the embodied energy down to almost zero, while this is an impossibility for the batteries. Never mind that in addition, the toxicity of the materials is another thing to consider.

Generating electricity is less robust and more expensive

While a pedal powered machine is the most robust and resilient energy source around if you power devices mechanically, this advantage is lost when you start generating electricity. Few people can manufacture batteries themselves, so you remain dependent on a regular supply of replacement batteries.

Furthermore, the electronic parts of the machine (voltage regulator, motor/generator, converter) can break down and are not easy to make or repair yourself either - contrary to old-fashioned pedal powered machines, which can be fixed yourself with readily available materials. Mechanical pedal powered machines are generally even easier to repair and maintain than bicycles.

The extra components also make pedal generators more expensive. The commercially available models sell for $700 to more than $1000, not including the necessary replacements of the battery over time. Even if you make your own pedal power generator, the costs add up. The 2008 book ‘The Human-Powered Home: Choosing Muscles Over Motors’, which has plans for several kinds of pedal powered machines, estimates the costs of a DIY generator at about $50 (using scavenged parts) to $350 (using new parts), not including a bicycle stand and replacement batteries. Another source estimates the cost at $600.

The mechanical pedal powered machines in the book can be built for $10 to $50 (the washing machine being more expensive at $100), everything included. While the only commercially available mechanical pedal powered machine today is very expensive too (the Fender Blender sells for $1,700), the high cost is almost entirely due to the steel frame - which, as mentioned, could easily be replaced by the frame of an old exercise bike, or built oneself from scavenged materials. Moreover, there are no additional costs for replacement batteries and the machine is built to last for a very long time.


One way to solve the large energy losses of pedal power generators is not to produce electricity at all and power devices mechanically, whenever possible. Another way - the only way for devices that cannot be powered via a direct mechanical connection because they do not rely on rotary motion - is to make the generation of electricity more efficient.

This can be done by building a pedal powered generator from scratch instead of using a road bicycle, and/or by ditching one or several electronic components in the power transmission chain. Read more:

Our bike generator connected to a 12V DC Air compressor. Image: Marie Verdeil.
Our bike generator connected to a 12V DC Air compressor. Image: Marie Verdeil.
View original image View dithered image

Sources (in order of importance)

Pedal Power in Work, Leisure and Transportation”, edited by James McCullagh, Rodale Press, 1977. Still the best resource on pedal powered machines.

The Human-Powered Home: Choosing Muscles Over Motors, Tamara Dean, New Society Publishers, 2008. Very good book on human powered machines, both hand and foot powered. Includes half a dozen plans to convert bicycles into stationary pedal powered machines.

Bicycling Science, Third Edition, David Gordon Wilson, 2004

The Dynapod: a pedal power unit” (pdf), Alex Weir, 1980. More here.

The use of pedal power for agriculture and transport in developing countries” (pdf), David Weightman, Lanchester Polytechnic, 1976

Design of a human-powered utility vehicle for developing communities”, Timothy J. Cyders, 2008

Supplement, Energy for rural development”, National Research Council, 1981

Tales from the Blue Ox, Dan Brett, 2003

Bicycles and tricycles”, Archibald Sharp, 1896

In search of the massless flywheel” (pdf), John S. Allen, Human Power (Fall/Winter 1991-1992)

Design and development of a human-powered machine for the manufacture of lime-flyash-sand bricks”, J.P.Modak & S.D.Moghe, Human Power (Spring 1998)

Human Powered Flywheel Motor: concept, design, dynamics and applications”, J.P.Modak, 2007

Modern mechanism: exhibiting the latest progress in machines, motors, and the transmission of power”, Benjamin Park, 1892

Make electricity while you exercise”, Mother Earth News, 2008

Luther’s tool grinders” (pdf, 5.8 MB), hand and foot powered grinders catalog. Hosted at Toolemera Blog.

Woodworkers’ tools and machines” (pdf, 29 MB), product catalogue no.25, 1884, Richard Melhuish Ltd., Tool and Machine Merchants, London. Hosted at Toolemera Blog.

Science & civilisation in China, Vol.5, Part 9”, Joseph Needham, 1987


To make a comment, please send an e-mail to solar (at) lowtechmagazine (dot) com. Your e-mail address is not used for other purposes, and will be deleted after the comment is published. If you don’t want your real name to be published, sign the e-mail with the name you want to appear.

Mr. S.

As for energy storage, wouldn’t making the pedal-powered machine actuate a vertical water-pump be ideal? No electronics, or chemical battery, and when power is needed water can be released from a water tower to generate current with some kind of turbine. Better yet, a windmill could be the primary power for the pump, with pedal power as a fallback for long windless periods.

@ Clifton

It’s a critical article, not a negative article. It points to several other articles on this blog to solve the issues raised.

Plus: a SHTF situation is not what this article is talking about.

charley hardman

i approach the efficiency topic from the other side. i loathe “working out”, lifting weights only to lower them. would like that converted into useful power along with the physical benefit. any true analysis must include wasteful expenditure of the subject that might be converted. caloric analysis must at least also subtract normal spending for mere existence (e.g., sitting in a chair reading book).

i suspect the most practical local solution is hydraulic power transmission (i.e., high pressure, little movement) to a dense mass (e.g., lead). don’t think water will cut it as per-home kinetic energy storage medium.


I had plans drawn out for a while now around the concept Mr. S explains. Many well designed off grid homes use elevated water so that they still have running water available, but they rely on using a generator once in a while to push the water. I always thought a small windmill could do the work of lifting the water into a small tank, and when electricity is desired, then this essentially turns into a microhydro system as water empties from the tank while spinning a series of turbines as it falls.

Kris, it’s nearly pointless to question the energy required to build this type of system. It’s obviously considerably less energy both upfront and continuing versus being grid tied (relying on miles of copper) and coal/nuclear/gas providing that power.

Colin Tonks

Hi Kris, this is an interesting piece of work and well put together. I especially enjoyed reading the history of the pedal powered machines.

Furthermore, I absolutely agree with you that humans generating energy, with a bicycle, is completely inefficient. In fact, the cost of the food required to pedal would be 150 times more expensive than the cost of buying the energy direct from a local supplier. So what is the point and why do we do this?

Well, there is one important point that I would like to make and that is the indirect benefits of this approach when applied in an educational environment. We’ve done hundreds of events using the technique of people generating their own electrical energy and then using it to do something fun. It does not matter if it’s a festival, school, corporate etc., we get the same reaction every time .. a new found appreciation for energy.

For many people, pedalling to boil a kettle of water or even lighting a light bulb is ‘enlightening’, it’s that relationship between what they are feeling in their body, the burning their muscles, the excitement of being in control to bring alive an electrical appliance that is stimulating and thought provoking.

I’m not saying that people who generate their our energy go away and become eco warrior and start wear heap shoes, it’s more of a subtle push in the right direction, a small step change in behaviour that would hopefully lead to a more sustainable existence. Thanks Colin

Chris Vernon

I’ve tinkered with bicycle electricity generation, thanks for this fascinating article.

Regarding the 10 to 35 percent lose in the battery and 5 to 15 percent loss in the DC/AC converter, for many applications an ultra-capacitor can be used to maintain/smooth the voltage and of course we should only be using DC loads! No point it going anywhere near AC. These two points improve the efficiency quite a bit, but fundamentally, I agree with your conclusions we should be powering devices mechanically.


A brilliant article, as they always are, but…

You say “An energy loss of 42 to 67.5 percent of naturally means that it takes 42 to 67.5 percent more effort”

Wrong maths! Say you had a 90% loss, you’d need to provide 10 times as much energy to get the same result. Or 900% more.

So a 67.5% loss means 100/(100-67.5)*100 = 3.07 times as much input energy is required, or 307% as the original, or 207% extra input energy.


I’ve been playing with the idea of using a bike and alternator to power an electric water heater. The electronics aren’t necessary then and alternators can come from junkyards for $35 or so. An RV 12-volt heating element costs less than $100. Any old water heater should work.

Of course, I don’t have a working prototype running yet. But I’m fairly close and the process hasn’t been particularly difficult so far.

Michael Dawson

Has anybody applied this kind of analysis to the so-called electric car? Where could one find that?

jason crowther

in order to re-gain the strength i lost due to head-injury, i spent much time on a rowing egometer. the appartus included a monitor measuring various outputs during a workout. eventually, i reached 46,400 watts/41 minutes. of course, these measurerments increased as my strength and endurance increased. i remember thinking that–maybe–i was onto something! could i use this system to increase my income through selling the power back to a local provider?


The assertion that “An energy loss of 42 to 67.5 percent of naturally means that it takes 42 to 67.5 percent more effort or time to power a device (say, a blender) via electricity compared to powering the same device mechanically” has a math error. The numbers to use are 1/(1-.42) at the low end and 1/(1-.675) at the high end. So you’d actually need 1.724 to 3.077 times more energy to compensate for a lost off between 42 to 67.5 percent.


To the author: I am a cyclist and I need to train indoors occasionally. I am thinking of finding a used DC generator and AC converter to run with my road bike. I understand that there is quite a bit of energy loss, if I try to store it. But could I just plug the AC current into my wall socket to slow or run my meter backward? I need to do the work out anyway, might as well do something with the energy.

Your response would be appreciated.


Michael Boswell

The advantage of simple pedal power is that almost all humans have legs and can generate power small machines. The best Australian example is the pedal wireless. This simple invention provided communications to regions of Australian were electricity supplies were non-existant. Which prior to World War 2, was most of the Australian land mass. Naturally, this two way radio service was public but no one ever listen to another conversation. Well, not much!

The other useful thing for human power machines is for things like sowing machines. I have heard many old ladies bemaon the disappearance of the old ‘pedal sowing machines". One control was the amount of power that you put into the machine. Electric ones have other means to vary power but one cannot feel the power going into it.

This is my first time here and I am subscribing.


I am interested in the use of pedal power while people are driving to generate electricity; though ineffecient, if a person is idle and not able to use time more efficiently any other way, this is a use of energy that otherwise cannot be used otherwise. Use of pedal power to charge batteries for use of power either in vehicle or for other purposes - and get needed exercise for our sedentary culture.


I’ve been intrigued by this “people power” concept for a few years. I don’t have any technical knowledge of how this works. It seems to me there should be some way to use human activity to generate power or to power devices in some way. There are some indirect benefits in the form of more exercise, particularly for combating obesity including childhood obesity. If there’s a people-powered device in every home, perhaps families could generate enough electricity to watch TV or charge their cellphones, etc. Parents could make sure their kids participate in generating enough power for the family to watch a program or play a video game. It could be a trade off: Sure you can play that video game, but you need to do some pedaling to make it happen! That kind of thing. It may not be efficient energy production but it has side benefits. Another application would be for emergencies, say when a hurricane knocks out power to your house. A bicycle-powered device could help pump out excess water from the basement, sound an alarm, provide lighting, or charge cellphones for emergency phone calls, etc. All things considered, it would seem that this concept has some potential, especially if you consider that millions of people drive to a fitness center (using energy to manufacture and power the car) and then turn on various electricity-driven exercise devices (like treadmills). Why not encourage energy production and/or create practical devices that are driven by human exercise in homes or at green fitness centers? There may be efficiency breakthroughs the more scientists and engineers explore how best to generate and harness energy produced by “people power.”


If you make an electricity generating device (I built a rowing machine) from scrap materials (- I got a usable battery from the scrapyard also) then the energy that has been put into making those parts is history you can’t change that but by using those parts to make electricity at least you are recouping some of it and balancing things in a positive direction overall

jane hansen

I am looking for the best way to use “human power” to run a sump pump in the event of a prolonged power outage. I had thought to use a bicycle to charge the back up battery…but may be there is a better way. Got ideas anyone???

Jane H.

Barrington, IL

Brian Hansen

Jane (#21): I think one message from this article is that using direct rotational energy is hard to beat for efficiency. From an engineering perspective this may be a principle in search of an application. When I think of a sump pump, I think of it as being asynchronous, running only when needed, for as long as needed. That scenario would suggest stored energy to run the pump.

It is possible, though, that it could be routinized, or perhaps scheduled, such that an individual could “ride” for 10 minutes every 6 hours, for instance. Or “ride” as long as needed when an alarm went off.

A direct mechanical linkage would present its own problems, of course. The bike has to be next to the pump.

Ssempa Nelson

The author is forgetting that by using a pedal generator, many liters of paraffin is not being used, reducing on carbon emission and also preventing the inhaling of the carbon by the people who use paraffin candles. Also measure the impact of using hydro power and the power generated by pedaling. If this is introduced in our developing countries many young children who read using candles will have a chance to access safe lights.


Interesting article. I’d be interested in a stationary pedal generator. My purpose would be to power radio equipment in the field. If I was to run it without a battery, I would need a flywheel to handle the transmitter’s peak power (100W PEP, or 250+W on peaks) for SSB transmission. This would be an ugly load to handle directly, so using a flywheel would make it easier to pedal. Would still need DC voltage regulation, but modern switchmode regulators can exceed 90% efficiency.

A more likely scenario is to augment my existing solar system, which provides 100W into a battery. Pedal power could be used when sustained power levels near or exceeding 100W are required, or at night/under cloud to keep the battery topped up. It’ also a way to maintain aerobic fitness while doing other things I enjoy. I am also a regular cyclist in the conventional sense (i.e. transport).

Obviously for my application, using direct mechanical transmission is not an option! :)

Steven Miller

The author brings up many good points all of which should be considered when designing pedal powered devices. I have tinkered a great deal over the years with human powered devices. Most of the inefiencies mentioned in the article can be reduced greatly by various means.

Obviously, a gallon of gasoline, ten pounds of coal or a small stack of firewood contains far more energy than a person could produce in days. The United States currently has 1.5 million pounds of proved accessible coal reserves for every person currently living in the country, billions of gallons of proved accessible oil reserves, and multiple trillions of cubic feet of proved accessible natural gas reserves. So for day to day energy needs there is only the slightest possibility that we are going to rely on humans for large scale power generation during any of our lifetimes.

However for emergency situations or inaccessible locations many modern electronic devices are efficient enough that several at a time can easily powered by one person with pedals or even hand cranks. I have several generations of hand cranked generators from early telephone magnetos to generators designed to power radios in WWII and up to Chinese military generators built in just the last decade. I also have built a permanent magnet based pedal powered generator that is extremely efficient that uses nothing but chain, sprockets an internal gear motor and a flywheel to spin the generator at the most efficient speed for power generation. It is mounted on an upright exercise bicycle. This can be used for producing power of up to approximately 180 watts at 12v DC continuously depending on the fitness level of the person at the pedals. This can directly power small electric motors with very little loss or to charge batteries. Unfortunately, I doubt whether there would be enough demand for this device at this time to justify production.

The author brings up many of the challenges that face designers of human powered devices, but his conclusions tend to lead one to believe that they will never be useful. To me this seems to be a short sighted argument that can be shown to be invalid by just one valid application. Human powered generation devices have been used in communications equipment for over 100 years. Just because the author does not know of a good use for such a device in his life does not mean that others do not have useful applications.

tom foxe

I don’t feel that this is a balanced article - it might just as well be headed “Don’t use a bicycle generator”. It assumes that a battery is needed, and seems to disregard the fact that electricity from the grid is even more inefficient, with long transmission lines dissipating most of the energy as heat.

I run a pedal-powered sound system using a “Super-capacitor” rather than a battery, and people LIKE IT ! It doesn’t produce music at a level which damages the ear, like pretty much every rock gig and festival does, and I don’t use an inverter because all my equipment runs on 12 volts DC, so it is relatively efficient. It can’t compare with mains power in terms of energy delivered, but it doesn’t need to - it gives me power as both electricity and freedom from the fat cats that own utility companies just to make money.


Very interested in this article. However, calling human power inefficient is puzzling to anyone at the gym watching the meters. I was lamenting how efficient we are. It’s a real challenge to burn off that $0.70 Snickers bar on the stationary bike. Also, I think it is an uneven comparison from generating instant use mechanical energy to battery stored electrical energy. Batteries have nothing to do with generating electricity, only storing it.

Sho Me

This is an interesting assessment of power loss in human powered electric generation, but the fundamental assumption that bike powered electricity generators are not sustainable is flawed. The flaw derives from a missing definition for “sustaintable.” None of our lives are sustainable. We will all be gone within most likely, at most 12 decades after our birth. Life on earth is most likely doomed to extinction sometime within the next billion years as the sun increases light output, because the sun itself is not a sustainable system.

Faults cited in the measure of undefined sustainability include the fact that it “takes energy to manufacture a pedal powered machine that does not take the intermediate step of generating electricity. This concern lies mainly with the production of steel.”

Again, nothing in modern human culture is sustainable by that measure. Metals are part of every “sustainable” system proposed by modern green culture. Even those sustainable hemp grocery bags are woven on automated metal machines from hemp harvested with energy intensive metal tractors.

Even if were were to prefer a culture based on products made only from annual plants, the culture to preserve tenure to that land is not sustainable until all peoples anywhere in the world - and maybe even the animals - agree to a unified system of land holding. That is sustainable only in one place - in your dreams. And your dreams are not sustainable because (I borrowed this line from Zimmy) you’ve got some big dreams baby but in order to dream you’ve got to still be asleep. When you gonna wake up?

That said, the article is an interesting study of an interesting idea. Two places pedal powered electricity might be practical - in terms of modern culture - would be in exercise machinery and in low-powered living, either long term or in temporary settings. For exercise bikes, I would as soon generate electricity for my laptop - which generally uses less than 60 watts, which I can generate from the approximately 100 watts I produce sitting on a trainer.

Likewise, in a low-power living arrangement - maybe off the grid no matter what my motivation (sustainable?, aesthetic?, lux camping?, socio-political isolation?) I can use a bicycle generator to charge electronic devices, supplementing other power sources including solar or wind.

Overall, I agree the limitations of pedal powered electricity make it impractical. Friends have tried various setups in their off-the-grid homes and quickly given up. But the advent of new, affordable portable high-density batteries and the proliferation of portable electronic devices - including cameras, tablets and especially low-power LED lighting - have brought new possibilities for pedal powered electric generation


What about pneumatics? How about using some pedal powered device to compress air in an air tank, and then use use the compressed air to drive tools or a motor.


First, thanks for the reminder on flywheels. Having been ignored for obvious reasons by all and sundry in the UK I recently decided to use a bicycle generation arrangement to demonstrate what I found back in 2009. The article is spot-on in explaining the inefficiencies but that is the area in which I’ve been working. The term “counter torque” is not mentioned, but most of the angst of human-powered devices could be directed at that one term. Many will know this, but the increase in applied force required as the electrical load increases is due to the magnetic effect that the generator stator has on the rotor. I’ve found a way of duplicating that effect in reverse which, of course, changes the playing field entirely. Regards, @withthechange.


I think the term “sustainability” should be more rigorously bond to scientific connotations. This article does not make sense in this regard: of course takes energy to build any sort of things!

By the same token many anti-environmentalists (eventually backed by oil companies) declare solar power not sustainable. Sure it takes energy to produce solar panels or to build a human powered generator, but then they work out of free solar energy or human sweating!

Stating human powered generators are not sustainable is unbelievably wrong! The author maybe prefers nuclear, burning down Amazonia forest or oil burning? Really?

On the engineering side, I am happy to announce that the above mentioned efficiency losses have been reduced by a great amount in the last century, i.e. a DC-DC converter with above 95% efficiency fits on a finger tip and costs few dollars!

Besides, modern cars and in general internal combustion engines’ efficiency is far less than 50%. So what? Let’s dump it and get back riding horses?

Power losses in state-of-the-art turbines for energy production are comparable with the numbers stated in the article, plus they burn oil or gas, producing combustion gases which are not the best thing ever in terms of sustainability. Also these plants use tons of fresh water in the cycle.

I guess for comparable efficiency, at the very least the wastes of a cyclist are by far more eco friendly and sustainable than the ones from production plants!

In conclusion, the article goes quite deeply into explaining technical issues related to human power generation systems but on the minus side, the conclusions are not realistic as they rely on not updated data, math is flawed somewhere and the concept of sustainability here stated is remarkably misunderstood.

Furthermore, before declaring pedal energy not sustainable, the author should have at least mentioned and compared it to actual production systems in order to provide a thorough analysis.


Quick release hubs make changing tires easy and quick so get a second rear wheel with the smoothest cruiser tire you can find and fill it with brine. You could also increase the flywheel mass by putting washers around the spokes and gluing them to the rim.


This article looks at this idea the wrong way. As mentioned by someone else, many people use exercise equipment everyday to keep fit so why not capture that energy rather than just let it go to waste.

A gym full of this type of equipment that could discount your membership depending on how much energy you generate would be awesome and give good incentive.

Inefficiencies do not compare to years of people generating electricity doing what they already do multiple times a week and significantly reducing energy consumption from the fossil fuel powered grid for that gym/business. Sounds like it should be a local/government council initiative.

Will G.

I think that there is no clearcut good idea/bad idea on this topic. Like so many things in life it has to do with what your use is. I use the Pedal-A-Watt brand stand because I work out alot. It allows me to charge all my mobile devices and have power left over to run my laptop for hours on end. So, for me it works. Is it cost efficient compared to grid poweer? No. Is it awesome to have some electricity when the power goes out? Yes!


Why does everyone assume you would be converting DC to AC, regulate it and convert back to AC? That would be highly inefficient. Same with solar panels… I take my power straight from the DC Side.

Modern switch mode power supplies are self regulated, a higher voltage results in a lower current, which in turn would ease up the strain on the person driving the bike. I personally find that just by pedaling fast my torque capability goes down quite a lot and the inverse is true, I can keep a good torque at low speed. This is because there is less energy wasted reversing the pedaling direction (up and down, the full weight of the leg/foot, etc).

With that said, a regulator could be used directly to charge a cell phone, laptop or anything with a built in battery with a single stage regulator. A fan would also be useful while working out.

Lastly the flywheel… If done properly the electric generator can work as a flywheel as more energy is taken as the speed increases. It could be set to only generate during peak speed to smooth out the rotation pretty much in the way a flywheel does.

@ tom foxe

The use of 12V DC improves the sustainability of bike powered generators a lot, see this newer article (related to solar energy, but just as useful for bike power):


I came across something called the Gravity Light the other day (literally a light that is powered by a descending weight that spins a generator) and it made me think of the very first comment for this article posted by Mr. S. Could this concept be applied instead of using water? I can only imagine that the weight required to produce enough power to light more than one bulb would be sizeable, possibly in the order of tonnes and of course construction of the contraption that would hold/lower the weight could make the whole exercise pointless… but would it work? I mean you might have to empty the container at the bottom, lift it to the top and fill it up again but maybe it avoids the requirement of a massive reservoir as mentioned in a subsequent post.

Tying in the bicycle end of things, I’m imagining using a bicycle to either lift the weight or power a conveyor belt that fills an empty container while its at the top. -shrug-

Side note: the light is part of an initiative to help people living in poverty reduce their dependence on kerosene for lighting their homes which seems like an alright idea. And no, I have no affiliation with the project!


This is a very good essay on the mechanics of pedal-power, and the cost of the equipment, and a good analysis of economy of this in comparison to the cost of electricity produced at utility-scale. But misleading because the greater consideration is the cost of the labor compared to the cost of labor for utilizing other off-grid energy resources. Pedal-power can still be transforming for hundreds of millions of people, because energy self-reliance at the small-community scale can be transforming for communities, and for humanity and life on Earth if applied universally. The key is to understand that the cost of fuel is very high off-grid for hundreds of millions of poor persons, especially for lighting and cooking, and freeing any significant part of the labor/price involved in acquiring that energy is a great boon that multiplies exponentially if that labor and/or money saved can be kept in the community to establish a self-reliant local energy marketplace as the foundation for a genuine local marketplace, which is one in which local people develop local resources for local consumption. The task of establishing this kind of virtuous cycle of supply and consumption is crucial to achieving the self-reliance that is the task before the world now, thus to empower and impel people to care for each other and the ground beneath our feet. Another very great advantage of pedal-power is that it permits employing the unemployed including the poorest of the poor and permits these people to afford to access the energy produced. Marketplaces that begin on that foundation leave no one behind. I’ve done the math on the cost of food for the calories for pedallers to produce electricity for lights and other small devices, which reveals a great gain when compared to the labor needed to earn money for kerosene. Today, small solar powered devices are imported, but this exports large amounts of money and undercuts the possibility of establishing local energy marketplaces. The bottom line is how much money can be kept in the community to fund the development of all human and natural resources with that development enhanced by a renewable-energy foundation for the local marketplace. The ultimate solution in this regard is reconomy


Pedal 8 hrs a day for 25 years or work a job 8 hrs a day for 2-4 weeks and then buy a 1KW solar panel and it generates for 25 years. Pedal a bike to heat water? Heat water on a wood stove that is heating your house in the winter and use a simple solar heater to heat in the summer, fall and spring. Decrease need for electricity by making appliances run direct off pedal (Much like the direct mechanical that you mentioned.

Solar ovens can cook and heat a lot of things several months of the year. If you have something to burn, look into rocket stove to cook with and heat water in times without sunshine. Propane backup for heat, cooking and running a generator for lazy or extreme emergencies. Underground cellar room can provide a lot of things including a place that is cool in summer and moderate in winter.

If you really want to make a pedal electric system, consider a very large flywheel in the neighborhood of 10 feet diameter. Then have a smaller wheel that has several alternators directly on the small wheel.


I can see the argument that using human power to generate electricity is a waste of time/effort/energy. On the flip side, as many have pointed out in the comments, there are situations or personal circumstances where using a pedal generator is a good thing. If I’m going to be pedaling my exercise bike anyway, why just have the resistance generate useless heat when instead I can use it to charge my cell phone and/or power a 32 inch TV? I purchased a new cell phone last December and have yet to charge it with a wall outlet - relying on my pedal generator and portable battery banks that I have charged with it if I travel for any period of time. My design is simple to build and does not rely on a 12v battery for storage. A stationary bike, e-bike motor, bridge rectifier and a charge controller and you’re set to charge most mobile devices - see for details and videos.

{ANS} a39

tom foxe, transmission line losses are only like 4.7% to 6%. That’s not hardy most of the energy. LOL!

Han-Lin Yong

Heating would be more efficient than producing electricity. Muscles are about 20 to 25% efficient with the rest body heat. That means for every 100W from pedalling, you produce 400 to 500W in total. Three people would make it comparable to a heater in terms of power output. A very fit person can produce 300W (1200 to 1500W in total) or more for one hour!

A fan is needed to keep you from overheating.

I think if you want to prevent jerky pedalling without a flywheel, you have to learn to pedal smoothly and possibly have it properly adjusted and with the right gear ratios. My pedalling on a trainer wasn’t jerky. My friend who tried it out was too jerky.

Other alternatives to exercise bikes include elliptical trainers and treadmills.

Roland Burton

First, somebody should mention that I can buy electricity at $0.10 per KWH. Or I could create the same amount of energy by pedalling for 10 hours, if I could generate 100 useful watts for 10 hours. Not likely I would make my own, if commercial power is available.

Second, operating a blender from a bike is more efficient if you don’t involve electricity, and if all you want to operate is a blender. But I haven’t figured out how to generate lighting with a bicycle unless I use electricity. Anybody?


If you want to turn human effort into energy, and you’re already using fossil fuels or are connected to the grid, the best way is to install weatherstripping, insulation etc. Or climb up a ladder and replace your inefficient lamp with an LED. You can save a LOT of energy with this kind of activity.

For transportation, bicycles can save a lot of energy. Or, at least, I think so. It may be, at least in places where there are a lot of automobiles, that medical care for injuries consumes more power than a cyclist could ever save. I suspect this may be the case for me, since I’ve had thousands of dollars worth of medical care to put my hand back together after being hit by a car. In any case, it isn’t that you’re using human power, it’s that bicycles don’t need much power. Almost all of the energy saved could be obtained with an electric bike. Furthermore, beyond a reasonable amount of exercise to maintain health, additional food will be necessary. If that comes from industrial farming, I suspect the energy advantage goes away. Rowing boats can also save a lot of energy, but again it’s because of the low power consumption required, not where that power is coming from. If you put an electric trolling motor on your rowboat, you’ll probably conserve as much as by rowing, at least assuming you get your necessary exercise elsewhere.

The magnitude of human physical power is tiny compared to what most of us in the developed world use from other sources. For instance, according to Wikipedia, per capita electrical power in the US is almost 1400 Watts, and just over 300 Watts for the world as a whole. I doubt if most of us could put out 100 Watts for a couple of hours, and then you have to subtract for inefficiencies in your system.

As someone else mentioned, comparing charging batteries with some gadget that uses muscular power immediately isn’t equivalent. What about using the electricity immediately?

When it comes to flywheels, a much lighter flywheel can be used if you spin it faster. Assuming you have an efficient transmission, at least, to spin it with. Typically, an exercise bike flywheel doesn’t turn much faster than a bicycle wheel. Spin it at 2,000 rpm instead of 200 and you’ll only need 1 percent of the weight, assuming the same diameter. Probably it would be better to use a smaller diameter and 5 or 10 percent of the weight, but it illustrates the idea.

There was much discussion in the article about the energy cost of making batteries, and the necessity of replacing them now and then. I wonder if those calculations were done for batteries made with recycled materials. My guess is it makes a big difference. Also, I wonder how different battery chemistries compare in energy to manufacture. I suspect a lithium iron phosphate battery would last longer than a lead acid batttery.

Someone mentioned human waste being more eco friendly than

Having said all this, I see some places where human power has some real advantages. For instance, if you want to read at night when you’ve been hiking. All that’s required is a very low power LED, less than a watt. Or maybe you want to be sure your bicycle has a headlight even when you’ve forgotten to change the battery.

Someone mentioned pumping water during power outages. It would probably be best to look into large, manual bilge pumps meant for good sized boats. Best to figure out how to power them with your legs, though. Alternatively, it’s said that the best bilge pump in the world is a terrified person with a bucket.


I would like to know if you had calculated how much embodied energy required by other energy generator plants. Such nuclear power plant, solar power plant, etc…

Clifton Austin Register

A very negative article. Does this guy get his money from the oil companies? I was looking for something to keep me alive during a long SHTF situation. With months or years to pedal a bike so I can listen to the ham radio or play computer games as I wait for the potatos to grow, I don’t need 99.99% return on my investment of effort. I need something cheap, fixable, available and able to last for a long time. This article is “You can’t get anywhere that way”. In a situation like in the movie(Jimmy Stewart), Flight of the Phoenix, whatever works will be just fine.


Like most of the articles I’ve read on this site, I thought this was a very interesting read and definitely worth sharing. However, I did have a few comments.

Pedal power may not be a particularly efficient use of chemical (food) energy, unless we assume that this chemical energy must be consumed one way or another. The primary example is fitness. In this case, we are simply harnessing an energy source that will be there either way, but getting a few watts of useful power out for our effort. Most people who use exercise bikes simply use a friction load, which turns all of that power to heat.

One way to get more of this “incidental power” for useful application is to skip the battery entirely, as some others have also mentioned. Let’s say your TV will always be on, drawing power, while you work out. You might as well drive your TV with the incidental power from your workout instead of from the coal power plant feeding the electrical grid.

In the article, some of the efficiency numbers stated seemed quite low to me. I’ve been researching charge efficiencies for modern Li-Ion battery chemistries and they are MUCH better than the lead-acid “worst case scenario” presented here. Also, switched mode power supplies (SMPS) can be >90% efficient. For example, the well-respected SMPS manufacturer MeanWell has a line of DC-DC converters (the LDH-45A) that can be up to 95% efficient while maintaining constant current with a boost topology.

I agree with another commentor that we should generally stay away from AC loads. This is quite easy nowadays, since a large proportion of our appliances (modern TVs, computers, even lights) are actually DC-powered and must use an often-external AC-DC converter. Remove the converter and drive it with the appropriate DC voltage. Even when an internal converter is involved, many AC SMPS can accept an albeit high voltage DC input, since they rectify the AC anyways. Of course, it’s always best to make sure the inlet components and rectifier can handle that. Or build your own appliances and make them suit your needs!

Some mentioned pumped storage…I do not think this compares well in terms of efficiency or practicality compared to directly powering an appliance or charging a Li-Ion battery. Pumped storage is an inefficient way to get electricity, and it generally makes the most sense on large grids with appropriate geology, such as mountains. You would be very surprised how little energy can be stored by raising a few liters of water even several meters into the air. However, if you want to use that to provide pressure to your faucets, showers, irrigation, etc, then I think it’s a fantastic idea. Why drive a motor pump when you can directly pump the water?

So while I agree that using pedal power to replace a traditional power source and spending your time and energy just to make power is not sustainable nor cost-effective, I think that collecting the “waste energy” from your workout and putting it to good use is an excellent idea and is highly sustainable. When you think of it as recapturing a waste product, it is essentially free power that you can use for something meaningful, or just to displace other energy usage to watch TV while you ride!

kenneth torino

I would suggest updating the electrical power generated portion of the article to include K-Tor’s pedal powered generators that have been available for the last several years. They address a number of your concerns, being designed from scratch to generate power and do not use a bicycle to generate power. Also there is little steel or weight to them reducing the energy used in making steel.

I found your article very interesting. thanks for the research and writing it. Ken

John White

What if a exercise bikes, rowing machines and other gym equipment could be fitted with dynamos and the combined energy produced could be put back into the national grid, much like solar power can be? Is this something that could be a viable solution. Im thinking spin classes must produce some serious wattage?

jon deaux

My ex-wife left an exercise bike with a cast-iron flywheel at our son’s house when she left the state. I’ve been toying with the idea of using the flywheel as a pulley to drive an alternator from a salvage yard to charge a 12v deep cycle trolling motor battery for emergency power after a hurricane.

I agree with you that for day-to-day use, bicycle powered generators or alternators are not efficient, but this would be a short-term backup until the power grid was restored.

Gregory Vanderlaan

There are Many Exercise Bicycles Currently in Use at Health Clubs. The People Ride those Bicycles AND GENERATE NO POWER AT ALL… A Total Waste… Especially when You Consider the Massive Electricity BILLS That a Health Club Has Heating the Swimming Pools… At Humboldt State University the Campus Center for Appropriate Technology Built a SIX Bicycle Power Plant for Use at Rock Concerts and Lectures… It’s GREAT!


It is probably inefficient or unsustainable if you are a fat slimy westerner that is used to air-conditioned rooms and pizza deliveries.

Theo Schmidt

The author writes: “Pedaling a modern stationary bicycle to produce electricity might be a great work-out, but in many cases, it is not sustainable. While humans are rather inefficient engines converting food into work, this is not the problem we want to address here; people have to move in order to stay healthy, so we might as well use that energy to operate machinery.”

I agree, as do several comments, converting food energy to mechanical work and even more to electricity has a very low efficiency, and is not suitable for continuous operation at high power levels and is highly unsustainable depending on the food eaten. If you eat things that grow anyway without fertiliser, like local garden apples, this energy is “free”, but if you eat industrial food or food imported from far away or from heated or illuminated greenhouses, or grown with imported fertilisers, it can contain a huge amount of gray energy.

People usually think that human power in moderation, also like cycling or walking, doesn’t involve eating more, but this is untrue. For each and every Joule expended as mechanical work, at least 4-5 Joules must be eaten as food. This usually goes unnoticed because a large person must eat about 1 Megajoule as food each day just for living at rest, without any extra muscular work. Also, many people eat too much anyway and are happy to lose fat and for the health benefits.

Therefore low-power and/or occasional human-powered devices are fine, especially when they replace things with much less efficiency, e.g. pedalling a short distance instead of motoring with a multi-ton car a longer distance.

However, “professional” human power exerted all day at high power levels, such as cycle-couriering or some of the applications mentioned, is probably more efficient when motor-assisted, unless the food is “free” energy-wise.