Department of Energy Argonne National Laboratory Office of Science NEWTON's Homepage NEWTON's Homepage
NEWTON, Ask A Scientist!
NEWTON Home Page NEWTON Teachers Visit Our Archives Ask A Question How To Ask A Question Question of the Week Our Expert Scientists Volunteer at NEWTON! Frequently Asked Questions Referencing NEWTON About NEWTON About Ask A Scientist Education At Argonne Perpetual Energy-Motion-Buoyancy Device
Name: Edward
Status: Student
Grade: Other
Location: NY
Country: United States
Date: Spring 2010

Say you have a large, hollow, very rigid, water-tight steel ball. When you place it in water it floats. Now you fill it with sand so it is denser than water. Now when you place it in water, it sinks. You drop it into a deep part of the ocean, say 10,000 feet deep. When it sinks to the bottom, you devise a mechanism to expel much of the sand. (For example you could turn on a motor to push out the sand. You might need to replace the void with some compressed air you have in a tank waiting at the bottom. But for the moment, let's ignore this problem.) Once the sand is expelled (and replaced with air), the steel ball starts to rise (float) with a calculable force.

Would it not be theoretically possible to hook up that steel ball to say a generator and capture some of that energy as it rises? That energy (work) would be the buoyant force times the distance traveled (10,000 feet). (Of course, the drag of the generator would have to be less than the buoyant force so that the ball would keep rising to the surface.) Once it comes to the surface, it would require relatively little energy (work) to refill the ball with sand and drop it in the ocean again. Once it is at the bottom, it would likewise require relatively little work to expel the sand. Then on the way up, you recapture some of the buoyant force, as before. The amount of work (energy) to fill and empty the ball is fixed.

The amount to potentially be recovered can be increased arbitrarily. (Say you drop it into 20,000 or 30,000 feet deep water instead of 10,000.) The concept is to use gravity (and buoyancy) to "generate" this energy. And this energy captured could be greater than the (relatively little) energy required to generate it. I know it's not possible to get more energy (= work) out of a machine than you put into it (and with losses you always get less), so I must be missing some physical principal in this experiment. Do you have any idea what I am missing?

You are missing the energy required to expel sand at the bottom of the sea. This energy is considerable, because it is lifting a column of water nearly two miles high.

Tim Mooney

The problem with your scheme is this: how would you fill the float with air when it is at the bottom of the sea? You consider using a compressed air tank. How much air would you need? At what pressure? (What is the pressure of the water at 10,000 ft deep?) How much energy would it require to compress that much air to that pressure?

Richard Barrans Jr., Ph.D., M.Ed.
Department of Physics and Astronomy
University of Wyoming

Hi Edward,

From your thorough description I can tell you have thought through many aspects of this machine. However, there is one area where I think you are overlooking a major energy requirement.

To simplify your machine I just considered a sphere you could fill with water. The metal sphere and other equipment would cause the water-filled sphere to sink. At the sea floor you could then activate a pump to empty the sphere. For this you would need a pump adequate to exceed the water pressure at the depth you were operating. You could size the pump to empty the sphere in a certain amount of time and then from the rate of pumping and the required pumping pressure you could determine the energy that would require. The energy at this step, whether it be to pump water or in some way to remove sand would require work / energy you are not accounting for.

As I see it, as you use energy to empty the sphere you are transferring or storing potential energy in the sphere. Then when you release the sphere you could recover energy in some manner like you describe. However, as with all other machines none of the operations are 100% efficient so energy would be lost at many steps and you would expend more energy than you would produce.

You could consider a machine like this similar to a pumped storage or compressed air system where you use energy (to pump the sphere empty) at off peak times when it is cheap and then generate energy during peak load times, however there is always a net loss of energy with such a system.

Carlton Schroeder
Eagle River, WI

Hi Edward,

Well, I am going to get right to the point here.... There is no free lunch! Your suggested device attempts to violate the First Law of Thermodynamics, which is an expression of the principle of conservation of energy, and states that energy can be transformed (changed from one form to another), BUT IT CAN NEVER BE CREATED OR DESTROYED.

Your suggested device is attempting to create energy, and that is impossible. There has never (and I repeat...NEVER) been an exception ever found that has been proven to violate the First Law.

In fact, if you were to do a rigorous analysis of the actual energy needed as an "input" to your system, such as that needed for the pump to remove sand from your metal ball, you will find that it takes more energy (probably FAR more energy) than the buoyancy of the ball will produce. As you go deeper, there is more buoyancy, but this is cancelled by the increased amount of energy needed to expel the sand at greater pressures.

Every year, for well over a hundred years, the US Patent Office gets applications for similar devices that inventors claim can get something for nothing, and none have ever been successfully demonstrated. So, put your trust in the good old First law of Thermodynamics. It has never, ever been proven wrong! Detailed analysis of every "free energy" or "perpetual motion machine" that has ever been conceived, has shown that every single one cannot work.


Bob Wilson

Hi Edward,

There's a lot of interest in hydrokinetics as a clean, renewable way to capture energy. This includes various kinds of buoys, planes/wings, and other devices. Most of them rely on tides or waves, which is a key difference in source of energy, but the kind of equipment envisioned is very similar to what you are proposing.

In terms of the physics, your concept is actually similar to hydroelectric energy -- but instead of using gravity and water, you are using gravity and sand/material. With hydroelectric power, water flows from a higher altitude to a lower one - flow converts the potential energy to kinetic energy, and generators convert the kinetic energy to electricity. Rather that directing the material through pipes and generators, you are using a steel ball and a tethered generator. But the conversion path of potential energy to kinetic to electrical is similar.

In your case, I suggest you do a calculation of the energy it takes to perform all the work you are talking about, how much energy can be stored and delivered, and then compare it with the installed (capital) cost of the system. If it turns out to be favorable compared to alternatives, you may have something.

Unfortunately, there are lots of ways to capture energy from the environment -- figuring out sources and methods is not the hard part. The hard part is finding ways to capture and deliver the energy economically. If a method is not affordable compared to alternatives, then it won't be of great interest or value to the world.

There are a few more difficulties I would like to add to your list that you did not mention. First, you would need to ship all that sand/material to your deep-water buoy. Have you determined how much material would you need, and how will it get to the buoy? This is non-trivial in terms of energy costs, and also brings up sustainability questions Will to be mining this material? To produce energy on societally important scales, a large amount of material will be necessary. Second, you would need to get the energy from the deepwater site to where people use it. Have you considered how you will store the energy that's generated on the buoy, and then how will you transport it? This is an issue that the hydrokinetic energy folks have thought about, so you can rely on their work for a head start. Other than that, the cost of the equipment you suggest will be high, and maintenance costs will be high too (corrosion and weathering are big problems at sea). All of these issues will be important in judging the merit of your concept.

In summary, there's capturable energy all around us. But that is only part of the problem. You have to store and transport it to where the energy will be used, and you have to generate, store, and delivery the energy in a cost effective way. The next step for your idea may be to put some real numbers to the concept, and then figure out the actual costs and production -- then you can tell if the idea has any value.

Hope this helps,
Burr Zimmerman

You are not "missing" anything. You understand correctly how the machine will work. But you err in thinking that the machine is somehow getting more energy out than is being put in.

Your machine converts the potential energy of a load of sand at sea level, being pulled by gravity downward to the bottom of the ocean, into useful work. It is gravity, pulling on the sand, that is the source of the energy. Your machine is capturing the energy as the ball moves upwards instead of moving downwards, but either is possible by adjusting the buoyancy, and you get the same amount of generated energy out of the sand.

You could make a similar machine that operates on land in almost the same way: you could put an elevator on the side of a tall mountain. When you put rocks in at the top, the elevator will descend and generate power. At the bottom you dump out the rocks, and the elevator will go back up again. Or a conveyor belt would work too.

The same principle, of stuff "falling" from a high place to a low place to generate power, is used in a hydroelectric dam. There, falling water is used instead of falling sand, and the water rotates a turbine, instead of making a ball move up and down.

Robert Erck

You have a lot of machinery operating in your device:

1. The "mechanism" to expel the sand you assume contributes a negligible amount of work -- not so.

2. Then after raising the sphere to the surface you fill it again, with no energy cost.

3. You cannot use buoyancy vs. gravity the way you suggest. It costs energy to submerge the sphere and you cannot regain that when the sphere re-surfaces, because there is turbulence and friction that resists the motion of the sphere no matter whether the sphere is rising or sinking. Many have tried and none have succeeded in making a perpetual motion machine. In fact, the U.S. patent office has terminated accepting such patents, out of hand. It is all those little "losses" that foil such attempts.

Vince Calder

Click here to return to the Engineering Archives

NEWTON is an electronic community for Science, Math, and Computer Science K-12 Educators, sponsored and operated by Argonne National Laboratory's Educational Programs, Andrew Skipor, Ph.D., Head of Educational Programs.

For assistance with NEWTON contact a System Operator (, or at Argonne's Educational Programs

Educational Programs
Building 360
9700 S. Cass Ave.
Argonne, Illinois
60439-4845, USA
Update: June 2012
Weclome To Newton

Argonne National Laboratory