Free Energy From Air? Sorry, no.

After href="">the
fiasco that was my flame against the downwind faster than the wind
vehicle, you might think that I'd be afraid of touching on more air-powered
perpetual motion. You'd be wrong :-). I'm not afraid to make a fool of myself
if I stand a chance of learning something in the process - and in this case,
it's so obviously bogus that even if I was afraid, the sheer stupidity here
would be more than enough to paper over my anxieties. Take a look at this -
the good part comes towards the end.

What this video is about is one of the more novel approaches that
I've seen towards emission-free vehicles. Instead of using electricity
provided by a battery, an Indian company has developed a car that operates
on compressed air. It drives the pistons of an engine by allowing the
compressed air to decompress, and then it uses the motion produced to drive
the wheels of a car (for motion), and a small electrical generator (for
the vehicles electronics, instruments, and accessories.) It's a really
clever idea. And there's absolutely nothing magical about it: a compressed gas
tank is just another way of storing energy: it takes energy to compress the
gas into the tank; some of that energy is given back when the gas is
uncompressed. The neat thing about the compressed air vehicle is that
the gas tank weighs less than the batteries of an electric or hybrid,
and it doesn't have all of the nasty caustic chemicals that are in
a typical battery. The tank can be filled in under a minute at a
compressed gas station, or overnight by plugging it in and running
an electric compressor. And it can go for around 200 miles at
50-60 mph on a single tank. So it's a reasonably practical car. Certainly
not what you'd want for long-haul driving, but for commuters, it's
pretty terrific.

Of course, marketing drones are never satisfied with a simple good
idea. Just being good isn't good enough.

Now, I'm all for being creative about mocking idiots. It's
fun coming up with ways of rephrasing their bullshit to demonstrate
how stupid it is. But in this case, I just can't do it. I can't
do anything better than what they've done themselves. In the smarmy
tones of the voiceover guy, starting around the 2:45 mark, they
conclude the video with the following:

The air car isn't cost free to operate, because it does take some energy to
compress air - but interestingly, MDI has created a generator powered
by compressed air. Which presents a tantalizing possibility: what if
that generator was onboard the car? Then one day, perhaps the
compressed air that runs the cars will also run a generator to
compress its own air. A car that runs on air, and constantly
refuels itself! Round a round, a perfect circle. Perpetual motion.
A no-cost fillup ever. Not one iota of pollutants, ever. And a cute
car. All for about $15,000. That's a future car.

They admit that they're trying to sell you a perpetual
motion machine! And they don't understand that there's anything wrong
with that. How could I possibly say anything to make them look

For the sake of pedantry, I'll walk you through the problem with
this, and all other perpetual motion machines.

It's an unfortunate fact of our universe, but any time you change energy
from one form to another, you lose some. No matter what you do, you
always lose some. It comes down to thermodynamics, which is how
physics describes energetic interactions. The laws of thermodynamics can,
non-mathematically, be sumarized as: "You can't win", "You can't even break
even", and "You can't quit the game".

The first rule tells you that you'll never get out more
energy than you put in. So the perpetual motion story doesn't work,
even under the first rule. You have some quantity of energy
stored in the compressed gas. You use some of that energy
to make the car move, and you use some of it to run the compressor.
You're going to run out: you're putting less energy back in to the
air tank that you're taking out of it, because part of it is moving
the car. So even if you had a perfectly efficient system,
it wouldn't matter. In fact, with a perfectly efficient compressor,
running the compressor would be a no-op - it would do exactly nothing.
Every bit of energy that it drew from the compressed air in the tank would
be returned to the tank as compressed air - so it would be as if it weren't
part of the system. But the car would still be draining energy.

The second rule tells you that you can't even get out as much energy as
you put in. Any time you use energy to do work, any time you transfer energy
from one from to another, some of the energy gets wasted. So even if you never
wanted to move the car - you just wanted to keep the tank full of compressed
gas - you'd lose. When you take the energy in the compressed gas, and you use
it to spin a generator, what you're doing is changing energy from its stored
form in the compressed gas, to a mechanical form which pushes the generator,
to an electrical form. Every step of that involves loss: decompressing the
air, pushing the generator, generating electricity from the generator,
using the electricity to produce mechanical energy to drive the compressor,
and using the mechanical energy to compress the air. So just the closed
cycle: cycling the air out of the tank, into the generator, and then using the resulting electricity
to drive a compressor to refill the tank - will end up, in short order, with
an empty tank.

And the third rule? It says you're stuck: you can't create a system
which doesn't have to follow the first two rules. If you're in this universe,
you're pretty well stuck with this. There is no free energy. There is no
such thing as perpetual motion. You'll always get less energy out of a system
than you put in.

If you look at the way we really produce energy, we use systems where
losing a whole of energy is just fine. Part of the reason that oil is so
damned valuable is because it contains a huge amount of energy for
its mass, and it's easy to extract that energy. The fancy way of saying that
is that gasoline has a tremendous energy density. There's roughly 46
megajoules of energy per kilogram of gasoline: that's a hell of lot of energy!
There's enough energy in it that it more than makes up for the effort of
carting it around, even if we use it very inefficiently. Typical gasoline
engines operate at less than 20% efficiency - meaning that the amount
of energy from the gasoline that actual gets translated into mechanical energy
moving a car - is less than one fifth of the amount of energy released by
burning the gas. Even with 4/5ths of the energy going to waste, gasoline is a
remarkably efficient energy carrier for us.

The challenge in all alternative fuel sources for vehicles is that
we need to find ways of storing energy that give us something like the
energy density of gasoline, and which don't cost a ridiculous amount
of energy to "fill up". The very best electrical batteries that we've
devised have an energy density of just 2.5 million joules per kilogram - a
bit under 1/20th the energy density of gasoline!

Lots of companies are looking at really interesting ideas for how to do a
better job storing energy for driving vehicles. Compressed gas is a good idea;
at the pressure that's being looked at for cars (around 4,500 pounds per
square inch), it's got a respectable energy density: about 4 million joules
per kilogram. And the engines that are built to run off it get around 18-20%
efficiency - roughly the same as gasoline engines. In terms of energy density,
it's still pretty awful compared to gasoline - but it's significantly better
than even the best battery! And there's good reason to believe that we can
work out storage for gas at higher pressures than that, and that we can boost
the efficiency of compressed gas engines by at least a little bit. So this is,
at least potentially, a serious, viable technology.

But it's not free energy. And when you tell lies about a technology,
getting people to believe that in things that are too good to be true, all you
do is set things up for failure. If someone can produce a really practical
compressed gas car for a reasonable price, and it's 20% efficient, and
requires you to refill the compressed air tank by plugging it in in your
garage every night, it would be absolutely brilliant. People like me
will beat a path to you door to buy one! But if you've told people that once
they buy it, it'll be free and perfect and will generate its own fuel by
magic, then they'll be serious pissed off at the electric bills that they're
paying to refuel their supposedly free-energy car - and you'll have successfully
transformed it in the minds of your customers from something brilliant to
something incredibly disappointing.

More like this

What's the deal with the Chevy Volt? Well, obviously, it is a cool car. A plug-in hybrid. The problem is in how to quantify its efficiency. Normal hybrids (the non-plug in type) have only one type of energy input, gasoline. The Volt can take gasoline or electricity input. This makes it…
Here's a quick bit of obnoxious bad math. I saw this myself in a link to an AP article via, and a reader sent me a link to the same story via CNN. It's yet another example of what I call a metric error: that is, the use of a measurement in a way that makes it appear to mean something…
This was on my 'to do' list, but Tom at Swans on Tea beat me to it. Basically, this grocery store has these plates that when depressed produce electrical energy. Tom does a good job pointing out that this is not free energy (the original article says this also). Clearly, the energy comes from…
This has been mentioned elsewhere - like on the Machinist blog on Salon (where I first saw it) - but I can't resist saying something about it myself. And I'll also chip in a little bit of originality, by also criticizing some of the people that I've seen criticizing it. The story is, there's a…

[fail]maybe they were planning on having vents on it to catch all the air that is blowing by as they drive and forcing it into tanks[/fail]

Their idea works find as long as you always drive down steep hills, with a tail wind

... or not.

I don't know which is sadder,
* the marketing people who may actually believe this and never cleared it past the engineers.
* that the bulk of general public who won't understand why this can never work.

I was already going to point out that the claims about fast refueling seemed suspect (ask any first-term physics student about the gas laws and ask any diver about the practical implications for filling tanks too quickly) when the perpetual-motion part hit me right between the eyes. Ow. Wasn't expecting that. How can anyone write that stuff?


Yeah, exactly.

I know that there's been work on rapid refueling of air-cars. My gas-law-foo is rather weak; from what I understand, you can refill a compressed gas tank in roughly the same amount of time that it takes to refill a gasoline car, and that the heat produced isn't all that different from the heat of a running engine in a gasoline car. I should probably sit down with some paper and a pencil and work it out :-).

On the subject of perpetual motion, a fun challenge is to explain to someone with not much background in physics or biology why a species that gets 100% of its sustenance from cannibalism is an impossibility. It of course boils down to a perpetual motion machine, but articulating the energy sources and costs in an understandable way is interesting.

This actually came up in a conversation with my wife, who fits the bill as someone with a weak background in physics and biology :) I eventually convinced her, but not before going through some weird thought experiment involving an impenetrable sphere floating in space filled with 100 rabbits...

Compressed gas is a good idea; at the pressure that's being looked at for cars (around 4,500 pounds per square inch)

Eh, uh, not to be a party-pooper, but I don't want to be anywhere near a cylinder of 4500 psi compressed air travelling at 60mph in close proximity to a bunch of metal boxes driven by slightly-advanced primates, also traveling at 60mph and possibly having their own cylinder of 4500 psi compressed air... Is there really a way to make that safe?? ? (well, "safe-ish" I mean, as in no worse than the current danger...)

Beyond all the BS about perpetual motion, cars running on compressed air are neither new, innovative, or useful.

Compressed air is a terrible store of energy, and is also incredibly dangerous - guess what happens when one of those tanks ruptures!

Also, don't be fooled by the other marketing language - they said a range of 200 miles AND UP TO 50 to 60 miles an hour. NOT at the same time! I expect the 200 miles is mostly downhill.

All garbage, 100%.

By Anonymous (not verified) on 02 Nov 2009 #permalink

FWIW, compressed air (and hydrogen) aren't energy sources- they are just ways to transfer energy around from place to place. There aren't any hydrogen wells, and there aren't any compressed air wells. You still need to somehow use energy to compress the air in the first place- and that method is likely to be even more inefficient.

The 20% energy efficiency for the car has to be marked down significantly if you consider the inefficiencies in the original generation source and power line transmissions. In other words, powering a car on compressed air may well be worse in terms of greenhouse gas emissions than running it on gas.

@Tim Howland: Very worthwhile to point out -- however, there is a potential advantage to a "green" energy storage medium even if it is not actually an energy source. If you have a way of producing massive amounts of energy off-site without emitting greenhouse gasses (*cough*nuclear*cough*) then hydrogen -- or I suppose compressed air -- frees you from having to put a nuclear reactor or whatever inside every single vehicle. You lose energy, but you gain the advantage of using an off-site energy source.

If you are not a fan of nuclear, same thing could be said about wind, solar, whatever.

Of course, if you are getting your electricity from coal or natural gas, then yeah, that ain't very green.

I'm wondering if the guy interviewed claimed these things or an over-enthusiastic documentarian jumped to those conclusions by himself.

@James Sweet:
That's a good point. The energy storage capabilities of hydrogen/compressed air/batteries becomes especially important when it comes to non-nuclear energy, and it is essentially the biggest hurdle we have to implementing non-nuclear power (it is also one of the reasons that people who have actually given a good look at power generation options realise that the knee-jerk reaction against nuclear energy is rubbish). One of the main advantages of fossil fuels is how reliable and controllable their power generation is. That is an aspect that nuclear reactors share (so a storage medium is only required for transferring that power to off-grid items like cars), but we unfortunately do not control when the sun shines and when the wind blows. Therefore, adequate solar and wind power needs a reliable storage medium.

The whole perpetual motion thing is a red herring, and to concentrate on that bit of copywriters' idiocy is to avoid the fact that these cars do's not an Indian company,it's a French company run by an ex-formula 1 engineer who invented and developed the engine.Tata has lisenced the technology to produce the cars in India. Incidentally, if James Sweet is nervous about driving alongside a vehicle filled with compressed air, contained in specially engineered carbon-fibre tanks, then why isn't he bothered by driving alongside vehicles carrying gallons of highly flammable gasoline held in thin skinned metal tanks?

I believe I ran into this design sometime last year, and worked out the energy storage of the primary air storage tanks. It was substantially less than a gallon of gasoline equivalent, and the power density was nothing to be awed by either. In general, compressed energy storage is comparable to lead-acid batteries in mass efficiency and inferior in volumetric efficiency. It's possible that this device has advantages over a streamlined conventional golf cart, but I'm not sure what those advantages would be.

On the plus side, not really all that dangerous unless you drop the air tank in a fire (which will cause it to heat up and eventually explode); the energy storage isn't really sufficient to turn the tank (or car) into hazardous shrapnel.

I'm still waiting for my clockwork car that has a mainspring which stores energy equivalent to a tank of gas.

I could refuel it by fitting the key to a windmill.

I'm a bit puzzled about the energy density of compressed air given in the otherwise fine article. I tried doing the math and checking some sources (for example but couldn't come up with anything better than about 0.5 MJ/kg for air (T=300K) at around 4500 psi (310 bar)... which is nowhere near the claimed 4 MJ/kg. Of course this doesn't make the compressed air car look any better.

Aitch: When a bottle of compressed gas fails catastrophically, it explodes and it (or its parts) becomes a deadly projectile. When a gasoline tank ruptures, the contents spill and may ignite if there is a spark or sufficient heat source nearby. But the whole vehicle never explodes (like in the movies). For gasoline to burn, it has to evaporate and mix with oxygen. Thus, the combustion takes place only at the surface of the liquid spill, where fuel evaporates and mixes with air. The whole contents of the tank never burn at once (explode). It can still be a hell of a fire, but it's no comparison to the explosion of gas bottle.
Also, Diesel fuel is almost impossible to ignite under ambient conditions.

By Philippe Leick (not verified) on 02 Nov 2009 #permalink

I'm with @Philippe above. The safety factors are a major problem. That whole perpetual motion thing is a red herring.

By Sweetwater Tom (not verified) on 02 Nov 2009 #permalink

The novel version of The Running Man has "air cars" in it which work as described in this video. (Only the police get the gas-powered cars to chase people down.)

By Ergo Ratio (not verified) on 02 Nov 2009 #permalink

By the way, wasn't there also one of the worst abuses of the word "ironically" (ironically, the inventor shares the birthday with Jules Verne...)?

By Daniele A. Gewurz (not verified) on 02 Nov 2009 #permalink

Are you sure the video isn't a parody?

I think the final claim that its you can build a "perpetual motion machine" (the use of this term) makes me feel its a parody. A snake-oil salesman will, I would think, avoid this term.

On the plus side, not really all that dangerous unless you drop the air tank in a fire (which will cause it to heat up and eventually explode); the energy storage isn't really sufficient to turn the tank (or car) into hazardous shrapnel.

truck and tractor tires rupturing have been known to kill people standing nearby, so i'm not taking any chances with considerably higher pressures in metal containers, myself.

By Nomen Nescio (not verified) on 02 Nov 2009 #permalink

The gas laws aren't a problem with filling up the tank quickly. The car has a small tank to be filled to 4500 psi. The filling station has a very large tank, already at 4500 psi. Connect one to the other and the small tank fills very quickly to almost 4500 psi (perhaps a smidge cooler than the big tank) and drives off. Then the compressor on the big tank notices that the pressure is a little less than 4500 psi and starts up pushing air in, making the big tank a little warmer than it was.

Now if you want to compress the air from scratch every time a small tank comes in for a refill, then you have gas law problems.

By Keith Harwood (not verified) on 02 Nov 2009 #permalink

I've seen an air powered vehicle at work! It was a high school experiment inspired by a scrap-hound instructor's latest finds. I don't remember details about the exact components, it probably wasn't the ideal setup anyway. But it was good food for thought.

It did okay, for a while. Then things got really, really cold! The cooling of the depressurizing tank dropped the pressure significantly more than the air released. The air wasn't totally dry, apparently, because what moisture was left eventually iced up in the lines until they were nearly plugged. Regardless, by then the drivetrain was showing frost and turning stiffly with the cold-altered tolerances. It definitely wasn't an ideal setup, just something put together from repurposed surplus shop treasures. But it did demonstrate that any moisture whatsoever could become a major problem. And that waste heat is a whole lot easier to deal with than waste cold. But there's a silver lining: The compressed air powered cars of the future will have air conditioning that'll put anything we have now to shame!

I don't have any math at hand, but I believe much of the energy isn't actually stored in the compressed gas itself. Compressed air requires heating to get most out of it. This heat, I suppose, comes from the environment (presumably the car wouldn't work too well in a colder climate). Suppose the heat produced by compression is captured and used? Of course it's not possible to exceed 100% efficiency by draining energy from the atmosphere, but the ratio could be high.

@19, 23:

Is video editing somehow a novel and nigh-miraculous technology?

(Besides, the technology is perfectly legitimate, so why should we be surprised that Kaku praised it?)

Really, this car is fake/vapoware. Many peoples calculated real distance it could go (as "volume of cylinder * number of cylinders * RPM" is volume of air per second, with assumption of "ideal" situation, that final pressure is not higher than atmospheric one, which is too good to be true) and found that declared amount of air at declared pressure gives about 1 or 2 miles, not 200 miles :(

By Lev Serebryakov (not verified) on 02 Nov 2009 #permalink

I was wondering what the energy density of liquid nitrogen is for this kind of application. I might take a look at it tomorrow, but I'd imagine that it might push the energy density up quite a bit.

Or not. Just googled this:

I can assure you that the French company MDI that has developed the air car has never claimed perpetual motion. They cannot be held responsible for what people say about them on Youtube.
The cars really work and of course they need to be filled up with compressed air now and again. It takes hours to fill up the tanks of an air station with compressed air, but only a couple of minutes to move that stock of air from the air station to the vehicle tank. So the claims of filling up your tank within three minutes (or less for the tiny Airpod) are not at all unjustified!
I invite you to have a look at my site where you will find all information, pictures, videos, links and the latest news about air cars. Enjoy!

if James Sweet is nervous about driving alongside a vehicle filled with compressed air, contained in specially engineered carbon-fibre tanks, then why isn't he bothered by driving alongside vehicles carrying gallons of highly flammable gasoline held in thin skinned metal tanks

Because liquid gasoline is not explosive. It's flammable, but not explosive.

I am not saying I couldn't be assured it was safe, but given the fact that -- as Nomen Nescio said -- rubber tires at <100 psi have been known to explode and kill people, I'm not going to just take for granted that it's safe. If the cylinder gets "dented" in a high-speed collision, what are the odds that that will compromise the structural integrity of the tank and that it will have a chance at rupturing?

If I am standing six feet from a tank of compressed air at 4500 psi when it ruptures, that's probably the last blog comment for me. On the other hand, if I am standing six feet from a car's gas tank when it catches fire, I just move away from it. Even if I am inside a car when the gas tank catches fire -- a terrifying prospect -- if I am conscious and able to free myself and open the doors and leave, it is not a problem.

Gar, HTML fail for me.

What got cut out was something to the effect of, I could be assured that compressed air was safe, but I am not taking it for granted. It is just one of many hurdles that such a car would have to overcome. Who knows, this could turn out to be a great idea, for all I know...

If I am standing six feet from a tank of compressed air at 4500 psi when it ruptures, that's probably the last blog comment for me.

At 4500 psi, is there any chance the vessel would fail like a rocket, rather than a bomb? That would certainly make for some interesting accident scenarios.

Maybe that could suggest a fail-safe mechanism- use an airbag-like trigger to jettison the entire tank like a compressed air rocket, driving it free of the incipient collision.

Of course, then you'd have to worry about it coming back down... a parachute perhaps?

...then again, we haven't yet considered tunnels and overpasses. ;)

@39, etc:

Looking roughly at the amount of energy involved in a tank like that, it's pretty significant. But if you consider pressure-release mechanisms in other high-pressure applications,
it should be possible to put a safety pressure release on a tank that would dissipate the pressure without an explosion.

We've worked out some pretty amazingly effective ways of dissipating kinetic energy in high speed collisions without killing passengers, using structural supports, crumple zones, airbags, etc. The amount of energy in one of these tanks is not all that dramatically different. If everything else could be made to work, I'm pretty sure a competent engineer
could design a system that monitored the tank for any signs of stress, and rapidly but safely decompressed the tank when necessary.

For me, the real question about these things is energy density and efficiency. Looking around the web, I've seen the energy density of a compressed air tank as ranging from 1/2 megajoule/kg up to 5 mJ/kg, and efficiency of compressed air engines ranging from 5% to 20%.

If you can get both of those towards the top of the range, you've got the potential to produce a viable vehicle. If you can't, then the whole thing is completely bogus. Unfortunately, while I can do some back of the envelope math on how much energy you could store in a tank of gas, I don't know enough about how much safety equipment you'd need to be able to work out the mass of the vehicle; nor do I know enough about how a compressed gas engine works to be able to work out *its* mass or what a reasonable expectation for its efficiency would be.

If anyone knows how to do that math - please feel free to come forward and share. If it passes a careful proofreading, I'll promote it to a fresh post with full attribution.

i assume the tanks would have built-in failure points; areas of wall, or welds, made deliberately weaker than the rest of the tanks so that they'd rupture in known and predictable ways.

that's not to say they'd rupture non-violently, of course, or that such designed failure points will always turn out to work as intended. a few years of bumping and shaking along random road conditions can leave a pressure tank in much different shape than it left the factory, after all. steam boilers have had fusible spots and overpressure valves since the mid-1800's, but still manage to kill people on occasion.

as well, at that kinds of pressure a rupture at any point in the pressure system could be dangerous. what's the design pressure for the engine inlet, i wonder? and where's the regulator situated --- hopefully at the tank itself?

By Nomen Nescio (not verified) on 03 Nov 2009 #permalink

In defense of MDI and Michio Kaku, I'm pretty sure this is a clip from a Discovery Channel show called "Future Cars". It is not marketing and it is (sadly) not (intentional) parody. As far as I'm concerned, all the misinformation is the fault of a moronic "writer" at the Discovery Channel who is more interested in filling air time with gee-whiz technobabble than with learning about and teaching why this tech might actually be interesting.

@30: I would imagine the rocket thing is not only possible, it is *probably* the more likely scenario. My understanding is that if you puncture a cylinder of compressed gas, that's exactly what happens. An explosion requires the pressure to exceed the integrity of the vessel, and I don't know enough about it to know if blunt impact could make that happen suddenly or not.

@MarkCC: Hmm, well, how come there have been all sorts of recorded incidents of punctured air cylinders getting launched like a rocket and resulting in a Really Bad Day for anybody who is in the way?

I'm not concerned about a gradual increase in pressure resulting in rupture; that's an easy problem to solve. I'm concerned about a sudden compromise of the vessel's structural integrity. If you've got a valve that makes sure it doesn't go above 4500 psi, and the container is rated for, say, 9000 psi, that's all fine and good, but what happens when the container gets its warranty voided in 0.1 seconds at 60mph?

Even if you have a release valve that somehow "knows" the structural integrity has been compromised and goes into rapid release mode, there is a limit to how quickly you can release the gas without... turning the cylinder into a missile, which gets us back to square one.

You have an interesting idea about crumple zones/airbags, etc... I suppose we could imagine protecting the cylinder on all sides with airbags, thus ensuring the cylinder never gets enough of a blunt impact to cause it to rupture. I'd still be a little concerned about puncture... Just thinking worst-case scenario, if a gas-powered vehicle involved in an accident gets impaled by something, the damage is still limited to the occupants. If your 4500psi cylinder gets impaled, though, it could potentially go shooting out of the car at high speed...

Nomen makes an interesting point about pre-set rupture points. I suppose if you could design the vessel so that a compromise of structural integrity always resulted in a violent explosion that was directed straight up, that would be a safe way to dissipate the energy...?

The first half of the following is from Beyond Tomorrow, which airs on Discovery's "The Science Channel", and discusses some of the issues mentioned, above.

Notable in that it also lacks the "perpetual motion" comment.


I suspect that the maker had "already developed a generator that runs on compressed air" because their refill business required a big air reservoir but electricity was only available a few hours a day in their locale (a common scenario in India)so the reservoir had to also power a generator making electricity to run valves and controls. It is not uncommon to see marketing people go off on wildly unsupportable flights.

By mike falkoff (not verified) on 03 Nov 2009 #permalink

@34 The Mythbusters already did that test, "puncturing" a scuba tank with a bullet, and did get as a result something more like a rocket than like an explosion.

Maybe if we ask nicely they can try coliding tanks sometime.

The Mythbusters already did that test, "puncturing" a scuba tank with a bullet, and did get as a result something more like a rocket than like an explosion.

Yes, I recall that episode, but I was resisting the temptation to mention it because I know some SBers are really down on the Mythbusters... heh...

The Mythbusters was with a bullet, though, not with blunt impact. (Well, I guess it depends on your definition of blunt, but...) I know if you poke a hole in a cylinder of compressed gas, you get a rocket. I know that if you overpressurize a cylinder of compressed gas (by a LOT) it can rupture and go off like a bomb.

But I'm not really sure what would happen if, say, you put a big dent in the cylinder, or bend it in half until it cracks. It depends on how big the hole gets and how fast. Heh, since you are a Mythbusters fan, think back to the episode about explosive decompression on a plane... A tiny hole = a lot of wind. A big hole all at once = boom.

There's roughly 46 megajoules of energy per kilogram of gasoline


The very best electrical batteries that we've devised have an energy density of just 2.5 million joules per kilogram - a bit under 1/20th the energy density of gasoline!

Since this is a math-oriented blog, I'll nitpick and point out that 2.5 megajoules/kg is a bit over 1/20th the energy density of gasoline.


Wrap the cylinders in semi-airtight kevlar, with a folded blowout section. On cylinder fail (catastrophic rip, puncture, valve failure) the kevlar inflates like an airbag.

If you wanted to be really tricky you could set it up to deliberately do this in a collision, with the vehicle suddenly having airbags on all sides, and a ruptured/useless airtank.

Canmore, Alberta (the town just west of where I live) has a couple of compressed air locomotives on display. These locomotives were used in the local coal mines and "refueled" from a pipe running alongside the the tracks. In a gassy coal mine the only other alternative was horses. When my father was taking Mining Engineering in the late 1940s, he had a summer job driving a horse in a coal mine. The air locomotives look like they were made in the early 20th century. The tanks were made of a bunch of curved pieces of metal bolted together. I think these locomotives were used until the mines closed about 40 years ago.

I am surprised that the efficiency of a compressed air motor is so low since there isn't a lot of waste heat generated. I suspect the 5% figure is for entire process from burning fuel, through compression to the work done by the air motor. Compressing air is inefficient, especially if you do it in a hurry or in a single stage. You wind up heating the gas considerably. This would reduce the amount of energy you could recapture with a regenerative braking system.

By Militant Agnostic (not verified) on 03 Nov 2009 #permalink

I still remember how excited I was as a kid when I came up with the idea of mounting a wind-powered generator on my bicycle, to be attached to an electric motor, so that once I put the bike into motion, the wind generated by the movement of the bike would generate the power needed to keep the bike moving. Damn 2nd law!

By Richard Bell (not verified) on 04 Nov 2009 #permalink

@34 The Mythbusters already did that test, "puncturing" a scuba tank with a bullet, and did get as a result something more like a rocket than like an explosion.

But this wouldn't be a loose tank. It would be attached (presumably) to a 2000 lb car. Even if the tank ruptured, it probably wouldn't move the car much. I'd be more worried about someone in the way of the venting gas.

By Brer Scientist (not verified) on 04 Nov 2009 #permalink

Mark, your description of the third law of thermodynamics is pretty weird. The third law says that as a system's temperature goes to absolute zero, its entropy decreases to some constant (which for some classes of systems is zero).

It's possible, I believe, to coax out of this a theorem of the form "with such-and-such kinds of system, you cannot get the temperature right down to absolute zero by doing such-and-such kinds of things", which then lets you state the three laws as (1) you can't win, you can at best break even, (2) you can only break even at absolute zero, and (3) you can't get to absolute zero.

But I really don't see any way that the third law can possibly be interpreted as saying that the first two laws really are laws, which is basically what you've said it says.


First, the way I described the three laws isn't original - it's something that I picked up somewhere. If I remembered where, I'd cite the original author.

The meaning of that statement of the third law isn't "The first two are really laws". In fact, your restatement contains the kernel of what the silly form of the statement tries to say. The first law says you can't win. The second law says that if you've got any energy at all, you can't even break even. And the third law can be interpreted as saying that you can't create a system with no energy - and since any system with energy is bound by the first two laws, you can't create a system which escapes the limitations of the first two laws.

@38: I suspect the reason for the air-powered generator was actually to be onboard the car, for use powering the onboard electronics.

Ah, a very sensible interpretation.

So, I think all we have to do to get a perpetual motion machine is take a conventional gas-powered vehicle with a particularly efficient alternator, replace the engine with an electric engine (but keep the alternator!) and then run the electrical engine off the same battery the alternator is hooked to... right?

That's pretty much the gas-powered vehicle equivalent of the narrator's moronic idea... heh.

Are any of you guys engineers? Anyone who has mastered basic Thermodynamics would know that compressed air could be used as an energy storage device but to be truly efficient from a volume point of view very high pressures would be needed. Safety is an issue. There have been cases where a commercial tank of high pressure oxygen has been dropped in such a way that the valve end was broken off. The cylinder became a rocket engine capable of penetrating concrete walls. That's why the tanks have a protector sleeve which covers the valve. The whole idea of efficiency involves the details. If, by some means, you could compress very cold air, then, by some means heat the compressed air to a high temperature, you would have a heat engine capable of extracting heat energy from the environment. This would be a heat engine. It would always reject more heat than it converted to work but, if the heat, the energy source, and the cold, the heat sink, could be easily obtained, say a hot spring next to a glacier it might be possible to produce a net work output. Go take a course in thermodynamics.

Assuming I remember correctly, the high pressure tanks are designed to fracture rather than puncture. That is to say, if there was a critical failure of the tank, the tank should rip open along a seam like a pair of tight pants splitting as you bend over to pick up your keys. This is supposed to result in a quick and safe release of the compressed gas.

But yeah, perpetual motion machine...

NB way back when, compressed air was occasionally used to power small locomotives that were used for switching duties inside buildings.