New form of mechanical advantage: Off season research potential

While working with Dave(in the video) we ended up working and finishing the prototype of this Ice piston. Ice expands with a force of ~43500 psi and until now has not been used in machinery. Please enjoy the video and attempt to play with this on your own. If any community can innovate on this, it is this one.

Here is the video in question:

Here is the ice piston cycling:

Reminds me of a wax motor.

Very similar. But we don’t need to use anything more complex or difficult to get a hold of than water and something else that wont freeze(anti freeze is easy to get a hold of). The biggest hope with this is that it becomes adopted by other countries struggling with power creation as something that can be made by a household.

In a country struggling with power creation, where will they get the power to freeze and thaw the ice?

Well for places within the arctic circles should be able to accomplish this easily. For those in more tropical climates I recommend looking into Einstein’s only patent for solar refrigeration. The link is Wikipedia but it does a good job explaining it.

The other place this would be of great use is space, the moon, and any other rock out there without an atmosphere.

You’re going to have a hard time getting me to believe that expanding ice has never been used to purposefully split a rock in a quarry, or do other useful work.

That said, if it’s really easier in some contexts to both freeze/thaw (chill/heat) a medium that expands/contracts, than it is to use more common hydraulic methods to modulate pressure/force, give yourself a pat on the back; and hire me to take care of your pool when you are rich.

Have you seen the video yet? Splitting rocks and concrete was talked about. And as for the context that makes this more viable than common hydraulics? That would be space and anywhere cold. Not to control machinery (Although, it would be amazing to see and could possibly even be better than current methods). but to generate electricity. With that PSI behind it you are looking at using torque instead of speed. Before you tear it to pieces please watch the video, and even try it at home. That is what we want to see, people using this and making it better.

Using a refrigerator, even an Einstein refrigerator (which is only really distinctive in that it has no moving solid parts) to drive this is unlikely to be useful due to the inefficiencies. If you coupled this with a ratchet, you could probably use this in areas with many freeze-thaw cycles during the winter to lift heavy loads a relatively short distance. Switching the “hot/cold” as described in the video is essentially making this into a [strike]Stirling[/strike]Rankine engine, with the working phase change being the freezing and thawing of ice. There are a number of gas/liquid Rankine model engines that can work MUCH more quickly (e.g. the steam turbine).

The big advantage of this system is the very large pressure that can be generated. This pressure is NOT going to be efficiently utilized in a plastic tube, but in a thick-walled steel or even harder metal piston. Further, the high pressure comes at a price - the high heat of latent fusion at the water/ice state change. As I recall from my undergraduate physics, this is about 80 cal/g at standard pressure (1 atmosphere). (Confirmed online - ~ 336 J, which is 80.3 cal). That is, the amount of energy it takes to freeze one thimbleful of water is equal to the amount of energy required to lift 34 kg (about 75#) by 1 meter (39"). I strongly suspect that at the pressures cited in the video, there is a significantly higher latent heat of fusion.

Don’t misunderstand me - I’m not saying that there is no point in pursuing this technology, merely that it is not going to be a magic bullet that will solve the world’s energy problems with a few dozen hours of work. There probably are a few high-force niches where it can be a game-changer.

Nope -I didn’t watch the video. You wrote that the pressure generated by freezing water had never been used in a machine before. Does the video contradict you?

And as for the context that makes this more viable than common hydraulics? That would be space and anywhere cold.
There is no free lunch. In cold places, it will be difficult to prevent premature cooling and/or reheat the medium, in hot places the reverse will be true.

Before you tear it to pieces please watch the video, and even try it at home.
I went to school so that I could answer some applied-physics questions after only hearing them and thinking about them. :wink: After thinking about this, I didn’t tear anything to pieces. If you can come up with a commercially viable way to exploit the volume change water or some other material(s) goes through when it is heated/cooled, I’m serious - Go for it.

You can certainly use this to generate a lot of hydraulic pressure without a hydraulic pump. So it’d be extremely useful in, say, a MacGuyver episode. In fact it probably has already been used in one.

You’re never going to use this for practical power generation, however. The thermodynamics there are pretty clear. As I see it, you have a few problems with this:

  1. It’s inefficient. The Carnot Cycle is provably the most efficient way of turning heat into work. Every other method of turning heat into work is guaranteed to be less efficient. The important equation is:
Eff. = Work /  HeatIn = 1 - (TempCold/TempHot)

Those temperatures are absolute, so Kelvin or Rankine. In the above video, you’re talking about something like 273K (0C) and, at most, 373K. So the best you can do is 27% efficiency. Which isn’t bad, but you have to start with boiling water. On the plus side, the cycle probably gets pretty close to Carnot Efficiency, since the heat transfer is so slow and you’d get such little power out of it.
2. You won’t get much power out of this. You’re not going to get ice to form quickly or melt quickly.
3. You still have to melt the ice afterwards if you want to do anything else. This is not free energy. Space and/or the Arctic can be a pretty big heat sink, but you still have to make heat to have liquid water to start with. If you’re going to all that trouble to melt the water, why not melt it a little more into steam and have yourself a steam engine or stirling engine with your limitless cold as a heatsink? Then you can work higher temperature differentials and likely get better efficiency.

This isn’t to say you can’t do clever things with ice to make power generation more efficient. In hot climates, large buildings often use off-peak ice storage to make air conditioning cheaper. The system will create ice at night when electricity is cheaper and often cleaner*. Then during the day, the ice is used as a very effective heat sink for the air conditioning system.

*Cleaner because less coal-fired, gas-fired, etc. plants that have to be ramped up to cover peak electrical loads.

I think that’s commonly referred to as a steam engine. Or turbine. Or maybe it was a stirling engine…

Candidate application #1:

Antarctica is not a particularly friendly place for man-made structures. Every time a snow storm or blizzard blows through (a single storm can last for weeks at a time), a new layer of snow and ice is deposited on the surface; this causes the rest of the ice underneath to slowly compress and sink, flowing (as any glacier does) downhill towards the ocean.

The obvious difficulties of drilling a foundation all the way to bedrock are therefore trivial by comparison to the immense shear forces that the slowly flowing glacier would impart on any support columns that we could possibly build. Simply put, there is just no way for a research station deep in the heart of the continent to be founded on anything but the glacial ice itself!

This conclusion leaves us with a new problem, however. In addition to the buildup of ice and snow constantly raising the surface of the glacier relative to the foundation we built, we also observe that any research station we can build will slowly melt the ice beneath its foundation by exerting significant pressure and conducting heat from the occupant quarters. We can minimize all three effects with careful engineering (for example, a station in the shape of an airfoil can accelerate the air moving over/under it, thereby discouraging the buildup of snow), but the fact remains that unless we do something about it, our research station will inevitably sink into the ice and become buried as time goes on.

How do we solve this problem, then?


Every time we feel that our building has sunk too deep into the ice, we lift the entire thing.*

This is currently implemented in a variety of ways at different facilities, but there’s no reason why we couldn’t just use the outside temperature to do the work for us. A ratcheting piston could easily be designed that requires nothing more than a source of liquid water to function; since the production of heat is already the primary purpose of an Antarctic station’s generators, this would be similarly straightforward to implement in practice. The process might go something like this:

  1. The valves on all the lift pistons are opened.
  2. Each piston is already full of ice from the last time the station was raised, so this ice is either removed altogether or melted with electric heating elements (pistons must be designed to accommodate the chosen method).
  3. Each piston is topped off with liquid water (about 4 degrees Celsius would be ideal).
  4. All piston valves are closed back up.
  5. Let the Antarctic temperatures do their thing; the ice will proceed to freeze and expand, lifting the entire research station above it.
  6. …?
  7. PROFIT!

*For more information about the challenges of South Pole architecture, check this out:

It doesn’t thaw enough in Antarctica to get much “free” work from the freeze/thaw cycle. If you’re going to spend enough energy to melt a lot of ice from Antarctic temperatures, it would probably be easier to make the whole research station light enough that it is positively buoyant in ice, and has a cylindrical cross section (not necessarily circular, just all faces vertical). This could probably be done with lightweight concrete as the filler and insulator. First of all, this would slow the rate of settlement. Then if you do need a correction, melt a thin layer around the station and float upwards or downwards as appropriate.

I applaud your enthusiasm and thinking-without-boundaries.

But… On my grandfather’s farm, my father and I used a sturdy, modest-size, non-hydraulic jack once. My father called it a “railroad jack”.

We wanted to use it to straighten the frame of a car, but instead we lifted the second floor of the building we were trying to use as a brace for the jack to push against. Oops!

That was a simple, portable, one-person, mechanical device. I suppose we could have used ice instead of that jack, but using ice would have been a huge pain.

I think the folks in the Antarctic might prefer to use an Antarctic version of a railroad jack (plus some shims) to lift heavy loads.


I believe in the first season, he uses the principle to escape from a locked freezer. He melted ice with the lightbulb and guided the trickling stream into the lock. When the water refroze, the pressure of expansion broke the lock.
Still waiting on the Mythbusters to test it.

This phenomenon could be helpful in certain circumstances. There are more obscure and inefficient machines that have found their niche.