Carlos Barrera geared turbine engine
    Ambri - Liquid Metal Battery for Grid Storage Update

    We’ve been somewhat short on content for a while, but at this moment Mark Goldes and Ken Rauen are coming back to trying to get free work from ambient energy (and of course free money from donations). I don’t particularly want rants on the status of Aesop as replies here – we all know that a lot of money has disappeared without any working product and that the promises to get it right this time thus have a hollow ring, but let’s try to follow the rabbit-hole and see where it leads instead. Mark Goldes replied politely to my comment on PESN about this, so let’s try to keep the replies polite as well. Argue with the ideas, not the people.


    The start-point here is what Aesop Institute says at which is unfortunately coming through Scribd so might not display on some peoples’ screens. Oh well – the gist is that Ken Rauen has thought up a gas-cycle that works using a heat source but no heat sink, and has developed the original 10-piston idea to one that only needs 4 pistons. Also worth looking at is Ken Rauen’s ideas on this from 10 years ago at (or do your own search). Worth noting that it was said to work at that time, and that in the intervening years no-one has verified that it works.

    Let’s start with some concepts of temperature. In a gas, the macroscopic energy stored at a certain temperature depends on the heat capacity of that gas. This heat capacity depends on the translation, rotational and vibrational velocity of the gas molecules, and thus different gases will contain a varying amount of joules of energy per litre when they are the same measured temperature. In a gas-cycle engine, we don’t really need to worry too much about things other than the velocity of the gas molecules and the gas will be the same at all points in the process (no chemical combinations), and we could use a monatomic gas for simplicity. Pressure is just molecules hitting the walls of the container, and they will recoil on average at the same speed that they hit the walls unless the walls are either moving or are hotter/cooler than the gas. At atmospheric pressure, any one molecule in air at STP collides at around 7GHz, so we don’t normally notice the random variations in pressure caused by the individual collisions.The directions are random, and the speeds follow a Boltzmann distribution. Incidentally it’s worth noting that the smaller the microphone and the lower the air-pressure, the more you notice the random variations as white noise – although this makes an effective lower limit on a useful size for a spy microphone before the noise gets too great, this can itself be used as a source of energy from the ambient.

    By Newton’s laws, which are accurate enough at the ~500m/s of a gas molecule at STP (and still apply relativistically though they become a bit more complex), collisions will conserve energy and momentum. Although in reality the walls are also made of atoms and will thus exchange energy with the gas, let’s have an ideal piston and cylinder to start with that does not alter the gas energy in any way – the real situation introduces losses so an ideal container will be the best system possible.

    Compressing the gas puts energy in since the wall is moving and the molecules will rebound with a higher kinetic energy, and we measure this increased kinetic energy as higher temperature and pressure. Similarly we get that energy out again by the expansion of the cylinder and the gas will cool down – all nice and perfect and ideal at the moment and those laws of conservation of energy and momentum will always apply even as we use real-world rather than ideal conditions.

    If you put energy in as heat, then that is shared amongst the gas molecules’ kinetic energy and will increase the measured pressure. You can get this energy out either as joules of heat or as joules of work (or a combination of both) but providing those laws of conservation still hold then in any cycle you can’t get out more than you’ve put in. The important word here is cycle, since any piston engine will run a cycle and will be at particular conditions at any specified point.

    So, can you produce a “cold point” using pistons such that that gives you a flow of heat energy from ambient, and from that flow of energy you can extract the work? To create a “cold spot” you can use adiabatic expansion – where we start at ambient temperature this will cost us work. The amount of work we can then get by using up the flow of energy from ambient to this “cold spot” is exactly the amount of work we’ve just put in – and that’s if it’s all ideal. This is the big Ooops in the idea, and why it can’t work to get energy from the ambient. Note that Ken Rauen built and tested a motor that used the Proell effect a decade ago. He said it worked, but I’ve never seen anyone confirm that. If it worked, it should just keep working (and giving you work out and cooling down) until it wore out. Obviously it didn’t. If it had done, then he would have demonstrated it widely (as would Mark Goldes) and have got a Nobel prize at least. The Proell effect seems to be a non-effect. By building it better with lower friction and better insulation he could make it spin for a longer time before it stopped, but the same can also be said about magnet motors. You can get it arbitrarily close to 100% if you spend enough time and (someone else’s) money, but you can’t get that over-100% out of it.

    To analyse any energy machine, we need to look at what the energy flows are in joules at all points – if we look at temperatures then we can make errors, but count the joules in and out and things become obvious. It also helps if, instead of using equations naked, you put in actual amounts of gas etc. that you want to use so that you have real numbers to work with.

    So far I haven’t mentioned the laws of thermodynamics. I don’t need to – this works on simple mechanics of collisions but just happens to end up agreeing with thermodynamics. What this says is that any idea of trying to use ambient energy (single heat source/sink) in a motor with pistons just isn’t going to work – while momentum is a conserved quantity you can’t do it with pistons or macroscopic structures. In a real situation, it won’t even break even – the losses will stop it dead. I had a discussion with Bill a while back on one of Tesla’s ideas that used a similar proposition of creating a cool sink and then having an almost Carnot-efficiency motor using the difference between that and ambient and thus converting all the input energy to work – again it seems logical until you go work out what happens to the energy flows.

    OK, so that’s knocked out the idea of a piston engine possibly using ambient energy to do useful work for us. Are there sneakier ways? About a year and a half ago I put up one idea here that skews the probabilities of passing a barrier. Nope, I haven’t managed to build one of these and see if it really works or not, though I have thought of a possible way to get the accuracy needed. One of these days…. Since it would do a ridiculously small amount of work and is difficult to make, it’s only value is to annoy the people who say it can’t work and to satisfy myself that it does. OK – if it does….

    Note that if you are a bit creative then the ambient temperature can be split into two sections. One is the ambient conductive energy of the air or other material substance, and the other is the temperature of the sky. Say ambient air temperature is 20°C. I’ve measured blue-sky temperature (using an IR thermometer) at -35°C in summer here. There is a temperature difference there we could exploit with a piston engine with atmosphere as the hot side and a radiator with a parabolic mirror pointed at the sky as the cold side. Clouds, by the way, measure somewhat warmer and could be somewhere round -2°C, but that varies a lot. You’d need a big radiator and mirror to get any reasonable amount of power, and the whole thing will be somewhat large for the power produced. It’ll be free energy from ambient, but would probably cost you more in depreciation and maintenance than using power from the wall.

    Energy moves around. In the course of energy moving from one location to another we can get out some work. Something will move from one place to another, or something gets heated, or something changes shape. The sorts of things we loosely call “using energy”. Since mass/energy is a conserved quantity, of course we can’t use it up. We can’t make it or destroy it, we can only move it from one place to another. At the end of the work you’ve done, exactly the same amount of mass/energy remains as you started with, but it’s in a different configuration (more spread-out, and thus higher entropy). As I’ve said before, there’s a bit of a language problem that affects our thinking on work and energy.

    One important point that isn’t much noticed is that all flows of heat are in two directions (or more). While the hot body will radiate more to the cold body, that cold body is still radiating energy to the hotter one – replace the cold body with a colder one and you can measure the increased heat loss from the hotter body. Thus even at ambient temperature, there must be a heat flow in both directions and thus if you can interrupt just one direction of flow you have the energy-flow needed to get work out. Take a while thinking about that one before you automatically reject it. Without that radiation, the IR thermometers wouldn’t work.

    Let’s say we have a white-hot radiating body and we take some selected spectrum of that suitable for photoelectricity to illuminate a PV. The PV gives us electricity – we know that works. If instead of the PV we had a photoelectric effect that worked on long-wave IR, and again use the right spectrum for it, then again we’d expect it to work. If the PV worked on the IR we get at around 300K, then we could expect an electrical output from that, too, and that’s around ambient temperature. By it’s nature it won’t be a high voltage per layer (of the order of maybe 10mV) but you could have a lot of layers in series. We may not be able to yet make such a semiconductor layer but it should at some point be possible. If it only works at a higher temperature than ambient (say 300°C), it would still be almost 100% efficient at converting heat to electricity which would be worth it. Maybe it’s actually possible to use ambient energy to do work, providing we don’t try to do it using thermodynamics. Just don’t expect to be able to do it with a piston engine and a single heatsink – it needs something sneakier.

    Carlos Barrera geared turbine engine
    Ambri - Liquid Metal Battery for Grid Storage Update