(updated 26 Feb and 9th Mar 2017) Since Mark decided to repeat a previous article in this series, it’s maybe time to put the whole series into one article here. If you want a shorter version, it’s maybe better to look at http://revolution-green.com/back-to-work-again/ first and then come back here to get the historical stuff. There are a lot of words here to get through, after all. I have been somewhat surprised at the lack of response overall, though. After all the fraudulent claims of overunity and Perpetual Motion we’ve covered here, I put the logic of how to make a real Perpetual Motion system up and some examples of how to design devices that actually do this around 9 months ago and it seems no-one experimenting in Free Energy noticed that it does all they wish their (non-working) machines should do. I still get people asking about buoyancy systems and magnetic motors, and if you understand the physics here and how Energy moves around and how Work is done then it’s very easy to show that nearly all the “suppressed technologies” can’t work. There are a couple of “lost technologies” that may have used these principles, in particular the Manelas device which was said to deliver 60W or so and was seen to be around 5°C below ambient temperature.
Put in a nutshell, nearly all the systems we use to produce Work do that by a movement of energy from one place to another, since that’s how we know we can get Work done. We measure the Work in newton-metres, a force times a distance, so we are predisposed by the language to think that way. This linear movement can only ever deliver a set amount of work for a set amount of energy that moves, and that observation is encapsulated in the 2nd Law of Thermodynamics.
It is however possible to make the energy travel in a loop, and then we can deliver an infinite amount of work for the same set amount of energy, providing the work done does not store energy.
If you understand that sentence, then there’s no need to read any further. The rest of the explanations here merely add some meat to the bones and say the same thing in different ways in the hope of finding the explanation that makes sense to you, the reader.
There is a lot of energy available in ambient temperatures so, even with work that stores energy, we can produce such work without needing fuel of any type. This energy-loop is conceptually quite simple to set up, as well, but does need some fabrication techniques that are *difficult* to do in a home workshop and where the kit is expensive to buy.
Over the past few years my time in the workshop was diminished because there was a need to care for my mum, whose Alzheimer’s disease got gradually worse. The last year was pretty bad. For this reason, I couldn’t put this theory into practice. In the meantime, I put the theory up in case any of the Free Energy experimenters wanted to make something that actually worked for a change. This year, however, I will have the time and a lot of help in getting the fabrication done, and we should thus see Free Work from a practical device. There are of course some others working along the same lines, and so we may see several alternate systems this year. I’m not telling you my design here, but instead the principles required, since there are diverse ways of achieving a working design and your idea may be better than mine.
I was intending to use this dissertation to apply for a DPhil, but it seems that there are residence requirements and that I’d have to take the course (and there isn’t a course because this is heretical philosophy) so I can’t annoy academia by a direct attack on the sacred cow of 2LoT. As such, I’ve just sent the document to a few people who can use it and I’ll publish it here. Despite being somewhat heretical, there have been no visits from MiB or others trying to suppress the ideas, and most of this has been on R-G for quite a while without any government reaction.
Thanks to Asterix, who pointed out an inadequate explanation of one point. This is explained by applying Conservation of Energy considerations. Mike Frost also engaged in the conversation, though maybe didn’t get the definitions I stated. Ken Rauen contributed to the thinking here and even though his engine ideas will not perform as he calculates, his intuition was correct (but joining with Mark Goldes was a baaad idea). To get something that works, you need to get both the maths and the physics correct, and his design doesn’t loop energy. Phil Hardcastle’s Sebithenco idea, using a standard thermionic valve (tube, for the US speakers), produces a small but verifiable amount of power from a single heat-sink in an experiment that is easy to replicate. To develop a new idea, it’s necessary to stand up and say what you think, and then see what the sceptical response is. Most of the time, the sceptics will be right and point out something that’s been missed, but sometimes the objections are not valid. Discussions are essential, so thanks to those that actually discussed the idea rather than dismissed it.
Thanks also to Abd, who demonstrated that I haven’t explained the basics and definitions well-enough for someone who hasn’t been immersed in 2LoT as long as I have. Since I didn’t define what work is normally understood to be, my redefinitions of the subdivisions of work didn’t make sense as to why I was doing this. I also didn’t define how we normally use a heat engine to get work out of a difference in temperature, so the point about using only a single heat-sink, and why this is important, is too easy to miss. I have a tendency to skip over intermediate steps in logic that seem obvious to me, and I’ve thus added in some extra paragraphs to cover the steps I realise I’ve missed. There may still be others I haven’t noticed, so if I’m made aware of non-sequiturs I’ll try to fix them.
OK, that’s enough foreword, and let’s get to the explanations themselves.
A dissertation on Work, Energy, and Perpetual Motion
“But if your theory is found to be against the second law of thermodynamics I can give you no hope; there is nothing for it but to collapse in deepest humiliation. ”
— Sir Arthur Stanley Eddington, The Nature of the Physical World (1927)
This work is a summation of essays published on the net over the last few years, on a website (Revolution-Green.com) that started by debunking claims for “Free Energy” machines. Seeing all the frauds made me want an easy way to show absolutely whether such a machine would work or not, and in the process I examined the confusions between energy, work and power in common parlance, and often in more scientific discussions too. I also saw the blind-spot in the scientific consensus, since if there is a wave there is energy flowing, and a flow of energy is necessary to do work – energy in the common meaning is in fact a relative term. Why then could we not use the energy-flows in the environment to do work without needing to burn fuel? Though this appears to be forbidden by the 2nd Law of Thermodynamics in a lot of circumstances, we have a flow of energy and there seems to be no bar to doing so except for that Law. I needed to explain the concepts in a way that ordinary people should be able to understand, and thus stop wasting their time trying to recreate some “suppressed” Perpetual Motion machine that never worked. In the course of that, I found that a Perpetual Motion machine was not actually forbidden by nature itself, but was simply postulated as impossible because no-one had managed to build one that worked. This is logically indefensible. Understanding the reasons that energy moves around and what work is enables the design of a real device, and I shall endeavour here to explain the logic and show a practical and proven design. Since this is basically a work of philosophy there aren’t many equations, and because of the original audience there are a lot more words than absolutely needed, with repetitions now and again. There’s not really much need for the glossary or appendix, either, since I would expect all people reading this text to know all the basics or know how to search for them.
We’ll start with the paradox. There is the principle of Conservation of Mass/Energy, and since mass and energy are equivalent via E=MC² then in a closed universe we can’t change the amount of “stuff” we’ve got (mass/energy) but whatever we do we will end up with no more and no less. If we then look at a normal physics or engineering explanation of a process, we are told that we start with X joules of energy and we can do a little less than X joules of work, and then we have no energy left. This obviously doesn’t agree with the aforementioned CoE (Conservation of Energy). We really have a semantics problem, since the words we are using are not adequate. This will shortly be addressed, but before that I need to point out a misconception in our models for thermodynamics that, if it were true, would itself make such perpetual motion impossible.
2nd Law of Thermodynamics (2LoT)
Clausius statement: Heat can never pass from a colder to a warmer body without some other change, connected therewith, occurring at the same time.
Kelvin statement: It is impossible, by means of inanimate material agency, to derive mechanical effect from any portion of matter by cooling it below the temperature of the coldest of the surrounding objects.
Planck’s proposition: It is impossible to construct an engine which will work in a complete cycle, and produce no effect except the raising of a weight and cooling of a heat reservoir.
Back in 1824, Sadi Carnot worked out the start of the theory of thermodynamics that we still use (because it works for classical devices). He regarded heat as a fluid, that will flow from hotter to colder and not the other way. That unidirectionality of the flow of heat is critical to how we think of thermodynamics to this day, but it is not an accurate model.
Carnot envisioned a heat engine that took heat energy from a hotter heat-sink and, by using alternate adiabatic (where no heat is transferred) and isothermal (where heat is transferred) changes of volume and pressure of a gas, the engine delivered work equivalent to the difference in temperature between the hot heat-sink and a cooler heat-sink. In this way, an idealised Carnot engine would translate all the heat-energy difference between the two heat-sinks into available mechanical work. Although this ideal engine cannot be physically realised, since real devices have losses, all subsequent devices have been considered as having a hot sink and a cold sink and can be compared to this “Carnot efficiency” to determine their departure from the ideal. Devices that use a single heat-sink only are not considered as possible, since there’s nowhere for the heat to go to. Though such devices exist and work, analysis of them is performed as if they have two heat-sinks available, and so the loophole in 2LoT is never seen.
These days we have more understanding of atomic theory and that heat is actually just random kinetic energy of molecules. See https://en.wikipedia.org/wiki/Kinetic_theory_of_gases for a good explanation, and see https://en.wikipedia.org/wiki/Stefan%E2%80%93Boltzmann_law for what happens with radiated heat. When we have real molecules exchanging momentum/energy in collisions, it should be pretty obvious that (a) regarding heat as a fluid is an approximation and (b) heat is not unidirectional from hot to cold but instead multi-directional and a colder body can and does radiate/conduct heat to a hotter body. The large number of random transactions in any reasonably-sized system means that the net flow of energy is always from hotter to colder, of course, but if you replace the colder body with one that is colder still, you’ll see that the hotter body cools more quickly, and that is because it’s not receiving as much heat energy from the colder body. It does not (and cannot) “know” what other bodies are in its vicinity to radiate to and cannot thus adjust the amount of energy it is radiating to make the calculations give the right answer. That would imply an awareness of what is receiving the heat that is radiated and would of course also break causality since those other bodies could be very far away.
It’s thus time to discard that idea of heat as a fluid and use a model that corresponds to reality. In reality, if we have two bodies of the same temperature (or a system in thermal equilibrium) then we can easily calculate how much energy is passing between those bodies. The Stefan-Boltzmann law gives us the radiated emissions from each body, and if there is a conductive path as well then it should be obvious that the amount of energy conducted out of a body is related to its absolute temperature, and that Newton’s law of conducted heat simply gives us the net flow by looking at the difference in temperature. If we have a flow of energy, then we should be able to get work out of that flow, providing we use the correct techniques. All that’s stopping us doing this is a semantic problem and a belief that 2LoT can never be broken – and of course that the device that actually does this is somewhat difficult to make. Later on in this dissertation I shall show that such devices have been made and do function, and that the 2LoT thus has a useful (and proven) loophole which will give us usable power from a single heat-sink (the environment) which can reasonably be regarded as Perpetual Motion of the second type, given how large that heat-sink is.
When a body is in thermal equilibrium with its environment it is still radiating and conducting energy away at the same rate, but its temperature does not change because it is receiving an equal amount of energy from the environment. Above absolute zero, therefore, there is always an energy flow available. Whereas with conduction it would be difficult to tap this energy flow since it is only visible and available at the atomic scale, with photons we have at least two methods of converting these to electricity. With electricity, we can do work.
Since a body above absolute zero radiates heat (Stefan-Boltzmann rule) and we can intercept that radiation and convert it to electricity, it should be obvious that though the Clausius statement is invalid in that heat is radiated from a colder body to a warmer body without anything else happening, it doesn’t itself say much about what happens and it is the interpretations of it that actually are inadequate since they only consider heat energy travelling between two heatsinks from hotter to colder. The Kelvin statement is falsified, and Planck’s proposition can be shown by experiment to be wrong. A single heat-sink will radiate energy, and we can get work from that radiation even if it is the coldest body in the local group. Further down in this text, we’ll see some practical methods for doing this.
The semantics problem
One of the big problems in this field is that the language we use is not precise enough, and Energy and Work (which have the same units) are often used interchangeably. I’ll thus start by defining the language I’ll be using here, since a degree of pedantry is needed.
I’m going to separate the words out. We have two forms of energy here (kinetic and potential) and we also have work. Potential energy (PE) includes mass, energy that is stored in springs of some sort, gravitational potential (may be stored as mass, but I’ll skip that question for now), etc.. Kinetic energy (KE) is stored as things that are moving, so we include photons, moving masses and other unbound energy not stored as mass. Work is on the other hand a bit trickier. Whenever we do work, at the end of it things are just in a different configuration than before. Some work goes into kinetic energy and/or potential energy, some of it is simply that a lump of stuff is a different shape or location. Hammering a lump of iron into a sword (or ploughshare) takes a lot of work, but at the end of it we have the same lump of iron in a different shape. Work is not a conserved quantity, and that is an important observation that is central to this thesis. It is so important that I’ll add it in bold:
Work is not a conserved quantity.
Generally, when we’re talking of how much energy we’ve got, we don’t look at the whole amount available, since that would produce ridiculous numbers when we count up the number of joules of mass/energy in the various bits of stuff we’re looking at. Instead, what we look at is a local excess of energy over another place (generally the environment) and call this what we’ve got. The total energy in a litre of Diesel is massive, but we look at what we get from combustion of it (that converts a very small quantity of mass into KE in the form of heat) and how much more heat we have than the ambient temperature. To convert that heat energy into work we need to let that excess KE move into the ambient and we harness that movement to do the work of moving our car from one place to another. Since we start from stationary and end stationary (and normally end up in the same parking-spot at the end of the day), in fact all that KE we’ve liberated by burning ends up as heat in the atmosphere and we’ve actually done no work at all, not even Displacement Work. It gets a bit complex when you really follow where all the energy goes to and what has really happened. The simple statement of “we had 10 joules of energy and performed 9 joules of work with it” is imprecise. We need some words that specify where that energy actually went.
I’ll try to restate all this a bit more simply, though. The sum of KE (energy of movement) and PE (energy stored as easily-convertible mass) remains constant no matter what you do. We get work done through the movement of energy from one place/body to another. Energy is conserved, but work is not , even though we use the same units to measure it. The work that is done can be divided three ways into KEWork (where the energy is stored as kinetic energy), PEWork (where the energy is stored as potential energy, which is some form of mass), and Displacement Work where no energy is stored in the work done and can thus not be recovered either.
KE is kinetic energy, which is energy of movement.
PE is potential energy, which is stored in a way that can be released. I see the energy stored as mass since that is the only available store.
KEWork is work that is done where the energy goes into kinetic energy of something.
PEWork is work that is done where the energy is stored in some way as potential energy in the system.
Displacement Work results in a different configuration of the system, but does not store any energy nor use energy to do it.
A real system that turns available PE or KE into work will normally produce all 3 types of work in varying proportions. In the course of this the energy will move, and it is from this movement of energy that we get the work done. It’s worth pointing out that what we see and measure are frame-dependent and that observers in different frames of reference will disagree on the absolute values but will agree on the amount by which the values have changed. It is thus the changes that are important here, though normally we’ll agree on the absolute values too since we’ll be in the same frame of reference.
KEWork is actually just KE that is on the output side of the process we’re considering, and so is the same as KE. Ditto for PE and PEWork. Whether we call it energy or work is a matter of perspective. However, when we hammer that lump of iron, the KE that we put in all ends up as heat energy in the environment, which is also KE. We’ve just changed unidirectional KE into random-direction KE, and we normally call this energy “losses”. We can’t however lose energy, since it’s still there, and it’s simply in a form that we previously didn’t know how to re-use.
If you burn fuel to get that energy excess, then allow that excess to go to the environment through a heat engine, and then do Displacement Work with it, then all that fuel energy goes into heat in the environment, which is random directions of KE (so in fact the energy from the burnt fuel translates into KEWork, and that energy doesn’t actually disappear). Since this is in general low-grade heat, we have no heat-engines that will use that slight difference of energy from the environment to do any more work with it. Since we are taught that we can only get work out from a difference of temperature, we don’t see that the temperature itself is a store of energy that can be utilised, providing we have a method using only one heat-sink.
Work is the output energy of a process, when the configuration of mass/energy is different than before we started the process. Energy will naturally move from a higher concentration to a lower one, and not the other way round. If we want to make a concentration of energy, we have to do work to move it, and of course since work is energy then we’re just using an even higher level of directional energy to do that moving. This seems an intractable problem since we always need a higher level of energy in order to increase the density of a lower level, which is why so far we use the conversion of PE into KE (or mass to energy, in other words) to power our society. This can give us the higher density of unidirectional energy that we need to get work done.
A local concentration of kinetic energy will naturally spread out until the energy density is even and without any local high concentrations. A hotspot in any object will spread out until the temperature becomes even throughout the object, given enough time. Potential energy will also head towards the lowest possible level, releasing kinetic energy in the process. This can be seen using water – pour a glass of it into a bowl and you end up with a flat surface where no point is higher than another. Pour the glass of water through a turbine and you can get work done in the process, but since you may choose not do this the work obtained can be anywhere from zero to (almost) the available excess energy when the glassful gets poured. This natural spreading of energy is because if we add energy in one place in a container (say as hot gas) then each hot molecule will undergo a random walk that will, over time, give an equal probability of being in all places it can physically get to. Much the same using photons, except that they do not appear to collide with each other but can be reflected by mirrors, so may not be as evenly distributed. After enough time, though, the absorption and re-emission of photons in random directions will still result in an even distribution of energy.
Part of Quantum Theory says that there is a residual amount of kinetic energy left even at absolute zero, and that things will thus still move. Some people think therefore that this Zero-Point Energy (ZPE) should be able to be tapped. There’s some logic there, in that if something is moving then we should be able to make it do some work, but in the case of ZPE I suspect it’s a problem of how we measure things, and that that that energy is not actually available, but is instead imaginary. Just because we measure something to be in a slightly different position does not mean that it’s actually moved, but that since our measurements require us to use some sort of particle to hit the thing we can’t really be totally sure of the measurements. There’s an underlying uncertainty of where the fundamental particle actually is, which is a probability function. ZPE therefore seems to me extremely unlikely to be a source of new energy and thus to break CoE.
Since I’ve also pointed out that work can be subdivided into stored KE, stored PE and displacement, and it should be pretty obvious that a simple displacement is a zero-energy transaction, there is however a chance that we can get the displacement-type work (which is often what we want) for free. That displacement is a zero-energy transaction has been known since Newton, since he said that a body in motion would continue in that motion unless there was a force acting on it. So – if you lift something up you’ve put in work which is stored as gravitational potential energy, and if you let it down again you can get that energy out again as work, but if you move it sideways then no work is done except against friction, and that can be reduced arbitrarily asymptotic to zero. If you watch a pendulum swinging, you see gravity doing work accelerating the pendulum, and then in slowing it again, and yet over a cycle very little work is actually done, and that work is in heating the air as the stored energy in the system dissipates. This is an example of Displacement Work that can easily be analysed as not needing any energy to perform in loss-free conditions. It also highlights the need to look at a complete cycle in a cyclic system, in order that we don’t think work is being produced when it isn’t.
Still, what is work, really? We can define it as a force times a distance, or as a second’s worth of voltage times amps, and the unit of work is the joule in SI units. What is really happening is that our initial store of energy as PE or KE is divided up into an output amount of PE or KE that may be in different locations and may be associated with different objects than it was at the start of the process. What has changed is the configuration of the energy. Some of the initial energy may have gone into a form where we’ve had problems before in re-using it, such as heat from friction that we normally dissipate into the environment in order to stop the device from overheating. KEWork and PEWork are thus no different from KE and PE, and we only give them a different name in order to specify that they are the output side of what we are doing. Displacement work is just a shuffle of the positions of things, and would not be in the definition of work at all except that it’s normally the desired result of using (often stated as expending) energy – we want that sword to be re-shaped in order to plough the land. Language can be a bit difficult, when the roots of it go back so far. Hammering that lump of steel into the right shape will bring on a sweat, and you can’t deny that the blacksmith is working hard, but the end result has no more energy embodied than the initial lump of steel.
In looking at a purported Free Energy (or perpetual-motion) machine we should thus look at the total interplay over a cycle between KE, PE and the three kinds of work (Stored KE, stored PE, displacement). Losses almost always end as heat, which is KE. Count the joules in and out and decide how to assign them in those 5 buckets. All the historical examples I’ve looked at show no extra joules out than was put in, with the illusion being created by hidden power sources if it appears to work at first look. Conservation of mass/energy seems to be absolutely true with no exceptions.
We know that solar panels work well. We also know that if we have any cyclic motion or wave then we can rectify it and get some work done – the energy in that wave gets diverted into a path of our choice and released elsewhere after having done some work. One thing to remember is that the total energy doesn’t change – if we take some in one place it has to turn up in another. All we need to look for is a flow of energy that exists and to divert it so that it does what we want done before we let it go again. Displacement-type work does not take energy to perform, though there is an element of borrowing and returning from the energy bank. Energy is put in to start the motion, then the motion continues until we stop it and get the energy out again. That sword-to-ploughshare change is displacement-type work, in that no excess energy remains after the work has been done. The vast majority of what is normally regarded as work is displacement, where at the end of a cycle there is no work (KEwork or PEwork) actually done. Displacement work does not store any energy, and we should therefore be able to do it without using a movement of energy – all the work we do put in ends up dissipated into the random KE of heat. At least we should be able to do it with a draw on the energy bank followed by a return of that energy.
At this point, we’ve shown that work is done when we harness a movement of energy, normally from one place to another where we can quantify the energy more easily but actually it’s just the movement itself that is important. We’ve shown that energy is not used up by doing work. We have also shown that there are such energy flows even in a thermal equilibrium situation. If we can see an energy flow and have the means to harness it to do work, it doesn’t matter whether the source of that flow is a higher or lower temperature than our equipment – this simply is not relevant. Work cannot be done by something in the past or future, or at a distance, but can only be done by a movement of energy here and now. At its heart, we see that though Displacement Work is simply a different configuration of where mass and energy are, KEWork is the same thing as Kinetic Energy and PEWork is the same thing as Potential Energy. Work (both KEWork and PEWork) is therefore simply a transformation of that mass/energy into another form and must have a distinct place and time for that transformation in order that mass/energy is conserved, whereas Displacement Work does not involve such a transformation and does not require energy to do it. Any flow of energy can be diverted to do work (of any type) for us if we use the right method, and it does not matter where that energy comes from or is going to but simply that the energy flow is available at the place and time where the work is done.
The majority of the “energy sources” and devices we normally use start by releasing PE from a fuel of some sort. This then creates an energy-flow to the environment that is then utilised to perform the work we want done, using a heat-engine that requires two heatsinks to function. Renewables utilise an existent flow of energy in the natural world, and divert this energy for our use without needing us to burn fuel to create the energy-flow. I have shown here that there is in fact an energy-flow available even in a system in thermal equilibrium, that we haven’t noticed because the net flow is zero. The 2LoT tells us that we can’t utilise this energy flow because the net flow is zero, yet we already have devices that can utilise this flow and produce electricity from it. We just need to recognise what is happening and improve the delivery of power from real devices.
The stage is set for the logic showing the loophole in 2LoT, and therefore why Eddington’s pronouncement from 90 years ago is demonstrably incorrect.
The logical case for Perpetual Motion
We now need to build the logical case for how to bypass 2LoT and to get usable work from ambient temperatures. This is set up as numbered assumptions and observations, which all need to be correct in order that the deductions in the conclusions are true.
1: E=mc² applies, and energy and mass are thus different forms of the same stuff.
2: Conservation of Mass/Energy universally applies, so the amount of mass/energy we have is constant. This also implies that ZPE is not available, and that we can only juggle with the energy that is available. No Free Energy, in other words.
3: Conservation of Momentum universally applies. Even if we can’t see it, every action has an equal and opposite reaction. This is important when considering heat conduction. This may be not totally true at very low accelerations where momentum may be effectively quantised, but for normal situations the deviations are calculated to be too small to be measurable. See http://physicsfromtheedge.blogspot.fr/ for a mind-blowing theory that may well be truth.
4: Work is not a conserved quantity. This can be seen in action in any office….
5: There is a hidden assumption of causality, in that the cause always happens before the effect. Although in quantum theory this may be slightly violated, the timescales are very short and related to the uncertainty in measuring exactly where something is when something happens. Such violations will not affect the somewhat larger systems we need for a real-life system.
1: Kinetic energy is energy of motion. Heat is simply random directionality KE. It will spread out to fill the available volume and tend towards an even distribution. We cannot totally confine it or stop it from heading towards that even distribution (though we can slow it down). A system of bodies at varying temperatures will change in the direction of having everything at the same temperature in thermal equilibrium. When such a set of bodies is in thermal equilibrium, each must be receiving exactly as much energy as it is putting out as radiation and conduction. The Stefan-Bolzmann law tells us that each body will be radiating energy no matter what the environmental temperature is.
2: Potential energy is stored as mass, and will tend toward a local minimum with a release of kinetic energy.
3: Matter is particulate – there is a scale at which this must be taken into account in the calculations. At a large enough scale (practically all mechanical devices we can make in the workshop with manual tools) we can consider matter as continuous because the numbers of atoms are so large that statistical calculations are closer to exact than we can physically measure. At a small-enough scale we are instead dealing with individual energy transactions and statistical mathematics no longer applies (and will give the wrong answers). At atomic scales, Newton’s laws will apply and within our frame of reference a slower and less-energetic particle can lose energy to a more-energetic particle. You can set up a demonstration of this on a snooker-table, with a fast ball down the length and a slow ball from the side. If you get the timing right, the slow ball will stop and the fast ball will gain speed as it is deflected.
4: Thermodynamics regards heat as a fluid that only flows from a hotter body to a colder body. This is indeed what we measure to happen, but it is not a correct model. This is a basic flaw in the derivation of the theory of thermodynamics, and is shown to be wrong by the Stefan-Boltzmann rules for radiation which show that every body above absolute zero will radiate according to the fourth power of its temperature. Two bodies in thermal equilibrium with each other must be radiating energy, and must therefore each be receiving as much energy as they are radiating. Radiated heat thus is always bidirectional (in fact omnidirectional) in its nature.
5: Conservation of Momentum also implies that conducted heat is also bidirectional in its nature. Heat will only pass in one direction when one of the bodies is at absolute zero temperature, and real matter will not achieve absolute zero. I can’t however think of a way to use the logical energy flows in heat conduction to produce a usable unidirectional energy flow external to the body. CoM thus makes macroscopic mechanical gas-based engines (and similar ideas) subject to 2LoT, unable to exceed Carnot efficiency.
6: Work can be performed only by harnessing the movement of energy from one place to another. Energy may however move without performing work. The amount of energy that moves defines the maximum amount of work that may be performed. If 1 joule of energy is moving from one location to another, then up to 1 joule of work may be done between those two locations. To get continuous work done, you need a continuous flow of energy. Classically, to produce a local concentration of energy, you need to either burn fuel which releases mass as energy, use an available energy-flow such as solar power to collect energy, or do work on the energy that is there to push it where you need it. 1 joule of work will produce at maximum 1 joule of local excess energy (from which you can at maximum extract 1 joule of work subsequently). This is why a perpetual motion machine is considered impossible. Kinetic energy of heat naturally disperses rather than concentrates itself, so the net heat flow will not naturally move from a colder to a hotter body, although we have shown already that there is an energy flow in both directions.
7: Thermodynamics does not apply to electricity, but only to the heat that is transferred. Converting the kinetic energy of heat (either as photons or as velocity of molecules) into electricity removes it from the thermodynamic equations. This is maybe the most contentious observation, but a photon incident on a photoelectric layer in a photovoltaic cell (PV) is converted 100% to electricity since energy is conserved (though note that excess photon energy over the PV voltage is normally converted to heat in the PV). The percentage of conversions is a statistical problem, but if a photon is converted then all of its energy becomes electrical – we’re dealing with a single energy transaction at a time, even though we’re also dealing with a large number of them. If that harvested energy is transferred down the wires to a load, then by Conservation of Energy it must be removed from the thermodynamic equations of the PV cell. Where heat-energy is converted into electrical energy, and that electrical energy is taken down some wire to somewhere else, that heat-energy disappears from the system which thus becomes colder than it was.
Similarly, even though it’s not quantised as far as I know, the kinetic energy of a molecule may be converted partially to electricity by a piezo crystal or a small-enough mechanical conversion, and the electricity thus produced may be used elsewhere. From an energy viewpoint, we are simply diverting some of that incident energy along a different path. This mechanical KE conversion system cannot be 100% effective since the incident molecule needs to be cleared from the receptor area – if you stop it dead it’s going to just sit there in the way. Once that bit of random heat energy is converted to electrical energy, which is not random but has a direction (so statistical mechanics no longer applies), we can direct it to wherever we want using wires and can do work with it, provided we take it away immediately so that the reverse reaction cannot occur. For most work, that directed electrical energy will get turned back into random heat energy again, but it may be stored in a battery or in lifting weights, compressing springs etc.. After a while, we’ll use that stored energy and it will again probably end up as heat in the environment.
8: The natural state for any body is to be at absolute zero. If it has more energy than this it will either radiate it away or conduct it away because such random kinetic energy will spread out. The only thing that stops the body reaching absolute zero is the inflow of heat from other bodies that are also getting rid of their heat at the rate appropriate to their temperature and composition. This is maybe the most startling observation, but can be seen to be true and that the Stefan-Boltzmann relationship states that this will happen (and also gives you the ability to say exactly at what rate). It should take no work to allow a body to cool down, but should deliver work when doing so. The work we put into refrigeration is effectively wasted.
9: When we convert the KE of a photon or molecule into electricity, we take that energy out of the local environment and move it along wires to somewhere else where we can use it to do work. In this case, the PV or other device must appear to be colder than the environment since the heat or energy that goes in is not re-radiated from the local space but reappears elsewhere. Like a cold body, a larger amount of heat goes into the device than comes out of it as heat, with the balance as electricity that is used elsewhere. This cooling effect is of course measurable. The reason that PVs and nantenna arrays have not in general been tested for this cooling effect is that it is obvious that it should happen from Conservation of Energy, yet nevertheless it hasn’t been remarked upon that this is against 2LoT. Maybe that quote from Eddington has something to do with this omission.
Assume we have a PV cell in the sunlight, and that we are not drawing any current from it. Assume the air temperature is 20°C and that we measure the PV temperature as 40°C, and that we have incident sunlight at 1kW/m² and that the PV is 20% efficient (so outputs 200W/m²). If we start to draw that 200W/m² from the PV, then its temperature will drop by 4°C to 36°C. This is an easy test of the cooling-power of a PV. Of course, I’ve made some approximations here and you won’t get exactly these figures, but it helps to put a number on things. Here, the PV is above the air temperature because it cannot convert all the LWIR it receives, so that is converted to heat in the PV as is the balance of the photon energy above the bandgap. If we used a different band-gap voltage for the PV then the balance will change.
If your logical box extends to enclose both the source of the energy and where it is used, then 2LoT is satisfied, but if you only consider either the location of the conversion or the location of the use of that energy you instead will see energy disappearing from the system or appearing where the work is done. The wires produce a physical distance between the cause and the effect, and of course storage of that energy can put a distance in time between generation and use of the energy. Overall, the laws are still applicable if we make the logical box big enough in time and space (so this isn’t really breaking 2LoT) but locally we have produced useful work without needing any fuel.
1: For macroscopic mechanical devices, whose working dimensions are large relative to either the atoms of which they are made or to the working fluid used, the statistics of random collisions mean that the net energy flow will always be from the hotter to the cooler and thus that 2LoT will apply. We thus can’t get any Free Work out of them, and real machines will only do a bit less work than we put in to start them because there are always losses. This eliminates most of the traditional Free Energy ideas (and the majority of new designs too) as viable ways of doing the job of giving us Free Work.
2: If we utilise the quantum properties of photons, where if an interaction happens (photoelectric effect or antenna) then the photon energy is converted totally or partially to electrical energy, then we can convert heat or light directly to electricity. We do not need two heat-sinks for this, since a body naturally radiates and so we simply intercept this flow. An example from real life is a solar panel. These can be driven from LEDs (at any physical temperature, and that can be below the temperature of the PV) as well as sunlight, and then it should be obvious that this is not a process that is covered by either thermodynamics in general or 2LoT in particular. It is a quantum process. It takes in random-direction KE and outputs unidirectional KE.
We can also use a rectenna to convert the THz region of the IR spectrum into available electricity, using the wave-nature of photons. These have been manufactured and tested. Although when we leave the nantenna open-circuit it will be in thermal equilibrium with its environment, once we start to draw current it will appear colder than the environment. Heat from the environment will thus flow in, and flow out again as electricity to drive a load. When used in the load, it will immediately or eventually be returned to the same environment it came from, to be harvested again by the nantenna. The same energy is recycled again and again to give continuous work output as long as the devices used do not wear out. This is a Perpetual Motion system of the second kind.
3: We should be able to make a mechanical device that similarly converts gas-pressure to electricity if we make its scale comparable to the mean free path of the gas we are using. For air at STP this means around 0.07 micron in size and capable of resolving impacts in excess of 7GHz. A section of piezo-electric material could be used, since a diaphragm and coil arrangement would require very small feature-sizes. A MIM diode could be fabricated to rectify the signal produced. Difficult but not impossible using modern fabrication techniques, and we can always change the gas used to use a somewhat larger feature size. Each collision of a gas molecule would result in a small electric charge being produced and taken out along the wires. We can reasonably expect to harvest in excess of 1kW/m² from such a device using atmospheric pressure and temperature (ambient energy), given that around 10 times this is available. Note that we calculate the energy in a volume of gas as PV (pressure times volume), and that that energy is actually available to use to do work, with the right conversion. Changing the scale of the machine to somewhere comparable to the mean free path of the gas makes this possible. There is currently no extant device that will do this, but there is no good reason why this cannot be built if desired or that, once made, that it would not work as specified. Again, this would allow the same energy to be recycled, and on each pass through the system that energy would do more work. This is thus also a Perpetual Motion device of the second kind.
4: There will be other methods possible that I either haven’t thought of or haven’t mentioned. In order to get continuous work out, we need to have a continuous loop of energy. Here, I have concentrated on the simple methods where the reason for the energy-loop is obvious, and where the devices have either been built and tested or where the technology is available to make them. The losses in a normal machine are actually simply changing unidirectional KE to omnidirectional KE (heat) that simple classical machines can no longer use. The energy is however still there, and we need to simply change the omnidirectional energy back into unidirectional energy. Any device that does that will give us an effective perpetual motion system, since we can’t destroy the energy itself.
5: For those who look to the far-distant future where the whole universe is in thermal equilibrium, it should be noted that that would not stop us either collecting energy together or doing work, though of course it would be a bit more difficult than today. Where there is an energy-flow, we can still harvest work from it, and at anywhere above absolute zero there is always an energy flow.
To restate the conclusion in a different way:
We are surrounded by a vast amount of kinetic energy that is moving in random directions such that the vector sum is substantially zero. If we use either the photoelectric effect, or a mechanical engine on a small-enough scale, to convert the random directions into a unidirectional energy flow along a path we choose, then we can perform continuous work using that flow without needing to burn fuel of any sort. Since this principle of Perpetual Motion has been demonstrated with a nantenna array, it is not impossible as has been held for centuries. Looping the available energy in order to achieve continuous work is in fact the ultimate renewable power source, and may be used anywhere on the globe. At this point in time the available methods of tapping this resource are low-power, but I am working with friends to improve the current levels of a few mW to designs that can potentially supply in the kW range.
We have been taught that only a linear movement of energy can be used to do work, and that thus the amount of work that can be done is limited by the excess energy available. In fact, it is possible to generate a loop of energy and then the amount of work that can be done is unlimited, except in so far as the energy is stored as KEWork and PEWork. Work itself is only a different configuration of energy in a system, and if we can direct the paths along which the available energy is allowed to move then we can do any work we want to without needing to add extra energy into the system. The most convenient way of directing energy is as electricity, which we can store and release as desired and which can be moved along wires from the source to the destination.
The obvious practical device is a rectenna that is tuned to the IR frequency required. These are very small, so are normally referred to nantennae. A reference on how these are manufactured is in the appendix, so there’s no need to explain manufacture here but simply what they are.
A nantenna is an antenna tuned to the IR band of interest (around 10 micron wavelength for room-temperature) and combined with a MIM diode to rectify it. These nantennae are fabricated in an array with connecting wires to deliver the electricity generated. With only a few watts/m² available within the bandwidth of the nantenna array, the available power out is not much – refer to the appendix for measurements.
If we get power out of the nantenna array, it follows that heat is being taken out of the environment (by Conservation of Energy) and that the nantenna array will be colder than the local environment. While we take power, the nantenna array is a cold-sink relative to the environment. Heat goes in to the nantenna array and comes out as electricity. All the heat that is thus transformed (100%) will end up as electricity in the wires, since energy is conserved. In contrast, if you look at a Peltier block or other thermoelectric generator, you’ll see that it needs to have one face hotter and the other face colder, and that the electricity is here produced as a consequence of the movement of net heat from hotter to colder. The Peltier block is thus limited by the Carnot limit and 2LoT, and so far they are way off that efficiency anyway. The nantenna, as we’ve seen, needs only one radiator of EM energy in the waveband it is tuned to, and though it also won’t translate all incident photons to electricity (thus won’t reach 100% efficiency in real terms), the efficiency is not limited except by how good the design is.
The energy-loop can be diagrammed thus:
Environmental heat => IR radiation => nantenna array => electricity => work in load => heat energy => environmental heat
The “work in load” step may be modified by storage of the electrical power in a battery, or the work may involve lifting something against gravity, but after some time the stored energy will be released back into the environment unless we send it off the Earth, for example as a laser beam. The load here is anything we want to do; it could be a simple resistor that heats up, it could be the motor in your electric car, or your mobile phone – anything that uses electrical power. Exceptions are easy to put forward, but each step is allowed in both theory and practice, and for most work we want to do the diagram is what would happen.
If we put the nantenna array in a closed insulated container, and take the electrical output to a resistor or other load such as a motor or lamp, then we will see that the temperature of the container will drop at a rate equivalent to the power delivered to the load. This is precisely what 2LoT says cannot happen, since the only results are that the container cools down and work is done outside it.
If you enclose the resistor or load in a second insulated container, then that container will heat up as the first cools down, and after enough time you will be able to run a heat engine between the two containers. Interestingly, if we regard the two containers as an isolated system, then the total entropy of that system will also decrease as the temperature-difference between the containers increases.
If we leave the nantenna in an open environment, and connect the wires at some large distance to a motor that does some work such as lifting a weight or pumping water, then that work will continue to be done as long as the motor doesn’t wear out, or the nantenna does not corrode or otherwise fail. In a practical sense, this is perpetual motion. The energy comes from the environment, is converted to electricity and then work, and at some point will return to the environment as heat which can then rejoin the cycle. If you want a shorter cycle, then put a fan on the motor and enclose it in the insulated container, when it can be seen that the available energy is simply cycled between heat energy and electrical energy – work is done but the container neither heats up nor cools down. The fan will continue to run until it wears out.
It should be obvious that however you state 2LoT, this system breaks it. It follows from this that 2LoT has a loophole that we have the means to exploit, and that a practical perpetual motion device that also delivers power to the user (the nantenna array) is in fact possible and has been built and tested. Practical details of manufacturing and testing are in the appendix. The key point in breaking 2LoT is the conversion of energy in random directions to energy with a single direction, which requires engineering on a microscopic scale.
A second type of device is also obvious, in that we need to make a PV using a semiconductor with a bandgap of around the same energy as the thermal energy at room temperature. Somewhere in the band 10-100meV would function, and the standard methods of making a PV should be applicable here as well. It is somewhat difficult to calculate beforehand what the actual output of this would be, since there is some confusion over the emission of IR wavelengths within a lattice – we only measure the emissions outside the lattice. It seems reasonable to suppose that IR is transmitted and received within the lattice in very near-field conditions, and so the actual emission rate may be a lot higher than expected. We will find this out by experiment. For far-field, though, a bandgap of 25meV will see around 138mW/m² available, and may convert 20% of that into electricity. This device is planned to be made during 2017, and the data it produces will be used in the next design.
A third type of device has been mentioned earlier, in that a piezo-electric device which is smaller than the mean-free-path of the gas it is immersed in will be able to resolve each individual collision of a gas molecule. For air at STP this is around 70nm diameter, but we could use a heavier noble gas such as Argon, Krypton or Xenon or alternatively Sulphur Hexafluoride in order to be able to use a larger diameter, or alternatively we can run the gas at a lower pressure. Electrodes on the piezo would take the signal to a fast MIM diode (or equivalent) for rectification, and so for each collision we’d get a DC electrical pulse that will add to the output from an array of similar devices. It is reasonable to expect somewhere around 1kW/m² from such a device. Gas molecules hitting the piezo will rebound with a lower velocity and we’d measure that as being cooler. If we are using other than air for the device, or other than atmospheric pressure, we’d need a hermetic seal to hold the working gas in and a heat-exchanger system to keep the working gas at near-ambient temperatures. Since I published this idea here in 2013, it’s not going to be patentable so I don’t expect it will get built until we have enough money to do it ourselves.
A large amount of power is used in cooling. If we use one of these devices for cooling, we actually get power out instead. Your refrigerator now becomes a power source in your house, rather than a power drain, though of course the energy it delivers comes from the heat in the food so will be intermittent. Producing liquid air products will also produce power rather than use it. This may cause some big changes in the ways we use “energy” since what we are really using is work, and work is free providing you can loop energy. It may take a while for the language to catch up, though.
It has been shown that for the heat-engines we know the 2LoT describes the limitations very well. While we have feature-sizes of the devices large relative to atomic size, we will not be able to make a device that will break 2LoT. On a macroscopic scale, heat will always flow from a hotter body to a cooler body, and we will be limited to devices that require a hotter side and a colder side to perform work by the controlled transfer of heat from the hotter to the colder heat-sink. The energy flows in a linear fashion from hot to cold, and will not go in the reverse direction without us needing to put work in. If we have a certain amount of excess energy in a location relative to another location or the environment, we can do only up to that amount of work by allowing that movement, but no more.
If however we use either the quantum nature of photons or the particulate nature of matter, and change the scale of our manufacture to suit, we can instead use a single heat-sink (or in fact gas pressure rather than gas-flow) to perform work, and instead of the energy flowing in a linear fashion it will be looped on itself, in a logical way if not actual. We convert the environmental random-directional KE to unidirectional KE to do work, and normally the work results in random-directional KE again. The same quantity of energy can give us a continuous delivery of work for as long as the device lasts, as long as that output energy is not stored. Meantime, the energy store in the environment is vast and is replenished by the Sun, so once we can use this we won’t run out of available energy. I have shown that practically a nantenna array will do this, and that they have been made and tested, and I have also given two other examples of designs that should do the same job but have not yet been tested. Though the currently-available devices only deliver a small amount of power, once the possibility of them is realised we should soon get much-improved power delivery. While they are regarded as impossible and as fraudulent if they are claimed, few will invest money into the development.
If we develop this as a technology, then we will no longer need to burn fuel to provide power. This is the ultimate carbon-free power, in that once the device is made it will continue to deliver power indefinitely. In hot climates, it can be used as air-cooling and will deliver power in the process. In cold climates, it can move the available energy from outside the homes to the inside and thus keep people warm without needing any fuel to do so. Energy is conserved, but work is not.
The central point is in understanding how and why energy moves the way it does. For heat energy, it is a random walk and so after enough time it will have equal probability of being everywhere accessible to it. Even when it is at this even concentration, though, there is still a measurable flow of energy, and all we need to do work is a flow of energy. At thermal equilibrium, the flows are just equal and opposite, and so the simple devices we’ve made for a net positive flow won’t work. We need instead devices that allow the heat energy to loop and thus deliver continuous work output. These devices already exist, but only deliver a few microwatts to milliwatts (if a large device) and are not yet useful.
Widespread use of such devices at the kW scale and above will take heat out of the local environment, but this will also be returned to the same environment after maybe a delay through accumulation in batteries to deal with peak loads. The net result will be zero, unlike burning fuel which returns energy to the environment that may have been stored millions of years ago. Since taking a large amount of heat from the environment will get more difficult the larger the amount, because of icing-up and suchlike, I do not expect these devices will be used for power-levels in the many tens of kW, though it’s possible that in a car or truck there will be enough airflow to sustain enough heat-transfer to actually run the vehicle or at least a substantial percentage of the load. We will still need other power-sources, but not to the same extent as today. It’s likely that, with the backbone of continually available (though variable with temperature) power that the other renewable power sources may be able to supply the rest of the power needs of our society. If these devices are used in conjunction with some heat-source, they will be almost 100% efficient in converting heat to electricity, unlike other heat engines. The only losses will be in the wiring resistance, thus losing some heat outside the system, since energy is conserved.
Since Perpetual Motion machines are in fact possible, and have been made and shown to work, we should direct resources into research into how to deliver a higher power-level. We do not need to have two heatsinks in order to convert heat to electricity, but we have the technology now to do this using a single heatsink (though currently at a low power). We need to improve the technology, and this does not need new physics but simply another look at what we already know.
2LoT – Second Law of Thermodynamics
CoE – Conservation of energy
CoM – Conservation of momentum
Heat engine – a device that utilises the difference of temperature of an energy-source and an energy- sink to produce output work. The thermodynamic efficiency is limited to the Carnot efficiency, but since an optimal zero-loss Carnot engine will in fact convert the energy difference totally into work, that Carnot limit of thermodynamic efficiency is simply equivalent to CoE. You don’t get more energy out than you put in.
IR – Infra-red light, radiated by all bodies above absolute zero temperature.
KE – Kinetic energy
MIM – Metal-Insulator-Metal diode or tunnelling-device
PE – Potential energy
Perpetual Motion of the second type – where a device uses no fuel or other obvious source of power yet runs itself and a further load as well. Long considered impossible, but here shown to be not only possible but manufactured and proved to work..
PV – Photovoltaic cell, which produces electrical power from incident visible and IR light.
Stefan-Boltzmann law – this states that the radiation from a hot body is proportional to the fourth power of the absolute temperature. See http://hyperphysics.phy-astr.gsu.edu/hbase/thermo/stefan.html for details.
STP – Standard temperature and pressure of 0°C and 100kPa
ZPE – Zero-point energy of a particle, which is the energy that it retains in its ground-state at absolute zero temperature. This may be only theoretical, and not practically available. In any case, to date no-one has managed to extract it definitively.
1: A dissertation on how to make and use a nantenna array
Title of dissertation:
IR DETECTION AND ENERGY HARVESTING USING ANTENNA COUPLED MIM TUNNEL DIODES
Filiz Yesilkoy, PhD Dissertation, 2012
Dissertation directed by: Professor Martin Peckerar , Department of Electrical and Computer Engineering
The infrared (IR) spectrum lies between the microwave and optical frequency ranges, which are well suited for communication and energy harvesting purposes, respectively. The long wavelength IR (LWIR) spectrum, corresponding to wavelengths from 8μm to 15μm, includes the thermal radiation emitted by objects at room temperature and the Earth’s terrestrial radiation. Therefore, LWIR detectors are very appealing for thermal imaging purposes. Thermal detectors developed so far either demand cryogenic operation for fast detection, or they rely on the accumulation of thermal energy in their mass and subsequent measurable changes in material properties. Therefore, they are relatively slow. Quantum detectors, allow for tunable and instantaneous detection but are expensive and require complex processes for fabrication. Bolometer detectors are simple and cheap but do not allow for tunability or for rapid detection.
Harvesting the LWIR radiation energy sourced by the Earth’s heating/cooling cycle is very important for the development of mobile energy resources. While speed is not as significant an issue here, conversion efficiency is an eminent problem for cheap, large area energy transduction. This dissertation addresses the development of tunable, fast, and low cost wave detectors that can operate at room temperature and, when produced in large array format, can harvest Earth’s terrestrial radiation energy.
You can download the full dissertation at:
A search will bring up a lot of similar devices, since the principle of operation is well-known.
List of relevant articles on R-G
http://revolution-green.com/free-energy-by-simon-derricutt/ The first article, that introduced the piezoelectric idea to get Free Work from atmospheric pressure. About 4 years ago, but the ideas predate that by decades.
http://revolution-green.com/robert-murray-smith_ambient-energy/ RMS’s replication of the Lovell device, which shows that it’s not that difficult to make a device that harvests ambient energy from room-temperature, and uses only a single heat-sink.
http://revolution-green.com/paradigm-change/ gives some details of Dan Sheehan’s work. This has gone dark since, so I don’t know what progress he’s made.
http://revolution-green.com/thoughts-proell-effect-similar-ideas/ which explores Ken Rauen’s ideas and his intended method of beating 2LoT. At the time, I said it wouldn’t work based on Conservation of Momentum considerations, but now I would say he doesn’t loop the energy. He did however point me at MerCaT devices, which are PVs that work at IR energies and are commercially available. Devices that loop energy exist, but the output is somewhat too low to be useful. Ken came close to seeing that an energy-loop was possible but chose the wrong way to implement it.
http://revolution-green.com/simon-derricut-free-energy-philosophy/ is where Mark took a comment and turned it into an article. Just about the start of seeing Free Work as a concept.
http://revolution-green.com/some-philosophy-on-work-and-how-we-might-get-it-for-free/ is where the Free Work idea was really put forward, and also the invention of 3 different kinds of work so we could better analyse where the energy was flowing and what was being done.
http://revolution-green.com/infrared-pv-gives-power-at-night-too/ shows a nantenna array and what it can do. Again, this is a real device that loops energy, and is a perpetual motion device. Since the reasons for it working are well-known in theory, and it’s only the manufacture that is a bit expensive and difficult, no-one realised that it is disallowed from working by thermodynamics theory. Luckily, being against theory is not a bar to something actually working experimentally.
http://revolution-green.com/some-energy-basics/ explains how energy moves, and in the comments there’s the first set of points (where all must be correct in order to achieve the conclusions) that specifies how to actually make a Free Work system. One needed a bit of expansion to show why it was logical to remove the electricity generated by a PV or nantenna from the heat equations.
http://revolution-green.com/new-year-simon/ was where the underlying error in the formulation of the 2LoT was exposed, in that the heat was modelled as a fluid that would only flow from hotter to colder. Here I show that it flows in all directions possible, and that just because we normally measure the net flow doesn’t mean that that is the reality. Again, Mark took a comment and made an article of it.
Most of these articles have further explanations in the comments section. Sometimes it takes many words to transfer an idea, and there are certainly a lot of words here.
Damn! The MiB just turned up!