This month, Talkshop contributor ‘Lucy Skywalker’ is travelling to Germany to attend a seminar being run by Roderick Graeff, the precision engineering company owner who conducted experiments to prove the existence of a temperature gradient in an isolated fluid column at equilibrium situated in a gravitational field. Graeff’s results matched his predictions, which ultimately derive from the theoretical work done by Johan Josef Loschmidt in the 1870’s. Maxwell rebutted Loschmidt through reference to his own formulation of the second law, a circular argument by assertion. Boltzmann attempted several proofs, but admitted none of them was satisfactory. Loschmidt gained support from Laplace and Lagrange. The issue was never settled because experimental equipment of the necessary sensitivity was not available. Graeff has changed that.
Controversy arises out of the apparent contradiction of the second law of thermodynamics, detailed in the well commented Loschmidt thread here at the Talkshop in January. Strong contributions were made by physicists and mathematicians, and it is worth a read. You will also find links to relevant material, including Graeff’s experimental methodology and results, criticism of these from Dan Sheehan (1,2,3), who convened a ‘second law violations’ conference in 2002, and various papers sprinkled through the comments cited by protagonists from both sides of the debate.
The long running arguments against the gravito-thermal effect made by Team WUWT (Anthony Watts, Willis Eschenbach, Leif Svalgaard, Robert Brown, and Ira Glickstein) since Nikolov and Zeller published their original conference poster (which doesn’t rely on Loschmidt’s Gravito-Thermal theory anyway) are currently resting on two threads which attack Hans Jelbring’s 2003 and 2012 papers, (without actually critiquing the gedanken experiments he set up), in which they didn’t square up to important scientific arguments from our side of the debate, as this comment from ‘Trick’ regarding one of them along with my response exemplifies:
tallbloke says:
May 2, 2012 at 11:51 pm
Trick says:
May 1, 2012 at 2:46 pm
Willis says at 12:51pm:
“Robert and I were discussing nothing but Jelbring’s gedanken experiment. I discussed it here, Robert discussed it there …we’ve conclusively shown it to be another failed perpetual motion machine.”
Ahhh… the non-GHG ideal gas column contest restarts. Team Tallbloke & Trick et. al. vs. Team Willis & Robert et. al.
Willis means Fig. 1 in link “there” as “it”, so interested readers can therein discover in l-o-o-o-n-g comments that Team Willis & Robert did NOT conclusively show “it” (Fig.1) “…to be another failed perpetual motion machine.”
Willis asks of our Team:
Q: “Cite?”
A: Again, the conclusive citation being a 2004 peer reviewed published paper by Verkley & Gerkema 1st cited by poster Roberto Caballero in “there” thread:
http://journals.ametsoc.org/doi/pdf/10.1175/1520-0469(2004)061%3C0931%3AOMEP%3E2.0.CO%3B2
Robert’s Fig. 1 is exactly the construct in Verkley & Gerkema Part 2b where their paper rigorously shows the temperature profile (of Fig. 1 in the “there” WUWT post) in z is non-isothermal, isentropic in compliance with 0th, 1st & 2nd thermo laws. No perpetual motion. Robert manifestly misses this by classically assuming T(z) constant with dz to perform the integration right before his eqn. 6.
Assuming that T is constant to perform the dz integration then writing T is shown manifestly constant in z by the integration doesn’t cut it.
Verkley et. al. 2004 paper cited above still stands as doing the T(z) integration properly based on Bohren&Albrecht 1998 text cited therein. Team Willis & Robert et. al. have NOT cited any later rigorous proof conclusively showing Robert’s Fig 1 (“it”) “another failed perpetual motion machine”.
Trick, thank you for staying on the case with this important issue. I fear that the opposing team believe the issue can be settled by shouting longest and loudest and most repetitively, whilst intermingling the most ad hominem arguments into their voluminous output, rather than by addressing the actual science.
They also studiously ignore the experimental evidence and ignore all calls for replication at an accredited lab. Instead, they dismiss the experimenter as a “crank”. This is just more ad hominem attack which attempts to divert attention from the key issue. One of our group is heading to Germany in a fortnight to meet the experimenter at a seminar he is giving. I will be publishing a full report afterwards. prior to that, I will be making a call for questions to put to him, so please visit my site and pitch in.
__________________________________________
So here is that call for questions. This is our chance to get answers from the man who has taken Albert Einstein’s injunction regarding the necessity for theory to be confirmed or disproven by experiment seriously. As Einstein said in private correspondence with Robert Millikan, June 1921:
Experimentum summas Judex – Experiment is the final arbiter.
This point was reiterated by Raghu Singh in the thread on his gravity theory:
A new theory should not be judged by an older theory. A new theory is initially judged by whether: it is internally, physically consistent; it re-discovers older predictions; and it discovers at least one new prediction. The model meets these criteria. Only observations and experiments may falsify a theory. A theory confirmed yesterday could be falsified tomorrow.
This is as true for Maxwell’s incomplete formulation of the second law of thermodynamics as it is for Einstein’s theory of relativity. Science progresses by investigating, understanding, and integrating novel results into the corpus of knowledge. Not by ignoring results and hurling ad hominem abuse at proponents of new theory. Not by misdirecting debate by setting up straw man arguments and misinterpretations of those theories’ content. That way lies dogma and ignorance. The high road to knowledge lies up the steep hill of the learning curve, not in the mire of self satisfied reflection on the past merit of flawed or incomplete theory. Strongly contested debate over scientific ideas is healthy for science, provided it is conducted fairly and with good grace. I hope the protagonists from the other side of this debate will step up their game and address the science instead of attacking the scientists.
Tallbloke's Talkshop 
No doubt I’m being thick, but I cannot see any connection between figure 4 and “temperature gradient in an isolated fluid column at equilibrium”.
How can a “speed of temperature change” be involved in a fluid at equilibrium?
[Reply] Read the paper.
The fact that the Venus/Earth temperature ratio, at points of equal pressure over the range of Earth tropospheric pressures (excepting only at points inside the Venus cloud layer), is a constant that is precisely and entirely due only to the ratio of the two planets’ distances from the Sun, IS the definitive empirical evidence–of two detailed atmospheres, and thus the largest, and clearest, experimental test–revealing the fundamental physical mechanism of warming of planetary atmospheres, by direct absorption of radiative (infrared) energy from the Sun. My Venus/Earth comparison establishes: 1) The tropospheres of both Venus and Earth must directly absorb, and be fundamentally warmed only by, the same physical fraction of the incident solar energy, whose intensity varies with solar distance r as one over r-squared (in which case, as I have shown at the above link, the temperature ratio is necessarily what it is observed to be); 2) The tropospheres, especially Earth’s, are not fundamentally (i.e., globally) warmed from the surface at all, as most scientists currently believe, as any such warming would cause the Venus/Earth temperature ratio to be other than what it is, due only to the solar distances of the two planets; 3) The Standard Atmosphere properly represents the stable equilibrium state of Earth’s troposphere, while Venus’s atmosphere is always in its equilibrium state (this means the Earth is not subject to “runaway” global warming OR cooling–meaning no “ice ages” due only to undirected physical processes, and no need for auxiliarly theories, like the physically suspect Milankovitch correlation-model theory, to “explain” them); 4) There is no “greenhouse effect”, of increased global mean surface temperature, with an atmospheric increase in carbon dioxide or any other so-called “greenhouse gas”; 5) There is no albedo effect, either from the planetary surface or from cloud tops, upon the equilibrium atmospheric temperature, since the great differences in these variables for Venus and Earth have no effect upon the Venus/Earth temperature ratio (and this is also the obvious physical consequence of atmospheric warming solely by direct absorption of solar IR); and 6) the proper handling of the Stefan-Boltzmann formula in analyzing the problem of atmospheric warming (it applies only in the case of warming only by incident radiation, assumes that all absorbed radiation is transformed into heat energy, and it cannot be applied anywhere within a system in which other forms of heat transfer–conduction and convection–are occurring).
My Venus/Earth comparison also shows that the temperature lapse rate structure of the troposphere predominates over all other atmospheric conditions, including night and day, that it fundamentally and strictly governs atmospheric temperatures, and that the lapse rate structure does NOT depend upon gross vertical convection, especially from the surface, since each layer of the atmosphere is directly warmed by the Sun. The term “adiabatic lapse rate” should be replaced by “hydrostatic lapse rate” (as I have consistently emphasized) to be consistent with the fundamental physics–of real gases in a gravitational field–behind it.
Steveta
I recommend strongly that you read Graeff’s paper. For me, going that extra mile to study the paper was the eye-opener. It will give you a much better answer to your question than addressing it directly out of context will do. The diagram Rog reproduced here is towards the end of the paper.
If you still need clarification, ask again.
I’ve also essentially retyped Graeff’s paper, with some minor editing to clarify issues, for the climate wiki that is still technically under wraps but if you google sensibly you will find it…
You will also see Graeff’s website referenced on the paper. Well, here it is. I think it’s badly designed as a website, but it does hold more interesting info including Graeff’s experiments with a column of air, and references to his book My Path to Peaceful Energy. This book is an absolute cracker. You can see a much deeper context for Graeff’s work and experimental designs. He’s actually done a lot of different experiments, all together showing unmistakeable evidence of gravity-induced temperature gradients. I’m going to do a review of it for Amazon as soon as I’ve finished it. Was halfway through when the invite arrived, and I thought, I need to have as many fair questions as possible…
Hi Harry,
Whereas all or nearly all radiative energy entering Venus’ atmosphere from the Sun is absorbed above the surface, this is clearly not the case on Earth, where a proportion of the radiative energy from the sun gets all the way through the atmosphere, and many metres of water, before heating the bed of lakes and the shallows around coasts producing measurable convection.
Have you had further thoughts on that since we last exchanged comments?
Cheers
TB
Hi Lucy: Personally, I want to concentrate on Graeff’s experimental work determining the gradient. I think that the waters will be muddied if we discuss his theoretical considerations of the possibility of deriving useful work from the gradient in the same context.
We need to keep consideration of his experimental work separate from his theoretical work or opponents of the Loschmidt effect will use his theoretical work to make an a priori dismissal of both his theoretical work and his experimental results.
TB wrote:
Hi Lucy: Personally, I want to concentrate on Graeff’s experimental work determining the gradient. I think that the waters will be muddied if we discuss his theoretical considerations of the possibility of deriving useful work from the gradient in the same context.
TB, drop me a line. I’ve been going the other direction.
For the William James pragmatists about a “truth”:
This point was reiterated by Raghu Singh in the thread on his gravity theory:
A new theory is initially judged by whether: it is internally, physically consistent; it re-discovers older predictions; and it discovers at least one new prediction.
James just added the concept of “cash value”, i.e. you could make some use or “profit” from it.
The profit here is another smack against CAGW taxes etc. based on settled science and certainty.
I must admit I, and I suspect others, have gotten lost in the argument, especially in the contradiction of the 2nd Law. Rog, if your upcoming report could give an elevator explanation of the conflict and the experimental test meaning and signficance, I, for one would be helped.
Oh, yeah, the shouting from WUWT: another Team in action. Ain’t science wonderfully non-emotional! So unlike human behaviour, especially the school-age stage.
[Reply] Indeedy. For a precis of the issues, hit the link to the Loschmidt thread and have a read. Or download the Graeff pdf and read the intro about Loschmidt and Maxwell.
TB, I’m slightly confused.
I too rank Graeff’s experimental, hands-on work as the most important work he has done. To me, his book amplifies that, because it helps me see the years of patient work and careful precision engineering behind his experiments. And there are lots of experiments he did, which his paper does not show.
Regarding his theoretical work, I am just trying to follow it and make sense of it for myself, as adequate explanation for the undisputed experimental results. I don’t know if it is “correct” though it may be, and it is certainly interesting and plausible and upheld by experimental results.
I’m not interested in looking at deriving “useful” work from Graeff’s “gravity machine” because it seems the amount is far, far too small to be useful. And even if it can be amplified in the future, right now, the issue to me is to show Graeff is essentially upholding the Second Law, but with a crucial modification, that makes sense of the adiabatic lapse rate as the effect of naturally balancing two forces, the gravitational warming effect, and convection equalizing pressure imbalances resulting from gravitational warming. Hence the troposphere is constantly “tropos”, turning. Then there is the extra warming effect of sunshine on tangible surfaces that warm adjacent air by conduction and cause the air to convect more vigorously.
To me, a formal reconciliation with, and modification of, the Second Law is needed to support Nikolov and Zeller, and indeed the fundamentals of Climate Science. And this has to be grounded in the experimental evidence.
Hope this clarifies things.
PS
It is perhaps worth emphasising, that it becomes clear from Graeff’s book that he didn’t expect to find a vertical temperature gradient that prevented him from equalizing temperature at top and bottom of his apparatus. The possibility had not even occurred to him. But when evidence appeared that suggested a measurable effect of gravity, he pursued the evidence and designed experiments to show up the effect more clearly and unequivocally; and to eliminate other possible causes. And he is begging for reproduction of his experiments.
It’s true that Graeff has been driven by a lifelong passion to discover peaceful sources of energy; yet it seems unlikely that his experiments will yield this dream, at least, not directly. But then, he has an excellent pedigree in this kind of experience. Kepler was driven by a passion to discover the harmony of the stars in astronomical mathematics, in a way that was commensurate with the correspondences he could see in his astrological work. Kepler did NOT manifest his original dreams; but he left a priceless legacy to Science, which centuries later could be used to discern mathematical harmonies beyond Kepler’s wildest dreams.
So – to emphasise – back to the crucial evidence of Graeff’s experiments.
Which are the laws governing energy interactions in the universe, negentropic life included? As far as we ignore, or worse, choose to ignore the general laws, we end by just groping in the dark. Perhaps Kepler was not wrong and we are not so “developed” after all, to unconsciously depreciate anything by calling it by names like “astrologist”. It takes time to get rid of self-conceit, and this is a pre-requisite of any science.
Think about this. You have two large parallel level plates at absolute temperature T in a gravitational field of strength g, separated by vacuum, difference in their elevation is d, the entire setup being isolated from its environment with good thermal insulation.
If it is left alone for a time long enough to settle into equilibrium, what will be the temperature difference between the two plates?
Perhaps I’m being thick but, why is that relevant when discussing a temperature gradient in a conductive and convective gas in a gravity field with a distant energy source?
Graeff’s experiments are designed to minimise convection (or show its effects by comparison) but keep conduction possible in the part being tested, while providing alternating encasing layers of insulation and high-conduction to approach an environment that is as close as possible to uniform temperature. The more I read his book, the better I feel I understand his engineering finesse. You have to imagine the atoms joshing each other and gravity effects being passed down the line.
So Berenyi P, I think the answer would be: same temperature on the tops of the plates, but these might be a tiny fraction cooler than the undersides of the plates.
But hey, can we stay with what is actually being measured – and why?
All I know is that when I cycle down a mountain the air around me gets warmer but I am further from the sun.
I also know local miners who tell me it gets hotter the deeper they go.
I’ve got to get my two cents in here since I was one of the bloodied remnants of this same discussion over at WUWT.
So many are wrong but at different levels:
Anthony Watts, Willis Eschenbach, Leif Svalgaard, Robert Brown, and Ira Glickstein are all both right and wrong. No there is no possible perpetual motion machine to extract free energy but also they are wrong for there is a natural temperature gradient.
The company with envisions of somehow extracting this difference will fail. They are correct on the gradient’s existence but wrong on their interpretation. It will take theoretically equal but, in reality, more energy just to try to extract it. The gradient being real will also apply to any machine, wires, apparatuses exactly cancelling what they are trying to take advantage of.
This is just another expression of potential energy. A stack of bricks high above with ropes and pulleys can extract energy from that potential, BUT, it takes exactly the same (really more due to loss) energy to raise the brick back up for the next cycle. To me I don’t care if you are doing it with ropes and pulleys, wires (Robert Brown), thermocouples, mirrors, (smoke?) the same principle will always apply as it should in physics.
Let me add my two cents into the discussion concerning the temperature gradient. Consider a steel ball located 2m above the floor. It is released and accelerates as it drops. Question: does the ball have more, less or the same energy at 2m above the floor, at 1m, or at the point of impact on the floor? The answer is that it has exactly, discounting for the effect of atmospheric friction, at every point. In like manner, the molecules of gas/liquid in the column in Graeff’s experiment have the same amount of energy regardless of their height. At the top the energy is in the form of potential energy, at the bottom in the form of kinetic energy. The increased kinetic energy is reflected in a higher temperature, and thus the gradient. The fluid at the top has just as much energy. Since an energy gradient does not exist, no work can be extracted.
Exactly you cannot deal with temperature in isolation, temperature varies in exact compensation for other changes in energy. If you added another force besides gravity the gradient would change.
Too bad his experimental apparatus is too large to place in a centrifuge and see how the gradient changes as force on the system varies. There may be other much weaker forces involved too such as electrostatic potential for a complete evaluation of the energy equilibrium of a tall column of atmosphere.
In the real world you have gravitational potential energy, kinetic energy of thermal motion, but kinetic energy of motion (wind, up draft down draft), plus potential energy due to ionization or other forms of energy change of the atoms. As you vary one form of energy and convert it to other forms all the other forms will adjust so the net total energy of the air parcel will remain constant unless you have some mechanism to add or remove energy.
We already freely accept the idea that a parcel of air can change temperature substantially due to the release of the energy of condensation or freezing, yet the total energy content of the air remains constant, only the form of the energy has changed not the quantity.
This is just an expansion of the same concept to include gravitational potential energy which in principle is no different than energy stored as latent heat of vaporization or solidification.
Daniel Sweger, well said and couldn’t agree more. Why some people think you can just ignore gravitational potential energy evades me… and more, some are even bona-fide physicists at major universities. Your steel ball example is great and so concise, thanks, I’ll store it away.
James says: May 4, 2012 at 12:08 am
I also know local miners who tell me it gets hotter the deeper they go.
Yes, absolutely. But one also has to consider why oceans are cold at the bottom… I had interesting thoughts on all this that I put on the Nikolov thread.
I’m a bit unhappy with the theory. It’s because it fails to take into account the variation of centripetal force with distance from the earth’s CoG.
Wayne and Dan
Reading Graeff’s amazing book, I think it’s not quite so simple. Let me explain.
Yes, potential energy does exchange for kinetic energy (heat gain). Therefore no contradiction of the Second Law. But… the heat difference energy can be harvested as electricity. But… the electricity is so tiny that nobody in their right mind would try to use it. But… the harvested electricity is there, indisputably, proved too many times to say it doesn’t exist.
It is the tiny quantity that has eluded discovery for so long. It took all of Graeff’s considerable engineering ingenuity to capture scientifically defensible records.
Only in meteorology is the effect amplified enough with altitude, to become real – the line of snow on the hills, heat down a mine, hot Jericho, flat bases to clouds, etc. Contract the scale to a laboratory and it’s much harder to see.
No wonder the cowboys are confused. Luckily I’m a terrier and once I’ve got the scent in my nostrils I don’t stop until I’ve dug out that fox 🙂
Larry: Too bad his experimental apparatus is too large to place in a centrifuge…
Yes. But I think this level of testing will come. Graeff says:
Mydogsgotnonose says:
I’m a bit unhappy with the theory. It’s because it fails to take into account the variation of centripetal force with distance from the earth’s CoG.
From terrier-with-nose: I respect your views generally, in fact there was a question I wanted to ask you but cannot remember 🙂
However, re Graeff, it seems that the variation is simply far, far too small to matter at the laboratory scale we are considering. And even if we take the whole atmospheric depth into account, up to troposphere, it is still tiny. Well, that ‘s what I understand.
Thank you for coming to rogers blog Lucy, your mind is needed in the pursuit of the intricacies of the plethora of new scientific endeavours that are currently coming to the fore.
Thanks Wayne! plethora, yes! I’m pretty gobsmacked myself, but looking forward hugely to this trip to Germany.
My questions for Graeff so far are:
1. Water: why is the measured temperature gradient higher than the calculated one? Application of the true (albeit modified) Second Law would posit that the measured gradient should be lower, not higher, than calculated.
2. Oceans: why are the ocean floors cold not warm?
3. Solids: have nonconducting solids been tested? Does conductivity obscure the Graeff effect?
4. Earth: can the Graeff effect explain (a) a hot earth core (b) vulcanism, when convective forces occasionally overcome the restraints of solid mass?
5. Atmosphere: the Graeff effect seems to prevail over convective balancing up to the tropopause, to yield the measured adiabatic lapse rate. But the Graeff effect does not seem to win out above the tropopause; instead we perhaps, perhaps! have the only region where the greenhouse gas effect prevails. Note: tropopause on all planets is at ~0.1 bar pressure.
6. Prof Sheehan: in 2005 he doubts whether Graeff’s temperature measurements are as accurate as stated. I personally doubt whether Sheehan’s doubt is justified, at least now in 2012; but is there a formal reply to Sheehan? This seems to be an important question to answer, especially for engineering-challenged theoreticians.
7. The “degrees of freedom” molecular issues that affect theoretical temperature gradient values (You need to read Graeff carefully here, preferably his book). I also need help understanding this so that I can “feel” it. And maybe, herein is contained the mystery of why measured temp gradient for water is higher than calculated gradient.
Here’s my first question for Graeff:
Has he had any replies to his calls for replication? Has he approached specific accredited labs to try to engage them? If so, what were their responses?
Lucy: “4. Earth: can the Graeff effect explain (a) a hot earth core (b) vulcanism, when convective forces occasionally overcome the restraints of solid mass?”
I just had to stop on this question. Since this has to do with the under-surface gravitational field I must ask if you were aware of what the gravitational field strength looks like underneath the surface?
Take a look at this quik and rough illustration I tossed together, the scale may not be correct… it was hand drawn:
That green curve is the gravitational field strength and a number of years ago I realized I had never asked that question myself and this what I came up with after reading Dr. Tatum’s great PDF’s and writing an integration of the Earth’s gravity. If the Earth was a hollow sphere must know that anywhere inside you would be weightless so instead of that big ‘V’ there would be two vertical lines instead right where the field reaches a maximum dropping immediately to zero (that black line at the bottom).
The whole point is a question as you posed must hinge somewhat on this fact but I for one am not sure exactly what the correct answer is. Don’t know if the field drops off fast enough to prevent such an effect under the surface. Thought you might find that interesting if you had never seen that curve… and, enjoy and have a safe trip, I have some friends from Germany and England on the way here as I write.
2. Oceans: why are the ocean floors cold not warm?
Because the Graeff effect is nonsense. Gravity can only warm if it is allowed to do work. On the ocean floor it isn’t because there is no convection. No water is removed so gravity can not pull any down. (there is some circulation but that’s a different matter)
Same with 5. There is no convection above the tropopuse because of the inversed temperature gradient.
Tallbloke
I shall certainly ask. To date, the answer is spelled out in a whole chapter of Graeff’s book. But here it is in a nutshell:
Igl appears to follow this pattern. (I phrased question 5 badly – I’m fully aware of no convection in either the stratosphere or the thermosphere, although convection exists in both troposphere and mesosphere, and this isn’t Graeff’s area anyway, it’s mine).
However there is a small number of people who do take Graeff’s work seriously, including Prof. Chuanping Liao whom I referenced up thread.
And I personally am planning to collect a group here who are interested in replicating Graeff. I already have one highly qualified collaborator and the possibility of two more, but this really has to wait until I return.
Thanks Wayne. That pic of gravity strength vs distance from the Earth’s centre of gravity, yes I knew about the linear drop under the earth’s surface. However there is a long way to go to the centre of the earth, and there is room for a lot of heating. Jericho shows it already. Deep mines become impossibly hot. Etc.
Igl,
I have examined the experimental and theoretical work of Dr Graeff carefully. I have studied and understand both the history, importance, theory, and practice of the Second Law of Thermodynamics. Even though Graeff challenges its current formulation, he challenges it to be modified, not abandoned. I can find no point in Graeff’s work where scientific method is violated.
You say “the Graeff effect is nonsense”. Please explain why, with evidence of your own scientific approach that culminates in this conclusion of yours.
Lucy
It is nonsense because the temperature in the mariana trench, 11 km, is even lower than at for instance 3 km depth, so obviously higher pressure does not make it warmer.
It’s nonsense because if it wasn’t you could simply put a well insulated pipe with a u-bend down to 1 km, pump water through it and the water coming back up would have warmed a lot.
It’s nonsense because there is no reason why gravity would create heat if it is not doing any work. There is no energy input.
Hi lgl:
From a classical physics point of view, gravity is a force. Forces don’t ‘do work’, they continuously act. Their action maintains gradients. What’s not to like?
No fluid in a gravity well is at equilibrum, each atom, molecule is independently affected/effected. as the atom drops, its velocity increases, it heats up, for each collision, it transfers heat, along with the direction it was heading, all simple 9 th grade physics, for more refer to your old physics/ and chemestry texts. Or if you can find the references they allude to, for better reading.
Hi tallbloke
http://wiki.answers.com/Q/When_does_the_force_do_work_on_an_object
Heh.
“If the force is unbalanced, it causes acceleration. This involves doing work to move it in a certain direction.”
I think someone should get around to hitting the ‘improve’ button on that ‘answer’.
In classical dynamics, bodies possess gravitational potential energy. When they accelerate towards the CoG, the amount of gravitational potential energy diminishes, as the amount of kinetic energy increases, due to increased momentum. Total energy is unaffected, as it must be in an energy conserving world. Gravity doesn’t ‘get used up’ while all this is happening.
If the person who provided that answer has a different scheme in mind, they should have provided a link to a full explanation of it.
“This involves doing work to move it in a certain direction.” which is exactly what happens when gravity is pulling air molecules to the ground. Work is done and there is a local adiabatic warming, but for this to happen air must be lifted from the surface another place by convection. This means energy is removed from one location and returned at another, net=0.
Ah, now I understand better where you are coming from, Igl.
I know that it’s near-freezing cold at the bottom of the oceans. I know the temperature gradient is warm at surface to cold at depth.
But I also know that Graeff has produced temperature gradients in the opposite direction under strict laboratory conditions that, if one checks Graeff’s full experiments, cannot seriously be attributed to anything else but gravity. However, the effects are tiny and are easily lost in convection that works in the opposite direction unless the convection is impeded.
The atmosphere average temperature is 15C at sea level; minus 50C at the tropopause. The difference in just the troposphere is already far more than the posited greenhouse effect of 33C. And the Moon’s average temperature bears out that the Earth’s atmospheric warming effect is even bigger in total. Jericho at 250m below sea level averages 8C warmer than all other nearby towns. Deep mines get unbearably hot. All atmospheric evidence suggests that temperature has a large dependence on altitude; Graeff simply nails this under lab conditions even though the difference he measures is of necessity tiny.
Negative temperature gradients show up for Graeff in both gases and liquids, so long as convection is impeded. In Nature, it appears that the atmosphere maintains Graeff’s direction of temperature gradient, but the oceans do not. This is what I am curious to understand better. Since there is no way I can pretend that Graeff’s experiments are duff, I have to think again about what is really going on in Nature. Too bad that so many dismiss him quickly. This does not change the thoroughness and depth of the science he’s been developing for years. As I ponder, understanding comes slowly, with occasional “aha” moments. The reality is subtle and deep.
lgl: So what happened when Neil Armstrong (or was it Buzz Aldrin) dropped the hammer and feather on the Moon? What got displaced where for zero net? Maybe I’m not understanding what you are driving at, but ALL air molecules are being pulled towards the ground equally. Gravity doesn’t rely on convection to do anything. Graeff suppressed convection in his fluid experiment anyway.
Lucy: Here’s my next question for Graeff. Can he help specify replication efforts by giving specs and sources for the best thermocouples and datalogging equipment?
In the experiment they need to turn the columns of water 60º on the x axis to see if the gradient reduces to half. Then to 90º to see of there is an instrument error (gradient ≠ 0), and finally, 180º to see if the gradient is restored to the original value with the sensors opposite in the y axis.
I’ve stayed silent.
Please back off from Graeff.
I reject the PDF. Why has no-one else spotted the problems?
On looking this is a come on about power from gravity.
dp, yep, ought to have been obvious to the experimenter, yet there is no sign of any attempt at validation or checking.
@ May 4, 2012 at 12:50 pm, Lucy Skywalker says:
“2. Oceans: why are the ocean floors cold not warm?”
Indeed, one does not need a supercomputer, spreadsheet on a desktop, or even a hand held calculator, to find that Graeff’s clearly expressed conclusion that there is a temperature rise of 0.04K/m in water is false.
If it were true, it would for instance predict that the temperature of the ocean at a depth of 2,500m would be about 2,500mx0.04K/m=100K above that at the surface. That prediction is a complete fail.
Also, the theory that predicts the same temperature gradient must therefore be wrong.
Why then, can he have performed a purportedly well (enough) controlled experiment that is supposed to confirm such an obviously wrong proposition? A question worth some further explanation, but requiring a somewhat more complicated answer.
On giving this a bit more thought a different experiment can be done. Place a sealed column of liquid in a centrifuge, evacuate the test chamber, and spin up the centrifuge to a stable RPM and hold it. Measure the temperature gradient continuously, giving it time to stabilize. Spin faster. Measure. Spin faster… You get it, I think.
The temperature gradient should change as a function of speed of the centrifuge. Somebody will surely want to see the centrifuge axis aligned with the Earth’s axis. Humor them.
Here’s another test. Put a fluid in a closed cylinder at 1 atm. Place the cylinder in many layers of insulation and evacuate the chamber. Measure the temperature gradient. Increase to 10 atm and repeat. Time must be allowed to bleed off the affects of heating by compression, of course. Measure the gradient. Try 100 atm. A thousand.
Or rent a cylinder of oxygen or other gas (acetylene comes with a convection inhibitor built into the bottle), ensure it is not leaking, instrument it, place it in multiple layers of insulation and leave it for a few months in an evacuated chamber, measuring the temperature all the while. A gradient should emerge over time.
My point is gravity is not needed to test any of this – all that is needed is to create a pressure gradient by any expediency.
Who would like to predict what the gradient over various pressures will look like when graphed?
@ May 5, 2012 at 3:58 am, tchannon says:
“I reject the PDF. Why has no-one else spotted the problems?
On looking this is a come on about power from gravity.”
Back in January I looked at the PDF and Graeff’s web site and came to the same conclusion, and then moved on.
dp says:
May 5, 2012 at 2:01 am
In the experiment they need to turn the columns of water 60º on the x axis to see if the gradient reduces to half. Then to 90º to see of there is an instrument error (gradient ≠ 0), and finally, 180º to see if the gradient is restored to the original value with the sensors opposite in the y axis.
tchannon says:
May 5, 2012 at 4:04 am
dp, yep, ought to have been obvious to the experimenter, yet there is no sign of any attempt at validation or checking.
I really wish people would go to the trouble of reading the material linked in the OP before they make comments like this. See the Sheehan critiques of methodology. Graeff performed exactly those tests in other experiments in order to validate his experimental equipment.
Here, let me spoon feed them to you:
Graeff is a precision engineer. He has spent many years of his life and large sums of his own money doing his utmost to eliminate experimental bias. Now you two jokers come along and falsify him from your keyboards in a matter of moments without spending five minutes surveying the material provided to you in the original post, implying along the way that Lucy Skywalker and I are gullible fools who don’t know the first thing about the scientific method.
Give us a feckin’ break. 😦
dp says:
May 5, 2012 at 6:44 am
On giving this a bit more thought a different experiment can be done. Place a sealed column of liquid in a centrifuge, evacuate the test chamber, and spin up the centrifuge to a stable RPM and hold it. Measure the temperature gradient continuously
How is this going to be done? Wifi senders on the thermocouples? How will you test the equipment to ensure spinning it up doesn’t affect the measuring equipment or the data senders? How will you prevent heat generated by the centrifuge bearings communicating along the frame of the rig? How will you prevent out of balance forces setting up vibration which will mix the test fluid.
Maybe rather than “a bit more thought” this needs *a lot* more thought.
Have you got the spare cash for this feckin great big centrifuge? If not, I suggest you carry on polishing your chair while Graeff does his stuff. 😉
Roy Martin says:
May 5, 2012 at 5:10 am
@ May 4, 2012 at 12:50 pm, Lucy Skywalker says:
“2. Oceans: why are the ocean floors cold not warm?”
Indeed, one does not need a supercomputer, spreadsheet on a desktop, or even a hand held calculator, to find that Graeff’s clearly expressed conclusion that there is a temperature rise of 0.04K/m in water is false.
If it were true, it would for instance predict that the temperature of the ocean at a depth of 2,500m would be about 2,500mx0.04K/m=100K above that at the surface. That prediction is a complete fail.
If you read the pdf, you’d discover Graeff suppressed convection and radiation by using powdered glass as a medium for his fluid to work in.
Warmed water molecules are more buoyant, they rise. That’s why you won’t see the gradient in the ocean. However, this doesn’t mean gravity isn’t acting on the fluid to affect the oceans temperature gradient. It could mean that the temperature gradient we see in the ocean from the bottom of the mixed layer to the thermocline isn’t as steep as it would otherwise be for example.
My approach is to be sceptical, but not dismissive. I think that’s the correct scientific attitude. Others may disagree.
@ May 4, 2012 at 12:50 pm, Lucy Skywalker says:
“6. Prof Sheehan: in 2005 he doubts whether Graeff’s temperature measurements are as accurate as stated. I personally doubt whether Sheehan’s doubt is justified, at least now in 2012; but is there a formal reply to Sheehan? This seems to be an important question to answer, especially for engineering-challenged theoreticians.”
I have not found the Sheehan paper you refer to, but as regards Graeff’s temperature measurements, they are probably good enough to give us an idea what the result really means.
As one of those who is not engineering-challenged, my interpretation from a close reading of the PDF and Graeff’s results is that the temperature differential at the tops of the two columns is due entirely to the flow of heat into and out of the system from the environment. To explain…
Graeff said on p.11:
“5. Consequences of the measured temperature gradients for the Second Law.
The brown curve 5 in Fig. 2 shows the temperature differences between the top of tube 1 and tube 2 with an absolute average of about .01 K.”
When you examine the brown curve in Fig. 2 closely, the first observation is that it is really not level at 0.01 K at all, there are many small rises and falls. (For those in the audience, get out your magnifying glass or enlarge the Fig. if you have to.) Also in Fig. 2, the actual thermocouple readings in the apparatus are shown as a group of smoothed lines, revealing that they all rose and fell together over a range of about 0.5 K during the experimental run.
It is very clear from a comparison of the brown curve and the smoothed lines that the only time the tube 1 – tube 2 differential was at the 0.01 K level was when the temperature was rising. It was zero or below when the temperature was falling, up to a bit after 28/12/06, and dropped to zero again briefly around 25/1/07, as the temperature was falling. When the rate of temperature increase fell off after about 25/2/07, the tube 1 – tube 2 differential fell to about 0.005 K. This shows that the tube 1 – tube 2 differential is a direct function of the rate of energy flow out of or into the system. It is not caused by a gravity effect at all.
I would take bets that if the whole of the apparatus was stabilized and held in a truly even, constant temperature environment, the tube 1 – tube 2 differential in temperatures would just disappear, and there would be no difference in temperature between the tops and the bottoms of the tubes either.
IMHO, the whole of the Graeff set up has all the earmarks of like an elborate set up of smoke-and- mirrors.
An experiment is just that, an experiment. If Graeff is careful and proper and the experiment is showing a reproducible effect present then so be it. Might be showing the 300-1000V gradient that is very real. Also it could be experimental error. It’s evidently not a gravitational potential energy effect at +40K/km in water. Maybe the graph is a typo. But I do think if this experiment were reproduced using a gas instead of liquid and scaled up a bit a gravitational gradient would appear close to the lapse rate. Also seems density gradient should play into the equation with solids showing basically zero, liquids a tiny effect, but gases showing the full scale effect and relatively quickly. Just saying.
Hi Wayne,
please read the Sheehan critiques linked in the OP and again above for dp and Tim Channon. Graeff tested solids. liquids and gases and his findings agree with your assumptions. Zero in solids, 0.04K/metre in liquid and 0.07K/metre in air. I don’t know where you picked up 40K/km but it’s 4K/km for fluid. The result for air of 0.07K/metre is close to the environmental lapse rate of ~6.5K/km.
Roy Martin says:
May 5, 2012 at 9:17 am
As one of those who is not engineering-challenged, my interpretation from a close reading of the PDF and Graeff’s results is that the temperature differential at the tops of the two columns is due entirely to the flow of heat into and out of the system from the environment.
1) Graeff is not engineering challenged, and neither am I.
2) Graeff ensured that throughout his tests, the external temperature gradient *ran the other way*.
Roy Martin says:
May 5, 2012 at 9:17 am
Quick answer, will return later. Problem here is Graeff’s presentation which is poor and is one reason I transposed his material to the wiki, to improve presentation.
The “brown line” cannot be seen in fig. 2, you have to look at the enlargement in fig.3. And Graeff’s numbering system is awkward to follow. However, if you do follow it all right through, you will see he refers to line no. 5 in fig.3 which is as he says. Moreover, since the brown line 5 is the difference between lines 1 and 3, you have additional check.
tb: Isn’t 0.04K/metre the same as 40K/km? That is why I said it might, or even must, be a typo. That’s a steep gradient.
“Come on about power from gravity”
Quick answer, will return later. It’s not a come on, and though it talks about the possibility of power from gravity, in reality the quantity measured is so small as to be useless from a practical point of view. But the measurements have been repeated many many times with many different experiments, and all held out.
My interest is that it requires a MODIFICATION not an abandonment of the Second Law; and that this modification is needed to validate Nikolov and Zeller.
You have to read Graeff carefully right through, not throw up your hands prematurely. And you have to read Graeff’s whole book, to appreciate the depth of integrity, expertise and patience from which he comes.
Wayne: ” Isn’t 0.04K/metre the same as 40K/km?”
Yes. I was referring to the theoretical gradient rather than what Graeff measured, my bad. If you read the Sheehan critiques, the discrepancy is mentioned there.
Lucy:” this modification is needed to validate Nikolov and Zeller.”
Careful now. Nikolov and Zeller haven’t said their theory depends on the validation of the Loschmidt effect or a modification of the second law. It is more significant for Hans Jelbring’s 2003 and 2012 papers, but here too, it depends on which version of the second law you are talking about. In terms of classical mechanics working on the second law stated in terms of energy, no modificaion is necessary. For Maxwell and Boltzmann type statistical mechanics it’s unclear, as Boltzmann himself wasn’t happy with his attempted disproofs of Loschmidt.
Furthermore, the second law as it stands talks about the isolation of the material from the environment. Since gravity pervades every terrestrial environment, the theoretical conditions of the second law cannot be strictly met anyway. So there is a case for saying that Graeff and Loschmidt are not challenging the second law at all, but more rigorously defining it, by making the gravitational effect explicit.
tb: Ok, I had read the Sheehan critique when this first was raised a few moth’s ago. they are measuring temperature with voltage differential in the test equipment with precision to +/- 10^-7 V. Great. And the good comparison to theoretic values for air was listed as 9 mK/m which is 9 K/km. So far so good. But that is the same argument I was giving Dr. Robert Brown in the wuwt arguments over a static lapse rate. He was saying that if a natural lapse was present a wire could be attached to the bottom and top of the column and free energy, electrically or thermally, could be extracted proving is was isothermal. I kept saying false, and he would never engage me. In fact totally ignored me. Power to Trick. My contention was that the wire or bar in a gravitational gradient would also show the same potential gradient, electrically or thermally, that the molecules within the column were experiencing. The column would be isentropic, even the wire, not isothermal. Sheehan is saying 0.04 K/m too which is 40 K/km and that is one big gradient and could very well be they are measuring, via the voltage differential, that effect I mentioned plus the natural voltage gradient vertically within our atmosphere since they are measuring with voltage and not thermometers.
There, I probably didn’t put enough words in my first comment so you could see clearly what I was leading to.
Wayne: Interesting. So maybe your hypothesis could be tested by raising and lowering the dataloggers and the connecting wires between the dataloggers and the thermistors?
Maybe this is a question Lucy should raise with Graeff.
Lucy Skywalker says:
May 5, 2012 at 9:59 pm
“The “brown line” cannot be seen in fig. 2, you have to look at the enlargement in fig.3”.
Lucy: the brown line is very clearly visible in Fig. 2, up to .01 K above zero, just where Graeff says it should be. The brown line 5 in Fig. 3 is a smoothed version of the above, and the smoothing obscures the interest characteristics I described.
tallbloke
“Graeff suppressed convection in his fluid experiment anyway.”
Yes, and if he succeeded in that he prevented gravity from doing any work and any warming.
Don’t mix this with what’s going on in the atmosphere. There gravity is allowed to do work and because most of the upflow is in a moist adiabat (ex. indian ocean) and much of the downflow is in a more dry adiabat (ex. sahara), I assume gravity is elevating the average surface temperature, but that’s not Graeffs tubes.
Lucy
Mines are hot because “Radioactive Decay Fuels Earth’s Inner Fires”
http://www.ouramazingplanet.com/1566-radioactive-decay-increases-earths-heat.html
lgl: According to classical considerations, gravity doesn’t ‘do work’ in any kinetic sense. Matter organises itself according to it’s inherent ratio of gravitational potential energy to kinetic energy. That ratio changes depending on its distance from the CoG. Total energy remains constant, all else being equal. Don’t ask me how matter does this, no-one knows. Gravity is a mysterious force. Classical mechanics is a human construct. So is relativity, which doesn’t explain what gravity is any better.
Mines are hot both because they are deeper in the crust, which has thermal gradient between the hot mantle and cooler surface, and because of the thermal gradient in the air, if Loschmidt is correct. Also because of higher air density having greater heat capacity.
“Wayne: Interesting. So maybe your hypothesis could be tested by raising and lowering the dataloggers and the connecting wires between the dataloggers and the thermistors?”
Yes. Something like that. To me they need to be VERY careful of anything depending on z differences. This is like measuring the redshift of e/m pointed upward but in this case that would be one of the errors. It was measured, I think it was at Princeton, in a high foyer at one of their halls, like five stories high. In Houston at the Hyatt the building is hollow. Every time I came out of my room on like the 24th floor it would take your breath away! I can’t remember the number of stories, 30 to 40, much like what houses the shuttle. Now that would be a great place for such an experiment. You would need two loggers, not running wires up and down the column, for that very fact it could impart part of the very effect you are trying to measure. One logger zeroed at the bottom, the other on the balcony at the very top. Measure. Switch the test equipment. Measure. That would zero any equipment errors.
Yes, I think you see what I am speaking of. Would be better if it had four layers of insulation and four thermistors between each layer. Might take eight. And seems the very outer layer would have to circulate to start the outer layer as being isothermal. Each layer going inward should show the growing gradient.
Wayne, Graeff eliminated instrumental bias by turning his tubes end for end. I can see from your considerations that might not eliminate everything. All the more reason for Lucy to take this matter up with Graeff at the seminar in Germany later this month. Perhaps bluetooth enabled sensors is the way to go: eliminate the wiring as much as possible.
tallbloke
Repeating, when there is convection gravity will do work and give adiabatic warming. When there is no convection the molecules will not move in the gravitational field (like you are describing) and there will be no warming.
If mines were not heated from the bottom, the gradient would reverse and they would have been cold. This study, http://nopr.niscair.res.in/handle/123456789/2506 actually disproves this whole nonsense. Air density decreases with depth in the warmest mines, where the gradient is greater than 34 K/km.
All, wordpress has started pre-ticking the followup email notification box. Don’t forget to untick it before you submit a comment if you want to avoid you inbox filling with junk.
Thanks to dp for the headsup
lgl: I didn’t say “no convection”, I said “suppressed convection”. Difference.
If gravity was to “do work”, then since energy isn’t created, gravity would “get used up”. So after deflecting the orbit of a large number of comets, we would float off into space from the gravity free Earth. Doesn’t happen, because gravity doesn’t “do work”. It is a continuously acting force, not an energy.
Tallbloke says:
May 5, 2012 at 8:56 pm
“My approach is to be sceptical, but not dismissive. I think that’s the correct scientific attitude. Others may disagree.”
No argument on that principle, but in this case my understanding of the distribution of ocean temperatures both vertically and laterally, and very slow rates of circulation in the deep ocean, does lead me to the point of being dismissive of the results of this Graeff experiment. The temperatures that would arise as a consequence of his results are so far removed from what is observed that the two appear to me to be quite irreconcilable.
If Lucy S. has the opportunity to put this question to Graeff I will be very interested to learn his response.
23:00 hours here, good night folks.
Hi Roy: Graeff got a bigger gradient than theory predicts. We don’t yet know why, but Wayne’s ideas above also need putting to Graeff. Please do take account of my earlier reply to you on the subject of the ocean gradient:
tallbloke
According to your physics I guess gravity does not do any work in hydro-electric power stations either?
James Buchanan says:
May 4, 2012 at 10:11 pm
No fluid in a gravity well is at equilibrum, each atom, molecule is independently affected/effected. as the atom drops, its velocity increases, it heats up, for each collision, it transfers heat, along with the direction it was heading, all simple 9 th grade physics, for more refer to your old physics/ and chemestry texts. Or if you can find the references they allude to, for better reading…
This looks sound and AFAICT Graeff takes this explanation further in his book, to explain his theoretical calculations (which come close to experimental measurements). Any URL’s James or anyone, to brush up on the relevant “9th grade physics”?
dp says: May 5, 2012 at 2:01 am
In the experiment they need to turn the columns of water 60º on the x axis to see if the gradient reduces to half. Then to 90º to see of there is an instrument error (gradient ≠ 0), and finally, 180º to see if the gradient is restored to the original value with the sensors opposite in the y axis.
Graeff’s book describes a version of the experiments which does the 180 degree turns. I shall say more on this when I finally do a review for Amazon.
Tallbloke, questions logged. And I shall modify mine in line with comments made here.
TB: Lucy:” this modification is needed to validate Nikolov and Zeller.”
Careful now. Nikolov and Zeller haven’t said their theory depends on the validation of the Loschmidt effect or a modification of the second law…
Point well made, and taken. Strictly speaking, N&Z do not need to be seen as a challenge to the Second Law at all. However, as popularly understood, as both Maxwell and Boltzmann believed (without clear evidence until now with Graeff), as certain formulations state, and as Anthony Watts and others believe, it is thought that the Second Law means that a vertical column of air in equilibrium must have the same temperature top and bottom.
Thanks for flagging this up. My poor language.
Everyone – thank you all for sharpening my own perspective and language – whether or not you agree with Graeff or even consider him worthy of investigation.
dp says: May 5, 2012 at 6:44 am
On giving this a bit more thought a different experiment can be done. Place a sealed column of liquid in a centrifuge…
Been done by Prof Chuanping, but without enough checks. See upthread. Maybe the experiment has been improved since that report.
Would you like to join a team to replicate Graeff’s work here, when I return from Germany?
lgl: Not just ‘my physics’.
http://en.wikipedia.org/wiki/Potential_energy
“The action of stretching the spring or lifting the mass requires energy to perform. The energy that went into lifting up the mass is stored in its position in the gravitational field, while similarly, the energy it took to stretch the spring is stored in the metal. According to the law of conservation of energy, energy cannot be created or destroyed; hence this energy cannot disappear. Instead, it is stored as potential energy.”
So energy is required to give an object potential energy, say by lifting it off the floor and putting it on the table. But Gravity does no work when that object is pushed off the edge of the table and falls. Instead, the object loses potential energy and gains kinetic energy (as it’s momentum increases under the acceleration it undergoes due to the gravitational *force*.
Roy Martin says: May 5, 2012 at 9:17 am |
…When you examine the brown curve in Fig. 2 closely, the first observation is that it is really not level at 0.01 K at all, there are many small rises and falls. (For those in the audience, get out your magnifying glass or enlarge the Fig. if you have to.) Also in Fig. 2, the actual thermocouple readings in the apparatus are shown as a group of smoothed lines, revealing that they all rose and fell together over a range of about 0.5 K during the experimental run….
Roy, you have to look again at this figure, and the following Fig 3. The smooth lines in Fig 2 are not what you say. They are the thermistor readings of the exterior layers of casing, following the scaling on the RHS not the LHS. And Fig 3 is not the same as the horizontal-ish lines on Fig 2. It is all badly labelled, and I had to try and try and try again before I’d understood it myself. That was one reason I decided to buy Graeff;s book, hoping it might shed light.
When I’ve done a review for Amazon, I’d like to try and do a page on Graeff on my GWT website. I did this for the wiki, to make Graeff’s diagrams clearer, among other things, but still I dare not release the wiki url here because some of my stuff there is not yet public or it hasn’t been ok’d with the author. Graeff doesn’t always sell himself well – which, curiously, is more evidence that he is trustworthy – having said which, I have not trusted but have checked the evidence of many experiments now.
http://en.wikipedia.org/wiki/Work_%28physics%29
Roger, I need to inject one additional piece of clarity to what I was saying above. That is about liquid or gases. I tend to completely agree with most commenting, I don’t see this in liquids either, well, some but not just half but more like more than one thousandth or one ten thousandth of what should be seen in gases.
Any velocity sorting which manifests as thermal gradient depends hugely on the molecules being mostly independent of each other so they can accelerate and this is not true in liquids to any degree, and molecules in solids of course can’t move at all so zero. It’s this inter-molecular attraction I see as preventing all but a very tiny gradient in a liquids case. I think most are right here, if it existed in liquids you would see this in deep lakes and oceans immediately and its not there, just the density gradient.
I forgot to tell you that in my examples above, that is speaking of gases.
wayne says: May 5, 2012 at 9:22 am
… I do think if this experiment were reproduced using a gas instead of liquid and scaled up a bit a gravitational gradient would appear close to the lapse rate. Also seems density gradient should play into the equation with solids showing basically zero, liquids a tiny effect, but gases showing the full scale effect and relatively quickly…
Graeff has done the experiment with air – also inhibiting convection. This gave results very close to his calculated “lapse rate” of 0.07K/m. That figure is clearly well over the atmospheric lapse rate – but it is a figure where convection has been inhibited.
Ah, another question coming…
Has Graeff tested air without inhibiting convection, and in this case, are his results close to atmospheric lapse rates?
My feeling (and that’s all it is at this point) is that in gases, the gravitational warming effect outweighs the gas’s capacity to convect – but in liquids, the convection capacity outweighs the gravitational effect, therefore ocean floors are cold. And I also wonder whether this notion of one factor outweighing the other differently in gas and in liquid, is related to what you note about density gradient.
Refining the above question, thanks Wayne. I take your interesting point about water’s theoretical gravitational temperature gradient of 40K/km (which I now suspect is overcome by convection in the unrestricted oceans) and apply it to air instead. Now Graeff’s theoretical and measured figure for air is ~0.07K/m which is 70K/km. But the measured adiabatic lapse rate in free air is more like 7K/km.
Roy Martin says: May 5, 2012 at 2:00 pm
Tallbloke says: “My approach is to be sceptical, but not dismissive. I think that’s the correct scientific attitude. Others may disagree.”
No argument on that principle, but in this case my understanding of the distribution of ocean temperatures both vertically and laterally, and very slow rates of circulation in the deep ocean, does lead me to the point of being dismissive of the results of this Graeff experiment. The temperatures that would arise as a consequence of his results are so far removed from what is observed that the two appear to me to be quite irreconcilable.
If Lucy S. has the opportunity to put this question to Graeff I will be very interested to learn his response.
I’m interested too. Will ask your question as stated here.
lgl says: May 5, 2012 at 1:47 pm
…If mines were not heated from the bottom, the gradient would reverse and they would have been cold. This study actually disproves this whole nonsense. Air density decreases with depth in the warmest mines, where the gradient is greater than 34 K/km
Great reference there Igl. It supports Graeff really neatly.
Your quote actually says Air density increases/decreases with depth in the warmest mines, where the [temperature] gradient is less/greater than 34 K/km.
This shows beautifully that in the convection-challenged environment of mines, you can indeed have air at the bottom at a lower density but hot. If it were in the open, it would rise with convection, and nullify much of that temperature gradient (which at >34K/km is far higher than the free air lapse rate of ~7K/km).
Thanks.
The temperature of a substance is determined by the kinetic energy of the molecules. In a gravity field, this will be increased as the molecules move lower in the field as potential energy is converted to kinetic energy. In the absence of conduction, convection or radiation this will lead to an increase in temperature. The process works in reverse as the molecules move higher in the field.
This is what Graeff has observed, the relationship between kinetic and potential energy in a gravity field, and the resulting effect on temperature.
Surely this result is not surprising. Molecules must gain kinetic energy as they move lower in a gravity field, just as larger objects gain kinetic energy. Molecules must lose kinetic energy as they move higher in a gravity field, just as larger object lose kinetic energy.
This will lead to a higher observed temperature for objects lower in the gravity field, unless the opposing forces are stronger. The idea that this cannot happen because it allows you to build a machine that violates thermodynamics is a nonsense
The energy for this machine comes from the increased energy of the molecules as they fall in the gravity field, which is no different than water falling to drive electrical generators. The energy to return the molecules back to their original position and energy comes from the sun, just like the energy to return the water to the reservoir.
Otherwise, if one was to build such a machine, as work was extracted from the system this would cool the molecules lower in the field and warm the molecules higher in the field, but the warming of the molecules higher in the field would be less than the cooling lower in the field due to the energy extracted as work.
Over time this would lead to a net reduction of the energy in the system, in keeping with the requirements of thermodynamics. Thus there is no perpetual motion and no violation of thermodynamics in Graeff’s observations. What is violated is some people’s interpretation of thermodynamics, but that is not the same as thermodynamics.
Tallbloke – my centrifuge ideas are not presented to be acted on but to be thought about. In each case I presented gravity was replaced with another device to produce different pressures. Thought experiments. Here’s another. Build two identical devices, place on in a cellar, the other in the top floor of a very tall building. Carry out the exact experiments as Graeff has done, note the gradients. Invert both experiments and again note the gradients.
Now reverse the locations of the two experiments and repeat the tests. This is still using gravity, but gravity at two different vertical locations.
I am going to be having hip replacement surgery and retiring shortly thereafter, but when healed and genteel, I will actually be in a position to help with this project. I am particularly interested in the notion that a gravity well is needed vs a simple pressure gradient. If a gravity well is indeed required then this is some serious physics of a new kind, heretofore undiscovered. If gravity is not needed then any well insulated room full of pressurized containers should be warmed by the contents, forever. That can’t happen.
Someone should contact these people and ask for some data, and if they would donate some space:
http://en.wikipedia.org/wiki/Neutrino_detector
To avoid spam, uncheck the “Notify me…” box before posting your comment.
For example, imagine that we turned off the sun and suspended radiation, conduction and convection.
What we have is air molecules at the surface with a temperature of about 300k moving around randomly about 500 m/s on average. Some molecules will shoot upwards and slow by 10 m/s/s. After 50 seconds their speed will be 0 and their kinetic energy will be 0, which corresponds to 0K.
At that point, the molecules then begin to fall back to earth. After 50 seconds they will be back at the surface and their speed will be 500 m/s and their temperature will be back to 300k.
In the absence of radiation, convection and conduction, any substance where the kinetic energy of the molecules is affected by gravity, there must be a temperature gradient due to:
1. conversion between kinetic and potential energy
2. temperature is determined by kinetic energy, not potential energy
dp says:
May 5, 2012 at 4:55 pm
Now reverse the locations of the two experiments and repeat the tests. This is still using gravity, but gravity at two different vertical locations.
==========
The difference in the field is not likely sufficient to show a difference between two locations on earth.
The centrifuge however would be an ideal test, because the gradient is a result of acceleration converting potential energy kinetic energy, while temperature depends on kinetic energy alone.
The problem is probably better understood if we don’t think of it in terms of gravity, but rather in terms of acceleration as Einstein did.
Place a column of water in a space ship with 1g of acceleration and observe the effects. The molecules will have their kinetic energy maximized in one direction, and their potential energy maximized in the opposite direction, with a temperature gradient created as a result.
However, at the same time convection will try and return the molecules with the higher kinetic energy to the opposite end of the column, which will seek to eliminate the gradient.
Another way to view this is that the movement of molecules (conduction) in a gravity field will set up a gradient, due to the conversion of potential energy to kinetic energy. However, the reduction in density due to the increase in kinetic energy will oppose this (convection).
Thus, a liquid of gas that did not change density with a change in temperature would be the most likely candidate to show the effects of gravity in creating a temperature gradient.
More revolutionary physics. This time centripetal force can do work, and air far from the center of gravity must be colder because of lower potential energy. – priceless.
Well said Ferd. you’ve not just grasped the essentials, you’ve put them better than anyone else I’ve seen. Clear and crisp.
Now can I recommend you buy Graeff’s book where you will appreciate how Graeff was inspired to modify his theory by allowing for the number of degrees of freedom in each material tested.
Here’s the first page of Graeff’s paper from 2006:
1. Do you feel that my experimental results would be of interest to you or
somebody in your department?
2. Can you think of any conventional explanation for the negative temperature
gradient measured?
3. Do you know of any theoretical treatise calculating the vertical gradient in
liquids?
4. Do you know of any Physics Department at any University which might be
interested to duplicate my measurements?
He wants intelligent criticism and replication, as per normal science.
That is indeed an interesting question. For discussion proposes lets call this the “ocean floor paradox”
Graeff measures theoretical gravitational temperature gradients in air and water. The adiabatic lapse rate in air approximately agrees with his measurements in a carefully insulated isothermal test environment, with convection suppressed.
This leads to the other unspoken question, is it fair to compare the temperature gradient in the ocean deep trenches with the temperature gradient in the atmosphere?
There are important differences between the gaseous envelope of the atmosphere and the liquid pool of the ocean.
1. First and perhaps the most obvious is that gas molecules have relatively long mean free paths at the sort of pressures that exist in our planetary atmosphere. The mean free path of the molecules in a dense liquid are much shorter. This gives the gravity field much more freedom of action to modify the kinetic energy of a gas molecule than a molecule of a dense viscous liquid.
2. There is a substantial difference in the viscosity of the two fluids, the viscosity of air is approximately 1.78×10−5 Pa.s, the viscosity of water 8.90 × 10−4 Pa·s
3. One is a polar solvent (water) the other is mostly an electrically neutral gas (at least in the troposphere).
4 They are vastly different in density and compressibility. The atmosphere easily compressing (changing density with height in the gravitational field), and as a result mean free paths for the molecules of the atmosphere increases substantially with altitude in the gravity field. Where water on the other hand is nearly incompressible at normal terrestrial pressures and the mean free path of motion of a water molecule is essentially the same from top to bottom of the ocean (ie very short).
5. The atmosphere is transparent to the suns energy and is heated through out its volume from top to bottom. The Ocean is essentially opaque to the suns energy with all of it being absorbed in the top microns to few hundred meters of depth depending on frequency.
6. The ocean has active cooling at its deepest points due to inflow of arctic/antarctic chilled waters or high saline content.
7. The combination of items 5, 6 above gives a very different energy input profile to the two fluids. The ocean is actively heated from the top by the sun and actively cooled at the bottom by cold arctic currents. The atmosphere is heated through out its depth by the sun but is most strongly heated at the bottom (surface), and is most actively cooled at its mean radiating altitude in the infrared of approximately 14,000+ ft altitude. LIkewise item 4 leads to far greater freedom of action for the gravitational effect in the gas than exists in the dense liquid)
8. There are very large differences in heat capacity between the two fluids. It takes much less energy change to modify the temperature or air a given amount than the energy required to modify the temperature of water the same amount.
I submit that the above differences (and others I might be unaware of) make them two very different thermal/physical environments and that the paradox may not be a valid comparison because of these differences.
If others have any other physical differences that should be considered in evaluating the validity of “ocean paradox” please add them to the list.
Larry
Larry thank you, that expands and clarifies what I’ve been reaching for. Plus, as Wayne suggested, and Graeff confirms theoretically, the gradient of air is 0.7K/m whereas water is only 0.4. Therefore, with all the other factors you mention, convection easily wins out over the small gravitational warming effect, to ensure it’s totally negated and hidden in the oceans – though visible in Graeff’s lab experiments.
If one thinks of a gradient as an infinite number of static points uniformly different from adjacent points it becomes possible to consider stacking separate containers, one above the other, to a height known to produce the “gravity well” effect. Each container is pressurized identically – the amount is not important as there is nothing sacred about it. If gravity is at play here the containers at the bottom of the stack should be warmer than those at the top.
The assumption in this experiment is that molecules at the bottom of the gradient have no capability to exchange information with molecules at the top, therefor there is no need to share a container. In fact the gradient is also unimportant. What is important is the location of the sample in the time frame of the gravity well. This is what Ferd alludes to with his reference to Einstein’s acceleration of time in the presence of gravity.
This produces paradoxes, of course. An insulated container protected from environmental air pressure changes should be gravitationally warmed by the simple act of lowering it down an elevator or mine shaft. This is going to be a hard sell.
Hi LS,
Your post of 4:18pm refers, specifically the paper you supplied a link to.
I consider the paper of extremely poor quality and content.
I supply two reasons:
1. The paper mentions varying geothermal gradients yet doesn’t try to explain the reasons for this happening on land. We know that the geothermal gradient on the Wits is 9 to 12 degrees Celsius per kilometre v 22 for the Bushveld igneous complex. The reason is the age of the rocks. The more recent the rocks were deposited, assuming an volcanic or magma genesis, the hotter the deposit.
2. The authors mention the deep gold mines of South Africa, Western Deep levels at 3,6km depth, yet fail to understand the mechanism that is used to cool the workings down. They talk about ‘refrigerated air’ without mentioning how that is achieved. This is glossing over one of the most important technical innovations in deep level mining of the late 1900s, pioneered by a British engineer, Dr Wheeler, which introduced the concept of pumping chilled ice down the mine instead of trying to cool the air before it is allowed to enter the mine.
To this day we measure the amount of cooling sent underground in terms of tons of ice. Dr Wheeler realised that it was better to pump chilled water with ice down the shafts instead of trying to cool air down before it enters the mine. One chilled cubic metre of water has the same cooling than 1000 cubic metres of air (assuming both air and water is at the same temperature) and in addition, extra cooling energy is obtained when ice turns to water.
A study of the subject of the ventilation of deep level mines will show that the gravity induced thermal gradient, also known as the adiabatic temperature increase is well understood and catered for in any deep level ventilation design and system.
JGM: Welcome, and thanks for your specialist knowledge. Could you link an online reference to the ” well understood adiabatic temperature increase”? Or provide a reference to any other publication which has details of how it is differentiated from other causes of heat in mines?
Thanks
Hi Jolly Green Man
Thanks for your comment. You may well be right that the paper is of overall poor content. I only read the abstract, which was sufficient for my particular purpose here, namely the realization that, far from debunking Graeff (the subject matter of the thread) as no doubt Igl had thought, its abstract seemed, to me, to provide implicit further support, in the way I spelled out.
I did find another fascinating reference to deep mines. It was a revelation to me, to read that the temperature gradient varied so much in deep mines, and that whereas limestone did well, coal got hot. This actually seems to put a spanner in the Graeff thesis, at least for the immediate moment.
I would therefore be interested to hear your take on this.
I’m not sure it has much to do with Graeff’s thesis, those highly veriable temperature gradients I would interpret to show that there are much larger heat flows involved in the deep rocks that would totally over whelm a small gravitational gradient.
For example here in Colorado we have lots of hot springs. In the town of Glenwood springs their hot spring pool has two temperature zones, the small “hot” pool has water at 104 deg F, where the larger pool is kept at a comfortable 90 Deg F. The heat provided by the local hot springs. A few miles south of town there is a mountain known as Red Mountain, in the winter time you can see bands across the mountain where all the snow melts off and on cold winter mornings you can see small amounts of steam rising from. In spite of that geological heat source the rock layers above and below those hot bands, are cool enough to allow snow to accumulate on the surface.
Limestone is a “wet rock” and is usually associated with ground water. There would be some evaporative cooling from that moisture not to mention the cooling power of any ground water flowing through the formation.
On the other hand coal is self heating, coal piles will spontaneously heat and eventually catch fire due to oxidation of the coal fines deep in the pile.
Geological heat from volcanic formations and the cooling effect of ground water would totally swamp a trivial temperature gradient such as Graeff supposes exists.
Not to mention the minor detail that all mines are actively ventilated to vent methane and CO2 buildups, and provide a clean breathable air to the miners.
To do a test involving a deep mine you would have to build a specially insulated air column who’s outer cover is actively held at a uniform temperature over its full length to isolate only the gravitational gradient if it exists. Given how difficult it was for Graeff to get an isothermal shell that would not be a trivial task.
Larry
I just had a light bulb moment regarding Graeff experiments and my comments above regarding why the Ocean Floor Paradox might exist.
It has been mentioned above that if you had a completely airless body and a single molecule at rest relative to the body, gravitational attraction between the molecule and the gravitational body would accelerate the molecule as it “fell” toward the body. It would gain kinetic energy (temperature) as it fell toward the body in direct proportion to its original gravitational potential energy it had in its rest position.
This in effect, is the idealized example. No intervening collisions between molecules of a gaseous atmosphere just a single gas molecule free falling toward the celestial body. ( visualize a single perfectly elastic rubber ball).
It is clear from simple high school Newtonian physics what would happen. You start off at position P0 with the molecule at rest (relative to the planet) some altitude above the planet surface (hence some unique and specific) gravitational potential energy in that system, and the kinetic energy of this hypothetical molecule would be absolute zero (if it is truly at rest).
Once you start the experiment that molecule would free fall toward the planet surface accelerating at the local gravitational constant acceleration determined only by the relative masses of the planet (molecules mass being so small it can be ignored). At the instant before the molecule collides with the planet surface Ps it would have a kinetic energy (temperature) exactly equal to its starting gravitational potential energy at P0, and relative to the surface 0 gravitational potential energy. You would have a perfect zero loss conversion from gravitational potential energy to thermal energy of motion (kinetic energy), with a lapse rate (delta T with altitude) directly proportional to the change in gravitational potential energy in that system with altitude.
This would only depend on the local gravitational constant acceleration which is totally dependent on the mass of the planet (ignoring the trivial mass of the gas molecule).
If, as I suggested earlier the observed gravitational warming is hindered by issues like mean free path, viscosity, etc. which interfere with the acceleration of the gas molecule toward or deceleration of the molecule away from the planet, than Graeff’s observed value should approach that ideal limit as the gas in his test apparatus approaches a vacuum (the idealized single molecule case).
That would imply that the effect would be more noticeable at low atmospheric pressures in the test apparatus where the molecules have longer mean free paths and experience fewer collisions with the test apparatus walls relative to their paths along the tube’s axis.
If you could some how constrain the gas molecules so that they never hit the side walls and only moved along the long axis of the test volume, they should approach that theoretical ideal Delta T with change in gravitational potential energy of the hypothetical single molecule case.
Hope that makes sense? This is one of those light bulb thoughts just as you are drifting off to sleep.
Larry
Minor addition, the density of the planet only determines the location of the surface reference point, but the mass determines the gravitational acceleration in that local system. If for example we had a planet with a very deep hole in it exactly in line with the free fall of that single molecule, the molecule would accelerate at the inverse square of its distance from the center of mass until it passed the surface of the planet and began to fall down that deep hole. The further into the planet it fell the lower the gravitational acceleration rate due to the over burden above it towards the surface canceling out the gravitational attraction of the mass still remaining below.
Larry
Hi Hotrod: Gedanken experiments have value, and inform theory. Theory has value and enables predictions. Here we have two conflicting theories. Maxwell and Boltzmann believed sensible heat would be isothermal in an isolated column at equilibrium due to the random distribution of molecules of different temperatures. Loschmidt and Lagrange and Laplace thought total *energy* would be randomly distributed, and due to the partition of potential and kinetic energy, that there would be a thermal gradient as you describe, due to changes in the ratio between the quantities of PE and KE at differerent altitudes.
Roderich Graeff didn’t just *suppose* there is a gradient as Loschmidt theorised. He set up experiments and tested the theory to see if it’s predicitons conformed to observations. That’s the point of this thread. As Einstein said: “Experimentum summas judex” – Experiment is the final arbiter. Maxwell and Boltzmann were theorists who didn’t make the step of confirming predictions by experiment. Part of the science is missing.
Several commenters here have dismissed Graeff’s results a priori. This isn’t good enough. Graeff’s experiment requires replication, and examination of possible instrumental bias. To dismiss his results on the basis that the ocean doesn’t exhibit a cool at the top warm at the bottom gradient is to ignore not only the buoyancy of warm water relative to cold, but also that the ocean is heated from the top, and that the atmosphere does in fact exhibit the gradient as predicted. Simply making an argument by assertion that we should “back off from Graeff” (and ignore his results) isn’t good enough either. I for one don’t want to remain in ignorance as to the truth of the matter. It is crucially important for the correct understanding of temperature distribution in climate systems.
I pointed to the Sheehan critique of Graeff’s methodology and results three times yesterday, when it became clear no-one had bothered to visit the links to it in the original post. Out of the several hundred people who visited this thread, and the dozens who commented, four people looked at the first page, two at the second, and one at the third. Now, one of our group is taking the trouble to travel to Germany to meet the experimenter and discover more about his methodology and experience. This is to be applauded. Here is a person not content to sit on a chair and draw conclusions from insufficient knowledge.
It’s easy to sit in a chair and pontificate from the basis of assumed knowledge or quick judgement. Much, much harder to determine the existence or otherwise of a subtle and difficult to measure effect by experiment.
We don’t know what gravity *is*. We don’t know what energy *is*. We can only observe how these ontological entities *act* in relation to each other, and try to draw out and construct, by inference and deduction, a framework of description to aid our attempt at understanding. The greatest impediment to this process is the high sounding authority of those like Robert Brown the Duke university physicist and Willis Eschenbach the construction manager who believe they already know the answer because Maxwell handed it to them without the need to follow Einstein’s injunction and perform confirming experiment.
Maxwell argued by assertion and appeal to his own formulation of the second law. It’s not good enough.
Experimentum summas judex.
Experiment is the final arbiter.
http://en.wikipedia.org/wiki/Geothermal_gradient
http://www.sci.uidaho.edu/scripter/geog100/lect/05-atmos-water-wx/05-part-7-atmos-lifting-fronts/ch5-part-7a-atmos-liftin.htm
Lucy, Larry, et al.
Wayne in a comment at 3:58 gave a link to a png showing decreasing gravitational field strength with depth below the Earth’s surface.
So perhaps the ocean floor paradox does not threaten to invalidate the gravitational thermal effect, since in the case of the ocean the molecules at the top of the ocean experience a greater gravitational attraction than the molecules at the bottom of the ocean.
In other words, in the case of the ocean, the gravitational effect is effectively in reverse, and we should expect the ocean surface to be warmer than the ocean floor.
J Martin: Don’t forget the crust, oceans and atmosphere are in an extremely thin layer compared to the radius of the Earth. The falloff in the gravitational force with altitude from the sea bed is negligible. So yes there is an effect due to the conditions you outline, but it is magnitudes smaller than the effect we are discussing.
Sorry for being late, and perhaps repetitive. Has Mr. Graeff turned is apparatus on its side or upside down? I did not see those options mentioned in his pdf.
Hi Terry, yes, he did the end for end thing to check his thermistors weren’t introducing bias. Sheehan says he didn’t test at other angles. Lucy will find out more in Germany.
Links to methodology critique by Sheehan:
Tallbloke
It is sometimes hard to determine exact meaning of comments on blogs, because of the lack of secondary information such as tone of voice, facial expression etc. But as I read your comments, it appears to me that you think I was minimizing Graeff’s experiment or even perhaps dismissing it.
I am not I think they are (assuming good procedure which they appear to exhibit) are a very important contribution. My use of the word “suppose” was simply recognition that his experiment is not widely accepted and to my knowledge has never been replicated so at the moment he is making an unverified assertion.
I believe it is a very important assertion and likely correct from what I understand of his experimental technique. My discussion of the ocean floor paradox is not to diminish his experiment but to try to explain what I think is an important aspect of the science. Farther up the thread this paradox was used to summarily dismiss the results of Graeff’s experiment. I think that is wrong but it might be very useful to understand the exact mechanics of gravitational heating of an atmosphere.
What I was trying to point out in my post last night, is that there is another test that can be performed in replicating Graeff’s experiment that might provide useful additional information. I also made a prediction that could be verified by that test.
That test would involve using different pressures for the interior volume of the test column to see if mean free path of the molecules is important to the process. If I am correct, it is and the more rarefied the gas in the test volume the more closely the results will approach the theoretical ideal of a perfect transfer of energy from gravitational potential energy to thermal kinetic energy along the columns length. If that is true, then the ocean floor paradox is not a paradox at all but a demonstration that the gravitational temperature gradient is not readily apparent in liquids and solids due to other aspects of the physics such as viscosity or lack of atomic mobility in the solid.
On the other hand I might be totally wrong and the experimental results might be immune to changes in the gas pressure or species, length to width ratio of the column or a dozen other complicating factors. Even if that is true we have learned another important fact.
If it is demonstrated that mean free path of the gas molecules is important, and that secondary processes like viscosity modify the result, then it is clear that the gravitational thermal gradient is only of significant consequence in a gaseous atmosphere, where its effects are easily over come by other forces in the deep ocean. That does not mean it does not exist in the deep ocean only that it is trivial compared to other processes at work in liquids. It would also show that there is not paradox in the deep ocean only a lack of understanding of the heating process.
I applaud Graeff’s attempt to experimentally verify what Loschmidt theorised. I would be doing the same experiment myself if I had the means. Since I don’t, I can best contribute by suggesting other variations of the experiment that might help eliminate objections to the theory (such as the apparent paradox of the deep ocean).
As is often the case in science a paradox is often not a paradox at all, but only demonstrates a lack of understanding on our part of some important but subtle issue at work.
Larry
Hotrod: my apologies, by no means all of my comment was aimed at you. And we all agree Graeff’s work needs replicating. Graeff himself was the first to say so.
Terry, yes Graeff has done 180°inversions, see upthread.
Larry
thanks, I think you’ve got some interesting ideas. It all takes time to filter out the rubbish and distil the essence. I looked up the densities of air, water, and steam at 100°C. Air is ~1.2 kg/m^3 at lab temperatures. Water is ~1000 kg/m^3. Steam is ~0.6 kg/m^3. Huge factor of expansion means that gravity has a lot more space-freedom to take effect, with gases, or so it seems to me. Also if you want to understand the difference between water and air, I think you need to come to terms with the issue of “degrees of freedom” in molecules that Graeff writes about. It repays the effort of study well. You start to get a feel for the difference between Newton’s apple falling and individual gas molecules falling, and how simple it is in the end to predict lapse rates. Actually it looks embarrassingly simple, in the end, to me. And this is perhaps the reason that so few academicians want to know. So many scientists can’t have missed something so obvious.
But they will be like those publishers who turned down J K Rowling. IMHO.
Hi Tallbloke,
My apologies for being so tardy with my reply, I was out, walking in a ‘deep dark wood’ with my grandson, and that takes preference in my life.
There is indeed a paucity of information regarding deep level mining ventilation on the web, which is not surprising, as mining has such bad press in the West. The Deep Greenies have vilified the mining profession with the result that there is now a shortage of BYT (bright young things) joining the ranks of our profession.
When I retired a few years ago, I started following the global warming polemic and realised that the field of mining ventilation offers quite a number of answers to the issues that were being debated. I searched the web for an up-to-date mining ventilation handbook with very little success. I did manage to track down an online handbook written by Malcolm J. McPherson and he addresses the topic of autocompresion by gravity. (I can’t find the link at the moment as I did it two or more years ago) A quote from his handbook follows.
….The process of gravitational compression, or autocompression, in the downcast shaft produces
an increase in temperature of the air…
For reference in his handbook on this specific chapter he mentions.
….Bibliography
Professor F. Baden Hinsley of Nottingham University, England, was the primary
architect of the development of mine ventilation thermodynamics. He wrote many papers
on the subject. Only the first of his major papers is listed here.
Hinsley, F.B. (1943). Airflow in Mines: A Thermodynamic Analysis. Proc. S. Wales Inst. of
Engineers, Vol LIX, No. 2.
Hall, C.J. (1981). Mine Ventilation Engineering. Published by the Society of Mining Engineers of
AIME.
Hemp, R. and Whillier A. (1982). Environmental Engineering in South African Mines. Chapters 2
and 16. Published by the Mine Ventilation Society of South Africa.
McPherson, M.J. (1967). Mine Ventilation Engineering. The Entropy approach. University of
Nottingham Mining Magazine, Vol. 19.
McPherson, M.J. and Robinson, G. (1980). Barometric Survey of Shafts at Boulby Mine,
Cleveland Potash, Ltd., Cleveland. Transactions of the Institution of Mining and Metallurgy,
Volume 89, 1980, ppA18 – A28…
Hope this helps.
Hi JollyGreenMan
I guess from your name that you (and I) are True Greens, rather than Quicksand Watermelons. If one actually follows the true green thread to its conclusion, one has to face the fact, whether one likes it or not, that there is a place for mining. So one might as well like it. And this needs green/sensitive input, but with active involvement from inside, as “Small is Beautiful” Schumacher did in the coal industry, rather than potshots from outside. Schumacher would I think turn in his grave to see what his name has been appropriated to. I tried to explain to Schumacher College about anthropogenic global warming as a dangerous betrayal of Science… a few bleeps then contemptible silence. Twice. Alas. Probably never heard of Nullius In Verba.
Now, my interest in the mining data that Igl flushed out, was only insofar as it related to Graeff and Nikolov/Zeller and the work of all these worthies saying that pressure/gravity induced warming is the major factor in Earth warmth. One thing that interested me was the statement that the density gradient was zero when the temperature gradient was 34K/km – despite far higher pressure at depth. Probably just a straight application of the Gas Laws, but it flagged up one enormous piece of evidence. 34K/km is huge. Adiabatic lapse rate in free air is ~7K/km. Graeff’s theoretical AND measured gradient is 70K/km as folk above rightly point out.
Larry has put forward some excellent thoughts as to why limestone mines are cool where coal mines are hot. Someone else said about snow bands on the mountains, corresponding to hot rock layers underneath. All fascinating, all building up the real Climate Science, which I think supports Graeff in geo-meteorology in the elements of Earth and Air, with a plausible exception for Water – and all with fascinating modifications. So, generally, descending into the Earth will increase temperature because of both increasing rock pressure and increasing air pressure, as per Graeff and N/Z. But the solid dimension allows huge local variations. And the constrictions of the mineshafts, lessening neutralization of the gravitational warmth by convection, also raises the lapse rate. Result: half the maximum air figure as calculated by Graeff. Huge compared with outside. But as your good geologists realize, directly attributable to pressure, not the fairy greenhouse gases.
Enough for now, eh!
I’m a little late. And I didn’t read all the comments carefully, but two points pop out at me.
1) This paper is trying to address gradients in perfectly insulated containers. As such, it has little to do with the real world.
* The atmosphere is heated at the bottom (by the warm ground that absorbed solar radiation) and cooled at the top (by IR radiation to space). As such, a temperature gradient is to be expected.
* The oceans are cooled at the poles, where dense, cold (~ 0C) water sinks to the bottom. There is a bit of heating from the bottom (geothermal), but that would only slowly warm the ~ 0C water that had settled to the bottom. And new, cold water is constantly convecting down to maintain the low temperatures at the bottom.
Since there are large energy inputs and outputs, the temperature gradients theorized for perfectly insulated containers are not really applicable to either the atmosphere or the oceans.
2) I see nothing in the paper about the thermometers other than vague references to “Type E thermocouples” and “thermistors”. They are claimed to be precise to 0.1 C, but I see no mention of their accuracy. I see no mention of calibrating the the thermometers in some uniform temperature bath to see if they indeed read the same.
I see no reason to conclude anything more than “The thermometers are not calibrated and initially are only accurate to +/- 0.3 C. They maintain their initial mis-calibrations quite consistently for several months.”
I suspect that even if the thermometers were embedded in a block of highly conducting metal, the readings would still differ by ~ 0.3 C. I would want to see information about the calibrations of the thermometers themselves before putting any stock in the small differences reported here.
Hi Tim,
1) It’s very important in the real world. If a gravitationally induced gradient in temperature underlies the lapse rate, the radiative greenhouse theory has suffered another body blow.
2) Read Graeff’s paper more carefully, it tells you about the precision of the measuring equipment. This is more important than the accuracy if you are measuring differences over time.
“4. Precision of measurements
Thermistors have a precision of only about +/- .1 K, not sufficient to measure gradients of
.01 K/m without making difficult additional calibrations. In the above measurements
thermistors are used not for reliably measuring gradients, but in order to establish the
changes of temperatures at different locations. The precision of measuring these
temperature changes is better than .001 K/hour.
Type E thermocouples are used to measure the temperature gradients as the difference of
the voltages between two thermocouple points. Connecting these in series, one obtains
thermopiles. The values reported here are more than 10 times the precision of individual
thermocouple measurements.
Actually, the precision of the absolute value of T(Gr) (20 or 30% higher or lower) is not as
important as deciding, whether the direction of the temperature gradient is positive or
negative. But this direction can be decided upon to a very great precision, because the
zero offset of the instrument can be determined measuring each value twice, the second
time with switched polarity.”
3) Read Sheehan’s critiques.
4) Apologies for the terse reply, but I’m getting tired of armchair critics who don’t read the presented material.
Lucy: “Graeff’s theoretical AND measured gradient is 70K/km” (for air)
Well, you’ve read his book and I haven’t, so maybe Graeff discusses the ‘degrees of freedom’ more there. I find the information in the paper inadequate, so if the book doesn’t cover it, I’d like to make that issue one of the top priorities for questions to him.
Here’s what he says in sections 2 and 3 of the paper:
“The potential energy
is converted only into an increase of their speed in their lateral downward direction while
no energy is used or distributed in accordance with the equipartition of energy to the
other degrees of freedom like the additional two lateral directions left to right and front to
back or towards the rotational energy in molecules with more than one atom.”
“Calculations, described in section 2, show that water, having a higher specific heat than
air and a greater number of degrees of freedom — 18 compared to only 5 for air — should
give a temperature difference of about -.04 K/m. “
Tim, I can sympathise with the problem of scanning a long thread. Got it all the time at WUWT. It is a weakness of science-by-blog, and in that sense I wish someone here would emulate Skeptical Science blog’s structure (NOT its contents!!) whereby the head post is altered in response to comments – and it is easily accessible later via a Blog Contents By Popularity of Issues page, whereas here and at WUWT it gets lost in the swamps of time.
DO read and re-read Graeff’s paper. DO pay close attention to his words and diagrams. DO realize that here is a retired engineer who designed and manufactured engineering goods with his own successful company, who in retirement has for over ten years been running these experiments, who was originally alerted by unexpected anomalies, who developed novel precision ways of testing those anomalies. DO realize that Graeff, together with Nikolov and Zeller’s rewriting of the Stefan-Boltzmann equation so as to apply correctly to sunlit spheres, have between them fashioned the true and necessary scientific nails for the coffin of the greenhouse gas warming stuff (that I used to believe too, long after becoming a climate skeptic). I say “nails for the coffin” because with both Graeff and N&Z, their theories MATCH their experimental results, with both precision, elegance and logic, in a way that Climate Science as-is does not begin to match. These findings are like Cinderella’s glass slipper. They fit. And continue to fit, in ongoing experiments. Eppur Si Muove.
Kelvin, Clark Maxwell, and the rest of the nineteenth century science magi did their best. And these two at least were also good men. I’ve taken the trouble to study and meet them across the centuries and appreciate where they too stood. Only then did I feel I had the right to join Graeff in his challenge to their detail (not the essence) of the Second Law.
When you can stand in Graeff’s shoes, then you can criticize. His paper has poor areas of presentation, which led me to buy his book, which helped immensely to clarify things, both humanly, engineering- and precision-wise, and scientifically.
Tallbloke
I am still wrestling with the “degrees of freedom” material, and why air has 5 and water 18 (I confirmed this at wikipedia). Graeff even in the book admits he maybe hasn’t explained clearly enough. He’s woken up some days with clear intuitions that work, and has had to try to explain to others later. But I see something of the essence of the issue. There is more than I’ve said below, but if I put it all in, it will distract and overwhelm too much.
When molecules are warmed classic style, let’s say by a Bunsen burner, they start to speed up in all directions. Now the directions in which they can speed up are precisely mathematically determined for each substance, and are called “degrees of freedom”. For instance, a single atom can speed up along three linear axes, x,y and z; it can also speed up in spin along (?) three more spin axes equivalent to x, y and z. Molecules get more complicated. There seems to be buzz factors between molecules too (?) I don’t fully understand. All I know is that it makes profound sense, and is relevant because gravity only operates along one of the “degrees of freedom” axes, whereas a Bunsen hits them all. This affects the relationship between the specific heat of the material as normally measured, and its “gravity-induced” specific heat, which has to be the normally-measured specific heat divided by the number of degrees of freedom.
Lucy: Thanks for the reply, and intuitively, it makes sense that gravity wouldn’t affect the torsional wiggling of molecules to any important extent, so maybe we can leave Graeff in peace on this issue.
Being triatomic, water is a good deal more complicated than diatomic molecules such as N2 and O2 and can vibrate in many more different ways.
Tallbloke,
Don’t get me wrong. I am impressed by the efforts Graeff made to insulate the columns. The alternating layers of insulation & thermal conductors is indeed close to ideal.
But there is more that could and should be done (or at least reported). It is not my fault that the details are not in the paper.
* Type E thermocouples are rated to +/- 1.7 C (about +/0 0.1 mV). Connecting them backwards in series will mostly cancel the signals, leaving the difference. But since the two thermocouples could produce thermoelectric voltages that are ~ 0.1 mV different even at the same temperature (because they are not manufactured identically) , they could still easily read ~ 1.7 C difference even with no temperature difference.
From what I can tell, his efforts to reverse the polarity are aimed at the voltmeter offset, not the thermocouples themselves. There should be a calibration measurement made with both thermometers at the same temperature (say in an ice-water bath or in a block of aluminum).
* It only takes ~ Q = k A Delta(T) / l to set up a temperature gradient. For the water in his experiment, this amounts to ~ (0.6 J/m*K) (0.0012 m^2) (0.03 K) / (0.85 m) = 25 microwatts. Yes, 25 uW at one end of the water column will match or overpower the effect he is trying to measure.
hmmm … thermistors typically have a resistance of ~ 3000 ohms. A typical DMM might use 100 uA to measure this resistance, generating I^2 * R = 30 uW of self-heating. Yes, the heat generated by one of the thermistors itself is about enough to swamp the desired signal (Assuming it was run continuously).
There would be heat transferred through the wires leading to the various thermometers. The point is that even minor details are enough to interfere with the measurements here.
Lucy,
“Degrees of freedom” are indeed tough to grasp. It doesn;t help that quantum mechanics rears its ugly head as well. For example, N2 has 5 degrees of freedom at room temperature, but 7 at higher temperatures! http://en.wikipedia.org/wiki/Heat_capacity#Example_of_temperature-dependent_specific_heat_capacity.2C_in_a_diatomic_gas
As for this being a “nail in the coffin”, I think there is still WAY more experimental work needed to overturn “textbook physics”. It’s great that people are taking such interest in physics, but you will have to excuse me if I don’t accept the word of someone who doesn’t understand “degrees of freedom” when it comes to making pronouncements on the foundations of thermodynamics.
One more comment, and then I may be done with this thread.
To set up a gradient of 7 K/km in air requires a heat flow of Q = k A (delta T) / l, which would be about 0.2 mW per square meter.
EVEN IF such a gradient were the equilibrium adiabatic condition (which I disagree with), then heating of the order of 0.0002 W/m^2 would be enough to mask this effect. Given that typical heat flows in the atmosphere are ~ 10,000 times larger, the N&K temperature gradient would never be observed other than in very careful laboratory settings.
Larry, Let me add one more thought to your “light bulb moment”.
* If there is a single atom in a column, and if the collisions with the base never transfer any energy, then your conclusion would apply. Of course, then the idea of “temperature” would hardly apply.
* Consider instead that the atom interacts with the base when it collides. The atom will leave the base with varying amounts of energy after each collision (ie it will have a MB distribution of energy). If the atom has above-average energy, then it will reach above-average altitude before gravity stops it. If the atom has below-average energy, then it will reach below-average altitude before gravity stops it. The atom reaches different heights during different trips. (This, in essence, is why the atmosphere is denser near the surface)
Now suppose we want the temperature at 1000 m altitude. The low energy trajectories never get there, so they are not counted (ie the “cold” atoms remove themselves from consideration). The high energy trajectories of the atom will get get there. Granted, they lost some energy on the way up, but they had above-average energy to begin with. It is completely plausible that the average KE at 1000 m is the same as it was at 0 m. The power of self-selection is considerable in this situation.
Earlier I wrote i rejected the PDF but I did not say anything about Loschmidt, just the PDF. This did not go down well. In effect kicking a design engineer does not go down well but these things happen, nothing I can do without breaking ethics.
My position on Loschmidt is I don’t know and if there is an effect it is very likely dominated by other factors. I go further and state I’d be surprised if gravity did nothing at all.
Several things have become clear.
So far as I can tell none of you are experienced with instrumentation.
A very obvious change is switch from the cheap domestic methods to serious. The critical measurement is being taken single ended which brings in many errors. Ever wondered why thermocouples have a cold junction carefully kept at a known temperature? Temperature measurement using thermocouples is a differential measurement between two thermocouples, hot and cold.
Obvious now. Don’t bother messing around trying to calibrate two, just calibrate one pair for a differential measurement.
Thermistors? Feasible if that is what is wanted, ways around most things. Historically I will use almost anything to measure temperature, if it works it works.
The following is hard to do.
What I think is needed is a new apparatus with very specific properties and for a reason.
Given the signal is large (at least to me) it can be physically much smaller.
I suggest truly good thermal insulation is considered, hard vacuum containers.
One of the very hard things is the interface connections, needs a lot of thought and care. (radio ain’t worth it, keep it simple)
This needs to be built as a package of a sensible size and weight.
It ought to be ovened, ie. elevated and controlled inner temperature.
The whole thing must be gas tight.
It must be sensibly robust, no liquid slopping.
The whole thing is battery powered, not as bad as it might seem. (it is insulated but might need forced circulation)
So why all the environmental stuff?
Difficult to do in the UK, but it is going to be driven up and down mountains repeatedly. or a deep mine shaft. How long it needs to stabilise, depends, can’t know now.
The wanted signal is the effect of gravity variation with altitude, small but definite.
I think this is too elegant an apparatus without outside funding. Moreover anything found has to be broken by the experimenters, actually try and falsify. This would take time and a great deal of patience. Needs discussing.
By a strangeness of fate the last design I contracted was a complete temperature measurement system, embedded.
I am all too aware the world at greatly magnified temperature is much the same as the world you know, gradients are everywhere, microcells are everywhere, things you don’t notice become huge. You are brave.
tchannon says:
May 8, 2012 at 2:26 am
Earlier I wrote i rejected the PDF but I did not say anything about Loschmidt, just the PDF. This did not go down well. In effect kicking a design engineer does not go down well but these things happen, nothing I can do without breaking ethics.
Hi Tim: It was a three line dismissal containing no supporting argument.
Several things have become clear.
So far as I can tell none of you are experienced with instrumentation.
Actually, the designers at CERN took note of my comments when we determined the angles of the drillings for the probes in the main body of one of the central components of the Large Hadron Collider which I machined for them. I didn’t have anything to do with the electronics though. I’m more a materials and geometry engineer.
The critical measurement is being taken single ended which brings in many errors. Ever wondered why thermocouples have a cold junction carefully kept at a known temperature?…Thermistors? Feasible if that is what is wanted
Graeff used both, at various heights in various parts of the test rig, according to the PDF.
From section 3.1:
“The temperatures inside the test setup are measured by thermocouples and by thermistors.
These are mounted at the tops and at the bottoms of the inner axes of the two glass tubes.
Additional sensors are mounted on the outside of these glass tubes and on the outside of
the two PVC tubes. The temperatures of the double wall aluminum housing are measured
3 cm below the top and above the bottom.”
Thanks for the rest of your comment, it’s good that more people are thinking about how to devise experiments to test the effect Loschmidt theorised, and Graeff pioneered the attempt to empirically determine. It’s worth noting that Graeff has also used other methods according to Sheehan’s critique. According to Lucy, Graeff is currently getting a 0.02K/m gradient in a column of air which doesn’t have convection suppressed.
Your proposed experimental setup won’t do the job though, because by putting the experiment inside a hard vacuum, sealed from the environment, and taking readings at different altitudes, you have eliminated the conditions under which Loschmidt supposed the effect would operate. In effect, you are setting up the same fallacy Robert Brown did with his ‘sealed jars’ gedanken experiment which he thought disproved the Jelbring hypothesis. You would, in theory, measure the change in gravitational force. But this is a tiny effect several magnitudes smaller than the proposed Loschmidt effect, which crucially depends on the changes in pressure caused by the action of gravity on the mass of the atmosphere. Your setup eliminates those changes in pressure.
I’ll do you a deal. You leave the mechanical design to the mechanical engineer, and I’ll leave the instrumentation to the electronics expert. We can argue over just how we interface the two. It’ll be just like old times with the CERN guys. 😉
Physics is far too important to leave to the physicists. 🙂
Hi Tim “TJFolkerts”
Thanks for weighing in again. I appreciate your not taking my word for things, but don’t just do it because I said I didn’t understand “degrees of freedom” too well, do it because it is good scientific practice. I’ve been working on those degrees of freedom, and understand them better now. But I could not, from Wikipedia (incl. your ref) or anywhere I could find online, find tools to verify that water has 18 degrees of freedom. Can you recommend something? I will of course ask at the seminar.
Thermocouples etc. I think your issues deserve raising. I will copy your words exactly so that my limits of understanding do not change your meaning. I understand that Graeff chose thermocouples to measure temperature gradient only, so it did not matter that there was no base measurement. Thermistor heat effect: in his book, Graeff said he had avoided thermistors earlier on, due to concern about this, but when he tested it, found the effect was negligible. Note, he only used thermistors on the outer layers, and in order to get a baseline against which to compare temperature fluctuations and especially negative gradients of the inner layers.
I disagree.
I see four factors at work (1) incoming SW radiation that is absorbed high up and translated into kinetic energy (2) surfaces absorbing radiated heat and heating adjacent air by conduction (3) convection, which, up to the tropopause, largely overcomes the effect of (4) gravity, which causes molecules to accelerate in freefall. You have to work with statistical probabilities of momentum being transferred in collisions so that the net effect resembles freefall – though this is not quite the same, as freefall momentum has to be distributed equally to all the degrees of freedom.
I’d always puzzled over the “W” shape of the atmospheric temperature profile. Now, in light of the above, it finally makes sense to me. All the atmosphere planets have a tropopause at ~0.1bar (except Mars which is altogether too low). Different factors dominate at different altitudes / pressures, under different conditions.
Where does the CERN thing come from?
Rog, you haven’t got what differential means. A direct measurement of difference, not computed from separate measurements using different thermometers.
tjfolkerts says:
May 7, 2012 at 4:59 pm
“…* It only takes ~ Q = k A Delta(T) / l to set up a temperature gradient. For the water in his experiment, this amounts to ~ (0.6 J/m*K) (0.0012 m^2) (0.03 K) / (0.85 m) = 25 microwatts. Yes, 25 uW at one end of the water column will match or overpower the effect he is trying to measure.
hmmm … thermistors typically have a resistance of ~ 3000 ohms. A typical DMM might use 100 uA to measure this resistance, generating I^2 * R = 30 uW of self-heating. Yes, the heat generated by one of the thermistors itself is about enough to swamp the desired signal (Assuming it was run continuously).
There would be heat transferred through the wires leading to the various thermometers. The point is that even minor details are enough to interfere with the measurements here.
I agree that all that is needed to get a false positive is a small source of extra energy being pushed into Graeff’s well designed experimental set-up which has not been considered. However, I think this could be the intrusion of Earth’s varying EM photon charge field which is adding the energy to the bottom of the system. Unfortunately, reversing the tubes will make no difference to the result as the energy will still be coming in at the ‘new’ bottom.
`Tim C, What do you mean “Where does the CERN thing come from?”? It’s an anecdote concerning the interaction of people with engineering expertise with people with instrumentation expertise. It hints at the benefit of collaboration across disciplines which respects the knowledge of other people.
“Rog, you haven’t got what differential means. A direct measurement of difference,”
On the contrary Tim, I know exactly what ‘differential’ means. And so does Graeff. As he tells us in the PDF
“Temperatures within the test setup
As already discussed in the section “Environmental influences” the smooth curves (9-14
in Figure 2) represent temperatures measured by the thermistors. In comparing the
measurements at different locations, one has to consider that the precision of a thermistor
amounts to only +/- .1 C. But the measurements are very constant over time, as indicated
by the smoothness of the curves, whereby the temperature change over time is measured
to a much greater precision than the absolute values.
9
This fact becomes very important, when one looks at long time periods, during which the
upper and the lower temperatures in a tube do not change. During these times one can
decide, whether a temperature gradient TGr exists under equilibrium conditions.
But, because the
temperature fluctuations in the environment are different at different heights of the test
setup and the various insulation materials are never totally identical over height, the
temperature changes and their timing may be different at the top from that of the bottom.
Therefore, T(Gr) fluctuates over time around the correct average value. In order to obtain
this correct value, one has to measure over longer time periods.
T(Gr) can be found very efficiently, when all measured temperature gradient values are
plotted as a function of the rate of the temperature change (Figure 4). We measured these
rates both at the top and at the bottom of the tubes, and found very similar results.”
Your earlier assertion that only the measurment at one end was used and that Graeff and I don’t understand differentials is falsified.
On another matter:
Do you accept that your proposed setup would measure the inverse square falloff of gravity, but not the Loschmidt effect. Or not?
I’ve not yet read thru all the subsequent comments, but let me observe here:
A column of gas is different from a solitary steel ball in a number of ways; e.g., in that as a unit it is at the median height of the column, not the top.
Depending on what means by “work”, it is possible to recover work from a steel ball “at” 2 m. above the floor, as it didn’t suddenly materialize there. Work was performed to raise it, and can be (mostly) recovered by having it (e.g.) weigh down the end of a rope wrapped round a pulley as it drops. For the column of air, the kinetic energy of the molecules is so much higher than their potential energy individually, unless they are very cold indeed, that it’s problematic to do a parallel operation; it would require, perhaps, extracting a molecule at 0K and dropping it through a vacuum column.
But if the column is in a vacuum chamber, allowing it to escape through a nozzle (at either end of the column) would perform work. Etc. In other words, gas has energy in gravitational potential, thermal, and kinetic (pressure) forms, and the kind of work you can get out of it is determined by the “sink” you arrange, which depends on the “frame”. In a sense, work is contextual and relative, as is the “energy” available in a particular physical frame. Change your definition/selection of frame boundaries, and you change the energy content and potential.
E=mc^2, but ‘c’ compared to what?
I keep getting a twitch when I contemplate such questions; temperature is a bulk effect, a measure of impacts between particles. Until impact occurs, there is no temperature. (Yet atoms are said by some, per S-B, to radiate depending on temperature, etc. This can’t be right. S-B is a bulk effect, defined for surfaces given specific constraints.) When the hammer/feather duo contacted the surface, momentum became heat. One micron above the surface, their temperatures surely were unchanged from the moment of release.
Say what? … because of higher potential energy.
There, fixed it for you.
Lucy,
I also saw this http://www.pha.jhu.edu/~broholm/l37/node5.html#SECTION00011030000000000000 with a discussion of experimental and theoretical confirmation of 18 degrees of freedom for water.
“I understand that Graeff chose thermocouples to measure temperature gradient only, so it did not matter that there was no base measurement.”
That is good experimental technique, but it is not sufficient. If the instruments used for the two measurements are not identical, then they will measure a difference even if there is none. It is very possible that the two thermocouples will measure different values even at the exact same temperature. This would give a value for the gradient even if the temperate was uniform.
Calibrating the thermometers should be done. Rotating the apparatus occasionally should be done. These simple checks would give a LOT of information.
“I see four factors at work (1) incoming SW radiation that is absorbed high up and translated into kinetic energy (2) surfaces absorbing radiated heat and heating adjacent air by conduction (3) convection, which, up to the tropopause, largely overcomes the effect of (4) gravity, which causes molecules to accelerate in freefall. “
But now you are working with a completely DIFFERENT situation. We are discussing the idea that EVEN WITH NO EXTERNAL HEATING OR COOLING, there would be a gradient. As I pointed out before, even very small energy fluxes are sufficient to change the gradient. So yes, these sources & sinks of energy can easily swamp any effect that might be hypothesized to exist merely from gravity.
Tim F
(It is confusing that we apparently have two “Tim”s in this discussion)
To all those concerned about notification spam: once you’ve made a comment with a “tick” on a thread, subsequent ticks have no effect. To stop notifications, you have to go to your WP control panel and cancel them for that specific thread, and then avoid ever requesting notifications there again.
My solution has been to filter incoming mail with “new comment” or “new post” into a dedicated folder, and sample and purge there when convenient, thus leaving my Inbox uncluttered. (Except, of course, by stuff from all the other sources I’ve injudiciously subscribed to in the last many years!!)
ferd, I’m really been enjoying your comments and thinking. But you spoiled it all by evoking that non-expansible gas. My brain pains trying to envisage it! How would the energy of impacts be increased to raise the temperature? Adding Higgs bosons from the Void to each molecule?
😉
Yes, an ordinary Thermos bottle of coffee or cold water, lowered quickly enough to render conductive losses immaterial, should be enough to measure it. With a radio thermometer immersed in the liquid?
😉
Yes, it seemed to me that the instance of Jericho (‘250m below sea level’) was a better “real-world” exemplar of the effect. Unless one were to posit/find that the ground (soil/rock) surface were at a different temperature to begin with, from the surrounding sea-level+ towns.
😦 🙂
Heh. ‘contemptible’, or contemptuous? Probably both, IMO: contemptible contemptuous silence …
🙂
😉
😀
tjfolkerts says:
May 8, 2012 at 1:32 pm
Calibrating the thermometers should be done. Rotating the apparatus occasionally should be done. These simple checks would give a LOT of information.
Yes. Graeff did this. READ THE SHEEHAN CRITIQUES
We are discussing the idea that EVEN WITH NO EXTERNAL HEATING OR COOLING, there would be a gradient. As I pointed out before, even very small energy fluxes are sufficient to change the gradient. So yes, these sources & sinks of energy can easily swamp any effect that might be hypothesized to exist merely from gravity.
This is why Graeff ensured there was always a reverse gradient around the test equipment.
Roger,
You said “This is why Graeff ensured there was always a reverse gradient around the test equipment.”
I was specifically responding to Lacy, who was discussing the actual atmosphere, where large sources and sinks of energy do indeed exist. Graeff did seem to do a pretty good job insulating the equipment.
When I look at the further comments, I do see some of the issues that I raised getting addressed. However, now we are looking at the results third-hand, (a report about a report about an experiment). Finally, I see the conclusions
>> “not yet conclusive”
>> “disputed by other researchers”
>> “should be replicated”
There seems to be a lot of work yet to be done. And a lot of details to be reported
* When the samples were flipped, how long did it take for the gradient to re-establish?
* When the samples were flipped, was the outer insulation also flipped?
* When the samples were flipped, were the thermometers flipped?
* What specifically were the locations of all the thermometers, and how were they wired?
There are lots mroe questions that could be raised, especially if one saw the actual apparatus.
This work is a HINT at some new physics.
[Reply] Yes Tim. This is why Lucy is travelling to Germany in a week’s time to talk to Graeff directly. I have been appealing for people to take an interest in replicating Graeff’s work for months.
Hi tjfolkerts,
I have been interested in Roderich’s work for a year and a half now, and admit to being totally stumped as to how to explain his results, either from a theoretical point of view, or finding a flaw in his experiment. If you can contribute to either, I would greatly appreciate it. It was discussion of Roderich’s work that brought me to this forum.
In Roderich’s favour I find:
1, He started out measuring the temperature difference just with thermocouples, so no external energy input,
2, He surrounds the inner chamber with both highly insulating layers, and with highly conductive layers,
3, The temperature differential at the outside of the setup is warm on top and cold on bottom, whereas the temperature differential between the top and bottom of the highly conductive layers is pretty much zero.
4, Hence there can be no transmission of an external temperature gradient to the inner chamber, which is cold on top and warm on bottom,
5, He does upside-down tests where the entire apparatus including thermocouples is inverted. The gradient restores itself on the timescale of about a day (depending on which setup he uses). This can be repeated as many times in a row as you like. That the thermocouples are inverted with the setup, and yet the gradient reestablishes itself, shows that it is not merely a bias of the measurement. I have stressed the importance of these tests to Roderich.
6, By trying different liquids and gases, he has achieved a result of -0.3 K/m. This gradient also flips with the upside-down test and the magnitude is far greater than instrumental noise,
7, The data is logged for months on end using a Keithley2700, corrected for meter bias.
8, Prof Liao repeated the experiment in a centrifuge and found gradients proportional to the ‘g’ of the centrifuge.
Counter thoughts are:
9, No real theory to explain the gradient. The argument about molecules slowing down with height is what brought me to Roderich’s work, but I (and very many others) have shown that this argument doesn’t work for a Maxwell-Boltzmann distribution (which is what one would expect of an insulated gas/liquid in equilibrium). The argument only works for non-MB distributions, such as you have when there is convection or winds bringing molecules vertically through a gravitational field. In contrast, Roderich finds much higher gradients when convection is suppressed.
10, An apparent breaking of 2LoT. I’m pretty sure that if the claim is correct, then you can indeed get work from it. Of course, if true then this would be amazing, and I try not to get carried away by such a thought. I have made several recommendations to Roderich to try and increase the power out of his machines, as he appreciates that you would need more power out to make them more convincing.
11, There is some temperature fluctuation in the environment. While this is only on the order of one or two K and seems to be quite stable for long periods, one can’t rest easy while there are heat flows around. I don’t know how this could generate the observed differential, but then I don’t know how anything else could either.
I have tried a quick and dirty centrifuge replication, but so far have only managed to measure wind-chill factor.
In summary, I can’t explain his results so would appreciate any insights. I look forward to Lucy summarising her visit.
Rog,
I’m trying to figure out how to explain what is a radical misunderstanding. Doesn’t look feasible so I’ll drop it for the time being.
Lake Vostok, Antarctica
Pressure: 350 atmospheres (eq. 3.5 km water)
Graeff temperature: 50(?)C/km*3,5km=175C
Actual temperature: -3C
lgl said:
“Lake Vostok, Antarctica
Pressure: 350 atmospheres (eq. 3.5 km water)
Graeff temperature: 50(?)C/km*3,5km=175C
Actual temperature: -3C”
I don’t think this is a counter-argument: Graeff doesn’t claim that the temperature is proportional to pressure, but that the temperature difference is proportional to height difference of the fluid under gravity. Height in this case I guess would be the height of the liquid water in the lake. Ice may have a totally different gradient (the gradient being material and presumably phase dependent). Seeing as the lake is surrounded by ice, one also has to take conduction into account and what the temperature of the ice above and below the lake is. The relevant question would be whether there is a temperature gradient in the centre of the volume of the lake? Average lake depth is 344 m so there should be room for a gradient to form. What the gradient ‘should’ be at a pressure of 350 atmospheres, and whether this is the same as for 1 atmosphere, is unknown.
Note also that Graeff measures gradients such as 50 K/km when convection is suppressed using for example glass powder. When the fluid is free to convect, the gradients become much less (<10 K/km).
br1
Well, it should have been 175C at the surface and increasing with depth, also with convection.
If it’s about height it’s even more strange. (0.000000something difference at 1 m)
Conduction is a good point, but if the energy flux is that low, this issue has no relevance with regards to the atmosphere anyway, with the ~500 W/m2 flux at the surface.
I think chemistry, perhaps electrochemistry, is the place to look.
b t w http://en.wikipedia.org/wiki/File:EarthGravityPREM.jpg
A few thoughts off-hand, and then I must leave this little thread.
EVEN IF IT WORKS, we are talking about microwatts. You would need millions of these to power even one house. Even with scaling up, this would be one of the most inefficient ways ever to generate useful energy.
EVEN IF IT WORKS, this is not what drives the gradients in the atmosphere or the oceans or mines. This effect is only expected to be seen in nearly perfectly insulated systems. Atmosphere and oceans and mines all have much larger energy sources/sinks involved, and the gradients are due to other reasons.
So from a practical perspective, I don’t think there is much future to this.
############################################
The signals are microvolts. There are vertical atmospheric electric fields on the order of 100 V/m. There are 60 Hz AC signals all over (50 Hz in Europe IIRC). Ground loops can introduce stray voltages. Any of these could induce currents and voltages in wires unless great care is taken to eliminate such factors.
The thermistors used in the experiment produce sufficient heat to generate such a gradient. The wires used to connect the thermocouples could transfer considerable thermal energy (metals are usually excellent thermal conductors). Stray RF/microwave/AC EM waves could provide some tiny heating.
The results contradict a time-tested law of physics. It is very uncommon to overturn such laws using basic equipment in a very mundane setting. It could happen, but the odds are long and require extremely careful proof.
So from a physics perspective, I don’t think there is much future to this.
[Reply] The gradient Graeff has isolated is not small. If it does represent something actually occuring in the free atmosphere, the other gradients we hypothesise about are much bigger than previously thought. This is not unimportant.
tjfolkerts says: May 8, 2012 at 4:16 pm
I was specifically responding to L[u]cy, who was discussing the actual atmosphere, where large sources and sinks of energy do indeed exist. Graeff did seem to do a pretty good job insulating the equipment.
Indeed. Greaff’s work is laboratory-limited and very precisely engineered. But not applied to climate science. That’s what I was doing, and it was pure speculation based on what I DO know, and where Graeff’s theory suddenly helps me make sense of a lot in basic Earth climate.
Sorry if that confused things. It’s because half my interest in Graeff is in putting climate science on a better footing – and suddenly I am getting “light bulb” moments of insight, often as I write to respond here.
[Reply] Eliciting light-bulb moments is what this blog is about. Keep ’em coming.
What Tim Folkert writes subsequently probably speaks for others too.
When I look at the further comments, I do see some of the issues that I raised getting addressed. However, now we are looking at the results third-hand, (a report about a report about an experiment). Finally, I see the conclusions
>> “not yet conclusive”
>> “disputed by other researchers”
>> “should be replicated”
There seems to be a lot of work yet to be done. And a lot of details to be reported
* When the samples were flipped, how long did it take for the gradient to re-establish?
* When the samples were flipped, was the outer insulation also flipped?
* When the samples were flipped, were the thermometers flipped?
* What specifically were the locations of all the thermometers, and how were they wired?
There are lots mroe questions that could be raised, especially if one saw the actual apparatus.
This work is a HINT at some new physics.
I cannot recomment enough that people asking questions like these obtain Graeff’s book for themselves. Dang it, anyone who’d like it but feels they cannot afford a copy, please let me know and I’ll post you one.
Graeff’s book answers almost all Tim’s queries above. I don’t want to keep saying this each time someone asks. Graeff’s results appear to be extremely conclusive – a lot more conclusive than you can see from just the one paper with URL here. His results are NOT disputed by other researchers. He wants to be replicated and begs for support in this. Who is willing to help with replication? This is what I intend to do when I return home. Please email me if you are interested.
There are lots of photographs and graphs in the book. Several more experiments. But most important, you come to meet Graeff the man who as an 11-year-old child was already wanting to test these things – and who survived the unthinkable horror of Hamburg being destroyed by fire-bombing, with huge loss of lives in horrible ways.
A major passion that drives Graeff is the hope of harnessing the “peaceful” “free” energy he has shown exists, so that wars need not be fought over this essential commodity. But he recognizes all too well that (a) he has only produced it in minute quantities albeit reliably and consistently (b) it has not yet caught on because people think it denies the hallowed Second Law – IT DOES NOT DENY THE SECOND LAW, but it does call for a modification.
Tim F
Thanks for ref to water’s 18 degrees of freedom. I’ve spent these last three hours getting acquainted with U, s, n, N, Cv (C), R, k_B. and many half-familiar concepts, including the edge of quantum mechanics, to my delight.
For years I’ve been self-taught, really very goal-oriented, so now this is on my horizon I shall master it as much as is needed. So as not to look an idiot! so as to understand people’s questions!
Roger,
The energies involved in these gradients are indeed small. I estimated we are discussing mW/m^2. The experiment in question involves microwatts of energy to maintain the gradient. Putting a 100 microwatt = 0.0001 J/s heater at the top of the column would destroy the observed gradient (and make the top warmer than the bottom). A good AA battery is ~ 10 W*hr of energy, so I could erase they gradient for 10^7 seconds = 1/3 of a year with the energy in a single small battery. Even if I am off by a couple orders of magnitude, we are talking about being able to “overpower” this generator with a tiny battery for a day.
Anyone else is free to try running the numbers themselves.
[Reply] While they are at it they might count the number of assumptions made. 😉
br 1 says:
6, By trying different liquids and gases, he has achieved a result of -0.3 K/m
I don’t recall any that high. That would be 300K/km.
9, No real theory to explain the gradient. The argument about molecules slowing down with height is what brought me to Roderich’s work, but I (and very many others) have shown that this argument doesn’t work for a Maxwell-Boltzmann distribution (which is what one would expect of an insulated gas/liquid in equilibrium). The argument only works for non-MB distributions, such as you have when there is convection or winds bringing molecules vertically through a gravitational field. In contrast, Roderich finds much higher gradients when convection is suppressed.
Graeff’s book does talk a lot about theory to explain the gradient, and how it came to him in successive intuitions. And it makes profound sense to me. Too much to repeat here.
10, An apparent breaking of 2LoT. I’m pretty sure that if the claim is correct, then you can indeed get work from it. Of course, if true then this would be amazing, and I try not to get carried away by such a thought. I have made several recommendations to Roderich to try and increase the power out of his machines, as he appreciates that you would need more power out to make them more convincing.
NO NO NO how many times must it be said, Graeff’s work does NOT break the hallowed 2LoT! Nor would I trust it if it claimed to do so. It does not claim to do so. But it does claim for a modification, to recognize the presence of an external force that is usually ignored at the molecular level ie gravity.
I don’t understand your reference to the Maxwell-Boltzmann distribution, apologies.
When you “made recommendations” had you read his book? He is all too aware that the electricity actually generated is microscopic. Nevertheless, the hope of future recognition and amplification of his work is what inspires and drives him. Not recognition for personal kudos or anything, heavens sake he is 84. No, he is passionate to help humankind find a peaceful source of energy. But Graeff’s results need understanding, replicating and accepting before work can proceed to try and develop amplifiers.
Lucy:
“I don’t recall any that high. That would be 300K/km.”
Well, it is not in his book (which I have bought and read). I went to his meeting (!) last year, and this was his latest result. I have only met Roderich once, but just from that one meeting would regard him as a friend, so I am on your and his side. My main problem is that I can’t reconcile his results with any physics I know.
“NO NO NO how many times must it be said, Graeff’s work does NOT break the hallowed 2LoT!”
ah, well that may be your opinion, but I don’t see it that way. According to his data, different materials have different gradients, and these gradients establish themselves over time. Hence it is possible in principle to run a Stirling engine from the temperature difference across two columns of different materials and extract work from a heat bath. Roderich himself claims to extract power from his thermocouples, and by implication extract power from a heat bath. No work is consumed as gravity does not run out, and yet work is produced.
“I don’t understand your reference to the Maxwell-Boltzmann distribution, apologies.”
MB distributions are meant to describe the range of velocities that molecules in a gas or liquid have in thermal equilibrium. With such a distribution of velocities, a column of gas will have the same temperature with height, even when under the influence of a gravitational field. You may find my posts on this thread from here on interesting:
In particular, the post after that one gives links to some results I got for the temperature of a MB distribution with height. My interest with DALR in that thread is a direct consequence of Graeff’s work. When the gas molecules do not have a MB distribution, then the gas can have a different temperature with height. Convection provides a non-MB distribution, as there are local net drifts in one direction, whereas a MB distribution has no net drift.
Tim C: No worries.
“Truth springs from argument amongst friends.”
– David Hume –
tjfolkerts:
“The signals are microvolts. There are vertical atmospheric electric fields on the order of 100 V/m. There are 60 Hz AC signals all over (50 Hz in Europe IIRC). Ground loops can introduce stray voltages. …Stray RF/microwave/AC EM waves could provide some tiny heating.”
Yes indeed. What gets me here though is that Roderich has many thermocouples placed at various points in the system, top and bottom and all through the different layers of insulation. The results seem consistent with real temperature measurements (for example, the outer thermocouples show a hot on top and cold on bottom trend, while the aluminium shell shows no temperature difference, the inner ones show cold on top. Intermediate thermocouples show intermediate gradients). He has also run many ‘double’ tests, where in the centre of the insulation there are two containers, one with convection suppressing powder, the other without. These produce different gradients, so if they are picking up some electric fields, they are not doing it equally (simply as a function of height, for example).
However, the point can’t be dismissed. How would you recommend going about testing such possibilities?
Some statements from Wikipedia articles below:
“The reason that it is used in both fluids is that changes in pressure result in warmer fluid residing under colder fluid- examples being the fact that air temperature drops as one climbs a mountain and water temperature can increase with depth in very deep ocean trenches and within the ocean mixed layer. When potential temperature is used instead, these apparently unstable conditions vanish.” …
and
“Potential temperature is a more dynamically important quantity than the actual temperature.” …
And what is ‘potential temperature’? The ‘effective’ temperature as seen from within a gravitational field with pressure and density gradients and heat capacity. In the oceans you also have to take into account the increasing salinity density with depth.
If you can, try to correlate some of this information:
http://en.wikipedia.org/wiki/Deep_ocean_water
http://en.wikipedia.org/wiki/Ocean_thermal_energy_conversion
http://en.wikipedia.org/wiki/Potential_temperature
Seems some trying to dismiss this experiment trivially might try doing some reading before you start making some ignorant and baseless assumptions.
Also the statement “…temperature can increase with depth in very deep ocean trenches…” is very curious. Could be this effect be showing the Loschmidt effect though countered by other factors? Further experiments will tell.
In a real setting three or four major effects all have to be taken into any equations showing what is occurring in the oceans or the atmosphere vertically.
I still say that this effect in liquids is so very slow due to the high viscosity, the inter-molecular attraction, that Graeff could be correct, his experiments took months to show this vertical sorting, but the gradients in gases it seems should be much faster even though the overall quantity of temperature gradient is lower if his experiments end up proving correct. Also in liquids the pressure differential with height is a counter-influence as compared to gases (our atmosphere) where pressure differences have a much higher gradient per height.
• The density of matter increases with depth into a gravity well except the case when water hits a maxima at 2°C.
• The potential energy of matter decreases with depth into a gravity well. If the kinetic energy did not increase to compensate any column of mass would have a natural increase in total energy with height per molecule. Entropy says this cannot be true.
• Liquids have higher density than gases, about 1000:1.
• Liquids have higher thermal conductivity than gases, about 10:1. Thermal conductivity is the tendency to equalize any differences in the kinetic energy spatially.
• Seawater has lower salinity with height. Salinity at great depths is about 3+% increasing the density with depth gradient even more than merely the gravitational compression.
There are more but all of these effects must be included in a natural settings of oceans and atmosphere before comparing to a pure and simplistic experiment as Graeff performed for the effect he is testing can easily be countered by other natural effects to the contrary. Right now I think I will stay neutral on liquids for the speed of the effect is so slow.
Igl:
“I think chemistry, perhaps electrochemistry, is the place to look.”
You could also be on to something, and is something I have thought about. But I can’t pin down what it is, and why the gradient reasserts itself when the whole setup is inverted.
The closest I got was that for a water measurement, imagine the container is very slightly leaky. It might be leaky due to the way the thermocouples are fed into the container. This leak will result in loss of water vapour through the leaky cap of the container and could conceivably produce a temperature profile as the ‘hot’ vapour escapes from the top of the system. When the container is inverted, the cap would be underneath and could leak water, which would reduce the partial pressure in any space at the other end of the container, which could cause evaporative cooling. So you could conceivably get a cool temperature ‘on top’, no matter which way up the container is.
However, that argument won’t work for the air experiments which show an even higher gradient!
br1 says:
May 8, 2012 at 10:23 pm
Many thanks for your input, much appreciated. I’m taking my time to reply as you raise several issues for which I need to inform myself better before I feel I can respond adequately; also I have other biz to attend to, but will be back 🙂
Trick says:
May 3, 2012 at 7:04 pm
Willis says at 6:26pm:
“I give up, Trick. Come back when you have learned the difference between convective mixing and molecular diffusion.”
Ok, I have learned and come back. But I already knew free convective mixing occurs away from the surfaces in Fig. 1 and molecular diffusion dominates near the surfaces because of the boundary layer effects driving fluid velocities down.
Willis says:
“See the part about “common misconception” there? …If you don’t understand what “common misconception” means, please come back and ask.”
Exactly Willis. I’ll ask & as usual Verkley et. al. paper gives the answer. The common misconception being talked about in your clip is THE big deal now settled by Verkley et. al. paper part 2b, continuing past your clip with more of the paper’s points:
“Exner pointed out that the confusion arose from defining the problem in an inconsistent way…(Maxwell) discusses the classical formulation of the problem and its answer (the profile will be isothermal) but then he argues that convective motions…will be important…. Bohren and Albrecht…consider an ideal gas in a gravitational field, and seek the state of maximal entropy… result in an isentropic profile. This can be regarded as a confirmation of Maxwell’s idea… This brings us outside the domain of classical thermodynamics, and hence one can expect that the temperature profile will no longer be isothermal; we will derive below what profile forms the outcome.”
In part 2b, Verkley paper derives “what profile forms the outcome” showing the common misconception of an isothermal profile must be replaced by a non-isothermal, isentropic profile.
Hi Tallbloke,
Could you post a link to the Verkley paper, please? I would like to familiarise myself with all the approaches to this problem.
Sure, here you go:
Click to access verkely2004.pdf
I suggest the notion of gradients can be removed from the equation. In the tests, two gradients exist – one is the gradient of acceleration of time in the presence of gravity with a length of the test subject, and the other is the pressure gradient created as a consequence of using a column of fluid in the presence of gravity. Using non-columnar test subjects (think dimensionless – analogous to an isotropic radiator in radio) at different vertical locations in the gravity well should produce the same results. Consequences of the test subject being dimensionless require some clever thought but that is what this entire process is about.
I am also curious as to why the test subject must be a fluid vs a solid except that it has been convenient in Graeff’s analysis to generate the gradient in the test subject.
br 1 – Glad you have met Graeff. And thanks for making me think – though I ended up feeling satisfied with Graeff and not sure I need your theories.
The 0.3K/m temperature gradient you said Graeff reached, I still doubt you have the decimal point right, unless Graeff was testing Xenon. This is Graeff’s theoretical temperature gradient for Xenon, but only Xenon – due to high atomic weight and being a monatomic gas with only 3 degrees of freedom.
The Stirling engine is surely the wrong engine, even in theory. We are talking about electricity flow not a heat pump, and the current Graeff can generate is of the order of 10^-12 watts. Nanoscopic. The Second Law is not violated – this is what happens: (1) gravity sets up a thermal gradient (2) this causes a microscopic electric current to flow through a wire connecting the hot and cold extremes (3) by conservation of energy, the thermal gradient is lessened in consequence (4) but this thermal gradient will tend to be reinstated, just as, for instance, the ventilating fins at the back of the fridge dissipate the heat extracted, allowing more to be extracted.
Maxwell-Boltzmann distribution…
I looked it up. It seems to me that the Maxwell-Boltzmann distribution curve is unnecessary to involve here. I start with the fact that Graeff has consistently produced results that will not go away, albeit tiny. And I progress to the fact that Graeff’s theory and maths make total sense to me. I’ve improved my understanding of thermodynamics to include degrees of freedom and a whole way of talking about heat (“thermodynamics”) that is not the language of my studies 50 years ago, although it was already developed by Maxwell and many others of that time and ilk.
Tallbloke,
You are mus-interpretting Verkley once again
In 1a) we have the standard isothermal result for a system in equilibrium.
In 1b) he is looking specifically at systems with active convection. For instance, we could set up two tall, insulated pipes connected at the top and bottom. Then set up a small fan to circulate the air. The air rising up one pipe will cool as it expands; the air falling in the other pipe will warm on the way down as is compresses. As long as you keep the convection going, this non-isothermal condition will continue, and will be the “dynamic equilibrium” solution.
The convection must be rapid enough so that we can ignore conduction (and/or IR radiation and/or evaporation/condensation) as a means of significant energy transfer.
1c) is in between. Convection is present, but conduction (and/or IR radiation and/or evaporation/condensation) becomes significant.
I leave you with one key quote from the paper:
Of course, the actual atmosphere is subject to processes like convective mixing. They prevent the atmosphere from ever coming close to thermodynamic equilibrium, that is, the ultimate state of maximal entropy.
In other words, maximum entropy thermodynamic equilibrium isothermal.
Oops … that last line should read:
In other words, maximum entropy = thermodynamic equilibrium = isothermal.
Lucy says: “(4) but this thermal gradient will tend to be reinstated [by gravity]”
And there is the problem. You had just invoked Conservation of Energy, and I will invoke it now. Where does this energy come from to reinstate the thermal gradient? Gravity is not doing work.
BTW, any method of extracting energy from the system will run into similar problems. The thermocouple is one way to extract “high quality” energy from the “low quality” thermal energy. A sterling engine would be a different way to achieve the same result. Just make the columns tall enough to get a 20 K temperature difference between he tops of two different columns and it should be easy to run a sterling engine.
tjfolkerts says:
May 9, 2012 at 7:10 pm
Tallbloke,
You are mus-interpretting Verkley once again
In 1a) we have the standard isothermal result for a system in equilibrium.
In 1b) he is looking specifically at systems with active convection. For instance, we could set up two tall, insulated pipes connected at the top and bottom. Then set up a small fan to circulate the air. The air rising up one pipe will cool as it expands; the air falling in the other pipe will warm on the way down as is compresses. As long as you keep the convection going, this non-isothermal condition will continue, and will be the “dynamic equilibrium” solution.
The convection must be rapid enough so that we can ignore conduction (and/or IR radiation and/or evaporation/condensation) as a means of significant energy transfer.
1c) is in between. Convection is present, but conduction (and/or IR radiation and/or evaporation/condensation) becomes significant.
2b or not 2b? That is the question.
Lucy:
“Graeff has consistently produced results that will not go away”
I agree, and that is what I find fascinating about it.
Unfortunately (for me), I’m still trying to find a satisfactory explanation. But I take that as fun!
Be sure to ask Graeff about his 0.3 K/m when you meet him 😉
Lightbulb moment, when I take Verkley together with Graeff and Trick who says:
Yeeees. Now it aaaall seems to make sense. Verkley says, IIUC, in air you get a balance between isothermal and isentropic, a result that agrees well apparently with measured adiabatic rates. Now add Trick’s details “convective mixing occurs away from the surfaces… and molecular diffusion dominates near the surfaces” only translate all the above now into Graeff’s experiments.
OTOH, we have the thermo-gravity effect: fallen molecules will buzz faster with more kinetic energy, AND get squashed in tighter by gravity pressure of molecules above
OTOH, we have the opposite effect of convection: bottom-heated molecules will want to expand, will gang up in packets, and rise where they can expand more freely, whereupon they cool down again of course by both having expanded and having lost kinetic energy
Where the fluid molecules are freer to mass together and “be themself”, groupthink away from the surfaces, density issues win out – top-bottom in oceans shows a positive heat gradient.
Where the fluid molecules are close to lots of “surface” (fine glass powder) the thermo-gravitational effect dominates, and top-bottom shows a negative heat gradient quite close to Graeff’s theoretical value.
In the open atmosphere, the balance points are constantly fluctuating, owing to day/night, rotation, poles/tropics, in addition to the above forces of dynamic equilibrium
H’mmmm… where else do we see dynamic equilibrium… electrons spinning? planets rotating?? the whole of astrophysics???
Tim folkerts: Gravity is not doing work.
Well, something is doing the work every time Graeff inverts his apparatus and the gradient reestablishes itself. That’s why I chose the refrigerator analogy.
Just make the columns tall enough to get a 20 K temperature difference between he tops of two different columns and it should be easy to run a sterling engine.
Double problem in practice (a) Height needed = 500m (b) the effect is still slow to build up.
I’d like to think Graeff’s gravity-harvesting could be made to work. After all, both hydroelectric power and tidal power are forms of harvesting gravitational effects. But it is a loooooooong way from producing any usable power.
Lucy:
“Where the fluid molecules are close to lots of “surface” (fine glass powder) the thermo-gravitational effect dominates”
Now that raises a thought – I wonder if capillary effects might have anything to do with it. I don’t know the fine details of Roderich’s setup. Imagine the inner container is filled with powder, but not quite filled with water. The water will try to ‘climb up’ the powder by capillary action. This could indeed produce a cold-on-top temperature gradient, as the capillary action is similar to convection for these purposes.
It would eventually come to a stop though, whereas he has apparently measured a consistent gradient for months on end, but it is a thought I didn’t have before.
He has also measured gradients in non-powder setups, which counters the above argument, but these are ‘small’.
Trick says: May 1, 2012 at 2:46 pm
Willis says at 12:51pm:…
Trick says: May 3, 2012 at 7:04 pm
Willis says at 6:26pm:…
Where are these quotes from, Tallbloke?
[Reply] WUWT’s recent Svensmark thread. Starting here: http://wattsupwiththat.com/2012/04/24/svensmarks-cosmic-jackpot-evidence-of-nearby-supernovae-affecting-life-on-earth/#comment-973163
Lucy Skywalker says:
May 9, 2012 at 11:19 pm
“…I’d like to think Graeff’s gravity-harvesting could be made to work. After all, both hydroelectric power and tidal power are forms of harvesting gravitational effects. But it is a loooooooong way from producing any usable power.
Hi Lucy, don’t think gravity-harvesting would work as the temperature difference which drives the thermocouples warms the cool end of the cylinder while while cooling the warm end. My best guess is that it is the Earth’s EM photon charge field which causes the Graeff observation, and it may be better to find a way of tapping directly into this low level energy source directly, rather than using it by proxy.
BTW, the renewable energy driving hydro comes from the solar EM photon charge field which drives the water cycle which in turn creates the potential energy to be used for generation of electricity. I think that Mile Mathis is also right in concluding that tides are driven by the charge fields of both sun and moon and not, as is currently assumed by ignorant physicists, by gravity. Otherwise the vast amount of energy involved in the tidal work we observe would be an example of a perpetuum mobile!
A couple of observations.
1) A lot of hydro schemes don’t ‘create’ energy at all. They soak up spare capacity from fossil powered stations pumping water up a hill overnight. Then the water runs back down through the turbines to generate additional power at peak demand times.
2) The Moon is receding from the Earth. The standard explanation is the friction of tides slowing the earth, and causing a transfer of angular momentum to the Lunar orbit. Tim Cullen positid the possibility that it is something to do with the shedding of material from the lunar surface in its magnetotail into Earth’s magnetosphere.
3) The idea of a perpetuum mobile arises out of the assumption that the orbiting bodies conserve their angular momentum. It goes back to Newton’s conception of the ‘innate force’ belonging to planets. There are problems with this in terms of accelerations between perihelion and aphelion according to Miles Mathis. There are further problems when you consider the age of the solar system, and the ongoing synchronicity of the orbits via perturbation theory which hasn’t (and can’t) solve the many body problem.
4) My working hypothesis is that electrodynamic coupling and cybernetic feedback does exist between the planets orbital parameters and the solar output via solar wind strength. If so, the variation in the water cycle is driven by the Sun’s activity, but in a direct sense predominantly, through evaporation, changing cloud albedo, surface temperature etc. Miles’ ‘fundamental E/M field’ may have a place too.
Tenuc:
“My best guess is that it is the Earth’s EM photon charge field which causes the Graeff observation”
Not sure I follow you here. What exactly do you mean by ‘the Earth’s EM photon charge field’? Is this just the emitted EM radiation due to surface temperature?
The entire setup is in Roderich’s basement (hence the fairly stable temperature), and the thermistors/thermocouples consistently measure higher temperature on top of the outside of the setup (that is, the room temperature increases with height). One would then expect the ceiling temperature to be greater than the floor temperature, so any photon flux will heat up the top of the setup more than the bottom. This sounds like the opposite to what you describe.
Inside the insulation, the temperature is cold on top and warm on bottom. How would any photon field get inside the setup, and not affect the outside? Remember also that there is an aluminium shell to get through! (which by the way, has the same temperature on top as on bottom)
Just uploaded a review of Graeff’s book for Amazon UK.
I hope this starts to give people here a better idea of its worth.
tjfolkerts:
“The signals are microvolts. There are vertical atmospheric electric fields on the order of 100 V/m.”
The inner chamber is within an aluminium shell, and the measured temperature gradient of that shell is zero. DC atmospheric fields inside that shell should be well shielded, and the thermocouples inside the shell should not be affected. The thermocouple wires all enter the setup at the same point. How could the atmospheric electric field affect thermocouples and thermistors?
“There are 60 Hz AC signals all over (50 Hz in Europe IIRC). Ground loops can introduce stray voltages. Any of these could induce currents and voltages in wires unless great care is taken to eliminate such factors.”
50 Hz and ground loops can provide stray signal, but how could they explain a consistent offset away from zero which takes a day to re-establish itself when the system is inverted?
“The thermistors used in the experiment produce sufficient heat to generate such a gradient.”
No, they are only on for a few microseconds once an hour. The data is logged at one point per hour, and the monitor currents are off in between. Roderich tried first with thermocouples only because he was aware of this objection, but added thermistors when he found that they made no difference to the result, added negligible energy (as they are nearly always off), and added a *validation* to the gradient as he could now measure it in different ways, both of which agreed with eachother.
“The wires used to connect the thermocouples could transfer considerable thermal energy (metals are usually excellent thermal conductors).”
The wires to all thermocouples/thermistors enter the insulation at the same point, hence there is no different external thermal condition. Any different thermal transport would be related to *internal* temperature differences (which shouldn’t exist in theory, so would still need explaining!). Thermocouple wire is very thin. If there was any significant thermal transport, how could it cause an inversion in measured temperature when the setup is inverted?
“Stray RF/microwave/AC EM waves could provide some tiny heating.”
The thermocouple wires are all very tightly bound together so there are no significant antenna loops. The measured values are consistently one-sided (ie, it is not like there is significant ‘noise’ giving zero-crossings). And again, when the setup is inverted it takes about a day for the gradient to re-establish itself. If it was RF pickup, it should either invert instantly when the setup is inverted, or else stay one-sided, it would not take a day to re-establish itself. Again, the inner chamber is within an aluminium shell which wil to some extent act as a Faraday cage (at least more so than for the outer thermocouples). If there was pickup which somehow heated the thermocouples, thus heating the fluid, then a persistent gradient which inverted when the setup inverted would only be explained by having different pickup in different thermocouple joints (which are in the same electric circuit) which caused convection, yet Roderich measures higher gradients when there is glass powder in the container which suppresses convection.
So in summary, I still don’t get it. However, I respect your suggestions and this has been puzzling me for over a year now, so any light you can shed would be appreciated.
br1
Thanks for those details. There are certainly ways to minimize the problems I mentioned. The sorts of things you mention are important and are part of good experimental technique. Without actually seeing the set-up, it would be nearly impossible to guess ahead of time which might be important (or to guess what preventive actions the experimenter might have implemented).
##################
One other thing that keeps jumping out at me is statements like “yet Roderich measures higher gradients when there is glass powder in the container which suppresses convection.”
There should be no convection in these tubes (either with liquids or with gases). Bulk movement of fluids is only expected if there is some energy supply (either differential heating or something like a fan). Without such energy input, viscous forces within the fluid and drag between the fluid and the walls should eliminate bulk motion in fairly short order.
tjfolkerts:
“There should be no convection in these tubes (either with liquids or with gases). Bulk movement of fluids is only expected if there is some energy supply (either differential heating or something like a fan). Without such energy input, viscous forces within the fluid and drag between the fluid and the walls should eliminate bulk motion in fairly short order.”
good point again.
Roderich was explaining it by admitting that the values of gradient he was measuring were greater than the ALR and hence are unstable. That his scale is that of a small bottle (he says he has measured a differential in containers as small as 333 ml plastic drink bottles) maybe should not make much difference? That is, let’s put a very gentle heater at the bottom of the bottle so that the fluid for sure convects – would the gradient be that of the ALR or not? Which would imply that any gradient greater than the ALR would be unstable, even in a small volume? I’m sure scale does become important at some stage, but I don’t know what the volume/surface area value might be.
The thought that an insulated and static (stagnant) gas/fluid should ‘self-convect’ goes against physics training, but that is the hypothesis put forward. Maybe that in itself can be independently measured – suspended particles monitored by a CCD camera? That would require some care, as the container still needs to be well insulated, and even illumination for the camera could cause heating. I have a hard time visualising how a stagnant fluid can self-convect – would the motion need to be the full-volume convection cell type flow, or might there be lots of little convection cells dispersed throughout the fluid. It all sounds rather unlikely, but there must be some way to get an answer to all this.
Seeing as Lucy is heading over to Roderich in the next week, what experimental details would you ask her to check out, that may clear up some of the issues you have raised?
Lucy,
Your findings and comments from your trip in detail are eagerly awaited.
HI TB, I am back from York, and the Bronte part of the world. I would like to put some facts on the table. Here is an excerpt from a handbook prepared by AMC, a well-known and respected Australian mining consulting company: –
….Auto Compression: – As air travels down the intake airways form the surface, its elevation
decreases. There is a corresponding conversion of potential energy into enthalpy. The
magnitude of the change in enthalpy can be estimated using the steady flow energy equation
for a flow from a higher elevation (Z1) to a lower one (Z2), assuming no heat flow and no work
done:
H2 −H1 = g(Z1 − Z2 ) Equation 2-3 Auto Compression
Where:
H = enthalpy (J/kg)
Z = elevation (m)
g = acceleration due to gravity (9.81 m/s2)
The enthalpy thus increases by 9.81 kJ/kg for every 1,000m decrease in elevation. For dry air,
the thermal capacity is 1.005 kJ/°C and the theoretical dry bulb temperature increase is 9.81
kJ/kg/1,000m ÷ 1.005 kJ/kg°C = 9.76 °C/ 1,000m. In other words, the temperature of dry air
flowing down a dry 1,000m shaft into a mine would increase by 9.76 °C (assuming there is no
heat exchange between the air and the rock surrounding the shaft). The following points should
be noted:
• Autocompression is not strictly speaking a heat source (it results from a conversion of
energy, rather than from the addition of an external heat source).
• Autocompression causes the air temperature to increase, therefore as the mining depth
increases, the ventilation air has less ability to remove heat.
• The temperature rise due to autocompression is independent of the airflow rate. In
contrast, as the airflow increases, the temperature rise due to other sources of heat
decreases.
It is also important to note that water temperature will increase with depth. If the water is
contained in pipes then this increase in temperature is in the order of 0.20C per 1000m. If
the water is free flowing then this temperature increases to 2.340C per 1000m…
I think this excerpt provides some answers.
Hi TB,
further details about adiabatic compression can be found by looking up these two terms in Wikipedia:
– Bergwinds
– Foehn winds
Both describe the same phenomena, but because of the different geographic locations the hot winds on the lee side of a tall mountain or plateau the naming changes. You get the same thing in Spanish. On the coast of South America, on the border between Chile and Peru, the pacific ocean changes sex, from El Mar to La Mar!
Hi JGM: Thanks for transcribing that, very interesting.
There is actually quite a lot on the net too. For example
http://www.ilo.org/safework_bookshelf/english?content&nd=857170928
The mine heat load includes the effects of auto-compression of the air in the intake airways (the conversion of potential energy to enthalpy as the air flows down into the mine)
Thanks again for drawing attention to this, it seems that the gravity effect is well understood by mining engineers, if not by physicists. 😉
The key point here is that this is an adiabatic process, not the heat of compression produced by actively pressurizing a gas.
Good, that’s settled then, without having to invoke degrees of freedom and all that jazz!
All that jazz comes into play when Graeff tries to explain why he is getting a bigger gradient than that predicted by the mining engineers.
tallbloke says: “Thanks again for drawing attention to this, it seems that the gravity effect is well understood by mining engineers, if not by physicists. ”
That “gravity effect” is indeed well-known in physics. In fact, that effect is really the “compression effect”, and not related to gravity per se. If you don’t recognize the enormous conceptual difference between the two cases, then you need to review the science and the situation.
In one case (Graeff’s experiments), the gas is sitting at rest and is allowed to come to thermal equilibrium with the gas at the other end of the container, a process that could take hours (or months, depending on the size of the gas containers). In the other case (mine ventilation systems, airplane cabins pressurization systems, updrafts in the atmosphere), gas is “quickly” compressed (or expanded), and NOT allowed to come to thermal equilibrium.
So yes, when air is pumped down a mine ventilation system, the air WILL get warmer as it goes down the duct — because it is actively moving and getting compressed, NOT specifically because it is going down. I could get the same heating effect pumping gas into a container with no change in elevation. But if the low pressure cool gas and the high pressure warm gas are brought into contact, they will equilibrate to the same temperature. So if I take a warm, high pressure gas cylinder and set it in a room, the gas inside and out will settle at the same final temperature as the room air. Similarly, the warm gas at the bottom of a tube and the cool gas at the top of the tube should exchange energy until they are the same temperature.
Hi TimF: “Similarly, the warm gas at the bottom of a tube and the cool gas at the top of the tube should exchange energy until they are the same temperature.”
Well that’s the whole point isn’t it? Loschmidt’s theory and Graeff’s experimental evidence says otherwise.
“when air is pumped down a mine ventilation system, the air WILL get warmer as it goes down the duct — because it is actively moving and getting compressed”
I suggest you look at the equations in Jolly Green Man’s quote again.
“mine ventilation systems, airplane cabins pressurization systems”
Bzzzt. Conflation of different scenarios.
“I could get the same heating effect pumping gas into a container with no change in elevation.”
Prove it by experiment.
>>I could get the same heating effect pumping gas into a container with no change in elevation.”
>Prove it by experiment.
Really? I thought “gases get warm when compressed and they cool when they expand” was so obvious that it didn’t need further comment. This is proven with every refrigerator in the world!
>>Similarly, the warm gas at the bottom of a tube and the cool gas at the top of the
>>tube should exchange energy until they are the same temperature.”
>Well that’s the whole point isn’t it? Loschmidt’s theory and Graeff’s experimental
>evidence says otherwise.
Yes, that IS the whole point. Which is why mine ventilation and pressurized cabins and oceans and Foehn winds are irrelevant to his work. Standard physics agrees that all of these SHOULD have temperature differences, but that Graeff’s experiments should not.
Mines and pressurized airplanes and Foehn winds are all related to “potential temperature” mentioned earlier in the thread.
So when low pressure air around an airplane is compressed quickly to 1 atm inside the cabin, the temperature of that air goes up, but the potential temperature stays the same. When air pumped down into a mine (in an insulated duct) increases in pressure from 1 atm to 1.1 atm, the temperature increases the same, but the potential temperature increases. (The same would be true pumping air down to a submerged submarine, even though the water temperature is actually decreasing as you go down). When you let air out of a tire, the temperature drops, but the potential temperature stays the same.
###################################################
The Graeff experiment is basically trying to determine whether actual temperature should be constant within a column of gas or if potential temperature should stay constant. Classical thermodynamics puts its money on isothermal being the condition for thermodynamic equilibrium.
QUESTIONS FOR DR GRAEFF – my latest
Note: Graeff’s involvement is (a) in the lab (b) from an engineering background (c) to challenge the Second Law as currently formulated (d) with the future possibility of an alternative heat/energy source. I suspect that all climate science issues may be up to us to answer for ourselves…
1. Water: why is the measured temperature gradient (0.05K/m) higher than the calculated one (0.04K/m)? Application of the (modified) Second Law would posit that the measured gradient should be lower, not higher, than calculated.
2. Oceans: why are the ocean floors cold not warm?
[ROY MARTIN: my understanding of the distribution of ocean temperatures both vertically and laterally, and very slow rates of circulation in the deep ocean, does lead me to the point of being dismissive of the results of this Graeff experiment. The temperatures that would arise as a consequence of his results are so far removed from what is observed that the two appear to me to be quite irreconcilable.
ANNE: I suspect that we have two opposite forces at work in the whole enterprise, Gravity which causes the bottom layer to heat up, and Convection whereby the hot lower layer tries to expand and rise up. It would then appear that in the case of the oceans, Convection wins.]
3. Solids: have nonconducting solids been tested? Does conductivity obscure the Graeff effect?
4. Earth: does the Graeff effect explain (a) a hot earth core (b) vulcanism, when convective forces occasionally overcome the restraints of solid mass (c) surface much cooler than hot depths should allow, if the heat was not just gravity/pressure induced? (and that’s allowing for poor conduction). And what about the varying effects within mines, whereby T(gradient) up to 34K/km goes with increasing density, but decreasing density over 34K/km?
5. In 2005, Prof Sheehan doubts whether Graeff’s temperature measurements are as accurate as stated. Is there a formal reply to Sheehan?
6. The “degrees of freedom” issue affects theoretical temperature gradients but is a strange new concept to grasp, involving the “equipartition theorem” and the difference between falling molecules hitting each other and Newton’s apple falling. Is any more help available? Where can one check that water has 18 d.o.f. – and maybe work out others?
7. [Tallbloke] What replies have there been to Graeff’s calls for replication? Has he approached specific accredited labs to try to engage them? If so, how many, and what were their responses?
8. [Tallbloke] Can Graeff help specify replication efforts by giving specs and sources for the best thermocouples and datalogging equipment?
9. Air: Graeff tested air “without inhibiting convection” and obtained a temp gradient of -0.02K/m. How does one explain the difference between his results and atmospheric lapse rates which are of the order of 0.007K/m (5K/km wet-10K/km dry)?
[ANNE: My reading of the temperature profile is now as follows: the gravity effect prevails over convection up to the tropopause, to yield the measured adiabatic lapse rate. But not above the tropopause; instead we perhaps, perhaps! have the one true region where the greenhouse gas effect prevails. Note: tropopause on all planets is at ~0.1 bar pressure.]
10. TIM FOLKERTS: The alternating layers of insulation & thermal conductors is indeed close to ideal. But there is more that could and should be done (or at least reported).
* Type E thermocouples are rated to +/- 1.7 C (about +/- 0 0.1 mV). Connecting them backwards in series will mostly cancel the signals, leaving the difference. But since the two thermocouples could produce thermoelectric voltages that are ~ 0.1 mV different even at the same temperature (because they are not manufactured identically) , they could still easily read ~ 1.7 C difference even with no temperature difference. From what I can tell, Graeff’s efforts to reverse the polarity are aimed at the voltmeter offset, not the thermocouples themselves. There should be a calibration measurement made with both thermometers at the same temperature (say in an ice-water bath or in a block of aluminum).
* It only takes ~ Q = k A Delta(T) / l to set up a temperature gradient. For the water in his experiment, this amounts to ~ (0.6 J/m*K) (0.0012 m^2) (0.03 K) / (0.85 m) = 25 microwatts. Yes, 25 uW at one end of the water column will match or overpower the effect he is trying to measure..
There would be heat transferred through the wires leading to the various thermometers.
TF’s QUESTION: How is the accuracy of the thermocouples’ temperature gradient detection tested?
11. TIM FOLKERTS: To set up a gradient of 7 K/km in air requires a heat flow of Q = k A (delta T) / l, which would be about 0.2 mW per square meter. Even if such a gradient were the equilibrium adiabatic condition (which I disagree with), then heating of the order of 0.0002 W/m^2 would be enough to mask this effect. Given that typical heat flows in the atmosphere are ~ 10,000 times larger, the N&K temperature gradient would never be observed other than in very careful laboratory settings.
TF’s QUESTION: How do we compare Graeff’s laboratory settings with climate science material? Important to consider, as the “Graeff effect” overturns 150 years of climate science basics, if gravity is the prime reason for a warm Earth surface and a hot Earth interior.
12. “BR 1” has met Graeff, and reports that he has achieved a temperature gradient of 0.3K/m. Is this true?
Tim Folkerts
I’ve found I have to think molecules, to see how all the different conditions fit together quite naturally. You have to see “pressure” and “density” and “temperature” and “convection” and “conduction” as molecular response to local conditions. You have to think statistically as Maxwell realized. And you have to grasp the difference between air molecules falling under gravity, and Newton’s apple falling – linked by their degrees of freedom. And as Tallbloke hints clearly, you must take note of Graeff’s experimental results, and the rigor resulting from a lifetime of designing and manufacturing machines. You can’t sell engineered equipment if it doesn’t work. Graeff is self-funded in retirement, like so many of us climate skeptics, because he’s passionate about what he’s found. If you are really interested to explore, I can send you a copy of the book on condition you mail it back to me after, if it doesn’t speak to you. I don’t really want to be writing here what the book says better.
Lucy
Thanks for the offer for the book, but I will pass. If I really want one I can buy one.
If you are not familiar with the equipartition theorem, I would suggest that as a good next step in your quest for understanding of statistical mechanics. It helps explain, among other things, why the large apple is different from the small air molecules.
Tim
TJ, with all Graeff’s material I am now familiar with the equipartition theorem. Perfectly sensible. It helps explain, as you say.
Last post here I expect. Next step is a new article when I come back.
Reblogged this on acckkii.
[On WUWT] tallbloke says:
May 26, 2012 at 3:32 am
Lucy Skywalker says:
May 26, 2012 at 1:01 am
… but in the end, was (willingly) seduced instead to attend the 2012 seminar of Dr Roderich Graeff on his 14 “retirement” years of experimental and theoretical work challenging the currently-accepted formulation of the Second Law of Thermodynamics.
In my view the law is correctly formulated, but incorrectly interpreted by some when specifying experiments which can test it. For a column of gas or liquid to be free from from the influence of external forces, it would have to be removed from any gravitational field present around and through it.
Will write it up asap and I hope Willis you can have a look when it appears as I’d appreciate your reactions. It will however have to appear, initially at least, on Tallbloke’s blog, unless Anthony can accept it directly… but I really hope Anthony will pick it up and republish. The work is stunning and replicable “at home” and important for Climate Science
Seconded, the issue deserves wider exposure and discussion than can be obtained through the Talkshop alone.
[…] post from ‘Lucy Skywalker’ who has recently returned from a trip to Germany where she attended a seminar given by Roderich Graeff, the engineering concern owner who has been experimenting with equipment […]
I realise that this is something of a late addition to this set of comments but, being somewhat frustrated by the rather dubious analysis of a gravity induced atmospheric temperature gradient, and not finding an adequate analysis of the problem via an internet trawl, I decided that the best way of resolving this matter (at least to my own satisfaction) was to examine the statistical mechanics of the problem from first principles. Having achieved this, the result is a small piece of analysis that may be worthy of sharing. The relevant pdf file can be found here:
Click to access statmechi.pdf
In reality, this analysis assumes very little about the (isolated) system under study – namely, that the number of gas molecules is constant in the system; and that energy is conserved. Everything else follows from this (and an assumption of ergodic mixing in phase space, of course). In particular, it does not assume from the outset that system is immersed in a heat bath at uniform temperature.
The conclusion is, however, that in an isolated system a gravitational field does not induce a temperature gradient in a gas.
I know that Graeff reports having observed a temperature gradient. However, I am not familiar his experiments. All I can say is that unless wish to overturn the underpinnings of statistical mechanics, it is difficult to reconcile those reports with theory.
Thanks for your efforts Keith. One of the problems I’ve been trying to unravel from the start of all this is the question of why stat mech doesn’t find a gradient, where classical mech does.
It seems to revolve around the average total energy owned by molecules at different altitudes, and an assumption that they start their journey at the surface.
Classical mechanics is embedded in statistical mechanics, of course. And, so, the two must ultimately agree. Although the expression for the energy of a gas molecule used in statistical mechanics ignores a small ‘interaction’ term between gas molecules, in reality gas molecules must occasionally collide if thermal equilibrium is to be reached. For the purpose of calculating the statistical mechanics, this interaction term is normally ignored in the maths (largely because it is very small, or that the particles are, for the most part very far apart) – but its existence is implicit in the ergodic assumption underpinning the statistical description.
If one focuses on a single gas molecule then, yes, in the absence of collisions, the molecule will have a higher kinetic energy at the bottom of the gas column than at the top of its trajectory. Via the (weak) interaction term, however, the kinetic energy of the gas molecules is eventually distributed such that the average kinetic energy of a gas molecule is the same throughout the gas column – even though all gas molecules follow the paths dictated by classical mechanics.
One way of exploring this notion is to conduct a numerical experiment in which one numerically integrates the classical equations of motion of all of the gas molecules that interact via a weak interaction term – i.e. they occasionally collide. Although one could not examine a very large ensemble, an ensemble consisting of, say, a few hundred gas molecules should be feasible. One should be able to demonstrate that kinetic energy is eventually distributed among the gas molecules in just the way described by statistical mechanics.
Since this is an issue that appears to be the source of some not inconsiderable controversy, I may spend a little time conducting just such a numerical test which, although replicating the paths implicit in the classical equations of motion of each gas molecule, should replicate the conclusion of statistical mechanics once an equilibrium state has been reached.
Hi Keith G,
If you read the Lucy Skywalker threads on Graeff, you will see that I also have found that gravity does not produce a temperature gradient for an isolated gas. This is after writing a 1D kinetic gas model, a 2D kinetic gas model, and approaching the question from different theoretical viewpoints.
However, I am still very interested in Graeff’s results – I have not found a decisive flaw in the experiment! It would be great if Lucy could replicate the result.
Keith G:
“One way of exploring this notion is to conduct a numerical experiment in which one numerically integrates the classical equations of motion of all of the gas molecules that interact via a weak interaction term – i.e. they occasionally collide. Although one could not examine a very large ensemble, an ensemble consisting of, say, a few hundred gas molecules should be feasible.”
Check out this, and surrounding comments – is this the type of thing you mean?
Since then I have totally rewritten the algorithm to make it an event-based simulation, and by running it in C++ have managed to run 10,000 molecules. I haven’t made that publicly available yet. However, the principles can be seen with a few hundred molecules.
Have you seen this paper?
Click to access s-velasco.pdf
Hi br1
Had a quick glance and, yes, I’m pretty sure it is. I’m in Hong Kong and its getting late, so I’ll have a closer look tomorrow.
Maybe it is time to say in public what I and more than one other spotted rather early to do with Graeff’s self report.
It is only necessary to reverse the thermocouple connections, to turn an external temperature gradient impressed on an internal object upside down according to measurement.
Graeff is very insistent on using thermocouples. No sanity check has been shown.
This is a very common connection mistake even when terminals and thermocouple cables are coded but coding is often poor or non-existent. Moreover Graeff reports manufacturing his own thermocouple wiring.
Contrast this with the myriad of resistance change sensors where there is no ambiguity. (thermistors however have both positive and negative kinds)
At no point does Graeff mention an externally controlled environment nor arranging for the external thermal gradient to invert.
tchannon:
“It is only necessary to reverse the thermocouple connections, to turn an external temperature gradient impressed on an internal object upside down according to measurement.”
That he simply got the connections the wrong way round is something a replication would sort out, but for a start it doesn’t seem compatible with his measurements shown here:
There is an aluminium shell inside the insulation in order to equalise temperature – the measured gradient on this is close to zero. The gradient within that shell is higher. Surely this is impossible no matter which way around the thermocouples are?
Furthermore, he has measured the gradient using both thermistors and thermocouples in the central flask on very many different occasions and in different setups. It seems hard to think they were all connected the wrong way around (unless there is experimenter bias subconsciously choosing the ‘correct’ result). Any other suggestion goes towards direct fraud. However a discussion around that would be pure speculation, so let’s hope a replication is on the way.
However, I like your proposal to invert the ambient gradient. Nice in theory, though not easy in practice.
He has of course inverted the entire setup and found that the gradient re-established itself, but if the thermocouple connections were wrong then the gradient would still be that imposed by the environment and the result would be reduced to being trivial. Which is of course your point.
br1:
I’ve now had a closer look at the comments in the referenced post. It looks like you have successfully carried out the numerical integration experiment and verified the statistical mechanical prediction of isothermal temperature. No need, I would think, for me to spend time on replicating this.
Just to emphasise what you have done: by a direct integration of the classical equations of motion of an ensemble of gas molecules, each of which interacts weakly with other proximate gas molecules via some form of interaction potential – i.e. they occasionally collide – you have produced an equilibrium state with statistical properties the same as those predicted by statistical mechanics. In other words, classical mechanics and statistical mechanical are in agreement, as expected.
In a way, it really couldn’t be otherwise since statistical mechanics can be viewed as just a very efficient way of solving the classical equations of motion for a large ensemble of particles.
As for the Velasco paper: Velasco et al examine the case where the number of particles in the ensemble is ‘small’. For any practical experiment on a sample of a real gas, the number of particles will be astronomically large so one would not expect any of their conclusions for finite samples to be relevant to any real experiment conducted at ordinary macro scales. As the authors point out, in the large N limit, the micro-canonical conclusions are the same as the canonical result.
Or to put it another way: although the system is not immersed in a heat bath – it remains isolated at all times – at equilibrium each gas molecule finds itself immersed in a large heat bath consisting of all of the other molecules in the system. Since each molecule finds itself in the same position, the average energy of each gas molecule is forced to be the same at equilibrium.
As the number of gas molecules in the system is reduced, the finite size of this ‘effective heat bath’ surrounding each gas molecule becomes more significant. In particular, for if the number of particles is in some sense ‘small’, any interaction between a gas molecule and its surrounding ‘effective heat bath’ does not leave the statistical properties of the ‘effective heat bath’ unchanged. For numerical experiments with small numbers of particles, this may become an issue – but I would have thought that a few tens of thousand particles would have been large enough sample for this finite size effect to be immaterial to the outcome.
Hi Keith G,
That sounds like a good summary – I agree with everything you wrote.
Keith G,
“For numerical experiments with small numbers of particles, this may become an issue – but I would have thought that a few tens of thousand particles would have been large enough sample for this finite size effect to be immaterial to the outcome.”
Exactly right, though even half a thousand particles gives an almost isothermal profile. I plotted up some relevant figures here:
br1,
Very nice! The comparison of temperature profiles vs Velasco’s finite N theory was rather impressive.
And I rather liked the spontaneous generation of convection in your ‘heated’ gas columns – a neat example of spontaneous symmetry breaking.
Still, this doesn’t explain Graeff’s results. Perhaps his experiments are revealing ‘new physics’. But on the basis of theory and numerical experiment, I think it more likely that his experiments are not showing quite what he thinks they do. Could be wrong, of course – observation always trumps theory – but without harder evidence I think it more plausible that theory is right in this case.
This report — “Gravity increased by lunar surface temperature”, http://dx.doi.org/10.7392/Physics.70081909 — suggests the reverse may be true — namely that increased surface temperature creates conditions required for object motion previously attributed to a “gravitational force”.
Distance to a reflector on the surface of a spherical rotating orbital body?
Also, the temperature variation of the lunar surface is only in a very thin layer because the surface is fine dust in a high vacuum which is an excellent thermal insulator. The mass (kg) of this layer is minuscule.
To make life even more fun, so what about the phase lead/lag?
As air molecules move between collisions, they follow elliptical arcs due to the influence of gravity. As they do so, some of their kinetic energy becomes potential energy, or vice versa. Their kinetic energy is related to their temperature. (Their temperature is actually their energy per degree of freedom.) A loss of kinetic energy is a loss of temperature, i.e. a gain in potential energy is associated with a loss of temperature.
The potential energy of an air molecule is mgh where m is its mass, g is gravity and h is height.
The potential energy gradient is therefore mgh/h = mg.
This can be easily calculated. m for an air molecule is about 29 atomic mass units, is somewhere between 28 for a nitrogen molecule and 32 for an oxygen, but weighted towards nitrogen due to its abundance. g is gravity: 9.81 m/s/s.
Knowing the energy gradient then allows us to calculate the temperature gradient. It requires a change of energy units from (for example) Joules to kelvin. If you perform the calculation, you will get about 10 degrees per kilometer for dry air and 6 degree per kilometer is damp air. The reason for the difference is the division by the number of degrees of freedom. ~5 for dry air (three translational and two rotational), more for damp air (due to the vibrational modes, H-O-H bonds flexing).
If you perform the calculations, it is worth checking your conversion constant by calculating an air molecules kinetic energy 1/2 m v2 in Kelvin. With a mass of 29 AMU and a velocity of the speed of sound, you should get an energy of room temperature.
protheric: Welcome, and thanks for that interesting comment. Try searching for ‘Loschmidt’ using the search box on the left.