[SPLIT]Older LENR Experiments were bad, good... in general

    Quote from THHuxley

    Well, I may be using the word loosely but then I don't think I've mentioned "logical fallacy" here.


    The expression "appeal to authority" describes a logical fallacy. It has been defined since ancient times. It means something specific. Your are using the term incorrectly. What you describe is not a fallacy at all. Citing authoritative sources and experts is a logical, approved argument that supports an assertion. You have the notion that there is something incorrect or illogical about such citations. You are wrong.


    Quote from THHuxley

    The problem with appeal to authority on an issue like this is that the authorities tend to be biassed.


    No more than non-authorities. Bias is a problem, but there is no reason to assume it is more common among experts than non-experts. It is a separate issue. So, it has nothing to do with an appeal to authority, and when such appeal is valid (and not fallacious) the likelihood of bias does not make it less valid.


    There is no reason to think that cold fusion researchers are more biased than people who oppose cold fusion. On the contrary, most of the people who oppose it have no knowledge of the field and have never bothered to read anything, therefore they are biased by definition. (People who support cold fusion yet who know nothing about it also biased, in the other direction.) We know that opponents have read nothing because their papers filled with factual errors and gross misunderstandings. For example, they think that recombination might be a problem, and some of them do not realize that Miles measured helium at levels 1000 times below atmospheric concentration, which rules out a leak as the source of the helium. People who say such things have no idea what they are talking about. They confuse the issue, mar the conversation with nonsense, and waste everyone's time.

    lenr-canr.org/acrobat/MarwanJanewlookat.pdf


    Quote

    Nearly fifteen years later, Shanahan followed this up with a commentary comprised of unsubstantiated blanket statements critical of the field. He reasons by syllogism from particular examples (often misunderstood) to general conclusions that clearly cannot apply in all of the examples of anomalous heat production observed in a wide variety of experimental configurations involving different kinds of calorimeters, e.g. isoperibolic, Seebeck, and mass flow. To explain the excess heat in these experiments, Shanahan invokes what he calls a Calibration Constant Shift (CCS). This CCS is nothing more than a hypothesis and should be stated as such (CCSH). There is no experimental evidence that it occurs, especially at the level of ±780 mW stated by Shanahan. Furthermore, Shanahan does not specify mechanisms by which a calorimeter thermal calibration can change in such a way that, just during the periods of putative excess thermal power production, the calibration constant is different from its initial and final calibrated value. He employs the calibration constant shift hypothesis (CCSH), unquantified, with the logic that if this can happen in one experiment or calorimeter type, then it must be presumed to happen in all. To dispel this notion, the excess heat results obtained using two completely different types of calorimeters will be discussed.


    I've highlighted the two key arguments from Marwan. The first is specious, and wrong:
    (1) Shanahan's hypothesis economically explains some results. That does not prove it, but it is some evidence.
    (2) It is surely up to those who claim highly interesting new results to show that they are not artifacts, and respond to all criticism by defending their claims - if necessary with additional experiments.


    The second is more interesting. And subtle. CCSH would apply here to some classes of calorimetric experiments, not necessarily to all. But CCSH could in principle apply to any calorimetric experiment where results depend on calibration and there is a plausible argument that calibration and experimental conditions are different in a way that could affect results.


    Put that way CCSH sounds just like "hey - look - you have calibrate properly!" Obvious. Which is why it is subtle. He is perhaps saying that for some classes of experiments there is a systematic error mechanism that has not been considered by the LENR paper authors are subsequent analysis.


    I believe that Marwan is that consideration which might refute Shanahan and show that CCSH could not account for the results of F&P style papers.


    I think there is some possible confusion here. Is Marwan claiming that no F&P style experiment could suffer from CCSH style errors? Or that some such experiments do not?


    Sorry, i'll go on reading:


    Quote

    The excess power measurements done at China Lake used an isoperibolic-type calorimeter. Periodic calibrations over a five-year period showed no significant changes in the heat transfer coefficients for the China Lake calorimeters.5 In addition, the isoperibolic calorimeters used by Miles at the New Hydrogen Energy Laboratory (NHE) in Japan incorporated an automated Joule heat pulse. The calorimeter was calibrated at least once every second day. From this, the coefficients of thermal calibration are deduced by backwards integration fitting of the calorimeter response to this known input thermal power pulse. Calibrations were performed before, after and during the production of excess thermal power


    Marwan is arguing that two sets of experiments are protected from calibration errors in the way that he suggests here. Perhaps before going further it would be helpful to get Kirk's comments on these two sets of experiments. Does he believe CCSH might apply here, or not? This does not refute Shanahan's contention that some results can be explained by CCSH.


    It is important neither to over-estimate nor to under-estimate the scope of critiques: and I'm personally unclear here. Give me a bit longer and I'll have worked out exactly what the CCSH argument is.

    Cold fusion researchers are not more biased than their opponents? ROTFWL! Well, thing is, opponents don't have to prove that cold fusion doesn't exist and of course, such a proof of a negative is impossible-- opponents only have to prove that some specific experiments under consideration are error prone.


    Jed:

    Quote

    I guarantee you that every mistake you come up with just thinking about this work was addressed by the authors, and prevented.


    Well, since you thought Rossi's claimed high power output was "irrefutable" on "first principles," perhaps it is justified to give your guarantee a jaundiced view.

    Quote from THHuxley

    It is surely up to those who claim highly interesting new results to show that they are not artifacts, and respond to all criticism by defending their claims - if necessary with additional experiments.


    When the "criticism" describes a phenomenon which is physically impossible, it is not a plausible artifact. It is nonsense. There can be no response to that. The CCSH magically pops up whenever there is excess heat, and then goes away when the heat stops. No plausible or even comprehensible reason is given for it. It just happens, just at the right moment to explain away the excess heat. Then it vanishes without a trace! It happens in many different types of calorimeters which are physically dissimilar and based on different physical principles. It is magically correlated with deuterium and high loading, and high current density, for no conceivable reason.


    These are not interesting new results by the way. People have been doing calorimetry in electrochemistry for 100 years or more. If the CCSH existed they would have seen it long ago. The only thing new about these experiments is the presence of deuterium instead of hydrogen, and very high loading.


    This is the very definition of nonsense in physics. It is a classic example of hand-waving and making up ad hoc bullshit that explains nothing, like Rossi's magical endothermic machine that swallows up 1 MW of heat continuously for years. There is not enough time before the heat death of the universe to answer all the bullshit that people like Shanahan and Rossi come up with.

    Quote from Mary Yugo

    Cold fusion researchers are not more biased than their opponents? ROTFWL!


    Since you have not read the cold fusion literature, and you know nothing about it, you are not in a position to dispute this assertion.


    You admit that you know nothing about the subject -- heck, you brag about it! And yet you say this sort of thing. You are biased. Speaking from willful, proud ignorance is the very definition of bias.

    Quote

    It would be nearly impossible to obtain these conclusions if the excess power was due to Shanahan’s random CCSH. Furthermore, SRI obtained very similar conclusions using a totally different type of calorimeter over this same time period.4,5 The SRI calorimeter was based upon mass flow in which the thermal efficiency reflects the fraction of the total heat removed by convective flow, i.e., Φ = ܳ஼௢௡௩௘ ௖௧௜௢௡ [ܳ஼௢௡ௗ௨௖௧௜ ௢௡ + ܳோ௔ௗ௜௔௧௜ ௢௡ ⁄ ] (1) A Mass Flow Calorimeter designed with high thermal efficiency, Φ, can operate as a first principles device with no calorimeter specific calibrations. Nevertheless, the calorimeter was 4 periodically calibrated using an internal resistor. The maximum error was determined to be ±50 mW. For a mass flow calorimeter with Φ = 99%, only 1% of the measured heat output is subject to the vagaries of geometric effects on conduction and radiation. The remaining 99% is determined solely from temperature, mass flow rate and the heat capacity of the convecting fluid. None of these measurements are subject to calibration drift and can be measured and calibrated independent of the calorimeter. Thus the CCSH can account for an excess power of at most (and actually much less than) 1% of the output power in the example given. Reported excess power numbers are typically >10% of the input electrical power. The CCSH can thus be shown quantitatively to fail in all cases of excess power reported in mass flow calorimeters. The SRI results typically yielded 5 to 10% excess power with a maximum of 28% excess power; the excess power was 1–5 W/cm3 on the average; the initiation time was on the order of 300 h for 1–4 mm Pd rods; the threshold current density ranged from 100–400 mA/cm2 ; and the success rate varied greatly with the source of the palladium.6,7


    OK. So there is here the artful use of "random CCSH" without, I'm afraid Jed, any justification. But that is polemic, not the arguments.


    There are three arguments: one is that CCSH could not apply given all the conditions in this experiment. That is not made explicit and I don't understand it, but it could maybe be argued.


    The second is that two different types of calorimetry show positive results an CCSH could not apply to both. i don't see that. the eror could surely be systematic and rely on features in common, or even systematic and both types gives positive errors coincidentally. Not a big coincidence.


    The third is that the calibration errors here are bounded and the results are some 10X larger than this. But this third argument uses "SRI" results. is this a third example, different from China Lake and Japan as above? Or not?


    It matters because it is easy to argue against something by cherry picking different examples, and showing that one bit of it fails for each. I think what we have here is a possibly correct claim that CCSH does not apply to the SRI results. I can't see, if that is correct, how it could be extended to other experiments?

    Quote from Jed

    The CCSH magically pops up whenever there is excess heat, and then goes away when the heat stops. No plausible or even comprehensible reason is given for it. It just happens, just at the right moment to explain away the excess heat. Then it vanishes without a trace! It happens in many different types of calorimeters which are physically dissimilar and based on different physical principles. It is magically correlated with deuterium and high loading, and high current density, for no conceivable reason.


    That is a very good point. But I'm not sure it is true. Part of the magic might be experiment selection - which i'm afraid we would need some time to consider. It is quite subtle. The other part of the magic might be if there is a reason here - which there might be. Kirk may even have ideas about this - but if he does not I don't think you can reject this on those grounds just because the reason has not been discovered. You would have to show there on general principles there could be no such reason.


    Remember, CCSH does not have to effect every single experiment. If it could affect some class of them, but not others, we could look at the others to see how convincing they are confident that one source of artifacts has been eliminated.


    Marwan's argument here is remarkably simplistic.

    Quote

    like Rossi's magical endothermic machine that swallows up 1 MW of heat continuously for years.


    Rossi's endothermic machine is physically impossible, short of melting ice cubes etc which is physically unfeasible. And whatever, such a process that is also industrially useful as Rossi claims is highly unlikely. This has all the hallmarks of post facto rationalisation.


    Shanahan's argument is a decent critique which should be addressed in detail. Could there be some mechanism that gave rise to this behaviour? Can we rule it out of some classes of LENR experiments? Which classes should we view as giving contaminated data because this mechanism is possible? These are proper questions with proper answers, not magic.

    Quote from THHuxley

    Perhaps before going further it would be helpful to get Kirk's comments on these two sets of experiments.



    Summarizing:


    1) the "CCSH" is NOT my proposition. It is the strawman argument put forth by the people who didn't like my proposal. As such it is wrong, all agree on that. So let's drop the "CSSH". I'm not defending *it*.
    2) All F&P-type electrolysis experiments, to my knowledge, have two critical experimental factors in common that allow the calibration to shift due to the onset of at-the-electrode recombination. They are a) the electrolysis gases are not kept separate from each other and b) all the penetrations to the cell are co-located in the "top" of the cell. 'b' concentrates heat loss pathways in one area which creates the need to model the cell/calorimeter differently than with a 'normal' calorimetric setup (which assumes all is homogeneous). 'a' allows the recombination to occur 'bubble-to-bubble' so to speak, and to exceed the electrochemically driven recombination caused by dissolved oxygen reaching the opposing electrode (that process is limited to about 2% or less). There is a crucial paper by Oriani where he placed plastic covers (cups, loose fit, not tight) over each electrode to keep the gases separate (separate vent lines in an open cell) - he got no excess heat. He then removed the covers and saw excess heat. Of note is that when he removed the covers he removed one vent line. As I recall, this had a predictable impact on calibration constants, which I had to estimate from the reported data, he didn't give them straight up.
    3) Different calorimeters make no difference as long as "2" remains true. The penetrations that give the heat loss pathways I mentioned also penetrate the calorimeter boundary. The more efficient a calorimeter is however, the smaller the effect should be, given no other errors present. Viewed another way, a mass flow calorimeter has a calibration equation (most usually) given by y = mx + b. So does a Seebeck calorimeter, with b typically set at 0. So does an isoperibolic calorimeter. Any of them can show an apparent signal (theoretically, + or -) if the cal constants shift. If their design remains typical, the heat losses can change if the heat distribution changes in the cell, and that drives the creation of an apparent excess heat signal. If you switch to a different equation form, you still have the same problem. (And claiming your calorimeter is so good it doesn't require calibration is just setting the constants to 1 arbitrarily.)


    Jed's comments reflect his chosen lack of knowledge about what I wrote and as such should always be assumed to be incorrect from the start. He simply refuses to believe. Storms is the same way, based on extensive email communications on the subject where he proved he understood the implications of my model, but just chose to refuse to believe it.

    Quote from Kirk

    2) All F&P-type electrolysis experiments, to my knowledge, have two critical experimental factors in common that allow the calibration to shift due to the onset of at-the-electrode recombination. They are a) the electrolysis gases are not kept separate from each other and b) all the penetrations to the cell are co-located in the "top" of the cell. 'b' concentrates heat loss pathways in one area which creates the need to model the cell/calorimeter differently than with a 'normal' calorimetric setup (which assumes all is homogeneous). 'a' allows the recombination to occur 'bubble-to-bubble' so to speak, and to exceed the electrochemically driven recombination caused by dissolved oxygen reaching the opposing electrode (that process is limited to about 2% or less). There is a crucial paper by Oriani where he placed plastic covers (cups, loose fit, not tight) over each electrode to keep the gases separate (separate vent lines in an open cell) - he got no excess heat. He then removed the covers and saw excess heat. Of note is that when he removed the covers he removed one vent line. As I recall, this had a predictable impact on calibration constants, which I had to estimate from the reported data, he didn't give them straight up.


    Excellent. So we have an explicit suggestion from Kirk of a plausible artifact that satisfies:

    • Many F&P style experiments have it
    • It is not usual in calorimetric systems
    • It could give rise to false positive results in a way that pre and post calibration cannot detect, because the error is linked to the precise conditions of an active test (at least I think that is right).

    Jed, I'm sure that those arguing these experiments are good evidence will be able to say which of these propositions is false, or else identify the F&P style experiments which do not have these specific conditions?


    It seems on this concrete matter that agreement should not be far away.

    Quote

    I guess if you now think it's a decent critique, you've worked out what it is exactly...Could you THH explain it in simple terms, please?... I find KirkShananana's explanation very opaque.


    I too. But what I extracted above is much better. Let's run with that and I can explain it further if you need that (I think it is pretty clear)?

    So to sum up, from your bullet points, it's something that happens in a calorimeter that cause false positives.


    Is an actual mechanism hypothesized? Something that could be tested for? Does H & O recombination have something to do with it?

    Quote from zeus46

    Is an actual mechanism hypothesized?



    After many, many hours of running with a regular F&P setup, or almost immediately with a "Szpak" co-deposition process, the electrode surface develops a 'special active state' that allows for the chemical recombination of hydrogen and oxygen (from bubbles) on the electrode surface. This reduces the exiting gas quantity in an open cell, but may increase entrained electrolyte microdroplets due to the observation that there are now 'micro-explosions' occurring at the electrode in a fashion similar to depth charges. In a closed cell, this reduces the amount of reaction at the gas space catalyst and places it at the electrode instead. The special active state clearly fosters said recombination, perhaps by causing H2 bubbles to adhere more strongly to the electrode so that they can coalese with impinging O2 bubbles OR by fostering adhesion of impinging O2 bubbles which then migrate to an H2 bubble and coalese. The metal surface then catalyzes H2 + O2 recombination (just as the recombination catalyst does). These microexplosions produce shock waves which can be picked up by a piezoelectric transducer, and can cause physical damage to the electrode or its support at the point of explosion. Because of the difference in heat capture efficiency in the cell, the heat that at one time formed in the lower efficeincy region is now registered in the calorimeter output at a higher efficiency. Because the calibration 'bumps up' the registered signal to equate it to 'power out', the more strongly measured signal is registed with a slight 'magnification' in the output. Thus, not only is more heat attributed to output, it is now incorrectly 'bumped up' using the calibration constants derived from a different steady state configuration. The more efficient the calorimeter over all, the smaller the effect.

    Got it.


    So when the H & O recombine, they release heat, when this normally happens at the top of the cell, some of this heat is able to escape (through wires etc). In theory, if recombaintion happens lower down in the cell, the heat can't escape through the wires - so can be confused with excess heat?


    Edit: What exactly is this "special active state" that allows recombination on the electrode surface?

    That is very clear Kirk. I notice this hypothesis shares many elements with typical LENR hypotheses (not surprisingly since it is trying to explain the same data)!


    It makes an immediate qualitative prediction. which is good. I'm not sure if that is strong enough to make it falsifiable, but maybe. In any case it can be compared with "vanilla LENR".

    Some details I forgot:


    The effect is expected to be very sensitive to isotope effects. The viscosity of heavy water is quite different from that of light water. the chemical bonds made by D are stronger than those made by H. The will impact the adhesiveness of the surface to bubbles. Likewise, using differernt metals and electrolyte makeup will tinker with the relevant parameters needed to be achieved by the special active state. The length of time required to form it suggests it may be quite unstable (non-co-dep processes). Altering the surface roughness would in principle also foster bubble adhesion. The ultimate of that would be dendritic structures or nanocrystalline ones. Thus it is quite reasonable to assume light hydrogen can produce the effect under some conditions (as yet to be defined).

    Quote from JedRothwell

    kirkshanahan wrote:


    I was addressing Mr. Huxley's comments, not yours. He may be under the impression that researchers claim excess heat with only 1% apparent excess heat. I do not know of any examples of that in the literature.


    Some of the noise may be reduced here if you stop assuming that THH is terminally clueless. If he makes some statement that could be interpreted that way, it could be simply incomplete expression.


    Imagine that there is a panel meeting to consider the reality of cold fusion. They will see the evidence we present and will hear our arguments. They will also, as people do, assess the probity and sobriety of the witnesses. There can be a reason to question the expertise or knowledge of another witness, but be careful about it. It can reduce the impact of whatever truth you have to present.


    Quote

    The only thing excruciatingly clear from your papers is that your claims are wrong, and imaginary. If you wish to address the assertions made by Marwan et al. I would be willing to upload your response, but I will not bother to address your comments myself. I have wasted too much time doing this in the past.


    This is a great "put up or shut up." Kirk has written responses, but they can be hard to find. Kirk, Jed will do it if you give him permission, if not, let me know, I'll, ah, remind him.


    At this point, much of Kirk's argument is on the level of "I told them but they won't listen." The journal editors were unfair, etc. It's ironic, eh? Actually, Kirk's last paper would not have passed an ordinary peer review. However, it is probably the best skeptical response received by the Journal of Environmental Monitoring. So they decided to print it. And then the response by the phalanx of cold fusion researchers contained at least one error or "unfair" statement about CCSH. OMG! Call the Fairness Police!


    Yet, Kirk has not actually established what he claims, AFAIK. Where is the careful review of this? It's a bit of a moving target, more than one theory bundled together under the general subject of "They Must Be Doing Something Wrong," which I call the Garwin theory.


    There are different types of calorimetry used in cold fusion experiments, so Shanahan puts together a hodgepodge of "possible artifacts."


    Quote


    This was an unsatisfactory answer. THH comes back with the answer I would expect. Jed assumes a large leak such that the helium would equilibrate at ambient or at least rise far toward it. I think there is a far better answer embedded in a consideration of what THH wrote:


    Quote


    This is an overgeneralization. Miles was not operating under stressful conditions.


    It is not possible to investigate "all low level leakage modes." This kind of problem is addressed with correlation. When I first was considering the leakage possibility, with regard to the correlation with excess heat, there was an obvious possibility. "Excess heat" sounds like "hotter," and maybe a hotter cell leaks more. So ... what about this? What do we know or what can we reasonably infer?


    1. Excess heat is not a temperature effect; the actual temperature of the cell and the excess heat do not necessarily vary together. There may be hotter cells with no excess heat and cooler cells with more. Some experiments run at constant temperature, excess heat is measured by the reduction in power necessary to maintain that constant temperature.


    2. Helium leakage would vary greatly with the exact nature of the seals and cell materials. The heat/helium ratio would be very unlikely to match across many varied experiments. Essentially, the leakage hypothesis is reasonable if a single experiment is being considered, not if many experiments are being considered.


    3. Some experiments measure helium above ambient. An example is Apicella et al (2005). This work should include measurement of ambient helium, of course. Apicella et al was not formally published, but Violante does answer questions.


    4. The Case study (which is not FPHE and may show very different helium trapping behavior) showed helium rising toward ambient and continuing above it with no decline in rate of increase, as would be the case with ambient. The concern there would be helium storage, and in discussions with McKubre about this, he said that their not having analyzed the cell materials afterwards was a shortcoming. Did they analyze them before? I'm not sure. If not, it would be an obvious oversight.


    5. It is extremely unlikely that a leakage source for the helium would cause the results to cluster around 60% of the theoretical energy conversion ratio (ordinary electrolytic studies), or the ratio itself (electrolytic studies plus anodic reversal).


    6. Hydrogen controls do not show helium generation. Why not? Would those cells leak similarly?


    Basically, at this point, Occam's Razor applies. The evidence is quite enough to demonstrate, by preponderance of the evidence, that the FP Heat Effect is real and is nuclear in nature. That's rebuttable, but there is extremely little contrary experimental evidence that is not consistent with this conclusion (as to heat/helium). Those of us who want this nailed down -- and I include myself in that group -- may be happy to know that this work is under way, after a decade of practically no work on heat/helium.


    This work can be done better than before (i.e, it is now understood that anodic reversal probably releases all the helium), and if anyone thinks there is some major artifact and especially a systematic one, this would be the time to suggest how it could be tested! I can get the message to them.


    By the way, how would anodic reversal increase leakage to produce a release of helium that is the other 40%? Because there are only two experiments where anodic reversal was used in connection with helium measurement, that, by itself, is not terribly strong, just something additional to notice.

    Quote from zeus46

    So when the H & O recombine, they release heat, when this normally happens at the top of the cell, some of this heat is able to escape (through wires etc). In theory, if recombaintion happens lower down in the cell, the heat can't escape through the wires - so can be confused with excess heat?


    Basically that's it...(for a closed cell...in an open cell the H2 and O2 usually just go out the vent.)


    Quote from zeus46

    Edit: What exactly is this "special active state" that allows recombination on the electrode surface?



    That's the $89 million question...

    I think the recombination alters things as follows:


    Open cell: (large direct effect) alters exiting gas amount. S claims can also increase the amount of entrained (as water droplets/vapour) H2O lost. I'm not certain which of the F&P style open experiments are broken by this - it depends whether they distinguish between H2O and H2+O2 lost when measuring this.


    Closed cell. Smaller effect dependent on cell efficiency. Recombination alters temperature profile moving heat from the top of cell (where there is more thermal leakage) to the electrode. I'm not quite sure how this works - what is the "space catalyst"? Surely the gas in the cell is much less thermally conductive than the liquid? OK - I'm less clear here.

Subscribe to our newsletter

It's sent once a month, you can unsubscribe at anytime!

View archive of previous newsletters

* indicates required

Your email address will be used to send you email newsletters only. See our Privacy Policy for more information.

Our Partners

Supporting researchers for over 20 years
Want to Advertise or Sponsor LENR Forum?
CLICK HERE to contact us.