Flagged Revisions installed. Unapproved pages display a Red unchecked notice under the title. Trolls attack here by creating and archiving pages with offensive content. To verify an archived page, check the original URL. Questions about administration? Contact User talk:Abd. Limited privacy on this site, see CFC:Limited privacy

Talk:Wikiversity/Cold fusion/Excess Heat

From CFC
Jump to: navigation, search


In his review of Pownall's thesis, Krivit writes:

“According to the principle of the conservation of energy,” he writes, “the amount of heat leaving the cell as measured by the calorimeter should equal the amount of energy applied to the cell by the electric current. This is exactly what Pons and Fleischmann found whilst the cell was ‘charging’ for days or months.”

Pownall makes a very astute point that has been underappreciated in most reviews of this history. Fleischmann and Pons' cells ran for weeks on end while the deuterium was loading into the palladium, all the while being in energy balance.

“Then, suddenly,” Pownall says, “usually because something upset the equilibrium of the cell, the power measured by the calorimeter shot up. The power coming from the cell was 10, 20, up to 50% higher than the power applied by the electric current.”

Notice that the excess power shoots up after the cathode is charged up, and as soon as the applied current has nowhere else to go but to start evolving gas bubbles on the electrode. The bubbles, of course, are what is disturbing the equilibrium of the cell, by introducing fluctuations in the ohmic resistance as a result of the formation of bubbles on the cathode. We know from simple AC circuit analysis that AC noise power begins to be injected, at a rate proportional to the (reported) DC power (which is in turn proportional to the drive current) and proportional to the (unreported) square of the fluctuations in the ohmic resistance. <comment originally added to resource page by Caprice, 17:58, 16 January 2011.

Caprice, you have consistently, so far, ignored a basic experimental fact. Yes, excess power shoots up, in some cases, when cell current is increased. But sometimes excess power shoots up when cell current is decreased, and the power burst you are looking at is the exception, rather than the rule. Many other times, with what appeared to be equally loaded palladium, and the same current excursion, and certainly the bubbling would be about equal (the difference between subcritical loading and excess heat loading is probably pretty small), there is no excess heat. You are reporting as fact what is contrary to experiment, i.e., you state that it is the "bubbles" "disturbing the equilibrium of the cell," then, you fail to notice -- I'm presuming you are looking at the famous plot of a single power burst from McKubre's work -- that the excess power does not persist in this excursion, even though the bubbling would certainly persist. --Abd 18:13, 16 January 2011 (UTC)
  • What I'm consistently looking at is the chart that McKubre consistently exhibits, which is reproduced as Figure 5 on page 12 of Pownall's thesis. —Caprice 18:28, 16 January 2011 (UTC)

Actually, I corrected what is an error above, before your comment caused an edit conflict. So I'll leave it. Figure 5 doesn't show the no-excess-power periods, but conclusively refutes the power supply noise theory as being a major contributor to observed excess heat. I wrote the following before seeing your comment above.

Look at Figure 5, we've discussed this experiment before. The deuterium cell power does track the input power, in this case. That is not necessarily what happens in all cases. Missing from your understanding, Caprice, is that the hydrogen cell is in series with the deuterium cell in this particular work. If there were noise power, it would affect both cells equally. So why would this power not show up in the hydrogen cell as well as in the deuterium cell, if power supply noise were the source of it?
Notice that at higher current, the hydrogen data becomes noisier, and so does the deuterium data. Could that be due to bubble noise? Maybe. Looks to me like there is a little excess power in the hydrogen cell, at that higher current, but very noisy. Notice that 0.5 A/cm^2, there is a very clear excess power signal with deuterium, and hydrogen remains solidly at zero excess power.
And you have completely missed the point of the Pownall paper. It's not about the "reality" of cold fusion, it's about how the scientific process worked. You are, here, over twenty years later, coming up with speculative hypotheses. If they were real, it would have been the duty of the scientific community to identify and demonstrate the artifacts, twenty years ago, instead of massively rejecting the heat data because excess heat is allegedly impossible, which is, in fact, exactly what happened. But they are not real, and you are looking at the data that demonstrates that they aren't real, i.e., are perhaps theoretically possible, but then they are easily ruled out by actual data. Already published.
Workers like McKubre have seen these phenomena over and over in their labs, and they have worked diligently to try to find artifacts; but, early on, they learned that the objections were like Hydra: cut off one head, another appeared, with no end in sight.
Early on, a major -- and extremely valid -- objection was the lack of ash. Because neutrons and tritium were expected from fusion, and if there were helium, gammas were expected, all of which are easily detectable, "no ash" generally meant no neutrons and no tritium and no gamma rays, a stand-in for helium. And this lack of ash was pointed out over and over. But when helium was identified and show to be well-correlated with excess heat at a value consistent with fusion, this was ignored. CF researchers, somewhere around there, abandoned the effort to convince the skeptics, because
  • Enough evidence had been provided, and
  • They now had proof that the rejection was not merely normal, functional skepticism.
  • And so they could waste scarce time and resources, without any gain at all.
At this point the hunt became, not one for proof, but for characterization and optimization. Until the physicists, in larger numbers, become involved, I consider it unlikely that the mechanism will be uncovered. But, as Beaudette points out with radium, nobody rejected Curie based on no mechanism being proposed for the mysterious self-warming material! At that point, however, to follow Pownall, there was no massive research establishment, spending a trillion dollars, pursuing an alternate research path that, if cold fusion is real, may turn out to be the biggest scientific boondoggle of all time. And ... it's real, but we don't know if it will ever be practical. Beaudette also makes that point.
Time to start learning, Barry, you are jumping the gun. Don't jump guns, you end up shooting yourself in the foot. --Abd 18:40, 16 January 2011 (UTC)
  • Why would putting the cells in series inject the same AC noise power into the Hydrogen cell as the Deuterium one? I don't apprehend your model for that. —Caprice 19:13, 16 January 2011 (UTC)
  • It would not be exactly the same power. It would be the same current, at all times, in both cells. The voltage, however, would vary. However, the overall resistance of the cells doesn't vary that much.
  • Suppose a resistance increase in one cell causes a decrease in current, so the power supply voltage ramps up to restore the current. Same current now, higher voltage, therefore higher power. But then it ramps back down because of a resistance decrease. If this happens quickly enough to be missed by the average voltage measurements McKubre uses, then we can understand this as a source of unrecognized power input, and in fact, we can model missed power as being the result of bursts like this, that happen too rapidly to be caught by the averaging that McKubre uses. The cell resistances, however, remain in roughly the same ratio all through this. If the cells have equal resistance, average, roughly half the increased voltage will appear over each cell. Since the current is the same for each cell, at all times, thus the noise power is divided between the two cells, they should both show the same apparent excess heat from power supply noise. (If the resistance change is a major fraction of the overall resistance of both cells, this gets more complicated.)
  • However, I thought you'd ask about his. In fact, the cell resistances will not be equal. Hydrogen and deuterium have somewhat different electrochemical behavior. However, the actual experimental data shows no apparent excess power in the hydrogen cell, during the 0.5 A/cm^2 period, while there is substantial excess power in the deuterium cell. Look at the graph. The behavior is drastically different. Further, the conclusion of no significant noise power is bolstered by the behavior of deuterium cells, where bubbling would be identical, during periods when there is high current but no excess heat. The high current would mean the same bubbling as the cell shown in the graph, yet these cells, when they are "dead," show no excess heat.
  • If we look at the period where current has increased to about 2/3 A/cm~2, we can see some possible increased excess energy in the hydrogen data. I'll estimate this as about 30 milliwatts, but it only reaches that level after more than 10 hours or so, there is no immediate excess heat apparent. Let's assume, however, that the excess heat in the first 10 hours is 10 milliwatts, and that this is entirely noise power, neglecting the possibility of small excess heat from hydrogen. (In the 30 milliwatts, we may be seeing what many have reported, small amounts of apparent excess heat from light water controls, they are not necessarily a "clean baseline," as Fleischmann noted at one point.)
  • From an estimate of two to one distribution of excess power, of 10 milliwatts, we might explain 20 milliwatts of the deuterium results of 500 milliwatts by power supply noise. That's with optimistic assumptions. Barry, my conclusion: it's not worth the effort to research it more closely, but you are welcome to. However, what I've seen is that you only do research to find what's wrong with McKubre's work, not what's wrong with your own. You neglected the distribution of power between the two cells, entirely, it looks like. You are doing research with a desired conclusion in mind, which is exactly how scientists fool themselves. They select data and arguments that support their desired conclusion, and thus neglect the balance, and disagree with others who might have more evidence in mind. Not to mention disagreeing with "true believers." --Abd 19:59, 16 January 2011 (UTC)
  • It's hardly any effort. You already have the theorem for a constant current, I, into a fluctuating resistance, R±r. Now we have the same constant current, I, into a pair of resistors in series: RD±rD and RH±rH. Now we have αD = rD/RD and αH = rH/RH. So PAC,H = αH²PDC,H and PAC,D = αD²PDC,D. So putting the cells in series doesn't really provide much of a control. Deuterium bubbles are twice as dense as Hydrogen bubbles, and have half the buoyancy. So the neither the average ohmic resistance nor the perturbations in the ohmic resistance will be the same in the two cells. The AC power injected into each of the two cells only has the current, I, in common. All the other parameters are idiosyncratic to the differing materials in the respective cells. —Caprice 20:27, 16 January 2011 (UTC)
I believe my analysis, above, is incorrect. Here is why:
  • When the resistance of cell A increases, without that of B changing, the immediate effect is that power to both cells decreases, because current decreases. The instantaneous voltage across the cell series remains the same. The total power drops.
  • Because the ratio of R(A) to R(B) has increased, the fraction of voltage across A has increased and the fraction of voltage across B has decreased.
  • The power supply will then slew to restore the set current. This will restore the power in B to the original power. However, it will increase the power in A. The net effect on B is a reduction of power for a time; on A it will be a period of decreased power, with a period of increased power. The net effect on A then depends on the time behavior of the resistance noise.
  • My error on this has no relationship to measured evolved heat, and that excess evolved heat is found to be zero, or very low, with control cells remains a demonstration that power supply noise, causing misestimation of power supply input, is not a major contributing factor to excess heat.
  • That Barry's formulae show no dependence of noise power error caused by bubble noise, on the frequency of the bubble noise and its relationship to slew rate, still demonstrates to me that the formulae are not correct. This is obvious, because with slew rate very high compared to resistance rate of change, there is no error in current, current remains constant, and voltage changes only directly with the resistance change, and averaging voltage measurements and multiplying them by the set current will give correct input power.
  • In the other direction, with very low slew rate, the supply will not respond to high-frequency noise in the resistance, so, again, there will be no squared noise power, as the total voltage will not change, only the current will change, and average power will be correctly calculated from the product of the measured voltages and currents. (I'm not here considering the effect of the cells being in series.)
  • McKubre assumes a high slew rate with respect to the resistance noise, that's what he thinks of as being "reasonable design." Granted, I can see a possible error here, if the voltage noise is high as a percentage of the DC voltage. I tentatively reject this for two reasons:
  • It would be obviously visible with a single glance at the power supply voltage with an oscilloscope, and it's inconceivable -- i.e., highly improbable -- to me that McKubre would not have looked. Nor is it conceivable to me that others would likewise not have looked. I'll be looking, and I'd have looked regardless of this discussion, and input power is not a critical factor for me! I'll already have the scope set up, for other reasons, and it's just a matter of moving a test lead and tweaking a dial.
  • The estimation of input power by McKubre is confirmed by the periods of zero excess heat, even with high current and thus high bubbling, plus it is more roughly confirmed by helium. (Helium could not have been measured accurately enough in those experiments to give more than perhaps 10% accuracy in estimating excess heat, at best.) --Abd 22:06, 16 January 2011 (UTC)
  • When the resistance of cell A increases, without that of B changing, the immediate effect is that power to both cells decreases, because current decreases. The instantaneous voltage across the cell series remains the same. The total power drops.
Correct. So far, so good. You have modeled the first half of a complete cycle. Now compute the second half of the AC waveform (for both cells). In which cell is there a quadratic term that doesn't cancel, but adds net noise power? In which cell is there no quadratic term and perfect cancellation of the power fluctuations from the two halves of one complete cycle? —Caprice 06:48, 17 January 2011 (UTC)

Error in report of CF cell behavior.[edit]

Barry wrote, responding to Pownall's description of Fleischmann's results, Notice that the excess power shoots up after the cathode is charged up, and as soon as the applied current has nowhere else to go but to start evolving gas bubbles on the electrode.

No. That's a radically incorrect description of the behavior of the cells. The cell has been up to maximum loading, or close, for a long time. There is no sudden transition to bubbling from no bubbling. We are, here, reviewing Excess Heat, primarily, so I'll recommend that Barry look at Figure 1.1 in Beaudette, page 7 in the book.
To explain the graph, the drops in temperature are due to the addition of heavy water to replace evaporated heavy water. Note that this is raw data, not calculated excess heat, so "misting" is irrelevant to this. Over a period of four days, as cell voltage is overall decreasing, with small increases each day when the heavy water is added, there is increasing heat (temperature, which in a Fleschmann cell is generally proportional to power). Voltage is declining as temperature increases. This is a constant-current power supply. At this point, bubbling rates would not be changing significantly.
CF cells are erratic, generally, in how they behave. If excess heat had been correlated well with bubbling, i.e., only appeared with high bubbling, as a result of power supply noise, many researchers would immediately have seen the "excess heat." Just crank up the power! The noise problem would have been resolved long ago. Again, see Figure 4.1 and 4.2, page 47 of the book. After 65 days of electrolysis, output power shoots up for about two days. There would be no change in bubble evolution rate at this point.
At some point, we will need to address "heat after death." For some examples, see Figures 15.1 and 15.2, pages 215 and 217 in the book. With heat after death, there is no input current. The current has been shut off. The most common attempt to explain this has been as being due to the "cigarette lighter effect," i.e, that the stored deuterium is being released in the cell and is burning, effectively. We'll need to look at that. Such an effect, in a relatively enclosed cell like a Fleischmann cell, if it started, would rapidly self-extinguish, as evolving deuterium gas and heavy water vapor from the hypothesized recombination forces any remaining oxygen out of the cell. In some reported cases of heat after death, the total power release after the current was shut off was many times the possible storage of energy through any known mechanism. As Arthur C. Clarke notes, something strange is going on, for sure.
  • I hate to be the one to break the shocking news to you, Abd, but you are not a credible source of reliable scientific theories, mathematical models, technical calculations, data analysis, or testable predictions. —Caprice 06:01, 17 January 2011 (UTC)
  • I'm an "informed student," and "expert" only in a relative sense, as to what is in the literature (but only some of it, the literature on cold fusion is huge). This is a Talk page, where participants in a course may freely comment, within Wikiversity behavioral guidelines, which are normally generous. If I put unreliable opinion on resource pages, other participants are completely welcome to point it out, correct it, etc. But ... neither is Barry a reliable source on this topic, neither as to what he lists (which is original research, generally), nor as to what I'm actually working with here, what is reported in the literature, combined with some personal communication with authors, etc., and some level of analysis (more general than the "data analysis" Barry mentions.) If there is error in the above comments, resulting from my alleged "unreliability," Barry is welcome to point it out, specifically. Simply pointing to my alleged lack of credibility isn't helpful.
  • Barry has not signed up as a participant. He's welcome to. In addition, other Wikiversity scholars are welcome to comment on what's in the course, on the Talk pages. If there are errors on resource pages, inappropriate material, etc., of course all Wikiversitans may participate in fixing them. But the purpose of this course is to provide a path to knowledge of what researchers in the field know, as to a lay understanding. Beaudette will take us up to about 2000-2002, and we will, I assume, create a Cold fusion/Recent history course to bring us up-to-date. The field has shifted, cold fusion is now routinely accepted in journals that are willing to publish anything on the topic. I also just noticed a review at [1], a chapter in Models of the atomic nucleus, subtitled Unification through a lattice of nucleons, (Norman D. Cook, Springer, 2010). Notice the claim, in the book introduction, writing about LENR, CMNS, and CANR, This branch of nuclear physics is the progeny of the controversial "cold fusion" phenomena, first reported in 1989, but now widely understood to have a solid empirical foundation. --Abd 17:19, 17 January 2011 (UTC)
  • Are you saying that modeling the energy left behind when moisture is vented away as mist (rather than as vapor) is original research? Are you saying that modeling burst noise from fluctuations in ohmic resistance is original research? Really? There is plenty of literature on both topics. It's just that there's not very much mention of it or attention paid to it in the CF literature. —Caprice 17:27, 17 January 2011 (UTC)
  • Yes, "mist" is original research, as presented by you. However, it's also trivial, and there is no controversy over the bare fact that mist would produce the error you predict, if it escapes unvaporized, but is treated as having vaporized. I got the answer to your question about sweat vs. urine, immediately. And more.
  • Let's get this clear: original research is permitted on Wikiversity. However, it should not be treated the same as what has been accepted, as shown by peer-reviewed publication, or, as a lesser standard, as published by a known expert.
  • With mist, the problem is that the practical application of the mist problem to real researhttp://moultonlava.blogspot.com/ch depends upon your speculations. You are correct in noting that the Miles-Fleischmann model assumes no mist escapes, without explicitly stating that. This is a valid criticism. But it also only applies to a certain class of experiments, and we have no evidence that any work where misting might have actually affected results was, in fact, emitting mist. It's speculation. You are welcome to note the lacuna, I intend to, to the extent possible, remedy it. If it's not possible, it nevertheless doesn't affect a major portion of the excess heat work. For example, it has no bearing on closed-cell work, such as that of McKubre, if the cells were truly closed, and it has no bearing on the Zhang work, because mist would not escape the calorimeter unvaporized. By design.
  • As to AC power noise, the use of controls and "dead cells" -- deuterium cells that show no excess heat, and cells that do, but don't show it during current excursions that would also produce bubble noise -- rules out significant AC power noise as a problem. I've also explained that major power supply noise would be very unlikely to escape attention. McKubre's comment about AC noise is possibly incorrect, as to theory, but quite correct as to the actual situation. (And understanding it with respect to theory might involve knowing much more about the nature of bubble noise. Frequency, contrary to your calculations, most certainly does matter.) In other words, at best you have discovered what could be a technical error with no consequence, and that is what the experimental evidence shows. You have neglected the experimental evidence that confirms the calorimetry. You have neglected that we are looking at a huge body of experimental evidence, and only some of it might be subject to this error. The general rejection of experimental evidence in favor of theory is what you've been doing, Barry, and this is a known and widely-discussed problem in this field. --Abd 18:39, 17 January 2011 (UTC)
Kirk Shanahan proposed mist five or six years ago. All I did was find a video that showed the mist in a close-up under bright sunlight. Today I was looking at some web pages of another amateur experimenter, Jean-Louis Naudin, who was replicating the Mizuno and Ohmori CFR design in which the dissociated gases are recombined directly in a spectacular brilliant plasma under water. Naudin also assumes all the lost electrolyte is in the vapor phase. But if about half of it is in the liquid phase, then his energy books would be in balance. One thing about Naudin's work is that he expressly includes the AC power, and you can see the dramatic fluctuations in voltage and current in his very high-energy plasma cells. —Caprice 19:17, 17 January 2011 (UTC)
The Naudin work is one where the mist criticism is quite cogent. Plasma electrolysis is very different from what happens in a Pons-Fleischmann cell. Kowalski addressed mist in the context of plasma electrolysis, as you noted before. Again, Naudin has violent conditions in the cell and AC noise would be prominent. That video you showed was of open electrolysis. If course there would be splash from the bubbles, of course there would be mist! (I believe the rate of gas evolution there was much higher than in a CF cell, and, remember, as you have been pointing out, deuterium bubbles are much less buoyant!) There was no calorimetry there, no confinement that would be likely to result in the mist depositing on the cell wall or being vaporized again before leaving the cell. Much CF work does detailed accounting for all the heavy water. Yes, Shanahan proposed mist, as I recall, but doesn't seem to have pursued it, which indicates to me that he didn't find that it had legs. --Abd 19:33, 17 January 2011 (UTC)
  • Naudin weighs the cell, before and after, to determine how many grams of water are lost. Presumably he weighs it quickly after the power is turned off, to avoid any further loss through evaporation. Naudin doesn't vent the D2 and O2 gases in his plasma cell; rather they are recombined chemically, while still under water, in that spectacular flaming plasma. All the moisture he loses is either as water vapor or as mist entrained in vapor. —Caprice 19:54, 17 January 2011 (UTC)
  • Fine. What does this have to do with the book, "Excess Heat"? Naudin's work isn't covered, I don't think any plasma electrolysis is. Let's get a little discipline here! --Abd 20:14, 17 January 2011 (UTC)
  • What does this have to do with the book, "Excess Heat"?
Beaudette entitled his book, Excess Heat, because it was about more heat coming out of the cells than the experimenters reckoned was going in. The mystery to be solved was where the excess heat was coming from. Naudin's work illustrates how, to this very day, people who have closely followed the literature are making the same sophomoric mistakes that Pons, Fleischman, Mizuno, Ohmori, and McKubre continue to make to this very day — leaving out of their energy budget such basic contributions as mist being mistaken for vapor and AC noise power being mistaken for audio frequency silence. —Caprice 20:25, 17 January 2011 (UTC)
Barry, you are speculating, behind all this. I have no idea of Naudin's qualification, nor any opinion about the quality of his research, and if he was doing plasma electrolysis and assuming only vapor escape, without having taken steps to nail down mist, he was indeed doing something stupid. But Beaudette doesn't cite Naudin, and if you can cite every stupid action taken by anybody anywhere as evidence about CF research, I could cite every stupid objection to cold fusion, including some by you. Not relevant here, and I'm not about to research Naudin to find out if he considered mist, anywhere. Please stop.
If Beaudette did cite Naudin, the place to cover Naudin would be the chapter where Beaudette cites him. Are you starting to get it? If plasma electrolysis is covered, but not Naudin, it might or might not be peripherally relevant. Okay? --Abd 20:45, 17 January 2011 (UTC)