(continued)
Quote[Kirk Shanahan] If you all choose to believe people who use strawman arguments to support their proposals, that’s your prerogative, but that action actually invalidates any derived conclusions, such as their supposed rebuttal. Net conclusion: no evidence of any true excess heat, ever (to date). (Rossi’s failure to do good calorimetry means his work doesn’t even get folded in…) (P.S. This also means there is no valid heat-He correlation, a fact Abd can’t seem to grasp.)
Kirk believes that he has found a systematic error, he calls it CCS. On the face of it, this is extremely unlikely, given the wide variety of calorimetric methods in use. There is no experimental evidence for his theory. That does not mean that it's wrong, but that nobody has tested it and confirmed what his theory might be thought to predict. Why not? Well, nobody is taking him seriously! But he wants me to take him seriously and discard a huge body of experimental work because it might be defective. In any case, he ends up with the classic argument, which Steve Jones also made, roughly.
If the heat evidence could be in error, then the heat/helium ratio is supposedly meaningless. (And tossed in there is that the helium measurements could also be in error. And that is actually a bit more likely than the heat being off, because there are people who are quite skilled at measuring heat, but measuring helium can be really, really difficult. The loss of precision in measuring heat/helium is mostly about the helium.)
However, Kirk doesn't understand the power of correlation, and his attack on heat/helium in the JEM paper was one of the dumbest arguments I've ever seen in a journal letter. Talk about poor review, but this was a Letter, not an actual paper of its own. It was so stupid that the scientists simply didn't understand it, and ignored it, just saying that his conclusion was wrong. I'm writing from memory here, so my apologies if something is not exactly straight. I covered this at length on [email protected], after first pointing this out to Kirk, privately, who blew me off with, "You will do anything to continue believing in cold fusion." So ... I published it.
Kirk digitized the data from a chart from Storms (2007) showing on the Y axis, He-4 atoms per watt-second, as I recall. On the x-axis was excess power. This was for a limited number of cells, reported by Miles and Bush and Lagowski. So Kirk took the points on that chart and calculated a correlation coefficient. It was very low. Why? It was because he was comparing heat/helium with heat (the sample interval was constant, so power was then in a constant relationship to energy or "heat"). If heat/helium is well-correlated, with a direct dependence, the ratio would be a constant. If not for noise, the correlation coefficient on that chart, then, with the ratio being plotted against the power, would be zero. That it was low showed correlation, not lack of correlation, but Shanahan thought the opposite, and this happens to us when we are sure we are correct. As soon as we find an argument that supports our position, we stop questioning it. At least that is the danger, eh, Kirk?
Krik might like to read my Current Science paper from last year, It is in the Special Section. This section has 34 invited papers, and, yes, it was invited. My paper first passed the section editor review, which was, indeed, easy, though Srinivasan was upset that I seemed to be deprecating tritium findings, where the Indians have done a lot of work. No, I'm merely pointing out that tritium is a very minor product, that if we want to look at the main show, we won't look at tritium, we will look at heat and helium and then we might start to look for signs of how the energy is being transferred from fusion to heat. This is one of the great mysteries of cold fusion, and it is still a mystery.
Many of those papers are reviews. To my knowledge, no critiques have been published. Maybe Kirk would like to write one? I can say that Current Science itself is not a 'cold fusion walled garden.' Once my paper was accepted for the section it then went for normal, anonymous peer review. The reviewer, rather obviously a physicist, I'd say, thought my paper was horrible and said so. So how did it end up getting published? Well, I looked at that response and said to myself, "Obviously, I failed to convince him. So I rewrote the paper! And he was bowled over. He was effusive. He helped write the conclusion. That, Kirk, is all part of my training, how to turn situations around. I could have been sputtering how unfair it was, I could have complained to the section editors and maybe they'd have found another reviewer. Maybe. Time was getting short. But I didn't go there, and the result was, undoubtedly, a better paper.
Here is a core paragraph from the paper:
QuoteUltimately, Miles reported 33 results from double-blind helium analysis. In 12 samples taken with no heat, none showed helium above measurement background. In 21 cells with heat, 18 showed helium and, generally, more the heat, more the helium produced6 . (Of the three major outliers, one was a cell where calorimetry error was reasonably suspected. The other two involved the only Pd–Ce alloy cathode used.)
Correlation. The zero-heat results are a crucial part of this data set. All other conditions were the same, and many of the measurements were from the same cell, merely samples taken at different times when different XP was being measured. This is controlled experiment then, with the no-heat cells being controls. (There is a hidden variable, the conditions in the cell causing the anomalous heat).
This actually confirms the heat measurements! To negate this conclusion, some other condition must be found that would explain the correlation. For example, suppose heat in the cell caused seals to leak. This could potentially create such a correlation. However, there are some problems with this.
One of the major ones is that it would not be heat that would change seals, it would be temperature. And these experiments were largely run at the same temperature. In some experiments, in fact, temperature is held constant and heating power to do that is varied as needed to maintain the temperature, so anomalous heat in the cell will show up as a reduction in heating power. But even if cells get hotter, it's only a few degrees. Not enough to make a drastic difference in leakage. This would also not be consistent, and then, that the ratio turns out to be consistent with the deuterium fusion value would have to be a rather amazing coincidence, eh?
The correlation has been measured under many very different experimental conditions. It remains the same.
Yes. The correlation is not perfect. There are those three outliers. And one of the items on a research agenda is looking at Pd-Ce cathodes. Miles wanted to do it but couldn't get the funding. Maybe something about the surface of a Pd-Ce cathode is an efficient helium trap. That would be easy to test!
The field of cold fusion moved on, cheerfully ignoring Kirk Shanahan, but he's certain they are wrong. However, it takes money to test hypothesis. Where will the money come from the test Kirk's CCS?
And it's all been finessed by heat/helium, which shows, already, that the heat results are at least roughly correct. Yes. A confirms B and B confirms A, and that can seem circular, but that's how correlation works. This does not show cause. In this case, heat is not causing helium and helium is not causing heat, but a process is causing both, together, it appears. (The alternative would be that heat is causing helium, as in the leakage idea, and helium causing heat at the levels involved is not going to get any serious consideration, for obvious reasons.) I can think of no way, so far, that heat could cause the helium results, given the actual experimental conditions.
Maybe Kirk can, and if he'd express it, it is not impossible that a test for it would be included in the work in progress.
Or he can continue to sputter, uselessly.