How many times has the Pons-Fleischmann Anomalous Heating Event been replicated in peer reviewed journals?

Quote from JedRothwell

Do not mix up apples and oranges. These are different tallies of different things.

Storms listed 180 labs in Table 2 of his first book. That is the number of labs that replicated excess heat. The number that replicated tritium and other effects are listed in other tables. Some labs replicated both heat and tritium. There were actually more than 180 labs, but Storms did not include them all in his database. 180 is enough to be sure, in any case. Five would be enough, in my opinion.

The number of experimental runs greatly exceeds the number of labs, because some labs run hundreds or thousands of tests, typically in an array.

Not all of these labs published papers in peer-reviewed journals. On the other hand, some published more than one paper. I list most of the peer-reviewed papers that are mainly about heat here:

http://lenr-canr.org/acrobat/RothwellJtallyofcol.pdf

This is a tally by D. Britz, not me.

Regarding the "essentially zero replications as proposed by the hyperskeptics" that has no basis in realty. Anyone can go to an academic library at a university and find hundreds of papers describing replications of cold fusion. If you want to argue that every single one of these replications was a mistake, you are arguing that the experimental method of science does not work. If that were true, we humans would still be living in caves. Some number of experiments are likely to be wrong, because people make mistakes. But people do not always make mistakes, day after day, year after year. Nothing would work if they did. When you get enough positive replications, done by a large number of people, the likelihood that every one of them is wrong is roughly comparable to the likelihood that today 1,000 electrochemists will simultaneously lose control of their cars and crash into telephone poles.

Display More

So for these multiply replicated results the issue is not are they replicable but what do the results tell us?

We have anomalous FPHE, always a small percentage of calibration constants, with CCS/ATER or some other calorimetry-related artifact on the table. For a novel heat source to be more plausible here, the issue is why we never get replicable larger results.

For the film results the issue is various experimental lacunae that can lead to them from other causes. We know they are sensitive to that.

For the tritium results these are all at level comparable with atmospheric He and therefore a complex analysis must be done to determine cause which takes into account all data and experiment selection mechanisms (e.g. how are leaks detected and dealt with in the data analysis).

Each separate set of results has different issues, and in all cases the issues are difficult because the data is marginal. This is unexpected given cold fusion, because the sensitivity of different methods of detection would be expected to be very different. For example, for radiation not to be much more sensitive than heat or tritium detection we need the LENR mechanism to alter reaction pathways from normal to ones that very largely favour stable results and non-gamma-producing reaction paths.

None of that is impossible, because LENR is not understood, nor are hypotheses coherent with other data, and therefore mechanisms can be hypothesised that fit whatever LENR experimental data exists.

But, it means that these replications do not constitute, for the wider scientific community, strong evidence of LENR.

There is a strange usage of the term "replication" around here. Most scientists use the term to mean that if you follow the same recipe, you get the same dish. Hundreds of observations of "excess heat" and a smattering of other things in experiments with different techniques and equipment do not constitute replications. They do suggest that something interesting may be going on and provide incentive to do more work. But until you can follow someone's detailed description of an experiment by using the same apparatus, methodology, and analysis and get a comparable result, the experiment has not been replicated.

Jed's constant refrain that it would take a boatload of money and two years for even the surviving masters of the field to produce results suggests that people cannot even replicate their own work, much less anyone else's.

Quote from JedRothwell

This is a tally by D. Britz, not me.

The tally developed by Britz was based on what was claimed in the paper he was summarizing, and does not imply any actual validity to the claim. Likewise, CFers routinely take any claim whatsoever to have replicated as 'one more for the count'.

In fact to be valid the experimental report must be analyzed for errors, and if any are found, it must be rejected, not automatically folded into the 'positive' count. So, my detection of a systematic error as published in 2002 invalidated the presumed validity of any prior claim to have observed excess heat until such time as the CCS problem was shown not to apply. To date, no one has done that. (Instead the CFers have attempted to 'prove' with faulty logic what I suggest can't happen, when it clearly can.) Thus ALL claims to excess heat observations are suspect and can't be counted as observations of real excess heat. Excess heat claims constitute the largest bulk of claims to have observed CF or LENR.

And there are problems with other blocks of claims as well. 4He for example. The number of times the claim has been shown to be greater than possible lab air 4He content is zero. Thus ALL 4He claims are suspect until proven otherwise. Like excess heat claims, the data may exist to do so so, but no CFer has ever actually done so.

The primary failure of the CF community is in their unwillingness to accurately assess the errors of their technology. They routinely underestimate error bars based on incomplete error analysis. They then claim they have high signal-to-noise ratios, but in fact they are working 'in the noise'. This is one of the prime characteristics of pathological scientists.

Quote from maryyugo

The arguments are that the experiments are "noisy" and the signal is not significantly stronger than the noise and that in almost three (1989-2017) decades, this has not significantly improved.

There are two things wrong with that.

First, it isn't a bit true. Results from 100 W with no input power are much stronger than, say, 50 mW excess with 4 W of input.

Second, even if it were true, it is not a valid metric for judging an experimental result. Many scientific discoveries remained difficult to reproduce or measure for decades, yet no one denied they are real.

Quote from interested observer

There is a strange usage of the term "replication" around here. Most scientists use the term to mean that if you follow the same recipe, you get the same dish. ... suggests that people cannot even replicate their own work, much less anyone else's.

Exactly. I have been wondering if I should point this out. The easiest way to clarify the discussion is to start talking about 'control' instead of 'replication'. To be convincing, you have to demonstrate at least partial control over the effect. What has been done so far is just barely beginning to reach that point. The fact that they have worked for decades to control the reaction and have failed is usually taken to mean the premises used to design experiments are incorrect. That's why I presented the ATER mechanism, to give the CFers an alternative starting point. Of course it's not nuclear, so they simply ignore it, rather than trying to use it.

Quote from kirkshanahan

The tally developed by Britz was based on what was claimed in the paper he was summarizing, and does not imply any actual validity to the claim.

Actually it was biased against the claims. As I point out in my analysis, several authors stated clearly that their results were positive, but Britz put them down as negative. Despite that, I reported his totals.

But -- as always -- you are missing the point. If hundreds of researchers could make thousands of mistakes over 20 years, doing what they were trained to do in their own specialty, then science would not work. No experiment would be meaningful. I am not exaggerating when I say that civilization itself would not exist.

Of course individuals can be wrong. Or they can be crackpots, like you, who imagine they know more about this subject than people like Fleischmann and Bockris, or Faraday, for that matter. You are a legend in your own mind. Alas, your claims cannot be tested or falsified, and you have not addressed any of the reasons given by the experts in the Marwan paper showing that you are wrong. But, such considerations never faze a crackpot!

I suggest you write papers showing errors in Einstein's theories. That is the usual target of deluded crackpots like you.

Quote from THHuxleynew

For the tritium results these are all at level comparable with atmospheric He and therefore a complex analysis must be done to determine cause which takes into account all data and experiment selection mechanisms (e.g. how are leaks detected and dealt with in the data analysis).

This must be a mistake. Tritium results have nothing to do with atmospheric helium. In any case:

Tritium typically ranges from 50 times background to 10E14 times background.

Helium is sometimes lower than atmospheric background. So low, in fact, that it cannot be from a leak or it would be random. In other cases it is higher, for two reasons. Either because it was produced at high rates and it climbed above background, or because the cell was initially filled with enough helium to make it higher than atmospheric background.

There is no selection bias in most studies. All of the results, including negative ones, are listed in the reports. Especially F. Will's tritium studies, where they went to a lot of trouble running many deliberate blanks.

Quote from JedRothwell

Results from 100 W with no input power are much stronger than, say, 50 mW excess with 4 W of input.

If you will cite a reference that I can look at, I'll let you know why this is likely wrong.

Quote from JedRothwell

But -- as always -- you are missing the point. If hundreds of researchers could make thousands of mistakes over 20 years, doing what they were trained to do in their own specialty, then science would not work

And as always, you are missing the point. Systematic errors are a well-known problem in science, and yes, they will affect 'hundreds of researchers' 'thousands' of times as long as those researchers are unaware of the problem. In fact, it's quite possible 'excess heat' claims represent a good case history of just such an occurrence. Still, science works, because there is usually some small inconsistency that gives the error away, or an alternate method to get the same result is devised that comes up with a different answer. Then science self-corrects, and people go back and reconsider prior results to see if they can be salvaged or not, while continuing on with experimental designs that don't have the error in them anymore. The failure of the CF community to do this brands them as pathological scientists.

Quote from JedRothwell

Helium is sometimes lower than atmospheric background. So low, in fact, that it cannot be from a leak or it would be random. In other cases it is higher, for two reasons. Either because it was produced at high rates and it climbed above background, or because the cell was initially filled with enough helium to make it higher than atmospheric background.

Not sometimes, almost always (all of Miles' he results are at the ppb levels for example). But of course the real problem here is the idea that the air in the lab is the same as the air in the middle of a corn field in Nebraska or the center of the Amazon jungle in South America. Labs have lots of people who use 4He for many different purposes. There are actual reports (from Miles I believe) of 70 ppm 4He in his lab air. I have information that McK's lab had 150 ppm at one point. So, the use of the 5.22 ppm number is nothing but a red herring. CF researchers *must* collect and measure *lab* air samples during CF experiments to establish the contamination potential in their experiments. It is these numbers which sometimes are alluded to, but almost never published, and then not in concert with the related CF cell exit gas sample analyses.

Quote from kirkshanahan

If you will cite a reference that I can look at, I'll let you know why this is likely wrong.

Give me a break. You know damn well what I am talking about. I am not going to keep giving you and Yugo the same information time after time, just to have you ignore it.

Hint: see my video and accompanying documentation. Look for it yourself.

Quote from maryyugo

Want to point me to that link again please?

Why the hell should I tell you the same thing again and again and again? You will never read it.

As I said, see the video.

So, no thanks to Jed, I believe he is referring to this paper: http://lenr-canr.org/acrobat/RouletteTresultsofi.pdf

This details experiments carried out by Pons and associates while at the IRMA lab (Toyota funded) in 1996. Note that this is prior to my CCS publication, so this paper will NOT consider the CCS at all.

Based on this paper, I can say Jed’s comment (“Results from 100 W with no input power”) makes no sense. The paper describes 7 experiments conducted in a new calorimeter, 4 gave no excess, one is listed as ‘Variable’ % excess power with 30.5 MJ of “total energy”, presumably ‘produced’. (Based on the only data table in the paper, which Jed uses in his video.) One gave 150% excess power with 294 MJ produced, and the 4^th in sequence gave 250% excess and 102 MJ produced. There is a line in the table with the label “Pwrexcess/W/4.2hr” which is unclear to me. The values listed are (-0.1, -0.6, 101, 17.3, 13.8, 74.5, 39.4). This is the only line of data that has “W” in the label, so it this the 100W Jed refers to? (If so it refers to run 3, which we have no information on besides what is in the table.)

The paper has several figures of data from experiment 4 (the one that produced 250% excess and 102 MJ). Presumably, the input power profiles are similar for all the experiments (hopefully?). You have to be careful in reading them because some of them have a 1/100 factor involved. The input power however, is never 0, as Jed claims. Figure 3 gives cell voltage (divided by 100) and Figure 4 give the current. During the 1^st 35 hours, the Pin is low, Pin = 0.21A * 0.8*100 V = 17 W. Then the amps jumps up to ~0.9, and the voltage does some strange things. First it starts to increase in a slow, non-linear fashion until the 48^th hr, where it jumps up to about 84V, so Pin is .9*84 = 75.6 W. But this is certainly not “0”. So Jed’s comment seems to be unrelated to this run at least.

THEREFORE due to lack of data and/or misstatements by Jed, I cannot confirm “100W excess power with zero input”. This is a typical result when examining cold fusion reports in detail.

Care to clarify Jed?

More confusion… Figure 5 gives “Cell Temperature”, apparently in degrees C, but the values plotted range from 0 to 1.0. So, I have no idea what this is. No mention of a 100 factor, but maybe we should multiply by 100 again?

The cell has three thermistors installed at different vertical heights (in a typical F&P cell design). I believe the calorimetric method is a type of isoperibolic calorimetry (i.e. Pout = k * (Tcell – Tbath)). They show the ‘excess power derived for each thermistor. They also note that depending on input power regime a different thermistor is used. In the 4^th run, the bottom one was not valid. Figure 7 and Figure 8 show this for the top and middle thermistor, and both show ~100W signals lasting ~25 hours, with a lead-in ramp up lasting about 16 hours and a following ramp down lasting about 7 hours or so. Figure 9 for the bottom thermistor shows a flat-lined 0 value signal.

Figure 10 shows the % excess heat but the peak value is ~ .16% of Pin (again no 100X mentioned but maybe this is actually 16%? Still, not 150%...).

Figure 11 shows accumulated excess energy and ends at ~310 MJ, but the table lists that run as producing 101 MJ. (Integrated Pin over that period (16+25+7 hrs) is 75W * 48 hrs * 3600 sec/hr = 13.13 MJ.)

So again I can’t make heads or tails from this. What gives Jed? Are you really claiming this paper shows 100W power out with no input? How do you do that?

BTW, there are no calibration equations or coefficients given, so I can’t easily estimate the possible CCS magnitude.

Show me where I’m wrong here folks….