Uploaded Letters from Martin Fleischmann to Melvin Miles

I do not think so. See the debate between Morrison and Fleischmann. I do not know any experts in calorimetry, nuclear reactions or electrochemistry who agree with Morrison. None of his claims has any merit in my opinion. If you disagree, you are wrong. Wrong with 99.99% probability. It is very unlikely you will find a problem in experiments that have analyzed this carefully by this many experts, in peer-review and elsewhere.

I raised a specific point (that condensed liquid egress from these cells during the boil-off phase was very possible, and would match the results). You no doubt can refute it if you are 99.99% certain?

Re peer review. You are guilty of overconfidence. Peer review catches many issues (when well done, which is not universally the case). However it could never catch all issues, nor is it any statement that a paper's conclusions or arguments are correct. The whole point of science is that interesting papers contribute - they are not the 99.99% final word.

Quote from JedRothwell

Rats avoid warm water. (Really.) .......

That RAT scenario was, is, and will always be a JOKE, Jed.

The swimming-pool evaporation rate was serious.

Quote from JedRothwell

Yes, I can. Fleischmann gave a number of reasons why this could not have happened. For example, they inventoried the salt left in the cell and found that no significant amount left. And they ran blanks where the energy balance was zero. I gave you the list of reasons why this did not happen. I did not see a response from you. If you did respond, and you still think there is a problem, I would say it is 99.99% certain you are wrong.

To put it another way, these methods are industry standard and have been in use for decades. Some of them have been used since medieval times, in distillation. It is very unlikely you discovered that standard techniques that have been used millions of times do not actually work. If you did that, you would win a Nobel prize.

Along the same lines, it is very unlikely that Shanahan has discovered errors in textbook calorimetry that other people have overlooked since the 1840s.

Jed, please read the discussion here carefully. You will note that I am considering vapour that condenses before it exists the cell as entrained liquid-phase D2O. This would not alter the salt balance. So that check fails. And since this is a logical point I can say 100% that if D2O evaporates or boils from the liquid and then condenses, forming entrained droplets, no significant amount of salt will leave the cell, because the vapour-phase steam cannot contain salt. Effectively, the exiting liquid is being distilled. This is exactly what happens in a kettle and is why you see visible steam a little way from the spout - after steam has condensed. In a kettle the water coming out of the spout is also pretty well pure H2O with no salt content. Whether the precise configuration used in this experiment would encourage such in-air condensation within the calorimetric envelope of the experiment - so that heat given up by condensing steam stays in the vessel - would need some investigation. I did not read in F&P's paper the details of this. Did you?

Along the same lines, the above argument shows how easy it is for something to be overlooked by industry standard methods - if doing calorimetry on an open fast boiling vessel can be viewed as industry-standard.

When dealing with anomalies, whether from LENR or CCS/ATER, the one thing it is unwise to do is to rely on the received wisdom of ages without careful checking. in this specific case the salt measurement is not a sufficient check for the reasons given above.

Shanahan has not discovered new errors in calorimetry, merely noted an error not considered in these cases. Most calorimetrists would agree that change in cell temperature distribution can alter calibration but consider it is not usually a problem because (a) stirring can reduce it and (b) multiple isothermal barriers can eliminate the effect of in-cell temperature changes on efficiency. They would look at the details of specific experiments to check, with controls as needed. In this case the anomalous issue is ATER which calorimetrists, as you point out, think does not happen because they do not consider the unusual possibility of catalytic effects in a specially prepared transition-metal electrode. If ATER exists it increases the possibility of anomalous in-cell temperature variation in precisely the experiments that are considered good candidates for LENR excess heat anomalies and the effect (ATER) would be every bit as elusive as LENR, therefore it is understandable that it is not commonly understood.

Is Shanahan's hypothesis here correct? I don't know, but were I wanting to validate electrochemical excess heat as proof of an anomaly I would have to consider it very seriously, with additional instrumentation (quite easy) to monitor possible ATER, or with 99% efficient calorimeters. With less than 1% heat lost the error due to CCS/ATER would be no more than 1% and a typical (?) 10% LENR excess heat signal could safely be detected.

Marwan et al point out that high-efficiency calorimeters cannot therefore suffer CCS. They are right. They do not indicate which of the classic results therefore remain safe when CCS is considered, as would be a logical next investigation, for someone seeking as they do to defend these results from critique.

Quote from JedRothwell

I believe he agrees that it works when there is no excess heat. For example, a blank run with Pt-H does not produce heat, so he says the calorimetry is working. I do not see how the choice of metal and electrolyte could trigger the effects he describes. I also do not see how his hypothesis would avoid producing a false endothermic result as often as a false exothermic one. And I do not see how this could have been happening since Faraday's time, but no one noticed it. However, I am not prepared to debate this in detail. See Marwan if you wish to know why Shanahan is wrong. Or read any textbook.

Jed, you are at liberty to leave this debate. Let me point out I have read Marwan et al and the critique therein of CCS/ATER is highly partial and based on assumption (ATER can't occur) and straw men (CCS does not significantly apply to 99% efficiency calorimeters). This has been considered on another thread but I find it interesting and will happily look at it more. Perhaps we could itemise all the rebuttal points made and consider the limitations and strengths of each in turn.

For your point here, a good one, why does CCS not deliver negative results, the answer is that for CCS to deliver surprising results we need some way in which the temperature distribution inside the cell varies significantly between control and active runs. ATER provides this, and with the design of all (to my knowledge) cells showing consistent strong LENR readings it will result in over-reading if it applies. I know of no (unexpected to classic calorimetry) mechanism that would deliver negative results in active as compared with control CF experiments. The fact that D often shows ATER when H does not is a one-off fact, and if Shanahan is correct possibly the reason why CF was initially seen. For a negative result you'd need ATER in the control experiment with no ATER in the active one. Since ATER comes from special preparation of a special electrode (like LENR) it is difficult to see how you can control for it when looking for LENR. You can add instrumentation to detect CCS, so it is not a problem for an experimenter who admits ATER/CCS might exist.

No-one is saying that CCS/ATER applies to all electrolytic cell experiments. (At least I'm not saying that). Shanahan will say that CCS can in principle apply to some extent to all experiments, which is a different statement. For example, as above, boiling open vessels as used by F&P showing high apparent excess heat coincident with fast boiling can possibly be explained by EEL (entrained evaporated liquid) as I suggested above. Perhaps, when all these red herrings are eliminated, some other cells show real LENR. I cannot for the life of me see why somone serious in investigating LENR would not look with an open mind at other hypotheses for anomalies: yet Marwan et al, from my reading of their paper and comparison with Shanahan's offerings, have not done this. I can see it is easy for them to dismiss Shanahan, they think with high probability he is wrong. Unfortunately that looks to the outside world very like the many scientists who think with high probability that LENR is wrong and therefore do not bother to consider it.

It is, I would hope, purely a feature of internet blogs that such a partisan analysis is common. Marwan et al shows my hopes here are not, in at least one case, realistic. That paper contains some classic blog-style dismissal. For one specific look at the way they argue that CCS cannot apply because it does not apply to ultra-high efficiency calorimeters. That is a specious argument: it can apply to the majority of experiments where claimed excess heat + calorimeter absolute efficiency < 100%.

Shanahan is correct in viewing Marwan et al as poor rebuttal of his ideas, where in fcat CCS/ATER is significant in any of the classic CF results or no.

A note on scientific argument. Nothing is certain when logical arguments get applied to the real world. All experimenters must evaluate results based on a balance of probabilities. When results are highly variable, with no lab rat experiment, and do not have any mechanism supported by other experiment, the probability that they are based on some not understood methodological anomaly, like CCS/ATER or EEL, is larger than normal, and must therefore be considered more carefully than normal. This is the same impetus given to LENR. Where anomalies are seen we relax our assumptions about what is likely in order to try and find the anomaly. I'm all for that, but all against it being done for LENR and not done for CCS/ATER, EEL, etc. That path (preferential interest in a single anomaly mechanism) leads to pseudo-science.

Quote from Shane D.

Within this perspective, I find it interesting that KS claims all CF calorimetry is/was wrong. I wonder what F would have said, were he still around to read that? SRI reads LF, and I am sure they took exception to that blanket statement also. Maybe KS did not really mean to say it that way.

I don't think you have reached the pages I quoted in my first post in this thread (where I tried to focus on Fleischmann, and not Mizuno), but the answer is right there. He got so irritated at my work he threw it away. And I note that I wasn't even talking about his work in 2002-3.

My blanket statement stands as stated. To my knowledge, to this day, no one has done the F&P CF cell experiment calorimetry right. They all assume no effect from a heterogeneous temperature distribution, and thus mathematically treat the system as if it were totally homogeneous. But CCS in this case comes from shifting around the heat. As long as they have to potential to generate heat in an unexpected place (by recombination most likely), that assumption will cause difficulties.

If I had been in communication with F, I would have pointed out what I did in my whitepaper (2012 I think?). The comments F makes in his letters pretty much convince me he would have not reacted well.

Quote from AlainCo

it is funny that we replicate more and more, and in fact the more it is replicated, the more it need to be replicated, while being less and less replicated...

Sometime I think there is a logical problem in the way this affair is managed? don't you think so?

I agree, there is a logical problem. It boils down to the old definition of insanity, which is doing the same thing over and over and expecting a different result.

The correct approach to all this is to run experiments probing what effect varying select variables have, and when they have none, dropping them out of consideration, and adding new ones in. 'ATER' gives the CFers a whole new set of variables to explore if they chose to.

"I raised a specific point (that condensed liquid egress from these cells during the boil-off phase was very possible, and would match the results). You no doubt can refute it if you are 99.99% certain?"

"Yes, I can. Fleischmann gave a number of reasons why this could not have happened. For example, they inventoried the salt left in the cell and found that no significant amount left. And they ran blanks where the energy balance was zero. I gave you the list of reasons why this did not happen. I did not see a response from you. If you did respond, and you still think there is a problem, I would say it is 99.99% certain you are wrong."

The 2004 publication by Szpak, Mosier-Boss, Miles, and Fleishmann (SMMF) reports a 7% excess in the water collected from the cell's exit. It amounted to about 0.5 cc of water. In my comment on that paper in 2005 I pointed that out and suggested it was attributable to entrained microdroplets. SMMF responded in peer review that 'that was just noise'. The question then regarding their supposed measurement of electrolyte salt concentration is: What is the error on that and how does it impact conclusions? I doubt they did it accurately enough to account for a loss on the order of 7-10%.

The point being that this is their SOP, errors on the order of 10% or so. The CCS error that I found in Ed Storms work was a 1% level effect but it gave heat errors of 780mW (and possibly larger). They need to do their work at the 1% level and document that. Otherwise they are just blowing smoke.

Quote from JedRothwell

Shanahan has not addressed these issues as far as I know.

Yes, I have. Every time you've raised them.

And, Jed, I answer your issues too.

I addressed that earlier. That would not alter the energy balance. The heat of vaporization would be returned to the cell. The headspace is small and very little of it would radiate out.

That is the whole point. During the boil-off phase F calculates his very large enthalpy by including the heat of vaporisation in the energy balance: as would be correct for what he assumes: vapour egress from the cell.

He knows that liquid egress would not give this large additional enthalpy which is why he checks for liquid egress by looking at salt loss in cell, which he finds insignificant.

That, however does not apply to recondensed liquid. Note that vapourise + recondense is enthalpy-neutral, so as long a the recondensing occurs inside the calorimetry envelope (my supposition above) this is that same for the enthalpy balance as if it were entrained droplets direct from the surface.

Furthermore, if this were to cause false excess heat with Pd-D, it would also happen with Pt-H calibration. So, there would be spurious excess heat from headspace condensation during the calibration. That does not happen

When did he calibrate under extreme boil-off conditions? He does not. Furthermore as KS and I have pointed ot may times here the physical properties of D2 & H2 are very different, so you cannot assume they will behave identically as regards this phenomenon.

Yes, I know. That is impossible for three reasons:

1. There is no conceivable reason why this would happen with Pd-D and not Pt-H or, for that matter, a joule heater during calibration.

Shanahan has addressed this in detail with the supposition that ATER is an unusual catalytic effect occurring in active environments in the electrodes caused by changes during electrolysis (and also possibly previous electrode state). It would certainly depend on electrode metal, and on D vs H since these two have very different physical properties. I realise that it would be speculative to put forward an exact mechanism at this stage: just as it is speculative (and difficult) to find an exact mechanism that fits LENR.

2. Excess heat is observed with cells where the temperature distribution within the cell cannot possibly affect the calorimetry, because the heat is measured outside the cell. Several centimeters outside of it, in some cases, such as with Seebeck and flow calorimeters. It is also impossible when the temperature is measured outside the cell with a copper sheath. And it is impossible when the temperature in the cells is measured in multiple locations in the cell and shown to be uniform. Those three conditions cover just about every cell I have ever heard of. I do not recall seeing a cell that would be fooled by this, even if it could occur.

This would be a strong argument except that the conditions you need are not often stated to exist. A good isothermal barrier with no leakage would solve the problem, however CF cells always have some leakage from the electrode connections and (if open) the vent. They also (if closed) have a recombiner in the headspace above the electrolyte. If open they have other issues, notably the one I address above, which is why quite correctly McKubre used closed cells for greater accuracy. Researchers sometimes make multiple temperature measurements in the liquid. This however does not catch the mechanism here, a change in temperature differential between the recombiner and the liquid. I'd agree that it is possible (for example in high efficiency calorimeters) to bound any such effect. The question is which experiments cannot be so bounded. You would need to look at each experiment series to check. If this were done, those that survived this check would be a stronger set of data to carry forward.

3. In some instances, the location of the heat within the cell is moved, deliberately, by a much longer distance to a greater extent than the CCS hypothesis could ever produce (if it were real). This happens, for example, when calibrating with a joule heater inside a cell that also has an anode and cathode.

If the heater remains in the electrolyte there is no such issue, as above.

It also happens with a recombiner in the headspace. This change in the location of heat production never produces a measurable effect.

That makes my point. Calibration is done via electrolysis with the recombiner, and any effect is lumped into the calibration coefficients. It is only when ATER occurs that there is a measurable change.

It is indeed fortunate that cold fusion has been replicated thousands of times, often with great uniformity, at places like Toyota. By your standard that makes it more of a science than, say, Tokamak plasma fusion, or the confirmation of the top quark, or most robot exploration of Mars, which are one-off events, done by only one group. Your standard is new and unprecedented. I suspect you invented it just for cold fusion, although it does not, in fact, fit. It resembles your recent assertion that Mizuno was depending on memory for his descriptions of the heat-after-death event. You made that up, and it wasn't a bit true.

Those are good examples.

In both LHC and Mars Rover single experiment results are not trusted. There is too much possibility of some unexpected error, or in the case of Mars Rovers a sensor behaving badly. Both conduct multiple experiments where the same phenomena can be observed via multiple independent pathways; it is these agreeing that allows confidence. Also (with a bit less confidence) there are sometimes two Rovers, but that does not knock on the head some systematic error mechanism affecting both, if the experiments are identical.

LHC is studying high energy physics along with many other devices. Results from LHC can (and are) combined with evidence from other separated experiments. There is no question of a single experiment: even of a single type of experiment being trusted for anything surprising, though when a single experiment delivers results in line with other experiments it will tend to be believed because of the additional validation.

When single experiments do produce surprising results, as with FTL neutrinos, what happens?

(1) Massive efforts from the group doing the original experiment to publish every detail

(2) Massive efforts from everyone else (and the original team) to check further and look for errors not originally noticed.

(3) Massive efforts from theoreticians to explain the results (under hypothesis that results are real).

(4) Other groups try to replicate.

A classic example of how science responds to very unexpected observations.

For FTL neutrinos you will notice it was (2) that solved the mystery.

The 2004 publication by Szpak, Mosier-Boss, Miles, and Fleishmann (SMMF) reports a 7% excess in the water collected from the cell's exit. It amounted to about 0.5 cc of water. In my comment on that paper in 2005 I pointed that out and suggested it was attributable to entrained microdroplets. SMMF responded in per review that 'that was just noise'. The question then regarding their supposed measurement of electrolyte salt concentration is: What is the error on that and how does it impact conclusions? I doubt they did it accurately enough to account for a loss on the order of 7-10%.

It is important not to generalise from one experiment to another, each of these experiments can have different characteristics, and also different measurements made. Jed has here generalised from the fact that specific checks are sometimes done to assume that with 99.9% certainty those checks are always done. Adequately.

This is a possible example of the same phenomena. Without details of the salt measurments and the error bars of those we cannot be sure, but your figures here (which we do know) make it not impossible in that case salt measurement would not have been accurate enough. We (I think?) no such details of the salt measurements from F, and therefore must trust his judgement, ill-advised because everyone makes mistakes. In any case, as I've pointed out above, salt measurements do not limit the amount of liquid phase electrolyte leaving a cell because it can be either direct (carrying salt) or recondensed (with no salt).

Quote from THHuxleynew

We (I think?) [know] no such details of the salt measurements from F, and therefore must trust his judgement, ill-advised because everyone makes mistakes.

The problem with that is that CF researchers have shown an almost universal tendency to overestimate the accuracy and precision of their measurements. Thus F would say "I measured that and it showed nothing." when in fact his error bars would preclude drawing that consideration if they had been accurately determined and properly utilized.