(Deleted comment.)
Cambridge University Professor Huw Price on the ‘Reputation Trap’ of Cold Fusion (Update: Response in Popular Mechanics)
-
-
Quote
How is a claim of four orders of magnitude fewer pits when H2O is used than when D2O is used a weak finding? It sounds like a strong finding to me, if true. For various reasons I agree with Abd that their work is likely to be solid work, and I and disagree with you that we have a weak finding. I disagree with Abd about the likelihood of this stuff being a side channel to LENR. But that's a different discussion.
Well without looking in detail at the paper I could not know, there are a number of mundane ways this result could happen. The most obvious (but because of that not the most likely) is tritium contamination of the D2O. That is easily controlled for - so a even halfway decent experiment would eliminate that.
If you post this "very strong evidence" paper reference I will say more.
-
Quote
You and Abd do not agree at all. Abd's saying the research by Mosier-Boss et al. is likely to be solid, but the phenomenon under study is likely to be a side channel in LENR.
I think it likely that Abd and I agree in evaluation of the paper evidence. However Abd (for other reasons) is convinced that LENR happens, quite often, in solid lattices. Therefore he will see a nuclear interpretation of marginal evidence as normal, and quite likely, whereas I am not so convinced, it is for me extraordinary and something you would consider only when other more likely alternatives have been definitively ruled out.
In this debate if you become a believer subsequent interpretation of marginal evidence will be biassed that way: you will naturally not have the due caution that anyone usually has encountering claims of something extraordinary and unproven.
-
Here is another CR39 paper with strong evidence of LENR tracks, i.e. Nuclear tracks far above background levels.
http://www.lenr-canr.org/acrobat/OrianiRAreproducib.pdf
- 43 different control Runs showed a mean track densities varying from 6 to 16 tracks / cm2. The highest recorded number was 24.
- in 25 actual experiments 15 experiments had average more than 100 tracks/cm2. One had track number so great that counting was impractical.
- typically tracks where also found in groups with much higher density than the average indicated above.Clusters "can not be generated by a sequential decay of ordinary radionuclides dissolved in the electrolyte. Such a source of charged particles could not remain stationary in the convection currents caused by the bubbling during electrolysis long enough to produce a radial distribution of elliptical etch pits whose axes intersect at one common point. Such clusters furnish additional evidence that a nuclear reaction of unknown nature can develop during electrolysis."
-
Oystla
I read briefly the paper, such evidence is mounting I would say although I can quite see Thomas's position as being valuable, although there is I think a great risk of bias on both sides, good to see reasoned debate in the ascendancy though. I think Huw Price would be confident mainstream scientists may be showing at least guarded interest.
Extract:
"A technique has been developed that consistently produces evidence that a nuclear process can accompany electrolysis of solutions of lithium salts in either heavy or light water. The evidence is in the form of nuclear damage trails made visible by the etching of CR39 plastic chips. The damage trails can begin at either external surface of the chip placed within the cell during electrolysis as well as within the interior if the thickness of the chips. It is demonstrated that the nuclear damage trails could not have been caused by the decay of ordinary radionuclides contaminating anything in the experimental procedure. Rather, the nuclear damage is caused by a nuclear process of currently unknown nature".If that demonstration could be repeated and verified in other labs, would that be considered some kind of 'tipping point' for LENR? or would the critics, perhaps reasonably, set the bar even higher?
-
Frank, one thing is certain in my mind
The human kind will use LENR commercially long before LENR is generally accepeted science and long before we understand the core of LENR.
just as we did with fire.....
but "should I deny myself a good dinner, just because I don't understand the science of digestion?"
-
Well without looking in detail at the paper I could not know, there are a number of mundane ways this result could happen. The most obvious (but because of that not the most likely) is tritium contamination of the D2O. That is easily controlled for - so a even halfway decent experiment would eliminate that.
I recall them reporting on tritium; hopefully this recollection is not mistaken.Perhaps there were mundane ways in which their results went back to error. A challenge would be to formulate a hypothesis and to get it through peer review, convincing specialists that the hypothesis is plausible. Even if this is done in a second-rate journal with a low impact factor. Note that Mosier-Boss et al. don't mention LENR, and they have published in a low-profile journal. They have followed your recommendation.
They do lots of controls.
Suppose you came up with a hypothesis involving mundane error that seemed plausible to you, but you did not have the time or energy to write it up and get it through peer review. A reasonable person, also lacking expertise in the specialties this particular research, could legitimately ask, "Ok. Tom thinks this hypothesis is plausible. But is it? What would a specialist think of it? What would Mosier-Boss et al. say? Perhaps they would have a quick reply."
-
Quote
Perhaps there were mundane ways in which their results went back to error. A challenge would be to formulate a hypothesis and to get it through peer review, convincing specialists that the hypothesis is plausible. Even if this is done in a second-rate journal with a low impact factor. Note that Mosier-Boss et al. don't mention LENR, and they have published in a low-profile journal. They have followed your recommendation. They do lots of controls.Suppose you came up with a hypothesis involving mundane error that seemed plausible to you, but you did not have the time or energy to write it up and get it through peer review. A reasonable person, also lacking expertise in the specialties this particular research, could legitimately ask, "Ok. Tom thinks this hypothesis is plausible. But is it? What would a specialist think of it? What would Mosier-Boss et al. say? Perhaps they would have a quick reply."
There is an inherent problem with the MB results. They do a lot of controls because the results are very not straightforward and the possible artifacts are large in number. So a refutation would not likely be - "I know the artifact is this" but just "I challenge you to prove the artifact could not be something vague". I've only read one paper which was clearly flaky, with a very complex set of assumptions and bad reporting of the actual evidence.
The general problem with all this evidence is because the key observation is radiation so weak that it can only be captured by an integrating radiation meter. This is inherently weak evidence. If there are really nuclear products from LENR than you'd expect to be able to get more radiation, properly measurable, by altering reaction rates. after all, LENR is billed as having a very variable reaction rate...
As soon as we have direct detection then we can also resolve time and the range of artifacts reduces enormously.
is there one paper that in particular has convincing data properly reported? I'll look at it.
BTW - what would you do, publishing a paper which said - "hey, look, this claimed evidence is very weak" - and cites a whole load of obvious general reasons for this. It would not be very interesting! I'm finding it hard to motivate myself doing this here and the time required for a decent paper is 100X higher.
-
There is an inherent problem with the MB results. They do a lot of controls because the results are very not straightforward and the possible artifacts are large in number. So a refutation would not likely be - "I know the artifact is this" but just "I challenge you to prove the artifact could not be something vague". I've only read one paper which was clearly flaky, with a very complex set of assumptions and bad reporting of the actual evidence.
Sounds like shoddy science. The referees should be very sympathetic with your position, and your job should be an easy one. List out all of the things that are bad about it: you didn't do proper controls; what you're detecting is not straightforward; there are a large number of possible artifacts (e.g., a, b, c, d); your signal is very weak, etc. See what the referees and Mosier-Boss et al. say in return.The general problem with all this evidence is because the key observation is radiation so weak that it can only be captured by an integrating radiation meter. This is inherently weak evidence.
Perhaps we're thinking of different things by the word "weak". When you say this, what do you have in mind? Are you thinking of "weak," in the sense of not statistically significant, or "weak," in the sense of you can't heat your home with it? They claimed to see four orders of magnitude more pits when D2O was used than when H2O was used. What is the statistical significance of a 1000-fold increase? Note that control runs show almost no pits, which gives us a sense of the experimental uncertainty.When I think of "weak," I think of "not statistically significant," or "hardly statistically significant." That does not appear to apply to their work; or does it?
If there are really nuclear products from LENR than you'd expect to be able to get more radiation, properly measurable, by altering reaction rates. after all, LENR is billed as having a very variable reaction rate...
Perhaps it's hard to alter reaction rates in their Pd/D co-deposition experimental setup. Also, this might not be LENR -- Ed Storms doesn't think it is more than a secondary hot fusion process, in the manner of piezonuclear fusion. But perhaps it is LENR. I think it probably is. Whatever it is, it seems like an anomaly. What's going on with it? Note that there have been experiments by other groups that have reported a much higher charged particle current.As soon as we have direct detection then we can also resolve time and the range of artifacts reduces enormously.
Yes, I agree that online detection would be great.BTW - what would you do, publishing a paper which said - "hey, look, this claimed evidence is very weak" - and cites a whole load of obvious general reasons for this. It would not be very interesting! I'm finding it hard to motivate myself doing this here and the time required for a decent paper is 100X higher.
Like I said, the way you make it out, it sounds like shoddy science. What referee would disagree with you?My unstated thesis is that we've been casting aspersions on their work, without going to the trouble of actually putting together a proper criticism. Mosier-Boss et al. do not have a chance to rebut. Hobbyists on sites like this have little more than intuition to go on to know whether your generalizations are well-founded or not. I think it would be a great learning process for you to try to put your hunches about this body of work into a rigorous criticism -- perhaps you would discover that it is not so straightforward to deconstruct their work. Or perhaps I'm wrong, and it would be quite easy, once the time and effort have been invested.
-
Quote
Perhaps it's hard to alter reaction rates in their Pd/D co-deposition experimental setup. Also, this might not be LENR -- Ed Storms doesn't think it is more than a secondary hot fusion process, in the manner of piezonuclear fusion. But perhaps it is LENR. I think it probably is. Whatever it is, it seems like an anomaly. What's going on with it? Note that there have been experiments by other groups that have reported a much higher charged particle current.
Perhaps when the effect always alters so as to be difficult to measure - that is because it is not a real effect?
QuoteMy unstated thesis is that we've been casting aspersions on their work, without going to the trouble of actually putting together a proper criticism. Mosier-Boss et al. do not have a chance to rebut. Hobbyists on sites like this have little more than intuition to go on to know whether your generalizations are well-founded or not. I think it would be a great learning process for you to try to put your hunches about this body of work into a rigorous criticism -- perhaps you would discover that it is not so straightforward to deconstruct their work. Or perhaps I'm wrong, and it would be quite easy, once the time and effort have been invested.
i read one paper. I have a proper criticism of it. But i'd rather look at this evidence that is so compelling people here keep quoting. Which paper is it? I'll do a proper written critique (which takes more time) only of the "best".
-
Perhaps when the effect always alters so as to be difficult to measure - that is because it is not a real effect?
How is four orders of magnitude more pits when D2O versus H2O is used difficult to measure? It sounds straightforward to measure. You can probably easily see the difference with your own eyes when looking in a microscope.i read one paper. I have a proper criticism of it. But i'd rather look at this evidence that is so compelling people here keep quoting. Which paper is it?
If you read the 2009 paper, which is the most recent I have seen, you read the paper claiming the four-order-magnitude difference in pits.I'll do a proper written critique (which takes more time) only of the "best".
No idea which one is the best. Perhaps they're all of the same quality. Who am I to say? I know little about CR-39 detection beyond what someone learns from reading Wikipedia and from reading those and similar papers. As far as I can tell there's more or less a single progression of experiments that build upon one another in the case of Mosier-Boss et al. Your call entirely. I assume the most recent paper would have incorporated feedback from previous iterations.I doubt my judgment of what's best in this series of papers would mean a whole lot. It would be like me opining on what the best paper is in a series of astronomy papers from a certain group.
Probably a good idea to get in touch with Pam Mosier-Boss directly to make sure that their current situation allows them to reply.
-
Well, if you are saying (as some do here) that these papers provide good evidence for LENR presumably you know which one is most persuasive in that respect? I don't want to spend a lot of time on one only to be told the others are more important...
-
Well, if you are saying (as some do here) that these papers provide good evidence for LENR presumably you know which one is most persuasive in that respect? I don't want to spend a lot of time on one only to be told the others are more important...
Your call, Tom. To me the work of this group taken together provides persuasive evidence of nuclear phenomena being triggered in an electrochemical cell. The 2009 paper is a representative specimen, and it has incorporated changes from previous rounds. And I get the impression that this team is competent and that their results are solid. I don't think this series of papers are "killer" papers, but of course I don't think there are any killer papers. I do get the impression that there are other groups that are doing work of a similar quality.I wouldn't put too much thought into it -- if you've got some problems written down already, perhaps get in touch with Pam Mosier-Boss and take it from there. Maybe things don't progress beyond that point. Maybe they do.
-
Eric & Thomas
And if this is an inescapable characteristic of the real world, what can or should be done, if anything?
Looking forward to the results of Thomas's discourse with Pam Mosier-Boss and the discussion of similar papers. My question would be if the phenomena could be attributed to 'background noise' i.e. radon, then what controls were there to eliminate this?
-
Quote
The 2009 paper is a representative specimen, and it has incorporated changes from previous rounds. And I get the impression that this team is competent and that their results are solid. I don't think this series of papers are "killer" papers, but of course I don't think there are any killer papers. I do get the impression that there are other groups that are doing work of a similar quality.
The three papers I had downloaded were from slightly earlier - so please give me a precise 2009 reference.
-
Eric & Thomas
See:
http://www.newenergytimes.com/…tingScienceColdFusion.pdf
refers and
http://peswiki.com/index.php/A…la_Mosier-Boss_experimentNot quite sure if this is the original paper though.
-
Frank,
This paper Below discussed and evaluated background noise (like Radon) in every detail. He proved that background noise could not explain the results of the actual experiments.
http://www.lenr-canr.org/acrobat/OrianiRAreproducib.pdf
- 43 different control Runs showed a mean track densities varying from 6 to 16 tracks / cm2. The highest recorded number was 24.
- in 25 actual experiments 15 experiments had average more than 100 tracks/cm2. One had track number so great that counting was impractical.
- typically tracks where also found in groups with much higher density than the average indicated above.
-
Frank - the two refs you give are both reviews. One by an MSc student, one a web page.
That does not help me much. Maybe the best M-B writeup is the one i've looked at (which does do a lot of comparative testing, and is complex) it is just it does not give the quantitative results I'd need to evaluate its significance. All the results are processed, and the precise processing method is not given (also it loses information, so for example the pit number is classified into "no pits" or "pits" according to some threshold which is not stated, to separate signal from background). The quantitative results are not comparative between signal and background or no signal control, but between signal and a radioactive source.
The fact that the paper does this very complex comparison of pit shape etc is telling, because if the pits were significantly above control that would not be needed, although a correct set of controls still would be needed.
I heat this 1000X greater number of pits than background. Could the people who know what this is please reference the source data?
Tom
-
I'll look at Oriani instead if you like - I'm just reluctant to give up on M-B when everyone was quoting them as important.
-
Abd:
I realise that you are a good advocate, and good at arguing on the internet. But really, facts trump rhetoric and your comments here ignore or select facts so much as to be biassed.
I'll go through, showing where we agree, where not.
QuoteClaims about experience are, in fact, evidence, and is pretty much the only kind of evidence admissible in court. Claims about interpretation are a lesser kind of evidence, generally admissible only if the person is expert on the topic.Rossi is expert on what he's doing. He may not be expert on, say, the physics of it.
His experience is running demos that apparently show excess heat but actually have consistent errors which are consistently not corrected when pointed out. If what he is doing, is those demos, then he is clearly either not an expert, or he is dishonest. The Lugano test calculation error is major, well advertised, and certainly Rossi himself, using the same setup, would either make the same error or know a very bad error was made by the testers (from reactor color alone, which he saw). Since the test he has quoted its results as definitive. So: either he is incompetent at what he does (the tests/demos) , or he is dishonest, or both.Or, if you reckon what Rossi does is to generate internet PR and accumulate funds then I'd say he is very competent at what he does, no physics competence is required, and his honesty or not is less immediately linked to his competence at his work.
This analysis does show that any self-respecting LENR advocate should not be using Rossi's claims as evidence for LENR.
QuoteAs well, evidence is not the same thing as "proof." Evidence can be impeached, but to deny that evidence is evidence, I only see among debunkers and pseudoskeptics.It is not found among genuine skeptics.
This is disingenuous.
The issue is the strength of the evidence, and its bearing on the hypothesis "LENR exists". Given the context, weak evidence (in the normal meaning of the word) will not alter rational probability estimation for "LENR exists" and therefore is the same as no evidence. The word "proof" may be used popularly to indicate evidence strong enough to move rational estimation of the probability of the LENR hypothesis. But many people, such as me or Joshua Cude, who you might reckon are debunkers, would want evidence strong enough to be relevant to the question and know that in science "proof" is strictly never possible.
Why is weak evidence the same as no evidence in this case? It is because all experiments are potentially subject to systematic and interpretation bias, that ones supporting a given case can always be selected when the evidence for is weak, as here. Therefore given a hypothesis H, and a group of people selecting experiments that support H, you will expect to find a lot of weak evidence (even reproducible weak evidence when there are subtle systematic errors).
This only applies when the experimental data is all either marginal, or non-reproducible, and that fits LENR evidence.
It means what you would normally do, which is to look for a whole load of independent weak evidence and say that together it is stronger, is logically incorrect.
QuoteThis is supremely ironic. An anonymous internet troll is making an argument based on the incredibility of a witness who is a real person, with a real identity and much at stake.Personally, I don't consider Rossi reliable, but what he says is what he says, and if he says, over and over, that he's sitting in a shipping container at a customer installation of a test system, he *probably* is. But until it actually makes a difference, I'm not going to form definitive conclusions.
See above. No definite conclusions are needed, however evaluating his [Rossi's] evidence - the claims - as supporting LENR (which I believe implicitly or explicitly your stance does) is incorrect. In fact, the proper thing to do would be simply to say that thus far there is no evidence from Rossi that bears on the issue of whether LENR exists or not. That is not "forming definitive conclusions" and I hope you can say that, and be consistent in doing so.
Quote
Again and again, we are told that Rossi could easily arrange a definitive test. The fact is that those familiar with cold fusion have been saying this since 2011. Rossi obviously hasn't wanted one, and Rossi's "rules" probably made it impossible for Lugano to be at all definitive.
Since here you are analysing the Rossi information, it would be better for you to do it correctly. The testers in the Lugano experiment made a basic error that turned COP = 1 (to within experimental error) into COP = 3, and generated an apparent convincing "acceleration" in COP with temperature. That is fact. The conditions that allowed this error were mandated by Rossi. Rossi was also present for a week during the test, at start and end. He must have been aware of the reactor color (indicating much lower temperature than stated by the testers). Since the calorimetry method used was unusual, and clearly followed by Rossi, it would be highly surprising if the testers had not compared notes with Rossi on the way they calculated temperature. Even if they had not, when he got the report he must have known the temperatures were not as stated, or else he must have consistently made the same mistake himself during internal tests - he has a TC measuring temperature directly (used for the black box control) so this indicates an extreme disconnect between claims and reality.This paints a picture of more than secrecy. It shows that Rossi must have a strong predilection for experimental errors that inflate COP (whether this is due to dishonesty is not something that needs to be elaborated - there are many lone inventors who are honestly deluded about the capability of their inventions). The testers believed, after two sets of tests, that the Lugano wrong results were definitive. I don't see "rules" inflating the COP since they could have done the correct calculation from the same data and produced a correct null report.
