Tuesday, May 21, 2013

Tommaso Dorigo impressed by a cold fusion paper

...but the paper is 100% crackpottery...

In his text "Is Cold Fusion For Real?", Tommaso Dorigo seems highly impressed by the following new Italian-Swedish preprint about cold fusion:
Indication of anomalous heat energy production in a reactor device
They claim that an Andrea-Rossi-style tube with nickel and hydrogen produced 10+ times more energy per liter of fuel than any known chemical reaction, as measured by thermal imaging cameras during 96- and 116-hour experimental runs.

Image credit: Rossi, Kullander, Essén and the e-Cat, retrieved from energydigital.com.

Dorigo says that "the conclusions of the tests are at the very least startling". He "continue[s] to believe in the scam hypothesis, but [he] must admit that this study impressed [him] for its reported result." Also, he must say that "[he] will from now on follow more closely the developing story of Rossi's E-CAT...".

Congratulations, Mr Rossi. You have clearly earned a fan who thinks that cold fusion is more realistic than supersymmetry! If you really believe so, Mr Dorigo, then you are an unhinged lunatic.

First, let's discuss the sociology. What I find remarkable is how easy it is for the authors to make folks like Dorigo repeat some self-congratulatory words such as the "third-party [investigation]" that appears in the first sentence of Dorigo's article.

If you look for a rather detailed, enlightened, entertaining, and completely reasonable criticism of the new paper (with discussions of mistakes, possible ways to do such an experiment to be convincing, and differences between science and pseudoscience, not to mention some useful links), see The E-Cat is back, and people are still falling for it! by Ethan Siegel (Starts With a Bang).

If you look at the previous publication record of the seven authors via the arXiv (click at the author names here), you will see that only two of them have previously published something. Bo Höistad appeared among dozens of authors in an (unknown) technical design report on a tracker in the KLOE-2 experiment. Hanno Essén of KTH Mechanics, Stockholm has published a dozen or two dozens of preprints on magnetism and fluid dynamics, none of which is really well-known and all of which look like papers by an engineer attempting to "do" physics as his hobby – which is what they are, after all.

Given the large amount of money that is circulating in the cold fusion business today, one must say that all the data are consistent with the seven authors' being inserted just as puppets who don't have to be afraid of their scientific credibility because they don't have much to lose (relatively to what they can be promised from the Rossi industry). So I have a problem with the adjective "third-party" in Dorigo's article. There doesn't seem to be solid evidence that this is independent from Rossi's organizations; the presence of the Italian co-authors makes the independence even less likely.

Update: Hours after this blog entry was written, Steven Krivit explained that this paper doesn't describe any independent test; instead, the authors were just witnesses of a Rossi demonstration. Essén, the only co-author with at least 2 other papers, admitted he didn't replicate the experiment. On the picture at the top of this blog entry, he is shown together with Rossi, in a position clearly indicating that Rossi is a teacher and Essén is an obedient disciple.

Dorigo himself may be rather positive because all the Italians form a family of a sort. A trailer for Mafia II, an amazingly realistic but too linear PC game.

A technical or sociological detail that doesn't prove that the preprint is rubbish but it's always a brightly shining and blinking "red light" for me is that the physics.gen-ph preprint was delivered as PDF only and it wasn't written in \(\rm \TeX\) but rather in Microsoft Word (on a Mac). That makes it very likely that the authors don't actually know \(\rm \TeX\) and most of such authors don't really know physics well, either.

Tommaso Dorigo often likes to paint himself as a person who is willing to think independently except that the extent to which he simply copied the self-congratulatory decoration that was added to this paper suggests that he isn't really thinking critically.

Now, somewhat more technically.

The authors are clearly ignorant not just about \(\rm \TeX\); they seem to be unfamiliar with the basics of the scientific notation and the rudimentary methodologies that are needed in experiments. For example, the units of dimensionful quantities are always written as \([\rm W]\) including the square brackets. No, if we perform dimensional analysis and want to say something about the units of a quantity, we may use the symbol \([l]\) or \([{\rm length}]\) to describe the units of length. But we never actually write meters – the particular units themselves – into the square brackets! This is a detail but a strong indication that the authors haven't really gone through some relevant undergraduate freshman courses of physics.

Perhaps more importantly, the paper never discusses the error margins properly. You may easily check that they never talk about anything such a "systematic error" and when they talk about errors at all, they're just "optional". Sometimes they add them to the results, sometimes they don't, and if they do add them, it seems that they just make the numbers up. For example, on pages 22-23, they say that the effective power consumption has error of 10% because "errors of this extent are commonly accepted in calorimetric measurements". Wow. There's surely no "one-size-fits-all" error for calorimetric measurements. The error depends on what you measure, how you measure it, and almost all the details in your experiment. The magnitude of a systematic error of your apparatus actually has to be measured and calculated for your case specifically; you can't copy a "general figure" from completely different papers describing different apparatuses. This sentence by itself shows that they don't know how to do experiments well.

But on page 22, you see an example of the "key calculation" meant to show that the "reactor" produces lots of energy. Over the 116 hours of the experiment discussed here, the gadget was consuming 283.5 watts from the grid and producing 810 watts (much more), we hear. An anomalous production of heat, they happily announce. Those 810 watts are claimed to be determined from the thermal radiation that the reactor was emitting; the 283.5 watts are calculated as 35% of 810 watts.

Where does the figure 35% come from? The resistors (electric heaters of a sort) were on for "about" 35% of the time and off for "about" 65% of the time. Couldn't one just measure the precise time during which something was turned on and off? Are we supposed to think that they used a gadget to measure time that uses "one percent" as the unit of time? Don't you agree that this claim about "65% off" is just a potential lie to mask that the energy was coming in all the time? See also this comment for a convincing indication that they measured the incoming energy completely incorrectly (even though this should be a very easy task!).

OK, let's not think about details. It's plausible that the average consumption in those 116 hours was close to 283.5 watts. The produced average power 816 watts is much higher and it would be enough for a proof that some very strong anomalous production of energy is taking place. Where does the figure 816 watts come from?

It comes from equation (24); the power was calculated from the emitted thermal radiation. The numerical value 816 watts whose error is estimated as just 2% is the sum 741.3+17+58 watts. Clearly, the only large contribution that affects the overall qualitative result is the term 741.3 watts. Where does this come from?

It comes from Table 8 on page 21 where it's calculated as the sum of radiated 459.8 watts and 281.5 watts participating in convection. The latter figure, 281.5 watts, contradicts some calculation resulting in equation (17) on page 12 where convection was claimed to give about 466 watts. I was trying to check where various numbers were coming from but almost nothing seems to be consistent with anything else. The paper looks like an incoherent pile of rubbish to me.

Moreover, the gadget seems to depend on the power outlet and has a nonzero consumption. So you could expect that its power production will also depend on whether or not the resistor coils are turned on or turned off. However, this dependence of the produced power on time isn't discussed anywhere in the paper.

Let's ignore the differences between 281.5 and 466 watts. They still see lots of radiating energy that is produced, don't they? Well, they surely claim so. In equation (8) on page 10, they boast to have produced 1568 watts (during the other, 96-hour experiment) which is 1609 watts minus the small contribution from the room. The numerical value 1609 watts is computed in equation (5) and nearby equations as the Stefan-Boltzmann power corresponding to the (fourth-power-based) mean temperature around 723 kelvins (450 Celsius degrees) multiplied by the area of the Rossi cylindrical "black box" (whose bases are overlooked).

The emissivity is set to one i.e. they assume the "reactor" to be a black body. This choice is labeled "conservative". Except that the truth seems to be going exactly in the opposite direction. The actual emissivity is lower than one and it's the coefficient multiplying the fourth power of the absolute temperature to get the power. Because they seem to calculate the power from the measured temperature (the infrared camera is claimed to give the right temperature and automatically adjust the observed radiation for emissivity etc.; see page 7 of the paper), the actual power is actually much lower than [the calculated figure] 1609 watts. The emissivity of metals at similar reasonable temperatures seems to be 0.2 or so – something of this order – which reduces 1609 watts to something like 300 watts, pretty much equal to the consumption.

Pretty much every hint that I have looked at in the paper suggesting that they produce some energy that exceeds the electricity consumption from the power outlet seems to be plagued by similarly basic errors.

Recall that I believe that the error in virtually all Rossi's presentations is that he assumes that the boiling point of water is always 100 °C, regardless of the pressure, and he "takes credit" for the evaporation of lots of water that actually stays mostly liquid because it's below the boiling point which is above 100 °C if the pressure is elevated (and it is elevated in his setup); even visually, it's clear that liters of water can't be getting vaporized because the steam would look much more intense than the feeble traces of vapor coming from his "miracle gadget". All these errors seem completely elementary to me. It's hard to say whether the authors see them or not. I guess that they're training themselves to overlook them because they are afraid that by admitting they have believed and promoted something that so silly that clearly doesn't work, they would look like complete idiots – and indeed, they look like complete idiots even if they try to obscure the evidence.

To summarize, the preprint is complete rubbish and the authors are probably linked to Andrea Rossi personally but that doesn't prevent the loudest blogger of the LHC's CMS Collaboration to partially endorse this preprint – without even attempting to read it because "this is not [his] field of research" – and suggest even though he hasn't looked at this paper at least to see that it's pure trash (and it's very easy to see), he will more closely follow cold fusion because of that. It's so easy to propagate lies and stupidity in this world especially because most people are even more stupid, mindless sheep than Tommaso Dorigo.

At any rate, I am amazed by Dorigo's claim "not to be an expert" itself. He is an experimental subnuclear physicist and this is a claimed groundbreaking paper in experimental nuclear physics. I have not been an experimenter at all but I see nothing in the paper that I could be misunderstanding because of an insufficient background. You see that even people claiming to be "scientists" often don't behave as scientists. Without even trying to study something, they just uncritically endorse some ambitious claims. And in many cases, their "being a scientist" is exploited in the promotion of a nonsense even though they clearly failed to evaluate the issue scientifically. Note that it's enough to find a few dozens of such "scientists" who haven't performed even the basic checks and the media often claim a "scientific consensus" even though the strength of the scientific evidence behind the claim is exactly zero.

With this extremely sloppy attitude, you can't be surprised that Mr Dorigo and others have no problem to deny string theory or other basic pillars of modern science. This particular chap denies completely basic insights into nuclear physics as well – and never hesitates to use a paper not knowing how numbers with units are written as evidence that nuclear physics fails.
See also a dozen of previous articles on this blog that mention Rossi and fusion.
Concerning somewhat more realistic sources of energy, see this new wind turbine with funnels (via Joseph S.). They claim to concentrate the wind so that 1 kWh only costs $0.20 or so – almost competitive with coal etc. at the Czech prices. The gadget looks silly but I do think that the concentration of wind energy (and similarly, possibly, concentration of solar energy with mirrors) is a largely unused way to make these "renewable" sources cheaper and more productive.


  1. Well, Mr Dorrigo is careful to cover his arse by saying that he still believes in the scam hypothesis but this is not good enough for our host who continues to insist that anybody stupid enough to report on such matters at all should be the subject of ridicule.

    Is our host quibbling over minor details? I thought the description of the November test was quite interesting:

    "The performance of this device was such that the
    reactor was destroyed, melting the internal steel cylinder and the surrounding ceramic layers."

    Sound pretty hot to me, if true. The ceramic layers are described as consisting of silicon nitride (MP 1900C) and corundum (MP 2000C).

  2. I think so, too, Casper. He is just covering his buttocks because he sort of feels that physicists around him could humiliate him but if he thought what he really thinks, he would say that it's an almost proof of cold fusion because they say so.

    Well, with enough electricity sent in, one may surely destroy a "reactor".

  3. I personally wouldn't dismiss something just because it wasn't written in LaTex; also I believe string theory is no more than science fiction that has already been disproven by the discovery of the Higgs.
    In any case, this experiment took place outside of Rossi's headquarters, so if the E-Cat took power from external sources, it would be extremely easy to detect, more so in the next experiment that's going to last 6 months.
    That doesn't mean I don't share your concerns, but I think this report is a great first step that merits further research into this criminally neglected field of science.

  4. I doubt the cold fusion, if nothing else because these claims have been around for quite some time, always resurface during an energy crisis, and never produce anything useful.
    What I would take issue with in this post is the last paragraph. String theory as a pillar of modern science? You are kidding right? One of the problems of modern science, particularly physics, is that it has become dominated by a "consensus" of whomever shouts louder, or gets more academic appointments in a wave of latest fashion. This is how you can see big bang theory as a pillar of modern science.
    There is a real problem with modern day science, and cold fusion is one of the manifestations. Global wartming is another.

  5. I don’t get it. Lubos. Andrea Rossi is a fraud and Dorrigo is incompetent. Those facts are well established. Why say more?

  6. The alternative theory: Tommaso was feeling bored, so he dangles a big juicy steak in front of the Luboscage and the cat jumps out roaring. It's Punch and Judy, Lemmon and Matthau, now with extra science content thrown in. Entertaining as heck, plus you learn something while munching popcorn in your seat.

  7. Thanks, Luboš, for the insight! Please excuse my ignorance, but could you explain in simple terms (I am no scientist) why their assumption of bigger emission area (they assume the full square area of the reactor to face the camera instead of calculating in the cylindrical shape), and higher emissivity (they assume black body emissivity instead of the 0.2 emissivity of steel) is not conservative (as they claim), and would result in much higher energy emission (as you tell) instead of the exact opposite as the common sense would tell? I admit, my common sense may be broken, but I would tell that when I measure certain heat flow by the camera, and assume it comes from a bigger area than it does in reality, and with higher emissivity than it really does, I get in fact very much underestimated values than the contrary. Am I really wrong, and if so then why?

  8. Couldn't one just measure the precise time during which something was turned on and off?

    Maybe because the time wasn't exactly the same every time the power switched on and off. The turns were automatic, there was a control box that switched the power by itself during the 119 hours.

    Since you have a PhD and you wrote so many papers, your poor brain should be able to elaborate a simple concept like this, or is it too difficult compared to universe's physics?

    I see, you didn't figure out the issue because you were distracted by the font of the paper (not LaTex standard). I have a suggestion for you and your future career: look to the wood and not to the trees.

  9. Back when the Fleischmann–Pons cold fusion claims were in the news there were several attempts to replicate the experiment, including even by some skeptical physicists. You could tell which were the physicists which were not even from a distance according to one report I remember. The non-physicists were all crowded around their apparatus like in the picture above, while the physicists were behind lead shields.

  10. sure,
    they want to proove the most revolutionary invention and they can't measure time because the time isn't the same every time.
    less bullshits, more science, please.

  11. At that part of the calculation, emissivity = 1 is actually conservative because the detector converts radiative power to temperature, they illustrate this on pg. 7.

    Later on they then convert the reported temperatures back to the overall power, I have not read the details there yet, but it looks like they determine an emissivity of ~0.80 based on some blackbody control samples attached to the cylinder.

  12. No, Arun, I think that you are confused.

    They may be saying how they would be calculating the temperature - incorrectly, assuming epsilon=1 - but they explicitly say on page 7 that the infrared camera gadget is actually measuring (plus calculating) the temperature correctly, using its own method to determine epsilon.

    Unlike the cranks, the camera does these things correctly, i.e. assuming (the indirectly determined) epsilon which is of order epsilon=0.1-0.2 for unoxidized metals. This camera directly gives them the temperature and they use it to calculate the power using the completely wrong epsilon=1. They never do the opposite by themselves as a part of the calculation.

    If the metal is mostly unoxidized, epsilon is f order 0.1-0.2. It's only closer to one if these metals are oxidized. Note that emissivity is equal to the absorptivity by Kirchhoff's law.

  13. you are probably right but after reading this:

    "Dorigo himself may be rather positive because all the Italians form a family of a sort. A trailer for Mafia II, an amazingly realistic but too linear PC game."

    let me say that your post doesn't deserve to be read :)

  14. I guess you are not a specialist in thermal imaging, but there's no such a thing like emissivity auto calibration. You can't just point a thermocamera at one thing and know its emissivity. You need to measure it separately and then use that value for the temperature measurement. What they did in the experiment should be corrent.

    I have a lot other reasons to think that the paper they have written is totally crap, but the emissivity part is not the problem.

  15. The temperature which the IR camera assigns to the two areas is 564.1 °C and 511.7 °C, respectively –these values being much higher than those of the adjacent areas.

    If the unadjusted imaging camera assigns much higher temperatures to the spots with e=0.8 and 0.95 than the rest, it means that the rest is emitting much less radiation. That's why it's producing a lower apparent temperature. If it is emitting less radiation, it means that the emissivity is lower than 0.8 and 0.95, respectively.

    And indeed, the emissivity may be substantially lower than 0.8. Emissivity of nickel starts at 0.04 or 0.05


    When the sigma*T^4, assuming that the T is right, is multiplied by the right emissivity of the bulk which is much smaller than one, one gets radiated power equal to those 300 watts or so that is going to the device from the power outlet.

  16. It's not a joke.
    Maybe it's only because I'm Italian, but... fuck you.
    If you wanted to be somehow friendly, well, you did it wrong.

  17. The point of that exercise on pg. 7 is that we can assume that the true amount of radiation being emitted from the adjacent regions 17, 18, 19, 20, 21 is approximately the same.

    Given that they are the same, it makes sense that the detector reports a higher temperature in regions 18 and 20 when we tell the detector to assume a lower emissivity in 18 and 20, because it calculates temperature from power. (For constant Power: T^4 is inversely proportional to emissivity.)

    When it comes to the later calculation of Power from Temperature, I understand your point about the emissivity of bare metals, but these tested cylinders are (a) black-painted metals and (b) the e=0.80 was determined by a calibration against 2cm white disks with known emissivity of 0.95, which as the document says, come from the same manufacturer as the IR camera. (picture of them on pg. 21).

  18. I think Lumo you are wrong on this issue of epsilon. The camera doesn't know about temperatures but can measure power. If you use a higher epsilon (1 being the highest) than the real one you are actually underestimating the temperature (derive from Stephan-Boltzman). The camera gives temperature as a proxy for power. If you use the wrong epsilon in the setting of the camera, let's say 1 instead of 0.1 you are underestimating the temperature by a factor of 10, so 5000 K is reported as 500 K. Then when you use the reading of 500 K to calculate the power using Stephan-Boltzman again (after averaging over many areas) reintroducing the same value for epsilon=1 would overestimate power but because the temperature was underestimated by the same factor, everything is all right and the radiation power is estimated correctly. It is still a lower limit of total power given that some power would be in other forms (like convection).

  19. Marcel van VelzenMay 21, 2013, 11:41:00 PM

    Although it sounds logical at first, it is actually the other way
    around. The physicists should be all crowded around the "reactor device" like
    in the
    picture above, while the non-physicists should be standing behind lead

  20. Thanks for the reply, Luboš! I am still rather confused, though.
    Perhaps I misunderstand the definition of black body (it is a few decades since
    my last physics courses), but I understand that, at a black body, 100% of
    energy produced is radiated by its surface, while at a gray body it is just a
    fraction of the produced energy that leaves the surface by radiation. I
    also understand that Stefan-Bolzmann's Law defines the radiated energy
    (exitance) and not the produced one. And since I guess we are trying to find
    out the produced energy and not the radiated one, assuming a black body where
    produced E = radiated E seems to be the most conservative assumption we can do
    in this case. In other words, by counting with ε = 1, they
    assumed all the produced energy was radiated by the surface, while if they took
    ε < 1, they would assume the actually
    produced energy was even bigger than the one measured (which is
    definitely less conservative, though true). Am I wrong here?

    But anyway, when they speak about ε = 1, they speak about adjusting the camera software, getting so the lowest possible temperature values. If they used any value lower than 1, the software would show higher temperatures, and it would result in higher power emissions, isn't it that?

    And lastly - they measured also an identical reactor without a charge, and did not find any excess heat. If the ε was at the root of the excess heat, wouldn't they meaqsure it at the dummy reactor too?

  21. I think Lumo you are wrong on this issue of epsilon. The camera doesn't know about temperatures but can measure power. If you use a higher epsilon (1 being the highest) than the real one you are actually underestimating the temperature (derived from Stephan-Boltzman). The camera gives temperature as a proxy for power. If you use the wrong epsilon in the setting of the camera, let's say 1 instead of 0.1 you are underestimating the temperature by a factor of 10, so 5000 K is reported as 500 K. Then when you use the reading of 500 K to calculate the power using Stephan-Boltzman again (after averaging over many areas) reintroducing the same value for epsilon=1 would overestimate power but because the temperature was underestimated by the same factor, everything is all right and the radiation power is estimated correctly. It is still a lower limit of total power given that some power would be in other forms (like convection).

  22. Deleting my comments?

  23. It is patently obvious that you have NOT read the paper, or only skimmed it due to your *belief* that this is a scam.

    1) you state, "Emissivity of nickel starts at 0.04 or 0.05 and even black nickel has epsilon below 0.5."

    The emissivity of Nickel has nothing to do with it. The outer cylinder is steel, not Nickel. So why even mention the emissivity of nickel here? You are either ignorant of the details of the test, or are intentionally misleading people.
    2) In addition, the steel cylinder is PAINTED, as was CLEARLY stated in the paper on pg16:

    "Another critical issue of the December test that was dealt with in this trial is the evaluation of the
    emissivity of the E-Cat HT2’s coat of paint. For this purpose, self-adhesive samples were used: white disks of approximately 2 cm in diameter (henceforth: dots) having a known emissivity of 0.95, provided by the same firm that manufactures the IR cameras..."

    These disks are used as CONTROLS to help validate the emissivity values used. I would think that a scientist would at least read the paper CAREFULLY before attempting to criticize it.

  24. Sorry, but In my view any reference to Rossi that does not explicitly cite his work as fraudulent can be dismissed out of hand.
    Perhaps some purpose is served, however, by your careful analysis of this additional piece of crap. It does make interesting reading and I agree with it 100%, as usual.

  25. Tommaso is a jerk; that’s all.

  26. Bernie KoppenhoferMay 22, 2013, 3:02:00 AM

    I stopped reading when you called a respected colleague "you are an unhinged lunatic".

  27. Yes, this could be a hoax. But if so, it is because the actual input power was not correctly measure.

    Sorry Lubos, but trying to discredit on the basis of emissivity is silly. The device obviously has a matt finish and is not made of a low emissivity metal like tungsten. Also, you are incorrect about the emissivity of non-oxidized metals having emissivity < 0.1. That is only the case for *polished* metals. Brushed stainless steel, for example, has emissivity of 0.85. Oxidized ("black") nickel has emissivity of 0.59-0.86.

    In order to see >800C infrared *plus* white-hot visible light sustained from this object (with surface area of ~0.35 sq. *meters* !) at 300W input, would require epsilon <0.01.

    You might be right this is a hoax. But your emissivity analysis is ridiculous.

  28. Aside from the calorimetry issues,
    If the reaction is p + 58Ni,
    then should one not observe emission of [1.301 MeV] γ rays?

  29. No, gsantosi, that's a collection of wrong claims. When the emissivity is 0.1, it doesn't change the temperature by the factor of 10. It changes the temperature by the factor of 10^{1/4}, the fourth root of ten, which is "mere" 1.78.

    At any rate, these folks played with the imagining device in an incorrect way and assigning wrong emissivity to the bulk of the "miracle gadget" even though they could have simply measured it.

  30. There exists no evidence that they performed any step of this sort correctly.

  31. Friendly to whom? No, I didn't have any intent to be friendly to Rossi, the authors, and their mindless fans, so as you can see, nothing I did was wrong.

  32. In Tommaso's case, this is confirmation bias. Deep down, he believes in it; an Italian is promoting it, and here is a paper claiming it is real. A few big names were drawn in several years ago, including Julian Schwinger. I haven't looked into it, but on the surface don't believe it, and Rossi doesn't pass the smell test.

  33. Sure. It is ludicrous to think that the people around the "reactor" would be healthy if nuclear reactions were really going on.

  34. Ok, you are right about the 10^(1/4) (it was not my main point anyway, but I was careless there) but the rest of my claim is right, that the overestimation in temperature is compensated by the understimation in power. So it shows that you are analysis of the source of the paper "error" is incorrect.

  35. I agree with your emissivity figures etc. If they couldn't measure even the incoming electric energy, that would be really, really bad. Couldn't they just ask their electric utility how much they pay bills? Every other employee can measure it in the electric utility companies. You know, they have these electric meters for that.

    The temperature 800 C measured - probably at places with a high emissivity - wasn't the temperature of the whole surface of the device. The electricity was clearly used non-uniformly and the bulk of the surface had temperatures much lower than 800 Celsius degrees. There is no contradiction.

  36. And they actually measured emissivity using a factory calibration dot in the second part of the paper.

  37. Right, Gordon. I think that there's a lot of this hypocrisy everywhere. Many people are thinking so many right things and so many wrong things but they're creating a completely different impression in order to improve their situation in the society. In some cases, this adjustment makes them sound more right than what they are; in others, it makes them sound more wrong than what they are.

  38. It doesn't matter for the big picture whether it's nickel or steel which is why I talked about "metals" in the main text. Polished steel (sheet) starts at 0.07 as well, see


    If someone writes lots of words such as "controls", it doesn't mean that he is doing things right. They are not.

  39. It is not the emissivity of nickel but that of a black painted ceramic material that is relevant. Lubos, it is clear that epsilon doesn't matter if you keep it at a consistent value when you calculate power from the temperature readings. It goes like this T_measured=(P_emitted*epsilon1)^1/4, P_estimated=epsilon_material*T_measured^4=epsilon_material*P_emitted/epsilon1=P_measured if epsilon_material=epsilon1. You should concede.

  40. These are just words. Bring me the gadget and I will measure the incoming and outgoing energy flows correctly for you. You - and these authors - say lots and lots of semicoherent things but I see no reason to think that they did the things correctly.

  41. I agree with you and you are obviously right, don't be confused. If you understimate the temperature setting the camera at high emissivity, considering that the temperature is at the forth square in the formula, you obviously understimate the power, even if emissivity is part of the formula. Can't stand such arrogance sometimes.

  42. I am closing this thread because I got bombed by visits by tons of Italian crackpot fans (this blog entry was linked to at several powerful enough Italian servers) who would be wasting too much time with their mostly idiotic comments.

  43. The paper is total BS. As an italian, I am deeply ashamed of this ridiculous approach to science, and can find no excuses for italian idiots who support it "because he's italian".

    There are real scientists who can do real science in Italy, and charlatans like Andrea Rossi destroy their reputation. This crook is the reason we have to read about "mafia" in a scientific post: it's his fault - and of the many like him - that we can't get rid of this humiliating tag.

    Please believe that Italy's real scientific community conforms to much higher standards.

  44. Dear Prof. Motl, let me tell you that your statement that "all the Italians form a family of a sort" which I presume has to be referred to the Italian scientific community is light-years away from reality. In fact, apart from the very few people who do research on LENR (and not even all of them) I bet that at least 90% of Italian physicists are 1)skeptic about Rossi; 2)would be extremely pissed off if Rossi is right.
    Btw, given the offensive nature of your post (and don't tell me the BS that your post is not offensive to Italian scientists), I would like to remind you that there were at least 5 Italian nobel prize laureates in Physics vs. a much lower number from your country. Perhaps you will be a nobel prize in the future, though I doubt so.

  45. I've been working with thermal cameras for 12 years Mr Motl (software engineer).

    My eyebrows were raised by this statement, "As it was not possible to measure the emissivity of the coating used in this analysis, it was decided to conservatively assume a value of ε = 1"

    The phrase "conservative" is a complete misnomer here.

    They also make no mention of having an actual black body in the scene in order to verify the accuracy of the camera's calibration. This is also extremely problematic as it takes the accuracy of the calibration for granted.

    An interesting undergraduate science project though.

  46. Also would like to take issue with your statement that String Theory is a "pillar of modern science". Until it makes new predictions that are verified by experiment, it cannot be considered a pillar!

  47. They don't say if the camera itself had high temperature calibration tables loaded. Most thermal cameras will measure, say -20C to 150C accurately with one set, but need a second set to measure 150+. The calibration curves across the whole range aren't accurate enough for normal use if you try to generate them across the entire range.

  48. Um... wait. Did you guys even check the Rossi's background? The guy is a professional swindler -- check the Wikipedia article on him. I am shocked that anyone would consider him seriously, based on his history. Why do we even have an article here?

  49. No, Dr Rasenna, I didn't refer to the Italian scientific community. I referred to the Italian nation - (mostly) men who just want to influence things - and by the special "family" status, I meant the cohesive forces that have created the Italian mafia, as the video should have clarified. (Watch the video. It says, for example, that "this family will be more important for you than your own family" etc.)

    A huge percentage of the fanatical supporters of Rossi et al. are Italians so the evidence is rather strong that the nationalism and the desire to be unified has something to do with the support.

  50. Thanks a lot for your comments, especially because people with your experience with this particular technology are rather rare.

  51. The camera reads the right flux but has no way to know the true temp unless it knows the avg emmisivity from 7.5-13um. If you tell it the emmissivity is low then it will report a much higher temp because you are telling it, hey dummy camera that flux you read is really corresponds to something much hotter. In the end if you assume it's 1 and it is really lower then you've underestimated the temp.

    The camera is calibrated for an epsilon of 1 at the factory.
    Additionally it is only the last surface that matters not the bulk. The last surface is a high emmisivity paint clearly stated. So it's probably a good estimate to use 1 as well.

    I'm sorry but your readers are right you are incorrect about the camera being used incorrectly. Last but not least the testers calibrated their predictions with boiling water to dummy check it. This was revealed in the interviews.

    It's ok to be wrong it's not ok to be so confident that you are right.

    Unfortunately this shows us that you are not trustworthy.

  52. Your analysis of the camera is way wrong, the testers were correct. If this is a fraud it's most certainly not an obvious fraud.

  53. Today calibration over wide temps is pretty common for ir cameras. And they did say that it was calibrated with the manufacturer. Oh and did I mention they cross checked with a thermocouple.

  54. Your right, and this moron Lubos is wrong and pretty soon he's going to feel really dumb when he realizes it

  55. Dear Ivo, there may be other ways to release the energy but the radiated energy was overestimated because for a known temperature, the radiated energy is proportional to epsilon and they overstated the value of epsilon of the bulk of the surface of the "reactor".

    Concerning the dummy reactor, I just think that it's a lie or a result of different resistance i.e. different energy input or whatever that was (deliberately) not mentioned. At any rate, it is a scam.

  56. It's all about money.

    1 - Mafia. Italian mafia is really a small thing compared to japanese, russian, chinese, etc. USA have instituzionalized its mafia callnig it lobbyism. The same is going on in Europe. Where is money, there is robbery and fraud. This is Marx.

    2 - Fascism. Fascism is the other face of the capitalistic medal. One face is democracy. When a capitalistic crisis arrives, bourgeoise switch to fascism, like in stalinism, nazism, etc. Italian fascism was also a little thing, because Italy is a little actor in the capital's arena. As a czech, with an interesting past in the horror of state capitalism (not communism), you should have studied critics of political economy. Italy is a little player in comparison.

    3 - Rossi. When he discovered, in the 80s, a supposed method to extract oil from exhausted fluids, italian government abruptly changed waste disposal laws, and he became a fraudster overnight. They spoiled him of its money, and sent him in prison. I don't know if Ecat is a scam, but his secretive behavior is reasonable to me. Is extremely improbable he'll succeed in obtaining international patents, exactly due to the economic interests I explained above.

    4 - Scams. Deceiving made by MIT to obfuscate early "cold fusion" work is well documented. Also big science publications have a record of countless errors and frauds on their side, likely to protect specific economic intersts.

    5 - Physics. Standard model is, in my opinion, full of inconsistencies and errors, undisclosed by the (economic) establishment of academic researchers and professionals. I find very intersting the critic carried on by Guglinsky on Ross's blog: http://www.journal-of-nuclear-physics.com/?p=802. Search for his paper and his comments in the pages of the post.

    6 - The system. Obviously, all this inconclusive talks about cold fusion would be wiped out if we had a production system more advanced than capitalism, because all science and tecnologies would be transparent and open source for the entire mankind.

  57. Your comment is mostly composed of Marxist lies and delusions - after all, you seem to be openly defending Marx.

    1) There may be other mafias today but Mafia as organization focusing on protection racketeering was still invented in Sicily, in mid 19th century, and nowhere else.


    Mafia has nothing to do with lobbying - the former is harmful to innocent members of the society,the latter is mostly beneficial.

    2) Fascism isn't another mode of capitalism. It has nothing to do with capitalism. Fascism is a branch of socialism. It was pioneered by Mussollini who had previously been a member of the socialist party, thought that it wasn't good enough, so he created a socialism on steroids. It was adopted by many people for the same reasons as socialism - it was meant to unify a nation regardless of the class, see


    3) Rossi deserved the punishments for his frauds in the 1980s and more than he received. He should be already arrested for what he has already done economically in the context of his fusion scam and he hasn't been partly because criminals like himself partly control the Italian judicial system.

    4) Cold fusion isn't bullshit because of anything that MIT says. Cold fusion is bullshit because it contradicts everything we have known about atoms and nuclei for more than 50 years.

    5) The Standard Model is consistent as an effective field theory - a theory of nearly everything, Lisa Randall says - and your pet authors are cranks.

    6) Capitalism isn't perfect but it's the most perfect system one can have. After all, the imperfections play a very important role in the long run and in various situations, too.

  58. I never said that I believe to Ecat. Decades ago (Bordiga), was scientifically established that your country regime was a state capitalism: money, banks, market, wage work are all capitalistic factors. Stalin, a state capitalist head, killed ALL the bolscheviks. Mussolini and Hitler seized the power with the help of worldwide bankers. But, from your answers now I'm sure you NEVER studied or understood history and philosophy, so it is useless to insist. Really crazy and DEEPLY ignorant guy..

  59. Lubos, excuse me for the insults. I'm stupid and I'm sorry. excuse me. Only I cannot accept you can't recognize that the worst horrors of the world were made in the name of the capital and imperialism.

  60. OK, enough about the emissivity... I design thermal imaging cameras for a living and the assumptions and reasoning on page no 7 are in fact absolutely correct for setting up the most conservative measurement possible. Setting the cameras emissivity setting to 1 yields the lowest possible calculated temperatures by the camera. Assuming the object emissivity is less than 1, which *you* say yourself it *must* be, even as low as .1-.2, wasn't it, would result in a *higher* calculated temperature and hence even more energy emitted from the device, which actually helps their case even more!

    The one thing they should have done was to have a known emissivity radiator to check the camera calibration with, but now they used a themocouple which isn't optimal, but give a pretty damn good indication if the camera is reasonably acurately calibrated or not. Which it was, which is what could be expected since the caamera is operated in almost ideal ambient temperatures.

  61. This is what I have always thought. We will know this stuff works when the scientists testing it are all dead... or at least pining for the fjords as Monty Python would have it. But my training was many decades ago and I went into other non-scientific fields so it is nice to know something I remember is still correct.

  62. Brian G ValentineMay 30, 2013, 6:29:00 AM

    Having no luck finding the possibility of an inexhaustible supply of energy by fusion in a condensed phase, and not much luck finding an AIDS vaccine, some people went on to demonstrate that the human race will be extinguished anyway because of ACO2 in the air ...

    By the way I don't think crack pottery today is any worse than it was in the 19th century. I'll bet that about the same percentage of science in the 19th century was crack pot as it is today.

  63. Lubos, I don't disagree with your opinions here, but: you have to use "all" instead of "physics" in the arxiv searches. If you do you find more papers by that author. If you use Inspire you get 165 hits.

  64. Dear Rien, thanks - which author do you exactly talk about?

  65. Sorry, I meant Hoistad. He's an experimental nuclear physicist.

  66. Dear Lubos,

    I was reading with a certain interest your opinion on the E-cat when I ran into your statement about Italians "forming a family of a sort". I think it would be time to get rid of such clichés. Mentioning the Mafia PC game wasn't fun either. National lobbying applies to many other countries in the scientific community.


    Andrea Migliori

  67. Dear Andrea,

    thanks for your feedback but I really can't believe that you're serious if you're serious. Is it un-PC in Italy to mention the mafia?

    You must be joking, right? It's a historical fact that just like an Italian, Galileo Galilei, invented the scientific method as we know it, the Italians also invented the mafia, based mainly on protection racketeering, and other Italians, led by duce, pioneered the fascism. It's also a historical fact that the Mafia PC games were created by Czech developers and I am not ashamed of this fact, either. ;-)

    You can try to deny such things but you won't succeed because it is utterly ludicrous to deny them. They're not just historical facts but also an important part of the contemporary events. Just search for Italian Mafia at Google News


    to see that there's absolutely nothing fake about such "cliches". On a daily basis, one readers about prosecutors murdered in the mafia style, Italian politicians linked to the Mafia, renewable energy barons who are key players in the Mafia (I rarely run such purely political stories but this one was even on TRF), Mario Balotelli (athlete) accused of running a drug unit in the Mafia, and so on, and so on. You're insane if you want to prevent people from mentioning that there exists something such as the Italian mafia or even mentioning it.

    Your claim that "national lobbying applies to many other countries in the scientific community" only proves that you are thinking in the mafia ganster's way. Supporting fraudsters such as Rossi has nothing to do with a national lobbying in science; it is a proof of the collapse of the internal Italian checks and balances that should have proven a long time ago that he is a fraudster who would already be arrested. This didn't happen exactly because science plays no important role in Italy. For the same reason, the Italian public doesn't hesitate to endorse the medieval witch hunts against the seismologists - and do tons of other scarily idiotic things.

    The Italian nation inherited lots of fame and sophistication from the Roman Empire and the tradition of the reneissance and other things - which is why it was behind lots of marvelous contributions to the mankind as well as some of the scariest ones. One shouldn't deny or try to deny any of these things. Attempts to deny its own history combined with the prosecution of some of its best people combined with the will to celebrate (and worship as great scientists) some of its most reckless and unscrupulous fraudsters and 3rd class crackpots only shows how deeply the bulk of the Italian nation has degenerated.

    Best wishes

  68. Absolutely right. Even then, searching arXiv will ONLY return the (non-peer reviewed) papers that authors have opted to post there, it often doesn't correspond to even a fraction of the peer reviewed papers published - for instance, even at my early-career researcher stage, I have 7 papers on arXiv, but 33 published through the peer review process.
    Searching for the authors with Scopus, one of the two main search engines used by physicists (the other is Web of Science, both unfortunately require you to have an login, usually through a university or research lab subscription) gives you the full publication story. I found that, for instance, Giuseppe Levi of Bologna University has, while not a long publishing record in a particular specialty, a number of peer-reviewed papers published. Hell, it's even on the Bologna University website. As rien points out, Prof. Bo Höistad of Uppsala University has over a hundred publications in nuclear physics journals listed on Scopus, and nuclear physics papers ALWAYS have dozens of authors. Have you seen how many coauthors are on the paper announcing the (likely) Higgs particle discovery from CERN?

    It's clear that at least Höistad is a serious nuclear physicist with a long publication track record, and a number of the other authors are at least associate professors in top quality universities (e.g. KTH in Stockholm), even where they haven't published extensively.

  69. Hi, you may count papers with 20 or more co-authors as "Bo Hoistad's papers" but I don't. Just look at Google Scholar


    and find a paper with at most 4 co-authors which would mean at least something.

    I've never used Web Of Science although it's known to me - and known to be completely unusable - and I've never heard of Scopus at all but be sure that Google Scholar is the most inclusive search engine of this kind while high-energy physicists, astrophysicists, and related disciplines mostly use INSPIRE for these tasks:


  70. Hey "Lošer Motl", fuck you!

  71. Cold Fusion Online Training, ONLINE TRAINING – IT SUPPORT – CORPORATE TRAINING http://www.21cssindia.com/courses/cold-fusion-online-training-92.html The 21st Century Software Solutions of India offers one of the Largest conglomerations of Software Training, IT Support, Corporate Training institute in India - +919000444287 - +917386622889 - Visakhapatnam,Hyderabad Cold Fusion Online Training, Cold Fusion Training, Cold Fusion, Cold Fusion Online Training| Cold Fusion Training| Cold Fusion| Courses at 21st Century Software Solutions

    Talend Online Training -Hyperion Online Training - IBM Unica Online Training - Siteminder Online Training - SharePoint Online Training - Informatica Online Training - SalesForce Online Training - Many more… | Call Us +917386622889 - +919000444287 - contact@21cssindia.com

    Visit: http://www.21cssindia.com/courses.html