[Technical Thread] Brightness of the reactor glow in the Lugano pictures and reactor temperature

    ".......compatible with the known surface temperature of the alumina (780C)."


    780 degC??


    And what would be wrong with the analysis of Bob Higgins, which concluded 1100 degC?


    https://drive.google.com/file/…4cOM2Zl9FWDFWSUpXc0U/view


    The only thing we can be certain of wrt the Lugano test is that the alumina surface temperature was not as high as 1400 degC.


    But that does not mean that COP was lower than stated in the Lugano report. It could be even higher. Too many unkown parameters. Like energy spectrum of LENR and related alumina transmittance.


    Use of IR thermal measurments is not suitable to evaluate energy released from LENR.


    One known data point is that the core of the Lugano reactor reached a temperature over 1450C because the nickel powder melted.

    Axil, may be you are right, in which case a large important part of core LENR energy must have been radiated through the alumina without heating it, i.e in the part of spectrum of high alumina transmittance and outside the IR camera spectral range.

    Quote from oystla

    Axil, may be you are right, in which case a large important part of core LENR energy must have been radiated through the alumina without heating it, i.e in the part of spectrum of high alumina transmittance and outside the IR camera spectral range.


    See

    LENR reactors need magnetic confinement


    Hydrogen Rydberg matter can escape from the reactor if the reactor is transparent to it. There is a great deal of energy invested in producing HRM. Losing it is detrimental to keeping the LENR reaction going. There is also experimentation that indicates that this lack of confinement can occur.

    Quote from Thomas Clarke

    These estimates for the wire temperature (unless you believe the strange "shadows" argument) seem perfectly plausible, and compatible with the known surface temperature of the alumina (780C).


    The temperature gradient can be related to the heat flux precisely if the wire helix OD / reactor diameter is known. (The fins complicate things but can be well approximated, since the ratio applies inside a log and therefore does not matter much). However unfortunately the power out comes from both heat flux and radaition from alumina surface, and also direct radiated power out from the wire, so things remain unclear.


    I'm a bit confused by your remark.


    The last plot is about the core temperature, not the wire temperature. I assumed that the core is made of Inconel as well. The case of a dummy core wrapped in radiating coils would be different.


    Also the "shadow argument" was an answer to H-G's objection that shadows are impossible. That simple model indicates that shadows seem possible provided the core is a bit warmer — as expected.


    I tend to agree with the authors of the report when they say that we do see shadows, and that this indicates that the reactor core is likely warmer thatn the heating wire it's wrapped in. That doesn't necessarily imply that there is excess energy production. For example, a skeptic might imagine a case where one of the heating wires is not wrapped helically but is goes through the center of the reactor, providing a "fake heat source".


    Let's not forget that the numbers, calculations, curves, etc. on this thread are preliminary exploration. All of this needs to be cleaned up, verified and written up properly.


    Ideally, we would get an expert in radiative heat transfer to look at the Lugano report and at the various objections. Does anyone know how we might go about this?


    Otherwise, I managed to locate the Olympus Stylus VH-520 I had. This low-end camera has four built-in color temperature settings: "Daylight fluorescent", "Tungsten incandescent light", "Cloudy weather and "Fine weather". I derived transformation matrices between each of those and validated the results using a different set of images, see here: http://i.imgur.com/mfK2B7m.jpeg The columns are almost identical, demonstrating that the effect of the camera color setting can be reproduced. The RMS error is 0.5 to 6.7%. Later I will use these matrices to repeat the Lugano calculations for different possible color settings. This will also provide some uncertainty values. The behaviour may be different for images containing a bright <1500K blackbody. Pointing the camera at wood burning in a woodstove shows the pinky haze similar to the MFMP, but only under some color settings, and I haven't applied the color balance transformation to those images yet.


    Other pictures of glowing blackbodies at a known temperature are welcome. I did look through some of the MFMP videos but all I see is a quarter screen of very low-res Google hangouts, and the reactor is usually completely saturated.


    We can deduce quite a bit more if willing to think. I answer your question below.


    oystla - you have two well written papers saying different things:


    [1] https://drive.google.com/file/…4cOM2Zl9FWDFWSUpXc0U/view


    [2] http://lenr-canr.org/acrobat/ClarkeTcommentont.pdf


    That is pretty normal. And you are not willing to do the work to reconcile the two? I'm not sure until you have done this you are wise to take any view on the matter. My paper was actually written after Bob's, and I pointed out to him that his analysis was incomplete. Since then I notice that he has added to his paper, and it will need rereading.


    So from [2] I solve this:


    R(erep, Trep) = R(ereal, Treal)


    where:
    erep = report emissivity (Bob and I agree 0.4)
    Trep = report calculated by Optris temperature (Bob and I agree 1400C)
    ereal = real band emissivity (Bob calls this effective emissivity). We both agree that this is got by integrating spectral response of Optris * spectral emissivity of alumina, both reach a view that this is around 0.9 based on numerical integration.


    R is the Planck function evaluated for a given temperature and weighted by the alumina band (effective) emissivity and the Optris camera sensitivity.


    The difference in emissivity values thus means that we need to find the temperature Treal at which the radiant Plank power over the IR band from T=1400C would reduce by a factor of 0.4/0.9.


    I calculated this numerically from the Planck curve. You may prefer the equivalent but using a readily available black body band power calculator entering the 7u - 13u bandwidth. You first work out the radiance in this band from 1400C. Then you multiply this by 0.4/0.9. Then you determine the temperature in the calculator that will give the reduced radiance. Please let me know your results? It is a basic sanity check - not as precise as numerical calculation that includes the IR camera bolometer spectral response but a decent approximation.


    I'll give another "analytic" solution for a ball park result. The two temperatures, 1053K and 1673K have a ratio of 0.63. The required ratio is 0.444. solving 0.63^x = 0.444 we get the power law relationship needed for Planck to make this:
    x = 1.8


    These are ball park figures, I have not done it accurately, but you can see (again use Planck calculator) that this power relationship is about right for the IR band at these temperatures.


    Now, while I await your calculations using a web BB calculator (please do this) let us look at how Bob reaches his 1100C (1373K) temperature. That, BTW, would imply a power relationship of roughly x = log(0.444)/log(0.801) = 3.65. Given these are ballpark figures that looks remarkably close to the well known Planck curve total power.


    So - before decoding Bob's paper - his estimate seems to be based on assumption that the Planck band power varies with T in the same way as the Planck total (black body) power. That is very wrong.


    Quote

    If, instead the single value emissivity parameter was iterated using the curve of Figure 8, then the actual temperature may have been found near 1100°C.


    This appears to be Bob's explanation of his method for calculating the temperature. (p5, col 2, para 3).


    Also:

    Quote

    The use of the reduced temperature estimated in this paper (1130°C)


    So Bob's exact estimate would appear to be 1130C. (p5, clo 2, para 4).


    Now Figure 8 is captioned: "Figure 8: Single value emissivity vs. temperature including spectral response of a microbolometer FPA"


    This does not help with how he gets 1130C - since the graph does not include the contribution of the Planck curve, and the text does not say how he calculates this as obviously he must.


    Let us do a sligtly better calculation based on Bob's data;
    Trep = 1410
    Treal = 1130
    erep = 0.4
    ereal = (from Figure 8 at 1130C) 0.901
    0.4/0.901= 0.444
    1407/1683 = 0.836
    log(0.444)/log(0.836) = 4.5


    So Bob appears to reckon a roughly T^4 dependence of band power (7-13um) on temperature in K. That is clearly wrong, and indeed playing with a BB band radiance calculator shows the dependence at these temperatures in the IR band to be around around 1.9, consistent with my numbers. (You'd expect this. 4 is the total power. 1 would be the power at very long wavelengths a long way from the Planck peak. We have here wavelengths 7-13u with a peak of around 2-4u.).


    Oystla - I did put similar calculations (though not comparing with Bob's report) on the other thread. There is nothing mysterious about my numbers, anyone with a Planck power calculator can generate the same result.


    Bob made the same mistake - assuming that the IR band power scales with temperature as T^4 - in an earlier versions of the paper. In that as well he was unclear about exactly how he calculated Treal from Trep - but the numbers speak for themselves. There is a good rule here - state precisely how you work things out. Anything not precisely stated may prove to be a mistake - much more likley so than when worked out properly. That is why I included all my code...


    Best wishes, Tom

    Thomas,


    If you have issues with the paper of Higgins, you should challenge him, not me. He is frequently on this forum.


    And Nice to hear you think your own paper is well written, good for you ;)


    Anyhow, I find the paper of Higgins a litle more convincing. It's just the way he builds his case to prove his point ;)


    Yes, he estimated 1130 degC, and the figure 8 was carefully deduced.


    And, yes Stephan Boltzmann law will apply ( T^4), but of course if a single emissivity Value is used it must be based on a weighted average of the whole band of the emitted spectrum.


    Another indication of temperature actually being at least 1100 degC, is the Ash sintering indicated from Axil. And It's likely that an important part of the energy have been emitted from the core in the spectrum of high alumina transmittance, meaning the temperature could be 1400 degC or higher in the core.


    Anyhow, my conclusion is that IR was a terrible choice of deducing energy released, as long a LENR energy spectrum is unkown.

    Quote

    Another indication of temperature actually being 1100 degC, is the Ash sintering indicated from Axil.


    Sorry, I made mistake here. The nickel was melted so the temperature of the reactor core was at least 1450C.

    We seem to have wandered off the path again..


    @Antoine10FF, you say that your model assumed an Inconel core tube. How does a "dummy" core (alumina ceramic) look? I believe that the core tube in indeed alumina, although the contents ("fuel") may alter significantly the attributes compared to a hollow alumina tube.


    In my estimation, the coil windings are wrapped slightly tight around an alumina core tube (They probably loosen a bit at operating temperature, possibly considerably if the coil wire ends are captive). This should improve heat transfer to the core by some degree of conduction. I am also of the opinion that an air gap exists between the inside of the outer tube and the outside of the coil windings, roughly 1 mm, (possibly as low as 0.5 mm) which mean heat must primarily radiate to the outer housing, and this air gap and radiative zone is therefore insulating to some substantial degree. The end caps seem to have the wires encapsulated at least for that 4 cm long section, or at least the wires fit fairly snuggly in grooves in the caps to admit the wires between the caps and the core tube (IMO). Something like 30% of the total heat generated by the coil wires is produced in the straight sections of the coil wires (both ends combined), based on some modelling. In this case the heat transfer to the caps is primarily conductive, at least until that is no longer a physical possibility.


    I hope this helps your model, and if you have the time and are willing, I would be interested to how the above version of construction would affect your model.

    Good question.


    However, the big particle does not seem to be solid nickel, simply pure nickel (at least where they tested it). It looks sponge-like.
    I would suspect if the temperature was close to 1450°C, there would not have been a bunch of particles so much as a solid stick. With all sorts of added extras, perhaps that might not be the case.


    Note that the nickel fuel particle tested is nearly as big, and there are several others nearly as big in at least one dimension. It wouldn't take more than a couple of those together to sinter into a rather large particle. The transformation into Ni62 is another story. (I am familiar with your opinions in that regard).


    Did you see that Rossi has recently confirmed the ash was re-tested, and the Ni62 was confirmed (no word on the % unfortunately).
    http://www.journal-of-nuclear-physics.com/?p=892&cpage=55#comment-1150286

    Quote from Paradigmnoia

    We seem to have wandered off the path again..


    Yes.


    Quote


    @Antoine10FF, you say that your model assumed an Inconel core tube. How does a "dummy" core (alumina ceramic) look? I believe that the core tube in indeed alumina, although the contents ("fuel") may alter significantly the attributes compared to a hollow alumina tube.


    Quote


    I hope this helps your model, and if you have the time and are willing, I would be interested to how the above version of construction would affect your model.


    I will look at that when I get back to this, but probably not before
    Sunday.


    So if I understood correctly you are talking about an alumina cylinder
    (inner diameter D1, outer diameter D2), containing the fuel, wrapped
    in three Inconel coils made of wire of thickness d with N turns per
    resistor over the length of the core (20cm). This is contained in the
    outer alumina cylinder (inner diameter D3, outer diameter D3) with the
    (neglected) fins.


    Would you mind repeating the dimensions?


    Also, what about visible emissions from the contents of the inner
    alumina tube? For a metal inner tube, only visible thermal radiation
    from the tube itself needs to be considered. But for an alumina tube,
    transparency must be taken into account.

    @Antoine10FF,
    No rush required. Thanks for considering it.
    For clarity, Inner core tube, ID 7 mm (D1), OD 10 mm (D2);
    Outer tube (minus fins) OD 20 mm (D4), ID 16 mm (D3) [maybe 15 mm?]
    Lets call the wire diameter 1.5 mm for simplicity, it should be close enough for now. Lets say that the coil is tightly wound around D2 for now.
    [These D1 etc, are not all as pictured in the earlier document.]


    The problem of the inside of the inner cylinder is a bit murky. We could assume a reflective inner surface coating or an opaque inner tube. I'm not sure about that. It might be worth looking at both. I will guess that it is not transparent (with fuel in it), and it might cancel itself out if it is modeled that way anyways (IE: the dummy version).

    Quote from Paradigmnoia

    Let's call the wire diameter 1.5 mm for simplicity, it should be close enough for now. Let's say that the coil is tightly wound around D2 for now.


    I'm not sure about the tightly wound part. If you look at figure 10 from the Lugano report, you can kind of see a faintly glowing shape that seems to be wrapped around the core.
    Some creative adjustment of color hues makes it more obvious:



    If this is one heater coil, it only makes three turns.

    I have not continued in this analysis because of lack of time, and lack of perceived merit [from my perspective] given the unknowns. Here are some observations to add to your list:

    • For photometry with a camera, you have to know the exact camera that was used for the "visible" light photographs. Many camera manufacturers have allowed the red band of their filters to extend into the infrared for improved low light sensitivity. The "purple fringe" is caused by IR light leak of the blue filters in the camera - also related to the longer wavelength used for the IR blocking filter in the camera (if it even had one). It is hard to do photometry without knowing just what camera and sensor are being used and what the filter line-up was for the optics.
    • The other thing that confounds visible light photometry analysis is the transparency of the alumina and how it differs in the visible band and within the Optris band. This also affects the radiation calculation. As the blackbody spectrum shifts into the visible spectrum, the higher temperature heater coils will emit more radiation directly in the visible and it will pass through the alumina. Without having multi-band information, calculating this flux is painful or impossible. It is also hard to estimate transparency for a cast alumina - their's was NOT "pure" alumina as the Lugano analysis suggested. [Pure alumina would have to be fired at a temperature that would have destroyed the heater wire. Cast-able aluminas are only about 70-80% alumina. The XRD technique used to analyze the material of the cast section probably would have only seen the crystalline phases and missed the filler/binders.]
    • What has bothered me about the 780C estimate for the temperature is that it is inconsistent with the apparent brightness that is observed. What happens when the blackbody gets hotter is that not only does the spectrum peak shift toward the visible at higher temperatures, but it also gets brighter from increased radiation. Both factors contribute to increased perceived brightness of the blackbody object. I have run my alumina reactors with the temperature measured using thermocouples [easier to do with no ridges]. The brightness and apparent color at 780C for my alumina reactor was no where close to what I perceived from the Lugano photos, suggesting that the calculated estimate for the Lugano temperature of 780C was too low for some reason not considered.
    • In the Lugano reactor, the transparency affects the amount of power radiated from the hotter [than the surface] heater coils and the hotter core of the reactor. Radiation will be the dominant path for energy exit. So a different calculation must be used for radiation and a different temperature (the heater wire temperature, the core temperature) for the visible spectrum. Convection, on the other hand, must be calculated on the basis of the outer skin temperature [which may be in the 780C range] - higher at the root and colder at the tips. But convection was not the dominant heat release path. My point here is that because of the transparency in the visible range, a single temperature cannot be used for both convection and radiation calculations. An effective or composite single temperature, if used, will require nonlinear compositing as the temperature rises, and the compositing will be VERY device and materials specific. Believing that a single temperature could be used for both radiation and convection in the Lugano calculations was probably another mistake.

    Because I had insufficient information to continue with analysis, and because personal experience suggested that the reactor was too bright to have been at 780C [for a single effective temperature] as others suggested, I decided that continued analysis would be of little value. As it stands, the actual Lugano numbers have been appropriately called into question. I considered that new data was needed before I could justify a particular composite temperature estimate. While re-creating the Lugano setup could be done [and there is some MFMP data on this that should be examined], I do not consider that having better data from a dummy re-creation of sufficient value for the effort required.


    All of this has made a difference in my work. It has caused me to believe that credible quantified excess heat data can only be secured by using a well calibrated calorimeter.

    Bob, Thanks for your interesting and considered views above.


    You are answering a slightly different (though valuable) question from that originally at issue.


    indeed the surface temperature estimate and the core (wire) temperature estimate are likely different. You are then hoping to find a "composite" radiant power temperature measurement which I agree is not possible with any accuracy.


    Your paper quoted by Oystla contains a conclusion for surface temperature estimate from IR thermography alone, it does not attempt a "composite" estimate. The surface temperature is a soluble problem, and a correct result of some interest, but your estimate is based I have shown on an incorrect calculation as above and should therefore be changed, or if you feel the surface temperature to be wholly uninteresting you could withdraw the estimate. Of course I'm happy to be corrected if wrong in this statement.


    The surface temperature is calculable from IR thermography and precise at 780C. Your paper states that this same IR thermography surface temperature estimate is in fact 1130C (without giving clear reasons), and this is obviously confusing a lot of people, including Oystla here. Therefore I'd like you to acknowledge this - that my calculation for 780C is correct - or at least that your's is wrong as it so appears to be.


    The point is that this surface temperature calculation is definite. It depends only on things that are well understood and the values used are inherently accurate. Surface roughness etc cannot alter a very high emissivity significantly, and fins (which also do not alter it much), have a precise calculable affect.


    The second matter relates not to surface temperature but to emitted power, and wire temperature. I'm happy to (broadly) agree with all your points, indeed I summarise them and a few others as issues in my paper. I agree the wire could be a decent amount hotter than the surface and while I'm reluctant to use photos to estimate temperature for very obvious reasons they are consistent with an 1100C wire temperature. I don't think this judgement is at all certain, but it is possible. Although it is work that has not been done a limit could be put on the wire temperature from the geometry, specific heat capacity of alumina, and known heat transport through the alumina surface. That is sort of interesting, but can only be done if we have clear understanding of the alumina surface temperature.


    As for output power this is complex because there are so many uncertainties in the radiant output power both up and down from the vanilla "alumina total emissivity" guess. Just to list them, given that over a significant part of the spectrum the alumina is translucent we have:
    (1) different temperature of radiating inner surface (+)
    (2) different effective surface area of inner radiating surface (reduces above + a lot, but since the rest of the area is still alumina does not actually make a -)
    (3) different total emissivity of inner radiating surface (could be + or -)


    So while I agree with the point you make about possibly hotter inner wire surface, the thermographic data for temperature is the one part of the jigsaw that is well nailed down (when correctly calculated) with less uncertainty than anything else. Given alumina is translucent in visible wavelengths clearly what the wire looks like can have no direct bearing on the alumina surface temperature, and so your argument there for being hesitant about the surface temperature does not hold.


    The calculation is a simple one. You could do it roughly using a web band emissivity calculator - no need to follow my numerical integration.


    When dealing with complex problems it is proper to emphasise uncertainty and say you can't be sure. It is also proper, and useful, to be clear about the things you can be sure about (in this case the surface temperature, though since you did not actually to the precise calculation in your paper maybe this for you represents further work, and therefore you should acknowledge that you have no information contrary to my easily checkable estimate). There is even then some uncertainty, with different alumina microstructure altering slightly the band emissivity, but we both know that at these high values characteristic of an insulator that is not likely to change much.


    best wishes, Tom

    @Thomas Clarke
    Yes, Tom, it would probably be good if I went back to that paper that was written over a year ago and just take out that section. The paper was really about emissivity. At the time it was written, I had not threaded through the other issues yet. In some sense, revising the paper now may not have much impact because the paper in its original form is in the ether. Do you think it would have value to revise the paper in that way today?


    I have come to think about this reactor more like an ordinary incandescent light bulb. In the light bulb, you have total power output that comes from envelope convection and radiation at its temperature + heat conducted through its base + power radiated through the partly transparent envelope (which depends largely on filament temperature and spectral transmission of the envelope).


    In the case of the Lugano experiment, what I haven't threaded through is how use of data from the dummy run may have "calibrated out" (cancelled) some of the errors in the actual experiment evaluation, providing a more correct answer than absolute calculations would have allowed.

    @Antoine10FF
    Have you considered that the pitch you see evidence for in your analyzed photo is the "beat" between the pitch of the ridges on the outer tube and the pitch of the heater coil?


    The heater coil is 99% likely to be as shown in the patent drawings - a triple wound 3-phase coil that is wound with about 1-2 wire thickness spacing.

    Quote from BobHiggins

    In the case of the Lugano experiment, what I haven't threaded through is how use of data from the dummy run may have "calibrated out" (cancelled) some of the errors in the actual experiment evaluation, providing a more correct answer than absolute calculations would have allowed.


    To add to this, I am struggling to understand how the temperature of the live run could have been or be known with certainty given that the power-in for the dummy dummy run only went up to 500 W (p. 7), well below the 900+ W used in the live run.


    In addition, the Levi et al. say this about the calibration (p. 7):


    Quote

    “Dots” of known emissivity, necessary to subsequent data acquisition, were placed in various places on the cable rods. It was not possible to perform this operation on the dummy reactor itself (and a fortiori on the E- Cat), because the temperatures attained by the reactor were much greater than those sustainable by the dots. We also found that the ridges made thermal contact with any thermocouple probe placed on the outer surface of the reactor extremely critical, making any direct temperature measurement with the required precision impossible.


    Here we learn that even in the dummy run the dots were not used on the body of the reactor but instead on the cable rods. My reading of this section as a whole is that considerable uncertainty was introduced into the Optris temperature calculation for the live run as a result of an inadequate calibration.

Subscribe to our newsletter

It's sent once a month, you can unsubscribe at anytime!

View archive of previous newsletters

* indicates required

Your email address will be used to send you email newsletters only. See our Privacy Policy for more information.

Our Partners

Supporting researchers for over 20 years
Want to Advertise or Sponsor LENR Forum?
CLICK HERE to contact us.