Mizuno reports increased excess heat

  • Of course. In fact with claimed R20 levels of power you don't need a box.


    The R20 results in the paper comprise one sample datapoint, and anecdotal accounts. Ignoring the anecdotal evidence, the one datapoint needs to be validated,

    now THHnew said that the R20 results were based on one single point


    This is patently untrue.


    The spreadsheets for these results contain hundreds of data points.


    THHnew alleges a processing error


    but cannot identify where it might be.


    Is it in the anemometer the thermistor or in the calculation..


    why would Mizuno suddenly change his processing for R20


    Its the same calorimeter setup.


    clacualtion Output = mx Cp x delta T


    where is the error


    Pin point it please


    No vague handwaving.. some other error...etc


  • Well, as someone reading the posts written here about airflow to which you replied above, and questioned, I am capable of doing this. I note that after questioning my competence to do these airflow calculations, and then caliming my bounds on airflow were wrong, you have still not substantiated this and now are asking different questions, about the R20. I take from this that in fact you agree with my bound of approximately 18% for airflow, and am happy therefore to change topic.


    Where is Mizuno's processing error in R20? Again, you are showing a misunderstanding of uncertainty. I do not know that Mizuno has any error in generating those results: obviously many here are hoping they are correct. I am just pointing out that there could be such an error. I have even volunteered one possibility. Since the R20 has a different heater the much better results could be if the (new) heater input power was wrongly calculated, for example by using the same formula wrt voltage as used for the previous heater, when in fact the heater resistance was much lower, thus explaining the huge apparent COP. Of course it could be something else, or no error. More detailed data would help to exclude possible errors. For example, I & V measurements for the input would preclude such an error due to the wrong voltage to power conversion being used.


    You will realise from the fullness with which I have answered your many and detailed questions in this thread that I consider no question inconvenient. Also, that rhetorical questions are much less helpful than real questions. Rhetorical questions just add to noise and serve little purpose.

  • why would Mizuno suddenly change his processing for R20


    Its the same calorimeter setup.


    clacualtion Output = mx Cp x delta T


    where is the error


    The most likely error here is not in the output power calculation, as you suppose, but in the input power calculation, as I have stated repeatedly above. Since COP = output / input obviously either error matters.

  • now THHnew said that the R20 results were based on one single point


    This is patently untrue.


    The spreadsheets for these results contain hundreds of data points.


    I apologise to those careful readers of this thread who realise this has been asked and answered before. And to those who are just bored with this exchange. Still, it may help to explain further.


    The R20 results indeed have many points - all logged identically, for both calibration and control. The issue is that this was just one experiment, and the calculations for it are not laid out as fully as for the detailed R19 data. Mizuno in the paper calls this a sample, indicating that it is an initial indication rather than a considered set of experiments written up fully. That therefore makes mistakes more likely, as does the change in design of the heater, which may not have been considered when doing the calculations. We cannot however know without more data which no doubt will be forthcoming since this reactor shows such extraordinary results it will be measured every possible way by many parties, if Mizuno allows this. I have no reason to think he would not allow it.

  • and now are asking different questions, about the R20


    I have asked those way back

    you never answered


    The topic is "Mizuno reports increased excess heat"


    Let me remind you of that


    Errors in the input?

    but in the input power calculation


    The processing changed where?


    Calculation Input = mx Cp x delta T.


    where cp? m? deltaT??


    Please be specific "The most likely error..


    Justify likely... is this just posturing... 90% likely 99% likely?

    just one of those THHnew vague mathematical terms??

    Mizuno in the paper calls this a sample, indicating that it is an initial indication rather than a considered set of experiments written up fully


    Another one of THHnew's many assumptions.

    why would Mizuno change his processing of data with the same calorimeter setup?

  • Could THHnew following his exhaustive analysis of this matter and his assertions of 24% -64% error in the anemometer readings

    show exactly what his assertions are based on by filling in the following datasheet.



    I recommend readers of this thread to follow the entire argument. After a very useful contribution from Bernard above, pointing out that turbulent flow in pipes takes a pipe length to develop that can be estimated, and that this estimate is much longer than the 66mm pipe, I agree that the 20% error from the turbulent flow calculations was not a reasonable bound, and that this resolved the difference between turbulent flow calculations and the paper traversal measurements. Which is why my error bound is now 18%, not depending on turbulent flow calculations. This comes with the insight that (contrary to what is implied in the paper) the pipe is not long enough to influence the airflow velocity profile much.


    I also refer back to my request that RB, who I think believes this 18% error bound to be wrong, should therefore show is his (tighter) error bound. RB - you could do this by putting error bounds on each of the numbers in your spreadsheet, and then calculating overall bound. That would possibly give us a more accurate bound than I have given. And I'd be happy to comment. In any case a difference between your bound and mine could then be related to different assumptions we make. For the details of my calculations I have given them recently above and am unwilling to repeat for reasons of thread noise.


    (Technically a tighter error bound does not make my bound wrong, just conservative. To show my bound wrong RB would need to argue that in fact it is too tight and there are extra errors I have not considered etc).

  • Basically THHnew indications and assertions are based on many assumptions.


    The most basic assumption was well detailed in 2017..


    There MUST been an error somewhere,


    The major error's that I have seen are in THHnew's calculations.

    '

    as in


    THHuxleynew wrote:

    That is because at the temperatures I did this calculation (380C reactor vs 80C wall) the re-radiation is\\

    less that 10% of the radiation due to the T^4 factor for relatively small gaps.

    Of course THHnew alleges that he forgot the correct formula..


    In view of this forgetfulness I will forget his assertions and assumptions


    although this critical stance might be useful

    in practice it is not

    because the THHNew critique is so garbled and unclear



  • RB: no bound

    Mizuno: no bound

    THH: some aspects bounded

    I just need your assumed data

    minus the assertions THHnew


    Then a numerical comparison of error =THHassumed expected minus THHassumed Mizuno can be made.


    The reason for this is that THHnew made many assertions about the foil relection until he did one clear calculation

    which showed that he was using the wrong formula for many posts

    Numbers minus THHnew words are the clearest indicator for me.

    Of course Thhnew may be correct..about the anemometry..

    but I just need to check...


  • gerold: do you want info on the ones I rely on?


    FWIW - I agree with those rather complex radiation formulae, though the argument has moved on since then. RB sometimes reiterates past matters and mixes them up with current.

    I would like to come up with a simplified mathematical model of the internal heater and the outer cylinder. I assume that since the pressure is that low 1-3 mbars heat transfer and convection can be neglected. Therefore we have only radiation between heater and outer wall. Please correct me if i am wrong. If you have relyable info i can then crosdcheck what i have available. Thx

  • I would like to come up with a simplified mathematical model of the internal heater and the outer cylinder. I assume that since the pressure is that low 1-3 mbars heat transfer and convection can be neglected. Therefore we have only radiation between heater and outer wall. Please correct me if i am wrong. If you have relyable info i can then crosdcheck what i have available. Thx


    Agreed. There is some question maybe as to the geometry of the internal heater - which is bendable but the ones we have seen are longer than the cylinder. Just a caution - I have not followed that much.


    Those calculations will work for that problem, I see the uncertainty as being working out the (at temperature) emissivity of the heater, and the emissivity of the inner wall of the reactor, since these can change with surface effects. Use concentric cylinder inside cylinder (ignoring end walls not that significant).


    THH

  • Are these the formulae you used to calculate 80C for the aluminium foil temperature on the inside of the box?


    Or did you use another formula?


    I used another method entirely.


    I took the R value given by Jed, and the box surface area from the paper, and the box heat loss in Watts from the graph in the paper, to determine the expected difference in temperature across the insulation. It was a ballpark calculation.


    But I'm not sure this is current...

  • Agreed. There is some question maybe as to the geometry of the internal heater - which is bendable but the ones we have seen are longer than the cylinder. Just a caution - I have not followed that much.


    Those calculations will work for that problem, I see the uncertainty as being working out the (at temperature) emissivity of the heater, and the emissivity of the inner wall of the reactor, since these can change with surface effects. Use concentric cylinder inside cylinder (ignoring end walls not that significant).


    THH

    My intention is already to work on a next iteration of the design. Make it more compact by using standard components from swagelok or other manufacturer, max. diam. of reactor tube therefore 28mm - SS made inert by Si CVD coating, avoid welding, use of a high performane (ceramic) glow plug as internal heater, catalyst - Ni foam galvanized with paladium in a tubular shape attached to glow plug. For this I would like to come up with a mathematical (simplified) model.

  • In fact the safe approach, which any really high quality scientist would provide, would be to assume nothing, and question everything, including one's own expertise. That would mean lots of external advisors called in.


    Some people would reckon you would also want a magician, to check for possible trickery.


    Sometimes, people who have spent time training themselves in the use of high-precision experimental devices may forget that such devices are not always the required tools. For the R20, perhaps, a fuse, a thermometer, a Rolodex and a phone might be appropriate tools to apply to the problem.


    At the power levels and excess heat reported for the R20, I believe independent measurement would require little expertise. My point is, if these results are personally observed by two or four people, those people would be able to bring in others with more expertise; then, if those people could not explain what they saw, they would be able to bring in people with more reputation, and so on. As for THHuxleynew's mention of a magician, or, more generally, a professional debunker, I'd imagine that if the R20 passed through a couple of rounds of Rolodex, such a professional would be brought in prior to any announcements, because people with reputations don't like being fooled in public.