Mizuno reports increased excess heat

    Here are some of the gadgets and clutter in Mizuno's lab. The $16,000 power supply is on the lower left. There are shelves full of miscellaneous stuff in the background. If you go through and throw away something that has been sitting there for 20 years, the very next day you will find you need one. It's practically a law of nature. There are instruments on the shelves from the 1950s.


    All of this came crashing down in the recent earthquake. You can imagine what that did.



    Researchers tend to buy things without regard to cost. They spend money like water. You have to expect that. I mentioned that when Edison got support from millionaire George Harrington, he immediately spent three times more than he was budgeted. R. Conot described this in his book, "A Streak of Luck," p. 45:


    [Edison] lacked the training, aptitude, and interest necessary for operating a business. He did not know how to estimate and balance income and outlay. He did not, in truth, want to know, for he was a compulsive spender. He was not profligate in his personal habits, though he did not live spartanly, either. But when it came to books, or chemicals, or machinery, or anything to do with experimentation, he indulged without regard to cost. When he received a large amount of money he went on a spree-it was his way of rewarding himself. To pay accumulated bills provided no gratification, so they were left to yellow till payment was forced by threat of court action. Edison's concept of balancing the books was to develop whatever income was needed to cover expenditures. . . .


    Harrington sent one man to audit the books in May, and another two months later to take charge of them.


    "About time!" Edison ejaculated. "That department is in a very slack condition! You cannot expect a man to invent and work night and day and then be worried to a point of exasperation about how to obtain money to pay bills. If I keep on in this way six months longer I shall be completely broken down in health and mind."




    He expected people to hand over the modern equivalent of millions of dollars, no questions asked, and don't ask how it was spent. Just send more money and stay out of the way. Which, in Edison's case, was a great idea, because he turned millions into billions. But it made his venture capitalist backers nervous! His backers eventually included the biggest names on Wall Street.

    Quote from robwoudenberg

    It would probably be useful to have a replica from the Master himself so see how well repeatability is.


    He has been replicating this for years. From reactor R16 to R20. Repeatability is excellent. Someone else in Japan already replicated, and reports it is working. Other people need to try this now. People skilled in the art.


    I do not think anyone should prepare a reactor, put deuterium into the reactants, and mail the reactor or reactants to anyone. I fear the reactants might self-heat and go out of control. It may be unlikely, but if it happened it could cause a catastrophic accident in an airplane flying from Japan to the U.S. We should not take any chances until this reaction is fully understood. I am confident that the reaction can be made safe and controllable, but that will take billions of dollars of R&D. Right now, mailing prepared and deuterated reactants is like mailing a fully charged battery or a can of gasoline. That's dangerous, and it is against postal regulations.

    Quote from JedRothwell

    He has been replicating this for years. From reactor R16 to R20. Repeatability is excellent. Someone else in Japan already replicated, and reports it is working. Other people need to try this now. People skilled in the art.


    I do not think anyone should prepare a reactor, put deuterium into the reactants, and mail the reactor or reactants to anyone. I fear the reactants might self-heat and go out of control. It may be unlikely, but if it happened it could cause a catastrophic accident in an airplane flying from Japan to the U.S. We should not take any chances until this reaction is fully understood. I am confident that the reaction can be made safe and controllable, but that will take billions of dollars of R&D. Right now, mailing prepared and deuterated reactants is like mailing a fully charged battery or a can of gasoline. That's dangerous, and it is against postal regulations.


    I'll have a look at Mizuno's previous papers, thanks for pointing out R16 - R20 were replications.

    As for transport: one could fill the replicated reactors with Argon during transport and leave the Deuterium loading at the testing site. Debatable, I know.
    The big puzzle is how to come to a stage where the outside world can be convinced something serious has been found and that it has been already replicated by several independent sources. If that can be done in a strategic way, professional money will be spend.

    Jed,


    Can you point us to a schematic of the vacuum/deuterium supply plus exhaust to the RGA/mass spec. From the photo, it appears that there is only a single gas/vacuum outlet on the conflat cylinder.


    I assume the unit gets pumped down before bakeout from the single line. It is then baked out under vacuum several times until there is no water in the RGA/mass spec (which is output from the turbopump).


    The question I am having is how does Mizuno keep the unit in a D2 atmosphere at relatively stable partial vacuums, i.e. 2 Pa, or 300 Pa or whatever. I have natural concern about air being drawn into the conflat under vacuum and then catalytically supporting combustion of the D2 on the hot metal surfaces. It is my hope that the schematic can rule that out, or instead monitoring output from the RGA/mass spec showing only very small amounts of H2O or D2O in the output.


    If the D2 and the turbo pump are controlled by valves, how are they regulated, i.e. automatically or by hand? Alternatively, is the unit pumped down, then loaded with D2 for the time necessary to achieve optimal D2 loading in the mesh, and then the valve sealed off while the pressure gauge monitors the slow yet steady rise as there is microleakage through the seals or outgassing.


    Note: a relatively simple way to measure the amount of microleakage is to pump the unit down to highest vacuum under operating temperature, close off all valves, and then measure the rate of rise of the vacuum over the fixed volume in the conflat cylinder. It can then be readily established that the quantity of O2/air and the D2O that could have possibly entered the conflat is limited to that which would cause the pressure to rise. Once the leakage rate is established, one can put an upper bound on the maximum heat from chemical reaction between O2 and D2 in the unit, thereby ruling that out in the results.


    It would also be helpful if the unit is under continuous D2 supply if a rate of D2 added per unit time (i.e. mol/s) can be computed from the instrumentation so again to rule out chemical heat from the supply.


    Thank you (and thank you for translating Mizuno's work to us over the years). Good work Jed -- you are a benefit to the LENR scientific community.

    Quote from Alan Smith

    That's not a claim it is a fact that a solenoid coil carrying a current creates a magnetic field. You know that to be a given. You also (I hope) know that magnetic field strength drops very rapidly with increasing distance...I advanced the hypothesis based on my own observations that a stronger magnetic field is the cause of the improved cold fusion reaction rate. As you say that can be tested. AC or DC makes no difference to the facts I mentioned, but merely add the frisson of field polarity reversal.


    OK - just a note. The R20 (which I was talking about) has a sheath heater - see diagram. this is a rod which however wound cannot have much flux escaping to the mesh - which is in thermal contact with the reactor body some way away. Whereas the much lower performance R19 has a large diameter coil outside the reactor. S-S will not shield magnetic flux and so the outside coil will deliver much higher flux to the mesh than the sheat heater.


    You are confused. The airflow speed has nothing to do with the reactor temperature. It only affects the calorimeter air temperature. The reactor is a 20 kg stainless steel tube inside the calorimeter. The temperature inside the reactor is 317 deg C. It is heated by sheath heater and the cold fusion reaction. The temperature in the calorimeter chamber outside the reactor is 34 deg C. They are worlds apart.



    Jed - you have pointed out that the reactor is insulated so that 75%+ of heat is transferred to the air stream. If you slow that down, you reduce the amount of heat expelled. Therefore the reactor casing temperature has to increase. In the case of no airflow the casing temperature must increase enough that all the power escapes through the insulation.


    We all know this. Take a constant power source, blow on it, the temperature cools because of lower effective thermal conductivity to ambient.


    During the 50 W calibration, the temperature in the reactor is 28 deg C, which is much lower than 317 deg C. That difference is not used in the calorimetry, but it could be, and it does indicate the reactor is a lot hotter when it produces 250 W of excess heat


    You are confusing, here, temperature and power. The casing temperature is determined by the power flux (claimed 300W active) the thermal resistance between casing and the airflow, and the airflow temperature. With slower air the thermal resistance increases because the boundary layer heats up more, as does all the air in the box. In fact you know well that if most power leaves the calorimeter via air, slowing the air must increase the temperature of the exit air and therefore decrease the amount the air cools the reactor casing for any fixed casing temperature.


    My point above is that the reactor casing temperature (and therefore the gauze temperature) can be controlled by altering the airflow, as well as altering the heater power. You can get the same excess heat from R20 (see above - to possible magnetic effect from sheath heater) by reducing heater power but also reducing airflow to keep the reactor casing the same temperature.


    A little more reflection and you will see that any system of this type with a COP of 6 - 10 where power out is controlled by temperature, and temperature is controlled by combined power in + out, must be close to self-sustaining and therefore could be made self-sustaining by reducing cooling. It is also likely unstable. Which is why in this type of system by far the best test rig is to stabilise temperature with a feedback loop that alters the airflow.


    That is if it works as supposed.


    I'm surprised others have not picked this up re R20 because the above thermal analysis is true for all exothermic temperature sensitive reactions that have the type of dependence Mizuno states. Higher COP means more unstable. If 50W delivers 300W, then 50W worth more insulation will deliver 300W without external power input.


    I'm also pissed off that people on this site don't seem interested in dispassionate analysis and interpret everything in terms of whether it is for or against what is claimed. In this case I am interested, hence posting even though there is some hostility. For me, interest in how systems can be analysed trumps worries about do the or do they not work. I am really not thinking about that: if it does work then it will be very exciting and usher in a new world. If it does not, then there will be some error. I don't suppose anyone here dishonest, but the R20 data in particular is very sparse, and the R19 data does not have full details.


    RB seems to vary between thinking I am out to deny LENR, or wanting to replicate this for myself. Neither is true. But I'll certainly follow this story when there are developments, as there must be given the very large R20 claims and large enough R19 claims.


    Have fun.


    THH


    Jed - this is weird. I agree with you - too stable. I disagree with you R20 as stable as R19. That is - I realise M claims it is - but given the much higher COP that really does not make any sense. In the same system R19 could happily be stable. R20 would tend to be unstable with hysteresis. Maybe you mean that too are just using different words.


    Anyway, I distrust this work because I distrust those R20 results. The 40W -> 50W change making a large output change is unlikely, because the chance of the system being stable, and having that high gain, are pretty small. Generally, if there have been lots of tests, you'd expect noted instability and (therefore) self-sustaining operation. NB - I use instability in the technical sense to mean a system that can have exponential signal increase until it hits system limits (in this case of high or low output).

    Robert, there is a difference between a hot-wire anemometer and a mass flow sensor. These units I am suggesting detect the actual molar mass of molecules passing the sensor so regardless if the flow is laminar or turbulent, the meter still reads mass flow correctly. Yes as you suggest the absolute values need to be recalibrated but one could easily purchase an insertion mass flow meter such as those made by Sierra inst. which is relatively easy to have calibrations with atmospheric air. The total uncertainty for mass flow would be less than 1% and properly designed thermistors with triple point of water and gallium calibrations can give you about a 100C range with which you can measure with 10mK uncertainty. Mind you, we are looking for a 10C difference and our accuracy will be +/-0.01C, so when you couple this accurate temperature reading from the TPW/Ga calibrated NTC thermistors (http://www.mnv.com.sg/wp-conte…/2015/01/items-system.png) your total system uncertainty would be just under 4 sigmas from rejecting the null hypothesis. For a standard P-value of 0.05, you would be exceeding by a large margin and probably could even reach P=0.01


    THH,


    1) 3 kinds of heat transport: Conduction, Convection, and Radiation.


    2) If we reduce the convection, then it gets hotter until conduction and radiation bring it back into equilibrium.


    3) Radiation goes as T^4, so that using radiation works to keep unit temperature more stable than relying on conduction and convection.


    4) There is always a chance of thermal runaway, but as long as for each marginal increase in temperature (say of 1 degree) the heat transport away by radiation, conduction, and convection increases faster than the excess heat generated by the reaction, the reactor will be thermally stable.


    5) Insulating the unit further or reducing the convection makes it more unstable, but the stability margins can be calculated (as an engineering exercise) once the effect is proven to work stable within the lab enough that basic measurements can be done.


    6) Brillouin's E&M field pulse control mechanism appears to immediately increase or decrease rate of reaction excess heat, so would be more stable than a pure thermal control system (assuming it works and is proven).


    7) Although you get frustrated, I appreciate your input to the discussion. Please stay with us. I encourage everyone here to be extra polite and professional in using this blog to avoid losing contributors. It takes a few extra minutes to clean up your comments.

    Quote from JedRothwell

    He has been replicating this for years. From reactor R16 to R20. Repeatability is excellent. Someone else in Japan already replicated, and reports it is working. Other people need to try this now. People skilled in the art.


    I do not think anyone should prepare a reactor, put deuterium into the reactants, and mail the reactor or reactants to anyone. I fear the reactants might self-heat and go out of control. It may be unlikely, but if it happened it could cause a catastrophic accident in an airplane flying from Japan to the U.S. We should not take any chances until this reaction is fully understood. I am confident that the reaction can be made safe and controllable, but that will take billions of dollars of R&D. Right now, mailing prepared and deuterated reactants is like mailing a fully charged battery or a can of gasoline. That's dangerous, and it is against postal regulations.

    Jed I don't think such fear is well placed. A self-sustaining reaction would continue to heat the nickel mesh up to its melting point and above 700C the few grams of Ni would simply melt and the reaction would stop. Assuming packing for shipping was done properly, the worst that would happen is that a non-working reactor would arrive at its destination. Irregardless of this I agree that out of an abundance of caution charged units should never be transported.

    Quote from Daniel_G

    Jed I don't think such fear is well placed. A self-sustaining reaction would continue to heat the nickel mesh up to its melting point and above 700C the few grams of Ni would simply melt and the reaction would stop. Assuming packing for shipping was done properly, the worst that would happen is that a non-working reactor would arrive at its destination. Irregardless of this I agree that out of an abundance of caution charged units should never be transported.

    There's a good chance that plating the pall with something like hematite can help feed and use aluminum amalgam to help sustain equilibrium, but still a long way away.


    All correct anonymous. The setup as described is mostly convection cooled via airflow. Radiation cannot dissipate much - given the insulation - unless outside of insulation gets v host, which presumably it does not. So when the airflow does not cool it the main loss will be conductive - with perhaps a radiative channel between casing and insulation (foil on inner side of insulation will reduce radiation but not completely). But that air gap transfer is in series with the insulation thermal resistance that is definitely conductive (and therefore constant thermal resistance).


    This complex combination of mechanisms is going to give roughly linear power / casing temperature curve deltaT = K(Pin + Pout)


    Stability then comes from solving this curve with the output power curve. in terms of control theory you would view these two functions as being the forward and feedback gain.


    Pout = F(deltaT) where it appears F(0) = 0 and F is a power law (M says exponential) increasing function.


    This is stable at deltaT = T0 if the loop gain:


    KF'(T0) < 1


    Let us take a simple example of F(T) = CT (linear)


    Stable if KC < 1. In that case K(Pin+CT)=deltaT => deltaT = KPin/(1-KC), COP = 1/(1-KC)


    You can see that the COP increases rapidly as KC gets closer to 1. Although C is a function of the reactor, K can be changed by altering the insulation in the calorimeter.


    Also note that a temperature dependence of the type Mizuno states would mean KF'(T0) getting larger as the T0 gets higher, becoming unstable with runaway at some temperature.


    The temperature at which runaway occurs can then be moved up or down by decreasing or increasing insulation (count decreasing airflow as increasing insulation).


    For R20, with COP 6 - 10 we have KC close to 1 (actually it is the dynamic COP that matters, but not so much difference for the dependence curve M gives) so it is close to unstable.


    If however you wrapped a reactor temperature body control feedback loop round this system you could get unconditionally stable behaviour. The reactor casing heat capacity would provide a dominant pole to keep things smooth and any sort of control (e.g. bang-bang) would work fine. Think of it as being like a thermostatically controlled radiator except it is the fan speed that is controlled not the power input.

    Quote from THHuxleynew

    My point above is that the reactor casing temperature (and therefore the gauze temperature) can be controlled by altering the airflow, as well as altering the heater power

    Yes, the Mizuno experiment can be analyzed in the same way as semiconductor packaging and cooling. This is usually expressed by the thermal resistance (in degC per Watt) from junction to case (theta-jc) and from junction to ambient (theta-ja). Theta-ja = theta-jc + theta-ca.


    Theta-ja depends on the thermal pad, heatsink, and airflow across the heatsink. (In most cases, cooling by conduction and radiation are so low in comparison that they can be neglected.) The main design choices are in the heatsink design (area exposed to the airflow) and the airflow provided by the fan. The same junction temperature can be set by adjusting the junction power (say by changing clock rates) or by changing the cooling system to increase or decrease cooling.


    The THH suggestion to reduce airflow and heater power to keep case temperature constant seems like an excellent experiment once the original experiment has been replicated. As he suggests, it would show a range of COPs for the same experiment.

    Quote from Robert Horst

    Yes, the Mizuno experiment can be analyzed in the same way as semiconductor packaging and cooling. This is usually expressed by the thermal resistance (in degC per Watt) from junction to case (theta-jc) and from junction to ambient (theta-ja). Theta-ja = theta-jc + theta-ca.


    Theta-ja depends on the thermal pad, heatsink, and airflow across the heatsink. (In most cases, cooling by conduction and radiation are so low in comparison that they can be neglected.) The main design choices are in the heatsink design (area exposed to the airflow) and the airflow provided by the fan. The same junction temperature can be set by adjusting the junction power (say by changing clock rates) or by changing the cooling system to increase or decrease cooling.


    The THH suggestion to reduce airflow and heater power to keep case temperature constant seems like an excellent experiment once the original experiment has been replicated. As he suggests, it would show a range of COPs for the same experiment.


    Yes, and that sensitivity would be a sign that you did have heat generation independent of whether the calorimetry calculations are right or wrong.


    Furthermore, it would prevent the otherwise likely meltdown. That M has not experienced meltdown or "saturation" in R20 is a bit surprising and rather goes against the other things he has observed.

    • Official Post
    Quote from THHuxleynew

    Anyway - anyone interested in these (simple) stability issues PM me. I don't think this is a good place to discuss them - too much noise!


    @THH. Don't be a snowflake, we all love having you here. As for the noise, that is a characteristic of public spaces, they tend to foster hooliganism. But I (and the rest of the team) do like to think that worthy voices can be heard, and are to a certain extent protected.

    Quote from Alan Smith

    . the cheap way to build a modest-pressure/high vacuum system would be to use some stainless steel truck exhaust pipe- this is available in quite short pieces.


    I think this would be a bad idea. I think you need laboratory grade stainless steel. Otherwise you may bake it out and find all kinds of stuff coming out of the walls. Also, it has to be leak tight, with Swaglok valves. Suppose you take an empty reactor, and on Monday you pump it down to 1,207 Pa. On Friday it should still be very close to 1,207 Pa. Can you make truck exhaust pipe so leak tight?


    Also, do not use any pump but a laboratory grade turbomolecular one. Other kinds spew oil out and about. The pump is pulling stuff out of the reactor, so you might think that is not a problem, but eventually milligrams of oil may find their way back up the pipe into the cell, and they will clobber the experiment. Keep it clean, clean, clean.


    I recommend laboratory grade stuff only. The best you can get. It might not work even with that, but it is more likely to work.

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.