Replication of Jean-Louis Naudin's Replication of the Mizuno Experiment
Earthtech International
Scott Little and George Luce
15JUL03

INTRODUCTION

At the 7th International Cold Fusion conference in Vancouver in 1998, Ohmori and Mizuno reported¹ on a new electrolysis experiment in which a W cathode becomes incandescent under certain conditions. They made preliminary calorimetric measurements on cells operating in this mode and concluded that the phenomenon produced significantly more heat energy than the electrical energy required to stimulate it. 

During 1998-2000 we made an extensive effort to replicate their experiment².  Despite having the full cooperation of Dr. Mizuno, including his providing several W cathodes to us, all of our experiments failed to show excess power production.

Recently, Jean-Louis Naudin has revived this experiment and has published extensive and detailed reports of his work on his website³. Naudin's results indicate substantial excess power production with power output/input ratios typically around 1.6 but sometimes exceeding 2.5.

Stimulated by Naudin's apparent success with the experiment we embarked upon a campaign to replicate his work.

APPARATUS

Components are as follows:

Power Supply: A manual safety switch (not shown) assures that power is off during work on the test cell. Application of power during experimental runs is by computer-controlled solid-state relay. The switched mains power is applied to a 20-ampere variable autotransformer that is connected to a 3 KVA 120 to 240-volt step-up isolation transformer. The output of the isolation transformer is connected to a 25-ampere full-wave bridge rectifier (type BR2510) followed by a 3000 microfarad 450 volt electrolytic filter capacitor. The measured ripple is approximately 4.7 volts peak-to-peak at 250 volts with a 2.5-ampere load, and approximately 8.6 volts p-p with a 5-ampere load.

Power Meter: The DC output of the power supply is connected to a Clarke-Hess Model 2330 Watt Meter⁴, which continuously measures voltage, current, and power. The cell is connected to the power meter through a shielded cable. The power meter readings are transmitted to the data acquisition computer through a serial cable. Previous experiments using the Clarke-Hess power meter have shown it to be accurate and reliable on a wide variety of electrical waveforms⁵.

Stirrer: Fluid in the test cell is placed in gentle motion by a small stirrer, consisting of a 6.3 mm diameter fiberglass rod carrying a 38 mm long rod of 3 mm diameter Teflon placed in a cross-drilled hole about 6 mm from the bottom end of the fiberglass rod. The fiberglass rod is driven at 770 RPM by a small motor. Power input to the motor is approximately 800 milliwatts. The stirrer greatly improves the temperature uniformity in the solution.

Temperature Monitor: Fluid temperature is monitored with an Omega Type 450-AKT temperature indicator⁶ and a Chromel-Alumel (type K) thermocouple. The thermocouple is mounted on the same support rod as the separate thermistor probe. The thermocouple readings are only used to monitor the solution temperature before starting an experimental run. During experimental runs the data acquisition computer reads the thermistor probe.

Digital Scale: The test cell assembly sits on a Sartorius Model LC6200S digital scale⁷. The scale is connected to the data acquisition computer through a serial cable. The mass of the cell assembly is measured continuously. The scale has a resolution of 10 milligrams.

Data Acquisition Computer: A 386-40 computer using the DOS 6.00 operating system is the basis of the data acquisition system. Interface to the experimental arrangement is by means of a Computer Boards Model CIO-DAS1602/16 data acquisition card⁸. A QuickBasic program, written specifically for these experiments, provides the apparatus control and data processing.

Test Cell: The test cell consists of an anode, a cathode, and a vacuum-insulated glass Dewar flask. The anode is a 58 mm diameter, 44 mm tall, 3.75-turn helix of 1.58 mm type 308 stainless steel welding rod. The cathode is a 178 mm long, 3 mm diameter 2% thoriated tungsten rod. A sheath of Teflon or silicone rubber is placed over the rod so that approximately 2 cm of bare tungsten is exposed to the cell solution.

The anode, cathode, temperature probes, and stirrer mechanism are suspended from above the cell vessel by supports held in a laboratory ring stand. Nothing touches the cell vessel. The cathode mounting can be adjusted to compensate for the erosion and shortening of the exposed end of the cathode. The electrode sheath is trimmed as needed to provide the approximately 2 cm exposure of bare tungsten.

The thermistor probe, inside a protective thin-walled glass tube, is also suspended from the ring stand. The separate temperature monitor thermocouple is held adjacent to the thermistor tube.

The stirrer shaft, with its Teflon crosspiece is positioned so that it does not touch the cell elements or the cell vessel. The stirrer crosspiece is located below the anode helix.

The cell vessel is a commercial "thermos" flask of nominal one-liter capacity, with an O.D. of 102 mm, an I.D. of 88 mm, and a depth of about 240 to the center of its hemispherical bottom. The wetted portion of the interior cell wall has a mass of approximately 100 grams, determined from disassembly of a similar vessel. The cell vessel rests in a cushioned plastic ring. A removable wooden support vertically positions the vessel such that the anode and cathode are well immersed in the cell solution. This arrangement facilitates removal of the cell vessel for cleaning and for access to the cell elements. 

\

This photo shows the experimental setup described above.  The Clarke-Hess Watt Meter is in the lower-left corner.  The autotransformer and step-up transformer are just above it.  The rectifier and filter capacitor are in the fume hood on the left.  The cell and associated apparatus are in the center of the fume hood (note the Sartorius balance under the cell). The Omega temperature monitor is just to the right of the balance.   The stirrer power supply is in the fume hood on the right.  The data acquisition system is on the far right.  For this photo, we employed a transparent cell vessel so the glowing cathode would be visible.  However, the results presented below were obtained with a Dewar flask which can be seen sitting atop the stirrer power supply.

In this photo you can see the W cathode (center) surrounded by the stainless steel anode helix.  On the right is are the two temperature probes (thermistor in glass jacket and bare K thermocouple).  The Teflon stirrer and its fiberglass shaft are just to the left of the cathode.

PROCEDURE

The cell fluid is a 0.2M solution of K₂CO₃ in distilled water. The vessel is filled with approximately 600 milliliters of electrolyte solution to an index mark on its interior wall. This amount is sufficient to assure that the cell elements remain completely immersed during each experimental run. When needed, distilled water is added to bring the solution mass back to the nominal 600 gram value before starting a new run.

 Runs consist of a manually controlled preheat phase and computer-controlled heating, boiling, and cool-down phases.

 Because the temperature of the cell solution is close to room temperature when fresh solution is being used or when no run has been recently made, we have found it appropriate to preheat the solution to approximately 80 degrees C. This preheating is not a part of the energy determination. The stirrer is placed in operation before heating the solution. Starting at low power, the cell voltage is manually adjusted to keep within the current limitations of the power supply and the measuring equipment. When the desired starting temperature is reached, the manually applied power is turned off and the data acquisition is ready for a computer-controlled run. The variable autotransformer is manually reset to a position, which will provide the desired operating voltage.

 The computer-controlled run consists of a starting delay of 30 seconds to establish a baseline. The computer then switches on the mains power to the DC power supply. The plasma mode is established at this point, and the solution begins heating from the starting temperature to a value near the boiling point. As the run proceeds, a portion of the solution is boiled away.

 The data acquisition program controls the run length and measures cell current, voltage, power, mass, temperature, and room temperature. The program also records the date and time, calculates total cell input energy, and plots all variables to a screen display. Measurements are written to a disk file unique to each run, which permits later playback and analysis.

 After the power-on phase of the run is terminated, the program continues to measure cell parameters until the cell solution has cooled to the initial starting temperature. This last portion of the run is included so that we can compare our results to those obtained by Jean-Louis Naudin.

 Upon completion of the entire run, the program calculates and displays the measured temperatures, input energy, the output energy, and the cell efficiency. Except for a manual notation of the dial setting of the variable autotransformer, all control and computation functions are automatic. Later sections of this report discuss the experimental protocol. Some examples of the computer screen displays are given.

RESULTS

This is the screen display of a typical run.  The horizontal axis is time (1 min/div). Several parameters are plotted on the y-axis. In each case, the trace is color-coded to the numerical display above, which contains the vertical scale information. For example, Mass is plotted in red. In the Mass numerical display the designation (1300/1200) after the numerical value gives the range of the vertical axis for that parameter -- the bottom of the vertical scale is 1200 grams and the top is 1300 grams.  

Note that some of the numerical data in this display are "instantaneous" values and therefore represent the conditions at the very end of the entire run. Other data, such as T1 and T2, are "captured" values which are recorded by the program and used in the calculations.

A run begins with 30 seconds of quiescent time to establish baselines.  Note, for example, that cell temperature (yellow trace plotted on a scale from 50 to 100 C) starts at about 78 degrees and declines slightly during the 30 second quiescent period.  The vertical green line (Pin) represents the beginning of the power-on phase. In this run power was on for 270 seconds during which time the cell temperature rose to around 98 degrees. Note the white tick-marks along the horizontal axis.  These mark the times at which key parameters are recorded for use in the heat output calculations.

We made 15 runs using the Dewar flask. The results of these runs are tabulated below:

TABLE 1 -- MEASURED PARAMETERS

Note: Time = Second; Energy = Joules; Temperature = Degrees Celsius; Mass = Grams

The silicone rubber sheath was periodically trimmed to maintain the exposure of ~2 cm. tungsten. Distilled water was added to maintain the initial solution mass at approximately 600 grams. The runs in which the stirrer was turned off show irregular temperature decrease during cooldown. 

Plasma Duration is the length in seconds of the power-on portion of the run.  

Cell Voltage is the initial voltage applied to the cell to initiate the plasma.  Due to the internal impedance of the power supply, this voltage rises slightly as the cell current decreases.

Input Energy is the time integral of the instantaneous power readings by the Clarke-Hess Watt Meter:

Energy Input = S (Cell Current)*(Cell Voltage)*DTime

This parameter is calculated continuously during the run.

DTime is the time increment of the successive instantaneous measurements.

Initial Temperature is the temperature of the solution at the time of applying cell power.

Final Temperature is the maximum temperature of the solution during the power-on portion of the run.

Initial Mass is the solution mass at the time of applying cell power.

Final Mass is the solution mass at the time the cell power is turned off. We make this measurement shortly after power-off to allow settling of any vibration, for a more accurate weight determination.

Mass After Cooldown is the solution mass at the time the solution temperature cools to the Initial Temperature.

TABLE 2 -- VALUES CALCULATED FROM MEASURED PARAMETERS

Note: Time = Seconds; Energy = Joules

The values shown are those calculated by the QuickBasic data acquisition program during each experimental run. Note that the mass terms used in the computations are net of vessel and vessel support mass.

The columns in Table 2 are calculated as follows:

E Heating is the measure of the energy required to raise the solution from a lower temperature to a higher temperature. E Heating is calculated from the measured values of solution temperature and mass:

E Heating = (Initial Mass)*DT*4.186,

where the Initial Mass is the solution mass in grams at the start of the power-on portion of the run, and DT is the change in solution temperature from the initial value at power-on to the final temperature at the end of the power-on portion of the run. The constant 4.186 is the factor for conversion of gram-degrees to joules.

E Boiling is the measure of the energy required to vaporize the solution, converting the water at 100 degrees to steam at 100 degrees. E Boiling is calculated from values of solution mass during the power-on portion of the run:

E Boiling = (Initial Mass - Final Mass)*540*4.186,

where Initial Mass is the solution mass at the start of the power-on portion of the run, and Final Mass is the solution mass at the end of the power-on portion of the run. The constant 540 is the heat of vaporization of water in calories per gram.

E Glass is the measure of the energy required to raise the glass of the cell vessel from a lower temperature to a higher temperature. E Glass is calculated from the mass of the wetted portion of the cell vessel and the temperature of the solution:

E Glass = DT*(Wetted Portion Mass)*0.19*4.186,

where Delta Temperature is the change in solution temperature, from the initial value at power-on to the final temperature at the end of the power-on portion of the run. Wetted Portion Mass is the mass of the wetted portion of the cell vessel, as determined from disassembly and weighing of the corresponding part of a similar vessel. (This was found to be very close to 100 grams.) The constant 0.19 is the heat capacity of the glass (taken here as borosilicate) in calories per gram.  The use of the mass of only the wetted portion probably underestimates this energy because other parts of the vessel are heated to some degree by the reaction.

E Conduction is an estimate of the heat energy conducted and radiated from the cell during the power-on period, using data from the cooldown period. The average heat loss rate during the cooldown period is computed using the solution mass, change in temperature, and cooldown time. The average heat loss rate due to evaporation during the cooldown period is then computed from the change in solution mass and the cooldown time. The latter is subtracted from the former to yield the average heat loss rate due to processes other than evaporation, i.e. radiation and conduction. This difference value is then assumed equal to the average heat loss rate during the power-on period. It is multiplied by the duration of the power-on period to yield E Conduction.

 E Gas is the measure of the energy required to dissociate a given mass of solution. E gas is calculated as the time integral of the product of the instantaneous cell current and the dissociation voltage constant for water:

 E Gas = S (Instantaneous Current)*1.48*DTime,

where the Instantaneous Current is measured by the Clarke-Hess Watt Meter, and the constant 1.48 is the dissociation voltage of water. DTime is the time increment of the successive instantaneous measurements. The summation, S, of the instantaneous calculations is the time integral of dissociation power.

 E Cooldown is the measure of the energy lost by evaporation after the power is turned off.   E Cooldown is calculated on the basis of the loss of mass during the time that the solution is cooling from the Final Temperature back to the Initial Temperature:

 E Cooldown = (Final Mass - Mass After Cooldown)*540*4.186,

where Final Mass is the solution mass at the end of the power-on portion of the run, and Mass After Cooldown is the solution mass at the end of the cooldown period when the solution temperature has reached the Initial Temperature. The cooldown period follows the power-on period.  E Cooldown is included in this analysis to permit comparison of our work with that of Jean-Louis Naudin.

COP is calculated as the ratio of Output Energy to Input Energy:

COP = (E Heating + E Boiling + E Glass + E Gas)/Input Energy

 COP2 is calculated in the same manner as COP, but with the inclusion of E Conduction:

 COP2 = (E Heating + E Boiling + E Glass + E Gas + E Conduction)/Input Energy

 COP w/Cooldown is also calculated in a similar manner, but with the inclusion E Cooldown:

 COP w/Cooldown = (E Heating + E Boiling + E Glass + E Gas + E Conduction + E Cooldown)/Input Energy

 COP w/Cooldown is included in this analysis to permit comparison of our work with that of Jean-Louis Naudin.

The results are discussed in the sections that follow.

DISCUSSION

Measurement of the electrical input energy, while by no means a trivial part of this experiment, is a straightforward process involving a single trustworthy instrument with a proven track record of accurate measurements^2,4,5.   Measurement of the output energy is not so simple.   The basic strategy is to first record various heat-related parameters of the cell, then immediately apply electrical power to the cell for a certain period, and then immediately record the heat-related parameters at the end of the power-on period.  The total output energy can then be calculated from the changes in those heat-related parameters caused by the input energy.  

To understand this strategy completely, please consider an ideal cell which is closed (no evaporation) and perfectly isolated from its surroundings.  The input energy would go entirely into raising the temperature of the solution.  With a known mass of solution (of known specific heat), we would only need to record the initial temperature and final temperature (immediately after power-off) and the resulting heat energy would match the input energy perfectly.   If we now open the ideal cell so that evaporation can occur, then some of the input energy goes into heating the solution and the balance goes into evaporating water.   In this case we would need to measure initial solution mass and initial temperature, then apply the input energy, then immediately measure final solution mass and final temperature.  The heat output calculation now involves the sum of two parts, (1)  the product of the heat of vaporization and the evaporated mass and (2) the heat required to raise the solution from initial temperature to final temperature.  

In a real experiment, the cell is not perfectly isolated from its surroundings, the cell vessel has a non-zero heat capacity and, in this particular case, some of the input energy goes into dissociating water into H₂ and O₂.   As described above we have made an effort to compensate for several of these non-ideal effects.   We recognize that there are other effects that have not been addressed such as, (1) the specific heat of 0.2M K2CO3 is not exactly 4.186 joules/gm*K, (2) there are other items being heated by the input energy such as the anode, cathode, stirring rod, temperature probes, and more glass than the wetted portion counted above.  However, we feel certain that the sum of these additional effects is small. amounting to no more than a few percent of our total input energy.

Our final results, including corrections for the measured and estimated losses, are presented in the column labeled COP2 in Table 2 above.  The values are very close to 1.00.  In fact, the average COP2 of all the runs (excluding 9, 10, and 11 when the stirrer was off) is 0.98.

Cooldown:

The last column in Table 2 lists COP values obtained using the technique employed by Naudin⁹.   He adds to the total output energy the energy required to vaporize the water lost by evaporation during the cooldown period.  We do not understand his justification for this addition.  The heat energy that drives the evaporation during the cooldown period has already been counted.  It is the quantity E Heating described above.  To include any portion of it again is double-counting, which erroneously inflates the COP value.

CONCLUSION

Our experiments show no sign of the large excess power levels observed by Naudin.  In fact, our final results (COP2) average very close to unity, which indicates that there are no anomalous energy-producing reactions in this experiment.

REFERENCES

1  "Strong Excess Energy Evolution, New Element Production, and Electromagnetic Wave and/or Neutron Emission in Light Water Electrolysis with a Tungsten Cathode", T. Ohmori, T. Mizuno, p. 279, Proceedings of ICCF-7, Vancouver, B.C. April 19-24 1998

2   See numerous reports of our efforts on the Mizuno experiments at www.earthtech.org

3   Jean-Louis Naudin's website is http://jnaudin.free.fr

4    See http://www.clarke-hess.com/2330.html

5   See, for example, http://earthtech.org/capabilities/vwfc

6 http://www.omega.com/pptst/450_HANDHELD.html

7 http://www.sartorius.com

8  http://www.measurementcomputing.com

9  private communication with J.L. Naudin