Do it yourself experiment

The Enhanced CFR replication

created on September 12, 2005 - JLN Labs - March 21, 2006
Toutes les informations et schémas sont publiés gratuitement ( OpenSource ) et sont destinés à un usage personnel et non commercial
All informations and diagrams are published free (OpenSource) and are intended for a private use and a non commercial use.

The CFR project is a High Temperature Plasma Electrolysis fully based on the Tadahiko Mizuno experiment from the university of Hokkaido in Japan. This is a very interesting experiment and its implication can be a real breakthrough in the field of New and Clean energy source....  

You will find below an explanation to replicate yourself the enhanced CFR experiment. With such a design you will be able to conduct more accurate measurements with better efficiency and long tests runs.

The enhanced CFR is composed of a 2000 mL thermostatic reaction vessel filled with 800 mL of demineralized water and Potassium Carbonate ( K2CO3 ). The electrolyte solution commonly used is 0.5 molar ( 0.5 M ).

There are three temperature probes ( K probe or PT100 ). Two probes are used for measuring the temperature of the cooling water (Temp In and Temp Out ),
and one probe is used for measuring the temperature of the electrolyte solution. You need also to use a flowmeter to measure the cooling water flow.

The Cathode used is a 4 mm tungsten rod. The tungsten rod can be a pure tungsten rod or a Th-loaded tungsten TIG electrode (WT20) (with thorium oxide ThO2: 1.70% to 2.20% ) commonly used for plasma welding. The use of a Th-loaded rod increases the life of your cathod. The sputtering effect produced by the thermionic emission is lower with a Th-loaded rod than with a pure tungsten rod.

The anode used is composed of stainless steel mesh ( a grid ) maintained with stainless steel  rods. If you have planned to conduct some chemical analysis, I recommend you to use a grid made with platinum or nickel . All the wires connections are made with a 1.5 mm2 copper flexible wire gained with silicon.

I recommend strongly to use a high temperature ceramic insulator (ie: alumina) around the cathod. You may use a common ceramic fuse (ie: a 25A/220V fuse) maintained by a silicon tube as shown below :

To avoid projections of some drops of the electrolyte solution from the CFR during the plasma ignition sequence, I recommend you to put floating balls on the surface of the liquid (hollow floating balls; pp, 20mm, 2000 PK from Cole Parmer Instrument ).


Use a DC Power Supply which is able to give about 300 V DC at 20 A ( don't use AC voltage ). The voltage is tuned with an autotransformer (see below)

Disclaimer : The author assumes no liability for any incidental, consequential or other liability from the use of this information. All risks and damages, incidental or otherwise, arising from the use or misuse of the information contained herein are entirely the responsibility of the user. Although careful precaution has been taken in the preparation of this material. Be Carefull, you must conduct this test in a well ventiled room or better in open air, you must not smoke during the test. This experiment is not intended for the inexperienced. User of this document should be very carefull to try anything out ! If you do it, the risk of any results is just yours. I take no responsibility of anything that might happen, let it be of a wrong information or anything else.

Safety :

Test procedure :

  1. Switch on the fume hood,
  2. Set the autotransformer to 0 Volt and switch on the power supply,
  3. The voltmeter (set on DC) is connected at the input of the CFR cell and the DC current clamp is connected on the positive wire,
  4. Turn the knob of the autotransformer so as to get 30V DC on the CFR cell,
  5. Let the electrolysis warm up the cell up to 50°C,
  6. At 50°C drop the voltage to 0 Volt and switch off the main power supply,
  7. Wait 30 sec to exhaust the mixture of hydrogen/oxygen,
  8. Measure the temperature ( TSinp ) of the input of the cooling water,
  9. Measure the temperature ( TSout ) of the output of the cooling water.
  10. Measure the temperature ( Tr_initial ) of the electrolyte then, immediatly, switch on the power supply,
  11. Slowly, turn the autotransformer knob so as to get 200 VDC across the cell. Start the chronometer,
  12. Note the Voltage and Current values at the permanent regime (pink glow discharge),
  13. End the run after ~3 minutes. ( set the voltage to 0 Volt and switch off the power supply ). Stop the chronometer (time).
  14. Measure the temperature ( Tr_final ) of the electrolyte,
  15. Measure the flow of the cooling water (Flow in L/min),
  16. Measure the temperature ( TEinp ) of the input of the cooling water,
  17. Measure the temperature ( TEout ) of the output of the cooling water.


Energy INPUT ( Electrical, DC Voltage )

Winp (J) = Voltage(V) * Current(A) * time(s)

Energy OUTPUT (Thermal )

Wcooling (J) = 4.18 * Flow(L/min) * (1000/60) * time(s) * ( ((TSout+TEout)/2) – ((TSinp+TEinp)/2))

Wreactor (J) = 4.18 * Melectr(g) * ( Tr_final – Tr_initial )

Wout (J) = Wcooling (J) + Wreactor (J)

Efficiency (COP, Coefficient Of Performance )

COP = Wout / Winp

Important note : With such a design, a COP up to 2 and more can be obtained. If you conduct yourself the experiment, you will notice that the value of the COP will increase with the number of test runs. When the tungsten rod is brand new, the COP is always less than 1 but becomes greater than 1 when the tip of the tungsten rod becomes thinner with some small longitudinal cracks visible on it.

I wish you a good CFR replication experiment and a great COP,
I shall be glad if you could sent me some photos of your working Cold Fusion Reactor....

For more informations, please contact Jean-Louis Naudin :

Documents and references :

return to the CFR experimenters page

visitors since September 12, 2005