Unconventional electrolysis

  • Robert Horst

    Something worth pointing out regarding the design of such system (if these experiments are worth of it) is that depending on testing conditions and procedures used a thick black foam depositing very fine black conducting material (likely mostly iron with these electrodes) on everything it makes contact with can be produced, as can be seen in this video I posted earlier on, from about minute 0:40 :

    This appears to occur more easily when a large portion of the electrodes is immersed, the electrolyte is reused from prior tests and it's close to or above boiling temperature. It's not clear yet if it's a desirable thing, but I often observed plasma reactions to occur more easily under such conditions. More recently I've been mostly concerned with reproducing the resonating noise and I haven't seen the above effect occurring to the same extent.

  • Today I put the previously (re)made coil into test. In short, it seems that it brought visible changes in that the apparent discharge rate was much slower than in previous tests and produced louder resonant noise (see embedded video below), but it didn't seem to produce very bright discharges. The latter could have been due to experimental conditions and other differences out of my direct control.

    The text below is directly from my notes, so it might not match exactly the previous reporting style.


    2.1 Coil

    (Photo previously posted)

    Figure 1: The newly prepared coil.

    • 3.35mm outer 1.75mm inner, ~18 meters stranded wire, probably should be considered between AWG13 and AWG14. I think I recall it was marketed as 2.5mm2 wire.
      • Since I previously measured roughly 2.95 Ohms, that's 0.164 Ohm/m, higher than I thought
      • Length was measured as 36 x 50cm lengths, so it's an approximate measurement
    • This is shorter than I originally thought I had
    • Melting damage was sustained on part of the insulation which got damaged while I was undoing the previous coil, so it's compromised.
      • Hopefully this will still be fine. I will need to get new wire at some point and it will be probably worth thinking of getting proper magnet wire by then.
    • Obtained a 95-99 turns coil around a 2Kg steel dumbbell

    2.2 Other stuff

    • Electrodes
      • Same configuration as last time, not disassembled yet
      • They have been quickly washed under running hot water after the last experiment
      • Inner gap scraped and cleared with a mica spacer
      • Mica spacers on the top and the bottom of the gap respectively forming a 1.4mm and 0.4mm gap close to their position
      • One clip (insulated with folded mica spacers) at about the center to hold the electrodes together
      • Replaced insulation on the external surfaces of the electrodes on their bottom end
    • Jar
      • Cleaned with warm water and soap, then rinsed again
      • Some residues from the acidic solution remain on the internal walls and might need a deeper cleaning to be removed
    • Electrolyte
      • Starting with about 10ml distilled water
    • Instrumentation
      • Clamp meter and multimeter for real-time reference of experimental parameter
        • Will not be logged
      • AM transistor radio
        • Tuned to 1600 kHz
        • The audio output signal will be recorded at 192 kHz
      • Video camera
        • Panasonic SDR-H20 to be used as needed to produce video documentation of any interesting reaction

    Figure 2: The setup just before starting it.

    Figure 3: Closeup of the jar. It now has hard to remove deposit on its internal walls.

    Experimental notes

    • 09:35:38 Started Audacity
    • 09:36:08 Zeroed out clamp meter
    • 09:36:38 Experiment started
      • Turned PSU on
    • 09:37:01 12.07V, 0.06A
    • 09:37:08 Slight bubbling
    • 09:38:44 0.25 A
      • Current spontaneously increasing before adding HCl
    • 09:39:12 HCl added
      • Seems slow to start
    • 09:42:59 11.05A
    • 09:43:47 39.85A
      • Experiment paused
    • 09:44:40 Added HCl drop
      • Added some solution from old jar
      • Cleared gap with a mica spacer
    • 09:46:36 Restarted
      • Worked for a while at a low loud rate, but then failed and current went to 39A
      • Experiment paused
    • 09:49:15 Cleared gap
    • 09:49:35 Restarted and worked for a while, but then current went to 35A
      • Experiment paused
    • 09:57:25 Tried to run the electrodes outside jar for a while
      • Incandescent spot inside gap formed even though current is low

    Figure 4: The electrodes outside of the solution.

    Figure 5: Incandescent spot formed where presumably it is short-circuiting. Limited current though the electrodes at the time of the photo.

    Figure 6: A different view of the spot formed.

    • 10:03:23 Added water
    • 10:03:44 Restart attempt, but current went to 35A quickly
    • 10:06:00 Tried to restart outside solution, but it seems that inductor is overheating
    • 10:06:17 Seen enough for today
      • Experiment terminated
    • 10:15:43 Electrodes washed under warm water and put into another room to dry slowly


    • The rate of the discharges seemed lower on average, with a base frequency lower than 500 Hz and during a later attempt lower than about 150 Hz.
      • The associated sound was at times significantly louder than usual.
        • This makes me wonder if the louder the better or if it's unrelated with any positive effect this could have.
      • Upon closer analysis, harmonics or patterns very closely resembling harmonics clearly visible at low frequencies, can be seen even at very high frequencies (tens of kHz).
      • The lower frequency does not seem to have affected my AM transistor radio's capability of picking up this signal at about 1600 kHz as usual.

    Figure 7: Spectrum after the experiment began, with a base frequency of about 320 Hz.

    Figure 8: Spectrum with a base frequency of only 145 Hz. Many harmonics visible in the spectrum.

    • Reliability again an issue for prolonged operation (i.e. longer than a few minutes).
      • Accumulation of deposition materials into areas of low current density cause persistent short-circuits.
      • Will probably have to use o-rings and heat-resistant paint like magicsound did to prevent this.
    • No significant change in Geiger counts either up or down associated with the experiment has been observed.

    Figure 9: Geiger CPM for the latest 24 hours.

    Figure 10: Geiger CPM for the latest 2.4 hours. No significant changes observed, but possibly some brief CPS spikes occurred.


    • 001
      • The experiment since HCl was first added. It started slowly, perhaps due to adding slightly less than usual.
    • 002
      • Restart attempt. The discharge rate seemed low and the noise produced was relatively loud. At the end of the video I turn the PSU off due to a persistent short-circuit taking place.
    • 003
      • Another brief restart attempt where the discharge rate seems slightly lower than earlier.
    • 006
      • Attempting running the electrodes out of the jar. Numerous small quiet discharges occurring all over the gap inner surface can be clearly seen. About halfway into the video the aren't visible anymore due to the residual solution in the gap drying out.
    • 007
      • One of the last attempts trying to run the electrodes immersed in the solution. No significant noise production, failed attempt.
  • For the record and avoiding being called a hypocrite for not reporting this, yesterday I tried adding a couple KOH flakes to the solution (used/already prepared) and I couldn't manage to reproduce the resonant sound except for very partially until I added back relatively large amounts of HCl and replaced the evaporating water with iron chloride (I'm assuming that's what the acidic solution it mostly is) solution from another jar.

    Probably not much of general interest to report except usual experimental observations (see spoiler tag below), but here's a video of something that in retrospect might have been (slightly) more interesting than I assumed.


    The electrodes here were short-circuiting in an undesired manner outside the jar, causing an incandescent spot near a mica spacer. Upon rewatching this several times I think this was actually caused by a continuous arc and not just joule heating of a low resistance path, as it appeared to produce a sort of "glow" around the incandescent area. Still not sure, though. Anyway, since it was not operating intermittently and did not produce any RF noise at all, I don't feel this was a desired mode of operation.

    In practice this also means that if I were to install o-rings (proposed method for holding the electrodes together instead of clips) they could possibly be quickly damaged by the heat of such hot spots, which might also form when immersed in water as well.

    Experimental notes

    Misc observations

  • I'm not sure if this is the right place to inform about the traditional plasma-electrolysis (maybe everyone has already gone through the Naudin-pages) but here is one example anyway:

    Test 4

    To me the Naudin-pages are some of the best documented tests for those who want to try something.

  • Rjzk

    If I wanted or could do things traditionally (i.e. Mizuno-style plasma electrolysis) I wouldn't have called the thread "unconventional".

    The basic idea behind the tests performed in the past few weeks here has been trying to apply low-voltage, high current short pulses using readily available equipment and in the simplest manner possible. To do this I take advantage of electrodeposition through a pair of narrow-gap electrodes in order to cause short-circuits that take place at a relatively high rate due to the electrodeposited material that progressively electrically bridges both electrodes.

    I use 12V DC from an ordinary computer power supply, but since there's a coil/solenoid in series with the circuit also acting as a current limiting resistor, higher voltage spikes will be generated.

    Does this actually produce excess heat? Probably not, but perhaps things can be settled down once a test with proper input and output energy measurements will be done.

  • Rjzk

    If I wanted or could do things traditionally (i.e. Mizuno-style plasma electrolysis) I wouldn't have called the thread "unconventional".


    Sorry, my purpose was not to critisize your tests. It's nice that someone shares openly all the results to the interested ones.

    I just wanted to show that test as one example of getting good results. Maybe there is something that could be scaled down from 200V level to 12 V. In your case the max. power is ~ 120 W v.s. the Naudin case ~ 700W. And as you mention short-circuit that means that the voltage level requirement is quite low.

    One thing that has been bothering me is that already at that time Naudin could get > 2 COP with quite a simple arrangement, when e.g. for Brillouin it is hard to exceed that value with much more sophisticated system.

  • Rjzk

    A possible issue could be that Mizuno-type experiments aren't well suited for a commercial product. Putting aside industrially low temperatures for the heated fluid, I understand that one main problem is that the tungsten electrodes tend to wear up quickly, so eventually they won't be able to work as intended anymore.

    Back in 2012 Italian researcher Ugo Abundo came up with a variation which used a fluidized bed cathode (made of tungsten/tungsten-iron powder suspended in the electrolyte) which solved some of these problems. Power was conducted impulsively with discharges through the conductive slurry using rectified 120-220V DC. Later variations used deliberately pulsed power. Abundo later on went semi-commecial and apparently moved onto more complex systems that don't seem focused on heat production but I haven't followed his work closely since that.



    It would probably be easier for me if I also used a higher power supply voltage, but the point of these experiments besides testing some ideas I had was also checking out if something could be seen also using already available, zero/near-zero cost equipment and a variac just isn't in my case.

    Brillouin are passing intense, narrow current pulses through low-impedance hydrogen-loaded dry cores heated to a few hundred C, which supposedly can be used in an industrial setting and won't quickly wear down as in (dusty) plasma-based systems. I think they used to have electrolytic/"wet" systems but I haven't checked those early systems in detail.

    It seems that regardless of the experiment type, narrow intense pulses are part of what enables excess heat observations, but it's true that this means one has to be careful with input power measurements and the overall efficiency of the system before envisioning a cold fusion-powered world. On the other hand, there's a risk of throwing away interesting results by only considering the overall efficiency as worth of study.

  • magicsound

    Thanks for your continued efforts. Any word on your planned changes compared to last time? Earlier you mentioned about using zinc-plated mild steel electrodes.

    I tried making a web image search and I realized that mine initially had an appearance similar to this:

    I can make a photo tomorrow of fresh samples to make sure, but here are a couple older photos I made partially showing how they looked like on the top portion that was not immersed in the electrolyte:

    I understand from this link that this typical yellowish color is due to a zinc-chromate finish: https://www.finishing.com/113/53.shtml

    For the record, following early narrow-gap electrolysis tests in KOH electrolyte at the end of last November where I observed the electrodes to cause persistent outgassing (possibly Zn reacting with KOH forming H2 as you pointed out at that time) even after removing power, I completely removed such layer by immersion in liberally diluted 10% HCl (no electrolysis) for less than 30 minutes at ambient temperature. Outgassing in following tests (again using KOH electrolyte) was not observed after that.

    All recent tests with electrolysis-discharges in HCl solution have been made well after removing that surface layer, so if you want to reproduce the same testing conditions you might want to incorporate this preparation into your next experiment.

  • All recent tests with electrolysis-discharges in HCl solution have been made after removing that surface layer, so if you want to reproduce the same testing conditions you might want to incorporate this preparation into your next experiment.

    Yes, I did the same, but up to 1 hour was needed for the thicker zinc electroplate common on most (Chinese-made) hardware store parts.

    For the first test tomorrow, I want to see if the induced contact arcing will be sustainable while immersed in H2O. If as you surmised this results in minute particles of iron suspended in the water, formation of plasma channels might even become possible. I'll also look at the di/dt while changing the inductor core.

  • magicsound

    Some of the tests I did in the past weeks (e.g. as seen in the video I linked on the top of the page) seem to indicate that sustained arcing could be possible. As it appeared easier to observe after reusing the electrolyte from earlier testing it could mean that it's useful to have particles already suspended in some amount within it, in addition to those formed in the process. This condition could be seen as similar in a way to what Abundo did with the fluidized bed cathode in his Mizuno cell variation that I previously cited.

    I think in my case the particles might also come from the thermally decomposing iron chloride and possibly water cavitation on hot electrode portions and/or near the electric arcs formed, destroying the electrode itself and part of the loose deposition layer formed.

    A high water conductivity (which increases with temperature), also from circuit simulation, appears to works against the observation of discharges, but I suspect that using temperatures on average near or above the evaporation temperature of water makes them more easily observed due to the voids (=dielectric) formed by cavitation. Furthermore convection currents overall should help mobilizing the larger particles in the solution and to some extent inside the gap.

    Almost just for the sake of citing this, on a loosely related note, in his abandoned 2011 patent application Brian Ahern suggested that a colloidal solution of metal particles could be formed by energetic (a 50-500 mJ/pulse figure is quoted in the patent) high voltage discharges ablating electrode material in an aqueous cell.


    [...] Liquid dielectrics produce similar energy focusing capabilities as the ceramic matrices. Liquid systems provide a direct method for producing nanoparticles in situ. The high voltage discharges through a fluid ablate electrode materials that are rapidly quenched and suspended in the polar fluid. Once formed, the nanoparticles can be hydrated/deuterated by the ionization of the water during the discharges. As such, the high voltage pulses fill the H2O/D2O volume with a constellation of suspended particles filled with interstitial hydrogen (H)/deuterium (D) atoms. The particles stay in suspension due to Coulomb Repulsion as each particle is surrounded by polar water molecules that attach the oxygen to the metal cluster surface and has the two deuterium atoms from the D2O molecules facing out. The deuterium atoms have a net positive charge associated with them, so each metal cluster looks like a large positive ion that repels all the other such clusters. The nanoparticles remain equally spaced in the dielectric liquid due to this repulsion process that is very effective at small mass/charge ratios. The suspension of the nanoparticles in the polar water medium is referred as a colloidal suspension.

  • A possible issue could be that Mizuno-type experiments aren't well suited for a commercial product. Putting aside industrially low temperatures for the heated fluid, I understand that one main problem is that the tungsten electrodes tend to wear up quickly, so eventually they won't be able to work as intended anymore.

    The tungsten electrodes disintegrate in about 15 minutes. This could not possibly be made into a practical source of energy. See:


  • First test results with the cell empty of liquid. With the inductor steel core removed, peak-to-peak voltage of 538 volts is seen, with resonant ringing at 5.8 MHz. With the core inserted, the peak voltage is slightly lower and the ringing frequency is higher, at 6.4 MHz and is more quickly damped. This is counter-intuitive, as higher inductance should result in lower frequency.

    The electrodes latched shortly after these samples, and the circuit breaker popped at a peak current of ~48 amperes. Some smoke escaped from the current-limiting resistor, but no permanent damage was done. I'll reduce the circuit breaker from 40 amps to 30 amps for further protection, then resume testing with some water in the cell.

  • magicsound

    Do I understand correctly that you brought the electrodes in slight contact in a controlled manner, producing a resonant (self-oscillating?) electrical arc for a brief period of time until basically the electrodes welded together? (which would mean that I previously misunderstood what you planned to do as your first tests today)

    For what it's worth, a few months ago I tried contact separation tests using 5V, a similar inductor as the one I recently prepared and the previous chinese-made ATX power supply. Interestingly the presence of very limited amounts of distilled water in the contact region appeared to make the plasma formed brighter, but alkaline impurities (e.g. when using tap water) quickly damped the effect.

  • Do I understand correctly that you brought the electrodes in slight contact in a controlled manner, producing a resonant (self-oscillating?) electrical arc for a brief period of time until basically the electrodes welded together?


    Not exactly welded together, but stuck in a conducting state for long enough to pop the circuit breaker. The driving signal to the linear motor is a high pass filtered square wave, resulting in a pulse to separate the electrodes immediately after the one pushing them together. Increasing the drive pulse amplitude seems to effectively prevent the latch-up from happening, but may reduce the useful life of the mechanism.

    I've now reduced the circuit breaker, from 40a to 30a. The extreme transients I measured were also causing an over-scale error from the power analyzer, so I installed a .01 uF HV cap across the current shunt, which seems to have solved that issue. But the high peak current pulses are too short for the instrument to measure accurately, even at 1 meg sample/sec.

    Other issue are also appearing - the radiation measurements are being affected by RF from the arc, so I will probably need to put the entire cell apparatus in a Faraday cage! I'll be posting a video illustrating this problem in a bit. Once that is done I will put some water in the cell and crank it up just to see what happens.

  • Interesting; definitely worth making sure if it's just radio noise.

    Yes' I'm pretty sure. In the test just done I put a battery-powered GMC directly behind the cell, and it showed nothing above background, while the pancake detector in front of the cell showed about 5x background. So the first thing I need to check is grounding problems in the DAQ system.

    For this test I added 150 ml of deionized water to the cell. The arc in water is still bright blue, with none of the yellowish color seen in tests with HCl. What is interesting is that the waveform at the electrodes was substantially changed by the addition of water. Most of the time it looked like transient2 below. But occasionally something different happened - see transient3a.

    I took a close look at that one with the scope's analysis tools. The anode voltage initially went up to 524 volts with the characteristic curve seen in transient2. Then it abruptly dropped to negative 242 volts in just 36 nsec. Based on the traces in transient2b and c, that's a slope of -21kV/usec, extremely fast. It would be very hard to do this with a semiconductor switch, and it's not surprising that it can generate lots of RF. By integrating the area under the rising part of the curve and estimating the inductance of the coil, it might be possible to calculate the energy involved in this event.

    There's still a tendency of the electrodes to stick together, not surprising given that the clacker resembles a spot welder. Maybe some electrolytic pre-treatment would build up a layer of oxide enough to inhibit this tendency. So that will be my next step on Monday.

  • magicsound

    I haven't performed any particular calculation yet, but attached are the digitized versions of Transient2 and Transient3a.

    Good to have narrowed down the issue purely to a RF noise issue; so far I haven't had any. Arc discharges produced by rapidly putting in contact the electrodes in deionized water will likely be more intense on average than under typical conditions during the experiments with electrodeposition-short circuits in an acidic solution.

    At this point your latest test seems more similar to the contact-arc tests that I used to do quite some time ago or to Parkhomov's Woodpecker device.

    In that case:

    • Using graphite electrodes will prevent or strongly mitigate (depending on the impurities present) sticking issues;
    • A strong magnet strategically placed in proximity of the moving electrode might be able to pull it back when the electrode makes conduction with the other, producing a strong magnetic field due to the large currents involved;
    • As the water will become more conductive over time you might notice a decrease in discharge intensity.

    It might be interesting knowing if by increasing voltage the frequency of the RF signal increases and viceversa.

  • At this point your latest test seems more similar to the contact-arc tests that I used to do quite some time ago or to Parkhomov's Woodpecker device.

    I did think of the Woodpecker along the way, but my primary goal was (and remains) investigation of what's happening with your device. The important implication of what I saw is that there are apparently two distinct modes of wet arc discharge. Transient2 looks like a fairly long-lived contact event where most of the energy stored in the inductor is carried off through a relatively long (millisecond?) metal-to-metal connection.

    The transient3 form suggests rather large current density, far greater than a sub-mm2 contact could sustain. One other curious fact: the ringing seen in my earlier test is absent from both these wet discharge modes. Water has a dielectric constant of 80, so self-capacitance of the electrodes could become a factor at these short time scales.

  • magicsound

    A probably obvious observation could be that the shorter mode arises when the previous contact-arc events leave thin protrusions that can't support large amounts of current in the next one, which would make it interesting to check the behavior when a large amount of fine particles are present/put in the contact area, while still using distilled/deionized water. This might be easier to test by constraining the active volume with a test tube (e.g. immersed in the jar) rather than using the entire jar.

    I wonder if the ringing being present only under dry conditions is somehow related with me hearing the resonant acoustic noise more easily when the electrodes are only partially immersed in the solution or just wetted with the acidic electrolyte. However, so far my AM radio seemed to pick up resonant RF noise only when the electrodes are partially immersed in water, even if a similar noise could be heard acoustically outside of it. (possibly this could be related with broadband noise overpowering the resonant noise on the AM radio under those conditions; it's not exactly a precision instrument).