Unconventional electrolysis

  • To check if my logging system is working correctly I tried taking a couple photos to calculate a longer term CPM average from the displayed values on the Geiger counter (which hasn't been moved from its usual location)

    22:17:34 = 70 counts

    22:33:44 = 1576 counts

    1506 counts / 970 seconds = 1.55 CPS = 93.15 CPM average

    This seems to agree with the general trend shown in the latest values of this graph from the logging system. The daily variations appear to have changed character since I started the tests described in detail in the past few comments, although it can still be coincidental. Ambient temperatures have dropped a bit in the past few days.

    Besides this, I've been thinking about what effects would have using KOH electrolyte instead of Na2CO3 as previously suggested.

    An immediate effect would likely be that electrode corrosion will ramp up quite a bit. While this is usually an undesirable problem, in my view it would be the opposite here. I expect even more material to get removed at the interface. Furthermore, as the water in the alkaline solution will evaporate in the process and as more is added, locally high (and corrosive) KOH concentrations on the electrode surface should occur in a progressively increasing fashion.

    I tried looking at information on the subject of steel and caustic corrosion, and I collected some links.

  • For AC tests, you might try a development board with a stepper driver like the ST L6208. It can drive up to 52V, 2.8A RMS and 100 KHz.

    see: https://www.digikey.com/produc…6208PD/497-4136-ND/724253

    The ST motor development boards also have a nice motor control GUI that runs on a serial-connected PC.

    This may be a better choice than an RC motor ESC. Sometimes they have protection circuits to prevent them from driving currents to other than balanced 3-phase loads.

    But any high-current motor driver, will generate lots of EMI. Before investing in one of these, you might do some quick tests to make sure your Geiger counter is not too susceptible.

    Get a cordless drill, preferably one with a brushed motor, and then start and stop it close to your Geiger counter to make sure it does not react.

    I recently did an experiment with a scope connected via coax to a circuit in a sealed thick aluminum box. After getting funny readings, I disconnected the battery in the box and still got the funny readings. Then I noticed that flipping a light switch would cause the scope to trigger. Then when I tried the cordless drill within a few feet of the box, it went crazy with 10s of mV noise. Aluminum Faraday cages are not perfect shielding for RF, and they provide no magnetic shielding.

  • Robert Horst

    I haven't specifically tested yet how mine reacts with active power tools nearby (I could try one of these days), but it seems that these Geiger counters aren't particularly sensitive to EMI. Alan Smith posted some time back a video of a test he made with 40 kV arc discharges and the same model with the same GM tube:

    (From about minute 2:10)

    I can say from actual testing performed a few weeks ago that with low-voltage, high-current "bad contact" arc discharges at about 40-50 cm distance I'm not getting any immediate reaction from it.

    Currently I'm using the counter on a location about 3 meters away from where the tests are actually performed, so I would expect that any possible EMI problem would be greatly diminished.

    I'm keeping it that far because if there's any emission of high energy particles like some researchers are observing (e.g. Leif Holmlid, who sees muons) or if some kind of novel gaseous molecule is decaying with nuclear products in the environment, there's no real need to keep it close to the actual testing area (although a couple days ago I temporarily moved it closer to the such area and background values were about 20% higher there). I think if tritium gets produced it would too get everywhere under my testing conditions.

    As for the actual components for AC testing, I have been looking (again without any strong commitment for purchase yet) for something that could support 3~12V at at least 20A continuous and up to a 150 KHz tunable switching frequency, but these are probably unusual requirements for what these drivers are generally used for.

    They would also need strong protection against short circuiting faults and voltage spikes, which would drive costs and complexity up: the last thing I'd want is smoking an expensive driver with accidental and/or dumb mistakes.

    I guess I'll have to accept that with such drivers the experiments described here would have to be done slightly differently to avoid them.

  • How about a high current PWM LED controller?

    The MAX16818 is designed for up to 30A, 7-28V, and a switching frequency set by an external resistor from 125 KHz to 1.5 MHz. You could use poor filtering to get the PWM cycles to show up at the load.

    They have development kits, but only for low current. You would need some higher current MOSFETs to drive 20A. See:

    https://datasheets.maximintegrated.com/en/ds/MAX16818.pdf and


  • Robert Horst

    Not knowing in advance what would be the best AC frequency to use, ideally the frequency would range from a few Hz to somewhere arund 100-150 kHz. There have been suggestions of frequencies from "a few hundred Hz" to "100 kHz", but this very likely depends on the experiment type and actual configuration. As I would like to keep the equipment I might be getting flexible for usage with differing projects, LED controllers like that specific one which starts from 125 kHz would probably not be the best choice for me.

    In any case, I must stress again that I usually try to spend as little as possible (and to avoid spending anything at all if I can), so there's no need to invest too many resources at this time on searching specific components unless obvious choices (i.e. known in the field) exist. It is sufficient to have a general idea of what I need to be looking for when I'll have a clearer picture of what I actually need and decide to purchase something. The keywords "stepper motor controller" and "LED controllers" are already good enough suggestions which I didn't previously consider.

    On a different note, in the past hours I've recorded the highest peak so far for the Geiger counter in the current location. It's not clear yet if the base level has actually increased as I previously suspected or if it's just a one-off peak value.

    I will keep repeating the same experiments daily at about the same time of the day (09:00 UTC) over the next few days, in case there is indeed some sort of cumulative effect that is causing readings to progressively increase. If this is actually what is happening, eventually the recorded values should rise to a level where it's obvious that something strange is going on.

    EDIT: for those interested, in the next spoiler tag I added notes and photos from today's testing. Due to having reached the attachment number limit and not wanting to litter the forum with excessive posts, some photos have been left out, but I've left the caption I put in my notes.

  • After a few more days of testing I could not obtain any clear indication yet that the previous experiments have been increasing the background radiation level.

    I tried making a graph with a 24-hours rolling average of Geiger counter readings of the past 10 days or so to better show the long term trend and remove the 24-periodic signal. There was a kind of stepwise increase a few days ago, but it didn't increase anymore after that. The location of the counter has remained the same all along, except for a few checks after which the location was restored to its original place, returning short term readings to their prior values.

    For the sake of completeness (and to some extent, transparency), attached is a document with all the notes from the tests I've done so far in the previous days under the common theme of not immersing the electrodes in water, but just providing water/electrolyte solution as needed. I acknowledge that the tests are very crudely done.

    I'll continue similar testing after obtaining KOH and K2CO3, which might be either later today or next Monday.

  • I received the potassium electrolytes, which came with a chemical analysis sheet.


    Potassium Hydroxide flakes - KOH 90% 1 Kg

    Parameter Value Tolerance ± UM Method
    Appearance Flakes
    Concentration 90 Minimum % ASTM E291-86
    Chlorides (KCl) 100 Maximum ppm ASTM E291-86
    Iron (Fe) 15 Maximum ppm ASTM E291-86
    Sodium (NaOH) 1 Maximum %p. ISO 1550-73
    Nickel (Ni) 5 Maximum ppm ME 26011
    Carbonates (K2CO3) 0.5 Maximum %w. ASTM E291-86
    Sulphates (K2SO4) 20 Maximum ppm ASTM E291-86
    Silicates (SiO2) 50 Maximum ppm ISO 995-75


    Potassium carbonate - 1 Kg can

    Parameter Value Tolerance ± UM
    Total alkalinity (K2CO3) 99.5 Minimum %w.
    Free potash (KOH) 0.2 Maximum %w.
    Sodium (Na) 0.5 Maximum %w.
    Humidity (H2O) 0.5 Maximum %w.
    Chlorides (Cl) 30 Maximum ppm
    Iron (Fe) 2 Maximum ppm
    Insoluble in water 100 Maximum ppm
    Density, at 20°C 1.300 Typical Kg/l

    To put radiation-related issues back into perspective, I tried placing the bottles close to the Geiger counter for a few minutes, from which I logged these changes (due to Potassium-40 radioactivity). First with one bottle (KOH), then with both.

  • I did a test with KOH but I haven't noticed anything noteworthy yet (EDIT: other than it's crucial that an oxide layer forms between both electrodes in order for this kind of experiment to work). Due to possible immediate hazards I used limited amounts of it at a time, however. Also, the Geiger counter is still a few meters away since there I have weeks worth of data, and I am looking if there's anything that is increasing background measurements anyway, not trying to measure local gamma spikes. Notes and photos under the spoiler tag below.

    My suggestion (after a quick read of your very comprehensive notes) is to try potassium hydroxide. Just a hunch.

    Can you now reveal what was your hunch based on?

  • can - I suggested KOH simply because in electrolysis experiments I have done in the past it was always slightly more reactive in terms of higher electrolyte temperatures at lower currents. This could of course be merely chemistry at work - but if you look at some early papers on lenr-canr you will (from memory) see mentions of KOH as a material of interest.

  • Alan Smith

    To clarify I wasn't blaming you for the apparent lack of results with it; I planned getting one and/or the other at some point anyway (and the canisters also work better as a check source than bananas).

    On theoretical grounds KOH should be better than other choices as an electrolyte; for the same reason it's most often used in alkaline batteries and indeed I think was getting a higher than usual power supply load with it (a proxy for current until I get a suitable current clamp, which I should as soon as I can). As for whether I am getting higher temperatures for the same current, that's something I can't verify yet with my "setup".

    EDIT: here's an interesting document on the conductance of various chemicals https://www.emerson.com/docume…ed-chemicals-en-68896.pdf

  • To clarify I wasn't blaming you for the apparent lack of results with it; I planned getting one and/or the other at some point anyway (and the canisters also work better as a check source than bananas).

    Goodness me! I never thought you were blaming me for anything, and I never expected you to get instant results like J5 - I hope you weren't either. LENR takes a lot of time and patience to track down, in my case several years before I began to see any results that looked interesting. There are many many variables, and the experimental space is huge. Good luck, if you enjoy it keep at it and keep telling us about your work. Even null results can be useful.

  • Alan Smith

    Of course I was not expecting immediate results. There is certainly an enjoyable aspect to the testing, but at this level it's almost similar to gambling in many ways.

    Right now I'm keeping the previously treated electrodes on a very mildly heated plate, adding distilled water or KOH solution as the solution dries up and the electrodes seemingly slowly rust away (I guess some hydrogen must also be evolved in the process?).

    Coincidentally over the past day I've obtained the highest and most prolonged "background" peak over the past few days and certainly noticeably higher than it was yesterday even just after bringing the K electrolyte canisters in the testing room at about the same time of the day. Furthermore in some cases it seems as if after adding water electrolyte solution briefly increases short-term readings.

    All of this could of course be wishful thinking (i.e. previously mentioned the gambling factor); it's too easy to see patterns where there probably aren't.

    With the electrodes unpowered, after adding KOH solution, there appears to be a 300-325 mV voltage between them, with the top plate (which incidentally uses to be the cathode during active testing) at a negative voltage relative to the other. Galvanic corrosion should be occurring at some level.


    The graphs at the end of the document are certainly interesting. HCl even at only 10% wt. concentration (grocery-store/cleaning grade) is a better electrolyte than I imagined.


  • I found (not that much of a discovery since it should be expected) that the electrode assembly would work as a very leaky battery. At 5V it can be charged to somewhere above 2.5V, which quickly drop below 1.5V within a couple seconds, then it follows a discharge curve similar to this one I recorded earlier with a webcam and a multimeter:

    So I can imagine that the full discharge curve would be similar (on a much shorter term) to the typical ones for ordinary batteries and that the sharp drop on the above graph would be right after where normally batteries would be considered discharged:

    Which made me realize that if this was a real functioning LENR cell, then the operating conditions would be akin to those of an abused rechargeable battery (overcharging/overdischarging, possibly cycling back and forth between these states).

  • Alan Smith

    The behavior reminds me of that of a "dead" battery in that it will maintain a baseline voltage level and even slowly self-charge to some extent, presumably due to ongoing chemical reactions occurring within it. Don't capacitors discharge to 0V? (not that I have any real experience with their charge/discharge curves. I looked at diagrams on the internet and thought that batteries were closer in behavior)

    Earlier today I inadvertently displaced the top electrode, which caused it to short and lose its charge. After that moved it back into a correct non-shorting position and added KOH electrolyte, but didn't attempt to charge it up. Voltage over time went like this (spot measurements):

    EDIT: I made another test after short-circuiting the electrodes.

  • The behavior reminds me of that of a "dead" battery in that it will maintain a baseline voltage level and even slowly self-charge to some extent, presumably due to ongoing chemical reactions occurring within it.

    Ah- this suggests to me that you might have a photo-electric battery- The 3-400mV figure I have seen many times before - probably simple charge separation. If possible, put it in the dark, short it out briefly and leave it for a known period - preferably overnight. Next day measure again and then put it into some sunlight. If I am correct you will see at least a 200mV jump in output, but the current is always pretty negligible. (sadly)

  • Alan Smith

    It's going to rain for the next couple days and it has been raining for a while at my location. I've been using only moderate artificial light for the past day or so. However the electrodes are currently in a low temperature heater held at slightly above ambient temperature (around 45°C, I think). Not much of the already dim artificial light makes it there, but I guess that infrared radiation could also be considered light.

    From the graph I posted above it appears to go from 0 to 380 mV in about 90 seconds, no need to wait for hours. This wasn't clear from the thumbnail, so I'm posting the full graph below.


    And here is the source video. Here the electrodes were over the low-temp heater but not with the DIY infrared reflector in place:


    Over the course of about 60 minutes it will reach roughly 480 mV as long as the electrolyte doesn't completely evaporate (just traces seem sufficient) and the electrodes don't short.

    Adding more electrolyte causes a temporary slight voltage reduction (e.g. from 472 mV to 444 mV) but then it recovers. If I add distilled water, voltage decreases significantly.

  • If I understand correctly from previous posts, the cell under discussion uses an Italian 100-lire coin as anode and a stainless steel washer as cathode. From an online description, I think the 100-lire coin is made from Cu-Ni coinage alloy of around 70% Cu / 30% Ni. The ferritic stainless steel alloy composition is unknown but would typically be at least 82% Fe (430 Alloy).

    A very common connection in piping systems is copper to iron/steel pipe. Without the use of an anti-galvanic insulating adapter the steel pipe corrodes many times faster than iron/steel alone.

    The standard cell galvanic potentials are Cu = +0.34 V Ni = -0.24V and Fe = -0.44V. So the measured cell potential of ~0.5V is about what would be expected.

  • magicsound

    Pre-WW2 100-lire coins indeed contained Nickel, but later ones didn't (EDIT: this is not entirely correct, see next comment for clarification). The earlier version was composed of an alloy equivalent to SS302, while the latter to SS430.


    https://it.wikipedia.org/wiki/Acmonital (contains more information, in Italian)

    https://www.lamoneta.it/topic/…=comments#comment-1092380 (background, in Italian)

    Following the previous post I did more testing and indeed as Alan Smith suggested it seems that the assembly (also) acts as a capacitor, especially when the electrolyte solution has dried up (and the oxide layer/dielectric acts more as an insulator). So, in some ways this worked a bit like a partially self-recharging, battery-capacitor hybrid (if this makes any sense).

    At some point I tested also the voltage-resistance behavior (to the extent of what was possible with my equipment) and it looked as if the higher the voltage the higher the resistance, which seemed to suggest a capacitor-like behavior (larger separation -> larger resistance?).

    Unfortunately no real signal that could be clearly associated with the testing was measured at the Geiger counter located about 2.5-3 meters away. I'm still getting a periodic signal but no long term increase that could be indicative of some sort of activation of materials nearby. It's difficult to make out whether the smaller features are real or just noise.

    (the brief spike on the left was when I tried placing the canisters of K electrolyte close to the Geiger counter)

    A final test yesterday was using much larger amounts of (supersaturated) electrolyte in an attempt to replicate the methods used for the first tests (before I made this thread). I used K2CO3 instead of KOH (safer to use liberally) and I found that growing a thicker red iron oxide layer would be much easier this way. On one instance I observed a sort of heat burst, but that could have been due to higher current being passed through the thicker porous oxide layer formed without shorting the electrodes out. I'm not measuring temperatures yet, so this is just a subjective observation.

    (a thick, iron oxide-potassium carbonate dried "paste" is visible in the second photo, it's not just rust)

    I couldn't reproduce the grainy black layer of the opening post, which at this point might have been due to bicarbonate impurities left in my DIY sodium carbonate. It seems that electrolysis of NaHCO3 gives of carbon monoxide, a notoriously powerful reductant (source).

    A better idea would probably be forming such layer beforehand (possibly with other methods) rather than growing it on the spot with the electrodes on top of each other. Improvements would be also needed in order to apply a large current in a way that the magnetic field generated won't make the electrodes slide against each other (destroying the intermediate layer).

    In the end after not seeing any improvement (mainly due to brittleness of the oxide layer formed, even with K2CO3) and continuous attention needed and difficulties to making it work as intended, yesterday I teared the setup apart. The electrodes are still on the low-temperature heater slowly rusting away together with other pieces.

    Today I started a new one with slightly different materials and arrangement, but still with ferromagnetic steel and still in a sense "unconventional electrolysis", but more on this later.