can Verified User
  • Member since Jan 20th 2017

Posts by can

    Diluting the potassium carbonate–sodium bicarbonate (mostly decomposed to carbonate, I would think) solution for the most part mitigated the electrolyte accumulation issue, but the so-called mid-point voltage (voltage at which current is minimized) shifted above 50V, even at a temperature greater than 90 °C.



    Due to sodium ions, plasma color at the [still partially blocked, it seems] cathode even at 62V was intense yellow (see gif below). At the same time, slow-moving strings (?) could be seen moving upwards in its proximity, which I also observed in earlier tests.


    Alan Smith

    Here is relevant data from the indicated paper (I added Celsius scale on the right graph).



    From here it does not look that the solubility decreases above 427K, only that it increases slower. It could still be that at higher temperatures the solubility actually starts decreasing, though.


    It could also more simply be that the rapidly boiling water around the cathode causes precipitation and local accumulation of the electrolyte, and that conditions are such that the phenomenon became macroscopic.

    Actually every welding inverter provides more than sufficient energy source for such a fun..

    In the opening post where I wrote that as an introduction, I was referring to high-voltage power supplies commonly used for plasma electrolysis experiments (technically also known as "contact glow discharge electrolysis" or "CGDE"). VARIACs are often used and I could have got one, but besides maybe a fuse, they don't have any built-in safety feature.


    In principle it should be possible to use a welding power supply for the experiments described here, but many low-end cheaper models such as the one you linked don't allow to directly vary voltage, or only allow to vary it on a limited range; not from 0V.


    Keep in mind that the reaction observed here is not an electric arc and unlike them, it does not need current limiting to be properly and safely driven. I don't need tens–hundreds of amps, only a few at most, and during the glow plasma discharge typically much less than 1.


    A lot of the basic information you need can be found in this - now very old - 1902 - source book. The basic science remains pretty much unchanged, and it was written by Ahrennius, a master of the subject. I use it as a reference book for the basics and beyond. The physics and the methods might have changed in the last 120 years, but the behaviour of the materials remain the same.


    ofelectrochemistry handbook 1902..pdf

    I definitely need a basic electrochemistry read, but to clarify one point here: the electrolyte mixture, under cold conditions, appears to be completely dissolved. It's only when voltage is increased to a level close to where plasma electrolysis starts that it accumulates close to the cathode and onto it in solid form.


    I'm aware that during electrolysis the electrolyte concentration locally varies, so it could be dynamically exceeding the solubility threshold under the conditions tested.

    It turns out that Randell Mills also has worked (at least for a period) on plasma electrolysis and been suggesting that hydrinos are produced in these systems. In some of the published patent applications there are descriptions of plasma electrolysis devices, like for instance in the abandoned US20040247522A1 - Hydrogen power, plasma, and reactor for lasing, and power conversion. Below is a composite image with relevant excerpts and figure extracted from it.



    Interestingly, the use of vibrations for the cathode and/or the solution is suggested. I also thought of something along these lines (but haven't applied it yet); I think it could help forming a gaseous layer around the cathode and start plasma electrolysis at lower voltages in particular when the electrolyte solution is cold.


    But more than this, of more interest is the possibility of using more than one electrolyte, in particular the "catalysts" reportedly useful for producing hydrinos. This would mean for instance having not just K2CO3 (or KOH) but also other compounds-catalysts in solution at the same time, as this could plausibly help lowering the voltage from which the process starts. A lowering of the voltage of plasma generation under the presence of these catalysts has already been reported by Mills with gas discharge cells in some publications.


    There's a paper from Mills et al. where plasma electrolysis has been used; I'm not sure if there is more on the same topic:



    Along these ideas, I tried adding sodium bicarbonate to the previously made concentrated K2CO3 solution, hoping it would decompose to sodium carbonate at the higher than water-boiling temperatures reached (115–120 °C), but [perhaps not too unexpectedly] it made things worse, with more severe blocking of the cathode by electrolyte accumulation. I'm not entirely sure if this was because of the bicarbonate and conversion of potassium to bicarbonate or because it would not have helped anyway to add anything more under the close-to-saturation concentration conditions.



    These electrolyte "sticks" are what I kept frequently removing from the thin tungsten cathode. Sometimes, when crushed just after/during removal they produced sparks and emitted tiny but dense smoke plumes; I think elemental Na-K is formed under these electrolyte accumulation conditions due to the electrolyte itself getting electrolytically decomposed.


    These formations prevent plasma electrolysis from taking place; a form of normal electrolysis at high current occurs instead with them. When the hot electrolyte solution concentration is high enough under their presence, sparks and explosions from larger electrolyte formations start occurring, as I showed previously. More spark-like features started appearing today after I added NaHCO3.


    This might work better using LiOH–NaOH–KOH and other soluble hydroxides of the elements listed as hydrino catalysts, but it needs more professional equipment and testing conditions than what I can provide. Furthermore, the solubility of mixtures of common electrolytes is likely to be different than that of the separate electrolytes, and that needs to be investigated.

    Yes, that's what he's saying. There must be an 'excess' of hydrogen atoms for the process to have an effect.


    Quote

    [0205] The potential energy of H2O is 81.6 eV (Eq. (43)) [Mills GUT]. Then, by the same mechanism, the nascent H2O molecule (not hydrogen bonded in solid, liquid, or gaseous state) may serve as a catalyst (Eqs. (44-47)). The continuum radiation band at 10.1 nm and going to longer wavelengths for theoretically predicted transitions of H to lower-energy, so called “hydrino” states, was observed only arising from pulsed pinched hydrogen discharges first at BlackLight Power, Inc. (BLP) and reproduced at the Harvard Center for Astrophysics (CfA). Continuum radiation in the 10 to 30 nm region that matched predicted transitions of H to hydrino states, were observed only arising from pulsed pinched hydrogen discharges with metal oxides that are thermodynamically favorable to undergo H reduction to form HOH catalyst; whereas, those that are unfavorable did not show any continuum even though the low-melting point metals tested are very favorable to forming metal ion plasmas with strong short-wavelength continua in more powerful plasma sources.


    Again, seemingly no relation with Holmlid's research, although the iron oxide catalysts used are active when they are oxidized, and hydrogen admission will reduce them, so one could argue that similar processes are involved.

    As far as I am aware of, according to Mills rather than H2O on its own, it's the process that chemically forms H2O which is a hydrino catalyst (in the presence of hydrogen atoms), i.e. "nascent water".


    Right in the claim of the application I linked earlier: https://patents.google.com/patent/US20200002828A1/



    However, I don't know how this could be related to Holmlid's research.


    In the past it was shown by Holmlid et al. that styrene catalysts can form Rydberg K–OH2 (alkali–water) complexes, but I haven't found suggestions that these can also form Rydberg matter (or in other words, a phase composed of just these complexes): https://doi.org/10.1021/la034142t


    If RM could be formed from them, incorporation of other atoms and molecules could be possible similarly to how it occurs with alkali RM, but I'm not sure if this is much related with Alan's report above which seemed to only involve atomic hydrogen and water vapor.

    Rob Woudenberg

    I find unlikely that hydrinos and H(0) are completely different things, but Mills and Holmlid have focused their studies on different aspects of it, so either researcher's view might not be complete.


    I think Alan is suggesting that normally atomic hydrogen would recombine very quickly close to the production source, but sometimes it appears as if it is taking enough time to travel meters of tubing.

    Alan Smith

    I have never measured temperatures accurately, but in my crude testing I noticed that soot/carbon appears to be more readily gasified or combusted while the KFeO2 compound is being formed rather than after it has formed—at higher temperatures even. This could be a further indication that reactive species are at least transiently formed during the synthesis of this compound, and in the previously linked paper it is incidentally mentioned:


    Quote

    [...] The efficiency of alkali promoted catalysts for reactions involving hydrogen transfer leads us to suggest that it is the H(0) formed which is the source of reactive hydrogen in the catalyzed reactions.

    By random chance, while discussing about BrillightLightPower patent applications in the Telegram LENR-Forum channel, I found this interesting excerpt in one of them: https://patents.google.com/patent/US20200002828A1/ from paragraph [0310] :



    By substituting Na with K, this is chemically the same process for forming the active phase of the styrene catalysts used by Holmlid in his experiments. Basically it is saying that hydrinos—i.e. ultra-dense hydrogen—would be formed as such active phase is also formed.


    Since the compound decomposes with water at low temperatures, it should be possible to conceive a reactor where such phase is cyclically formed-destroyed since the reaction is reversible. This is also a suggestion being made in the same patent application, and a table lists it among many other reversible reactions that would also be forming hydrinos.




    The overall process described by Mills for forming hydrinos however is different than that suggested by Holmlid for forming RM and then UDH. From his latest publication with Kotarba and Stelmachowski, furthermore, Holmlid does not seem to think that Hydrinos are a proven concept:


    https://doi.org/10.1016/j.ijhydene.2021.02.221


    Quote

    [...] Other forms of hydrogen H have been proposed to exist but have not been convincingly observed or deeply studied. The most discussed case may be the hydrinos proposed by R. Mills [4] with very little experimental evidence. The proposed hydrinos have no resemblance to H(0). Further, based on quantum mechanical calculations a form of picometer-sized hydrogen molecule was proposed by Mayer and Reitz [5] to exist at high pressure. These proposed molecules are similar to H(0) in some respects, and may well exist, at least transiently.

    I haven't tested it long and deeply enough to verify how reliable it is, but it seems well-built. The floating outputs also make it more suitable for laboratory usage.


    I set it to 31V so that after arranging it in series with my bench power supply (0–31V selectable) I can more or less seamlessly adjust voltage in the 0–62V range.

    In order to extend the usable voltage range of my adjustable power supply I recently got a 36V 10A fixed voltage DC PSU. Surprisingly, not only it is adjustable in a relatively wide 26–41V range, but it also has floating outputs, which makes it more flexible than laptop/computer PSUs which are usually ground-referenced (I checked the ground connection and it was ok).



    Not much difference in reaction at higher voltages at least up to the maximum I could reach overall (72V), although the plasma starts to acquire a more violet hue reminiscent of potassium there. Cathode wear and reaction intensity (noise, etc) also start becoming significant and for practical purposes it might not be useful to go much higher since then the experiments would only last minutes.


    The applied current only increases after the minimum abruptly reached at about 31–35V (under hot K2CO3 electrolyte conditions at high concentrations depending on temperature; when cold this threshold is higher) and in one test I could plot a similar voltage-current curve:



    The highly dynamic conditions make it difficult to plot an accurate graph in this respect, since the reaction changes quickly with electrolyte temperature and concentration, and both change during operation given the relatively small amount of electrolyte solution used.




    Perhaps the most significant observation so far is that at the high electrolyte concentrations used, at the same voltages where plasma electrolysis occurs, the K2CO3 electrolyte appears to slowly accumulate on the surface of the cathode in relatively large amounts. Local temperature can occasionally even reach levels high enough to turn it incandescent and melt it, sometimes bursting with the emission of potassium-like light as shown previously. Possibly, electrolytic decomposition of the solid electrolyte to metal also occurs to some extent, while water electrolysis still continues.


    It would seem that this accumulation process is somewhat related to the plasma electrolysis phenomenon, i.e. that plasma electrolysis could be occurring because (or also because) of the strong increase in electrolyte concentration close to the cathode (K+ ions in solution, here). Since increasing the concentration of the electrolyte in the solution very easily lowers the voltage from which the electrolytic plasma process starts also (beyond conductivity), at least on an intuitive level this seems correct.



    Whether allowing the electrolyte to accumulate like this is advantageous, I'm not entirely sure. The cathodic plasma appears to be still occurring to some extent below the surface and on the exposed cathode portion, but not as with the clear surface.


    On their own, conditions where hydrogen atoms (from electrolysis) are forced through alkali as partially molten salts and as metals could be interesting, though. LENR in molten salts has indeed been reported in the past, although under different conditions than what I've tested so far. Just a couple examples from lenr-canr.org:

    It might be common knowledge, but I wasn't fully aware of this during the tests performed. It looks like the boiling point of highly concentrated aqueous hydroxide solutions like KOH is greatly increased relatively to that of plain water. Graph for KOH:



    Source: digitized and converted to Celsius from https://www.oxy.com/OurBusines…/KOH-Handbook-2018-ed.pdf


    For NaOH:



    Source: http://www.cheresources.com/in…_database/oxy_caustic.pdf


    And K2CO3 (this, I did notice after measuring 116 °C from the saturated solution I tested recently):



    Source: digitized from https://www.armandproducts.com…nt/pdfs/k2so3Complete.pdf

    Rob Woudenberg

    Technically speaking, the process I am referring about has the characteristics of a glow discharge. It occurs more or less homogeneously over the entire immersed cathode surface and also with fresh, non-sharp cathodes. I don't think there is any particular electric field enhancement effect going on with the cathode surface itself, but a related effect must be occurring close to it or at its interface when the density of ions in solution is increased (up to saturation). The small-area negative cathode under operating conditions must be somehow seeing a "wall" of positive charges, causing a homogeneously distributed large effective local electric potential.


    If H RM actually manages to be formed under these conditions, the large condensation energy to the ultra-dense form would likely be contributing energy to the observed plasma phenomenon.


    EDIT: if one wanted to involve RM further, a related hypothesis could be that high-excitation alkali RM formed in the same process causes intense electron emission, similarly to what happened in the early '90s in experiments with thermionic diodes by Holmlid et al. (example), promoting plasma formation. As for proving it, though...

    The explanation from https://doi.org/10.1016/j.ijhydene.2021.02.221 is a bit impractical and serves explanation only since the surface of the solid as shown has a certain high temperature that will make condensation of H RM almost impossible.

    Other than that it may indeed be the case that using the K doped iron oxides the production efficiency can be very high. So, that gives it indeed a simple control just by the supply of Hydrogen.


    If the relatively weakly bound alkali RM can be formed on the surface of the hot catalysts, it should not be an issue for H RM to also exist there with its bond energy of at least 9.4 eV/atom in its lowest excitation state. See also the conclusions of https://doi.org/10.1063/1.4947276 in reference to another effect:


    Quote

    Above the transition temperature, the superfluid and superconductive long chain-clusters H2N(0) have disappeared, and only the normal small clusters like H4(0) remain. The higher Rydberg matter level H(1) remains on the surface at high temperature, thus extensive desorption does not take place


    On the other hand, I see an issue for the superfluid form of H(0) to exist on the surface of the industrial alkali-promoted catalysts under regular operating conditions in light of this transition temperature. This point is not explained in the latest paper.


    I figured that the most ideal setup require three efficient elements present:

    • a provision to split molecular H into atomic H
    • a provision to release alkali metal atoms and excite them into Rydberg states or ions
    • a surface to allow condensation to UDH

    For all three we can give multiple examples.

    For what it's worth, it seems that a surface is not required for the condensation to UDH from H RM according to the latest publication (although it should still be required for the initial RM to form): https://doi.org/10.1016/j.ijhydene.2021.02.221


    Quote

    The final conversion step from ordinary hydrogen Rydberg matter H(l) to H(0) is spontaneous and does not require a solid surface [...]

    Quote

    H(0) is finally stabilized probably by emission of infrared and visible radiation [81,82], either in the form of free clusters in a vacuum or gas [83], or supported on suitable surfaces [79,84].


    This should mean that H RM may not necessarily condense to the ultra-dense form in close proximity to where it was first formed.


    The approach by using electrolysis is an intriguing one.

    I guess that the biggest challenge is to bring the alkali ions into an excited state. This will probably be possible at the boundaries of the plasma bubbles, where temperature might be high enough and free electrons are available. The plasma bubbles themselves will probably contain the required H atoms. The surrounding water could be useful to absorb the condensation energy that is released when UDH is formed. I am curious how you see this though.

    The hot plasma sheath surrounding the cathode is reportedly initiated by secondary electron emission from the impact of positively charged species on it (H, alkali) due to the applied electric field. As the cathode becomes hotter, thermionic emission should also start becoming important. The electron density should be high.


    The density and flux of H and alkali atoms near the plasma region is likely to be considerably higher than that attained on the surface of ordinary industrial catalysts under typical operating conditions. Neutral species in solution may also participate to the reaction.


    Reactive H species (radicals) assisting ordinary chemical reactions are also known to be formed under these conditions (try for example searching for "H·" or "radical" in this open-access paper: https://doi.org/10.1007/s11090-017-9804-z), which may be of the same nature of those that Holmlid et al. suggest to be due to UDH on the surface of industrial catalysts.


    Transmutations and excess heat have been sometimes reported in these experiments. If there is a common cause for LENR, it should apply also for plasma electrolysis.



    I should also add that from my own recent testing that I documented in another thread here on LENR-Forum, a visible plasma can be initiated at rather low voltages (less than 30V) if the concentration of alkali ions in the solution is high enough. Just a personal speculation, but I thought that perhaps there could be a link with how a high density of alkali atoms is suggested to be important for [eventually] the formation of H(0).


    (which leads to another question: how is the electrolytic plasma discharge actually initiated in the first place? High voltages are not strictly required as commonly thought).

    Alan Smith

    Since the solution is saturated, the pH should be already rather high and possibly the small amount of KOH added this way might not influence it too much. I don't have suitable means at the moment for measuring this, though.




    Speaking again of cathodic plasma electrolysis at lower voltages than usual, in a paper (attached here, from page 79+) Takaaki Matsumoto described the usage of K2CO3 as electrolyte at 1.5M concentration. He came up with this current-voltage diagram (annotations mine):



    Although his experiments have already been performed at lower voltages than other researchers, in my case with the saturated hot K2CO3 solution, Vd appeared to be significantly lower than his ~85V (but accurate measurements were difficult due to potassium carbonate blocking the cathode).



    One interesting thing mentioned in the paper, though, is:


    Quote

    Fig. 4 shows a sequence of pictures for the tiny sparks, in which they were accidentally separated from the surface of the palladium cathode by two explosions. The first small explosion took place in the gas atmosphere over the water level, which was triggered by the tiny sparks, as shown in Fig. 4(a). Due to mechanical shock from the explosion, a group of the tiny sparks were separated from the cathode, and were driven by the electrical potential towards the anode, as is shown in Fig. 4(b). The tiny sparks were decayed on the anode to trigger the second explosion, as shown in Fig.4(c). These pictures showed that the tiny sparks were negatively charged and could exist in water solution as independent bodies.


    I have never observed these in all the testing made so far, except for today once I used concentrated enough K2CO3 solution to produce deposits on the cathode that would glow and to some extent explode. I wonder if a form of this is what Matsumoto observed, since apparently during glow plasma electrolysis the electrolyte tends to concentrate close (and on) to the cathode, while this does not happen during normal electrolysis in the same electrolyte.

    Alan Smith

    Indeed, thick formations of apparently decomposing hot potassium carbonate would form around the cathode. Sometimes they would detach from it and burst at a distance, or deflate with the emission of hot particles. I imagine that the potassium vapor formed as suggested in the PDF you linked would rather quickly form KOH.



    I found interesting that heating of some sort was still going on and sufficient to cause K2CO3 to heat up that much in relatively large amounts.

    Since saturated KOH solution worked very well in previous tests, but was far too dangerous to use, I tried preparing a saturated K2CO3 solution and repeating cathodic plasma electrolysis with the same 1 mm tungsten needle used so far. Unfortunately overall it still does not work as good as with KOH, in that the characteristic breakdown (VB) and mid-point (VD) voltages are not decreased as much as with KOH.


    There appears to be a blocking effect during the cathodic plasma reaction where the potassium carbonate would get precipitated out of the solution and deposit on the surface of the cathode, preventing the reaction from proceeding as intended. The electrolyte temperature in the video below was above 95 °C— close to boiling—and voltage was switched between 19V and 50V.


    Temporarily restoring normal electrolysis (by lowering voltage) removes such solid K2CO3 layer and the intended operation when voltage is brought back to high levels. It could be an argument for pulsating input power.


    External Content youtu.be
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.


    EDIT: allowing such K2CO3 deposits to build up on the tungsten cathode at 50V during cathodic plasma electrolysis eventually shows interesting effects. Video thumbnail made at minute 2:21.


    External Content youtu.be
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.

    The method Holmlid is describing to generate UDH/UDD seems to have very limited options to control the amount of UDH/UDD. I am wondering whether pulsation would be helpful in doing so.

    In particular modulation of additional electromagnetic and/or electrostatic fields would be candidates of methods in my view.

    Looking at: [image]

    Extra stimulation (and control) of emitted electrons and alkali ions could be done by placing one or more extra electrode(s) to add a controllable additional electrostatic field near the surface of such catalyst.

    From the description given in the latest paper it sounds like the reactive hydrogen species adsorbed on the surface of alkali-promoted catalysts are already UDH in various forms and so that it is apparently very easily formed. If this is the case, the amount of UDH produced would simply be proportional to the amount of gas admitted over the catalysts.


    I guess the main issue will be not making this form of hydrogen react with ordinary molecules before it can transition to a stable form, which should be easier to accomplish in a vacuum.


    See this final excerpt from https://doi.org/10.1016/j.ijhydene.2021.02.221


    Quote

    The efficiency of alkali promoted catalysts for reactions involving hydrogen transfer leads us to suggest that it is the H(0) formed which is the source of reactive hydrogen in the catalyzed reactions. Condensed atomic hydrogen has in fact three different forms with different length and energy scales, as described by Hirsch [125]. At the two largest length scales, the bond energy for hydrogen atoms is a small fraction of what it is in the most densely bound form which we here call H(0). The second largest length scale is called ordinary Rydberg matter of hydrogen H(l), of which the lowest state is H(1). The largest length scale of hydrogen is superfluid and superconductive, with Rydberg electrons and very loosely bound hydrogen atoms. This means that H(0) becomes a storage phase of hydrogen atoms, and can produce loosely bound hydrogen atoms which will easily take part in chemical reactions.


    Otherwise, completely different methods which can provide a high density of excited alkali and hydrogen atoms should in principle be able to produce UDH. I think cathodic plasma electrolysis, which I have often toyed with, should work towards this, but as usual the main issue is suitable detection of the expected reaction products.

    I was curious and found that the plasma reaction appears to improve all the way to 10M KOH concentration (37.73g KOH in 67.2g tap water), although conductivity at very low voltages somewhat gets worse. So, it does not appear to be strictly a matter of electrolyte conductivity, although more testing would be needed to make sure, preferably under more controlled and safer testing conditions.


    The solution, which rapidly turned dark violet-red possibly from tungsten and iron from the electrodes used, is now way too dangerously concentrated; I'm not sure if I want to keep it around in my rather unsuitable testing environment.


    At this KOH concentration, there is visible cathodic plasma already at 28–29V, using the same 1 mm tungsten needle as mentioned earlier.


    External Content youtu.be
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.


    At higher voltages (up to 50V) the plasma looks "sharp" and at least on video it appears to emit a violet color at the highest steps. It turns more yellow as voltage is decreased. Sometimes such transition has been pointed out in related literature, but I don't recall where exactly.


    External Content youtu.be
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.


    5x slow-mo of the first few seconds of the above video:






    EDIT: from further testing, it appears that there are improvements all the way to KOH at saturation levels, albeit with diminishing returns. Using the usual 1mm tungsten needle, I managed to get the mid-point voltage to 31V, and at 50V the cathodic plasma reaction was relatively violent, causing electrolyte droplets to get out of the glass vessel. The solution turned completely opaque dark violet and not much could be seen of what was going on with the reaction except from hints at the surface.


    At higher voltages, it takes some time for the reaction to ramp up (or the electrode to heat up), and current increases similarly to what I showed earlier.


    I couldn't get full I-V data for this run, but I managed to get some around the mid-point voltage. After about 38V current increased in a strange way and I stopped there since reliable measurements were not possible due to a sort of hysteretic behavior as pointed out earlier too.



    Of course, at this point such experimentation becomes rather dangerous due to how caustic such concentrated KOH solution is. In the end, due to lack of suitable equipment and environment to handle this sort of experiment safely, I decided to dispose of the electrolyte solution, decontaminate the equipment used and wait for better times for more testing (or others with better equipment). I think I got good data and made interesting findings so far, though.

    When switching on the cathodic plasma from an off state, sometimes the process is not instantaneous. It appears as if heat takes a while to ramp up.


    This occurs more easily when conditions are such that the applied current is higher, for example using a larger cathode and larger immersion depth, or when electrolyte temperatures are somewhat lower than usual (which leads to a higher current draw). This makes me think that resistive heating of the cathode has an influence of some sort, although it's difficult to separate this variable from other ones under these conditions.


    In the gif animation below I used a 1 mm tungsten needle where this relatively quick ramping up effect is shown after switching from 19V to 50V.



    I also took current-voltage measurements with it in the higher voltage region. In the last power step current increased slightly, which seems consistent with the behavior visually observed. The behavior below 30V was strange this time and I mostly omitted it from the graph.