It seems like you would need a sustained short in order to dump energy into the inductor during the current pulse. Then when the high current clears the short, the inductor delivers the high voltage to cause the sparking.
There was some sparking for a while just before the current increased. I captured one event and will post images in a bit. On release of the short 21 volts was observed at the electrodes. Now that the cell is boiling, those events seem to have stopped.
Also Magicsound's inductor may have much higher L than CANs because the bolt he is using appears to be a much better core. The higher the inductance, the longer it takes for current ti build up, meaning that the short has to stay longer in order to charge up the inductor.
My inductor core is made tuneable for that reason. Once I find the physical settings for the cell, I can play with the core tuning to see if there's an effect.
I've added some summary notes on the experiment procedure to my original post #258 for reference.
Here are scope capture images of a shorting event.
My latest tests so far have been much shorter (minutes long, instead of hours) mainly due to reliability issues so I'm impressed yours is still running, although in this precise moment current appears to be low with seemingly not much going on. I'm wondering if the slope of the electrodes also affects this somewhat. Due to space limitations (as shown in the various media posted), I usually put them at an angle and in contact with the bottom of the jar(s) I've used so far. Possibly this could affect the way material deposits on the inner gap surface.
As a side note, I tried to manually sample Geiger readings from the dashboard of your live experiment. Readings seem consistently higher than they were at the beginning of the experiment, but the change is slight and so it could be due to daily natural variations or other external factors. I get large swings daily in my place because of those.
As for your experiment, perhaps you could try to reinvigorate it a bit with some more HCl addition. Hopefully at this point your testing location isn't smelling too much of bleach or pool water.
EDIT: I see that current sometimes shortly spikes to 2-3 A from its average value of about 0.5 A, so something seems to be still occurring.
EDIT2:
It seems like the experiment is supposed to have these states:
S0: Plating = high resistance, low current.
S1: Shorted = low resistance, high current. Inductor is charging up.
S2: Discharge = high resistance, high current, high voltage, . Indcuctor is discharged by sparking.
Sequence: S0 -> S1 -> S2 -> S0 ...
It seems like you are sometimes stuck in S1 (shorted with 10A current too low to break short), and sometimes in S0 (plating with microshorts cleared by a few amps of plating current).
Getting it to oscillate seems difficult.
The S1 problem may be solved by a higher current supply. The S0 problem might need mechanical or electronic mods to force it through the 3 states at a desired rate. An electrical change could prevent medium currents. It should plate slowly until shorted, then have a big current surge. It should not stay at a medium current where tiny shorts are self-clearing.
can There are still occasional shorting events as you noted, and the current is going up again. In fact there are some intense sparking events just now, with sustained current of 20+ amperes. I've turned the light off in the lab so they show up clearly on the camera.
But it has been a longer run than I expected, with much learned about the apparatus and the cell setup. I'm encouraged and look forward to further progress tomorrow.
Getting it to oscillate seems difficult.
The S1 problem may be solved by a higher current supply. The S0 problem might need mechanical or electronic mods to force it through the 3 states at a desired rate. An electrical change could prevent medium currents. It should plate slowly until shorted, then have a big current surge. It should not stay at a medium current where tiny shorts are self-clearing.
Yes, correct. The power supply is capable of 45 amps but there's a 0.2 ohm current limiting resistor in the circuit that I will bypass in a future test. Power is off now, but there was a lot of intense arcing during the last 15 minutes or so, with sustained current of 20 amps or more. I'll have a look at the electrodes tomorrow and expect to find areas where the steel melted and sustained an arc like a welding puddle. Hopefully they won't be stuck together....
I appreciate your input on mechanical stimulation to initiate the arc. It's not an easy engineering job and I encourage you to propose a practical design to try.
Intense prolonged arcs indeed seemed to get produced in the past few minutes. The resonant noise I got so far however is from much briefer events occurring at a high rate rather than these, which could be considered an additonal mode of operation to what Robert Horst listed above and require different conditions. Which mode of operation is "better" or could be more promising from the point of view of possible anomalies at this point it's not clear, though.
For what it's worth, in my latest tests S1 and S2 cycle as soon as I start adding HCl drops and they don't revert to the S0 state, but my experimental conditions are different on several aspects than magicsound's and are probably also why I have only been able to run my tests for a few minutes at most.
With magicsound putting so much effort on yesterday's experiment I had to somehow contribute too with one today. I was considering undoing my current coil and craft a better one with a different, hopefully better core composed of one single piece, but perhaps a more significant and useful test that could be done would be checking out if I could replicate the same resonating sound with an air-core coil, going along what Robert Horst suggested.
Preparations
Experimental notes
Videos
https://streamable.com/d369w
https://streamable.com/bc1mi
https://streamable.com/960gk
Observations
EDIT: about 10 minutes into the test (see previous experimental notes for reference):
This is when I ran the electrodes out of the water. The signal is different and with a much higher frequency. I changed the gain setting of the spectrogram to make it more visible:
EDIT2: Geiger readings. Again they decreased at about the time of the experiment, possibly due to having the window in the testing location open while it was running.
I appreciate your input on mechanical stimulation to initiate the arc. It's not an easy engineering job and I encourage you to propose a practical design to try.
I will work on a drawing of my thoughts on this.
To make a specific proposal, it would be helpful to know what material would be best to use for supporting and connecting hardware. Do you or CAN have any thoughts on this? What is the material of the electrodes? It looks like steel L-brackets. I assume you want to avoid stainless steel electrodes because electrolysis will produce hexavalent chromium which is extremely toxic and carcinogenic.
What is the material of the electrodes? It looks like steel L-brackets [...]
They are (in my case they were straight and have been bent to obtain a more convenient L-shape). I initially picked them on the basis of overall shape/size and ferromagnetism exhibited. I'm assuming that mine are composed of ferritic or martensitic steel, meaning that they're composed of mostly Iron, a significant percentage of Chromium and probably no Nickel.
The main rationale for this choice was initially wanting to form during narrow gap electrolysis (as described in the first posts of this thread. The experiments have lately morphed a bit from those) a metal oxide layer similar in composition to the Fe-Cr-K oxide catalysts used in many industrial processes and in experiments by other researchers. A second reason was that at the time I used to hold the electrodes together with Neodymium magnets, which could only be possible with ferromagnetic materials.
At the moment I don't see too many reasons for sticking with steel electrodes, but it still has its own advantages.
A possible idea could be using electrodes composed of different materials like Ni-Cu, Ni-Ti, Fe-Ni, any other combination or alloy of these, etc. Aluminium would probably dissolve too quickly in the acidic solution (but this could potentially be an advantage. It has to be tested).
Other candidate materials could probably be selected from the tables that have been posted and discussed in this thread, and specifically here and here:
Simon Brink "Subtle Atomics" Discussion Thread
I think it would be good to use readily available, economical materials. I thought steel would be a good candidate for this, but for the purposes of the reaction Nickel-containing austenitic steels would probably be better (minus potential dangers from hexavalent Chromium) than what I've used so far. A further idea for introducing other elements could be dissolving them in the acidic electrolyte solution beforehand rather than directly using them as electrode materials.
I assume you want to avoid stainless steel electrodes because electrolysis will produce hexavalent chromium which is extremely toxic and carcinogenic.
Thanks for this good tip. I did actually use SS brackets in my test yesterday. They are "hardware store grade", probably 409 alloy which is 10-12% Cr and less than 1% Ni. So maybe not the best choice but what I had on hand. I will use due caution in handling them and the electrolyte.
For the next run I'll use mild steel brackets which are zinc plated as bought. I'll sand them before use to remove the zinc, and to flatten the working faces.
can I measured the pH of my used electrolyte to be 1.3. I started with 150 ml of water, to which I added 9 ml of HCl during the test. But about half of the water boiled off, and some of the H+ will have escaped as gas during the electrolysis. So my molar ratio by volume is close to what you used, but the ending measured pH is somewhat higher (less acidic) than the calculated value.
The electrodes are pretty toasted. I rinsed the assembly and will let it dry a bit so I can collect the debris when taking it apart.
The paint on the outer surfaces didn't hold up well, especially the center section where it looks burnt. The rubber o-rings that hold the pieces together seem OK and worked well during the test.
QuoteCan said:
A possible idea could be using electrodes composed of different materials like Ni-Cu, Ni-Ti, Fe-Ni, any other combination or alloy of these, etc. Aluminium would probably dissolve too quickly in the acidic solution (but this could potentially be an advantage. It has to be tested).
Aluminum electrodes in HCl is probably very dangerous. The evolved H gas can be significant and obviously if sparks are being generated, easily ignited.
Interesting. Do you know in particular what material are those o-rings made of, or the specific application? You mentioned earlier that they are automotive-grade.
They are probably neoprene or nitrile (Buna-n). They came in a kit of different sizes from an auto parts store.
One reason I thought to use these is the Gough-Joule Effect, in which elastomers under tension become tighter when heated.
