Trying to See Thermionic Emission by Building a Flame Diode

Note: Build montage music is muted because KZbin is a bit trigger happy about copyrights.
In this episode, we try to build a diode using some scrap steel and fire! It doesn’t go all that well, but any day you get to play with the torch and turn metal bright red is a good day in my book.
Also, we’re now on Discord and Twitter, so join us!
Discord: / discord
Twitter: / usagielectric
Thanks for watching!

[@leosbagoftricks3732]

As soon as you scorch the wood, it turns to conductive carbon crud. The input impedance of the scope is probably around 1 megohm, so it doesn't take much bunt wood to register a large voltage on the scope. Plasma (fire) is conductive, you might play with this idea more, set up the experiment to eliminate the "burnt wood" conduction path so you can really see the plasma conduct.

[@UsagiElectric]

The wood was definitely the weakest point of this whole test. I think in the future, if I'm going to stick with the flame idea, I need to totally isolate the base between the two, so even if it gets charred, there's no way that it can conduct. Although, the best option is definitely trying to electrically heat the cathode without melting it. Side note, I love your projects! Both of your clock projects were absolutely fascinating and had some brilliant engineering in them. Looking forward to more videos in the future!

[@hinotori8232]

I suggest you to look up Nyle Steiner's "Flame triode" video here on YT. What you trying to achieve is definitely possible, with higher voltage and without salt.

[@MatsEngstrom]

I'm a bit skeptical to the idea that there was some capacitive coupling between the coil and the plate that cause the current to flow in the cold state. The capacitance would probably measure in attofarads and at a measly 50Hz the impedance would be more or less astronomical. Just guessing wildly it might be that the piece of wood is not totally super dry and actually conducts a bit between the mounting points. Mounting the parts on a piece of plastic maybe the cold state result would be different, but it would be hard to use the blowtorch then ;-)

[@rocketman221projects]

It looked like the wood was getting charred. That would turn it into a resistor, probably less than a couple hundred KΩ.

[@UsagiElectric]

You know, my father said the same thing to me after he watched the video too. The wood conducting is a definite possibility that I didn't think about at all!

[@UsagiElectric]

@@rocketman221projects That's definitely a possibility. The only thing that makes me think there might be a bit more going on is that the wood was already pretty charred by the time I hooked the scope up and the coupling affect grew immediately upon adding the flame, before the wood had a chance to char further. But, the wood itself could definitely have been conducting!

[@AccidentalScience]

The wood may conduct but remains not clear why the voltage raised when heating. This do not necessarily means it is a sign of thermoionic emission though.

[@brianhginc.2140]

Unfortunately it is a year late, but, if you placed a 10uf cap across the probe's GND and TIP, you might have seen a tiny build up of DC voltage as that cap would have shorted out the AC component of the signal leaking through your, most likely, wood and nails as it got hot and burnt.

[@UsagiElectric]

Holy cow, it's been a year since I posted this video! Man, time really flies. I've still been thinking about this video and I definitely want to revisit at some point. I'd like to eliminate the wood and nails aspect of it and really get a better setup to properly see if I was conducting through the wood or I just made a terrible air gap transformer. Some day I definitely will give this another shot though!

[@TheSonBAYBURTLU]

Since you heat both surfaces, it emits electrons on both sides. That's why a structure that conducts electricity is formed and you start to see the voltage in the transformer. It would be more accurate to experiment with DC volts. Besides, it would be more interesting to consider generating energy from heat instead of showing the diode effect. Thanks for the interesting experiment.

[@UsagiElectric]

Thank you! One of my goals is to revisit this, but instead generate the emission the proper way with a filament in a vacuum. I have some ideas about how to tackle it, but there's still a lot of unknowns that I'm thinking through. One day, I'll get back to this though!

[@graceoverall]

I think Unagi Electric would have been a better channel name since there ARE electric eels, but I've yet to hear of an electron bunny. 😋

[@T2D.SteveArcs]

I think this could be flame rectification, through the plasma channel of the flame

[@jhoughjr1]

seems it may be a thermocouple effect between the differing metals and not thermionic?

[@arjovenzia]

Im not sure its the same effect, but I know fire is conductive. I spent a fair while screwing around with high voltage power supplies, and eventually getting ~25mm arcs with a nice fat solid ac. I didnt have any end use for it other than electrifying a bunch of stuff, so I electrocuted alot of stuff. I noticed burning things increased my arc length. well, not exactly arc length, but it was clear a flame could increase the gap between the electrodes considerably, even if the arc itself got shorter. the largest flame I had access to was my Iroda gas torch, and could almost double the distance I could conduct between electrodes. This was all very unscientific, and at the end of a very deep rabbit hole of "Huh, thats cool, I wonder what happens if I..." combined with beer. I took no measurements and was using an obscene amount of voltage AND current (I soon got bored of skinny blue arcs and stopped chasing voltage and focused on jamming current into the system. I topped out at about 400w). so not exactly comparable. But aside from showing a few mates, I didnt do anything else with it. Very interesting demonstration, very cool. thanks

[@tstahlfsu]

Needs more Harbor Freight :-D

[@UsagiElectric]

You can never have enough Harbor Freight! (Mostly because their tools are wear items and need to be regularly replaced, haha.)

[@Kyle-sg4rm]

Using a Nernst lamp style oxide filament could also be interesting.

[@UsagiElectric]

Oh man, I hadn't heard of a Nernst lamp before, but that's actually a really interesting idea! I wonder if the ceramic filament in a Nernst lamp has enough emissivity to emit electrons? If it does, I think it could make a very interesting rectifier if placed close enough to the anode (and maybe the anode encapsulated the filament). That's a really awesome idea!

[@Kyle-sg4rm]

​@@UsagiElectric Do you reckon if it worked, it could it be efficient? I've done a fair bit of reading on Nernst lamps in general and when tested, the filaments were found to eventually stop working, but the reason for this was and as far as I know, still is, unknown - this could be worth investigating further for clues. The lamps did last a long time though - longer than most, or all of the Edison incandescent bulbs, taking into account variables such as wattage. It appears as though it was actually the preheating wire and the ballast, which failed first. And as the filament and heater were attached as one product, an entirely new lamp would be purchased, even though the old filament was still good. This shortcoming in design was probably a business decision, made to sell more of the things. The built-in preheating part of it is not actually necessary, at a basic level, because the filament can be heated by a small blowtorch, etc, until it is hot enough to conduct electricity. There may even be ways to rig that up for remote usage. And i'm guessing some other component, (a large resistor?) could replace the ballast, which I think was only there due to the filaments negative temperature coefficient of electrical resistance. These kinds of oxide filaments don't strictly need to be rare earth oxides either - it may be possible to use Aluminium oxide and/or some other more common metal oxides. The light that each oxide emits will range in colour. There is actually a video on here where a dude heats an Aluminium wire to incandescence in the presence of 'air' - as it begins to glow brightly, what appears to be happening is an Aluminium oxide layer has formed on the outside and is emitting incandescent light, while the inside of the wire, protected from oxidation by the outer Aluminium oxide layer, becomes molten Aluminium and so the wire sags until breaking. I saw great potential for Aluminium oxide to be used as a filament material. Actually making the filament is the other thing - but a bit of researching and mucking around could make it doable.

[@Kyle-sg4rm]

@@UsagiElectric Intuitively I feel that an oxide filament could be used for making a diode. The flame diode suggest/shows? that a vacuum is not essential and if other working principals can be determined, then maybe it could be engineered to work quite well. I've been meaning to try it and to simply just making a working filament, but things and stuff have got in the way. And my knowledge is patchy when it comes to electricity, so i'm a bit hesitant. I've got some links to more info on oxide filaments if you want to check them out?

[@UsagiElectric]

@@Kyle-sg4rm I actually played with the idea of heating a wire up in a weak vacuum to see if I could get any electron emission out of it as a sort of sequel to this episode, but it was a total failure. I tried steel wire and copper wire, and while I could get the wire glowing a nice orange, I was never able to get any electrons to flow to a positively charged plate. I think this came down to a few different things working against me. First, I was just using some scrap steel to work as the anode, and that’s a terrible anode material. Ideally, I should have been using something like pure nickel. Second, both copper and steel have really low emissivity. I really should have been using something like Tungsten, but then that requires a fairly strong vacuum. In indirectly heated vacuum tubes that have a separate cathode, the cathode material itself is usually nickel coated in something like barium oxide. And finally, a vacuum is usually necessary because air particles can block a large majority of the electrons trying to travel via thermionic emission. Now, an oxide filament is an idea that could work with some proper building. Nichrome wire could be used to pre-heat the oxide filament until it gets self-sustaining. Then, a nickel cathode could be tightly wound around the filament, encapsulating it and absorbing all the heat. If the nickel cathode is coated in some barium oxide, thorium oxide, calcium oxide, or even aluminum oxide, I think we can get some decent emission out of it (at least enough to be measured). Then a very tightly wound nickel anode and even without a vacuum, we might be able to get something working. But, if we’re going to coat the cathode in an oxide anyways, that seems a little redundant. If we could build a filament out of Aluminum oxide or calcium oxide or something, that could maybe make life a lot easier! I would be very interested in hearing any ideas you may have about this, I would absolutely love to give it a shot one of these days!

[@Kyle-sg4rm]

@@UsagiElectric All worthy considerations. I've been trying to find a page I found a while ago, which had quite specific details of how they made the Nernst oxide filaments. It was fairly simple, but I remember there being some stuff about the process which was important...I just don't remember exactly what it was! I'm not sure how important the binder is - someone on another yt vid mentioned using Nitrocellulose as a binder. I would guess that minimal amounts would be used and the mix would be pressed. Possibly heat pressed. And sintered, possibly in a oxy-hydrogen flame. The aims kinda need to be defined, because there are so many ways to approach it. It would be choice if the materials are ubiquitous in nature, not just stores. And if the manufacturing process is as simple as possible. Something someone could make at home, or even in the wilderness. Some grasses and such are high in Aluminium oxide/hydroxide and Silica. Sorghum being one of them. Potassium/Sodium salts can be removed and remaining calcium maybe dissolved in an acid - even Acetic acid. The most basic, yet still effective purification process. The main hurdle if approaching it like MacGyver, would probably be how to connect the filament... as platinum would work, but may not fit the brief and is not so available in nature... I was wondering about flaring the ends of the filament and if that might dissipate enough heat to make using copper a possibility... but then, lower heat = lower conductivity. So I dunno... There is not alot of info on these oxide filaments and some contradictory info eg; I just read that actually the filaments were actually removable and replaceable. A strange lack of info, given that they still have certain uses today, at least they have up until recent times. Here are a couple of links to some in action. And maybe some of my silly questions, etc, in the comments from ages ago: kzbin.info/www/bejne/l4rZiHZjos-Zmqs kzbin.info/www/bejne/j2amkIqYhp6dY7s

[@BluePawwolf]

i believe the problem comes from the ionization of the air molecules, this really do need a vacuum to work properly. also metals have different work function, im sure a cheap copper plate is enough as a cathode for this, just put the plates as near each other as you can. anyways, i really love these experimentation videos, they are fun. :)

[@BluePawwolf]

Also the problem with using salt is that you are blasting it away so that it starts to coat the plate, and the surface of the cathode and the anode should be a nice clean surface for best result. :)

[@UsagiElectric]

That definitely makes more sense! Experimenting with creating a vacuum is definitely something I want to try in the future as well, but I gotta work out how to get the cathode hot enough without a flame (also without having any tungsten filament hanging about that I can use). There's lots of good paths for experimentation here though!

[@boutinclovis4233]

The flame is made of lots of hot, tiny droplets in suspension, possibly reducing the isolation and increasing the capacitive coupling between the cathode and anode. This and the possible ionization is what probably caused the AC to appear on the scope. Maybe using an e-cig heater with its coil as the cathode would give a way to heat up without using a flame ? Also if you get your hand on some alkaline earth metal (strontium or baryum), you can put it on the coil and get a good electron emissivity. But without a proper vacuum, this kind of coating may react with the air and dissapear quite violently...

[@UsagiElectric]

@@boutinclovis4233 I initially thought that the flame is what was carrying the coupling effect between them, but the high coupling remained even after shutting the torch off. The e-cig heater is an excellent idea though! I have one more idea I'd like to try using some scraps I have in the garage first, but if that falls through, I'll definitely look into that and see what I can come up with. I mean, anything that might violently and explosively disappear in the air is high on my "to own" list, but until I figure out the whole vacuum aspect of things, that may have to wait, haha.

[@MatsEngstrom]

By accident I stumbled across this very relevant old blog post www.sparkbangbuzz.com/flame-amp/flameamp.htm - you might want to have look at it for further inspiration.

[@UsagiElectric]

Hi! I'm not sure why this comment isn't showing up on the actual video (at least it isn't for me). I can only find it in the KZbin Sutdio comments section, so I hope you can see this! At any rate, that's a fantastic resource! That's actually one of the sites I had originally gotten the idea from, but I forgot to bookmark it, so thank you so much for the link!

[@humanseagull2744]

This is the principle used as flame detection/safety within modern gas combi boilers (well over the last 30 years at least)

[@UsagiElectric]

Interesting! I never really knew how the flame sensors in a boiler worked, I'll to do a bit of reading on them!

[@T2D.SteveArcs]

You might have more luck with a big indirectly heated cathode and filament from a large tube and a higher plate voltage, also the gap between A & K should be as small as possible as im sure you know 👍 great videos mate 😎

[@jannepeltonen2036]

I think what you built is something between a capacitor and a transformer. You have a coil, and a plate, and air insulation between them. So there's going to be AC coupling - probably not purely capacitive and not purely inductuve. But the heating of the system probably reduced the impedance of it a lot - everything is moving more freely inside the heated metal, so the electromagnetic coupling is more efficient. But I don't see clear evidence of thermionic emission here. To rule out electromagnetic coupling, you really should have DC current coming in and see if that passes through.

[@UsagiElectric]

The more I think about it, the more I think you're right here! I've just built a really terrible transformer that requires a torch to work, haha. I need to build a good DC power supply that can give me some high voltage for the next test, which should be pretty fun!

[@stamasd8500]

Put a resistor in series with the plate and hook the oscilloscope probe to the ends of the resistor. What you want to be looking at is the current through the plate, not the voltage on it. The resistor will act as a current-to-voltage converter and you'll be able to see the rectified current.

[@Anthonytheredneck]

I had a taught what would happen if this was replicated but instead of adding the emission coating driectly to the cathode, and it to the fuel of a alcohol bunner. Test that, then maybe try building a catalytic style burner a device that operates practically flameless, and in a low oxygen environment.

[@duality4y]

that glowing yellow you see is ionized particles and hot ionized particles conduct electricity

[@UsagiElectric]

I really do need to revisit this one of these days! There's a whole lot going on here that I just don't understand, but it is an ultimate goal of mine to build a tube from scratch someday!

[@JorgheAyala]

Very educational video, there are many thing to learn from this experiment. The inductive reactance of the coils is the stored energy as a magnetic field, hence the voltage on the osciloscope. I wish you would remake this experiment with the ideas already pointed here, such as a blinded wood base, putting the anode and cathode closer and maybe a more pure environment in a circuit to measure the magnetic fields also. Great work, I'm subscribing!

[@Slowly_Going_Mad]

I've actually done this with a propane torch and full line voltage and the flame managed to pass enough current to dimly light a neon. Problem was that it was one with the dark glass (from a hot line tester) so couldn't tell if it was DC or not.

[@frankowalker4662]

Well done, I think you've built an AC voltage amplifier. LOL.

[@UsagiElectric]

Haha, I'm not sure what I built, but it sure wasn't what I intended to build. It was still a ton of fun to experiment though!

[@ecocrime8957]

When you trying connect to aliens on the lost planet. Just building a detector for a detector receiver is a bit nervous and sloppy)

[@shinedom]

Flame is sort of plasma, so is conductive in both directions?

[@AccidentalScience]

Why you measured voltage? Thermoionic emission should carry electric charges thus we should see current. I will try to replicate your experiment.

[@UsagiElectric]

The short answer is that I'm totally making this up as I go, haha. I actually based this a bit on what you can see and measure using a 6AL5 diode. Run voltage into the plate, and you can measure the same potential at the cathode with a voltmeter, but if you run voltage into the cathode, you won't see any potential at the plate. I figured rectifying a transformer was going to be the easiest way to demonstrate this, but as you can see in the video, there was so much more going on inside the goal of rectification got lost in the midst of other, weirder discoveries!

[@AccidentalScience]

@@UsagiElectric actually even with a diode I would have choosen to measure the current, providing a voltage source through a resistor. But that's not the point. I wonder why you read an increase in voltage and I'm not convinced it's because the charred wood (as someone else proposed). If it was a flow of charges through the carbon it wouldn't had changed with temperature. Carbon's resistivity raises with temperature, however since you read voltage there was no current, so no voltage drop.

[@UsagiElectric]

@@AccidentalScience The more I think about it, the more I think Janne Peltonen is the closest as to what is actually going on, in that I just made a really terrible air-gap transformer and the fire/heat just got everything moving more freely. That explains why we saw the peaks increase on the scope with the application of fire, and then they stayed increased even after the fire was removed.

[@sketchysticks1127]

dont know if any suggested it yet but make core of salt inside the wire. make a cylinder of paper fill with the desired salt wet the salt and then press it and let it dry then pull away the paper wrap a wire around it. i heard potasium salts is a better ion source for flame triodes. cheers!

[@UsagiElectric]

Oh, that's a very interesting idea! I had read that salt was good for promoting emission in flame diode but I could come up with a good way to get the salt on the wire. I really want to revisit this idea of thermionic emission again sometime in the future, so I'll definitely keep this in mind!

[@danek_hren]

This man went from building electronic projects to designing a working computer with more than 200 vacuum tubes. Amazing!

[@ericmoeller3634]

i hope your garage is not a part of the house

[@jasonl3445]

Awesome

[@UsagiElectric]

Thank you!

[@lunarology9158]

Emission of electrons

[@nakrul987]

even more saaalt!!!

[@homeopathicfossil-fuels4789]

My friend mentions salt vapor being accelerated by flame creating some sort of flow of energy, and that you would get a better result by putting the torch perpendicular to the plate

[@Broken_Yugo]

Most experiments into this effect I've seen used a soft flame in a fixed posistion and the elements were moved around to find proper positioning.

[@nacn5506]

Like! 👍

[@UsagiElectric]

Thanks!

[@dontbitemyface]

If you heat the plate from the other side does the voltage change? If you heat the element from the side instead of head on does it react different?

[@UsagiElectric]

I unfortunately don't have this setup to test anymore, but what was interesting is that after turning the torch off, the waveform shown on the scope didn't immediately decrease. So, I don't think the flame itself was carrying any electrons. Most likely, it was electromagnetic coupling, kind of like a capacitor or transformer. So, unfortunately, no real thermionic emission, but instead just the most inefficient yet coolest transformer ever (because anything that requires fire is cool in my book)!