I Built A DIY Vacuum Tube Diode

winter is coming in my part of the world which means that the shop is uninhabitably cold luckily last year I installed this fireplace which doesn't even come close to heating the whole building but does give me a cozy spot to warm up between adventuring into the colder recesses of the lab a couple of really cool things have happened since the last video update one is that the channel passed a thousand subscribers this is a pretty Niche area of Interest so it is cool to know that there are so many people keeping tabs on what's going on on the channel and welcome to all of the new subscribers the other cool update is uh well this that that is a homemade vacuum diode so that was me presenting my successful vacuum diode build at hackaday Super conference in November if you follow the patreon then you know that I submitted a talk sort of covering everything that I've learned this year on the channel about making your own vacuum tubes when the talk got accepted I decided I was going to drop everything and try to make something to take to the conference as a demo even if it didn't work and in some ways it didn't work having made a successful vacuum diode is sort of a positive spin on the fact that I made a failed vacuum fluorescent display that said it does work as a directly heated diode and the tube is still under vacuum it hasn't gone gassy in the weeks since I put it together regretfully I I didn't document this build as well as I could have because this was really a Hail Mary pass having a demo for super conference and I honestly didn't expect it to work in fact I think it probably only worked to spite me for not documenting it better the first thing that I did was to finish building the spot welder I say finished but there are still some bugs to work out I'm not sure whether or not I'm going to make a YouTube video about this build because I don't know how many videos YouTube really needs about microwave Transformer resistance welders but if you're interested leave a note in the comments and maybe I'll get around to doing that once I was able to weld the wire elements together I went about constructing the tube basically using the methods laid out by JD flyback on his YouTube channel here you can see my pass-through made from stainless steel wire with fine tungsten wire for the actual glass seal I'm using little pieces of nickel sheet as a sort of glue metal between the stainless steel and the tungsten to try to get them to weld together and in fact I'm not sure that I succeeded in this what I ended up doing was creating Little Nickel ferrules around the tungsten that sort of crimp it to the stainless steel and I believe I just sort of welded that ferrule to the stainless steel trapping the tungsten in place this ended up working but it's something that I'd like to improve going forward once that was put together I oxidized and then cleaned the tungsten Wire by just passing it through a torch flame until it was red hot a few times and then hitting it with a Q-tip and a little bit of electronics cleaner that whole assembly sat down into this piece of glass tubing that I necked down in a propane torch flame and then I heated that up and did the pinch this was literally my first attempt at creating this hermetic feed through and Against All Odds it actually worked and that's despite the fact that the tungsten really doesn't have that nice straw yellow color that seems to be characteristic of tungsten bore silicate seals anyway with the pinch made and no real way to vacuum test the stem I just forged ahead building the rest of the tube here are the rest of the elements while it in place on top of the stem that skull and crossed wrenches is called the Jolly Rancher it's the hackaday logo I thought it would be a nice touch for the conference I ended up having these laser cut by oshkot out of nickel sheet and I had them made in two different sizes and I think they cost a few bucks a piece for future projects I plan to produce things like this in the lab using a photo etching technique as you can see the anode has a healthy coating of white powder that's the phosphor that I got from phosphor Technologies I attempted to apply the phosphor by mixing it with mineral spirits and then painting it onto the anode allowing the mineral spirits to bake off and leave this sort of thin layer of phosphor powder behind and I think this would work at lower phosphor concentrations so that you get a thinner layer but ideally there would be some sort of binder there you can see in the more recent pictures of the tube that a lot of the phosphor ended up falling off of the anode I think the next time that I try this I'm going to use a little bit of nitrocellulose in acetone is my binder system and we'll see if that fares any better I think I actually heat damaged the phosphor during glass work and that's ultimately why this ended up being a diode and not a vacuum fluorescent display at any rate once that assembly was finished I slid the whole thing into another glass tube which I had flared so that it would fit over the base I took those two things to the torch and heated them and pulled them apart such that the flare and the base were melted together and the leads were exposed which meant that now the tube was sealed on one side and then I took the other side and heated it up and stretched it out to create a neck where I could perform the tip off and then of course the end of the tube would go into the O-ring fitting on the evacuation Port of the vacuum system and here it is on the pump which I'm running in essentially the most Reckless way possible I would not recommend that you endanger yourself or your turbo molecular pump in this way if you can at all avoid it but here we are pumping down past 10 to the minus 4 millibar heading towards 10 to the minus 5 while the tube was on the pump I did my best to bake it out I hit all of the glass envelope with a propane torch and then heated up all of the metal components with an induction heater and I even lit up the filament on a bench top Supply to Orange hot just enough to try to drive out any absorbed oxygen molecules that might be hanging out on the tungsten filament once the vacuum pressure stabilized and I had made all of the appropriate ritual sacrifices I attempted the tip off I heated everything up real nice with just a standard propane torch until the neck of the tube started to bend over and then I focused my heat on that section and gave it a couple twists before pulling it off and it stayed sealed I tried to do the tip off in such a way so that both the tube and the evacuation Port stayed sealed during tip-off and I did manage to do that which is good because it is a straight shot from the evacuation Port down to the turbo molecular pump and I really did not want to have some sort of blowout and once the tube was sealed I got to do the fun part which is to flash the getter and the getter flashed beautifully leaving a nice mirrored surface on the inside of the tube I used this mini induction heater that I bought on eBay for under two hundred dollars it's marketed as being a bolt heater for removing rusted Bolts from cars and it works great in this application of course as soon as the getter was flashed I took approximately a million pictures of the finished tube because I fully expected it to go gassy by the next day and I wanted to capture it while it still looked nice shockingly the next day it was still intact so I put it on the bench for some testing all right I'm playing with the tube and it's not a vfd but it is a diode so I have two bench supplies over here this is filament and this is plate the Plate's sitting at 60 volts right now and then we have the filament is voltage limited we've got it set up here in the clamp and this meter is reading the current in milliamps across the tube so this is current flowing from the filament to the plate through the vacuum and if I crank up the filament Supply there it goes if it sits here for a while we will get up to two milliamps of conductance through the tube uh and then of course we need to verify it's a diode right so we're going to reverse the banana plugs without shocking ourselves we've reversed the current across the tube we're going to do the same experiment again all right we're up to a little over two two and a quarter amps 10 volts across the filament and no current across the tube it's diode it's a diode all right we did it that's A working vacuum tube I know that I glossed over a lot of the details about the construction of this tube even though that's sort of what this entire YouTube channel is about and that's mostly because this was sort of a quick and dirty experiment and not really a demonstration of how I think this should be done once the vacuum bench is finished and I've worked The Kinks out of a couple of other tools in the shop I plan to make a video where we will attempt to build a triode and in that video I'll go into much more detail about each step of the build speaking of the vacuum bench we are getting dangerously close to having a working system in fact why don't I just walk over there and show you what we've got foreign so this is what the bench looks like at let's call it 98 complete there are a few little things missing I'm still waiting on stuff to ship not everything's wired together but all of the major components are here and most of the interconnect so I think it's a pretty reasonable time to give you a system tour and for that we're going to start underneath the bench with the roughing pump this is my Welch 1402 rotary vein pump its job is to provide backing pressure for the turbo molecular pump eventually it'll hook up to this Bellows but I don't have it hooked up yet because I plan to disassemble the whole system and give it a thorough cleaning before everything gets hooked up and actually runs right now I have the inlet capped off because dust and the outlet is waiting on a proper oil Mist Eliminator currently it is stopped up with a rubber glove tied around some shop towels as long as we're down here I might as well adjust the air compressor an air compressor isn't strictly necessary on a vacuum system but I'm using pneumatically controlled valves for all of my high vacuum valves which means that I need some source of compressed air to actuate the valves now I could use the shop air here in the lab but I like the idea of having a compressor on board the system that can move with it it also means that my vacuum valves are never competing for air with other tools that I might have running in the shop looking up at the bottom of the workbench we can see the rough vacuum end of the system this is the other end of that Bellows that's connected to the mechanical pump it feeds into this cross and this cross has a couple of valves connected to it this valve is used to vent the four line after we shut down all of the pumps so that oil isn't pulled from the mechanical pump up into the turbo pump or any other part of the high vacuum system where there might still be a vacuum this is one of the two valves that isolates the turbo pump and of course this is the turbo molecular pump itself on this arm of the Cross is another Bellows that bypasses the turbo pump and goes straight to the top of the system connecting to the evacuation port in the event that the evacuation Port needs to be isolated and then brought up to atmospheric pressure let's say I have a tube that doesn't pass leak check and I need to do some work on it and then put it back on the system it'll be necessary to rough down the evacuation Port before I expose it to the turbo again so this bellows in concert with a couple of other valves will allow me to isolate the turbo rough down the evacuation port and then open the valve between the evacuation port and the Turbo and put the whole system back into high vacuum now on the side of the pump body here we can see a few interesting things one is this wild electrical connector which hooks up to the controller which I luckily have on hand over here we have a grounding log down here we have some cooling fins which I'll have to hook up a fan to blow across and then here we have this weird little eighth inch npt plug now the reason this is here is it's for a bleed down valve this actually enters the pump between the turbo and the drag stages and allows you to flow nitrogen or some other dry gas Source into the pump to help slow it down when you shut off the system now I probably won't have a bleed down valve on my system but I don't want to permanently plug this thing up because in the future I may change my mind so right now this brass plug is is crammed in there and part of the reason why it's in there with a bunch of Teflon tape is because the threads on this thing is pretty gummed up and that wasn't my doing it was like that when I got it so my plan is actually to replace this plug with a few other fittings that I have here me show you this is an eighth inch npt to kf16 adapter that I have connected to a kf16 bleed down valve and this is what I'm going to put here in place of this eighth inch plug now currently the plug has a bunch of Teflon tape on it but I'm going to replace that with some apis on high vacuum wax hopefully once this is in place and covered in vacuum wax I'll never have to touch it again I could just take this clamp off and replace the bleed down valve with whatever I want whether that's a dry nitrogen Source or just a kf-16 end cap the turbo itself is mounted to the workbench a lot like a router would be mounted to a router table I had a quarter inch thick eight by eight inch aluminum plate machined with the bolt pattern for the iso 100 flange on the pump then I put the bolts for the claw clamps that hold the turbo pump to the reducing flange above it through that plate before putting them into the clamps that way the flange the pump and the clamps are all bolted to this plate and then the plate can be bolted to the top of the workbench here you can see where the reducing flange exits the top side of the workbench the plate is bolted to the workbench top with four bolts at the corners and I actually used a router to route a pocket in the top of the workbench the plate sits down in so that it's relatively flush with the top of the workbench I had a large enough hole made in the middle of the mounting plate so that if I wanted to move up in size from a kf-16 to say a kf-25 or a 40 or even a 50 there would be enough room to do that so now we've reached the top end of the system which spends most of its time at high vacuum this is where the evacuation Port is it's basically the business end here is the inlet to the turbo pump which we just saw coming up through the table and then it feeds into a cross now this cross has all of our gauges mounted on it that's an Edwards PR 10K pirani gauge as well as a Dunaway stockroom Bayard Alpert ion gauge these gauges basically belong to the turbo pump and what I mean by that is that above them is another valve that in concert with the valve on the outlet of the turbo pump can be closed to isolate the turbo pump from the rest of the system and when that happens these gauges get isolated with it because the ion gauge can't be operated at Rough vacuum or at atmospheric pressure so it has pretty similar operating constraints to the turbo pump so anytime that I want to isolate the turbo I basically want to isolate the ion gauge too otherwise I'd have to turn it off and let it cool down above that isolation valve is another cross and this is where our evacuation Port is the evacuation Port is where we're going to connect our devices that we want to pull down the vacuum as you can see I currently have the stem from the first successful tube still stuck in this Ultra torque connector actually this isn't an ultra torque connector this is a 16 millimeter ion gauge connector and it's way cheaper than an ultra tour and it connects nicely to the 16 millimeter outer diameter bore silica tubing however if we're evacuating a device that has a smaller evacuation stem we can pull off this kf-16 connector and we can replace it with one of these Ultra tours this is a quarter inch and this is uh probably 3 8 something like that also of note the kf-16 centering ring that I'm using directly under the evacuation Port has two features that are pretty important one the O-ring itself is viton not Muna in which means that it can be baked to a higher temperature when I'm baking out the tube secondly there is a mesh in the centering ring that will stop any large objects from falling through the evacuation port and down into the turbo just in case something should go catastrophically wrong during tube evacuation the evacuation Port is mounted on across so that I can have an auxiliary port on the side this port will be useful if I want to hook up some other instrument to this and just use the bench as the vacuum system or if I want to flow in some sort of gas to backfill the tube with say neon or r gone or hydrogen or something like that on this arm of the Cross is another T and connected to this T is another Edwards Perani gauge now the reason I have another pirani gauge that's monitoring the pressure of the system between these two valves is because if this portion of the system has to come up to Atmosphere while the turbo is isolated I'll need to rough this back down so I need some way to monitor the pressure in here to know that it's down to roughing pressure so that I can open this valve and then expose the turbo and the hot ion gauge to the volume inside of this portion of the system finally after this elbow is another valve which works with this one to isolate the top end of the system where the evacuation Port is and then that feeds down to the Bellows that we saw earlier that connects to the cross on the bottom end of the system the rough vacuum end of the system and finally we have a 19-inch instrument rack into which I've mounted the ion gauge controller which also acts as our pirani gauge controller for both of the Perani gauges now as well as giving me the analog readout here visually it also gives me an analog voltage readout for each of the vacuum gauges which I'm going to feed into a process controller which is going to live here the process controller is going to be a 2u rack mount PCB that has an Arduino on the back side and then on the front it has a number of buttons as well as a visual indicator of the pressure in every part of the vacuum system it's going to read all of the gauge values as well as user input from the button panel and it's going to control all of the valves as well as the pumps and make sure that no part of the system is exposed to too high or too low a pressure the process controller will be able to control all of the valves in the system using this little manifold each of these solenoid valves is controlled with 20 24 volts DC which I can get from the same meanwhile power supply that I'm using to power the turbo pump this manifold will get pressure from the half gallon oil-less compressor that I showed you on the bottom of the workbench and then we'll selectively allow that to go to each of the pneumatic valves you can see I had four valves on the system I only have three solenoids but there are two valves on the system the one on the top and bottom of the turbo pump that will never need to be actuated independently so that those two are going to be getting together on a single solenoid all right I think that's about it for this Fireside edition of signal ditch I want to give a shout out to everyone who is subscribed to me on patreon I had to buy a car recently because my trusty pickup truck finally died on me that means there's about 500 bucks a month which could have been going into this channel which now is going into paying off a high interest loan all of that is to say that the patreon money really has come in clutch in the past few months when it comes to buying odds and ends for the vacuum system just recently the patrons paid for the air compressor on the vacuum bench as well as the 19-inch equipment rack that's on top they also paid for the machined aluminum plate that mounts the turbo pump to the table that ain't nothing that's stuff I wouldn't have been able to afford to do if it weren't for all of the support that I get on patreon if you want to join the patreon and help out the channel you can find me over at patreon.com integratedtherm any amount of support gets you early access to the YouTube videos as well as the ability to message me there on patreon and talk about tube stuff I also post the occasional behind the scenes update or just interesting piece of research that I've dug up on vacuum tubes My triode Level supporters get their names added to the end of every YouTube video like you're seeing right now if you found this video entertaining I really would appreciate it if you'd hit that like button and feel free to subscribe to the channel for updates on how this project moves along we're pretty close to the point now where we can start seriously experimenting with different methods and materials for making electron devices in the home shop I've got about 10 different project ideas that I want to start as soon as the vacuum bench is ready but until then Happy hacking and I'll see you in the next one

Channel: Signal Ditch

Views: 5,622

Keywords:

Id: 8vipZG4cozA

Length: 21min 4sec (1264 seconds)

Published: Tue Dec 20 2022