A Vacuum Diode for Make: Magazine
Recorded: Sept. 14, 2026, 11:01 a.m.
| Original | Summarized |
A Vacuum Diode for Make: Magazine | Nick Poole Nick Poole Explore Website About CV Youtube BeanCounter GreatBigFactory RSS Search Open Sidebar DIY_Vacuum_Tubes A Vacuum Diode for Make: Magazine Scroll Down After my presentation at Hackaday Superconference 2022 I was asked by my friend David to write a piece for Make: Magazine. I originally wrote the article around the diode tube that I presented at Supercon. However, because I had built that tube hastily in the days before the conference, I didn’t have a lot of good photos of the process. I decided the best thing to do was to make another tube specifically for the Make: article and take care to document the process this time. Pins being prepared in my extremely high-tech clamping fixture Once the pins are prepared, we need a piece of glass to pinch them in. The glass needs to be a smaller diameter than the tubing for the envelope of the tube so that once it’s pinched, it still fits inside. That’s why the end is flared, to make it easier to join with the envelope. Behold. A bad, small, trumpet. To make the flare, I simply chucked up the tubing in a power drill and heated the end with a MAPP gas torch while turning it. I nudged it gently with a tool to get the end to flare. Ideally, I would do this in the glass lathe, but I haven’t made a collet to hold these small parts. After the flare is made, I pre-flatten it just enough to support the wires before the pinch. Here you can see the way that the pinch will fit in the envelope. To actually make the pinch, the pins are inserted into the flare and then the flare is heated to softness with an oxy-propane flame. I usually make a preliminary pinch, to capture the pins in place before heating the glass more thoroughly to make the final pinch. I’ve tried various tools to do the pinch. First I used smooth jawed pliers, but it’s difficult to make an even pinch that way because they don’t close parallel. Then I tried large tweezers, because they’re slightly more parallel but it’s difficult to put enough pressure on the pinch. I have my sights set on a pair of smooth jaw, brass parallel pliers which I’ve yet to try, but I suspect will work quite well. This seal may not be pretty, but it did work! Assembling the Active Elements Ready for action Now I just need to stuff all of this into a piece of glass tubing. Here’s an idea of how the flare fits into the envelope. Here's the fitup of the unwelded part While the piece is upright like this, the flare was tacked in place with a small oxy-propane flame. Then the piece was moved to the lathe so finish up the glass seal around the base. Careful heating and gentle pressure to prevent it from collapsing resulted in a serviceable joint. ...and now with the welded joint. This tubing is sized to fit into a quick connect on the vacuum system, but to aid the separation, I heat the tubing above the assembly and pull it to make a neck. This neck in the tubing will make it easier to close off after evacuation. Big Succ Pumping down to 10^-5 mbar During pumpdown, the tube is heated with a propane torch to drive off any adsorbed water. The internals are heated with an induction coil. Finally, the filament is powered to a dim glow to drive out oxygen. Signs of life! This is always satisfying And now the step where I can ruin everything extremely quickly! Gently heating the neck of the tube with a propane torch until it begins to soften and close in on itself, I use my other hand to twist and pull the tube away from the vacuum system. If all goes well, both pieces of glass will remain sealed, and the tip-off will be complete. Only one thing left to do and this tube is finished After the tube is tipped off, the only thing left to do is to fire the getter. To do this, I use an induction heater designed to loosening rusted bolts. I bought mine on eBay for under $200. All you have to do is put the coil around the tube and line it up so it’s concentric with the getter ring. After a few seconds, the ring will get red hot and you’ll see a mirror developing on the glass above it, this is the getter material. This mirror of reactive Barium metal will react with oxygen in the tube to keep the vacuum clean. The diode is complete! G(B)lowing Filament Look at that beautiful glow Of course, in the process of taking these photos, I got a little carried away with the filament current and tested it to destruction. That’s fine, I got the photos I needed. Besides, look at the beautiful Tungsten mirror it made: If you're not testing to destruction, are you even testing? I’m looking forward to building a curve tracer so that I can properly characterize these devices, but for now I’m just happy that several months later this tube still hasn’t leaked up. Share: Next Search Search for: Close Explore Site About About Nick Poole Contact nick [at] signalditch.com Close Sidebar Recent Updates Year in Review With Rising Tide Research Foundation September 5, 2026 Upgrades to the Vacuum Lab and Cleaning for HV September 2, 2026 Building a Private Lab and Workshop September 2, 2026 Projects Assembly GitHub Back to the top Nick Poole © 2026. Curious theme by JustGoodThemes. |
The process described details the fabrication of a vacuum diode tube, emphasizing the challenging steps involved in creating the base and achieving a high-vacuum seal. The initial critical step involves fabricating the base, specifically addressing how pins pass through from the atmosphere to the high-vacuum interior. The preferred method utilized by the author, Nick Poole, is the pinch method, which requires carefully selected materials and precise thermal treatment to achieve a robust seal between metal and borosilicate glass. To ensure a vacuum-tight seal, the feed-through metal, preferably Tungsten, must be meticulously prepared. This involves cleaning the Tungsten wire to remove impurities and treating it with a flame to generate a uniform oxide layer, which creates the necessary bond with the glass. The author details a specific cleaning protocol using alcohol and heptanes, followed by heating to a white-hot state and carefully removing loose oxide residue with a swab. Furthermore, the structure relies on spot-welding short sections of Tungsten onto longer Nickel wire, as Nickel is more mechanically forgiving. This setup allows the Tungsten to wet the glass while the Nickel provides mechanical integrity. The geometry of the tube envelope requires a flared end to facilitate joining with the outer envelope. This flare is achieved by heating the tubing with a MAPP gas torch while manipulating the tube, aiming to gently deform the end. The insertion of the pins into this flared section requires thermal expansion from an oxy-propane flame to achieve the final pinch. The author experimented with various tools, such as smooth jawed pliers and tweezers, seeking a method to create a parallel and even pinch, which is essential for a functional seal, ultimately suggesting smooth jaw brass parallel pliers might be ideal. The assembly of the active elements involves creating a base structure that will support the diode. This base incorporates a tungsten filament and a coil of wire, fashioned from Nickel wire, which serves as the plate. These components are spot-welded to the pins on the base, followed by thorough cleaning. To maintain the integrity of the vacuum, an evaporable getter, composed of Barium metal, is applied to the assembly. This getter is designed to sequester any oxygen migrating from the metal during operation. Once the base is prepared, it is fitted into the glass tubing. The separation process is managed by heating the tubing above the assembly, which allows the glass to soften under gentle pressure, enabling careful separation of the glass pieces while maintaining the integrity of the joint. An intermediate step involves heating the tubing to create a neck, which is intended to aid in the separation process after evacuation. The final stage involves evacuation, where the entire assembly is placed into a compression fitting system. During this pumpdown, the internals are heated with an induction coil, and the process is driven by heating adsorbed water, driving out oxygen from the filament, and ensuring the vacuum is maintained. After evacuation, the reactive Barium metal getter is activated using an induction heater to create a mirror on the glass, which reacts with residual oxygen to sustain the vacuum integrity. The successful completion of this procedure results in a functional diode, demonstrating the established principles of high-vacuum fabrication. |