Vacuum Tube Repair After A Spectacular Failure

[Eric] has an Atwater Kent 55C AM radio from the early 1900’s. He’s been trying to restore the radio to proper working condition. His most recent pain has been with the rectifier tube. The tube is supposed to have a complete vacuum inside, but that’s not the case here. When the tube is powered up, it glows a beautiful violet color. It may look pretty, but that’s indicative that gas has leaked into the tube. It needed to be replaced.

[Eric] had a tube that would serve as a good replacement, but it’s plug didn’t fit the socket properly. He was going to have to use this old broken tube to make an adapter. Rather than just tearing the old tube apart, he decided to have some fun with it first. He hooked it up to a variac, an ammeter, and a volt meter. Then he slowly increased the voltage to see what would happen. The result was visually stunning.

The tube starts out with the same violet/blue glowing [Eric] experienced previously. As the voltage increases, it gets more and more intense. Eventually we start to see some green colors mixing in with the violets. [Eric’s] reaction to this unexpected result is priceless. As the tube gets increasingly hot, the anode starts glowing an orange-red color. Finally, the filament starts to crackle like a sparkler before the tube just gives up and completely fails.

After the light show, [Eric] moves on to replacing the tube. He begins by tapping on the old tube’s socket with the end of a screwdriver. After much tapping, the glass starts to come lose from the socket. After a bit of wiggling and twisting the tube finally came free from the socket. [Eric] luckily had an unused octal socket that fit perfectly inside of the old socket. All he needed to do to build his adapter was to connect the four pins from the old adapter to the proper pins on the octal socket. Piece of cake.

…Or so [Eric] thought. After testing some new tubes with a tube tester, he realized he had soldered all four pins incorrectly. On top of that, he had super glued the adapter together. He eventually got the two pieces apart. This time he removed all of the unused pins from the octal socket so he wouldn’t get it wrong. Another run on the tube tester confirmed that everything looked good. After plugging the tube into the radio, it worked just as expected

If you need fabrication rather than repair, we’ve got you covered there as well. Check out [Charles Alexanian’s] process for making new vacuum tubes in his garage. Now if you just have too darn many of them around, you can always decorate your pad with ’em.

33 thoughts on “Vacuum Tube Repair After A Spectacular Failure

  1. Is the tube a mercury vapor tube? Some rectifiers were. The mercury vapor ionized and dropped the resistance of the tube. Once ionized, the mercury vapor glowed purple.

    1. The clue is the silvering on the glass. If gas gets in you get a loss of silver from a small plate due to some reaction which I have forgotten. I remember mercury vapor rectifiers, but I think they were much higher current and voltage than this one.

      1. The silver “getter” flashed on the glass, usually barium metal, is actually supposed to be there, it’s supposed to be shiny in a functional tube. In a tube with a significant air leak (which is more than the getter can compensate for) it won’t be shiny any more, it will turn a dull white color when it’s oxidised.

  2. Being im not very knowledgeable on tubes is it possible to actually repair it by finding the leak and restoring the vacuum or does the loss of vacuum result in internal damage? It’s age brings out the preservation in me :D

    1. It is possible but you’ll spend a lot of money and time doing it when a replacement tube is cheap. Very high power transmitter tubes are repaired but they’re extremely expensive in the first place so you can justify it.

      You need a very good vacuum for the tube to work properly and that requires significant effort . The getter material inside the tube will be nearly useless now as it’ll be mostly oxidised so you’ll need to replace that also.

      1. This “light show” is so awesome, it’s kinda tempting to see if tubes could intentionally be refilled with gas, like neon lights, but with all the cool metal-stuff inside that makes it look like the friggin’ starship enterprise going to warp-speed.

          1. Nice, thanks! I’m tempted to go check out part-numbers on the box full of tubes I’ve got. Though, as I recall, they were all from an old radio or two, so probably no v-regs necessary :/

  3. That slight bit of gas and the accompanying glow can actually serve a useful purpose. My first ham radio transmitters were WWII-era Command sets that’d been installed in bombers. They had two 1625s tubes (the 12-volt equivalent of an 807). I tuned up my radio by adjusting a large capacitor for minimum blue glow. No need for a meter in the plate circuit.

    I had to be very quick though. The tubes were rated for 600 volts on the plate, but to squeeze out a bit more power, I was running them at 750 volts (and briefly 1200 volts). Within a couple of seconds, the anode plate would begin to glow if it wasn’t properly tuned and after about 5 seconds the tube was kaput. I didn’t worry. I could order more from surplus stores for about 20 cents each.

    Those were the days. Transistor finals are so dull. They don’t glow or turn cherry red, they just get warm.

  4. This reminds me of my final year project at secondary school, where I spent a lot of time messing about with discharge lamps and spectrometers. The violet glow looks like hydrogen emission to me, though without a spectrometer it’s impossible to be sure, of course.

  5. The violet glow would be Nitrogen and the green is Oxygen. As the voltage is increased the electrons become more energetic and are able to ionize oxygen as well as nitrogen. There is not much Hydrogen in the air.

      1. Could be, but I suspected it was from the air that leaked in. The same sort of glow occurs during the auroras when charged particles (protons and electrons) from the sun during a solar storm travel along the earths magnetic field lines and strike the upper atmosphere.

  6. That tube is defiantly gassy. Not Mercury vapor. Interestingly enough the getter still looks good, but who knows what mix they were using. Very typical of rectifiers of that era. By the late 40’s they had them worked out pretty good and they approached what one might call bulletproof, or as much as could be said about a tube anyways. Those tubes were replaced by the 80 rectified, and later the 5Y3.

  7. Lame guy plus a bad potty mouth. I can’t believe it that he screwed it up with the pinout and breaking the glass. Rather than frying the tube a more interesting approach would have been trying to repair the old tube by trying to pump the gas it and redo the vacuum. But ok, he appears to be a red neck do not too many expectations…

    1. Tube was gassy, there was so much air leaked in that it could not produce any x-rays before arcing over. X-ray tubes need much over 30kV in a good vacuum, and in this video the tube was not even given 300 volt. Only normal vacuum tubes that could withstand enough voltage and have good enough vacuum are CRTs and high voltage rectifier tubes.

  8. i have bags of gassy and not working 845 and 211 valves sat on my desk that i cant decide what to do with. i was going to cut them all open and use them for insect displays but now i might see if i can link them up in a way they will sparkle away…
    any ideas?
    i have about 60.
    i also have about 10 faulty 833 valves.

    any ideas people?

  9. How faulty?
    833’s are just too cool to discard, some faults (ie H-C short) can be repaired or worked around with some clever kludge-fu and besides they would be a great conversation piece at Nerd Nights.

  10. Hate to tell you this, but you most likely destroyed a perfectly good tube.

    If you would have just plugged that tube in, and operated it normally, it most likely would have cleaned it’s self up reasonably well after several days, or weeks of operating.

    The getter (silver stuff inside the glass) was still in good condition, and no signs of consumption. So a few days of running at full temp, and the getter would collect most of the free gas in the tube.

    The gas only builds up during a slow leak when the tube isn’t being used. First time you use it after a long hiatus, then most tubes will be a bit “gassy” until they have ran long enough to clean themselves up.

    When most of the getter has been consumed, and the tube is still gassy, then yes, it’s time to be disposed of. On the other hand, not that tube.

    1. Nope. That tube was a gone’er. The heater was burned out. No amount of hoping would have saved it.
      The “getter” is not the silver stuff on the inside of the tube. The getter is a filament which, after the tube hase been constructed, is heated up to a tremendous heat, thus burning up all the oxygen in the tube, creating a vacuum. Once used, the getter itself is often burned open, never to pass a current again.

  11. I’ve seen this failure mode a few times. It should be noted that if you run it in the radio other components can also be damaged (eg other tubes, coupling capacitors, grid leak resistors)
    Its also a common failure mode on certain CRTs notably the ones from console games, something about running it with the tube vertical eventually damage them.

    1. Agreed. In fact, I just now finished the restoration of a Grinsby Grunow 1929 Majestic 130-A with the exact same fault in it’s 280 tube.
      What we are seeing is the HT on the plates of a tube with a burned out heater. Without the heater, the HT cannot disperse to the rest of the set. Remember, on tube rectifiers, it’s the heater winding which carries the B+ to the circuit.

  12. If the getter has not turned white and the filaments are good then put the tube in an oven at 100c for 24 hours. This will allow the gas to be scavenged. Then 30 mins with only the heaters but no B+ and you should have a good tube at the end. This is not a bad technique to clean up any NOS tube before using them. Also gives a longer life as frequently tubes will get gas in them over the years which is stored in a cold location will not be removed by the getter.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.