• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Violet Glow in Output Tubes

I am expermenting with a homebrew amp. The output stage is two Phillips 6L6WGB's in push pull. Recently I noticed that the tubes give off a faint violet glow that is only visible in dim light. I am not new to electronics (23 years in analog\RF) but am new to tube circuits. I associate a violet glow with pushing the tube and or causing some sort of ionization.

I established that the circuit is quite stable, there is no oscilation of any type. I removed the tubes from the amp and placed them in a test jig consisting of a Fairchild 6200B curve tracer and a Heath Kit power supply meant for powering tube circuits. Using this setup I determined that with plate voltages greater than about 200V-250V the glow would appear. The screen supply had some effect, but it was primarily under plate control. The control grid had practically no effect. The tube was stable in the test setup.

As the glow appears under conditions that are certainly within the limits of the tube I ask this of more experianced tube heads. Is a faint violet glow normal, at least in power tubes?

FYI the bias setup in the amp is 400V/70mA plate, 280/2.7mA screen, -18 Volts VGS
 
Doug,
There are two distinctly different blue glows associated with tubes.
1) If it is the glass envelope glowing, then sit back and enjoy the show. You have nothing to fear. It is nothing more than stray electrons striking the glass, causing it to fluoresce.
2) If the glow is between the elements, down deep within the tube, it is caused by gas molecules being tortured within an inch of their lives. This isn't such a good thing, as you want as near a perfect vacuum in the tube as you can get. If it gets bad, you're about to lose the tube--go ahead and remove it from service. If it's faint (very faint), it just goes with the territory. File it under the 'you can't get good help any more' category. Tube manufacturers aren't as careful as they used to be when there was sufficient competition to keep them on their toes.
One of those FYI things, which you may or may not already know if you're new to tubes: the silvery coating on the inside of the tube is called the getter. It's composed of various highly reactive metals (I can look up which ones if you're curious) which are there to scavenge the residual gas after the vacuum pumps do their work. Now, the getter stays there, doing its job, still scavenging throughout the life of the tube. An occasional gas molecule will sneak in where the pins (or leads from pins, in the case of, say, an octal tube) come through the glass. Or, oddly enough, the metal parts of the tube will outgas (particularly hydrogen, which is very difficult to purge). At any rate, if the getter in a tube ever turns white don't bother turning the circuit on--replace the tube--it's shot.
You can watch this for yourself by taking a dead tube and popping the glass (carefully, please--don't want no lawsuits around here). After a few minutes the getter will have turned from silver to white.

Grey
 
Grey,

Thank you for the very informative response. Another curious thing which I didn't mention is that it is primarily top and bottom, on one side of the tube, and does appear to be near the envelope. After reading you explaination I would say that the glow is in the glass, away from the electrodes. I now assume that is is the envelope flourescing. The getter is in fine shape on these tubes, they were NOS (date code '86) and now only have a few hours on them. Another point in favor of flourescing is that the tube exhibits no sign of break down. The curve tracer is a very sensitive measurement, if something was ionizing it should be the plate voltage causing that as it is the highest voltage. The jump in current draw should appear in the tracer.


Do you know of a SAFE way to flouresce the glass without using the tubes electrodes? UV light maybe? That would definatly prove that its the glass.

I've learned something new today, thanks!
 
Doug,
You can try UV, but I wouldn't expect it to produce the same result. One of the things that I'm bringing to the table from my geology background (ahem...some people would be surprised to learn that other fields are relevant to electronics...ahem) is that there are some minerals that fluoresce in shortwave UV, some in longwave, and some in both, not to mention phosphorescence, which isn't, I trust, at issue here. Glass is essentially a man-made mineral, although the physics behind fluorescence easily generalizes to many other fields. (Did you know that teeth fluoresce? Would you believe scorpions? Good way to find them in the field--portable UV lights at night.) The energies involved in beams of electrons would almost certainly be different, but a) I'm too lazy to drag out the books and do the calculations, and b) I need to fix the dryer pronto so that we can have some clean clothes around here.
You can play with it, but I wouldn't get obsessed unless you simply happen to be captivated by fluorescence. (I find it fascinating, myself.) From your description, it sounds as though you're fine.
Incidentally, a well-made tube has virtually infinite shelf life. It's only when they're put into service that the heat causes outgassing from the tube parts and minute leakages around the pins.
Hope this helps...

Grey
 
ppl,
Brown is okay. There are different combinations of metals that are used in getters, some are more silvery; some are brownish in color. With a brown getter, you'll notice that it gets concentric rainbow patterns as it ages, coming to look somewhat like a soap bubble. When the getter disappears or turns white, the tube is dead. You should replace it before it gets to that point, however.

Grey
 
I had a similar problem with with some 807's in an old RCA monoblock. In my case, the tube was oscillating well above the audio spectrum, the cause was loose sockets acting as small capacitors to the pins. I put in new sockets with snug pins and it was fine. The only tubes I know of that should glow violet are plasma rectifiers.
 
Frank,
In an ideal world, no electrons would escape the plate to strike the glass. Viewed strictly, it's inefficient to waste those electrons. However, given that there are inspection ports and sundry holes into which to insert tabs while assembling the tube, there are always going to be ways that the electrons can sneak past the plate. It's not that the tubes are supposed to glow, per se, it's just an incidental side effect of the way they're constructed.

Grey
 
Just a little to add to previous discussions on tubes going soft.

Grey has explained very well that a perfect tube should maintain its vacuum indefinitely. This is correct. But there are no perfect materials on earth. That is why God is in heaven!

In general terms, tubes with glass envelopes have a long shelf life of about 20+ years before they start to leak. This applies to tubes that have been sitting on the shelf and not turned on for 20 years.

Ceramic metal tubes used in transmitting equipment have a shelf life of 3 years. Although many CX tubes will perform to specifications well past this shelf life, manufacturers rarely recommend using these tubes beyond this.

There are ways of re-activating the getter. Detailed discussions on this topic will be beyond the scope of this forum.

In the early-mid eighties, I used to talk to
Getter Corporation somewhere in USA. This company used to manufacture getters only! I have lost contact with them since. In those days, we did not have the information explosion that is taking place today on the NET. Perhaps, one of our members could post a link to this company if it still exists.

Mohan
 
I had bought some EL509s to build a set of plasma tweeters, not having sockets i tried to quickly solder to the pins with a low temp solder. Aparently this let a little bit of air into the tube, when i fliped on the HV the whole insides of the tube lit up bright purple. It looked cool, but i lost a $24 tube which sucked.