• 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.

Fun with glow tubes, a success story

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Had a recent success moment I thought I would share with the community. I spend enough time asking for help, it’s time of give a little back. Plus, it gave me an opportunity to use a piece of test equipment I normally wouldn’t for tube circuits, but in this case it was the perfect fit. There might be some who are planning on using CCS-fed shunt regulators with glow tubes; this little lesson might come in handy for you.

Schematic of my amp is attached: http://www.just4sheep.com/public/300b/schematic.pdf Things are working very well, sound is superb. All glow tubes are firing, and all currents are as expected. Originally, my two power supply current sources were set for about 52mA on the first stage, and 65 mA for the driver stage. This appeared to work well, and gave me continuous glow tube currents of around 10mA, right about what I was looking for. Data sheet lists 5-40 mA as a suggested window.

Problem: the right channel, when the input glow tube would fire, would send a nasty pop to the speaker. Drove me nuts, cuts right to the soul, you know what I mean ?? Left channel was almost imperceptible. So, I set out to find the gremlin. Enter stage right the Astromed Dash-8Xe waveform recorder (sorry for the poor pdf quality). First, a shot of the quiet amp:

http://www.just4sheep.com/public/300b/left.pdf

It’s pretty clear that when the gas tube fires the initial swamp of current through the tube is significant, as it takes away from the current of the push-pull triodes. But this only lasts for a brief moment until the current stabilizes, after which all is good. You can also see the signal at the speaker terminals is almost nonexistent. I would also note that I did do simultaneous recordings of the output stage voltage and current, but did not include them for readability. But the output stage is active. The net noise from the firing glow tube is simply of no consequence.

Repeated these measurements on the popping amp:

http://www.just4sheep.com/public/300b/right-before.pdf

Here, you can see the significant pop at the speaker terminal, as a result of the push-pull triodes dropping to zero current. I noticed two things, which are interrelated. One, the glow tube is reaching a higher voltage before firing, 175V on a 0D3, versus 152V on the working amp. This also has the effect of causing the push-pull triodes to draw more current, since the voltage to that stage is increasing. It would sometimes get really bad, to the point where the stage is drawing 51.6mA from a 52mA current source, leaving a meager half a milliamp for the glow tube to fire (but it WOULD fire, eventually).

Well, that’s easy, I thought. Just select a glow tube with a low firing voltage. And it remedied the problem to an extent, but the pop was still there. In fact, I could measure a firing voltage of 155V on the test bench, but in circuit, it would still rise above that before firing.

Further study of the data showed the problematic stages had a slightly higher idle current given the same size cathode resistor. So while the glow tube would successfully fire, it wasn’t happy doing so with limited current, and would cause a collapse of the push pull triode current to compensate. So I simply increased the magnitude of each CCS (glad I kept those pots in there!) to 57mA-input and 75mA-driver:

http://www.just4sheep.com/public/300b/right-after.pdf

Much better! Now the glow tubes are firing very quickly, very quietly, and at very low voltages (155V and 305V, even when their bench-tested firing voltage was higher). It looks to me like the glow tube enjoys being fed with a CCS as long as you give it enough current headroom, making sure your rising voltage on that stage is not causing the stage to consume too much pre-fire current.

As left glow tube currents are now 15mA-input and 20mA-driver. Still well within the limits of the tubes. I could even increase it a little further if needed, but for now we’ll watch the progress. Would be interesting to hear if anyone has any data on glow tube current vs. life expectancy or noise. Maybe there’s an optimal current to run them. It does appear to me that when a NOS but old tube is operated for a while, it fires easier/smoother in subsequent cycles.

Now I have to stop futzing with this amp and just listen/enjoy.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.