04-22-2013, 06:37 PM
Hi Terry,
Yes, there is a subtle difference in operation between the 2A3 and the 6A3. Although as you say, the filament power used is the same (6.25 watts) the voltage along the length of the filament differs more than twice as much with the 6A3. Its a little bit difficult to explain but I will give it a try.
First think about a tube with an indirectly heated cathode with a heater inside. The voltage along the length of the cathode is the same at any point, which is why it is referred to as a unipotential cathode There is no change in cathode voltage caused by the heater voltage, the filament just raises the temperature and nothing else.
Now take the directly heated filament of the 6A3. Lets assume that we are using DC to light the filament, which in this case is also the cathode. Now we have a cathode which is not unipotential, the voltage varies across its length from 0 to +6.3 v. So at one end of the cathode you have 0 V and at the opposite end you have +6.3 V. This affects the bias of the tube since you effectively have 6.3 V more bias at the positive end of the filament than the negative end. So one side of the tube conducts more than the other. The usual bias of a 6A3 is about 45 v, so 6.3 is a significant difference. With a 2A3, the filament voltage and difference in bias is only 2.5 v and therefore has much less of an effect.
Now consider what happens when you light the filament with AC. Instead of the filament always being 0 and 6.3. the polarity reverses 60 times per second. So there is an AC component of the bias which creates the hum in the output. If you provide the ground return for the filament at the exact center tap point of the filament supply, the AC variation is minimized, now being + and - 3.15 v and it somewhat balances out. But the lower the filament voltage, the lower the AC variation and the resulting hum. So the 2A3 will always have a lower hum output than a 6A3. The lower the directly heated filament voltage, the closer you get to the ideal case of the unipotential cathode of a indirectly heated tube.
Sylvania, the creator of the 6A3 and the later octal equivalent 6B4G, must have realized the problem because they finally came up with the 6A5G. This tube is built just like its predecessors, but has tiny isolated cathodes built around the filament, which are connected to one of the filament pins. By grounding this filament terminal, there is no voltage difference across the cathode, no AC and therefore no hum. But by the time this tube came out, the 6L6 had already been developed which became much more popular as a high power output tube.
Mike, WA2YGA
Yes, there is a subtle difference in operation between the 2A3 and the 6A3. Although as you say, the filament power used is the same (6.25 watts) the voltage along the length of the filament differs more than twice as much with the 6A3. Its a little bit difficult to explain but I will give it a try.
First think about a tube with an indirectly heated cathode with a heater inside. The voltage along the length of the cathode is the same at any point, which is why it is referred to as a unipotential cathode There is no change in cathode voltage caused by the heater voltage, the filament just raises the temperature and nothing else.
Now take the directly heated filament of the 6A3. Lets assume that we are using DC to light the filament, which in this case is also the cathode. Now we have a cathode which is not unipotential, the voltage varies across its length from 0 to +6.3 v. So at one end of the cathode you have 0 V and at the opposite end you have +6.3 V. This affects the bias of the tube since you effectively have 6.3 V more bias at the positive end of the filament than the negative end. So one side of the tube conducts more than the other. The usual bias of a 6A3 is about 45 v, so 6.3 is a significant difference. With a 2A3, the filament voltage and difference in bias is only 2.5 v and therefore has much less of an effect.
Now consider what happens when you light the filament with AC. Instead of the filament always being 0 and 6.3. the polarity reverses 60 times per second. So there is an AC component of the bias which creates the hum in the output. If you provide the ground return for the filament at the exact center tap point of the filament supply, the AC variation is minimized, now being + and - 3.15 v and it somewhat balances out. But the lower the filament voltage, the lower the AC variation and the resulting hum. So the 2A3 will always have a lower hum output than a 6A3. The lower the directly heated filament voltage, the closer you get to the ideal case of the unipotential cathode of a indirectly heated tube.
Sylvania, the creator of the 6A3 and the later octal equivalent 6B4G, must have realized the problem because they finally came up with the 6A5G. This tube is built just like its predecessors, but has tiny isolated cathodes built around the filament, which are connected to one of the filament pins. By grounding this filament terminal, there is no voltage difference across the cathode, no AC and therefore no hum. But by the time this tube came out, the 6L6 had already been developed which became much more popular as a high power output tube.
Mike, WA2YGA
![[-]](https://philcoradio.com/phorum/images/bootbb/collapse.png "[-]")
