"The phone is a remarkably complex, simple device,
and very rarely ever needs repairs, once you fix them." - Dan/Panther

Main Menu

The gas tube

Started by Bill, September 22, 2008, 04:19:49 PM

Previous topic - Next topic


We've been discussing the role of the 426A gas tube found in a few 500 sets, and wondering what it did and how it works. I think the following covers it. This is kind of a long presentation, so I thought it appropriate to start a separate thread. Here's the scoop, assembled from many little bits and pieces.

In the most common telephone system, ringing voltage is AC superimposed on the DC line voltage. The ringer is connected to the line through a capacitor, so the positive and negative peaks of the ringing are passed to the ringer, but the ringer has no effect on the line's DC voltage.

Most ringers have two gongs. When not ringing, the clapper sits midway between them.  When the positive peak of the ring voltage comes through, the clapper is pulled one way and hits one gong. When the negative peak of the ringer voltage comes through, the clapper is pulled the other way, and hits the other gong.

However, there is no reason why the clapper has to rest in the middle. There is a special ringer called a biased ringer, in which the resting clapper is pressed against one gong. When the positive peak of the ringing voltage comes through, it pulls the clapper away from the rest position. It hits the opposite gong, and then falls back, hitting the first gong, and sound is heard from both gongs. When the negative peak comes through, the clapper is pulled even tighter against the resting gong, and no sound is heard. As before, the AC voltage still causes the ringer to sound.

Now suppose that we connect a biased ringer directly across the line. And suppose that the central office doesn't use AC for ringing, but sends out a series on pulses, either positive or negative. If the central office sends out positive pulses, the ringer will ring, but if it sends out negative pulses, the ringer will be silent. And now suppose that we take a second identical phone, and reverse the ringer connections inside it. Now if the central office sends out positive pulses, the first phone will ring, and if it sends out negative pulses, the second phone will ring. That's great – the central office now has a way to ring either phone independently, even though they are both on the same line. Think party line usage!

However, there is a flaw in the plan. In order to get this to happen, we had to connect the ringers directly across the line, and this connection will disturb the 48-volts DC that is also on the line. And we can't use a capacitor to isolate it. A capacitor conducts current in both directions, so if you pass a series of positive (or negative) ringing pulses through a capacitor, it turns them into a series of both positive and negative pulses, each having half the original amplitude. What we need is a device that will pass current in only one direction. Today, we would simply use a semiconductor diode, but in the early days of the 500-set, semiconductors did not exist. However, a gas tube conducts current in only one direction, and was readily available at the time. So the capacitor was removed, the gas tube was put in its place. The gas tube is a great blocking device in this application.

If positive pulses are sent, and the blocking device is set to pass positive pulses, then one of our two ringers (the one set to operate on positive pulses) will ring. The other ringer, with the reversed connections, will not. Furthermore, if we reverse the connections of the blocking device in the seocnd phone, so it passes negative pulses, and the ringer is reversed so it responds to negative pulses, and if the central office sends out negative pulses - then the second phone will ring and the first one will not. In other words, if the polarity of the incoming pulses matches the polarity of both the blocking device and the ringer, the ringer will ring. Thus the central office can choose which phone it will ring, by choosing which polarity of  pulses to send out.

This is perfect for a two-party line. And if the ringing voltage (the pulses) is sent between L1 and ground, we can have two parties there, and if the ringing voltage is sent between L2 and ground, we can have two more parties there.

I think this makes sense.




St Clair Shores, MI