W.E. 306

[poplar1]

Before the 306 tel set and 634AT subset, both with tubes, relays were used instead for 4-party selective ringing---where 4 different subscribers could be rung without disturbing the other parties. Examples are the 98A fiddleback, which otherwise looks like an 85A, and the 297G subset, which resembles a 295A but is longer  to make room for the relay. The 500 series with a 426 tube is coded 501. (Many 306s and 501s have been converted to 302s and 500s.)

I happened to run across a later version of this same ringing method. It is the Northern Telecom QSQE 2500 AX. Instead of connecting the slate and slate/red ringer leads to a capacitor (terminals A and K) as you normally would, these ringer leads are connected to B and D (spare terminals) with a diode between B and D.

Here is the link to the QSQE 2500 AX, submitted to the TCI library by our very own davePEI:

http://www.telephonecollectors.info/index.php/document-repository/doc_details/2399-2500-type-desk-and-wall-phones-ne-2500ax-tl


The direction of current through the diode, along with connecting the ringer to ground and either side (called Tip and Ring) of the line, makes 4 different ways of connecting the ringer. Thus, the correct party's phone will ring. The 4 parties are designated as Tip +, Tip -, Ring + and Ring -. A ringer connected through a relay, tube or diode will only respond to the type of ringing current that it is wired for. The central office sends out the correct ringing for the party selected. On non-dial phones, the operator would have buttons or a switch on her switchboard that she would set before ringing the line. Though somewhat more complicated, dial central offices did the same thing. In some COs, all the parties on the same line had the same phone number, except for the last digit or letter which would select the proper party.

[Bill]

That is interesting. The ringing current is sent out as AC, and most ringers expect to see full-cycle AC. But the presence of a diode implies that the ringer would work properly on half-wave-rectified AC. I would not have expected that.

I was under the impression that when AC was applied to the ringer, the armature (which normally rests in the center) was pulled to one side by the positive half of the AC waveform, and then pulled to the other side by the negative half of the AC waveform - thus striking both gongs alternately. With a diode in the circuit, one half of the waveform would be missing, and this might cause the armature to strike only one gong, albeit repeatedly.

Instead of a center-rest ringer, the phone must have contained a biased ringer, as described in Meyer 2nd Edition page 40. But a biased ringer, in conjunction with a rectified waveform, implies that the bias direction (or the ringer connections) had to be set to match the diode direction. If the match was backward, the armature would have been driven against the gong where it was already resting, resulting in no sound. Thus field installation was a little more complicated.

Am I missing something? Or have I learned something new about ringers?

Bill

[poplar1]

I have heard this referred to as superimposed ringing, meaning the AC ringing current is superimposed on the 48 Volts DC talk battery. However, it would seem that most ringing is this way, other than on ground start trunks; otherwise you wouldn't be able to trip the ringing (make it stop ringing) during the silent interval (between rings).

Even with regular divided ringing of a 302 or 534A subset, for example, you are supposed to connect the red ringer lead to ground for one party and the black ringer lead to ground for the other party, then the other ringer lead (through a capacitor) to tip for tip party or to ring for ring party.  This would seem to indicate that the tension on the clapper would help the ringer while, at the same time, it would make cross ringing (ringing when it should ring only the other party) less likely.

See also pulsating DC---not sure if this applies here but may say something about biased ringers.