There are quite a few examples of how to get an old phone ring. They usually require some 50 V alternating current at 15 - 20 Hz. I’m trying to create that from what I have:
The old telephone
An L298n motor driver
An Arduino
A 240 V to 40 V transformer for Christmas lights
My idea was to use the transformer, I pick it in parts and use only the two coils and the iron core. I connect it the other way around to raise 9 V to 54 V-ish. I would use a 9 V power source (6 * LR14 batteries) for the L298n. I’d create a pseudo AC with the L298n (a square wave) to go into the transformer.
Would I need to filter the square wave somehow to make it more sine like?
The original telephone bill was designed so it could be heard from anywhere in the house. You can vary the loudness by using a lower AC voltage.
The bell uses electromagnets, so the current is more important than the voltage. A square wave current will depend on the time spent at the top/bottom of the wave. Plan accordingly.
I am betting that 6-12 volts AC will be loud enough for your project.
And that is on an old telephone with real metal bells? I might try a computer power source, which is 19 V something. I tried to put 12 V DC on the wires. I heard something move, but it didn’t strike even once. Then I switched polarity and it moved again, but no bell strike. Perhaps with the right frequency it strikes the bells even with 12 V. And with 50 V it wakes the neighbour. I don’t have the H-bridge circuit ready yet.
The bell is mechanically resonant at the frequency it is designed to operate on, so about 20Hz. If you give it AC at the correct frequency it will ring loudly.
Yes, you are correct of course, but getting the frequency right is important. I was responding to your experiment with DC and your observation that it didn't ring. If you get the frequency right it will ring with the lowest possible voltage. Before you ask, I don't know what the lowest voltage is that will operate it, but I would guess maybe 25VAC might be OK.
My brother was a phone man for GTE back when. They had a product called a party line where several customers may be on the same circuit. A party line customer could (but were not supposed to) pick up the line and hear a phone call from another household on the same line. The rule was if you pick up to make a call and you hear a conversation, you should not eavesdrop, hang up and try again later. Each customer had a different phone number, and depending on which phone number was dialed to reach the party line, a different frequency was sent to ring. If you were standing near the phone that was not being called, you could hear the clacker in the phone buzzing a little but since it was not at the resonant frequency the bell would not ring. He had customers with as many as 2 parties on 1 line; they were phasing out party line system at the time. The bells would ring at either 50 or 60 Hz for the phone system he was working on. There are a bunch of different frequency bells. This thread has a variety mentioned. It is for the US, and I’m sure other countries had similar things going on. https://www.classicrotaryphones.com/forum/index.php?topic=11938.0
A H-bridge produces a higher AC voltage than the DC voltage it's powered with.
24V DC gives 48V peak to peak or almost 34V AC.
It should be possible to generate almost 70V AC with an L298.
Leo..
My first attempt is very promising. I get a weak ringing with a 12 V source. I adjust the frequency between 10 and 30 Hz with a trimpot. At 10 Hz it rings nicely, but there’s a lot of noise and just a little ringing from the bells, which means it’s just weak. Raising the frequency will make the bell strikes even weaker until they disappear. And then the movement disappears totally. I guess the sweet spot frequency will be higher, if I use a higher voltage power source. The eigenfrequency of the striker probably depends on how hard it strikes the bells and thus bounces off.
I must prepare a 19 V power source. That will probably give enough juice.
True. The output seems to be 12.8 V, when I set my multimeter to AC voltage.
If my memory is correct, the old CO (Central Office) or PBX ring voltage typically ranged from about 90 to 105 volts AC at a frequency of 20 to 30 Hz (some systems used up to 40 Hz). This high-voltage, low-frequency AC was superimposed on the idle line to signal an incoming call.
The subscriber loop itself was nominally 600 ohms and generally ungrounded. In long-haul or carrier systems the basic arrangement used four wires: one pair for transmit and one pair for receive. At each end of the line a hybrid (or "duplex" unit) converted the four-wire circuit to the familiar two-wire interface. The hybrid network inside a telephone set is often called the "network interface" or simply "the network."
Traditional telephone ringers were high-impedance AC devices. When the handset was lifted (off-hook), the DC resistance of the phone dropped significantly. This increase in loop current effectively shunted the line, which stopped the ringer and also signaled the CO that the subscriber had gone off-hook, completing the loop and initiating call supervision.