Hi Guys,
I finally got round to using a HBridge to drive a phone ringer solenoid,
and the results are reasonable with a 12 Volt supply.
The circuit is three chips and no transformers.

https://www.youtube.com/watch?v=g2SauckwBfU

Cheers, Brek.

Nothing quite like a mechanical bell is there.


Ben

Pretty good.

Now, if you can add a chip to control it so that it rings the phone with the official Oz ring cadence (400 ms on, 200 ms off, 400 ms on, 2000 ms off) at 16.66Hz (say 17Hz) a number of people including myself, would be interested in constructing it.

Easy with a micro, Interesting challenge for stock logic I’ll see what I can do.
The frequency that rings the bell best, I think depends on the circuit driving it.
In the video, I didn’t measure f, but I’m pretty sure it was higher (but can you tell)?

Three more logic chips... 555 driving a 200ms clock,
2x4017 cascaded decade counters.
Outputs 0,1,3,4. connected to the H-Bridge enable pin,
Output 5 of the second counter to reset both decade counters.

The frequency that rings the bell best, I think depends on the circuit driving it.

Yes, that's true. From distant memory, in exchanges the ringing voltage can be anywhere between 50 volts and 100 volts, and it's a sine wave, and the tolerated frequency range is 15Hz to 25Hz.

IIRC, 16.66Hz is the result of dividing our 50Hz mains frequency by 3, which was a convenience at the time.

The ranges came about because of the nature of exchanges over the decades, from manual switchboards with magneto ringers on them, to computerised equipment of recent times, and everything in between.

Back in the day, the exchange ringers were often massive motors with cams on them that generated/governed dial, busy and number unobtainable, tones as well as ringing the bells.

Example: https://www.youtube.com/watch?v=VzIXvO6RahQ&list=PLgNzI09bRXtlv9-M7mrABEALyhplLpd-G&index=39

There are chips that contain a pair of 555s., so with two decade counters added it’s 5 chips total.
But a single pic micro can do the job of 4 of those, and it’s down to two chips,
but not everyone can program the micro, so it’s your choice GTC.

Wow that takes me back a long time GTC, I worked for the British Post office (GPO Gods poor orphans) and it was standard to have 2 ringers per exchange in those days (1966). How things have changed eh.

Good work Brek.

I had seen H-bridge circuits and wondered if they could be used to ring a phone bell. You have shown that it does work.

You will be able to improve the ringing by rotating each of the bell gongs slightly so that they are just clear of the clapper when the armature is pushed to each side. The idea is that in ringing the clapper flexes slightly to just tap the gong at the end of each stroke.

I have covered ring frequencies in another post but 16-2/3Hz was the old standard and later it changed to 25Hz. The voltage was 75V to 90V. The bell motor in a 400AT can operate (just) at 50Hz but it sounds more of a buzz.

GTC is correct. The early exchanges had US made sub-cycle ring generators (a type of oscillating transformer) that output 90V at a third of the supply frequency. In the USA that was 60Hz so the output was 20Hz but here on a 50Hz supply the output was 16-2/3Hz. Later exchanges had motor-generators that were designed to output 25Hz.

Andrew

a single pic micro can do the job of 4 of those

A PIC micro would suit me fine.

Here in Australia the ringer voltage is 100 volts no less . Then 50 volts for the speach circuits. I'm Ex Telecom. I wasnt aware that you cant run a external bell on the newer systems. I had a massive external bell on mine when I was living in Kootingal near Tamworth. Late night calls did not make me a popular neighbour lol. I can also remember when we pulled the old gear out of the second storey the floor bowed up because it was prestressed to carry the weight of the old gear.
The hundred volts DC was delivered via contacts opening and closing via a Cam driven constantly via a electric motor .

Hi Andrew, I was hoping you’d show up again some time.
Where I was getting at earlier, you don’t have to be told what frequency the phone wants to ring at,
it’s easy to find it without knowing what the frequency is. It sounds like a phone only in a fairly tight margin.
I suspect at least the "first removed” harmonic also makes it sound like a phone.
It’s improved with a transformer 10x turns ratio to increase voltage, and I’ve used a 10R current limiting resistor on the TI H-Bridge.
That 120 Volts might be closer to 90 Volts (or less) when the ringer is drawing current.

GTC, you are a bit too late, but that’s not what I thought you’d say!
I have done the decade counters, but out of protoboard real estate for the 555.
(and have to wait till payday for the next electronics budget).
The idea was, I suppose, anyone should be able to use those parts.

I can do everything with the phone, but have not yet been able to put it all together, even in my mind.
When someone picks up the phone, you’d have to look for change of impedance to
turn off the ringer and connect the intercom circuit, but the first feedback part is something I haven’t done.

This shows improvement to the circuit, but not a phone ringing with it.
The 60’s transformer has both secondaries connected phase aligned to complete x10 turns ratio.
That is by push pulling only one side of the centre tap with the H-Bridge so only half of the primary turns are used.

The hand drawn transformer is a future one being pulled low either side of a centre tap,
where the centre tap is positive supply voltage.
https://www.youtube.com/watch?v=fvgiyTm0hYI

GTC, you are a bit too late

... either that or you are a bit too fast!

Let's see how it looks with multiple chips, and then perhaps you can produce a PIC design for those interested.

The pic with hardware PWM, feels like cheating,
but a benefit, the crystal derived timing for the durations you mentioned wouldn’t need any adjustment.

Back in the days of Strowger exchanges it was a relay in the final selector that sensed the loop on the line when the phone was picked up.