Thanks for that history RA. Your mention of tuned transformer in telephony jogged a vague memory from 40 years ago of reading about those.

The Spence Electronics box looks interesting, but the "standard" model strikes me as more a ring relayer than generator as it needs a PBX ring current as input.

The LED input model appears to be more like a generator?

Yes, thanks for the info RA, I'm always curious about apparatus that existed in those red brick buildings where the public were not allowed to peek in!

For those unaware, Western Electric and Standard Telephones and Cables (STC) were related companies. The font used for the printing of the names of both companies was the same. STC is now part of Nortel and STC Australia was swallowed by Alcatel in the early 1990s.

If you pull an Alcatel-built Telstra phone apart you can see STC stampings on the plastic mouldings. Those made by competitor Exicom had AWA stampings inside. The conservation of the logos of the original companies making the series T200 touchphones was simply a cost-cutting measure and one easy to get away with since most customers wouldn't resort to pulling their phones apart.

‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾
A valve a day keeps the transistor away...

Further to the question of how were ringing and tones produced, I have scanned two pictures of STC PAX equipment from a 1947 brochure. The associated text states that, “Ringing and tones are generated by means of a 2½ watt battery-driven ringing machine of the tone inductor type or alternatively by a ringing vibrator operating at the exchange voltage. Vibrators are employed only on the smaller units.”

The specifications are:

Ringing Current - 75 volts at 17/20 c/s interrupted at approximately 0.2 sec on; 0.4 sec off; 0.2 sec on; 2.0 sec off.

Dialling Tone - A low pitched continuous tone having a frequency of about 33 c/s.

Busy Tone - An interrupted tone (frequency of about 400 c/s) 0.75 sec on; 0.75 sec off.

Number Unobtainable Tone - A continuous high-pitched tone (frequency of about 400 c/s).

The first picture shows the ringing machine which seems to be a DC motor/generator that also drives two sets of cams through a reduction gearbox. The cams then operated spring-sets to get the required cadences. Assuming that the plug contact strip was the same as that on a relay set, then the whole unit would have been about 18-20 inches long. (And probably very heavy judging by the cast iron base plate; note the holes for the mounting bolts.)

The second picture shows a 25-line PAX with a vibrator arrangement in the bottom section of the cabinet. The original photograph is rather poor but the two relays at the top of each relay set appear to be the vibrators with the other relays probably generating the timing of tones. There seems to be a uniselector in the left hand set that would have produced the ringing cadence.

The AE sub-cycle ringer can be seen at:
http://www.telephonecollectors.info/index.php/bruce-crawford-library/doc_view/1509-ae-bulletin-444-sub-cycle.

The D-35237A unit was used by the PMG in Australia at one stage. I 'recovered' one from a PABX room in a disused building just before it was demolished. The building was part of what had been a WWII ammunition factory. Connected to a 240V/110V step-down transformer (borrowed from an old Bell and Howell 16mm projector) it worked for a while until the insulation on the main transformer broke down making the chassis 'live' and too dangerous to use.

Andrew

From GTC "The Spence Electronics box looks interesting, but the "standard" model strikes me as more a ring relayer than generator as it needs a PBX ring current as input."

The Spence unit is a ringing generator with a 25Hz oscillator section driving two FETs that then switch DC through a transformer. It requires a 27-48V DC supply usually taken from the PAX supply. The ring current from the PAX just gates the drive to the FETS through an optocoupler to get whatever cadence the PAX uses.

When I made contact with the bloke who made the unit in an attempt to get the circuit diagram, he told me that he had copied the basic design from a unit that Telecom once had. The main difficulty was getting a suitable transformer that would work efficiently at 25Hz. The 'standard' 50Hz mains transformer performs very poorly once the frequency is below 50Hz so he had to get a special batch made up. I didn't get the circuit diagram but was told how to connect the unit.

Andrew

While my main interest is old radios, I do have one phone, given to me with a radio I purchased.
I want to say that this thread is one of the most interesting ones I have seen for a long time
Thanks to all the knowledgeable contributors.
Harold

After some digging I found the 25Hz ringing converter that I previously mentioned. It was made in 1974 which was later than I thought but I have seen the same arrangement built into other power supplies for PMBX (cordless switchboards) from the 1960s.

Below are photographs of the unit which is about nine inches wide. The transformer appears to be custom made but does not have any identifying markings. It has a much heaver laminated core than the typical 5VA transformer. The capacitor is rated a 600VW. The circuit diagram was printed inside the case and could not be photographed so I had to redrawn it. It shows the diode in the primary side which means that the transformer is fed with pulsating DC at 50 pps.

This unit still works but having DC on the primary side rather upsets the safety breaker in the switchboard.

Andrew

Further to the 25Hz Ringing Converter, I found a reference to similar units used by the BPO in Britain in the 1960s on Bob Freshwater's excellent site. See:

http://www.britishtelephones.com/conring4.htm.

For the curious there was a description on how the unit worked.

"Consider the converter transformer as having two cores with identical secondary windings and a common primary winding. These cores saturate in turn on alternate positive half cycles of the half-wave rectified mains supply. During the time a core is saturated no energy is transferred to the output. During the time a core is not saturated (i.e. it is working normally) a pulse of energy is transferred to the output. The saturation pulses are at 50 c/s. The secondary windings are connected in series opposition relative to the primary winding, therefore these output pulses will be alternately in opposite senses, giving rise to an output at a frequency of 25 c/s. The optimum value of the capacitor which is connected to the secondary windings of the transformer is determined by the behaviour of the converter at low mains-input voltage and the starting of oscillations on load. Unlike a normal LC resonant circuit, the value of this capacitor does not influence the frequency of the output current.

When the loading on the output terminals is increased beyond full load, the output voltage falls rapidly and the production of 25 c/s ceases. On removing the excess load, the oscillations restart. This provides a useful self-protecting feature."

The circuit diagram for the converter is at another excellent site run by Sam Hallas:

http://www.samhallas.co.uk/repository/n_diagrams/0000/N651.pdf.

Details of an AE Ringing Converter of the vibrator type are at:

http://www.telephonecollectors.info/index.php/bruce-crawford-library/doc_view/1508-ae-bulletin-425-32a46-ringing-converter

Using a 48V DC input, it generated 65V at 20Hz with an output of 25W. Again note the size of the transformer to get the necessary low frequency efficiency.

Andrew

I was wondering how they did it, those intriguing circuits revealing!

I'm guessing the secondaries - on separate cores - have minute asymmetry which would initially allow Cap to douse saturation in weaker side by positive feedback (coupled +sec phase to other +sec phase?).

Mine incorporates a "113A Frequency Converter" potted module (no circuit available)

pictured in link

http://www.worthpoint.com/worthopedia/western-electric-co-113a-frequency-172937244.