Phone ring circuit

my

sonnected

Will

The ring generator needs to send a pulse train that matches the ring cadence you hope to simulate. For example the ring signal may be present for 2 seconds and then be away during the time between the pulses for maybe 4 seconds. The ring signal itself is typically a 20 Hertz sine wave signal with a peak to peak voltage of about 90 to 100 volts. This AC voltage rides on the nominal -48 volts of the phone line such that the average value level of the 20 Hertz sine wave is at -48 volts.

Yes it takes a more complicated circuit to produce a ring signal but be that as it may the ring signal is often generated by a 90 to 100 volt p-p oscillator and low output impedance driver and then AC coupled to the phone line. One method of coupling could use a transformer of the appropriate type.

- mkaras

Marcin,

What you're wanting to do is fairly complicated if you want to do it correctly. Making the phone ring is fairly easy. If you were to switch your A/B leads back and forth across a 50V source (at a 20Hz rate) then the phone would probably ring. This would apply a 50V rms square wave to the phone. Typically, an 86V rms sine wave is applied, and it is superimposed on the

However, as you've hinted at, you need to detect when the phone goes offhook both during the ringing voltage application and also when the ringing voltage has been removed. If you don't remove the ringing voltage soon enough (after the user has gone offhook during a ringing burst) you'll either blow up the phone, blow up the ringing circuit, expose the phone user to very loud signals, or all of the above. The circuit that does this detection is usually referred to as the "ring trip" circuit.

The modern central office uses a device called a SLIC (subscriber line interface circuit) to interface to each of its telephone lines. It provides the BORSHT functions (battery, overvoltage, ringing, supervision, hybrid, and test). Nowadays, it's the easiest way to accomplish all the things you want to do.

Here are a couple links to SLIC circuits.

Fig 2, in the st.com pdf, shows how ringing voltage is normally applied to the phone line. The leads they call "T" and "R" correspond to what you've called "A" and "B".

Here's another SLIC manufacturer.

Bob

my

sonnected

Will

"Normal" CO ringing is 20 Hz rms sinusoidal wave superimposed on CO battery (-48 V, derived from a string of 24 lead aacid batteries, so it really is about -52 volts). As a dear friend at Bell Labs used to say, "There are a million ways to solve any problem and here in the Bell system we have a million and one in the field!" Accordingly, there are lots of other low frequency ringing voltages and much higher voltages (for both ringing and battery) used on very long circuits.

The PBX business (very short loops within a single building or campus) took some money saving steps by reducing battery voltage, in some cases, to -24 V and reducing the ringing voltage. An interesting set of integrated circuits, know as "ringing SLICs" (subscriber line interface circuits) was developed for this market. These device generate 100 V p-p ringing out of the normal -48 V battery by carefully reversing the polarity of tip and ring. Generally they generate triangular waveshapes, which have the potential of causing interference with other lines (but far less potential than square waves!). In other cases they switch to higher voltage source to generate ringing. In either case, since the ringing is not superimposed on battery, alternate off-hook detection schemes are required. AFAIK, all these chips use external control for cadence.

Google "ringing SLICs" and you'll find lots of information from major venders like Legerity, Mitel, AMD, etc. The problems you'll have to deal with are off-hook detection and the fact that some devices (some modems, "electronic" phones, and some fax machines) look for ringing with a circuit that has a threshold above 50 volts.

E. Tappert

Dnia 25.12.2006 Bob napisa³/a:

... cut ...

Thanks for answer ! Now i have lecture for tommorow:)

Probably i will have some more questions after that, so i will post them.

This newsgroup is very usefull.

Thanks !

my

sonnected

Will

that's unlikely to work.

and

-50V | | [R] 1K 5W (I measures 50mA into a short here) | [L] | | | oa ob | ring | o--------- 50V (AC 15-20Hz) `--o_ ~-o-----. normal | | ----- gnd

use a low pass filter on the voltage across R (or the voltage at point a) and you'll have "hook" up/down detection

Bye. Jasen

Here is another: