In a message dated 11/15/2002, Colin Botha writes:
> In South Africa there is a rather upsetting requirement to make the PSTN > front end of a DAA or trunk interface circuit immune to these meter pulse > signals, which are applied to tip and ring at 8Vrms behind a 200Ohm > resistor. Most chipsets would fold under this onslaught and result in > broken/interrupted transmission if you should be on an international call, > say, with meter pulses pouring through at one second intervals, for > example. > > Is this requirement, or rather this practice, peculiar only to SA? or do > other countries have similar requirements? > > What is strange is that these signals originate in our Siemens exchanges, > which would seem to suggest that it is not necessarily of our own making, > or > maybe I am being naive here. > > Although it is not too difficult to provide either a low-pass filter on tip > and ring, or an LC notch filter across the line, the impedances of these > networks is such as to violate the requirements for the on-hook impedance > required by CLI (CID) specifications. > Hi Colin: First, just to clarify the "requirement," I do not believe that meter pulse immunity is a regulatory requirement in either the present TE-001 or the proposed draft of TSR-001 for South Africa. Of course, such a requirement could certainly be imposed by a customer in certain situations. Meter pulses (aka billing tones) used to be quite prevalent in Germany, and there used to be a regulatory requirement for immunity to 16 KHz tones launched at 10 VRMS. In Switzerland there was a similar requirement for immunity to 12 KHz tones at slightly lower levels. The really high levels are associated with older electromechanical exchanges. Modern electronic exchanges typically launch at much lower levels, which is better, but often still high enough to cause problems. Due to the RTTE directive, these requirements are no longer mandatory in Germany and Switzerland, but the tones are still present on many lines. For users who have trouble with meter pulses and are unwilling or unable to have the phone company remove them from the line, external filters are available from a variety of sources. Several other European countries used meter pulses to a lesser extent, as did Australia and China. In general, the use of meter pulses is being phased out, but they still appear in some parts of the world. If you find that in your application you really must remain immune to meter pulses, and an external filter is not a good solution due to CID or other issues, you can add an LC notch filter behind the switchhook relay in the DAA or trunk interface. This should help get around the CID impedance problem, and will place the filter in a more protected environment. There used to be special inductors available from various vendors that had the right combination of inductance, tolerance, and DC resistance to readily make a very nice notch filter using standard value capacitors. I do not know if these are still available. One last thing to note is that the presence of the filter (if there is a shunt leg in it, and not all filters have a shunt leg) will influence the return loss of the composite interface, so if return loss is an issue this should be taken into account in developing a solution. To be honest, I have not put a meter pulse filter in a new design for several years, since the usage of meter pulses seems to be declining. You may want to push back a little on whoever is imposing this requirement to see if it really is essential for this particular application. Joe Randolph Telecom Design Consultant Randolph Telecom, Inc. 781-721-2848 http://www.randolph-telecom.com
