"The phone is a remarkably complex, simple device,
and very rarely ever needs repairs, once you fix them." - Dan/Panther

Main Menu

Dial Speed and the Central Office

Started by Dan/Panther, March 25, 2012, 04:34:35 PM

Previous topic - Next topic


When I first joined CRP forum, We had several discussions about Dial timing and that it was customary for it to be 1 second for the turn of the dial , or roughly 1/10 second per pulse. I was thinking about that today, doesn't it work similar to a telegraph key and no matter the speed of the dial 10 clicks is 10 clicks ? You have a set of contacts that open a circuit. Which sends a current to the office end, and locks in relays to dial, so the speed wouldn't seem to me to be important. Just maybe a convenience of operation standard ?

The More People I meet, The More I Love, and MISS My Dog.  Dan Robinson


In general terms, the dial speed must be somewhat slower than the response time of the receiving equipment in the CO.  If you've watched the videos of the step-by-step office equipment, you know there was a lot going on mechanically in the CO as the dial was turned.  Relays were opening and closing and servos were moving contacts up and down and in an arc.

If the dial goes too fast, pulses will be missed, resulting in a misdial.

The goal was to set the dial speed standard as fast as realistically possible to minimize dialing time, but not so fast that errors were introduced.

Also, dial pulses start out at the dial as (approximately) square waves, but capacitance and inductance on the line to the CO distort the wave form depending on the quality of the line and distance.  (One reason operators' dials were set faster is that they were located in the exchange, so didn't suffer as much line loss getting to the switch.)

Since electronic switching was adopted, a lot of the restrictions imposed by the older CO equipment have disappeared, so it's theoretically possible to support faster dialing rates.

There are several switching documents in the TCI Library that go into much more detail, if you want to dive into the deep end.
Visit:         WE  500  Design_Line



I think it's become the standard because of the tolerances of modern phone exchange equipment at your local exchange (or CO, still getting used to the terminology!!), modern digital units will only take a 10pps dial speed, with some tolerance either way, and it varies form exchange to exchange...

For example, where we used to live, we had a VirginMedia cable phoneline (with an actual phone line, not an ATA/VOIP thing like is common over in the states), and my black WE500 dialled on that line perfectly and worked great, the ringer was strong and loud, but after we moved and got a new BT line (well, it's with a company called "Primus", but uses BT's hardware), the phone would not dial out, and I had to lube and adjust the dial speed before BT would accept it's pulsing, plus the ringer sounds weaker and not as strong, so I think that the tolerances and specifications are just down to what your local phone co's hardware will accept, if at all... :)


In the olden days, speed was very important because the dial was directly driving a mechanical switch at the other end, and the mechanical switch had speed limitations, that were made worse by the distance the phone was from the central office.

Today, the programmers of the newer ATAs and other electronic devices latched onto the 10 pulses per second "standard" and duplicated it into the programming of their devices as far as being able to translate pulses into other forms.

What they all seem to have forgotten is that no mechanical dial is always EXACTLY 10 pulses per second.  Even on the same dial, dialing a 2 can come out at a different speed than dialing a zero, and while dialing a zero, the speed on pulses 1 and 2 may be different than pulses 9 and 10.  The reason is the spring tension and the fact that any mechanical governor is not a precise thing.

All dials are different.  So, the mechanical switches could handle a wide range of variation in the pulses per second.  they were rather forgiving.

However, you now take a VoIP ATA, and they seem to be all programmed so tightly, because they are made for electronic phones that if they pulse at all, they pulse more precisely and with a better wave form than the old dials.

Wave form of the square wave is also affected by the combination of the inductance (L) and capacitance (C) of the circuitry in the legacy phones.  The resulting wave forms when observed on an oscilliscope are jagged and not a precise square wave.  Modern electronic phones produce a totally resistive (No L and no C)  load, so switching the load in and out with electronic pulsing produces a nice even square wave.

The electronic engineering and programming idiots in Tim-buc-tu that designed and wrote the programs for the newer devices also forgot to pay attention in class to the fact that older phones have the uneven square wave.  I have run into the wave-form problem on my Vonage service when I had a dial that was perfectly timed, but running from a phone that had an older subset with the dual purpose ringer and receiving capacitor.  The extra L and C was not allowing the phone to properly dial.

Hope that helps.....

-Bill G


Quote from: twocvbloke on March 25, 2012, 04:58:28 PM... at your local exchange (or CO, still getting used to the terminology!!)
We call them exchanges, too, so no need to translate.  ;)