|WikiProject Telecommunications||(Rated Start-class)|
Decoding the pulse train
How are pulses decoded at the telephone exchange? Any volunteers? Thanks in advance!22.214.171.124 20:52, 25 July 2005 (UTC)
- No idea how a modern exchange converts them into something sensible, but a crossbar or Strowger exchange literally thwocked the mechanical compenents of the switch around for each pulse, and then dropped onto the next number after a sufficient pause.
- Crossbar switches very rarely had the dial pulses operate the crossbar switches themselves directly. It was pretty much only a few small PBX systems which used that arrangement. With all regular crossbar systems the dialed digits were stored in a register until the system had enough digits to know where to route the call, then the control logic would select a free path through the variuous switch arrays to complete the connection. Originally, registers and markers which performed this function were made up entirely of electromechanical devices themselves, mostly a lot of relays. In later years, there were crossbar switches which used processor control to perform the function.
- With Strowger step-by-step switches, the basic system had each dial pulse train step a switch mechanism vertically, that switch would then rotate on the selected level to hunt for a free circuit to the next switch, and so on. The last two digits stepped the final selector in the chain vertically and rotary to select the required line. But there were also many cases with Strowger SxS where pulses from the subscriber's dial were stored in a register rather than stepping the switches directly. In England, for example, the six major urban areas of London, Birmingham, Edinburgh, Glasgow, Liverpool and Manchester used what was called a director system, in which the first three digits were stored and translated into some other digits in order to route the call to the required exchange within the city. The digits translated by the director were then pulsed into to the switches to step them. Similar arrangements were used in some parts of the U.S.A. British STD also had a register-translator arrangement, so after dialing the initial zero, the rest of the number dialed was not stepping the switches directly but being stored in a register in order to translate the number dialed into the digits necessary to route the call to the right part of the country (the rest of the digits were just repeated to the distant end though). 126.96.36.199 (talk) 11:02, 19 August 2012 (UTC)
- This was the time when transatlantic "phonecalls" were done by longwave radio... so it may sound stupidly crude, but thats pretty much how it worked. --Kiand 06:53, 4 January 2006 (UTC)
I don't exactly remember how the pulse relays worked in a crossbar exchange, but I just now briefly described the workings of digit receivers 1ESS switch from the 1970s. Perhaps others will want to ask more detailed questions and jog my memory a little more. Jim.henderson 03:17, 30 March 2007 (UTC)
Good info. I would say this article needs a lot of expansion, showing both classic and modern methods. (added stub status) mrh 20:22, 22 May 2006 (UTC)
- Many modern PSTN line cards still support pulse dialing. All of the current System X and System Y (Ericcson AXE10) switches within the U.K. do, for example, and although it's been some years since I left BT so haven't followed the precise details of upcoming changes, I'm led to believe that BT plans to continue supporting it on the new 21CN switches which are now being installed. 188.8.131.52 (talk) 11:05, 19 August 2012 (UTC)
I added this section which I removed from the Rotary phone article because it has a lot more to do with the technology of pulse dialing than it does to do with rotary phones. Cornlad 19:45, 9 August 2006 (UTC)
It was also possible to dial a number from a public "red box" and wait for the pickup. No coins inserted. You could then yell very loudly into the earpiece and the other end could hear you faintly. The Post Office then put buzzers on the line, until coins were dropped. The Australian PMG did not add buzzers until 1973 and I always rang for free!184.108.40.206 (talk) 11:33, 25 September 2015 (UTC)
Which one is good and fast, pulse or tone dialling? Why do phones still have a pulse option? It seems to me that it only prolongs the dialing process, what practical purpose does it serve?
- Tone dialing is faster. Phones used to have a pulse option so that they could be connected to exchanges, or private exchanges, that had not yet been converted to tone dialing.
- In Australia, phones with touch dial pads were sold so that ignorant people could pretend that they had a phone system like they saw in American movies. The touch dial pads electronically generated the pulse dial signals to work with the Australian phone system, and allowed the national telecom provider to continue the pretense that it had a modern phone system. Pulse/Tone dialling was only a small part of that harmless fraud. —Preceding unsigned comment added by 220.127.116.11 (talk) 23:18, 5 February 2010 (UTC)
- Post Office Telephones, at is was at the time, offered pushbutton dial phones in the U.K. in the early 1970's, but these were the same "store & forward" arrangement which sent pulse-dial digits. DTMF (TouchTone) versions didn't start to appear until the very late 1970's for use on some TXE (electronic) exchanges, but for most people tone dialing didn't become possible until System X started to be rolled out through the 1980's.
- In the U.S., DTMF dialing was available with electronic and crossbar exchanges considerably earlier, although there were still plenty of places which were pulse-dial only into the beginning of the 1980's. In some cases in North America, TouchTone converters which accepted DTMF digits and converted them to pulse digits were installed in Strowger SxS switches to enable subscribers' phones to use DTMF. That resulted in the time taken to actually dial the number by the subscriber being considerably less, but obivously there would then be a wait while the converters at the exchange pulsed out the digits, so the actual time taken from dialing the first digit on the phone to the call being switched through was only marginally quicker than if it had been dialed on a rotary phone directly. 18.104.22.168 (talk) 11:17, 19 August 2012 (UTC)
I think this part is incorrect....
"In rotary systems this" (The need to pause between numbers so the numbers don't bleed together) "is taken care of by having the user wait for the rotor to revolve back to the start position before the next digit can be dialed."
This is not so. The pause is taken care of by the user putting his or her finger in the dial rotor and turning it to dial the next number. When the user releases the rotor, it will switch the line on and off the correct number of times [b]as the rotor returns [/b] to the resting position (waiting for the next number to be dialed).
I used to tap-dial telephones all the time... tore a few rotary phones apart to see how they worked.
Eyknough 03:26, 29 March 2007 (UTC)
- It looks correct to me. After the last pulse is sent the dial continues to rotate without sending pulses until it reaches the resting position. The next digit can be dialed immediately the dial stops rotating. -- Nick 16:47, 31 March 2007 (UTC)
- I agree; what you say sounds correct. But there only needs to be a very short pause between pulses for the numbers to be distinguished. The article seemed to say that the pulses are sent as you rotate the dial, then the long pause waiting for the dial to return to resting postition is what is necessary to distinguish between numbers. It takes about 1/4 second pause between numbers to work effectively. This is long enough for one to dial the next number, no matter how quickly it is dialed. Eyknough 20:14, 3 April 2007 (UTC)
In order to clearly differentiate the pulse trains a pause of at least 2/10 sec (i. e. the time of two pulses) was enforced by the sigal giving mechanism. It produced two more pulses, i. e. for the number 5 then 7; the last two of them were suppressed by a special contact (nsr, see http://de.wikipedia.org/wiki/Nummernschalter#Die_Kontakte (German)). On the outside you could see that by the larger distance between the 1 and the finger stop than between the 1 and the 2 etc. Phones, where the 1 was diectly adjacent to the finger stop, didn't have this extra-pulses-mechanism, and were not so reliable in dialling- -- Fritz Jörn (talk) 11:59, 14 May 2011 (UTC) (Fritz@Joern.De)
- The norm for British & American dials (and almost all others, in fact) is that the correct number of pulses is generated as the dial returns to rest, its speed being controlled by the built-in governor in order to send the pulses at the correct rate (normally 10 pulses per second). There is a natural delay introduced between the last pulse (or only pulse in the case of the digit "1") being sent and dial reaching its natural position, then an aditional delay before any further can be sent by the user turning the dial clockwise for the next digit. So the inter-digit pause is comprised of BOTH the last part of the return movement of the dial and the time taken to "wind it up" ready for the next digit. In fact in many types of P.O. dial mechanism there was also a slight additional pause immediately upon releasing the dial and before the first pulse was sent, due to the design of the mechanism. In the SxS switches which predominated in the British network, the inter-digit pause had to be long enough to (a) cause the relays in the current stage to recognize that the dialing of the particular digit had finished, and (b) to allow sufficient time for that stage to find a vacant circuit to the next and for that stage to prepare itself ready to receive the next digit. 22.214.171.124 (talk) 11:26, 19 August 2012 (UTC)
- That would be paragraph 3, "Two exceptions to this are New Zealand, with [...] and Sweden, with [...]". No idea about Oslo, though. Sabik (talk) 12:44, 19 November 2007 (UTC)
Rationale for "Reversed numbers"
RE: (These were phased out between 1958 and 1994 as subscriber trunk dialling was introduced.)
The push button A and B phones were phased out in the 1960s bar perhaps one or two in the Highlands. This is presumably a typo.
126.96.36.199 22:16, 2 September 2007 (UTC) Peter Thurston
- I'm guessing somebody picked 1958 here for this being when STD was inaugurated from Bristol (it was the very end of 1958). But the pay-on-answer phone, designed from the outset to be STD-compatible to allow direct-dial STD calls did not appear until 1959. While the largest period of changeover was as STD was rolled out during the 1960's, there were still a significant number of the pre-pay A/B phones in use into the early 1970's in those areas which were still waiting for STD conversion (with the few remaining A/B phones themselves having to be converted for decimal currency). The 1994 date is probably just a typo: The very last A/B pre-pay phone was removed from service in a remote spot in the Scottish Highlands in 1984. 188.8.131.52 (talk) 11:43, 19 August 2012 (UTC)
I humbly think the article on "Rotary dial" is nicer. This one here should disappear under the "Rotary dial", or both be joined. Additional details: There had been speed dialling with 20 impulses a second. Where and when? Then: The actual electrical process was a complete shorting vs. complete opening of the line during the impulses, so as to give maximum signal rise and fall. A number 8 was 8 full openings of the line, for example. The distance of the 1 from the finger rest was exactly a 3 digit "angle" – at least in Germany with good phones –, as at the end of the dial train two impulses were generated but then supressed to form the extra 2 pulse pause between digits. Fritz Jörn (talk) 08:03, 30 December 2009 (UTC)--
- Where and when? In North America, some operator console dials were designed to run at 20 p.p.s. where they were always dialing into equipment such as a crossbar or electronic tandem switch which could easily accept the faster dial sequences. Naturally, that resulted in less time spent by the operators setting up a call, and less holding time for the registers in the equipment, which were only held for the duration of dialing, not for the entire call. I believe that a few countries tried 20 p.p.s. subscriber dials at one time, but the main limiting factor then becomes the length of the physical lines themselves and the distortion introduced to the pulses. 184.108.40.206 (talk) 11:36, 19 August 2012 (UTC)
This is wrong for the A and B button phones I knew, so I want to see more justification here.
There were a variety of A/B button phones made. Normally, the A button accepted payment, the B button was used for coin return.
I have never seen or heard of an A/B phone which did not allow dialing free. (Early A/B button phones allowed operator connection, payment was required to accept the call).
Some phones had timers or slow disconnect, the call disconnected if the fee was not paid, allowing you to confirm that the call had gone through to the right place. Other phones did not connect the microphone until the fee was paid, allowing you to shout through the earpiece.220.127.116.11 (talk) 23:41, 5 February 2010 (UTC)
- There seems to be a little confusion here. The standard A/B phone was arranged so that you got dial tone as soon as you picked up the handset, but contacts in the coin collector kept the dial shorted so as to prevent the dialing of a number. An extra set of contacts on the dial provided an exception to allow "0" to be dialed to reach an operator (some years later this was extended to permit the dialing of a "9" for 999 calls, and later still to allow "1" for 100 operator calls). At this point the transmitter in the handset was "live" to permit talking to an operator, and to allow incoming calls to function properly.
- Depositing the correct fee (2d. initially, rising to 4d. by the 1960's) caused the short across the dial to be removed to allow the caller to dial the required number. However, the first coin dropping into the mechanism also caused other contacts to short the transmitter. Thus the caller could hear ringing, busy tones, etc. and could hear the distant party answer to ensure that he had reached the required number, but could not talk to the distant end (nominally, as noted already). Upon pressing button "A" on the front of the box, the coins were deposited and the short removed from the transmitter to permit conversation to proceed. The short was also replaced across the dial to prevent the caller from simply hanging up for a moment and then making another call on the same money.
- In the event of an unsuccessful call, pressing button "B" on the side of the box caused the coins to be sent into the return chute, while at the same time operating a small spring-wound timer which opened the line for about 6 to 8 seconds in order to release any connection. It also restored the transmitter and dial contacts to normal, as with pressing button "A."
- There was another coin transmitter inside the box which was switched into circuit when the handset transmitter was shorted, and which allowed the operator to hear the bell and gong signals of deposited coins for a long-distance call.
- This basic method of operation remained the same from 1925 when the A/B box was introduced right through to the last one in use in 1984, just with changes to the dial for 999 and 100 as mentioned already, and the appropriate changes as the minimum call fee increased over time from the original 2d. to 3d., then to 4d. etc. There were a few minor variations, such as for A/B phones used in a handful of rural areas served by a UAX5 which required the dialing of some 0x code to reach an operator, but these were rare.
- So yes, if you just picked up the handset and did NOT deposit any coins, you could dial a local number by pulsing it out on the switchhook and make the call for free, since the transmitter remained in circuit. The transmitter wasn't disabled until the first coin was deposited. The fact that in normal operation you could make a call and hear the distant end, just not talk to the other party, also meant that it was possible to dial normally to the time, recorded weather announcements and so on and hear the announcement for free. Just make the deposit, dial the number, listen, then press button "B" to get your money back. 18.104.22.168 (talk) 12:04, 19 August 2012 (UTC)