Reference / Paper · 2019
The step-by-step telephone switching system: The connector switch
Connector operation in detail, with test-stand photo.
Reference / Paper · 2019
Connector operation in detail, with test-stand photo.
The step-by-step telephone switching system: The connector switch Douglas A. Kerr Issue 4 December 19, 2019 ABSTRACT AND INTRODUCTION The step-by-step telephone switching system (as it is known in the Bell Telephone System; the “Strowger” system elsewhere) was the earliest “mechanized” telephone switching system to receive broad acceptance, and it remained important for many decades. This is one of a series of articles on this system, and it describes the connector switch, which is used in the final stage of the switching network. The basic functions of the switch are summarized, and an illustrative schematic drawing is used as the basis for a detailed description of its operation. 1 GENERAL 1.1 The series of articles This article is one of a series. The “master” article, “The step-by-step telephone switching system: Overview”, by the same author, gives background on the historical development of the system, and then describes its overall architecture, scheme of operation, and the technical details of the unique type of switch used in the system. It also gives background on such telephone concepts as battery and ground; tip, ring, and sleeve; and the like. The other articles (including this one) describe in detail (including at the circuit level) the different switches used in the step by step system. In some cases, information given in the master article is repeated here for continuity. All the articles are indexed on, and available at, my site, The Pumpkin: http://dougkerr.net/pumpkin 1.2 Types of switches The step by step switching system in its most widely-used form uses three kinds of switch, all with essentially the same base mechanism but varying substantially in their complement of relays, function, and operation. The three types are the line finder, the selector, and the connector. The line finder serves to provide a connection from a subscriber line requesting service (the user lifts the handset) into the switching network itself, in the person of a selector switch (this one Copyright © 2018-2019 Douglas A. Kerr. May be reproduced and/or distributed but only intact, including this notice. Brief excerpts may be reproduced with credit. Step by step: The connector switch Page 2 being a so-called first selector). The selectors serve to advance the connection stage-by-stage through the “interior” of the switching network, each one in response to a successive digit of the dialed number, not including the last two. The connectors constitute the final stage of the overall switching network. After all but the last two digits are dialed, the connection has been extended to a connector switch, one which can access 100 line, those whose numbers are been dictated by the earlier dialed digits. The last two dialed digits move the connector to the corresponding line. The connector tests the line to determine if it is busy (that is, on an existing connection). If not, a connection is made to the line and the ringing signal applied. When the called station answers, the ringing signal is removed and a transmission path is completed between the calling station and the called station. Figure 1. Typical connector switch Step by step: The connector switch 2 Page 3 THE SWITCH ITSELF Figure 1 shows a typical connector switch, with an unwired bank assembly attached, in a test stand, used to hold the switch while it is being adjusted or tested. At the bottom we see the banks. Plural? Yes. When I introduced the concept of the Strowger bank and wiper system, I was intentionally vague about the nature of the bank “terminals”. It would be easy to assume that each “terminal” of the bank had three contacts (for tip, ring and sleeve) and that the wiper had three contact-making members to connect to them. But in reality, we can only readily have two contacts at a bank position. So in fact the “bank” of the switch is actually an assembly of two banks, as we see in some detail in figure 2. On the lower bank, at each position are two contacts, rather thin, lying opposite one another on opposite sides of a thin insulating phenolic sheet. At each position, these contact pairs carry the ring and tip leads of the circuit. There is a thicker phenolic sheet, not shown in the drawing, between these “sandwiches” at the various levels. On the upper bank, at each position, a single contact, a bit thicker than the contacts on the lower bank, projects out from between two thick phenolic sheets. At each position, this contact carries the sleeve lead of the circuit. Figure 2. Three-conductor bank assembly Step by step: The connector switch Page 4 For either bank, the entire stack of phenolic sheets and metal contacts, for all ten levels, is clamped together between steel pates at the top and bottom by long screws extending from top to bottom. What we have thought of as “the wiper” of the switch is actually two wipers, one running on the lower bank and one on the upper bank. Each wiper has an upper and lower contact “leaf”. On the bottom wiper, the two contact leaves are insulated from one another (since they will touch separate contacts, for the tip and ring leads). On the top wiper, the two leaves are electrically connected together, since they will be touching the top and bottom of a single contact, for the sleeve. The bank assembly is fastened to the switch, in rather precise alignment, by way of the threaded studs seen at the tips of the bank rods (with the nuts seen there). In figure 3 we see a “two-contact” wiper. Figure 3. Two-contact wiper As we see, the wiper (which of course moves around quite a bit) is connected to the switch circuitry with two very flexible cords. Their actual conductors use what is called “tinsel” construction. A very thin ribbon of conducting material is wound around a small fabric core, this whole thing then covered by a durable but flexible woven cover. This style of cord is able to withstand literally millions of cycles of flexure and twisting as the switch goes through its motions. Toward the top of the switch “chassis” we see seven relays. The different kinds of switch (line finder, selector, and connector) have different numbers of relays, and their functions vary between the type of switch. These relays are of a basic design long associated with the step by step system. Although, for example, the switching systems designed Step by step: The connector switch Page 5 by the Bell System and made by Western Electric used, from the outset, new types of relay, more compact and less costly to manufacture, step by step switches made by Western electric (as well as those made by Automatic Electric) continued to use the relay design we see here. Just above the bank assembly is the switch mechanism. Its heart is three electromagnets (generally just called “magnets”). One steps the switch in the vertical direction, one steps the switch in the rotary direction, and the third releases the “dogs” (retaining pawls) that hold the switch in place after it has stepped up and around, allowing a spring to rotate the shaft back to its home angular position, and then allowing the shaft to drop by gravity to its home position. There are also various contact assemblies that do things like detect if the switch is at its home position or not. These play various roles in the logic executed by the relays in controlling the switch movement and otherwise managing the emerging connection. In use, the switches (not including the banks and wipers) are each covered by the iconic “mailbox-shaped” sheet metal cover. 3 SWITCHING NETWORK ARCHITECTURE AND OPERATION I will assume that the reader is generally familiar with the principles of telephone system operation and with the basics of the step by step system ( perhaps from the prior articles in the series). To set a specific context for discussion of the role of the connector, we will consider a city in which there are several central offices, and in which a 5-digit numbering plan is used. The first digit identifies the central office in which the line with that number is served. We will imagine a call to number 5-2368. The digit “5” tells us that the line with that number is served by the “5” central office. And we assume that we are in fact looking at the “5” office, and that the calling line is also served by the “5” office (so this is an intraoffice call). The hyphen in the number as it is written here is of no technical significance, but is just used to help the subscriber in dealing with the number. I will omit it in some places in this discussion. Figure 4 shows the entire “train” through which the call is handled from one end to another. We see a line finder, three selectors (in successive stages), and a connector. A much more wide-ranging discussion of step by step system architecture is given in the companion article, “The step-by-step telephone switching system: Overview”. Step by step: The connector switch Page 6 Figure 4. 5-digit switch train Each consecutive digit dialed (for the first 3 digits) controls the operation of a selector in a consecutive stage of the switching network. Suppose we join the action after the subscriber has lifted the handset, the line has been found by a line finder, the subscriber’s line has been extended to the permanently-associated 1st selector and received dial tone from it, and the subscriber has so far dialed “523”. As a result, the connection has so far been extended to a connector, in particular a connector we can describe as being in the 523XX group. That means that it is only ever involved in calls to lines whose telephone numbers start with 523. And thus it is only ever reached by a caller starting his dialing with 523. At the connector, “talking battery” is fed, through a relay, upstream to the calling station. This relay will respond to the pulses of the digits dialed into the connector. The 6 pulses of the next digit dialed, 6, cause the wiper shaft of the switch mechanism to rise to level 6 of the contact bank. On this level of the contact bank there are 10 sets of terminals, arranged in an arc, so the wipers on the shaft could, if the shaft were to rotate, contact any one of them. These lead to the lines whose telephone numbers are 5-2361 through 5-2369 and finally 5-2360 (because movement over this arc of terminals will be directly caused by stepping by the pulses of the last digit, and of course dialed “0” is represented by 10 pulses). While the caller is dialing “6”, a timer (implemented by a relay with a “slow release” characteristic) observes the time between consecutive pulses. When there had been no pulse for, perhaps, 200 ms, the relay releases, in effect declaring, “I think there are going to be no more pulses for this digit”. So the vertical journey of this switch is complete. Someplace on level 6 of the bank is where the connection is going to. Step by step: The connector switch Page 7 When the next digit (8) begins, each pulse causes the shaft to step in the rotary direction by one step. During the receipt of the pulses of the second digit, another slow release relay observes the time between consecutive pulses. When there had been no pulse for, perhaps, 200 ms, this relay releases, in effect declaring, “I think there are going to be no more pulses for this digit” And in fact, at this time, the switch wipers are on the terminal (contact set) in the contact bank leading to the line whose number is 5-2368. At this point, the busy/idle status of the line is tested. If the line is busy, its sleeve lead will be grounded (a result of another connection to it). If it is not busy, the sleeve conductor goes to battery through the winding of a relay in the line’s line circuit. If the sleeve shows ground, a relay is operated that sends busy tone back over the connection to the calling subscriber. If the sleeve does not show ground, but shows battery, a relay operates that causes ringing voltage to be applied to the line. This is done through the coil of a “ringing trip” relay that does not respond to the AC component of the current flowing in the line through the station ringer, but does operate when the subscriber there answers and DC current flows in the line. When the ringing trip relay operates, it disconnects the called line from the ringing voltage and connects it to a second “battery feed” relay. A pair of capacitors provides for the propagation of voice frequency signals (“audio”) between the two lines during the conversation. I’ll leave discussion of the further adventures of the connector for the detailed circuit description. 4 THE CIRCUIT SCHEMATIC DRAWING 4.1 Our subject switch The circuit schematic drawing and associated circuit operation description do not necessarily represent exactly any of the numerous specific connector switch designs found in the wide range of step by step switching systems. All of them, however, follow almost identical principles, and the hypothetical one we will discuss uses representative circuitry and operating procedures. 4.2 Basic schematic drawing conventions The schematic drawing employs a system of notation introduced in the Bell Telephone System in the late 1950s, called detached contact notation. In it, the relay contacts are not shown in a form evocative of Step by step: The connector switch Page 8 actual physical contacts, with all the contacts on a certain relay all adjacent on the drawing, much as they are in real life. Rather, in this system, simple (easily drawn!) geometric symbols are used for the basic contact elements, what would be called in other contexts a “normally open” or “normally closed contact”. The contacts on a certain relay are not gathered together on the drawing, but rather are placed so as to allow the most clear portrayal of the circuit paths. The possibly many contacts of a relay, and its activating coil, are related by each being marked with the same symbol (which, by the way, in real equipment would likely also be marked on the relay itself). In this context, the two basic kinds of contact mentioned above are not called by the names I mentioned there. Rather, the contact type that in other contexts would be called “normally open” is called a make contact; the type that in other contexts would be called “normally closed” is called a break contact. Figure 5 shows the principles of this convention. Figure 5. Detached contact schematic symbols—relays In panel (a) we see a relay, M, under the older attached contact convention. This relay has a coil with a single winding and four contact “spring sets”, each of a different type. Each “spring” is identified by a number (from 1-10). The dashed line we see between the three spring sets shown above the coil emphasizes that the “moving springs” of all these spring sets Step by step: The connector switch Page 9 move together. Of course, spring 10 on the spring set shown below the coil moves at the same time, but we are expected to know that. Spring set 1-2 is a make contact (what would be called in other electrical work a “normally open” or “form A” contact). Spring set 3-4 is a break contact (a “normally closed” or “form B” contact). Spring set 5-6-7 (what would be called in other electrical work a “form C” contact) is called a transfer contact; It implies a break-before-make operation (so there is never, even momentarily, a path from spring 6 to spring 7. Spring set 5-6-7 is also called a transfer contact. It however implies a make-before-break operation. There is never, even momentarily, a path from spring 8 to one of the other springs, 9 or 10. This is often in fact spoken of in other electrical work as a “make before break”, or “form D”, contact. It is sometimes called a continuity transfer contact. In panel (b) we see this same relay portrayed under the detached contact convention. There I have purposely shown the spring sets “scattered” to remind us that they would not ordinarily be shown adjacent to the relay coil but would be placed on the drawing wherever the circuit paths through them would be easiest to follow. (In larger drawings, they may well appear on separate sheets.) We see that the coil has a simpler symbol, one not graphically evocative of its winding. As we see for contact 1-2, the symbol for a make contact is a simple cross, centered in the line representing the circuit path. For contact 3-4 we see the symbol for a break contact, a simple line across the circuit path. For the basic transfer contact (break-make) (5-6-7), we use a combination of those two symbols, usually adjacent, as we see here. (But if needed for clarity of the circuit paths, the two parts may be separated.) For the make-before-break contact (8-9-10), the portrayal is the same as for the break-before--make contact. There is nothing in the graphic representation that distinguishes the two forms of a transfer contact. In formal Bell Telephone Laboratories drawings, that distinction was provided in a separate “apparatus figure”. But in informal drawings, often some mark was applied, keyed to a note that told this was a make-before-break” contact). I do that using the symbol “#”. In some cases, there are two (in rare cases even more) windings on the coil. We must generally be aware of the relative “polarity” of the Step by step: The connector switch Page 10 windings, so the current through the two windings produces adding, not opposing, magnetic fields (or in some cases, produces opposing fields). In panel (c) we see a two-winding relay coil shown under the attached contact convention. In panel (d) we see that same coil under the detached contact convention. In both cases, the little half-moon marks show “corresponding” ends of the two windings. (But those were not always shown under the attached-contact convention.) Especially in the case of more modern relays whose physical construction is not that suggested by the symbol shown in panel (a), the contacts (rather than individual springs) are identified by number. In panel (a),I have shown these contact designations in brackets. In panel (e), we see the contact whose springs would be numbered 6-7-8, but as a contact would be numbered 3, identified by the contact number. 4.3 Identification In the schematic drawing in this article, the relays in the connector switch are identified with the designations (the letters A-G and K) most commonly used in actual practice.1 For reference in the discussion, the various contacts (not springs) of each relay are identified by fairly-arbitrary numbers; these do not necessarily follow the numbering system that would be found on the formal circuit schematic drawings (which would identify springs). I will often use a shorthand: to refer to the make aspect of contact number 2 on a relay I will refer to contact 2M, and to refer to a make contact I may refer to contact 3M. Once I get rolling, I might refer to contact 2M on relay K as just “K2M”. For a break contact, a typical reference would be “J2B”. 4.4 Simplifications A very few simplifications have been adopted in the drawing. For one thing, it omits various R-C networks used to limit the amplitude of the voltage spikes that occur when the circuit to a serious electromagnet is interrupted. Also eliminated (or in some cases simplified) are some circuit paths devoted to the monitoring of switch behavior by external circuitry. 1 Bell System step by step systems use this convention, inherited directly from the practice in systems made by Automatic Electric. It is based entirely on the physical locations of the relays in the switch. Other systems, originally designed in the Bell System, would typically use mnemonically-based relay designations, perhaps “BF” for the battery feed relay rather than “A”. Step by step: The connector switch 5 CIRCUIT DESCRIPTION 5.1 The circuit schematic drawing Page 11 Figure 8 is the circuit schematic drawing for the hypothetical connector switch being discussed. 5.2 Initial conditions With the switch idle, all relays, electromagnets, and contacts are released. The switch wiper shaft is in the idle (full down) position. Battery (-48 V DC) is standing on the ring conductor of the incoming circuit and ground on the tip conductor, both through windings of the A relay. The incoming sleeve conductor is open. 5.3 Seizure When the switch is seized (by the off-hook calling subscriber line, which has been extended through the prior switches in the train), the loop current through the calling station operates the A relay. Its contact 1M operates relay B. Relay B contact 2M grounds the incoming sleeve lead, which holds all the earlier switches “up”. 5.4 The first digit When the calling dial creates the first pulse of the first digit handled by the connector (the next-to-last digit of the called number), relay A releases momentarily. But because the B relay is slow release, it does not release. The path though A1B, B1M, and the vertical off normal (VON) switch contact 2B through the coil of relay C and the vertical magnet, to battery, is completed, Both relay C and the vertical magnet operate. The vertical magnet steps the switch shaft up one level. This results in the VON switch operating. It reroutes the path to the C relay and rotary magnet coils to now pass through C1M (complete since C is now operated). At the end of the pulse, relay A re-operates, breaking the path to relay C and the vertical magnet. The vertical magnet releases, but relay C does not, it being slow release. Subsequent pulses for this digit repeat this process. At the end of this dialed digit (after its last pulse), relay A remains operated for some while, and relay C is de-energized for that period. After perhaps 200 ms, relay C releases. Its contact 1, combined with VON being operated, reroutes the pulse path to go to the coil of relay E and, through some other contacts, through the rotary magnet to battery. Step by step: The connector switch Page 12 At this point, using our example of the dialed number being 2368, the switch shaft is at level 6 of the bank. 5.5 The second digit Much as we saw for the first digit, the first pulse of the second digit (the last digit of the called number) results in A being momentarily released. The path though A 1B, B1M, and VON contact 2M, then through the coil of relay E, then some other B contacts, and the vertical magnet, to battery, is completed, Both relay E and the rotary magnet operate. The rotary magnet steps the switch around by one position. Each subsequent pulse steps the shaft around one father position. Relay E, being slow release, returns operated until perhaps 200 ms after the end of the last pulse of this digit. 5.6 Busy test—called line busy As mentioned earlier, for a line that is busy (as a result of another connection) its sleeve lead is grounded. If the line is not busy, the sleeve shows battery in series with a relay in the line’s line circuit. When the shaft has made its final rotational step, the sleeve wiper contacts the sleeve contact on the called line’s terminal on the bank. Let us assume that the line of interest is in fact busy. Because the E relay is operated through the entire train of second digit pulses (and a bit after), the ground on the called line sleeve (S′′) passes through E 1M and K 4B to the coil of relay G, which operates. When E releases, a locking path for G is provided via K4B, E2aB, and G3 to the incoming sleeve (S), which has since seizure been grounded by B2M. G1M connects busy tone (BT) to lead R′, from which it passes through capacitor C1 to the ring of the incoming line (R), and thence to the calling subscriber. Relays D and A each, by virtue of their two windings, which are coupled, act as transformers and approximately result in the tone being balanced on the tip and ring conductors (desirable from a transmission standpoint). 5.7 Line idle If the line is not busy, relay G is not operated when the wipers have landed on the called line terminal. Shortly, relay E releases. Now the sleeve wiper will go though E2B to winding 1 of relay K, and through it to the incoming sleeve (S), which is grounded. Relay K operates. A holding path for K is established through K5M to the sleeve (grounded). Step by step: The connector switch Page 13 K makes a “connection” to the called line, through K1M and K2M. But this connection is at this point committed to applying ringing voltage to that line. Additionally, through K3M, the sleeve wiper is grounded. In the station’s line circuit, this operates the cutoff relay that removes from the line the line relay, which, when the line is idle, detects if the station is taken off-hook in order to place a call. And of course that ground on the sleeve “marks” the line as busy, for the information of other connectors that may later aspire to make a connection to it. Before I proceed with the story, let me digress and talk about ringing and ringing voltage. 5.8 Ringing and ringing voltage The basic signal used to operate telephone ringers is an AC signal at perhaps 75-90 V RMS at 20 Hz, to which is added a DC voltage of perhaps -48 volts. At the station, the coil of the ringer “motor“ (which is a bit like a polarity-sensitive relay) is connected across the tip and ring of the line through a capacitor. The main purpose of the capacitor is to prevent the flow of DC current through the ringer coil. This could falsely indicate to the central office that an idle station was “off-hook” and thus wanted to make a call. It would also “steal” some of the line current when the station was on a call, and we want all of it conserved to operate the station transmitter (microphone). Why the DC component of the ringing signal? When ringing is applied to the line, although there is a DC component to the applied voltage, because of the capacitor in series with the ringer coil, there is no DC component to the current that flows. But when the subscriber answers, the telephone set circuit proper is placed across the line. It of course has DC continuity, and so now there is a DC component to the current that flows (from the DC component of the ringing voltage). At the central office, when a line is rung, the ringing voltage is applied through a winding of a special kind of relay. It is very insensitive to the AC component of the current through it, but is sensitive to any DC component. Thus, with ringing voltage applied to the line, this relay does not operate until the station answers. Typically, the operation of this relay causes ringing voltage to be removed and the final “connection” to be made to the line (and we will see just that here). This action is called Step by step: The connector switch Page 14 “tripping the ringing”, and from that, the special relay is called a ring tripping (or just ring trip) relay. Of course, normally, the ringing signal is applied in repeated “bursts”, typically 2 seconds long separated by 4 seconds of no ringing voltage. But during these “silent” intervals, there is still voltage applied to the line, this time just -48 V DC. The result is that if the station is answered during the quiet interval, the ring trip relay still operates (from the DC current from the “silent interval” DC voltage), and ringing is tripped promptly. Going back to manual switching days, the matter of mechanically generating the repetitive “cadence” of ringing (rather than having the operator pull the ringing key intermittently), and to have the ringing automatically ended by a trip relay, was overall called machine ringing. And from that usage, the ringing voltage, with the DC component, and interrupted to form ringing and silent intervals, is often spoken of as machine ringing (MR). 5.9 Ringing the called line After the K relay has operated, the line having been found to be idle, a connection has been made to the line, but at this point only for ringing the line. The ringing voltage (MR) is applied to the ring wiper (and thus to the ring of the called line) through relay contact K6M and winding 1 of relay F (the ring trip relay), through F1B, and K1M. The “ground side” of the ringing voltage source (RG2) is connected to the tip of the line through F2B, K2M, and the tip wiper. Thus the tip is grounded during ringing. When the called station answers, as I discussed above, relay F (the ring trip relay) operates. F relay transfer contacts 1 and 2 transfer the wipers from the ringing voltage source to what will be their permanent home in the completed connection. There, battery is applied to the tip, and ground on the ring, through the windings of relay D. This of course provides “talking battery” to the called station. The flow of current through called station (which is now off hook) operates the D relay. One result is that, via D relay contacts 1 and 2, the polarity of the battery fed toward the calling station is reversed. This (spoken of as “reverse battery supervision”) is to advise various “upstream” entities that the called station has answered. Among other things, this can play a role in the operation of certain coin stations 2 RG: ringing generator ground. Step by step: The connector switch Page 15 that use a relatively-simplistic modus operandi. The details of this are beyond the scope of this article. We note that the polarity of the talking battery supplied to the called station is reversed. This, essentially “instant reverse battery supervision”, in effect would convey to any “entity” at the called end that the other end of the connection (the calling line) is off hook. 5.10 Audible ringing A bit of the ringing voltage (on the called station ring lead, R′′) is fed to the R′ lead through capacitor C3, and travels to the ring of the calling line through capacitor C1. This allows the calling subscriber to hear that the called line is being rung. This is called audible ringing (but sometimes ringing induction). But wait—the frequency of the ringing voltage proper is typically 20 Hz. At that frequency, the response of the telephone set is almost nil, and this is at the bottom limit of human hearing response as well. In the earlier days of telephone switching, AC ringing voltage was often generated in a number of ways that hardly generated a sine wave (normally a desirable property of an alternator). Thus the ringing voltage itself was high in harmonics, and these were readily heard. When motor-drive alternators were used to generate the ringing voltage, they would typically develop a fairly-sinusoidal waveform, and thus audible ringing performance was problematical. So ways were found to change the design of these alternators to intentionally generate a less-perfectly-sinusoidal waveform. In other cases, an external non-linear transformer was used for this purpose. Again, the transformer action of relays A and D approximately provides for the audible ringing signal to be balanced on the tip and ring conductors. 5.11 Release—calling party hangs up first. If, at the end of the conversation, the calling party hangs up first (as suggested in the “good old days” by etiquette authorities), opening the DC path at the station, and ceasing the flow of DC in the line, the A relay releases, which through A1M de-energizes B. In a short while, B releases. Contact B2M opens and removes the ground from the sleeve of incoming line. This causes all the previous switches to release, destroying the “calling line” portion of the connection. Step by step: The connector switch Page 16 In addition, the holding paths for relays K and F, which were completed to that sleeve, are now disabled. Those relays release, “turning loose” of the called line, ring, tip, and sleeve. Then, the path from ground through A1B, B1B, K5B, and VON 1M and through the release magnet to battery operates the release magnet. This allows the switch shaft to rotate back to its “zero” rotational position, and then to drop to the “fully home” position. That releases the VON contact set. Its contact de-energizing the release magnet (whose work is done). 1M opens, 5.12 But . . . If in fact when the calling station hangs up, the called station has not yet done so, what happens to the called station line when the connector completely “turns loose” of it? When ground is removed from the sleeve lead, the cutoff relay in the called line’s line circuit is released. The line relay is put back in place. Since the called station is still off-hook, the line relay is operated, just as if the subscriber had picked up the handset to place a call. A line finder is started and, when it “finds” the line, connects it to the associated first selector. The called line receives dial tone. Unless indeed the next thing the called subscriber planned to do was to make another call, this is of no use, and he hangs up, abandoning this fruitless partial connection. So, doesn’t this result in useless wear on the switches and some consumption of their traffic capacity? Yes. So why does it work that way? The reasons aren’t fully clear, and what I know about this is beyond the scope of this article. Nevertheless, in Appendix A I describe a cinematic convention (I hate it) that arose from this situation and another from a situation I will describe shortly. 5.13 Called party hangs up first Suppose that (contrary to the etiquette experts’ advice), at the end of the conversation, the called party hangs up first, and for some reason, the calling party doesn’t promptly follow suit. When the calling party hangs up, the D relay releases (immediately). Via its transfer contacts 1 and 2, the polarity of the battery fed to the calling line is restored to “normal” (indicating to interested entities upstream that the called party is now on-hook. That’s all? Yes, that’s all. The connection is not released? No. Step by step: The connector switch Page 17 In Appendix A, I describe a second part of the cinematic convention I mentioned before that results from this situation. 5.14 Back to the busy test There is an interesting wrinkle to the matter of testing the called line to see it is busy. I did not mention it earlier so as not to derail the point of the story. But it is interesting. Suppose that the first digit has been dialed and, after a brief pause, the second (and final) digit is being dialed. During this process, relay E is operated (it operates soon during the first dial pulse) and remain operated until a while after the dialing of this digit is done. So the entire time the second digit is being dialed, the sleeve wiper goes, through contacts E2M and K4B, through the coil of relay G to battery. Thus, as the wipers, heading toward the desired line terminals, pass over any line that is busy at the time, relay G will momentarily operate. The advantage of this, rather than waiting for E to release at the end of the second digit before testing the sleeve for ground, speeds up the process of reacting to the dialed line being busy. However, depending on where we are at the time in the “cadence” of the busy tone when G operates on a busy “earlier” line, this could cause busy tone to be briefly returned toward the calling station. Won’t that sound—well—odd to the calling subscriber? Well, this happens while a digit is being dialed. During that time, the receiver muting contact on the station dial mutes the receiver (to prevent the subscriber from hearing a loud series of clicks from the interruption of the line current to send the pulses forming the digits). So the calling subscriber won’t hear these anomalous bursts of busy tone at all. If this phenomenon had been considered a problem, a break contact on E in series with the busy tone feed would have eliminated it. But I have never seen that done. 6 HUNTING CONNECTORS 6.1 The rotary hunting connector Suppose a small law office has three telephone lines, but of course wants callers to dial a dingle directory number, which will result in the call being completed to one of the lines that is idle. This is the job of the rotary hunting connector. Its bank has four terminal contacts, T, R, S, and H. For our hypothetical law office, the Step by step: The connector switch Page 18 three lines have consecutive numbers, with all digits the same but the last, as for example (and from here on I will only speak of the last four digits) 4572, 4573, and 4574. The number listed in the directory, sometimes called the guide number, is 4572. On the connector bank, on the terminals for all but the last line, the H contact is connected to the S contact; for the last line, the H contact is open. In this switch, it is the H relay, not G, the looks for ground on the S contacts as the switch steps in the rotary direction. (Recall that the relay designations come from their locations in the switch, not from their function, and this relay ends up in a different spot in a rotary hunting connector, which has more relays altogether.) Suppose the first line (4572) is busy, and a caller dials 4572. When the connector lands on terminal 72, and since the sleeve is grounded, the H relay is operated, but here that does not result in busy tone being returned to the caller as the corresponding relay (G) would do in an ordinary connector. When the E relay releases to indicate that this digit is completed, relay G tests to see if the H terminal is grounded. It is, since the line being busy results in the S terminal being grounded and for this line the H contact is connected to it. The operation of relay G causes the connector mechanism to make an additional rotary step, landing on terminal 73. The circuit operation is very reminiscent of that used for rotary trunk hunting in a selector. Again, relay G tests for a ground on the H terminal. This line is idle, so the S terminal is at battery (through the cutoff relay) and so the H terminal is likewise at battery. Thus G does not operated, and there is no further rotary step taken. The battery on the sleeve operates relay K relay, which completes eh connection to the called line (here, 4573), and ringing begins on that line. Suppose that all three lines were busy. The process we described just above causes the switch to eventually step to terminal 74. Here, S is grounded, but not H. The result is that the switch does not step further, but relay J operates and the caller is sent busy tone. This scheme can accommodate up to 10 lines, in which case of course it is mandatory that the number of the first line (and thus the guide number) end with “1”. Step by step: The connector switch 6.2 6.2.1 Page 19 The level hunting connector Introduction But suppose that we have a department store with 35 lines (which actually go to a PBX at the store). But again, we want callers to be able to dial a guide number and end up connected to an idle one of the 30 lines. Providing for this is the job of the level hunting connector. There are two basic methods of operation for level hunting connectors. 6.2.2 Incremental level stepping This is the arrangement favored by Automatic Electric company. Imagine that the guide number is 8611. The digits “11” go into the connector as usual, and as usual, the switch wipers end up on terminal 11. The process proceeds as described just above for the rotary hunting connector, and if an idle line is found on this level, the call is completed to it. But here, the H contact is connected to the S contact at terminal 10 (since this is not in fact the last line of the group). So if that line is busy, the switch steps to rotary position 11 (where there is of course no terminal). A “fin” on the switch shaft operates a set of contacts, called the “11th step rotary contacts” (11SR). This sets of a chain of events. The vertical magnet is energized through a resistor, such that its pawl will be plunged into the vertical ratchet but not with enough force that it could lift the shaft if it were free to move (which it is not at this juncture). Then the release magnet is energized, which kicks out the double dog, which has been holding the shaft in position. The shaft