Reference / Paper · 1982
A History of Engineering and Science in the Bell System: Switching Technology (1925-1975)
Authoritative history of Bell switching including SxS, panel, crossbar and ESS.
Reference / Paper · 1982
Authoritative history of Bell switching including SxS, panel, crossbar and ESS.
Switching Technology (1925-1975) A History of Engineering and Science in the Bell System Switching Technology (1925-1975) Prepared by A. E. Joel, Jr. and Other Members of the Technical Staff, Bell Telephone Laboratories G. E. Schindler, Jr., Editor. Bell Telephone Laboratories, Incorporated Credits for figures taken from other than Bell Laboratories sources appear on page 601. Copyright @1982 Bell Telephone Laboratories, Inc. All rights reserved First Printing, 1982 International Standard Book Number: 0-932764-02-9 Library of Congress Catalog Card Number: 82-072517 Printed in the United States of America Contents Foreword ................................. .............................. ix Acknowledgments . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . .. . . . .. . . . . . . . . xiii 1. Setting and Scope ................................. ............. 1 I. The Setting, 1. 2. II. Scope, 2. State of the Art in 1925 .. .. .. . .. .. . .. .. .. .. .. .. .. .. .. .. .. .. .. .. 7 References, 9. 3. The Evolution of Electromechanical and Manual Switching Technology, 1925-1935 ...................... 11 I. Panel System Improvements, 11. II. Step-By-Step Improvements, 22. III. Apparatus and Equipment, 34. IV. Switchboard and Testboard Apparatus, 36. V. Signaling, 50. VI. Start of Operator Distance Dialing, 52. References, 54. 4. Introduction of Crossbar . .. .. . .. .. .. .. . .. . . . . . .. .. .. .. .. . .. . 59 I. Genesis of Crossbar in the Bell System, 59: The Coordinate Switching System, 61; Resumption of Work on the Crossbar Switch, 63; Switchboard Tandem Trunk Concentrator, 65; Other Apparatus Developments, 66. II. Local Crossbar, 67: No. 1 Crossbar System, 68; Crossbar Tandem, 73; No. 2 Crossbar System, 76; Early Crossbar CDOs and PBXs, 78. III. Toll Crossbar, 78. References, 84. v vi Engineering and Science in the Bell System 5. The Better Understanding of Switching . ... .. .. .. .. .. .. . 87 I. Early Research Contributions, 87. II. New System Ideas and Trades, 92: Automatic Ticketing, 92; Step-by-Step Senderization, 93; Higher Speed Step-by-Step, 93; Motorelay System, 93; The "Browser" Group, 94. III. Traffic Theory and Engineering, 95: Interoffice Networks, 96; Switching Center Networks, 100; Traffic Measurement (see also Chapter 11, section 3.4), 103;: Other Traffic Activities, 107. IV. Other Theoretical Studies, 110. V. From Boolean Algebra to Switching Algebra and Synthesis, 111. VI. The Teaching of Switching, 112. VII. Documentation, 113. VIII. Improvements in System Reliability, 113. References, 116. 6. Preparing for Full Automation .. .. .. .. .. . .. .. .. .. .. . . .. .. 121 I. lhe Switching Product Line of 1945, 121. II. Needs for Dial Service in the Smaller and Suburban Communities, 122. III. Nationwide Calling, 122:: Operator Distance Dialing, 123; Nationwide Numbering Plan, 123. IV. Longer Dist1mce Signaling, 128. V. Charge Recording Methods, 130:: Automatic Ticketing, 130; Automatic Message Accounting, 135; Centralization of AMA, 142; Automatic Number Identification, 148. References, 153. 7. No.5 Crossbar ................................................ 157 I. The Innovation, 157: The Ring Translator, 161; Other Inventions, 162. II. Media, Pa.-The First No.5 Crossbar, 164. III. Englewood-The Fir!it DOD Equipped Local Office, 165. IV. The Maturing of No. 5 Crossbar, 166. V. Packages and Smaller Versions, 168. VI. Non-New Jen;ey Locations for Switching Development, 170. References, 170. 8. Direct Distance Dialing .................................... 173 I. Nationwide Planning, 173. II. No. 4A Crossbar, 180: No. 4A and 4M Crossbar, 180; Card Translator, 181; Cost Reduction and Capacity Improvements, 185; III. The Full Stature of Crossbar Tandem, 188: Centralized Aultomatic Message Accounting (CAMA), 188; Toll Features, 190. IV. Auxiliary Senders, 190. V. International DDD (IDDD), 192. References, 196. 9. Beginnings of Eh~ctronic Switching .................... 199 I. Early Applications of Electronics, 199~ II. The Start of Systems Thinking, :zoo. III. Resear,:h System Experiments, 202. IV. The Start of Electronic Switching Development, 203: System Studies, 203; 1952-1954 Studies, 212; Exploratory Development, 214; Towards an Electronic Central Office--Pre-Morris, 225; Electronic PBX, 228; The Morris Field Trial, Contents vii 229; New Devices, 239. V. The Development of Electronic Switching, 249: From Morris to the Start of No. 1 ESS, 249; The No. 101 ESS Development, 252; Initial Development of No. 1 ESS, 257; UNICOM, 262; AUTOVON, 266. VI. Expansion of No. 1 ESS Applications, 269: Signal Processor and Capacity Improvement, 272; Centrex, 276; No. 1 ESS ADF, 278; 32 K Call Store, 281; Generic Programs and Feature Additions, 282; Remreed Network, 286; Centralized Maintenance, 288; Improved Trunk Circuits for No. 1 ESS, 290; The 1A Processor, 291; Improved Memory Systems, 295. VII. No. 2 ESS Local System Development, 296: The Initial System, 296; The 3A Central Control and No. 2B ESS, 298; Centrex and Other Feature Developments, 300. References, 302. 10. The Complete Electronic Switching Family .. . . . . . . . . 307 I. Traffic Service Position, 307: Early Studies, 307; Person-to-Person, Collect, Special Positions (PPCS), 309; TSP on Crossbar Tandem, 309; Traffic Service Position System (TSPS), 312; Hotel/Motel, 316; IDDD, 316; Integrated Circuits, 316; Remote Trunk Arrangement No. 1A and Position Subsystem No. 2A, 317; Automatic Coin Telephone Service (ACTS), 319. II. No. 4A Crossbar Electronic Translator System (4A/ETS), 319: Peripheral Bus Computer, 321; Common Channel Interoffice Signaling (CCIS)-Signal Transfer Point (STP), 321. III. Automatic Intercept System, 322: File Subsystem No. 2A, 327. IV. Toll Features on ESS, 328: HiLo Transmission, 329. V. Small Electronic Switching Systems, 330. References, 332. 11. The Service and Feature Era ...................... ........ 335 I. Introduction of New Services, 336: TOUCH-TONE* Service, 336; Wide Area Telephone Service (WATS), 342; INWATS, 343; Centrex (also see Chapter 13, section 3.1), 344; Automatic Identified Outward Dialing (AIOD) and Direct Inward Dialing (DID), 346; Data Services, 350; Emergency Service911, 352; Coin Service, 355; Common-Control Switching Arrangement (CCSA), 358; Expanded DOD, 359; Automated Intercept Service, 359; Mechanization of Service Evaluation, 361; Switching for PICTUREPHONE* Visual Telephone Service, 362; Electronics in Local Electromechanical Switching, 363; Other New Services, 366. II. Growth of Service, 368: Unigauge, 368; Foreign Area Translation, 369; All Number Calling, 370; New Numbering Plan, 370; Local/ Tandem/Toll, 371; No.5 Crossbar Route Translator, 372; Directory Assistance, 372; Charging, 378; Increased Crossbar System Capacity, 382; Party Lines and Concentrators, 382. III. Quality of Service, 385: DDD Improvements, 386; Main Distributing Frame (MDF), 388; Maintenance Improvement, 389; Traffic Measurement Systems, 400; New York Fire, 409. IV. New Features Made Possible by Electronic Switching, 411: Custom Calling Services, 411; Feature Growth, 412. References, 414. *Registered service mark of AT&T Co. viii Engineering and Science in the Bell System 12. Th1e Integrated Dilgital Toll Network . . . . . . . . . . . . . . . . . . . 421 I. No. 4 ESS-IntegratE~d Transmission and Switching, 421. II. Maintenant:e and Administrative Techniques, 427. III. Comm011 Channel Signaling, 4.30. IV. Stored-Program Controlled Network, 438. References, 440. 13. Private Branch Exchanges, Key Telephone Systems, and Special Systems .. . . .. . . .. . .. . . . . . .. . .. .. . .. 443 I. Beginning the Second Half Century, 443. II. Continued Evolution of PBX Systems, 444: The Thirties, 444; The Forties-War and Post-War, 447; The Fifties-Improvement, Change, Innovation, 451; The Sixties-Solid State and Super ServicE~s, 455; The Seventies-Realization of the Electronic Promise, 479. III. 1rechnology of Major PBX Service Innovations, 494: Centrex, 494; PBX Switchboard Innovations, 513; Innovations in Switching F·eatures, 519; TOUCH-TONE® Calling and PICTUREPHONE® Visual Telephone Service, 528. IV. Key Telephone Systems, 53'7: Some Comparisons with PBX Systems,. 537; Common Ancestry, 539; Divergence from PBX Technology, 539; Early Key Switching Systems, 540; The Key Telephone Systems Pattern Emerges, 543; Adding TOUCH-TONE Calling and PICTUREPHONE Visual Telephone Service to Key Telephone Systems, 549. V. Special Systems, 550: Telephone Answering Service, 553; Order-Taking Systems, 559; Systems for Dispatch and Emergency-Type Services, 566; Teletypewriter and Data Switching, 577; Switching of Network Facilities, 586. References, 590. Abbrevialtions and Acronyms 595 Credits . . .. . .. . . . . . . . . . . . . . .. .. .. .. .. .. .. .. . . . . . . . . . .. .. . . . .. . .. . . . . .. . 601 Chronology 605 Index .................................................................... 617 Foreword Very soon after Alexander Graham Bell invented the telephone, it became obvious that it would be impractical to extend wires from every telephone to all other telephones. To conserve copper and dollars, wires had to converge on central points where individual telephone-to-telephone connections would be made. That is, calls had to be switched. Moreover, another basic concept of telephony also quickly emerged-namely, the trade-off between switching and transmission. The transmission costs of wire, cable, and associated plant could be optimized by having switching offices distributed to serve centers of population. The number of switching offices could be adjusted with changes in transmission costs. A classic and continuing problem of switching, therefore, is to design switching systems with capabilities that enable them to be sited at demographically advantageous points, so that total network costs are kept as low as possible. Also, for public telephony the need for compatibility in the presence of technological growth was established very early. Every item of equipment must function accurately and reliably with every other item of equipment in a network, no matter how spread out nationally or internationally. In addition, the network must absorb a continuous stream of innovations so as to bring the benefits of technology to its users as rapidly as possible. In this book we see how these principles were applied during a significant period of technological change in Bell System history. The account begins in 1925, when various engineering and research groups of the Bell System were reorganized as Bell Telephone Laboratories. The nominal end point is 1975, which rounded out the first century of telephony. As it turns out, however, 1975 is a reasonably good year to mark the end of one era of telephone switching and the beginning of another: the rapidly emerging new era of a digital, stored-programcontrolled network. So, although we here concentrate primarily on the years 1925-1975, we have not hesitated to cover certain post1975 developments started prior to 1976, when we knew them to be significant to current trends. ix X Engineering and Science in the Bell System An earlier volume of this series (A History of Engineering and Science in the Bell System: The Early Years, 1875-1925), describes how techniques born of necessity changed the telephone from a novelty to a useful instrument of personal communication and of commerce. Switching technology during this period was characterized by the introduction and improvements in manual switchboards. In the years covered by this vollume, we see switching technology change from the final refinements in manual switching to work leading toward the introduction and complete take-over of automatic switching. In the process, the technology went through four successive phases: (1) indirect control, with switches requiring large mechanical motions, (2) common control, with switches requiring only very small motions, (3) stored-program control electronic switching, and, now (4) time division digital switching. At Bell Labs, each of these tedmologies moved from concepts to innovations, and the innovations were introduced to provide improved and expanded resources for the administration and operation of switching offices. The resulit has been high-quality and constantly improving telephone service and features, not only in the United States where the innovations were first employed, but throughout the world as these concepts were widely emulated. The first practical automatic switching system was the step-bystep system, so-called be·cause the switches move in step with, and under the direct control of, the telephone dial. With step-by-step, switch motions are limited by how fast (or slow) the customer dials, by how long it takes fo:r the pulses representing each digit to be received and detected in the central office and, of course, by mechanical inertia. Switching innovation progressed to the first or indirect control phase, the panel system, separating the control of switch mechanisms from the dialing process. With panel, nondecimal power-driven svvitches of large capacity were introduced, and the independence of dial and switches permitted translation between the information dialed by the customer and the control needs of the system. In the second phase, crossbar systems were designed and common control was first introduced. With common control the callhandling control was separated from the interconnection network and great improvements in flexibility were provided. Translations allowed complete flexibillity in the assignment of lines and trunks, numbering plan and routing flexibility was provided, and the trunk network became mo:re efficient and service again improved. In addition, the crossbar switches themselves introduced precious metal (platinum alloy) contacts into telephony which improved transmission performance. These relay-like contacts operated with very small motions and therefore with increased speed; this blended Foreword xi nicely with the common control equipment that controlled their operation. After World War II, the transistor and bulk electronic memory created tremendous new potentials for switching, and many exploratory developments were undertaken at Bell Labs to convert them into realities. The result was the third and entirely new approach to switching systems design-stored-program controland switching took its place at the frontiers of the new electronic technology in terms of sophisticated hardware and software, systems architectures, complexity and size. The relationship to computers is obvious, but the switching system designs were unique. In real time, electronic switching systems process telephone calls with an objective of two hours downtime in 40 years, or about one-thousandth of one percent, compared to the one percent performance of a typical commercial computer. Now more than 50 percent of Bell System calls are switched using stored-program techniques. The fourth era of switching technology, and the one currently in progress, is a period of further extending digital techniques into switching to encompass the interface with transmission as well and, indeed, to require the engineering of switching systems and transmission terminals carrying digital signals as single, integrated entities. The Bell System was the first to combine stored-program control and time division digital switching. Furthermore, a new high-speed digital signaling technique, known as common channel interoffice signaling (CCIS), was introduced in 1976 and soon will be serving 50 percent of all toll calls. Thus the foundation for future high-efficiency, low-cost telecommunications has been established. The new era of integrated digital switching and transmission will continue to extend service beyond the telephone to include more complex terminals and all forms of information-voice, data, and video. The advantages are greatly expanded services, new services, new and more powerful operational and management techniques, and sophisticated measurement and maintenance arrangements. And the network will continue to grow, both domestically with the cooperation of the independent telephone manufacturing and operating companies and internationally with the cooperation of foreign administrations. In all probability, the next fifty years will see changes take place even more rapidly than those of the first fifty years of Bell Laboratories and the first 100 years of the telephone. While telecommunication switching technology and its application will continue to play important roles, the technical and managerial expertise of Bell System people will be the most important ingredient in insuring this progress. xii Engineering and Science in the Bell System The contributions of many Bell System people are acknowledged in this book, but it is important to recognize that, as the technology has become more sophisticated, it has taken ever larger teams of designers and managers to convert inventions into developed and engineerable systems and, therefore, individual contributions are not as easily identified. This book is dedicated to the generations of engineers who made these accomplishments possible. On a personal note, it has been my pleasure to share with many colleagues in this, one of the great technological adventures of the twentieth century. By whatever measure--complexity, numbers of people, development costs, investment, or telephone traffic processed--this adventure is a huge one. The greatest satisfactions, however" come not from designing something big but from seeing the human achievements involved when individuals tackle and solve some of engineering's most challenging problems. W. 0. Fleckenstein Vice President, Switching Systems April1982 Acknowledgments Although I was the principal author of Chapters 1 through 12, a major Chapter (13) devoted to the development of PBXs and other customer-premises switching systems was prepared by Charles Breen who, prior to his retirement from Bell Laboratories, was associated for many years with the engineering of these systems. Other contributors to the customer-premises switching chapter were George M. Anderson, PBXs from 1970 to 1975, and Frank M. Fenton, key systems from 1970 to 1975. Frank F. Taylor prepared sections IV, V, and VI of Chapter 9, and was most helpful in editing the entire manuscript. WalterS. Hayward reviewed and contributed to section III of Chapter 5, and sections III and IV of Chapter 11. I would also like to acknowledge the considerable help of Janet Edwards, who keyboarded the manuscript and its numerous revisions into the Bell Labs UNIX* system word processor, Ruth L. Stumm, who searched diligently for illustrations and helped with proofreading, Terry Getz, who prepared the illustrations for publication, and Arlene M. Dagostino who took the words and illustrations and made them into a book. One could not prepare a book of this type without continuously monitoring and collecting data over a period of time. My wife, Rhoda, provided much encouragement, and my secretary, Frances Anderson has been most patient, participating in this process as well as typing all of the original manuscript drafts. All contributors thank George E. Schindler, Jr., our editor and mentor, whose encouragement and continuous help were important to the creation of this history. In this book, readers should note that numbers of installations and entities refer only to the period covered by this history. Such numbers will in many cases be different when updated through recent years. *UNIX is a trademark of Bell Laboratories. xiii xiv Engineering and Science in the Bell System Over many years, during which colleagues have provided insight into the developme nts with which they were engaged, I was privileged to observe their progress, reflect on these activities, and now to record in this volume the events as I recall them. The subjects covered have been chosen from a larger set. There are always questions of history that one cannot anticipate even in a most comprehensive volume. I hope readers will find the material included here interesting to peruse and useful for reference. A thorough reading should provide an understand ing of how the research and development activities of Bell Laboratories during its first 50 years have made available the switching techniques and technology for serving the nationwide network during a critical period of its growth. Amos E. Joel, Jr. The Mulberry office switchboards (originating in the background and terminating in the foreground) used with the first panel switching equipment in Newark, New Jersey in 1915. Although manual and semiautomatic switching were to remain in use for many years, it became apparent that full automatic switching and high-capacity transmission systems were the path to more and better telephone services. To reach that goal , manual systems had to be adapted to work compatibly with the new automatic systems. Moreover, telephone devices, logic circuits, numbering and routing plans, and administration and maintenance techniques all grew in complexity and ability to handle large amounts of local and toll traffic . Chapter 1 Setting and Scope By 1925, Bell System engineers had built a large body of knowledge concerning complex logic circuits and had begun to apply this knowledge to the design of automatic switching machines, and, subsequently, to the design of computers. Fifty years later, in 1975, a huge nationwide automated network was in place and working. The soundness of common-control principles and of stored-program control had been repeatedly demonstrated, and customers could look forward to a network of ever-greater machine intelligence and utility. I. THE SETTING If one were to attempt a list of the major technological achievements in telephony to date, three of them would surely be analog and digital multiplex transmission, common-control switching, and stored-program control (SPC) switching. Multiplex transmission greatly improves transmission efficiency by sending more than one signal (sometimes thousands or even tens of thousands) over a single transmission medium. Common-control switching and SPC, though less well known outside of engineering communities, are at least of equal importance in terms of economics and technical sophistication. Common-control switching and SPC are analogous to multiplex transmission in the sense that they greatly improve efficiency. Older systems tied up expensive equipment during the time of a telephone call, but common-control systems segregate equipment into a switching network and the control units. The switching network consists of paths for voice signals that of necessity have to be assigned for continuous use during calls because they carry instantaneous two-way telephone conversations. But the common-control units quickly "jump in and out of" the various stages of setting up and taking down a telephone connection, and so work quickly and with great efficiency. And SPC has added new dimensions of flexibility and speed. From these concepts and their application in many intricate ways throughout the Bell System, there have flowed enor1 2 Engineering and Science in the Bell System mous benefits to customers in terms of good service and new services at attractive costs. II. SCOPE A previous volume of this series, titled A History of Engineering and Science in fhe Bell System: The Early Years (1875-1925), covers important Bell System innovations in manual switching and also covers the beginnings of attempts to assist operators or to substitute for operators through the use of switching machines. From the earliest experiments with automatic switching, and extending for almost a quarter century, the precise directions the new switching arts should take were not entirely clear. Basic problems of the configuration and design of central office systems had to be solved. With time it was appreciated that large cities required switching capabilities different from those of less densely populated areas. Large local networks were put in service with manual techniques. Tandem offices (switching offices that serve other switching offices) afforded opportunities in these networks to try some of the new technologies of automated switching later to be applied more generally throughout the Bell System. The new techniques helped put to rest beliefs that semiautomatic techniques might be superior to fully automatic switching. On the other hand, the new techniques also showed the need to develop arrangements to handle calls between manual and automatic offices during the transition period. By 1925 the designers of switching systems had become most adept in devising complex logic circuits using general-purpose electromagnetic relays of telephone quality. Some of these circuits were the forerunners of those used by designers of computers, who began their work about 25 years later. Indeed, telephone engineers at Bell Laboratories applied their art to the early relay computers (see A History of Engineering and Science in the Bell System: National Service in War and Peace: 1925-1975, pages 133 through 174). During the second 50 years of telephone switching, covered by this volume, manual switching was extended by using and improving upon many of the successful techniques invented and applied by the pre-1925 pioneers. Long distance calling became possible without the delay caused by the awkward procedure of having the customer reach a second (toll) operator, hang up, and wait to be called back. The bulk of this volume" however, will of necessity be devoted to a comprehensive account of Bell System developments in automatic central office switching and in the types of switching used in private branch exchanges (PBXs) and key telephones-the latter two categories often characterized as customer-premises switching. Setting and Scope 3 The period covered, 1925 to 1975, is an especially important one since these 50 years saw the integration, automation, and great expansion of a nationwide telecommunications network. As mentioned above, this p~riod is also characterized by the two pioneering achievements of first introducing common-control electronic switching and, later, stored-program control electronic switching. In the area of customer-premises switching, automated PBXs and key telephone systems were gradually and ubiquitously introduced. Complex logic circuits, and particularly the skill with which they were applied to insure reliable service, were keystones to the development and expansion of the concepts of common-control switching. Prior to the adoption of these principles, system intelligence was in the logic circuits associated with the individual switching mechanisms that progressively established a connection. These circuits could not be reused until the switches became available for use on another call. Innovative techniques, however, were applied with increasing capability and skill by Bell Laboratories engineers, which meant that the logic circuits could gather facts and make the decisions needed to complete connections, and could do so "in common" for many calls in sequence without awaiting the completion of any one call. Furthermore, the decisions made by these control circuits could be based upon the prior knowledge. of the status of connections already established. In a word, the systems were acquiring a form of machine intelligence. These fundamentals of indirect and common control brought early, worldwide recognition to switching innovations of Bell Laboratories people. Although others had proposed aspects of such logic, Bell Labs engineers had devised the necessary and sufficient conditions for their application in large systems. Most of the systems were perceived as being required to fill the expected growth of the Bell System. These included switching systems not only for small, large, and medium-size local serving areas, but for tandem and toll service as well. The success of development techniques and the deployment of common control provided momentum towards allowing customers to dial more and more of their own telephone calls. New developments in signaling and the recording of charges provided the tools. Network plans first permitted operators to dial long distance calls, then led to a transition period during which direct distance dialing of long distance calls by customers became a reality. Important to this work were devices, common-control circuit configurations, and numbering and routing plans that today are integral to the nationwide network. At the end of this era, Bell engineers were able to look back on the automated network of switching systems as the largest distributed computer in the world. 4 Engineering and Science in the Bell System The application of common-control techniques approached a peak in the 1950s with the maturing of developments for the No. 4A and No.5 crossbar systems. At the same time, and drawing on the large base of electromechanical switching expertise, a new generation of switching systems techniques emerged, using the new disciplines of high-speed electronics as conceived and evolved in the research efforts at Bell Laboratories. (These innovations are to be covered in another volume of this series, dealing with research.) An exploratory development effort was started with the objective of applying electronics for the logic and memory and in the network portions of central office and PBX switching systems. The result was the subsequent development of the Morris (1960) space-division central office electronic system, named for Morris, Illinois, the site of the first field trial. This was followed in 1965 by No. 1 ESS (electronic switching system), also a space-division central office electronic system, and the No. 101 ESS (1963) time-division customer-premises switching system. The most important result of this exploratory effort was the emergence of the concept of storing the logic of the system actions in machine memory, a concept now widely recognized and used worldwide~ as stored-program control (SPC). The Morris development effmt proved the effectiveness of the concept at its initial implementation, and SPC subsequently ushered in a new era in switching. Progress in SPC technology has now made it economical to include SPC in the smaller switching and key telephone systems, such as for community dial offices and customer premises. The irtvention of SPC at Bell Laboratories was in part stimulated by the growing capability required by common-control systems, such as No . 5 crossbar, as described in Chapter 11 of this volume. These many requirements for extending service beyond "plain old telephone service" (POTS) made the switching systems of the growing network the locations for introducing many sophisticated features that offered new services. Associated with this growth was the realization of the need for many ancillary systems for traffic and service measurement and evaluation, number services, network management, and service maintenance. These adjuncts started small, but as their need was proved by their development and deployment, sometimes as small systems, they became a major class of systems known as operation support systems (OSS). This volume describes developments from the time when most calls were handled manually, when the rudimentary long distance network had just a few circuits, and when these circuits were available only after waiting in a queue. As the transmission art developed and the network grew, larger groups of circuits were provided at lower cost. The switching challenge was to bring the calls Setting and Scope 5 to the network and to learn to serve them on demand, with essentially no delay. Recent developme nts in switching, described in Chapter 12, have brought transmissio n and switching closer together technologically, since each serves digital signals that represent the speech message signals. Digital signals are switched using time-division techniques-tha t is, the distinguishing feature of a signal is its occurrence in a particular time slot, as distinct from space division, in which a signal can be identified by its physical location in an array of equipment. When the technology history of the next 50 years is written, the digitization of the network will be a major theme. Panel dial system Laboratory, New York City 1930. The direction of switching systems development was influenced by the need for automatic equipment in the large central offices served by the Bell System, and the panel system represented a major effort of the years immediately preceding and following the establishment of Bell Laboratories in 1925. Panel successfully divorced switch control from the dial-a necessary prelude to modem switching efficiency and flexibility. The last panel office in the Bell System was removed from service in 1982. Chapter 2 State of the Art in 1925 By 1925, the Bell System had taken a number of basic steps leading toward full automatic switching. Engineers had developed interfaces with manual systems, without which dial systems could not grow. The principle of indirect control of switches, initially with register-senders, was an established design concept. And experimenters probed the idea of a coordinate switch to decrease mechanical switch motions and thus speed servicing of telephone calls. The first half century of the telephone 1 was dominated by manual switching systems and equipment. Practical methods of operation were devised, and apparatus was developed for high reliability so as to keep maintenance costs low. The methods for operating manual switching equipment for both local and toll switching were constantly improved so that the least amount of labor and shortest call-completion times could be realized. Prior to 1925, manual toll switching was also improved to the extent possible with the limited amount of toll transmission facilities then available. This meant devi~ing methods for deaijng ~ith lar~e numbers of calls whose completion had to be delayed while awaiting the availability of transmission facilities. Long distance signaling methods were complex but, by today's standards, primitive. In automatic switching, Bell System developments were quite significant, although they started slowly. While the step-by-step system was being sold throughout the world, Western Electric Company engineers were devising switching mechanisms of greater size so as to provide better service for the larger central offices found in the Bell System. Initially, the approach was to use electromechanical switching as an aid to the operators in establishing connections rather than as a means to allow customers to dial their own calls. This semiautomatic method of operating was first thought to be a better mix between the labor costs and maintenance and operation 1 References are grouped at the end of each chapter. 7 8 Engineering and Science in the Bell System than full automatic operation. Several field trials of panel and rotary equipment devised by Western Electric engineers were tried in this country and abroad. (See discussions of these systems on pages 580 to 600 in ref. 1, A History of Engineering and Science in the Bell System: The Early Years . 1875-1925.) The application of these "machine" switches brought forth many new design concepts. Primary among these was the indirect control of switchE!S from register.. senders. As a result, it was possible to divorce the operation of the switches from the method used for placing information into the system and for making switch selections required to reach the desired destination. Another basic concept that accelerated the decision to accept full automatic operation was the addition of letters as well as numerals to the telephone dial (PENnsylvania 5000, etc.). As a result, there was greater acceptance of the possibility that customers could accurately dial six or seven digits, since the central office code, as well as the numerical digits, could be dialed from memory. (As discussed in Chapter 11, section 2.3, later human-factors studies showed that there was no need for letters in the dialing sequence.) To test the new switching systems in actual service meant that the Western Electric designers had to solve the basic problems of interconnection between manual and automatic switching systems. As a result, the call indicator--a method of displaying numbers on calls from automatic systems to operators' positions-was successfully tried, improved, and made ready for introduction. Also developed were switchboards allowing orerators to originate calls to be completed by automatic sysbems. These innovations were known as "semimechanical" systems. Many of them were tried in a "metropolitan tandem" system introduced into New York in 1920.3 By 1919, the Bell System had taken the major steps to go to full automatic switching. Both the panel and step-by-step systems were standardized, and Western Electric engineers set out to develop both systems for large-scale manufacture and installation. Thus, by 1925 the number of dial telephones in the Bell System exceeded the number in the U.S. independent telephone industry, which had started automation much earlier. Almost unnoticed during this period were the seeds planted for the next generation of automatic switching equipment (see Chapter 4, section 1). In 1913, patents were secured by Western Electric on coordinate switching mechanisms-methods for closing electrical contacts at points in an x-y coordinate array, rather than by moving a contact over an appreciable distance to make (establish) a desired connection. Over the next ten years, many improvements in coordinate designs were proposed, the purpose being to replace the rotary and Keith4 line switches used in the panel and step-by-step systems State of the Art in 1925 9 and the large vertical-motion switches of the panel system (see next Chapter, section 1). As more laboratory experience with these coordinate line switches was obtained, the designers proposed complete coordinate systems as improvem ents over the panel system designs. This work went forward until1923, when a complete coordinate system was designed and set up in the Western Electric laboratories at 463 West Street in New York City (see Chapter 4, section 1.1). In 1926, it was decided not to proceed with the further developm ent of this system since it could not compete economically with the panel system, which was still being improved and reduced in cost. REFERENCES, CHAPTER 2 1. Fagen, M. D., ed. A History of Engineering and Science in the Bell System: The Early Years (1875-1925). Murray Hill, New Jersey: Bell Telephone Laboratories, Incorporated, 1975, Chapter 6, pp. 467-714. 2. Ibid., p. 582. 3. Ibid., p. 584. 4. Smith, A. B., and Campbell, W. L. Automatic Telephony. New York: McGraw-Hill, 1914, p. 38. Incoming selector frames, panel office, Cortlandt Street, New York City, 1931. Installed primarily in larger cities, the panel system was the most sophisticated automatic switching system of its day. Innovations incorporated by Bell Laboratories included a decoder using large multicontact relays, trouble indicators with lamp or teletypewriter output, and line finders that removed the need for having an individual switch for every telephone line. One version of panel included a "call announcer"--a system of announcing station-code numbers by converting dialed digits into voice messages using strips of sound-on-film. By 1958, panel served over 3. 8 mill ion telephone lines in the Bell System. Chapter 3 The Evolution of Electromechanical and Manual Switching Technology 1925-1935 One of the first large switching jobs of the newly incorporated Bell Laboratories in 1925 was to make major improvements in the panel system-an effort