The History of Relays in Telephony
Tracing its125 Year Reign
The word relay derives from an idea in hunting, Old French relai, "Hounds placed along a line of chase (to replace those that tire)”. As "a squad of men to take a spell or turn of work at stated intervals," by 1808. As a type of footrace, it is attested from 1898. The electromagnetic instrument by name from 1838, originally in telegraphy.
​
Electro-mechanical relays were essential for circuit switching in the 19th and early 20th century. They supported telegraph signal repeating, telephone switchboards, automatic exchanges and control systems. The invention of the telegraph circa 1838, along with progress in understanding electromagnetism, paved the way for the telephone in 1876 (see Accidental Telephone). The relay as a switch was a crucial element in electrical systems design for nearly 125 years [Ref: Switch].
The historical evolution of the relay spans from first light to use in telegraphy, its widespread adoption in telephone switchboards/exchanges, its importance in computing and industrial applications of all sorts.
​
Though there are many applications for relays, this overview focuses on telegraphy, telephony and touches on computing. The key principles covered apply to all relays, regardless of their application or size. Although relay-like, large talking-path switches are not covered here (see Endnote A).​
​
What is a Relay?
The central element of a relay is its electromagnet, a coil of wire with an iron core (see Fig 1). Current passing through the coil creates a magnetic field and an attracting effect creates a switching action.
The figure shows 3 contacts, A, B and C, the switch part. With the coil not energized, A connects to B. When energized, a part of A moves connecting to C. When the current is removed a spring-like action returns contact A to touch B. It’s an electrically operated switch.
Fig 1, Relay fundamentals [Porter]
1.0 The Relay’s Founding Fathers: The Discoverers
The road to the relay’s invention was not a straight path. As with most important inventions, the idea developed after repeated experiments and tweaking of the concepts. Three men deserve credit as discoverers and three as inventors.
First let’s look at the discoverers/experimenters. See Fig 2. Foremost is Hans Orsted of Denmark. During a lecture demonstration in 1820, while setting up an apparatus, Orsted noticed that when he turned an electric current on, a compass needle held nearby deflected away from magnetic north, where it normally pointed. The compass needle barely moved, yet it was enough to spark his curiosity [Orsted].
Fig 2, The Discoverers
On July 21, 1820, Orsted published his results in a pamphlet. His results were mainly qualitative, but the effect was clear – an electric current generates a magnetic force. As a professor, he was no doubt influenced by William Gilbert's great 1600 treatise De Magnete. See Endnote E. For his discovery, the Royal Society of London awarded Orsted the Copley Medal in 1820.
​
Second in line was William Sturgeon an English electrical engineer, Fig 2, middle. [Sturgeon].
He was a hands‑on experimentalist and lecturer who published and demonstrated practical electromagnets in 1825. He constructed the first practical electromagnet. His 7-ounce (200-gram) magnet was able to support 9 pounds (4 kilograms) of iron using the current from a single cell [Fahie, Fig 10]. His work built on the earlier link between electricity and magnetism that Orsted had revealed.
Thirdly, Joseph Henry (Fig 2 right) built upon the work of William Sturgeon. He was an American physicist, inventor, and influential science administrator at the Smithsonian. Henry was interested in electromagnetism and liked to demonstrate its effects with science experiments. He created an electromagnet capable of lifting 750 pounds (340 kilograms), surpassing all previous designs [Henry].
Next, let’s look at 3 influential inventors as seen in Fig 3. Because Joseph Henry was an experimentalist and an inventor, let’s count him twice.
1.1 The Relay’s Founding Fathers: The Inventors
Fig 3, The Inventors
Invention tends to trail discovery, not lead it. At times, experimentalists are also product inventors. Think of Edison and his light bulb filament tests. Henry is in this category but nowhere near Edison. He was never granted a patent despite having invented an iron ore separator with Allen Penfield [Henry], invented the doorbell, co-invented the relay and laid the groundwork for the telegraph. The common unit of inductance, H, is named after Henry.​
Henry's invention of an oscillating machine was a significant advancement. He published a groundbreaking article in Silliman’s Journal of American Science, in 1831, titled "On a Reciprocating motion produced by Magnetic Attraction and Repulsion." [Silliman]
Henry describes a small “machine” that produces an oscillating (rocking up and down) motion using magnetic attraction and repulsion (Fig 4). This mechanism was the first of its kind. How does it work?
Fig 4, An oscillating electro-magnetic machine [from Silliman]
A single coil (with ends A and B) is hinged between two stationary magnets (C and D). As the coil-arm swings, say down on the right side, the 2 wires on that end touch the terminals (small cups of mercury) of battery F. The coil becomes magnetized, and side B is repelled (N-N) while side A is attracted (S-N) to their respective permanent magnets.
​
Cleverly, Henry’s design reverses the coil’s magnetic field once the arm’s 2 wires connect to battery G and the process repeats causing an oscillation. The motion is described reaching 75 cycles per minute.
Henry thought the device was a “toy” but in one version the swinging arm rang a bell. He acknowledged that the principle may be applied to some "useful purpose" in the future. The machine had some characteristics of relays; electromagnets, switching contacts, and “arm movements”.
According to the Chronology of Smithsonian History (title: Joseph Henry Demonstrates Primitive "Relay" Telegraph) – “In 1835, using a self-made electromagnet, Joseph Henry arranged a small intensity (electro-) magnet, which works well at low power over great distances, to control a much larger (electro-) magnet.” This is the definition of a relay in action. Apparently, there was a “make contact” on the first relay to energize a second electro-mechanical device. Henry did not patent his idea and there is no known image of his relay.
There is no evidence that Joseph Henry ever used the English word "relay" in his 1830s–1850s writings. Henry left three manuscript volumes (now at the Smithsonian) of his experiments, many on electricity/magnetism demonstrating relay fundamentals.
Incidentally, at times both Sturgeon and Henry used mercury in cups as relay terminals. Why? Reliable spring‑loaded contacts hadn’t yet been perfected, especially for experimental works.
It’s not unusual for brilliant concepts to emerge simultaneously throughout history. Following Henry, Edward Davy’s work emerged.
Davy’s contribution
In England, Edward Davy (1806-1885) independently built an “electric relay” for his telegraph system (not to be confused with the more famous Sir Humphry Davy). He described a device to “renew” a signal for long-distance transmission. He was an amateur experimentalist compared to Henry. There is no evidence that Edward Davy knew of Joseph Henry’s work on relays.
Davy invented his relay to “renew the current” along long-distance telegraph lines and to steer telegraph signals. Davy’s 1838 British patent GB183807719A (title: Telegraphs) provides a grand example of his relay invention. In 2025, this author visited the British Library in London to obtain a scan of this offline patent. Davy was not the first to experiment with the electric telegraph, see Endnote B1
​
While most sources claim Davy only used the term "electric re-newer," a review of the primary 1838 patent text reveals he used the word "relay" three times. See Fig 3 (center), for a quote of his first "relay claim" in the patent.
​
Fig 5 shows a figure from his patent. This is clearly a signal routing relay with two groups of “3 make” contacts. This figure is part of a much larger circuit that describes telegraph signal renewal and path steering.
Fig 5 – Davy’s multi-path switching relay 1838
Two electro-magnets are separately energized as needed to steer the A-C or D-F signals. In the figure, coil 3 is energized so that the arm’s right-side three metal contacts bridge line E1 to line E2 and D1/D2 and F1/F2. There are 18 contacts in total. This configuration was remarkably advanced for an early prototype. With a little tweaking this could easily be a 3-state relay: L-on, All-off, R-on. He was the first person to patent the idea of an electro-mechanical relay.​
​
Davy demonstrated a working model of his needle telegraph in 1837 at the Exeter Hall in central London along with relays.
What does history record regarding the Davy relay? According to J. J. Fahie’s research [Fahie] in 1884, the invention of the electro‑mechanical relay belongs jointly to Henry and Davy, rather than later telegraph pioneers such as Wheatstone, Cooke, Bain, Gale or Morse.
Fahie said, “There can be no doubt that to Davy belongs the credit of the discovery of the relay system. The Electric Telegraph Company bought his patent, chiefly because it covered the use of the relay.” The same source credits Joseph Henry as the relay co-inventor despite no patents in his name.
Vail’s contribution
The use of relays in telephone systems is largely based on their first use in telegraph systems. So, it’s fitting to understand what progress relay design made during the “golden age” (~1838-1866) of telegraph invention from basic electric telegraphs to mature international network infrastructure.
While Joseph Henry and Edward Davy co-invented the relay, Alfred Vail was a bridge. He was the practical translator, along with scientific collaborator Professor Leonard Gale, who helped Morse’s idea become technically viable. Although Morse is often credited as the “father of the telegraph,” historians note that Vail’s engineering skills during 1837 to 1848 were indispensable. Without Vail, Morse’s vision might never have left the laboratory.
From a Smithsonian document, summarized here: Although Morse held the principal patents, Alfred Vail’s practical engineering, shop work, and operational management were essential to turning the laboratory demonstrations into a functioning system. [Smith2]
Importantly, Vail’s expertise as a machinist helped improve relay designs. Fig 6 [Vail, pg 42] shows a beautiful relay designed by Vail in 1843. This version was designed to test relay operation speed. Could it keep up with a fast telegrapher’s code tapping?
Fig 6 – Vail relay wired for testing its operation speed
Notice all the hallmarks of a modern relay: coil A, iron core D, armature lever B, hinge point K, contact spring arm J, and contact points J-V. The wiring for this case is to test the speed of relay operation. With A energized, C/B is attracted to D thereby breaking the J-V connection and causing the relay to quickly oscillate and as Vail states “the motion is so rapid as to produce a humming noise.” [Vail, pg 42]
​​
Alfred Vail viewed leveraging electro-magnetics not merely as a utility, but as a profound leap in human capability. In his 1845 description, he reverently refers to the telegraph as "That last and most wondrous birth of this wonder-working age." Alexander Graham Bell would likely agree with these words given his intense interest in the telegraph starting circa 1872.
This concludes the coverage of the relay’s genealogy. Next, let’s examine the adoption and application of relays in early telephone switchboard systems.
1.2 Inventing the Telephone Switchboard
It was June 1876 when Alexander G. Bell gave a dazzling demonstration of his telephone at the Centennial Exhibition in Philadelphia.
Attendee Sir William Thomson (Lord Kelvin), a world-leading expert on electric theory, called it "the most wonderful thing I have seen in America." His widely enthusiastic promotion was crucial for Bell's immediate credibility, especially in Europe [Thomson].
Mechanic and inventor George Coy was inspired by Alexander Graham Bell's lecture at the Opera House in New Haven, Conn. on April 27, 1877. Within 9 months, Coy built the world’s first switchboard in New Haven, inaugurated January 1878 with 21 customers [U-Conn]. This was only 18 months since the Centennial demonstration. Figure 7 is a circuit diagram of Coy’s first board [Rhodes, pg 168]. In some ways it was very advanced as we shall see. Rhodes describes the operations to establish a new call as follows:​
“Subscriber A, on line 1, desiring to talk to subscriber B, on line 4, depresses his push button P-A, which momentarily opens the circuit through Line relay R-1, causing its armature to release and close a circuit to power annunciator (buzzer) A-1.”
The other operations are not covered here.
Fig 7, Circuit diagram of Coy's first switchboard
Note four important points:
• There is no “magneto generator” in the subscriber’s set to signal the operator
• There is a “Line relay” per subscriber, 8 total
• There is no battery in the subscriber's telephone
• The Line relay is always on unless the subscriber pushes button P-A, causing
the R-1 to release and power its paired annunciator (A-1)
It's surprising to see a Line relay powered by a central battery B-1. At the time of Coy’s switchboard, there were no hand cranked telephones to announce a new call. So, Coy used a DC operated Line relay to trigger a new call.
​
Soon there were more than 150 subscribers on twelve subscriber-lines (12 Line relays), and the ratio of calls per subscriber was constantly on the increase [de Land]. This is about 13 "party line subscribers" per landline. See Endnote C.
Why was the switchboard Line relay energized when the telephone was not in use? Operators “rang up” a called subscriber using a very loud buzzing sound heard in the primitive receiver. So, in this sense, the telephone was “always on” (zero privacy). This was soon remedied with the invention of the telephone ringer.
Figure 7A shows the primitive telephone that connected to the new switchboard. No dial, no bell, no battery. The microphone was also the receiver. In some ways, this arrangement was very advanced and forecast the concept of a "Common Battery" system, discussed below. ​
Fig 7A, Primitive telephone attached to Coy's Switchboard [de Land]
Rhodes notes that by 1881, at a National Telephone Exchange Association meeting, magneto calling was quickly replacing battery/line-relay methods. Figure 8 shows this new arrangement. Hand cranked magneto currents cause a switchboard ‘Drop’ to fall (operator, it’s me!) and when the operator inserts the plug under the Drop it is “cut off” and in some cases resets automatically. 500 subscribers meant 500 Drops. If the switchboard was a “multiple” (say 2), then 1,000 Drops were installed. Not cheap.
Fig 8, Announcing a new call with a ‘Drop’ [McMeen]
Because of the magneto operated Drop, the switchboard line relay was abandoned circa 1880. The Drop was relay-like but without contacts. However, the line relay would soon be revived...it's not dead yet.
Improving the switchboard’s indicators (alarms)
Several forms of switchboard alarms were designed and provided to be effective. The most accepted, at the time, was a lamp indicator proposed by J. J. O’Connell of Chicago, Illinois in 1891 [Kingsbury, pg 382].
Fig 9, O’Connell proposed lamp indicator circuit 1891
Operation is simple, see Figure 9. As the subscriber cranks the magneto, the polarized relay (a specialized Line relay) responds to, say the positive part of the sinewave only, and turns on the lamp and the “signal relay” during cranking. The signal relay operates a bell to catch the operator’s attention. Importantly, only one annunciator per switchboard and no expensive electro-mechanical drops. Polarized and 2-Coil Shunt Field relays are useful for "polarity logic" where a relay action's depend on the polarity of a coil's applied voltage [Palmer, Pg 63].
So, about 10 years after the line relay was dumped it came back to life, never leaving the stage again. O’Connell’s idea had legs and in 1895 J. E. Kingsbury was granted British patent 11,549 claiming a new form of fast acting relay (Fig 10) to replace the expensive polarized relay and with no magneto needed in the subscriber set. When engaged, the disc armature contacts the center pin, lights the lamp, and falls away via gravity when released (Fig 10 relay is flipped vertically when mounted).
Fig 10, Kingsbury patent figure (simplified) of Line relay in circuit
Advancing the art, Mark Edson (American Bell Telephone) filed an innovative patent US550260A in 1895. This was for his “Knife-edge” device as it was commonly called. It was a Line relay integrated with a lamp. See Figure 11.
Fig 11, From Edson’s patent, line relay + lamp
From the patent, “A great advantage of this construction is that the armature (9) resting on its sharp edge on the internal periphery of a circle considerably larger than itself is almost frictionless.” Hence, the name “Knife-edge” relay. This device saved switchboard space compared to using separate relays and lamps. Over time a second relay was introduced, the Cut-off (or CO) relay. See Endnote D for more on these relays.
​
The Knife-edge device or similar became practically mandatory on all switchboards and automatic systems starting ~1895.
Common battery system advantages
The invention of the common battery (CB) system by Hammond V. Hayes in 1892 (US patent US474323A) increased relay usage per line. Some advantages were:​
​
•Centralized power supply: One large, well‑maintained and charged battery at the exchange served all subscribers.
​
•Line relay operation: When a subscriber lifted the receiver, the loop closed, and current from the common battery flowed, operating the relay and powering a lamp (or a drop) at the operator’s board.
​
•Supervisory signaling: The same battery allowed the operator to detect when a call was answered or terminated.
​
•Ease of use: Subscribers no longer needed a local battery or had to crank a magneto; they simply lifted the receiver to signal the operator.
With the common battery approach, it was much easier to add relays as needed to an exchange. With growth comes new relay models with better costing, manufacturability, reliability and flexibility.
What voltages did relays operate with?
Early telegraph offices connected multiple 1.1V battery cells in series to achieve 3–12 V across the line, enough to actuate the electromagnets in sounders or needle instruments across short distance lines. The nominal battery voltage of a common battery switchboard exchange was 22-24 volts while the automatic used ~48 volts. Of course, worldwide, there are always exceptions.
​
Figs 12-14 reviews some typical relay models starting from 1902.
Fig 12, Early relay types used in switchboards
Fig 13, Key relay features and general purpose examples [Keister]
Fig 14, Flat Spring relay after engineer E.B. Craft (Source: author)
Relay design continued in the Bell System, and GPO in Britain, until the 1970’s. In 1956 Western Electric made 30 million per year. They manufactured 98 types and a total of about 6,400 different codes in 1956 [Mueller]. [Hawthorn] states that the “Hawthorne Works” complex employed 42,000 people in the 1920s. It housed the primary "Relay Assembly Department." For more insights, check out these interactive relay demos.
In Britain (1970) the GPO offered at least 550 different types based on their universal 3000 style relay [GPO]. So, we can only imagine that the worldwide total of varieties from many providers is likely at least 10,000 in 1970.
As an aside, it’s interesting to note that use of relays in North American exchanges grew from 8 (New Haven) to about 650 million by 1971. This is an annual compound growth rate (CAGR) of ~22% year or a doubling roughly every 3.5 years over 1878–1971. See Endnote B2. If relays hadn't been invented, telephone systems would have appeared much later, maybe with the invention of the vacuum tube circa 1907.
See a Western Electric AF22 Wire Spring relay in action below.
Note: Look for 2 "early-break-make" contacts and 1 "early-make-break" contact.
1.3 The Relay is king in an analog world
The relay was remarkably flexible to meet telephony’s wide design needs for ~90 years before digital methods took the stage. Some relay drawbacks are large size and slow response time compared to semiconductors. Nonetheless, they had no commercial competition (occasionally vacuum tubes), before ~1962 when digital logic chips were introduced by Fairchild Semiconductor. Of course, this progress was tied to the famous invention of the transistor at Bell Labs in 1948.
​
The Bell System's first first commercial use of transistors in an exchange was the #1 Electronic Switching System (#1 ESS, 10K lines) in 1965. It used 160,000 diodes and 55,000 discrete transistors in place of many relays (it still had 14,000 relays!) and other electro-mechanical devices [Ferguson].
​
Bell engineers begin replacing vacuum‑tube line repeaters with transistorized designs in 1952. Their T1 Carrier System also used discrete transistors in 1962. In 2026, telephony is purely digital and use of a relay is an exception.
​
Some key features of relays designed for telephony:
•Very low to moderate AC/DC currents, ~130 volts max
​
•Many contact form factors -- SPST, SPDT, DPDT, + ​​
​​
•Transfer types -- make, break, break-make, make-before-break, +
​​
​•Up to ~30 contacts with 100s of configurations
​​
​•Fast (3.3 ms) and slow (500 ms) acting (Bell System Wire Spring Relay Ref, 1953)​
​
​•Bidirectional for analog signals and without loss
Of course, no single relay needs to fulfill all these design needs, so various types were used as discussed perviously. Transistor designs struggle to meet even the basic requirements; for example, a make-before-break form-C switch requires over 12 components and was still not bidirectional.
​
That said, today general purpose relays (not specific for telephony) are a thriving market. See Endnote F for more on this.
​​​
1.4 The Relay as a logic element
Relays can be wired to be logical AND, OR, XOR or NEG operators so any conceivable logical function can be constructed. One of the first uses of relays doing math (Fig 7) in an exchange was the digit counter used in Panel offices in 1929. It used 18 relays of different types to count subscriber dial pulses.
George Stibitz (1904–1995), mathematician and researcher at Bell Labs, designed the first relay-based 1-bit adder in 1937. This proof-of-concept, built in his kitchen, started a “relay race” to construct more powerful relay-based computers and complex process control systems.
​
Stibitz’ invention extended what relays could do and new telephone exchange designs were the better for it. Figure 15 shows his original creation; 2 batteries, 2 'telephone' relays, 2 lamps, and 2 push button switches. Fig 16 is the schematic with a “truth table”. This was a so-called "half adder" since it lacked a carry in port. It's straightforward to convert this into a "full adder" with a carry-in bit using two or three relays per bit depending on the circuit design. So, an 8 bit adder could be built using only 16 relays, best case.
It’s surprising no one demonstrated this before 1937 given German mathematician Gottfried Leibniz’ [Leibniz] published discovery of binary math in 1703 and Englishman George Boole’s work, “An Investigation of the Laws of Thought” in 1854. Boole laid out the basics of what would be called Boolean algebra. In 1942 the Atanasoff–Berry Computer (ABC) was the first binary-based computer using vacuum tubes (~300) with 50-bit numbers.
Fig 15, George Stibitz’s 1-bit adder (Computer History Museum)
Fig 16, Schematic with two relays A and B [Kovalick]
Of course, 1-bit adders (with carry in and out) can be cascaded to make 8/16/32/64-bit adders and all sorts of math calculating circuits. One proof of concept is the Zuse Z3, 2,600 relays, 22-bit words, digital computer in 1941 [Zuse].
Claude Shannon, a Bell Labs mathematician, wrote his MIT Master’s Thesis paper in 1941 and it changed computer design forever- A Symbolic Analysis of Relay Switching Circuits. Some consider this to be the most important Thesis of all time [Fox]. The Bell System’s bible on relays and switching was [Keister], a book that this author as a teenager could not put down. See Fig 17 with 8 Line and 8 CO relays.
​
The Marker—of which there were several types—served as the brain of a Crossbar system. Containing up to 1,500 relays with hundreds of schematic pages [History2, pg 157], it orchestrates all aspects of call setup. Among tasks, it finds an available end-to-end path through the complex switching network and activates the cross-points. Figure 18, with 12 relays, illustrates a small portion of a Marker’s total schematic.
​
Before the integrated circuit, before the transistor, before the vacuum tube, relays really did change the world. And it all started with the twitch of a compass needle.
​
-- Al Kovalick
Fig 17, 40 relays in an 8-line DIY rotary telephone system (Kovalick)
Fig 18, Partial Marker relay schematic
ENDNOTES
Endnote A: Visit Calling315.com for an overview of 1D and 2D linear/rotational switches like Strowger, Panel, Rotary, and Crossbar that route calls. These switches, particularly Crossbar types, have relay-like features that aided their development and control. Sequence switches in Rotary and Panel exchanges also function similarly to relays.
Endnote B1: Baron Pavel L’vovitch Schilling (1786–1837) was a Russian-German diplomat, scientist, and inventor who likely created the earliest practical electric telegraph systems. This was long before Henry, Davy and Morse. Schilling's 1832 demonstration telegraph in St. Petersburg used eight wires: six for signaling, one wire for calling, and a common return. [Schilling].
Endnote B2: Research reveals that there were about 12.6 million telephone stations circa 1920 (AT&T Annual report 1920). Most subscribers then were on shared party lines. There were roughly 115 million stations circa 1971 in North America. This amount includes those in the Bell System (AT&T Annual report 1971), plus ~1,500 independent companies lines, plus private PBX/Centrex lines. Factoring that ~20% of public telephones were on party lines (sharing landlines) in 1971 there were about 93 million individual telephone lines stretching from exchanges to stations.
​
Source [Muller pg 229] states that, on average, there were ~7 relays per line (automatic: Step-by-Step and Crossbar) and ~2.5 relays per line (manual switchboard). There were only 11 manual Central Offices in 1970 out of 14,756 Central Office Codes according to [Pferd]. So, we can conclude there were about 650 (~93*7) million relays installed in exchanges circa 1971 just in North America.
(It started with 8 relays in 1878 at New Haven, Connecticut. It’s interesting that Connecticut ~ connect & cut, is precisely what a switchboard operator does.)
Endnote C: According to [Rhodes], Thomas Watson invented the hand cranked magneto in 1878, which the subscriber rotated rapidly to announce a new caller to the operator. The very first telephone sets did not have magnetos.
Unlike the first makeshift board, Coy worked with a “Mr. Snell” to build a second board using a much-improved design with 35 lines. It could support ~450 subscribers with many party lines. Snell was a local machinist in New Haven, Connecticut, who specialized in building equipment [de Land].
Endnote D: The Line relay has a close brother called the Cut-off (or CO) relay and both were included in most switchboards circa 1895 and all automatic exchanges. A 10K line step-by-step system in Chicago in 1903 [Smith] by the Automatic Electric Company (AEC) had one L and one CO per subscriber.
An energized CO relay indicates a busy line to a new caller. When the operator answers a call by inserting a plug into the jack, the Cut-off relay activates. It disconnects the Line (L) relay, which in turn causes the subscriber's line lamp to turn off. The so-called “sleeve conductor” (tip, ring, sleeve of a plug) is often a control lead for the CO relay.
As an aside, you may wonder why this author has a strong interest in Line relays. In a mature switchboard and automatic exchange there is an L and CO relay pair per subscriber. In a Number 5 Crossbar exchange there may be 60,000 relays for a 10K line office. So, ~33% of all the exchange’s relays are L and CO types. It's the L relay that starts off the dance to connect the two subscribers. This makes their design features (cost, size, power, reliability,…) paramount in any exchange.
(Incidentally, as a teenager I was invited to see my home line’s L/CO relay in the 552 Panel office on Otis Street in San Francisco. Frankly, I won't soon forget being allowed to tap “my L relay”, knowing its importance.)
Endnote E: William Gilbert (1544-1603), an English physician and natural philosopher is often called the father of magnetism. He coined the term electricus, now electricity.
In his landmark 1600 treatise De Magnete, Gilbert carefully distinguished between magnetism and static electricity, introduced experimental methods, and argued that the Earth itself behaves like a giant magnet.
His work provided the intellectual groundwork upon which Orsted, Sturgeon, Faraday, Henry and James Clerk Maxwell would help build the science of electromagnetism, ultimately leading to the technologies that power the modern world.
Endnote F: For reference, the market outlook for low power DC and AC relays (both solid state and electromechanical) is very healthy. Not for telephone exchanges but for industrial applications where relays have advantages over solid state.
​
The DC small power relay market size was valued at USD 1.2 Billion in 2024 and is forecasted to grow at a CAGR of 9.2% from 2026 to 2033, reaching USD 2.5 Billion by 2033. https://www.verifiedmarketreports.com/
​
The AC relay market size was valued at USD 4.19 billion in 2023 and is poised to USD 8.95 billion by 2032. https://www.skyquestt.com/
References
Brooks, John, "Telephone: The First Hundred Years", 1976
​
de Land, Fred, “Notes on the Development of Telephone Service IV”, Popular Science Monthly, Volume 70, March 1907
Fahie, J.J., “A history of electric telegraphy, to the year 1837”, published 1884, London.
Ferguson et al, “No. 1 ESS Apparatus and Equipment”, Bell System Technical Journal, Sept 1964.
Fox, Charles, “Computer Architecture: From the Stone Age to the Quantum Age”, San Francisco”, Pg 114, 2024
GPO, “3000 Type Relay Data Sheets”, March 1970, Britain (General Post Office)
Hawthorn Works: https://hawthorneworks.wordpress.com/2014/06/25/end-of-an-era/
Henry: https://edisontechcenter.org/JosephHenry.html
History, “A History of Engineering and Science in the Bell System, The Early Years (1875-1925)”, 1975, Bell Labs
History2: "A History of Engineering and Science in the Bell System Switching Technology (1925-1975)", 1982, Bell Labs.
Keister, Ritchie and Washburn; Members of Bell Laboratories, “The Design of Switching Circuits” ,1951
Kingsbury, J. E., “The Telephone and Telephone Exchanges”, 1915
Kovalick, Al, redrawn schematic after Kenioua
Kuhn, W, Critical Relays of the Telephone Systems, Bell Laboratories Record, October 1928
Leibniz, Gottfried, Explanation of Binary Arithmetic, 1703, translated from German, (Leibniz explicitly discusses binary addition), https://kastalia.medienhaus.udk-berlin.de/odl/Leibniz.pdf
McMeen, Samuel and Kempster Miller, “Telephony”, 1912
Mueller, "Relays in the Bell System: Facts and Figures," Bell Laboratories Record, June 1957
Orsted, J. C. (1820). "Experiments on the Effect of a Current of Electricity on the Magnetic Needle". Annals of Philosophy; or, Magazine of Chemistry, Mineralogy, Mechanics, Natural History, Agriculture, and the Arts. Vol. XVI. London: Baldwin, Cradock, and Joy. pp. 273–276
Palmer, R. W., “Relays in Automatic Telephony”, Sir Isaac Pitman and Sons, London, 1930
​
Pferd, W., “The Evolution and Special Features of Bell System Telephone Equipment Buildings”, BSTJ, Feb 1979
Porter, Harry, “The Design of a Relay-Based Computer”, 2007
Rhodes, Fredrick, “Beginnings of Telephony”, Harper and Brothers, 1929
Schilling: https://en.wikipedia.org/wiki/Schilling_telegraph
Silliman’s Journal of American Science, Volume 20, 1831, pages 340–348
Smith, Arthur Bessey, “The Early History of the Automatic Telephone”, circa 1907
Smith2: https://siarchives.si.edu/blog/forgotten-history-alfred-vail-and-samuel-morse
Sturgeon: https://www.britannica.com/biography/William-Sturgeon
Switch, “The Essential Electrical Switch” (short video tutorial)
Thomson, William, "Report on Bell’s Telephone," Journal and Proceedings of the Royal Society of New South Wales, Vol. XII, 1878, p. 4.
Vail, Alfred, “The American Electro Magnetic Telegraph”, Lea & Blanchard, 1845.