Strowgers first switch | Tribute to Relays

Strowger’s First Switch

Highlights

Switch category: Linear and rotary, two degrees of movement

Inventor: Almon B. Strowger

Important dates: 1891, fundamental switch, US patent 447,918 granted. November 3, 1892, the first Strowger exchange opened for public service at La Porte, Ind. 1901, formed the Automatic Electric Company.

Legacy: Almon Strowger is considered the father of the first practical automatic exchange switch. With excellent engineering and business support, his companies and successors pioneered improved switches, the rotary dial and many innovations in exchange technology.

Phased development

Prolific writer and engineer Kempster Miller [Miller] states that the ideas advanced by Connolly and McTighe’s patent of 1879 were improved upon by Almon Strowger. Interestingly, while Strowger’s first invention required five wires from each subscriber to the central office (CO), Connolly only required two. So, Connolly was advanced in this regard. See the backgrounder on the first Strowger exchange including a bit of history on Strowger’s motivation to invent the switch. There are other claims on being first. See Appendix B.

Before describing the switch workings, let's review some aspects of its phased development.

Fig 1, Major development phases of Strowger’s switch

Strowger produced several switch versions during the development phase. Fig 1 outlines the stages. Indicator #1 identifies his first switch patent, the most famous. This version, with a cylinder of terminals, is outlined in detail later in this section.

Strowger’s second switch version is identified by #2 in the figure. This version (US 486,909) was considerably different, yet influenced by, his first patent. It has a circular “Switch-Table”, as he calls it, with 1,000 contact points. The contacts are arranged in 10 concentric circles with 100 points per circle. The wiper pin (14) has radial and linear movement and was attached to arm (12). With two degrees of freedom, the wiper pin could reach any one of 1,000 terminals embedded into the Switch-Table. All of Strowger's versions were "one wire" switches, with one wiper contact per switch. See Appendix A.

Interestingly, it was this version that became the basis for the switch eventually used in the live exchange at La Porte. Identifier #3 shows a replica of the deployed switch. It was based on the second patent but only had 100 contact terminals arranged in one circle. So, no linear wiper movement was needed, only radial. There are no extant detailed documents explaining this final switch other than what appears in his second patent.

Alexander Keith (#4, Fig 1) was strongly influenced by Strowger’s first patent and developed an improved version. Keith’s switch became remarkably successful.

An early Almon Strowger prototype is shown in Fig 2 (left). Realizing that the switch would call for precise construction, Strowger hired a jeweler and his helper from Wichita, Kansas, at $5.00 each per day to make the first model [Smith]. Fig 2 (right) is a larger view of the replica switch.

As mentioned above, Strowger’s exchange switch needs five wires per subscriber when two was preferred. He was granted a patent (US 492,850) to reduce the per-subscriber wire count, but the ideas were never implemented.

original strowger switch prototype and replica used in La Porte

Fig 2, Strowger switches -- jeweler's prototype (left) and production version (right) [Archive]

The discussion to follow focuses on the first patented switch version for several reasons. First, there are several available fine images for this case. Second, this version provided inspiration to A. Keith and others for improving the first switch design. Third, the version used at La Porte was not improved upon. It was the end of the line for its style.

Some salient points of the first patent are:

Each switch has only one radial wiper arm with a contact. This wiper can be directed to any of 999 terminals to forge a talking path to any other phone.
Support for 3 digits; vertical is hundreds digit and the horizontal is tens and units.
Each subscriber has a dedicated switch.
To make a call the subscriber needed to push up to 3 buttons, usually multiple times each, to advance the switch mechanism to the called number. On hang-up, push the restore button. The rotary dial had not yet been invented.
The power to move the wiper vertical and horizontal comes from the subscriber’s ~50v battery. This size of battery is not practical at a subscriber station and Strowger was aware of this.
Each phone has 6 wires leading to the Central Office (5 plus earth ground). Strowger knew this was a problem.
There was no motor needed (good) as there were in the Lorimer, Panel and other switches.

Switch details

Let’s delve into the workings of this device. Fig 3 is an annotated version of Fig 1 from the Strowger patent.

Fig 3, Telephone and switch, annotated Strowger patent figure

The switch has 4 electromagnets and the diagram calls these out. Each of these is controlled by switches at the subscriber telephone set. The cylinder has 1,000 stationary contacts in 10 rows with 100 contacts per row. Note the five wires per subscriber to the exchange. The movable wiper is in the center axis of the cylinder and can turn 360 degrees to reach any of 100 terminals on a given row.

Cleverly, there are two means to ratchet horizontally, along a row. See Fig 4 for a closeup of this feature. The smaller ”tens ratchet” moves 36 degrees for each pulse from the tens button at the station. The larger “units ratchet” moves 3.6 degrees for each pulse from the units button. So, any of the 100 terminals on a row can be reached with a combination of tens digit and units digit pulsing.

Fig 4, tens digit and units digit horizontal ratcheting means

To reach terminal 15 on row 3 (phone # 315), the subscriber pushes the buttons seen in Fig 3:

Pushes the hundreds digit button 3 times (hundreds V ratchet)
Pushes the tens digit button once (tens H ratchet)
Pushes the units digit button five times (units H ratchet)
On hang-up, pushes the reset button once

The subscriber can also reach horizontal “position 15” by pushing the units button 15 times.

Likely no one would do this, but it achieves the same result without using the tens button.

The Strowger switch installed at La Porte did not have the vertical lift feature. It was one row of 100 terminals. Strowger was pragmatic and saw no need to make a 1,000 point switch (very difficult) when a 100 point switch would prove his concept.

Integrating the switch into the exchange

As an aside, let’s briefly look at how the switch would have been used at La Porte (except with a max of 99 subscribers, 2 digits). As discussed above, the actual switch was based on Strowger's second patent. Nonetheless, the same principles apply with either switch version especially since five subscriber wires (plus a common ground) were used to control the switch. Fig 5 simplifies the subscriber-to-office wiring showing a ground and 1 lead.

Fig 5 shows the case where subscriber #288 is calling #315. This website’s URL (calling315.com) is based on the example given in Strowger’s patent. Sub #288 dials their switch to position #315 and the talking path is created. It may be difficult to see in the figure; the station #288 switch has advanced so that its wiper connects to the talking line for #315.

Fig 5, from the patent, switching system example – calling 315

Every station’s talking wire goes to the other 99 switches in a 99-station exchange. The amount of wiring just to provision the talking paths is ~10,000 connections and this increases as the square of the number of subscribers. This is another reason why Strowger’s first attempt, at a switch and an exchange, was proof-of-concept. However, this milestone screamed to the world -- "The automatic telephone exchange has arrived."

So, the headline “Undertaker changes the world”, is not far from the truth. Sure, Strowger’s first switch needed some tweaking to make it feasible for wider usage. In a few years engineers, inspired by Mr. Strowger, took his ideas to new levels that did indeed change the world of telecommunications forever. On the Switch page, select Keith’s Strowger to learn more about this.

Acknowledgments

Thanks to Len Hicken for providing valuable insights related to Strowger's second patent.

Thanks also to Savannah Jackson, Assistant Director, La Porte County (Indiana) Historical Society Museum, for providing pictures and documents related to the first automatic exchange.

References

Archive: La Porte County (Indiana) Historical Society Museum Archive.

Miller, Kempster, Telephone Theory and Practice, 1933, New York, McGraw Hill.

Smith, Arthur Bessey, The Early History of the Automatic Telephone, circa 1907

Thanks to Stanley Swihart, Telephone History Institute, compiler and editor of [Smith] 1995

close up of strowger switch showing tens and units gear for dialed number

from the strowger switch patent, switching system example – calling 315

Appendix A

Some details from Strowger's second patent: US486,909

The Switch Table 3 is stationary. Arm 12/11 moves radially and wiper 14 moves linearly as required to make contact 15 with one of 1,000 terminals embedded in the table. Member 10 is for support. Spring 25 returns the wiper to its home position after a call completes.

The terminals 5 are located on 10 concentric circles, each with 100 terminals. The single wiper 14 makes the talking connection over the "Tel" lead. Most early switches had only one wiper with a common system-wide ground as the second lead for the talking path. This means proved noisy and soon all switches had 3 leads; so-called tip, ring and sleeve for control (TRS).

Lever arm 17 creates the linear movement of wiper 14. The full patent details the mechanism that creates the arm and wiper movements. A simplified version of this design (100 terminals, no linear movement for wiper 14) became the production switch for the La Porte, Indiana, first exchange.

Appendix B
The Dane Sinclair Switch

A six subscriber switch was the first attempt at an automatic telephone exchange in Great Britain. It was invented by Dane Sinclair, an engineer at the National Telephone Company in Scotland, and was used in an exchange installed at Coatbridge, near Glasgow. This was in 1886, six years before the first automatic exchange was set up in the USA in 1892.

With only six subscribers, all others needed operator assistance. It functioned using electro-magnets and clockwork mechanisms. Of course, supporting only six subscribers marked it as experimental. It was not scalable as Strowger's method that supported 99 subs. Nonetheless, it was a significant milestone and a beautifully engineered device.