The Slow Extinction of a Brilliant Invention—the Analog Telephone
Here are some of the reasons a familiar technology has bucked the “speedy obsolescence” trend for more than 100 years.
In today’s fast-paced world of electronic gadgets, devices often become obsolete within a few years. Plain Old Telephone Service (POTS) and the analog telephone have somehow endured for over 100 years. It is instead facing a slow demise.
Even as analog telephones are being replaced by digital IP phones, the old terminology has managed to endure. Although it is rare to come across a rotary phone, we still “dial.” There are no bells in phones any more, but we still say that they are “ringing.” We even have “ring tones,” most of which bear no resemblance to a ringing bell. A phone may not have cradle in which to rest the handset, but we still “hang up.”
If we take a step back and look at the underlying technology behind the analog phone, we find a design that is brilliant in its simplicity compared with today’s digital replacements.
Analog phones are connected to the telephone company central office using nothing more than a pair of wires, as shown in Figure 1.
These two wires are used to convey not only full-duplex voice, but also many different types of call progress signals. Beyond that, the same two wires supply electricity to power the phone itself. And, in an environmentally conscious way and in contrast with IP phones and nearly every other electronic gadget in our homes, the analog phone draws no current when it is not in use! It requires no local always-on power supply. It operates without a microprocessor, without the Internet, and without complex Voice-Over-IP networking protocols. It even operates when the power to your home goes out.
Simplicity and Ingenuity
Just as the paleontologist learns much by studying dinosaurs, we can learn a lesson or two about simplicity and ingenuity by studying the analog telephone.
When a phone is in the idle state, the telephone company central office supplies the phone with a 48 volt DC “battery” signal. To be specific, and for the purpose of further discussion, the phone’s two wires that connect it to the central office are referred to as “tip” and “ring.” The origin of the terminology comes from the jacks at the end of the cables that were used by human operators (a long time ago) to connect phone calls. The jacks had a “tip” conductor and a “ring” conductor, as Figure 2 shows.
So more specifically, during the idle state, the central office applies -48 VDC to ring with respect to tip.
Let’s go through the process of originating (placing) a call and terminating (answering) a call. To place a call, a person takes the phone handset off the phone’s cradle or hook, an action also known as going off-hook. By doing so, the user has closed a (hook) switch that connects the tip and ring leads through a resistor causing the phone to draw current through the “loop.” The loop refers to the circuit that forms from central office to the phone via the ring wire, through the phone, and back to the central office via the tip wire. The central office detects the current flow and applies a dial tone signal to the circuit. The person then dials the phone. In the old days, a rotary dial was used. For each digit to be dialed, the user had to rotate the spring-loaded dial and then release it. Upon releasing the dial, the phone would open-circuit the connection between tip and ring and close it over and over until the dial returned to its original orientation. The number of “breaks” or open circuit conditions was equal to the digit dialed plus one, except for the zero digit which caused eleven breaks. In the old days, each break would reposition a mechanical switch that facilitated the connection to the dialed party.
At the other end of the link, the dialed party’s phone rings. To make the phone ring, the central office at the other end applies a 20 Hz 90 volt AC signal between tip and ring, causing the phone’s bell to ring. This AC ring signal rides on top of the -48 V battery signal. When the receiving party goes off-hook, the central office detects the DC current flow in tip and ring, halts the ring signal, and an end-to-end voice connection was then established. At the end of the conversation, the parties would hang up and the circuit was released.
But that’s just the tip (no pun intended) of the iceberg with respect to all the signals that are carried over a single pair of wires. One example is hook-flash, which is used for three way calling, call hold, or call waiting. A hook flash is accomplished when a person who is already in an active phone call hangs up for a brief amount of time (less than 1.2 seconds) and quickly goes back off hook. This causes the tip to ring circuit to be opened long enough to alert the central office of the user’s intent, but not so long that it declares an on-hook.
Have you ever wondered how an old style answering machine knew when the other party went on hook so it didn’t record long periods of silence at the end of the call before deciding that the calling party had hung up? This is accomplished by “battery reversal.” During the active phone call, the “talk battery” polarity is negative, where the ring DC voltage is negative with respect to tip. When the other party hangs up, your central office reverses this battery polarity to your line, allowing your answering machine to detect the battery reversal, stop recording, and hang up.
Do you remember what a “party line” was? Party lines were used to allow two houses or apartments to share a single pair of wires! The two parties could not talk on the phone at the same time, but they did benefit from a discount on their monthly bill from the phone company. The shared line saved the phone company the cost of stringing additional copper wire pairs and the cost of an additional line circuit at the central office. In order to make a party line work, the central office needed the ability to determine (for billing purposes) which of the two parties placed each phone call. And when there was an incoming call, there needed to be a way for each party to distinguish between phone calls intended for him/her and those intended for the other party being served by the party line. The later was accomplished by using “distinctive ringing.” The normal ring that we are accustomed to hearing is a repeated cadence of two seconds of ring followed by four seconds of silence. By introducing a different cadence, the two parties could distinguish between calls for them and calls for their neighbor.
Less obvious is how the central office was able to distinguish between the two different parties going off hook. Earlier we discussed the “loop-start” circuit in which the off-hook condition causes a circuit to be formed connecting tip and ring through a resistor, and the central office detecting the resulting current flow. But if the pair of wires is shared between to subscribers, how can the central office determine which party went off hook?
To that end, we introduce the “ground-start” circuit. The ground start phone indicates that it is off hook by connecting the ring lead to earth ground via a resistor. The central office can then can detect current in the ring lead but none in the tip lead, thereby allowing it to distinguish the ground-start party from the loop-start party where the central office will detect equal but opposite current flow in the tip and ring wires. (Yes, we have introduced a third wire—the ground wire. But this wire does not have to travel from the phone to the central office, which could be miles away.)
How, you may ask, did the phone company prevent one party from snooping on the other party’s phone call? They didn’t. The best answer is common courtesy. If one party picked up the phone when the other party was already talking, he or she would hear the neighbor’s conversation. That was in fact the only way to know (short of a geeky customer monitoring line voltage) that the circuit was busy. The only way to truly prevent this was to pay the full monthly fee for a non-shared circuit. That business model sounds curiously like the model we see today distinguishing software and services by terms like “basic” and “professional” where the customer pays more to be able to use higher-end features.
The Workhorse Pay Phone
Have I convinced you yet that the analog telephone was brilliant? To anybody who has worked with the Voice-Over-IP protocols, reminiscing about the analog telephone is like a breath of fresh air. But there is still more to tell. What about the nearly extinct pay phone? It too used a single pair of wires. It had to perform the same functions as a normal phone, but even more. How did the central office know that the user had deposited coins, how many coins had been deposited, and what were the denominations of the coins? How did the central office instruct the phone to collect coins or refund them? In the spirit of the simplicity in the design of the analog phone, I will spare you the text and instead list the signals in Table 1, including those already discussed.
All this is done with a single pair of wires. Brilliant!
Scott Kurtz is the founder and president of DSP Soundware, a company dedicated to improving voice quality through the use of digital signal processing. Scott has over thirty years of experience in the fields of telecommunications and digital signal processing in both the commercial and defense industries. Scott earned his bachelor’s degree in electrical engineering from Lehigh University and his master’s degree in electrical engineering from Drexel University.