Feel The Noise
THE ART AND SCIENCE OF MAKING SOUND ALARMING
IN 1991 HARRY BARRY FOUND THE NOISEMAKER OF HIS DREAMS: a huge air-raid siren from the 195Os. It weighed more than .S1OOO pounds and was powered by a .33 I-cubic-inch V-8 Chrysler industrial gasoline engine. Barry, who has been collecting sirens, horns, and whistles since he was 15, had his latest treasure hauled from Detroit to his home in northeastern Pennsylvania. Might years later, in 1999, he invited Mric Larson, a professional pipeorgan restorer and fellow siren collector, to hear it in full song.
The two men towed the siren, which was mounted on a trailer, to a remote wooded area by a lake. Barry started the engine and let it warm up, and Larson noticed that the V-8 had no mufflers. In the Summer 2000 issue of Horn & Whistle , Larson told what happened next:
“When the engine had warmed up sufficiently, Harry then threw the clutch engaging the compressor and siren rotor plate. Although at this point the V-8 engine was probably only idling at 500 RPM, the siren began to produce sound, an enormously powerful and deep tone, not unlike the bass of a pipeorgan flue pipe but much louder and with more harmonic development. … Finally, Harry flipped a switch that activated a solenoid attached to the engine throttle.
“At this point, the engine began to pick up speed and reached top speed in about ten seconds. 1 stood at a respectful distance to the side and behind the huge siren and held my ears. As the machine approached rated speed, I could actually feel the pavement of the road vibrate.… At first I thought I was feeling mechanical vibration from the engine and the compressor, but the rubber tires of the trailer would have absorbed that. The vibration of the ground was strictly the result of the siren sound output. Finally Harry flipped the switch to the other position and the siren slowed down until again it reached idle speed.
“Then he invited me to try it. I stepped up on the trailer and went to the small control panel on the side of the engine compartment to flip the switch. Although I now wore Harry’s hearing protectors, it soon became evident that they weren’t sufficient and I slid a finger under each earmuff to hold my ears. At this point I was standing probably about seven feet behind the compressor-siren assembly. Suddenly everything else disppeared from existertee.… I looked briefly at my clothes and saw the fabric appeared blurry. My garments were vibrating at this same frequency! I even felt the sound in my beard and mustache.… My nose and the area just below my eyes began to ache slightly, and I noticed my eyes beginning to tear over.…
“I felt as though I could reach out and touch the sound, or lean on it or even sit on it, maybe even pick it up with a shovel or swim in it. It was that physically real, as though the sound itself had taken on a new and somewhat viscous, liquid form. Now, I had to figure out how to shut this thing down without removing my fingers from my ears. Finally, I reached up and hit the switch with my right elbow! Then I stepped down and back as the siren once more slowed to idle speed.” While every pastime has its own particular joys, very few collectors of stamps, coins, or shaving mugs can report such a visceral experience with their hobby.
Sirens are named, of course, after the characters in Greek mythology whose singing lured sailors to destruction on their island’s rocky shore. They first saw widespread use in lighthouses, as a means of signaling in foggy conditions. From this standpoint, one could hardly have chosen a worse name, since the original Sirens made sweet sounds in order to cause shipwrecks, whereas lighthouse sirens make harsh sounds in order to prevent them. The explanation for this anomaly is that the device originally called a siren had nothing to do with signaling. In fact, it was invented as a laboratory instrument.
For millennia, humans have built devices that produce sound by making something resonate, such as a string or a cavity. But sound can also be produced by generating the necessary pressure oscillations mechanically—for example, by sticking a card in the spokes of a spinning bicycle wheel. This principle came in handy when scientists studying what would later be called acoustics needed a way to measure and produce tones of specified frequencies.
The Scottish scientist John Robison invented such a device in the late 1700s. It was a pipe equipped with a stopcock that opened and closed 720 times per second. The first instrument to generate adjustable frequencies was invented in 1819 by Charles Cagniard de la Tour, a French scientist who gave the siren its name.
Cagniard’s version took the form of a brass cylinder. One end was sealed except for a tube that admitted compressed air or some other gas. At the other end were two disks with a very small gap between them. Each disk was pierced near its edge with a ring of holes, which were inclined relative to the axis of the cylinder, with one set slanting left and one slanting right. The inner disk was fixed, and the outer one was free to rotate.
As air passed through the fixed disk, the opposing directions of the holes forced the free disk to spin. As it spun, the holes in the two disks would periodically coincide, releasing tiny puffs of air. The rapid alternation between open and shut created sound. The faster the free disk spun, the higher the tone, with its frequency equaling the number of holes multiplied by the number of revolutions per second (which could be measured easily). Overtones at multiples of this fundamental frequency were also present. To measure the frequency of an external tone, the user simply cranked up the siren until the two were in unison.
The device also worked with a stream of water instead of air, even if it was fully submerged. Because of this “property of being sonorous in water,” Cagniard gave his invention the somewhat farfetched name of sirène . It relied solely on the stream of air to turn the disk, so the sound it made was soft at low frequencies and grating at high ones. Later models were modified to use an external source of power, such as a steam engine, to turn the disk, thus divorcing the pitch of the tone from its volume. This development raised the possibility of using a similar arrangement, greatly enlarged, as a signaling device—in lighthouses, for example. Since ancient times, seafaring nations had been looking for a way to assist sailors in foggy conditions, when lights could not be seen, and no completely satisfactory solution had yet emerged.
The earliest fog signal at an American lighthouse was a cannon installed at Boston Harbor in 1719. Cannon remained popular for nearly two centuries because they were simple and effective, though they could make excessive demands on the staff: In 1855 an unfortunate ex-Army sergeant detailed to Point Bonita lighthouse, just north of San Francisco, had to fire the cannon for three days and nights straight, with only two hours’ rest, when a heavy Golden Gate fog rolled in.
For various reasons, mostly having to do with cost, fog signals remained rare until the mid-nineteenth century, when increasing traffic and the greater speeds resulting from steam navigation made them necessary in more and more places. They had their problems, notably short range and the difficulty of determining how far away and in what direction they were. In heavy fogs it was not uncommon for a signal to be heard several miles to the north but not a quarter-mile to the south, or to be audible from three miles away but not from one or five.
A variety of sound-making devices were used in America over the years, including bells, whistles, cannon, drums, gongs, and a hornlike arrangement called the Daboll trumpet, which ran on compressed air and ranged up to 17 feet long, with a bell 38 inches in diameter. But sirens were the loudest—a quality that could be both good and bad, especially if people were sleeping nearby.
In 1900, when a siren was installed at the Pomham Rocks lighthouse on the Rhode Island coast, a newspaper headline declared it THE GREATEST NUISANCE IN THE HISTORY OF THE STATE . It was soon replaced with a bell. When a siren was installed at Great Captain Island, on the Connecticut shore near Greenwich, a reporter described the typical local resident as “a shrinking looking man or woman, whose hands clasp and unclasp nervously, who shivers every half minute as if from a blow, whose lips twitch and eyes roll, while the hair shows a tendency to uplift at regular intervals.” The siren was adjusted to reduce the noise level onshore.
Even when no one lived nearby except the keeper, sirens could create havoc. An 1890 visitor to Wisconsin’s isolated Pilot Island lighthouse wrote that when the siren was in use, “all the lights in the signal house must be hung by strings to prevent them from going out. The sound is so intense that no chicken can be hatched on the island, as the vibration kills them in the egg, and it causes milk to curdle in a few minutes.” At Point Reyes, California, vibrations loosened rocks from the cliff on which the lighthouse was situated. The signal had to be toned down and the cliffs reinforced to prevent a landslide.
The first steam-driven fog siren installed in an American lighthouse was an experimental model at Sandy Hook, New Jersey, in 1867. Five years later one of the proprietors of the Brown Brothers manufacturing firm, in New York City, patented a design that would be the pattern for all later sirens. The basic configuration is essentially a cylindrical version of Cagniard’s original laboratory instrument. The siren consists of two slotted or perforated cylinders, with one, called the stator, standing still, while the other, the rotor, rotates past it in close proximity. Compressed air or steam is forced through the slots of the spinning rotor, and the resulting rapid puffs are “chopped” by the stator. A horn at the mouth of the cylinders directs the sound in the desired direction. The faster the rotor spins, the higher the pitch, which is why the frequency of the siren’s wail rises during start-up and falls after the engine is shut off.
The simplest sirens were the hand-operated kind so often seen in old war movies and newsreels. These were barrelshaped and took power from a crank, which was geared to the rotor and to an internal blower fan. Even today many fire trucks still use mechanical sirens, although the blower and rotor are driven by an electric motor. The pitch is directly proportional to the speed of the rotor: An 11-slot rotor turning at 40 revolutions per second will give the siren a pitch of 440 cycles per second, the familiar A-440.
Most modern “sirens” on emergency vehicles, however, are not mechanical and thus are not true sirens. Today when a police car or an ambulance flashes past, you are probably hearing an electronic, digitized replica of a siren—what amounts to a very loud car radio. You can tell an electronic siren by the way it stops dead when it is switched off.
The greatest advance in noisemaking after the siren was the diaphone, which in its earliest form was patented in 1895 by the British (and later American) inventor Robert Hope-Jones. Like the original siren, it was not intended as a warning device. Hope-Jones had a passion for pipe organs, and he developed the diaphone as a new type of tone generator that would be particularly good at creating very low, very pure bass notes. Hope-Jones made extensive use of diaphones when he designed the original Wurlitzer organ.
Instead of having two cylinders, as in a siren, the diaphone is a slotted vertical cylinder with a free slotted piston inside it. The piston resembles a tin can open at one end. As it moves up and down when ;air is blown upon it, the piston resembles a tin can open at one end. As it moves up and down when air is blown upon it, the piston alternately blocks and opens the slots of the cylinder, and the cylinder, and the escaping air produces a sustained, unwavering note.
Most of today’s foghorns are the diaphones. They follow a modification patented in 1903 by John P. Northey, of Toronto. Northey enlarged the diaphone and introduced a key innovation by powering the piston’s stroke in both directions instead of just one. Northey also adjusted the timing of the valves so that the last few puffs of air or steam, after the motor had been turned off, would pass through the diaphone when the piston was moving more slowly. This created the brief final “grunt,” or low tone, that was characteristic of early diaphones. In 1929 Northey’s son Rodney designed a model in which the low tone, instead of being short, lasted as long as the high tone. This became the classic foghorn sound familiar to most people. The reason for a two-tone signal was that in certain atmospheric conditions, low tones carry better than high ones, so a two-tone signal had more chance of being heard.
Early sirens and diaphones in lighthouses were typically powered by steam engines. These generated a very loud sound, but they also could take as much as an hour to get up steam, and they required the keeper to maintain a supply of water. As the twentieth century wore on, most steam engines for fog signals were converted to internal combustion. By mid-century, radio was replacing audible fog signals, and today, with Global Positioning System monitors being standard shipboard equipment, the old fog siren is virtually extinct.
Sirens live on, though, as a means of alerting the general public in times of danger. In many suburban and rural communities, a siren summons members of the volunteer fire department. Civil defense is another important application. The type of siren that Harry Barry owns is a civil defense siren that originated in experiments conducted in early 1942 by Bell Laboratories.
The U.S. military had asked Bell scientists to develop an extremely loud noisemaker that could deafen or at least frighten enemy troops (it would be controlled remotely, of course). Bell’s prototype, nicknamed Big Bertha, used a 95-horsepower Ford flathead V-8 to drive an air compressor furnished by the American Locomotive Company (and originally built for railroad air brakes). A separate 20-horsepower Wisconsin Engine Company four-cylinder engine powered the rotor. The secret to Big Bertha’s extremely loud wail lay in its rotor and housing, which were designed to open and close the slots very fast and hold them wide open for relatively long periods. Bell Labs had discovered that much of a siren’s loudness results from overtones, or multiples of the main frequency; for example, if the main tone is 440 Hz, overtones will occur at 880, 1320, 1760, and so forth. More overtones were found to develop with a rapidly opening and closing chopper.
The Chrysler Corporation was also developing an air-raid siren early in World War II, and its researchers met with Bell engineers in March 1942. This led to production of the ChryslerBeIl Victory Siren, which was powered by an in-line, eightcylinder, 140-horsepower Chrysler automobile engine. The twostage blower produced five pounds per square inch (psi) of pressure, for a sound output of 137 decibels (dB) at a distance of 100 feet, with a frequency of 430 Hz. To give an idea of how loud 137 dB is, a jet aircraft engine at full throttle produces about 130 dB at 100 feet. A shotgun blast generates roughly 140 dB at the muzzle (every 10 decibels represents a doubling of loudness as perceived by the average listener).
Towns and cities across the United States and Canada ordered Victory sirens to warn citizens of air raids, tornadoes, and other disasters. Despite a 1942 price of $3,760 per siren (about $41,000 today), Detroit and Chicago bought 20 apiece. New York and Cincinnati purchased 10 apiece. In some instances the Chrysler-Bell Victory Sirens served their buyers into the early Cold War. None are known to survive today.
From 1952 through 1957 Chrysler produced an improved version of its World War II siren driven by a 180-horsepower hemi V-8. This was an industrial version of Chrysler’s famous automobile engine, and it powered the siren through an automotive fluid coupling. The hemi V-8 drove a three-stage centrifugal compressor at 4,500 rpm. The compressor housing was about 42 inches across and 18 inches deep. The entire output of the compressor section then entered a circular chamber to which six cast-aluminum horns were attached.
The horn throats, set at 60-degree intervals, formed a ring around the front of the siren. Just inside this circle was a round, half-ton, steel rotor plate with six ports. As this plate rotated, also at 4,500 rpm, the ports alternately opened and closed the throats of the six horns, creating a high-amplitude sound wave. The pressure inside the Chrysler siren was less than seven psi, but the compressor managed to move a huge quantity of air: 2,610 cubic feet per minute. Since the Chrysler siren’s blast was highly directional, many cities mounted their sirens on rotating turntables.
In cold climates, engine-block heaters and battery warmers helped keep sirens on full alert in winter. Since inactivity tended to make the system unreliable, sirens were usually tested for 15 minutes at least once a month. Sirens could be started and shut down remotely, by either dedicated telephone lines or radio signals. The early Victory sirens were also equipped with a seat and a control panel so that an operator could start and stop the device manually.
New warning methods have taken precedence in many situations, but sirens still have their uses. In Hawaii, for example, large sirens warn of tsunamis (tidal waves) and tropical storms. Fire departments stick to their mechanical sirens because they are louder than the electronic variety. With today’s motorists sealed into their cars with stereos blaring or cell phones in their ears, firefighters want to be sure they will be heard. While no one wants air-raid and civil defense signals to be as familiar as they were during the Cold War, it seems unlikely that the sound of a siren’s lusty wail will ever vanish completely.