Horn & Whistle - Issue 68

This article originally appeared in Horn & Whistle Magazine, Issue #68, Summer 1995.
Reprinted here courtesy of Horn & Whistle Magazine.

A HISTORY OF THE CHRYSLER SIREN

by Harry Barry

The Chrysler Air Raid Sirens were the most powerful warning devices ever made, producing a sound pressure level of 138 decibels (dB) at a distance of 100 feet. No other sirens, whistles, or horns made before or since are so powerful. They were powered by internal combustion engines instead of electric motors or steam, making them mobile and self-contained.


Fig. 1. The original Bell Labs experimental high-output siren, nicknamed "Big Bertha."

Development of the Chrysler siren began around January, 1942, when the E. D. Bullard Co. of San Francisco submitted an engine-driven centrifugal siren to the Chrysler Division to see whether the straight 6-cylinder Chrysler Royal engine could be adapted to it. The output of this siren was inadequate, and it was concluded that improvements would have to be made for it to be useful as an air raid warning signal. Earlier it had been determined that a sound pressure level of 120 to 140 dB at 100 feet was needed for an effective warning. Despite considerable effort, the Bullard-type siren did not attain the necessary output. The problem was still under consideration when the Chrysler Corporation was invited to send representatives to a meeting called by the Office of Civil Defense (OCD) in Washington, DC on February 25, 1942, to inspect a new type of high output siren. This device was the experimental result of new sound-producing principles developed by the Bell Telephone Laboratories under the direction of Dr. Harvey Fletcher. The Bell Labs siren (shown in Fig. 1) was brilliantly engineered and very powerful, producing a level of 137 dB at 100 feet. The following principles. quoted from Bell Labs' technical report, governed the design:

(1). The source of pressurized air should be connected directly to a horn or orifice by an opening whose area may be varied periodically with time.

(2). The period during which the port is being opened or closed should be short compared with the period during which the port remains open or closed. The total efficiency of a siren whose ports are either wide open or completely closed, and in which the opening and closing are completed in zero time, is just twice that of a siren whose port area is varied sinusoidally. This is typical of the relation between a square wave and a sine wave.

(3). The 'Q' of the horn should be large compared with unity. For an exponential horn, this means that the fundamental frequency of the siren should lie well above the low-frequency cutoff of the horn.

(4). The path of air through the siren should he as direct as possible to avoid turbulence, and throttling of the air through the port should be avoided.

In the Bell Labs design, the clearance between the rotor and the port faces is only 0.010 to 0.015 inches. If smaller, the chopper might rub against the ports, and if larger, the sound output drops to an unacceptably low level. The experimental Bell Labs siren used a compressor furnished by the American Locomotive Co. This blower was driven by a 95 HP Ford engine and delivered air at 5 PSIG. Tests showed that an additional 20 HP would be required to drive the chopper rotor at 5,000 RPM, and this was provided by a separate 20 HP Wisconsin engine. This siren produced a frequency of 440 Hz when the rotor was driven at 4,400 RPM. A great deal of ingenuity, experimentation, and thought went into the development of the high output Bell Labs siren. Regrettably, this one-of-a-kind prototype seems to have been scrapped in later years. It would have been a nice historical piece for a museum or collector.


Fig. 2. Chrysler's first production siren, the 'CHRYSLER-BELL VICTORY SIREN'.
Note New York City mayor Fiorello LaGuardia in the operator's seat.

Three different production air raid sirens were produced by Chrysler, based on the Bell Labs design. The first version was called the 'Chrysler-Bell Victory Siren'. Its development was begun in a meeting on March 7, 1942 between representatives from Bell Labs and Chrysler. Fig. 2 shows this siren. The first production Victory siren was completed and officially tested on June 4, 1942. These sirens sold for $3,760, a substantial amount in 1942. 120 of these Victory sirens were sold that year to 28 cities and locations in the U.S.A. and Canada. New York City got ten, Detroit twenty, Chicago twenty, and Cincinnati ten, to name a few. The Victory siren was powered by a Chrysler straight eight-cylinder gasoline engine of 324 cu. in. and 140 HP. The blower was a two-stage centrifugal unit producing 5 PSI for a sound output of 137 dB at a frequency of 430 Hz and a distance of 100 feet. I know of no Victory sirens surviving today.


Fig. 3. The first version of the Chrysler air raid siren.  Note the seat for the operator.
The control panel is on the engine shroud, facing the seat.

Fig. 3 shows the first Chrysler Air Raid Siren to carry that name. This version appears to be neater and more refined than the Victory siren. It had the same sound output, the engine was the same straight-eight, 324 cu. in. 140 hp gasoline power plant, and the blower was the same two-stage unit. These sirens were built some time after the Victory siren but before 1952. So far as I know, none of these survive today either.


Fig. 4. The final version of the Chrysler air raid siren. 
This was the three-stage V-8 powered model, built in quantity.

Fig. 4 shows the final version of the Chrysler Air Raid Siren, introduced in 1952. This is the one that most of us are familiar with. I believe that several hundred of these were built between 1952 and 1957. They were the official Civil Defense sirens used in large U.S. cities and other locations during the Cold War nuclear attack threat era of the 1950's and 1960's. They had new three-stage centrifugal blowers to compress the air to 6.95 PSI before it was chopped by the rotor. These sirens were also fitted with Chrysler's new industrial 331 cu. in. hemi-head 180 HP V-8 gasoline engine. This siren produced an incredible 138 dB at 100 feet, and the frequency was 460 Hz at 4.600 RPM. It could be remotely controlled from inside a building or other location.

All four models had six horn-loaded ports. The three production model Chryslers had engine block heaters and heated battery cases, so that the sirens were ready to run down to -25°F in all weather and climates. Because the sound output was highly directional, they rotated at 1.5 RPM on a belt-driven turntable for full 360° coverage. The Chrysler Bell Victory Siren and its successors had metal seats for the operators, as shown. The operator rode the siren around rather like a merry-go-round.

I know of eight surviving sirens of the final three-stage design. Horn & Whistle members account for four of them: I have two, Jim Pritchard has one, and Jim Carruthers has one. As I was growing up in Pittsburgh and attending school in the 1950's and 1960's, a three-stage Chrysler siren was tested for at least fifteen minutes every week. I remember its authoritative sound very well. That siren is now in my possession.

The 'Mystery Siren' described on page 7 of Horn & whistle issue No. 42 is an Army unit, perhaps one of a kind. The siren portion is a two-stage Chrysler and the engine is a Waukesha. Surely this combination was concocted by the U.S. Army. I have learned a great deal about Chrysler sirens since this article appeared in 1990, but I have never seen any documentation on this siren. It seems certain, however, that Chrysler did not install a Waukesha engine in place of one of their own.

For more reading on Chrysler sirens, see Horn & Whistle issues Nos. 10, 28, 37, and 49. I have acquired much additional material from the National Archives, the Henry Ford Museum, owner's manuals, and other sources, and will be pleased to go into much greater detail with those who are interested.

 

[ HOME ]