Going Out on a Limb

By Dr Eugene Patronis

The digital revolution began in earnest in the decade of the 1980s with the advent of personal computers in the market place. Since that time the lead-time between scientific innovation and technological application has been getting shorter and shorter with an ever-hastening pace until this very day. This means that scientists, engineers, and technicians must always be undergoing a continuing educational process in order to keep abreast of developments in their own and related fields of endeavor. I was recently reminded of this when Pat Brown asked me to comment on some work that I had done for the Human Engineering Laboratory at Aberdeen Proving Grounds that had its beginning in the late 1980s.

This was an instance when responding to a request for a proposal required going out on a limb to a certain extent. That is to say that the technology necessary to satisfy the requirements of the future contract were in the development phase but not quite ready for market. The contract had many facets of which there were two biggies. The first requirement involved recording with accuracy the sounds made in a 100-meter by 100-meter target area while undergoing live-fire bombardment by US Army 155 mm or USSR 200 mm howitzers. These recordings were to be made at the artillery range located at Fort Sill Oklahoma. The second requirement was to design and supervise the construction of a facility for realistically playing back these sounds to a small audience of test subjects while the test subjects were attempting to track and fire at targets projected on a large video screen. The object was to determine which subjects could or could not perform their designated tasks while being subjected to startle effects that were definitely audible to the ear but also, more importantly, produced nauseating thumps to the testee’s chest cavity! This facility was to be located at the site of the Human Engineering Laboratory in Aberdeen, Maryland. I had been following Sony’s work on rotary digital tape recorders or Rdats and knew that they would soon be available in Japan. I was also current on loudspeaker developments in the US and felt that after the lengthy process of submitting a proposal, going through a selection procedure, and being awarded a contract the chances were good that the technology required would be available. So, out on a limb I went and responded to the request for a proposal.

My proposal was accepted and after the passage of some time the contract was awarded. Surely enough in the meanwhile the Sony Rdat recorders became available even though the first few units had to be imported from Japan and modified to operate on 120 Volts AC as the domestic standard in Japan was for 100 Volts AC. These were 16 bit stereo recorders capable of flat response from 0 to 20 kHz.

The target area at Fort Sill was fitted with a large number of 1/2 “ instrumentation grade pressure sensitive condenser microphones arranged in a grid pattern. These microphones were operated with reduced polarization voltages in order to extend their linear behavior to higher values of absolute pressure. Each microphone had its own battery operated power supply as well as preamplifier along with a line driver for the long buried cable runs back to the recorders that were located in a remote protected area. The bad news is there were casualties among the microphone population! The good news is the fallen heroes recorded faithfully until the penultimate moment!! Three different types of artillery explosions were recorded consisting of airburst, point detonate, and delayed detonate for rounds that were allowed to first bury themselves in the ground. The acoustic spectral density function, a graph of acoustic energy per unit frequency interval versus frequency, for each of these types was slightly different with the latter two displaying more energy at lower frequencies. The common features were a function maximum in the range of 30 to 40 Hz and that nearly 90 % of the blast acoustic energy appeared below 300 Hz.

The spectral density information gleaned from the on site recordings served as the guideline for the design of the audio system in the reproduction facility. The audio system in brief consisted of identical left, center, and right forward channels plus surround rear channels. The total spectrum for the forward channels was divided into five parts consisting of a sub-bass, bass, mid-bass, mid-frequency, and high frequency sections. Practically all of the muscle had to be in the sub-bass and bass regions so that will be the focus for the remainder of the description.

The sub-bass loudspeakers in each front channel consisted of four servo driven Contrabass loudspeakers designed by Tom Danley. Each of these units had a complement of two 15 inch driven loudspeakers and two 18-inch passive radiators that were operated in the range of 15 Hz to 40 Hz. These four units were closely coupled atop a thick concrete floor so as to act as a single coherent unit operating in a half space configuration in this frequency range. The bass array for each front channel requires an extended description, as it was somewhat unique. The basic building unit for this array was an Electro-Voice model MTL-4 loudspeaker that consisted of a manifold arrangement of four 18 inch woofers. Those readers who are not familiar with this unit can download an engineering data sheet by going to www.electrovoice.com and searching through the EV archive of discontinued products. For lack of a drawing I will attempt to give a word description of the basic structure. This manifold basically consists of a combination of four inside out vented bass enclosures. Imagine a standard un-vented bass enclosure where the driven element is mounted such that the loudspeaker frame and magnet assembly are mounted outside of the enclosure at the center of the panel with the cone facing inward toward the enclosed volume. Now take four such identical units and arrange them so that one driven element is located at 12 o’clock, one at 3 o’clock, one at 6 o’clock, and one at 9 o’clock such that the magnet and frame structures are as close together as possible on the interior of the arrangement. Bolt this array to a common rigid panel so as to form a five-sided box with a common open area containing the loudspeaker frames and magnet structures. The final step is to provide a properly sized vent at each of the four corners on the unenclosed side of the structure. Now here comes the FM. Those who know me well know what FM stands for in this context. Those who do not must consult someone who does. The question at this point is how to array such loudspeaker units in a half space so as to cause them to couple as a single coherent unit over most of their frequency range of 40 Hz to 160 Hz knowing that the dimensions of each unit are 36x36x30 inches? The answer that I came up with was to manifold the manifolds. One unit is placed on the concrete floor facing the test subjects. The second unit is placed in front of but just to the left of the opening of the initial unit with this second unit facing to the right. The third unit is positioned so as to be a mirror image of the second unit. Finally, the fourth unit is elevated above units two and three, centered on and rests on them, and faces down. You may well ask what is accomplished by all of this? A single MTL-4 operated into a half space will produce a sound pressure level of 140 dB at one meter. The process of manifolding the manifolds of four such units adds an additional 12 dB to this value because of coherent addition of sound pressures at low frequencies. The power addition of three such channels operating simultaneously contributes an additional 4.8 dB in the low frequency range. The actual measured results turned out to be even better than expected. When the overall system was fully powered and reproducing explosions the test subjects could be subjected to bass thumps ranging from 150 to 160 dB. Just a little time exposure to that goes a long, long way.