Dimming the Din

Acoustics in the cockpit

June 1, 1995

As pilots, we tend to deal with aircraft noise by making an end-run around it. We embrace various noise-cutting paraphernalia — ear plugs and headsets — in an effort to reduce the fatiguing effects of a high-noise environment. Mostly, we look for ways of protecting ourselves from the din, rather than attempting to block it at the source.

And there's no denying it: Airplanes are noisy. Take a noise reading in the cockpit of a typical high-performance single at takeoff and you'll find the needle bouncing around the 100-decibel mark. For comparison, a noisy office records 60 dBA, and a subway station posts an 80-dBA figure. Pain sets in at 120 dBA.

About those measurements. Sound comes in all frequencies, from the thumping of a bass drum to the shrill of a factory whistle. We can separate sound-level readings by frequency range, usually stating the center point of the measurement for reference. Cycles per second, or Hertz, tell you the frequency range in use. A 60-Hz sound is quite low, like a bass guitar; 2,000 Hz (or, typically stated, 2 kHz) is at the high end of human speech; a 15-kHz tone resides at the top of human hearing range and will start to get the attention of neighborhood Labradors. Acoustic experts refer to the decibel scale with reference to a type of weighting. The so-called A weighting, the most common type of skewing used, averages the noise across the audible spectrum according to the human ear's sensitivity. Decibel measurements are also logarithmic: The difference between 60 dB and 70 dB is a tenfold increase. One decibel represents the smallest difference between sounds detect-able by the human ear.

Why is this so? As a rule airplanes are de-signed for the lightest possible weight, given a certain desired structural strength. Extras such as heavier skin, which would help deaden sound, are just so much excess weight to the designers. So, we have lightweight structures married to large-displacement engines turning low rpm and swinging a semi-balanced set of airfoils through the atmosphere. It's not exactly the recipe for serene smoothness and luxury-car silence. Too, the build quality of even the finest of light aircraft — with regard to how the doors fit the fuselage and the flushness of the windows — isn't much better than your typical Detroit taxicab.

As the fleet ages and owners seek ways to prolong the lives and utility of older airplanes, more emphasis has arrived at soundproofing's door. There are some new methods of taking the noise out of the cockpit without adding tremendously to weight and without creating some new under-the-upholstery mayhem. (Owners of airplanes whose insulation had trapped water against corrosion-prone skins or ready-to-rust steel structures know the potential for trouble all too well.)

It helps first to understand what it is that makes a light airplane so noisy. Chris Brunt Engineering, which performed testing for Aero Sound Shield, a soundproofing-kit maker, posts numbers for a Beech F33A. The firm has measured a great number of other aircraft, from basic trainers to jet transports, and helped Aero Sound Shield tailor soundproofing packages to their respective installations.

At takeoff power, the Beech in stock form posted an A-weighted sound-pressure level of 96 dB — that's noisy. Pinpointing the real noisemakers in the airplane proves simple when the noise levels are segregated by frequency. In the Beech, the highest levels occurred at 125 Hz, with a reading of 105 dB. Second worst in the spectrum was the 250-Hz post of 101 dB, followed by a 97-dB reading at 63 Hz. These are all quite low frequencies, well below even the deepest of human voices.

Brunt points out that the 125-Hz reading falls into the propeller's fundamental frequency, or the frequency at which parcels of air are thrown back by the prop. When any given panel is excited at its natural frequency, it begins to resonate, adding to the din. Different parts of the airframe resonate at different frequencies, Brunt says, but the worst offenders are the large panels in the fuselage sides, especially in airplanes whose bulkheads and longerons are far apart. Imagine that the larger the unsupported panel the greater its ability to transmit low-frequency noise, in the same way that a large woofer in a stereo system is capable of turning out deeper bass than a smaller speaker. Brunt says that many of the fuselage panels in light airplanes are by acoustical standards nearly transparent membranes.

So it's this relationship to the prop's fundamental frequency that's so important. Incidentally, you can find the fundamental frequency by multiplying the rpm by the number of blades and dividing the result by 60. A three-blade prop spinning 2,700 rpm, for instance, has a fundamental resonance at 135 Hz. Changing engine and prop rpm, then, can greatly influence the noise level, as nearly every one of us has noticed. Bring the prop's fundamental frequency below the resonance point of the fuselage and you'll cut the sound level dramatically.

In testing, the F33A's interior noise at 250 Hz fell from 95 dB at 2,500 engine rpm to 91 dB at 2,300 rpm, and further to 88 dB at 1,800 rpm. Down the scale at 125 Hz, the F33A be-came louder at 2,300 rpm versus 2,500 rpm, posting 103 dB and 100 dB, respectively; this shows the effects of moving the propeller fundamental frequency.

Engine vibration — from combustion events, internal imbalances, and exhaust pulses — joins the party in the lower frequency ranges as well. It's interesting to note that a four-cylinder engine with a two-blade prop and a six-cylinder engine mated to a three-blade prop place their natural vibration characteristics smack in the middle of the prop fundamental frequencies. From an acoustical point of view, it's almost as though the most common combinations are also the worst low-frequency noisemakers.

Noise readings at 500 Hz, Brunt says, are a combination of prop beat and irregular sound pulses from the engine conspiring to increase noise. At the lowest end of the scale, Brunt says, testing has shown that noise in the 63-Hz range comes mostly from structural members of the airplane.

Further up the audible scale, things improve in our tested Beech. For example, the 500-Hz center point posted an 88 dB reading; the 1 kHz point, 80 dB; and the 2 kHz point, a mere 72 dB. Beyond 2 kHz, Brunt says, the sound pressure levels fall into the low 60-dB range, and are primarily influenced by such things as wind noise and gyro whine. Even so, the levels in the upper frequency range are quite mild compared to those down low, and even lightweight headsets or ear plugs effectively mask this higher-pitched sound.

Noise enters the cabin not just from the metal bits, either. Windows transmit a significant portion of the racket, but limiting the intrusion is relatively easy. According to Brunt, testing in the Bonanza has shown that thicker windows (up to 1/2-inch) result in considerably less noise inside, and that the sloped windshield used in later-model Bonanzas is best of all. Further, Brunt notes, curved windows are generally quieter, for a given thickness and area, than are flat panes. And while window inserts are effective in cutting noise, they are not as successful as simply increasing the thickness of the window. That's because, to be most useful, the air gap between the panes needs to be 4 inches or more for the range of frequencies needing to be damped. Needless to say, few airplanes have room for 4-inch-thick side windows.

Okay, so the very act of producing thrust creates a level of noise akin to a heavy-metal concert. What do you do?

Aero Sound Shield's noise-reduction methods are quite clever. Company founder Olen Nelson says that he and Brunt spent considerable time testing various combinations of foam and fiberglass insulation before settling on a formula. In most cases, the soundproofing kit supplied by Aero Sound Shield consists of three layers of material — a fiberglass mat, a thick and quite dense foam pad, and another layer of glass, all wrapped in a sealed blanket of nonflammable aluminum-coated, fiber-reinforced mylar. (The aluminum coating, by the way, is too thin to be conductive, says Nelson.)

These blankets, which are precut for various locations in the fuselage, do two things. First, when held against the skin by the interior panels, the blanket has the effect of adding mass to the panel. Adding mass greatly decreases the natural resonance frequency of that panel; the objective is to get the resonance to a point that's so low it is either unexcitable by sound waves produced by the prop and engine or, failing that, so that it is unobjectionable to the human ear. Second, the fiberglass insulation acts to damp the sound waves that emanate from the panel; the next layer of thick foam tends to both absorb some of this residual noise and to pass some back to the outer layer of fiberglass. Finally, the inner layer of fiberglass acts to keep interior panels from acting like the exterior panels in the transmission of noise.

We saw several examples of Aero Sound Shield's kits, which are available for a wide variety of airplanes and are installed with supplemental type certificate (STC) approval. The kit for the short-body Beech Bonanza, for example, consists of more than 60 individual pieces. These can be either the aforementioned pillows or bits of the thick foam core where the entire assembly is too dense to fit. The kit appears complete, with bits of soundproofing for every part of the interior, including a pad to seal the area around the fuel selector box, inner firewall, and all access panels.

According to testing done by Brunt on the aforementioned F33A, application of the Sound Shield system resulted in some pretty impressive improvements. Overall noise on takeoff fell from 96 dBA to 89 dBA. On average, the treated airplane was 5 to 6 dBA quieter than stock, with typical improvements in the 125 Hz band of 10 to 12 dB. Aero Sound Shield estimates that the kit adds between 12 and 20 pounds to the airframe, depending upon model and the type of soundproofing previously installed.

We also had the opportunity to fly around in Nelson's kitted V35 Bonanza. We didn't wear headsets, and at the end of more than two hours' flying, a headache failed to materialize. Subjective impressions back up the test data by Brunt — the soundproofed airplane is significantly quieter than some much newer Bonanzas we've tried. (We nonetheless would not say that it's possible to fly four hours straight without headsets and not have slight discomfort.) But the point isn't necessarily to do away with headsets and the like. Rather, the noise reduction permits use of lightweight headsets or even simple ear plugs for passengers not wanting to don heavy headsets. What we sampled seems to be a genuine and cost-effective (around $1,000 plus about 20 hours' labor) method of plucking the aural pain out of the typical light-airplane cockpit. More involved, yes, but certainly better than a band-aid work-around to noise.

For more information, contact Aero Sound Shield, 2501 Airport Ave, Hangar 12, Santa Monica, California 90405; telephone 310/915-9101, fax 310/915-0131.