Sound, Sound Levels And Sound Meters

SOUND, SOUND LEVELS AND SOUND METERS

IN MOST CASES, THE MEASUREMENT OF SOUND IS SOMEWHAT UNSOUND

DAN HUNT

WHO WOULD HAVE ever thought, 10 years ago, that you'd be reading about a sound meter in a motorcycle magazine today? Times do change, and attitudes, too, and now silence is in style.

Bike clubs are faced with the prospect of enforcing new sound limits applying to racing. Indoor racing promoters are worried about their customer's ears. Manufacturers and distributors want to know whether their silencing systems will pass muster with the state.

So they need an instrument to measure sound. The kind used by such organizations as the California Highway Patrol is cumbersome and expensive—well beyond the means of small organizations.

A typical example of a meter for non-rigorous applications is the portable, 2-lb. device made by T & T Products Inc. They sell the device for $98.50 and also market it through the Ameri can Motorcycle Association, Box 231, Worthington, OH 43085.

It is handy. You point its internal microphone at whatever you want to measure and it gives you a rough idea of how loud it is. Notice I said “rough." Its claimed accuracy is plus or minus 1.5 decibel (db). This is not exactly laboratory standard, when you consider that a difference of only 5db in the 90db range constitutes an increase in sound energy of about 60 percent. But it doesn't really matter. Even the CHP has trouble getting decent readings from their fancy meter. After they spend hours setting up runway distances and throttle settings to eliminate errors from test sample to test sample, they have to contend with environmental factors.

First, distance is a major factor in measuring sound. You can get the same db reading from the average conversation at 3.25 feet as you get from the average automobile at 15 feet. Obviously, the automobile makes more noise, but the distance has equalized your subjective impression of the noise and fooled the db meter as well. Sound sources, including motorbikes, are directional in nature and can fool the measurer. They may be louder from one side than the other, etc.

Secondly, you have constant variation of the medium through which sound travels. Depending upon its condition, air may carry varying amounts of energy. Changes in barometric pressure, altitude, wind direction, air temperature, and the chemical composition of the air itself (high carbon monoxide or carbon dioxide content dampens sound) will affect your perception of a given sound and change the db reading at any given distance. This is the reason that mundane noises sound intriguingly different when you go camping in the mountains or when you are surrounded by fog.

The third set of factors affecting measurement of sound energy have to do with the spatial and textural characteristics of the area in which the sound is measured. Are there any walls nearby to increase the reading, or any porous areas (like woods) to decrease the reading? Does the natural resonance of the area just happen to coincide with the frequency emitted by the sound source, thus causing a disproportionate rise in sound in comparison to the actual sound energy emitted?

Variation of road surface, for example, can introduce striking errors in sound measurement. The difference in reflectivity of smooth and rough finished concrete can introduce more than 1 db of error.

What we are saying, in effect, is that truly scientific measurement of emitted sound level is almost impossible outside the laboratory for the simple reason that sound is too hard to control.

Since the objects we desire to measure (motorcycles, vans, trucks, etc.) are hard to stuff through most laboratory doors, we compromise by yielding to the sin of approximation. And considering the hassle and money saved, approximation isn't that sinful.

Basically what we are after is the range of sound level into which a motorcycle falls. It is not important that we know that a racing bike emits exactly 92.367db at 50 ft. It is more important that we know that the bike does not emit sound in the 100db, 110db or 120db brackets. At these levels, the sound becomes irritating, and the neighbors start complaining.

The TTP sound meter fills these requirements, as it is most accurate in the 70 to 100db range. It is this sound range where most of the arguable decisions will be made. Extreme pain to the ears, beginning at 100db or so, is not arguable. But the difference between 88 and 92db is arguable-until it is settled by a sound meter.

Tests with a calibrated audio generator indicate that the TTP can settle any argument of relative sound to within plus or minus 0.5db in the 70 to 100db range. Above or below that range the relative scale error at center scale increases to plus or minus 1db. (These measurements were taken with a 1000 Hz sine wave.) At 4000 Hz the error is somewhat greater.

As with any low cost amplification system, the frequency response of the TTP is not completely flat and tends to be more sensitive to higher frequencies than to low.

Best accuracy is available in the 0 to minus-10 range at center scale. If you desire to measure a sound in the 80 to 90db range, you turn the range selector to 90. If the sound was 88db, for example, the needle would point to minus-2. Subtract two from the selected range, 90, and you get 88. The four-decible "plus" range of the needle indicator should be used merely to tell you to select a higher range with the selector knob, as it is not very linear. There are 10 ranges available from 50 to 140db, allowing measurement of sounds from 40 to 144 db.

Two nice features: a battery check button jumps the needle into the "plus" section when the two 9V transistor radio batteries are in proper condition; a two-range filter provides two rates of indicator needle damping to either match fast peaks of sound or average these peaks over a few seconds.

While the metal casing of the TTP makes a functional, tough package for the electronics inside, it does require the user to take greater care while taking measurements. Remember, the microphone used to pick up the sound is located inside the case. Bang on the case or rustle its resonant surface with your hand and you may affect the sound reading at moderate levels.

Further, you must take care that the simple metal carry handle is not rattling or loose. It's a useless thing anyway and should be taped down or removed.

In an outdoor test, the height you hold the meter above the ground will affect the reading you get. It should be held at least three feet above the ground to minimize error from sound reflection.

The California Highway Patrol Engineering Section uses 50 feet from line of travel as its standard distance for measuring decibels. We are reproducing here a typical CHP report sheet

from their 1969 new model test series so you may see some of the considerations involved in a noise test.

Because of the many variables involved, it would be impossible for you to get exactly the same results per given test. But, if you are a sound scrutineer at a club event, it is only proper that your method be similar from week to week to be fair to your entrants.

If you tested in a clump of trees one week and passed a man's bike, and then tested in an open area the following week and excluded the same machine from competition for being 1 or 2db over the limit, your method would be highly suspect.

Used properly, with consistency from measurement to measurement, the portable sound meter is a valuable tool for the bike club or racing promoter.

It won't give a legally valid, absolute measure of sound energy by itself, but it will tell you whether a given rider and machine are in the ballpark...and should be racing there.