NOISE: An Exhausting Problem?

Competitors sometimes get caught by the noise regulations, often at the Castle Combe Sprint. Due to local sensitivity and planning restrictions, the limit applied at Combe is lower than at many other hillclimb and sprint venues – 105dBA at ¾ maximum revs at 0.5m, though it is the Blue Book norm for Saloon and Sports Car Races (Section J, Chart 5.18). This is an attempt to clarify some of the issues surrounding silencing and the characteristics of sound and its measurement.

The engine produces a series of explosive sounds (two per revolution in a four-cylinder engine), as each piston discharges the burnt gasses from the cylinder. Without a silencer this noise would be deafening. But the exhaust system, as well as reducing noise, can have a beneficial effect on the performance of an engine. In a properly designed manifold/collector system the outward flow of exhaust gas from one cylinder creates a suction effect which assists the flow of gas from the next cylinder in the sequence. This in turn helps to improve the flow of the incoming air/fuel mixture.

For this to work there has to be a relatively free flow of gas through the system as a whole, which means that the silencer must not introduce any obstruction. A silencer containing tubes and baffles is effective at reducing noise by sending the percussive waves through a sort of maze, but it impedes the flow of gas in the process. For maximum engine performance and efficiency, the ‘straight-through’ silencer is used, which consists of a single perforated tube running through the middle of a cylindrical box, surrounded by wadding. As the gas enters the silencer it expands through the perforations into the larger diameter of the silencer. It cools as it does so, and its rate of flow through the system slows considerably, thus reducing the noise impact. In addition, the pressure waves dissipate energy by internal friction within the absorbent wadding.

If you’ve ever blown over the neck of a bottle to create a sound, you will be familiar with the principle of the Helmholz resonator. The silencer box acts in a similar way, and absorbs sound energy at its tuned frequency. The size of the perforations and the dimensions of the box are critical in determining which frequencies are absorbed, and some silencers are subdivided into chambers of different sizes to absorb sound across a range of frequencies. For this reason it is important to use a silencer appropriate to the type of engine; for example, a silencer designed to be effective on an engine revving to 6000 rpm will not be much use on one revving to 12000 rpm. Clearly, if the wadding becomes compacted, choked with carbon, or burnt away it will lose efficiency.

Now, what of the dB? We probably all know that it stands for decibel, which is the standard unit for the measurement of relative sound pressure levels. Because the ear doesn’t respond in a linear way to sound levels, the decibel is a logarithmic unit. Sorry, this is going to get a bit technical now, but in mathematical terms, sound level change in dBs = 10 x log10 x the power ratio. For example, if one sound is twice the power of another, the log of that ratio (2) is 0.3010, and the difference in dBs is 3dB. It is a fact that 3dB is about the smallest change in sound level that the human ear can detect, so if you upgraded your 50 watt home stereo to a 100 watt system, you would only just be able to hear any difference in loudness! A difference in sound level that is perceived by the ear to be twice as load actually measures 10dB, a power ratio of x10.

When we talk of an absolute sound level of, say, 105dB, we are actually referring to a level of 105dB above the threshold of hearing, i.e. the quietest sound the ear can hear. In fact, the sensitivity of the ear is dependent on pitch. It is most sensitive to sounds around 1kHz (this is the frequency of the Greenwich Time Signal, the ‘pips’). It is much less sensitive to higher and lower frequencies. When measuring sound levels, it is important to take this into account, and there are three compensating standards, A, B and C. The most commonly used is the ‘A’ weighting standard, hence the specification of the Race Noise Limit as 105dBA.

Another important characteristic of sound is the way in which perceived sound level changes with distance from the source. It follows an Inverse Square Law, which means that for a doubling of the distance the intensity decreases by one quarter (=6dB). If you look at the table of alternative distance readings in the Blue Book Section E you will see this borne out: 2 metres is 4x the standard 0.5m, so the sound level is reduced by x16 = 12dB; 8m is 16x standard distance, the sound level is 256x less, which equates to a 24dB reduction.

The microphone in a Sound Level Meter is what’s known as omnidirectional, in other words, it responds equally to sounds coming from any direction. This means that it really doesn’t matter where it is actually pointing, it is the distance from the source that is important. However, the Noise Test requires the meter to be ‘at exhaust outlet level at an angle of 45° with the exhaust outlet’. This 45° angle could obviously be either direction in the horizontal plane, one of which might place the meter closer to the engine (especially in a rear-engine car), where it is going to pick up mechanical and induction noise, which could adversely affect the result. Other aspects of the installation can affect the noise reading: the proximity of steel or alloy panels can reflect or amplify noise, and can make their own contribution to the noise level by sympathetic vibration.

Allen Harris (2001)