We discovered much about noise and testing for noise.

We hired high quality test equipment that meets the necessary standards and guidelines as laid down in the appropriate test methods as required under legislation. We utilised sound engineers with NATA accreditation to check that we were in accordance with the methodology and train our test people. Our test operators were experienced laboratory personnel, accustomed to handling test equipment. We had a crew of 4 testing personnel, 3 data collection personnel and many marshals to ensure orderly conduct. We tested 124 motorcycles over 6 hours.

Performing the actual test presents a number of difficulties.

If the motorcycle is due for service, it is extremely difficult to stabilise the engine revs at the ESMP or "noise test revs". Even on a perfectly tuned machine, it is quite difficult to settle the revs to exactly the ESMP figure, especially for fuel injected engines, which are extremely sensitive when not under load.

Even experienced riders found stabilisation of revs difficult.

We found variation between the same model of motorcycle with the same exhaust system. This was for both stock and aftermarket exhausts. We could achieve repeatability of results with the same motorycle re-presented at different times. We have to conclude that the OEM tachometers are variable.

We went outside the scope of the Legislative test to probe the limitations of the test.

Different models of motorcycle exhibited a variation across a 500 RPM range centered on the ESMP figure. On some models, this variation was up to 6db(A). This is an issue with both multi-cylinder mid-capacity and large capacity V-twins. The relationship is not limited to the brand and model of aftermarket exhaust.

This variation can make the difference between a pass or fail of the EPA Regulation test.

Besides the variability of similar bikes with the same brand and type of exhaust, we found considerable perceptual differences.

The observation crew 20 to 30 meters away from the test area reported strong qualitative differences. Some exhausts gave a "soft" note and others a "harsh" note. The "harsh" notes often appeared to be louder than they tested. Similarly, some "soft" pleasant notes tested louder than expected.

This suggests that some noise complaints may be subjective "social complaints" about being able to hear an abrasive sound and not related to actual noise level at all.

Overwhelmingly, the aftermarket exhausts gave a distinctive "sound of a motorcycle" as opposed to the "sound of a car" and this correlated strongly with riders reasons for electing to utilise an aftermarket exhaust, in addition to lower cost.

Safety in traffic was cited as a prime reason for selection of exhaust type.

That the distinctive sound of a motorcycle assisted drivers of cars in locating them in traffic, by alerting them to look for "a motorcycle". The vast majority of riders did not select on the basis of absolute noise - quite the opposite - they wanted a distictive "note", but not one they wouldn't want living next door.

A motorcycle horn sounds the same as that fitted to a car and an inattentive driver who hears a horn sounding in traffic will look for a car, not a motorcycle.

There is no doubt that riders strongly regard a distinctive exhaust note as a prime tool for their safety in urban traffic.
The rider view is expressed as follows:-
Until the proportion of car-into-motorcycle crashes is reduced through improved driver behaviour, this additional noise is a cost the communty will have to bear.

The decision from riders is unequivocal, they dislike injuries.

This does not require an excessively noisy exhaust.

We concluded that the test methodology is fraught with variability and difficulty in obtaining repeatable results across different examples of the same equipment.