I worked next to a sidecar safety class yesterday and before my BRC started I had a short conversation with one of the side car coaches. When he explained what his class was going to be, I asked “Will there be any mufflers on anything on your range?”
His response was, “Yeah, like mine.”
His bike is a straight-piped Harley with a sidecar. When he’s demonstrating one of the exercises 100 yards from my range, I can’t hear any of my 11 motorcycles, what my co-coach or students are saying, or even my own thoughts. In some ways, it’s a great demonstration of how poorly the “loud pipes saves lives” delusion works, since his bike is pretty much non-existent when he is riding toward us and tears your ears off when he is riding away. Our training range is fairly open, so there are only a few reflective surfaces for sounds to be bounced back toward our location. Once again, the directional qualities of sound and exhaust pipes are most about telling people where you’ve been not where you are.
However, the comment that struck me most in that conversation was his comment as he walked away, “Gotta hear the motor.”
For the mechanically disinclined, the exhaust is the motor. It is the most obvious “expression” of the combustion part of internal combustion, after all. The problem with concentrating on the attenuated exhaust note is that it is all too capable of masking far more subtle, critical, and important engine noises. The ticking, whirring, and clatter of the valve train, for example, is easily overwhelmed and “masked” by the broadband and deafening noise output of a non-stock motorcycle exhaust system.
The masking effect of noise is fairly well understood and nicely diagramed in the drawing on the left. The “threshold of hearing” is frequency-dependent (your ability to hear sounds is directly related to the sound’s frequency content). As you can see from the illustration, our hearing mechanism is tuned to the general territory of human speech: 250-5kHz. Above and below those frequencies our hearing capabilities become severely limited. The poorly-and-improperly-applied “A-weighting curve” reflects the fact that at low volumes we are unable to identify low and high frequency sounds, for example. If you were to apply a noise filter shaped almost exactly the reverse of the “Threshold in Quiet” curve in the illustration, you’d have an A-weighted filter.
The point is, exhaust noise covers a pretty huge bandwidth, unaltered. I looked all over the web to see if anyone had published something resembling a “noise output spectrum analysis” of any sort of motorcycle exhaust. There are multiple sources with broadband comparison data like the chart at right. As far as I can tell, all of the charts I found are inappropriately “A-weighted,” so the data has been massaged to create a false impression that motorcycle noise is not particularly harmful, grossly disproportionate to vehicle size and utility, and outright idiotic as a lame attempt to provide public information. Honestly, the so-called “professional engineering” community (SAE, for example) should be shamed into public organizational suicide for their lame performance in this (and most) areas of noise measurement and control. The purpose of A-weighting is to obtain a typical human response measurement to low level noise signals. Low, in this case, was intended to be noise levels below 55dBSPL. No motorcycle is that quiet. Most of the noise signals measured and regulated are inappropriate for A-weighting, but A-weighting allows noise polluters considerable advantages.
This is, unfortunately, the closest thing I could find to a spectrum analysis of a motorcycle exhaust. The measurements were at 4M off axis of a car and motorcycle exhaust, A-weighted. Even with the weighting, you can see there is substantial low frequency noise output, especially for the motorcycle. That nasty hump around 100Hz would be about 40dB larger with C-weighting, which would give you a really dramatic picture of how hazardous a stock twin exhaust system is to your hearing. The real problem is that all of this normal broad band noise is bad enough, from a seat-of-the-pants diagnostic standpoint. Adding exhaust noise output will do the exact opposite of letting you “hear the motor.” Hearing the valve train, including the timing chain, the drive chain, the transmission gear whine and shifting, piston detonation, wheel bearing noise, etc is all made difficult-to-impossible by excess exhaust noise. I have yet to be impressed by the diagnostic or mechanical skills by any “loud pipes save lives” character. Their riding skills are even worse.