Follow Us Online! Website - Facebook - YouTube - Join Our Email List
Here is an example of an amplitude/response graph for a speaker with a somewhat irregular frequency response. Let’s call this speaker ‘A’.
Here is an example of an amplitude/response graph for a speaker with a somewhat irregular frequency response. Let’s call this speaker ‘A’.
The above measurement is taken free air in an anechoic chamber so all the peaks and valleys you see in the response graph are a product of the speaker and not the environment or enclosure. The peaks and valleys represent erratic amplitude at various frequencies.
You might be curious as to the cause of such uneven or erratic speaker behavior. There are a number of contributing factors.
One major cause is the speaker’s natural resonance. Just like a tuning fork, a drum skin or a guitar string, every object has a resonance. So the speaker, made up of the diaphragm, the voice coil, motor strength, spider and outer surround, has a complex but combined resonance that impacts the response.
Another major cause throughout the speaker’s range are the nodal or break-up modes where the speaker does not behave as a uniform diaphragm. At one frequency the speaker’s cone may warp much like a ‘Figure 8’. Looking at the speaker as if it’s a clock, at 9 and 3 the cone may be moving outward in unison but at 12 and 6 the cone may be moving inward or lagging in movement so that it is partially out of phase. This subtle twisting of the cone is referred to as the ‘conical break-up mode’. Another nodal pattern may be when the cone ripples from center to the outer rim similar to how the surface of water ripples away from the center when you throw in a stone. This subtle flexing is referred to as the ‘axial break-up mode’. At other frequencies you may see a combination of both of these described nodal patterns (conical and axial) and in varying multiples (number of twists or ripples).
Ripples can also travel outward from the center of the cone and reflect backwards so that the reflected energy is out of phase with and cancels out some of the direct or intended energy. Like I mentioned, these nodal patterns are complex and change behavior with frequency. And by the way, break-up modes get decidedly uglier as you increase the amplitude of the speaker. Since a speaker that reproduces 4000 Hz has to violently change directions 8000 times a second, there is no way you can see these rapid inconsistencies in the speaker cone behavior. But you can definitely hear them as they manifest in the frequency response and dispersion pattern. Here are eight graphic examples of various nodal patterns that can be part of a single speaker at different frequencies. A visual can be worth a thousand words.
So which speaker do you think will sound better? Certainly the speaker with the smoother response, like ‘B’, will sound more musical as it portrays every fundamental, harmonic and nuance in perfect balance and without false emphasis or dips that conceal vital information. Certainly speaker ‘B’ will cause less listening fatigue. Certainly speaker ‘B’ will give you the satisfaction of feeling you are hearing all the background detail at a lower listening level. Certainly speaker ‘B’ will deliver louder listening levels without irritation.
Here is another example of an amplitude/response graph for a speaker with an extremely linear (smooth or equal) response. Let’s call this speaker ‘B’.
In the next chapter of this article we’ll explore how some speaker builders achieve better results and why others do not.
Visit an Earmark and give our excellent speaker selection a listen, and look for continuing articles and blogs on this subject of Understanding Speakers.
David
