This article begins my series on understanding audio systems. This seems to be a bit of an advanced topic for many people, and I’ve had a lot of people ask me questions about it throughout the last couple of years. I’ve also come to realize that more advanced articles may be difficult to understand without a fundamental understanding of technical concepts. The understanding of a few of these technical and fundamental concepts related to audio will help you make more intelligent and informed purchasing and design decisions for both home and car audio. It is assumed that you already have a basic understanding of sound waves.  Today, we’ll start with frequency response.

What is frequency response?

In a nutshell, frequency response is a measurement of a loudspeaker’s ability to produce sounds of varying frequencies, depicting the magnitude at all frequencies given a linear input. This measurement is typically produced by way of a line graph, such as this one:

The above frequency response chart was taken of my 2012 Chevy Cruze with the base sound system.

In any case, a frequency response plot has two axes. On the vertical axis, you’ll see magnitude. Magnitude here is represented as “loudness” in decibels. It is dependent on microphone sensitivity, among a few other factors, so ignore the actual numbers in the above chart for the time being. On the horizontal axis, you have the frequency being presented.

Typically, frequency response charts range from 20Hz to 20,000hz (or 20KHz). The reason for this is that the human ear cannot perceive frequencies lower than 20hz. Anything lower than 20Hz is instead only sensed as pressure. The human ear also cannot perceive frequencies higher than 20KHz. To summarize, a frequency response chart presents the frequency range that the human ear can hear, and the amplitude at which each of those frequencies can be played by a given speaker.

 

What’s in a frequency?

A frequency or range of frequencies is used to play a particular tone or sound. It can also be referred to as pitch. Lower frequencies will be lower in pitch, while higher frequencies will be higher in pitch. I’ll provide a few examples so you can get a general understanding.

The human voice has a frequency range of 300Hz to 3,000Hz. 3,000Hz would be the highest registers of a female soprano voice, while 300Hz would be the lowest registers of the male bass voice. The lowest note on an electric bass guitar is 41hz. The highest note for a flute is around 2,000Hz. The highest note on an 88-key piano is around 4200Hz. Have you ever heard that high pitch whine emitted by a standard CRT (cathode ray tube) TV? I’m referring to those big, heavy, curved glass TVs from before the era of LCDs. That high pitch sound is around 16,700Hz. Here’s a page that shows you the typical frequency range of common instruments:

Frequency Range of Common Instruments

 

How does this help me?

Without getting into the technical aspect of speaker driver design, the basic concept is that speakers will produce frequencies based on a number of factors. These can be the crossover network that is designed, the size of the speaker, the material of the speaker, the size of the magnet, and the kind of box in which it is installed, just to name a few. Each speaker will have a different frequency response, and your goal is to make that frequency response look as flat as possible. While this is normally very difficult if not impossible to achieve with just a raw speaker driver, people often install equalizers that can adjust the frequency response of a given speaker.

For example, say you buy a set of speakers from the store, plug them into your home theater, and have a tool with which you can measure their frequency response. When you measure them, you might find that a region from 3,000Hz to 6,000Hz has a high peak due to a design compromise in the speaker. That peak would cause certain instruments or sounds to come off as much louder, thereby making the speaker sound very “bright,” but also fatiguing. Suppose that you bought speakers that had a large dip from 300Hz to 700Hz. That kind of a dip will cause male vocals to sound “thin”, nasally, or hollow.

The goal here is to achieve as flat of a frequency response as possible, so that the sound you’re hearing in your music will be as accurate to the recording source as possible, allowing you to hear it as if it was live. Any peaks or dips in the frequency response will cause the music to sound muddy, unclear, distorted, or fatiguing to listen to. In the upcoming articles, we’ll discuss additional concepts and ways to design a home or car audio system that will sound as true to the recording source as possible.

In loudspeaker design, unless for very specific reasons beyond this 101 course, targeting a flat frequency response will assure the most accurate representation of the signal being fed to your speakers and if the recording engineers have done their jobs right, the music! It is why flat on axis frequency response has become an effective mainstream standard for objective evaluation of loudspeakers. Straying from a targeted flay response can cause audible issues that can make a speaker muddy, fatiguing, etc. However advance designs and design techniques at times may stray from the rules to overcome other issues present. [Michael Z.].

 

Hit “like” if you like this article, and leave a comment if you have any questions!

– XTR