Loudspeaker Measurements – Stereophile

Stereophile Magazine

I always enjoy reading the equipment reviews in Stereophile Magazine, especially the speaker reviews. Speakers have always amazed me in how electromechanical energy is converted to sound energy and then interjected with our room through the speaker drivers. John Atkinson always does the testing using an oscilloscope and a DRA Labs MLSSA measurement system. John’s technical data usually follows the audio/music review by another reviewer for Stereophile.

Conclusion, First

When I get the magazine, the first thing I do is look at the conclusion John draws from his analysis of all the variables he has measured. It is found in the last two or three paragraphs of his review. I look for anything that jumped out at John and was annoying enough to cause him to mention it. If it a technical thing that I can not hear, I read it and acknowledge it, but don’t pay too much attention to it. If it is audible and John usually explains what the technical issue he discovered means to our ears.

Loudspeaker Impedance

The first measurement John always performs in that of the impedance of the loudspeaker. The impedance of the loudspeaker in simple terms is the electrical range from low to high that the speaker works and operates best in. Impedance covers the optimum operating range of all the drivers used in the single speaker cabinet. One speaker, John reviewed, had an impedance that was higher in the region covered by the tweeter over the other drivers. John’s testing equipment allows him to focus on each individual driver and its associated impedance within the speaker cluster. Nice!

Impedance Matching

Impedance matching is necessary with all speaker/amplifier combinations. The amplifier must have the proper current rate and level and feedback must be controlled and matched correctly to work within the impedance range of the speaker. If not, energy from the amplifier may not be available when the speaker requires it and then one will miss certain sounds completely, especially in the low end. The reverse is true where excess energy from the amplifier can exaggerate certain sounds, wanted or not.

Bad Vibrations

Vibrations within the cabinet itself are next examined with an accelerometer attached to the cabinet itself. Remember that our speaker cabinet is really a small room and is governed by small room acoustics. Back wave energy from the speaker cabinet drivers produce resonances within our speaker cabinet “room”. These resonances are dependent on the speaker cabinet’s internal volume and manifest themselves as cabinet structural resonances. These cabinet resonances may have a high enough magnitude to be heard mixed in with sound of the driver or drivers. I never really knew what those break points were. I am certain they vary with frequency.

Tweeter

John Atkinson then usually goes for the tweeter, kind of a top speaker down approach and measures the tweeters on axis response. This response measurement domain is averaged over a 30 degree, horizontal window. What do we want in our on axis response or our response within this 30 degree horizontal field?. Lets say that is our listening couch described in acoustic terms. We want a very even and balanced electrically and acoustically across horizontal window or listening couch.

Everything Represented

We need everything represented within that 30 degrees. All instruments and vocals must be present. All of it must have equal amounts of sound pressure behind it, so one particular vocal or instrument stands out. We always look at the vertical axis of measurement because a horizontal measurement is plagued by side wall reflections. Not so, in the lab situation, but in real life speaker placement positions, side wall reflections are sound stage and image destroying for our two channel audio presentation. Each driver in the speaker cabinet in given the same testing procedure for on axis sound distribution.

SMART Software

The SMART software can go into great detail on the analysis of the loud speaker. The software takes 10 different measurements centered on a vertical grid that outlines the reviewers ears. Listen to the interpretation by John Atkinson of this detail in the sound measurements.

“Other than a slight excess of energy at the bottom of the tweeter’s pass band, the treble is smooth, with a slight downward trend due to the increased absorptivity of the room’s furnishings in the top octaves. However, the lower mid range is shelved down, and while the low frequencies are aided by the lowest frequency resonances in my room, they are still a little light compared with the level of the treble.

Horizontal Dispersion

Another response measured is the horizontal dispersion of the loudspeaker. We must measure bass, mid range and tweeter the same way we do for vertical om axis distribution. Each speaker’s dispersion pattern has an impact on the final sound. One can have a bass and mid range that is balanced, wide, and evenly distributed throughout the horizontal plane. One can have in the same speaker, a tweeter that becomes directional as the frequency rises. This translates to a sound that will not fill a large room with the required amount of sound.

Measurements are critical with our loudspeakers and they can tell us a lot about the speaker’s ability to produce the sounds we expect the loud speaker to produce. There is the electrical matching that must be considered between the loudspeaker and the amplifier. Speakers with low impedance curves need stable, well damped amplifiers that can produce high levels of current. There are cabinet resonances that must be low enough in amplitude to not be audible and thus heard. Both vertical and horizontal dispersion patterns are analyzed within individual dispersion domains. All of these numbers have to be considered and agreed to before any purchase is made, weather it be an amplifier or speaker.