AN3340

Frequency response of a loudspeaker

Doc ID 18453 Rev 1

9/24

5

Frequency response of a loudspeaker

While the speaker electrical characteristics and the impedance measurement can be

performed using standard electrical equipment available in an electronics laboratory

(oscilloscope, function generator, current probe and so on), the frequency response

measurement needs a dedicated environment and specialized tools.

To determine the frequency response of a speaker it is necessary to measure the sound

pressure that the cone produces when a signal is applied to the voice coil.

To perform the sound pressure measurement it is necessary to use a microphone to convert

the sound back to an electrical signal. The microphone frequency response must be known

since this is also part of the measurement loop and must be subtracted from the overall

response.

The test should be performed in an acoustically shielded environment, that is, in an

anechoic chamber, to eliminate reflections and background noise.

If an anechoic chamber and the related tools (microphone, specialized software) are not

available then the speaker frequency response must be estimated using an alternative

method.

An accepted approach is based on the MLS procedure using Audio Precision equipment.

This allows the test to be performed without an anechoic chamber6.

With the MLS method the loudspeaker frequency response is measured in a simulated

“quasi-anechoic” chamber. This means that it can be done in a laboratory without particular

acoustic requirements.

However, even with this approach, the performance of the microphone must be known.

Figure 6 shows the frequency response measured by the two methods on different

speakers.

Figure 6.

Speaker frequency response

-80

-50

-70

-60

dBV

200

20k

500

1k

2k

5k

10k

Frequency (Hz)

100

dB

-30

-40

-50

-60

-70

-80

100

1k

10k

Frequency (Hz)

Speaker 1 measured in anechoic chamber

Speaker 2 measured using MLS method