Measuring Frequency Response

This document covers the practical methods undertaken to measure the frequency response of the infinity hall.

Using Room EQ Wizard a frequency response was taken to see how sound behaves within the sound hall with the current speaker configuration. This is to assess the response and look for psychoacoustic anomalies that need to be mitigated or reduced to the best of the engineers ability.

The first step was to set up the measurement microphone. The microphone was placed equidistantly between the front of house speakers, carried out by lining up the microphone with the middle of the stage. Using a laser measurer, tape was marked down in meter increments down the middle of the sound hall, starting from the front of the subs. The reason for not measuring from behind the system is that the reproduction of the sound won’t be perceived as it is when heard from the perspective of the audience and engineers. To achieve an accurate response of the room, the microphone and stand were angled directly upright, rather than aiming towards the system, so that the room acoustics weren’t manipulated by the system.

When taking a response measurement of the room it's important to make note and take into account the current acoustic treatment and reflective reductions on the room. In the sound hall, there are acoustic drapes that reduce wall reflections either side of the stage behind the system. The floor is shiny and smooth which will cause reflections, however when the hall is full of people this will be reduced. The back wall/door of the room will cause reflections due to not having treatment or drapes on it, however there is acoustic treatment on the back and side walls. To reduce back reflections, the system, currently configured in point source, can be angled down slightly.

The system is the d&b T10 tops configured in point source paired with 4 x B4 subwoofers. Operated with the Yamaha CL5 desk. The microphone used for measurement was the Klark Technik 6051 Condenser Measurement Microphone. The microphone has an intentionally flat frequency response so as not to colour the results of the room in its natural state.

A 20Hz-20kHz sweep was sent through the system to accurately measure the room’s response to all frequencies within the human hearing range. The first frequency response taken and analysed showed that 12kHz has a longer decay time when viewed in a waterfall graph on room eq wizard. On the CL5 desk 12kHz has been eq’ed down to -6dB. Although this is more drastic, a more even frequency response is predicted on the 12kHz decay. A second frequency response sweep was taken and analysed. As predicted, there is a clear reduction at 12k. There are accurate graphs showing the -6 dB reduction on the graph sweep.

A song was played through the system to test the different frequency bands, and using the Fletcher and Munson methodology of hearing response, the system was eq’ed to boost the bass, take out some of the mid and boost the hf. The song was then listened to again with an analytical ear and sounded more balanced and the frequencies appeared to blend more. The eq’ing is objectively seen as a correction on the system, however the perception of the resulting sound is somewhat subjective based on how trained each individual engineer's ear is. The Fletcher and Munson phenomenon is important to consider during live sound operation as loudness affects the human perception of different frequencies and frequency bands. For example, sound played at a lower dB volume will present the mid frequencies as more prominent, however once the dB level is increased, the bass and high frequencies will become more prominent. This is all based around the human ear and how it reacts to varying sound pressure levels.

During the process of Measure, Analyse, Process/Adjust, and repeat, when playing the song, the fader levels were adjusted. To get accurate results the faders must remain at a constant level. Due to this small oversight, the responses were retaken at a set volume before and after eq’ing the song for the music. PVC tape was used to keep the faders at a constant volume throughout the measurements.