In my previous article, entitled “Analyzing a Stereo’s Frequency Response and Decay Times”, I suggested measurement tips and the use of important variables to do with frequency response and decay times. This article begins with a measurement baseline of a 2-channel stereo with two subwoofers within a typical domestic living room (25’ x 11.25’ x 8’) equipped with bass traps and diffusers. In this setting, problem frequencies were identified and matched with the relevant acoustic solutions, remeasured, and fine-tuned with my ears. The improvements were not subtle.

Acoustic measurements were taken using a Dayton Audio OmniMic software and microphone and then imported into a spreadsheet for further data manipulation and analysis. Only frequencies and their harmonics that represent notes played by traditional acoustic instruments (i.e. Concert A = 440Hz*) were plotted on the charts.

There are likely several approaches to better sound quality, but the path I prefer to follow is illustrated below:

Establish a Baseline for Comparison Purposes

The frequency response chart in Figure 1 shows large bass loudness variations below 300Hz, moderate variations in the middle of the chart (e.g. between 300 – 1,800Hz), and flat, even loudness in the high frequencies.

The software’s relevant T40 decay time measurement, which denotes how long a sound takes to fall 40 decibels, is shown below:

Identify Problem Frequencies

A quick scan of each chart above shows measurements that fall outside the upper and lower target limits, which would indicate problem frequencies.  To decipher the measurements, I used a spreadsheet’s computational flexibility to help create additional metrics not offered by the measurement software. I needed to identify and keep track of problem frequencies, and the table below helped identify when the two channels were not identical. Using test sweep signals, I looked for deviations from symmetry between the left and right channels.

To keep track of the problem frequencies and what caused them (i.e. peaks/nulls/inter-channel differences), I created the chart below with both bass frequencies and their root causes on the horizontal and vertical axes, respectively. The sound arriving at a listener’s ears should be the same from the left and right channels, so the chart presents the reasons for deviations from symmetry. The chart lays out 7 root-cause variables, each on its own horizontal line, as follows:

• Line 1 (dark blue, bottom line): Inter-Channel Loudness Variance – the difference between the left and right channel is 3dB or more, making it audible.
• Line 2 (red, up one line): Left Channel Nulls – at least 1dB below the -3dB target curve.
• Line 3 (green): Right Channel Nulls – same as above.
• Line 4 (purple): Left Channel Peaks – at least 1dB higher than the +3dB target curve.
• Line 5 (light blue): Right Channel Peaks – same as above.
• Line 6 (orange): Left Channel Decay Time – when substantial deviation exists from the green coloured middle decay time target curve (e.g. too long or too short a decay time).
• Line 7 (grey, top line): Right Channel Decay Time – same as above.

Match Acoustic Solutions to the Problem Frequencies

To help identify the acoustic solutions to the problem frequencies, I created the chart below, which summarizes traditional and commercially available acoustic treatments and equalization (EQ), and the frequency ranges they roughly affect. These represent the tools I used to address the problem frequencies.

Focusing only on the bass region, Figure 4 shows methods that can be used to address bass irregularities in Figure 3, including how to apply EQ, and pressure- and velocity-type bass traps for bass frequencies below 300Hz. My treatment strategies and tactics included:

• Strategy (A): Absorb bass
  • Bass absorption comes in two types: pressure and velocity (force and speed, respectively).** Pressure absorbers use a membrane or diaphragm to absorb bass energy while velocity traps use fibreglass and friction. Pressure traps usually work below 100Hz and offer a narrow bandwidth (1-2 octaves) of effectiveness while velocity traps work above 100Hz and are effective across a broader frequency range. They are complimentary products and most home listening rooms will need a mix of both types. To handle the problem frequencies below 100Hz, I bought 9 membrane traps and placed them around the room where pressure was highest, such as near a subwoofer or by a concrete/brick support wall. These were very effective at lowering the long bass decay times. I also added more velocity bass traps to help with bass peaks and nulls.
• Strategy (B): With multiple subwoofers and bass traps in place, EQ the remaining bass frequency irregularities
  • Further correction was done via digital equalization (assuming a person is using digital musical sources). EQ typically consists of parametric filters to cut bass peaks or the more capable Finite Impulse Response (FIR) type filters. FIR filters are used in real-time and mixed with the musical signal sent to a DAC, which then converts the signal to analogue so we can hear it. I use Audiolense software to generate FIR corrective filters, but FIR capabilities are also available in AV receivers that source it from DIRAC, Audyssey, Anthem, and other suppliers.
• Strategy (C): Diffuse and/or reflect mid/high frequencies
  • To help achieve a natural tonal balance, I corrected both bass decay times that were too long and high frequency decay times that were too short. Diffusion/reflection tend not to reduce the decay times of the highs’ smaller wavelengths, but when upper frequency decay times are too long, absorption can be an effective tool. Removing hard Styrofoam-like diffusion from the front and back walls and replacing them with wooden diffusion helped lengthen the high frequency decay times.

In Figure 5, the top chart shows the ‘Before’ problem frequencies, while the bottom chart shows the ‘After’ results produced by my acoustic treatments. Pressure-based bass traps were very effective at reducing long decay times while velocity-type bass traps helped reduce nulls and peaks and decay times. Equalization using a FIR filter operating below 600Hz helped smooth the frequency response.

Iterative Measurements and Results of Acoustic Treatments and EQ

A room’s response to acoustic treatments can’t always be predicted accurately, so the placement of treatments becomes an iterative process (process of repetition), and measurements help immensely. The “After” results are shown in figures 6 and 7.

After the treatments were implemented, the same acoustic variables were remeasured. The following Table 2 provides a “Before” and “After” comparison.

Final Calibration Using Listening Tests

Once the measurements were as good as I could get them, it was up to my ears to do the fine-tuning. This elicited a couple of important discoveries: 1) the QRD diffusion on the sidewall’s first reflection point from the closest speaker reduced midrange clarity so was moved to the front wall, and, 2) the angled wooden board on the sidewall’s first reflection point from the far speaker needed more angling so the left ear heard less from the right speaker (same for the other ear), which resulted in better midrange detail.

The post-treatment sound quality improved in several ways: bass boominess had disappeared, making the lower midrange sound clearer; bass notes were more distinctive and easier to hear as one note decayed before the next note began; and wooden diffusion on the front and back walls seemed to increase the midrange’s decay times and, consequently, improve the sound’s “liveliness” factor. Without any prompting on my part, multiple visitors who heard the “After” system commented on how well-balanced it sounded.

Treating a room acoustically using software and measurements can be a prolonged, iterative process, where solving one problem can sometimes expose others. But, in the end, with patience, perseverance, and a strong desire for better sound quality, your efforts should pay impressive sonic dividends that’ll make you happy you went through it.

_{* “Concert A” or “A440” is the musical pitch above middle C used by an orchestra’s musicians to tune their instruments.}

_{** A pressure bass trap is placed where bass pressure (force) is greatest, i.e. against the wall. Where pressure is greatest, velocity is zero, and vice-versa. A velocity bass trap is placed some distance from the wall (velocity is zero at the wall surface) and uses a fibrous insulation-type material to convert bass waves into heat.}