May 29, 2011

Why you need a subwoofer for music

If you are aiming for the best bass you can get, you need at least one subwoofer, even in a two channel system. When this is suggested, a common response is "my mains have enough bass!" The reasons for using a sub aren't quite so obvious.

What is required to achieve exceptional bass:

  • adequate extension to reproduce bass as low as your source material demands
  • adequate output to achieve the highest level you expect in your listening position with headroom
  • independent level adjustment
  • adequate headroom to allow for a room target curve and some EQ
  • ability to place the bass speakers where they perform best 
  • multiple placements - from one to five units

Obviously the bass drivers included in your existing speakers can't meet all of those requirements. In-built bass drivers on your mains are far too limited to achieve the best that is possible. In most cases, sub will work best in locations that could not be used for a full range speaker.

I use the term "subwoofer" broadly here. An active woofer not intended for home theatre use can also be used.

The main reason for using subwoofers is that they can be placed to get the best out of the room. Once properly integrated, one can achieve a much smoother response that is free from peaks and dips that plague all rooms. It isn't difficult to remove peaks with EQ, but the unique benefit to a multi-sub arrangement is the elimination of dips.


May 27, 2011

Bass integration guide

In this article, I cover how to best set up the bass in your room. It has occurred to me that many audio enthusiasts pay careful attention to every aspect of their system while accepting mediocre performance in the bass. In response I've put together a process aimed at pursuing the best possible performance. It isn't for those willing to stop at good performance. I'm writing this for those who are passionate about bass. 

The first of this three part series is now online at Hifi Zine:
(June issue 2011)

The first part focuses on measuring the current level of performance in your room. Some performance targets are also introduced so that you can see the areas that need improvement.

Bass integration - part 2
(September issue 2011)

Part 2 covers acoustic treatment. How to modify your room for improved performance and how to choose bass traps.

Bass integration - part 3
(June issue 2012)

Part 3 covers speaker placement and system calibration.Of particular interest to many will be EQ.

May 24, 2011

Room measurement service

A new room measurement service in Melbourne

Following numerous requests, I'm now offering a room measurement service to Melbourne suburbs.

Update: I also provide a remote service that enables you to take your own measurements, and I will provide support and interpretation services.

More about the remote service >

When done right and correctly understood, measurements will show you exactly how you can achieve great bass in your room. If you want to learn how to take your own measurements, I can also show you how to do it effectively.

You can find out:
  • how your speakers and room are performing
  • how you can benefit from acoustic treatment
  • how you can benefit from EQ
  • how best to integrate your system with the room

I will explain the charts and what they mean.

How the process works

1. Define the goals

You explain what you would like to get out of the process. You might want to see how your system is performing. Or you may be after help with integrating your sub, which is more involved. If your expectations are a bit higher, you might want to go through a complete bass integration process.

2. Measurements

I will take measurements in your room to suit your goals.

3. Analysis

I will explain the charts and what they are indicating, along with pointing out areas for improvement.

4. Suggestions for improvement

Depending on what you are open to considering, I will discuss options to improve.

The process can become more involved than this, depending on your requirements. A complete bass integration may require more than one visit. 

What kind of measurements are involved?

I will take measurements that show the response of of the speakers independent of the room as well as in-room measurements. Frequency response is a start, but there is a lot of important information about a room contained in time domain measurements. Even if you have used EQ to get a flat response, it's more than likely there are substantial further improvements possible when you consider also the time domain.

What happens next?

What you do with the results is entirely up to you. You might satisfy your curiosity or you might want to do everything you possibly can to get great bass. Bass traps, multiple subs, active woofers, EQ, placement changes, phase alignment - those are all options I can help you with.

Room measurement case study >

Enquiries

Contact me to enquire further regarding this service
(Link will launch external contact form)

May 23, 2011

Cat: "I'm a person in a cat's body!"



Introducing Otis. He thinks he is human. A person in a cat's body! When I saw him sitting in my seat expecting a first class breakfast, I couldn't resist the Kodak moment ... even thought I had to put him up there three times to recreate it!


May 21, 2011

Unity horns + Acoustic Elegance TD18H + Maelstrom 18

This afternoon I had a chance to take some measurements of some speakers of great interest. One of very few Unity horns, along with some Acoustic Elegance TD18H woofers and a Maelstrom sub.

You can also see a pair of 6th order bandpass subs with 12" drivers. 

Nearfield measurement of the Maelstrom. It was converted to Aperiodic since the vents were tuned too high. 

Acoustic Elegance TD18H - nearfield measurement. 

All up, 3 subs and two active woofers. Quite a lot that can go wrong! There were some challenges in setting up EQ and integrating them. The Maelstrom sub near the couch was used for the bottom end. The bandpass subs were actually reducing the bottom end output due to interference. Compared to integrating a sealed sub with mains, there is a great deal more complexity.

May 19, 2011

The importance of room damping

Room damping or more specifically bass damping, is an essential part of accurate bass reproduction. All charts shown here are based on a REW simulation of a single subwoofer in the left corner in a small room 4.65m wide x 3.9m deep x 2.85m high. You quickly and easily generate similar results. They will differ based on the chosen listening position. Here we are focusing on the room itself, without any acoustic treatment.

What is bass damping?

Bass damping is simply absorption in the bass range. I refer to it as bass damping as we are dealing with room resonances that need to be damped. You can achieve it with room construction or added treatment. Here we will focus on the former. Plasterboard/drywall light framed walls work well as bass traps. Sound waves cause the boards to flex and in the process, energy is absorbed. The effectiveness is based on the depth of the air gap behind, the presence of insulation in the cavity and the mass and stiffness of the membrane.

Here you can see the impact of damping on the frequency response.


The top navy blue line is a room with minimal damping. That means a very low absorption coefficient of 0.1. You can see this is a poorly performing room with the worst peaks and dips. The tell tale sign is that they are all narrow Q (sharp and narrow). The best room has an enclosure with very high absorption coefficients (0.9). Each chart increases the absorption coefficient in increments of 0.1. A coefficient of 0.1 will absorb 10% of a sound wave. The black line represents the kind of damping you might see in an anechoic chamber. Obviously it isn't practical as a listening room due to likely cost and pragmatic issues. This level of damping would only be achieved with a great deal of added acoustic treatment.

Here are three selected rooms.


The navy blue line again shows the worst case scenario. The magenta line shows a room that is quite good and may represent many lightly constructed rooms with plasterboard/drywall and a timber floor. The green line shows a room that is very highly damped. It would probably require added bass traps, or special construction. 


In this waterfall, you can see just how bad this room is. Not only are the peaks very sharp, but they also decay at a very slow rate. After 300 ms they have only decayed by 10 dB. While this is a problem in the bass range, it is even worse above 100 Hz where the ear is more sensitive to time domain effects. The result would be a cavernous sound. Fortunately few rooms would be this bad.




In this waterfall, you can see much better performance. Here one of the limits of the simulation becomes clear. In a real room, the bass region below 80 Hz would be worse, but above it would probably be better. Still, it isn't too difficult to see how this is a big improvement.
 
With very high damping in this room you can see that the response is much flatter and any peaks and dips are much broader. The decay is very even and rapid. If a room were built to perform this well (not likely), very little extra effort would be required for exceptional performance. Unfortunately this level of damping requires a very well designed room and extensive added bass traps.

Bass traps vs room modification

It is important to combine acoustic treatment with a room that also provides damping within its envelope. The room damping tends to be restricted in bandwidth and is not likely to do all that is required. If it is left out, the additional space required for larger bass traps may not be practical. The best and most sensible approach is to combine the two different types of damping. With this approach in mind, the bass traps used should be of the porous broadband type. Membrane-based traps are not likely to have the required bandwidth.

What are room modes?

Room modes are a collection of resonances that are unique to each room. The are based primarily on the size and shape of the enclosing boundaries seen by sound waves. They present the greatest obstacle to accurate bass reproduction, and they cause problems in even the best sounding rooms. 

Types of modes

There are three types of modes - axial, tangential and oblique.

Axial modes

Axial modes tend to dominate the modal response. They are formed between opposite wall surfaces.

Tangential modes

Oblique modes

Oblique modes are less significant due to their longer path which causes their energy to dissipate.

Getting back to basics

To understand modes, it helps to consider a simpler situation than a normal room. First, let's consider a sine  wave in a free field environment with no boundaries (not even the ground).
At that one frequency, you have peaks (nodes) and dips (anti-nodes). If you were to move towards the source, you would notice the tone gets louder and softer. For simplicity the modes in the first three diagrams are shown as lines, but each line represents a sound wave with peaks and sips along its length as shown above.

Now let's consider what happens when we add a boundary. We'll add in the ground.


Now we have a secondary source with a longer path length. That path difference means that the reflected sound wave will interfere with the direct one. We can expect a dip caused by the difference in phase of the two sound waves. The higher the level of the reflected wave, the greater the impact on the response. This effect is referred to as SBIR (Speaker Boundary Interference Response). The first three reflections are covered there - those off the side and rear wall and the floor.

In a rectangular room there are 6 enclosing boundaries. Now consider the first three diagrams. In the midrange, those nodes and anti-nodes become closely spaced together so that they are reasonably smooth. It is in the lower midrange that there is a transition point where nodes and anti-nodes are spaced far enough apart that they no longer can offer a smooth response. This transition point is called the Shroeder frequency, and it will often be around 200 - 300 Hz. Below that point we see major peaks and dips in the response until we reach the lowest mode. As wavelengths approach and exceed the largest room dimension, we no longer encounter modes and the room becomes uniformly pressurised. The room provides acoustic boost below that point depending on the extent to which bass is either absorbed or transmitted through the envelope.

The importance of position

Room modes create a complex pattern of peaks and dips in a room. The existence of modes is determined by the room shape and its acoustic properties. The way in which those modes are excited depends on the location of the speaker and listening position. As a result, shifting the location of the speakers and listening position will change the response. Placing a subwoofer in the corner will tend to excite the most room modes and achieve the maximum output. This may or may not be a good thing. If you then sit against the back wall you are likely to experience strong peaks and a boomy sound. Shift your seat forward and the bass level and boom will be tamed. This is the simplest form of adjustment to suit the room, but you can certainly improve on that with more advanced techniques.


Here is a section through a room with a depth of 3.9m and a sub in the corner. Position 1 is with a couch against the wall, 2 is moving out and 3 is in the middle of the room. So you can see that if you sit against the wall in position 1, you will see a strong peak at 44 Hz and also one at 88 Hz. Moving out from the wall, both will decrease but in position 2 there will be a dip at the second order resonance. In the middle of the room (position 3), there will be a null at the first order resonance and a peak at the second. All this assumes a sub in the corner as shown to excite all these modes. This is in fact my current listening room. The first order axial mode for this wall pair is at 44 Hz. This can be calculated by this formula:

Frequency = 0.5 x 344 / Distance between walls in metres

The second order mode is a multiple as is the third and so on. In this particular room, in position 1 the 44 Hz mode is dominant with most sub and speaker positions. The result is a 15 dB peak which causes the room to boom.

A decay plot can show up the presence of modes in both frequency and time domain.


The top of the red plot shows the response of a full range speaker, while the purple plot shows the response after 150 ms. Where the gap between the two plots is small, that indicates modal ringing.

  1. 44 Hz - first order depth mode
  2. 61 Hz - first order height mode
  3. 84 Hz - second order depth mode
  4. 130 Hz - third order depth mode
Note that two of the four easiest to pick modes are depth modes. The position is against the wall (position 1 on the previous diagram). So this is not surprising. Harman Audio has a useful tool for quickly working out the modes that could possibly be excited in a room.


It is a handy tool to use in making sense of measurements. 

So what does all this mean? 

Room modes do two things. Firstly, they introduce peaks and dips to the response. Secondly, they cause ringing in the time domain so some frequencies stay in the room longer. This is a serious impediment to accurate bass reproduction. Understanding room modes is just the beginning. Dealing with them ideally involves a mix of acoustic treatment, placement, multiple bass sources and EQ.