Acoustics 101-Designing Sound Treatments

Front Broadband Absorbers and Bass Traps

Front Wall Absorbers and Bass Traps

Rear Corner Broadband Absorbers and Bass Trap

Introduction

Acoustics 101-Designing Porous, Resonant and Reflective Sound Treatments The last two posts introduced you to the nature of sound, it various parameters, the types of absorbers and diffusers and how to locate them together with your subs. Acoustics 101-Designing Sound Treatments will show you how to design those porous absorbers and diffusers and find the first points of reflection.

For those serious compulsive fiddlers you should at this point set yourself up with an acoustics analysis program and an inexpensive, but professionally calibrated, microphone. If you are new to this process I can highly recommend Room Acoustic Wizard (REW) (free) and a calibrated Dayton microphone from Cross Spectrum Labs ($75.00). Both of which I use and have considerable experience with.

NOTE: Make sure that any microphone you purchase supports loop back timing. I use the EMM6 Premium Plus not the USB version. You need loop back timing in order to allow you to correctly time each speaker.

Reflection Points

We still need to know how to find the first points of reflection. They may be found using one of three techniques:

Using a Computer Aided Design (CAD) program. – Fast if you are CAD proficient and have all your room elevations correctly scaled.
Using an acoustics program and a piece of string. – A little time consuming but works well.
Using a mirror and a helper. Quite fast and very simple. – Recommended for the beginner.

In all cases you need to determine and mark, the acoustic center of each of your speakers on the front of each speaker, after removing its cover. Well at least L, C and R.

Method 1

If you are CAD proficient then using correctly scaled room elevations and CAD is very fast. Just remember that that the angle of incidence equals the angle of reflection. Once they are all determined mark them and get ready to place absorption or diffusion at each location.

My A/V Room First (and second) Point Refection Paths For All Speakers. Reflections Not Shown If Absorbed.

Method 2

If you are familiar with an acoustic measuring system like REW you can use it to measure the the Impulse Response (IR) or Energy Time Curve (ETC) which will provide you with a view of all the rooms reflections at the MLP (see post 2) and below.

Acoustics 101-Designing Sound Treatments

Example Energy Time Curve – ETC

You may then select those having too high a level, and determine their increased path distance over the direct sound using the cursors in the ETC program. In this example there is a bad reflection 3.178mS or 3.58 feet after the direct sound. Now attach one end of a piece of string to the speakers acoustic center and the other end to the location of the measurement microphone. Adjust its length to that calculated by the ETC program plus the direct distance to the MLP for the speaker under review. Taking up the slack, find a wall, floor or ceiling point that the string will just touch. This is usually just one point. If it is more than one you can check each locations reflection using the mirror method described below. Mark each location where the string touches, you will be placing absorption or a diffuser at those points.

The String Method of Finding the 1st point Of Reflection

Method 3

For the mirror approach you need two people, one to sit at the MLP the other to hold and position the mirror. Now sit in the MLP and have your little helper position the mirror on each wall, ceiling (and floor) so that when you look into it you can see each speakers acoustic center marker. Mark the locations on the surface and you are now ready to place absorbers and/or diffusers at all those locations.

The Mirror Method of Finding the 1st Point Of Reflection

Absorber or Diffuser?

At this point the selection of the type and size of the treatment is up to you and your room design and layout. The simplest approach is to start of with at least 4″ porous absorbers that are no smaller than 2’x2′. Remember that sound doesn’t travel in a beam it spreads out so the area you treat must be sufficient to stop reflections over more than just a single point. Remember, you move your head and there are others in the room with you too. Areas next to and around speakers need to be especially deep, maybe 8″ and at least the height of the speaker.

Calculations and Mathematics

For many the math required to design acoustic treatments may be considered to be either tedious or beyond your skill set. So while I have provided some simple higher level formulae, in the interest of simplicity they may not take into account all parameters. Also the calculation process is often required to be iterative so using a free program, of which there are many, will make your life much easier, and is really a must for most of us. A selection of web sites that support the required calculation tools are provided during and at the end of this post.

Box Construction

All my porous and resonant boxes are constructed the same way.

Porous Absorbers

The four sides are cut from 1/2″ MDF. They are then glued and panel pinned together. The resulting frame is then glued and stapled to a 1/8″ ply backing.

Once the frame is filled with absorber an acoustically transparent cloth is stretched over the front and sides and stapled to the rear. Besides appearance this cloth is essential to prevent the release of fibers and aids in keeping the absorber in place.

For those with the tools and patience you can remove a large percentage of the sides of the box. In doing so you would expose a significant amount of the absorbers height area thereby increasing its absorption area and coefficient values. (You could of course directly wrap the absorption material especially if it is rigid. I found that technique difficult to do effectively and the panels were much more difficult to handle and install.)

EXAMPLE:

A 4″x12″x24″ (DxWxH) panel has an exposed surface area of 12×24=288sqinches. The exposed area of all the sides would be 2x(4×12) + 2x(4×24)=288sqinches, equal to that of the front surface area. So exposing the sides of a porous absorber can help particularly with overall random incidence room absorption.

Resonant Absorbers

The four sides are cut from 3/4″ MDF then glued and panel pinned together. The resulting frame is then glued and panel pinned to a 3/4″ MDF backing. All joints are then caulked with a sealant. You do not want this box panels to resonate, vibrate or leak air. Once fully assembled and the membrane or perforated panel is attached an acoustically transparent cloth is stretched over the front and sides and stapled to the rear making sure it is clear of the front panel or membrane. This cloth is only for decorative reasons and can be omitted.

It is essential that the front membrane or perforated panel is air tight to the box sides. This maybe achieved by using silicon (or similar) caulking on the surface of the side panels and then pinning a batton all around the edge which doubles as a spacer for the front cloth.

Qudratic Residue Diffusers.

Building these is quite time consuming but creating the dimensions is straightforward using one of the many programs. My favorite being QRDude. (Be sure to read all the web site articles on QRD’s before starting your design.) At least one vendor, Decware sells kits for small DIY QRD’s. The following diagram will give you an idea as to how to build one. Usually QRD’s are deployed in at least pairs and often in large groups. Pairs are often used in alternate alignments and groups are often set to a particular (Barker) sequence requiring at least one of the panels to be of a different well design.

Porous Absorber Design

General Wide Band

These are the most basic of absorbers and may be made from Fiberglass or Roxsul. If you use the rigid versions you can bond the fabric covering straight to the rear of the panel using an impact adhesive. However, I found these difficult to install. If like me you prefer additional mounting options build the simple wooden frame and use the flexible absorber. In my opinion you should not use absorbers that are less than 4″ thick due to their poor absorption at low mid frequencies. Thin 1″ panels are acceptable if you are only trying to prevent effects like flutter echo. Wide band panels make great 1st point absorbers unless you want to spend hours or lots of money either making or buying diffusers. Diffusers will generally reward you with a more lively room and a longer decay time. However, once installed you will need to check that they are not increasing the ITD reflections by using an acoustics program.

Basic Porous Absorber Panel – Style 1

Porous Absorber with Increased Surface Area – Style 2

The block approach (see opening images) in style 2 using 4″ Roxsul provides both an 8″ deep absorber and an increase in surface area of approximately 50%. Remember that absorption is proportional to both depth and surface area.

Style 1 – Porous Panel Absorption Frequency Response – No Air Space

Clearly even 4″ of Roxsul is only really effective down to about 125Hz.

Bass Traps

We saw from the last post that there are many styles of porous velocity bass traps. The four most important parameters are:

Location – the front surface must be in a high velocity area, so well away from the room boundary.
Depth – as deep as you can possibly make them with due consideration to item 3 below
Ease of sound pressure ingress
Area – you can never have too much bass trapping.

Corner Bass Trap – Style 1

Bass Trap – Style 2

For deep FILLED corner traps I recommend OC703. As we saw in post 2 Roxsul has a higher flow resistance. If you use Roxsul the second bass trap design is more appropriate once you get much deeper than one foot. See this posts opening images of my room.

Corner traps can easily be made by cutting triangular sections of 703 (or Roxsul) and just stacking them on top of each other to fill each room corner floor to ceiling. Ideally adding battens on either side of the front of the chunks to hold them in position and provide a surface to which to attach an open weave cloth covering, see below.

Wall Corner Bass trap

Ceiling Wall Bass trap – Rigid Fiberglass Chunks

This technique should also be applied to all the wall/ceiling junctions by installing a frame and filling it with absorber. These porous absorbers are generally not very deep being typically 12″ or so. See my room below. These will help reduce all three room modes.

Ceiling/wall corner porous absorber using Roxsul. The rear QRD and rear corner resonant panel absorbers prior to membrane installation.

NOTE: After installing all my resonant panels in both the front and rear of my A/V room, even though they all resonated as predicted, they provided insufficient absorption below 100Hz and were all ultimately replaced with wide band porous absorbers.

The depth of a porous bass trap should ideally equal 1/4 of the wavelength you are looking to absorb. If you room has its first standing wave at 40Hz, then using the formula 1125=wavelength(ft) x frequency(Hz), its wavelength is 28 feet and quarter wavelength is 7 feet. So we immediately see that even if the trap is 3.5 feet deep we aren’t very close to having the absorber in the highest velocity region. However, even a 2 foot deep trap will still provide useful absorption even as low as 20Hz, particularly if it is Style 2 above.

Typical 24″ Deep Bass Corner Trap Absorption Characteristic – No Air Space

Resonant Absorber Design

Membrane or Panel

There are two types of panel or membrane absorber, the ‘rigid’ or limp membrane.

These panels are pressure absorbers so need to be installed at room boundaries on the walls or in corners. It is essential that the cavitys are air tight and that the damping behind the panel doesn’t touch it and is not much more than 1/2″ behind it. This absorber is used to damp the panel, absorb the energy and stop the panels vibration. The flexible solid panel is often replaced with what is referred to as a limp membrane. These are panels made from materials like roofing felt, Revac and Dedpan. They have the added advantage that they are self damping and do not really require a damping membrane, but do still require absorption within the cavity.

The resonance frequency of both types may be calculated using the formula:

Frequency (Hz)= 60/sqrt(m.d) where m=the panels mass in Kg/sqmeter and d=the panels depth in meters, OR

Frequency (Hz)=170/sqrt(m.d) where m=the panels mass in lbs/sqft and d=the panels depth in inches.

Some typical panel masses in Kg/sqmeter are:

1/8″ hardboard 3
1/8″ plywood 2.3
12mm softboard 3
6mm plywood 3.3
8mm plywood 5.3
9mm MDF 6.5
12mm MDF 9.0
1/2″ standard drywall 9.3

EXAMPLES:

For an 1/8″ plywood panel with a sealed cavity depth of 75mm (3″) its resonance would be 60/sqrt (2.3 x 0.075)=144Hz.

For an 8mm plywood panel with a sealed cavity depth of 150mm (6″) its resonance would be 60/sqrt(5.3 x 0.15)=75Hz

For 1/2″ a standard drywall panel with a sealed cavity depth of 150mm (6″) its resonance would be 60/sqrt(9.3 x 0.15)=51Hz

Clearly panel absorbers have the advantage over porous absorbers in that they are relatively shallow in order to absorb low bass frequencies.

Typical Membrane Panel Absorption Frequency Response

This type of absorber is really only suitable to deal with specific room modes and should not be used for general broad band basss absorption.

Helmholtz

All of these resonators rely on the mass of an air pocket resonating with a much large volume of air. There are numerous styles and designs of this type of absorber from a single simple resonator with one opening to using perforated panels and slat openings.

Basic Helmholtz Resonator

Panel Helmholtz Resonator

For a single opening the approximate frequency of resonance (Hz)=55 x sqrt(A/L.V) where A=opening area in sqm, L=length of neck m, V=container volume cubic m .

For a perforated panel the frequency of resonance (Hz)=200 x sqrt(p/D.t) where p=% panel perforation, D=cavity depth in inches and t=(PT+0.8d) where PT=panel thickness in inches and d=the hole diameter in inches.

EXAMPLES:

Single opening: A=0.1sqm, L=0.1m, V=0.1cu.m. F=55xsqrt(0.1/0.1×0.1)=174Hz.
Perforated panel: p=5%, D=6″, PT=0.5″, d=1″, t=(0.5+0.8×1)=1.3. F=200xsqrt(5/6×1.3)=160Hz

Typical Helmholtz Panel Resonance Frequency Response

Similar to panel absorbers all resonant absorbers are essentially narrow band and are not generally suitable for dealing with large numbers of modes unless they are designed to have a wider response (low Q), which means a lower peak absorption coefficient. I do not personally advise this. If you have a large number of modes you should use broad band absorption with resonant absorbers for any problematic stronger standing waves.

The depth of the internal damping material and its spacing from the front panel affects the final performance of the absorber. It is generally better to fill the space keeping it about 1/2″ away from the perforated panel. The available calculators allow you to see the impact of this absorber on the absorbers frequency response for different depths and spacings.

Similar to the resonant panel they take up less depth than porous absorbers for a given frequency. Providing sufficient bass absorption is still a function of area so you need to cover a large area of your room with whatever bass absorber(s) you are using.

Diffuser Design

Although there are many types of diffuser design, in the interests of brevity and simplicity, we shall only review the design of the 1D QRD. If QRD design is new to you please read these technical articles before starting on their design and construction. Effectively defusing frequencies lower than 250HZ in a residential environment is generally not practical due to the size and quantity of diffusers required. A diffuser that can truly diffuse 250Hz would need to be at least 12″ deep.

My A/V Room rear 1D QRD using a Barker Sequence of +1,+1,-1,+1

Unless you are extremeley profficient at math I would strongly advise you to use one of the many free on line programs. These will allow you to calculate all the required dimensions of a QRD (or any other diffuser design). My personnel favorite is QRDude. I used this program to create the QRD’s that I used in my A/V room. Other diffuser types may be calculated from this link.

A diffusers frequency response is different to an absorber. It shows the degree of diffusion with frequency and if relevant, its absorption with frequency.

Image Credit – Vicoustics. PRD Multifuser DC2 – 2 Dimensional Acoustic Performance

The diffusers response often appears as polar plot showing how the sound energy is radiated from the diffusers surface over a 180 degree arc. Keeping the lobbing to a minimum is important in order to maintain a diffuse, uniform energy, sound field.

The following diffusion lobe patterns show how they change based upon the number of panels or periods (N) used. Note that QRD’s work best in correctly designed arrays and that a single panel of any size does not create any significant lobes.

Improvement In Lobes (dark line) By Using The Correct Sequence and Panel count

The way diffusers are designed, their quantity and array sequence affects how the above polar response lobes look, the range of frequencies over which it will effectively diffuse the incident sound, and how close you should sit to it. As described in post 2 these diffusers may also be 2D and QRdude offers you a design for that approach.

Using my design as an example: I required a 1D rear QRD that would diffuse the front L/C/R speakers, allow me to sit about 4’6″ away and have a lower cut of frequency of at least 1000Hz, (HF cut off=6800Hz). I also wanted it to cover the entire wall behind the seating positions to ensure good diffusion. The selected QRD was based upon a 17 well (1″) design that was 5.5″ deep and occupied 6.5 feet. This meant I could install 4 vertical sections and use the preferred Barker sequence of +1,+1,-1, +1 requiring one different set of well dimensions. This would help ensure a smooth dispersion characteristic.

QRDude User Interface. Setup for my 17 well, +1 panel.

Each of the side QRD’s are two stacked pairs of 1D panels. They are based on a 14 well (3/4″) sequence that is three inches deep that have a lower cut off frequency of 2000Hz, HF cut off of 9000Hz and allows you to sit within 1′ 7″ of them. Their function is to diffuse the rear left and right surrounds.

Rear Side 1D QRD Panels

At a later date, after further ETC measurements, I added Vicoustic Multifuser DC2 PRD 2D diffusers to the ceiling above the entire seating area to remove several strong reflections from the six rear speakers.

Vicoustic Multifuser DC2 PRD 2D ceiling Panels

Other Diffuser (and Absorber) Solutions and Services Suppliers

For those of you who are not in a position to build diffusers, or absorbers, here are a few links to established suppliers:

As I built most of my absorbers and diffusers I have only had experience with Vicoustics who supplied my rear ceiling 2D PRD diffusers. They are well built, light, easy to install and certainly did what they are advertised to do.

Other Acoustic Considerations

Absorber/Diffuser Distribution

Using a basic formula like Sabins formulae (see post 2) you can calculate the amount of absorption you need to achieve the decay time you are looking for. Although a good starting point this is only approximate. There are other much more detailed calculators on line that allow more accurate assessment of a rooms absorption requirements. Remember, you can never have too much bass absorption, and I mean never. So the skies the limit on how much LF absorption you can get into your room. When it comes to mid band and HF absorption you need to be more careful as to how much and how it is distributed. You are always wanting to create a diffuse sound field so do not put all your absorption at one point. Make smaller panels and spread them around the room, keeping things generally symmetrical about the MLP. They will be much more effective. I do not personally recommend porous, Helmholtz or diffusers panels smaller than 18″x18″ and membrane panels smaller than 24″x24″ or 18″x36″.

Edge Diffraction

An often overlooked acoustic issue is edge diffraction. When a sound travels around an abrupt corner as opposed to a curve it ‘sprays’ out in all directions creating a secondary point of radiation and multiple paths. This occurs at speaker and grill abrupt edges and any supporting woodwork abrupt edges. The reason why my Genelec 8030s and 8040s have rounded speaker case corners.

Acoustics 101-Designing Porous, Resonant and Reflective Sound Treatments

Edge Diffraction Effects

These effects may be mitigated by rounding the edges of all woodwork immediately around the speaker, including grill structures, mounting speakers flush with the rooms structure and placing large amounts of absorption immediately adjacent to the speaker. In my case there was a significant reduction in many refection levels during the ITD by applying these techniques.

Sound Flanking

Another often forgotten effect is flanking. This effect occurs when a sound enters a physical structure like a wall, is conducted by it, being radiated from the structure as it travels through it. For speakers the sound vibrations from the cabinet enter the structure to which it is mounted in or on and then be heard in another distant space or arrives at the MLP BEFORE the direct sound, and effect sometimes referred to as ‘early sound‘. Sound in solid materials travels MUCH faster than through air. In concrete or wood it travels almost TEN times faster. So for every foot a speakers direct sound travels through air, it travels ten feet in a typical structure. This can result in sound being heard from a radiating surface BEFORE the direct sound arrives at the MLP. Clearly this is to be prevented.

Flanking or Early Sound Path

We see that despite the physical sound path through the wall being more than twice that of the direct sound this flanking sound arrives far earlier than the direct sound and can impact the spatial qualities of the sound image.

Speaker Isolation

Flanking and in particular ‘early sound’ can be prevented by isolating the speaker cabinet from whatever it is mounted on or in. This isolation also reduces or prevents the structure from sympathetic vibrations from the low frequencies generated by LF units and in particular high power subs. The overall effect is to produce less structure vibration and a tighter and cleaner bass together with an improved sound stage (my personnel experience). It is important to note here that the method of isolation is important. Remember Newtons third law “to every action there is an equal and opposite reaction“. So the speakers resilient mount MUST NOT let the speaker move, or it will impact the effective cone excursions as the cabinet tries moves backwards and forwards in opposition to the cones movement. However, it must provide a sufficiently resilient mounting to prevent cabinet borne vibrations from entering the supporting structure. This is especially true for high power subs.

Homemade Speaker Isolation Pad Construction

There are many speaker isolation products available, many of which are quite expensive. Making effective isolators is very simple and requires little practical skills and financial outlay. See above diagrams.

NOTE: Please take great care if mounting speakers above head height, like in studio control rooms. Ensure that there is no possibility of the speaker working itself loose from its containment space and falling on somebody. This type of mounting maybe better left to professionals if you do not feel confident about your DIY skills!

Acoustic Resource Web Sites:

Post 1 – Acoustics 101 – Sound Basics

Post 2 – Acoustics 101 – Absorbers, Diffusers, Reflections, Room Dimensions and Subs

From Vinyl To Plastic

One man's quest for audio perfection

One man's quest for audio perfection

Acoustics 101-Designing Sound Treatments