Welcome to the Acoustics Page! 

HOME

Room Acoustics
Reflections
Diffusion
Standing Waves
Room Gain & Corner Loading
Subwoofer Placement
Optimum Dimensions
Homemade Traps and Diffusers

SoundProofing
Sound Transmission
Mechanical Sound Transmission Construction Tips
Acoustical Sound Transmission Construction Tips
HVAC Construction Tips


Room Acoustics
Reflections and standing waves are the dominant problems that destroy a natural sounding 3-dimensional image. 

Reflections

Acoustic absorbers (acoustic foam, drapes, heavily upholstered furniture, etc...) will help control reflections.  However, if we eliminate every reflection, we would essentially have an anechoic chamber which would not sound natural.  A certain amount of reflections are still acceptable for the room to sound "normal."  We primarily want to eliminate the initial early reflections.  Acoustic foam tiles, drapes and tapestries may be placed behind the speakers, behind the listening area, on the side walls and on the ceiling.  A carpeted room goes a long way toward controlling reflections.  Don't try to cover whole walls, etc...  Instead, strategically place individual tiles at points that reflect toward the listening area.  An easy way to visualize this is to have an assistant place a mirror on the side walls (ceiling, back wall, etc...), and move it around until you see the speaker's reflection.  Place acoustic absorbers at those locations.  Don't forget about ceiling reflections from the left and right speakers AND especially the center channel.  Sparsely place individual absorbers at various other locations.

Return

Diffusion

As stated before, too many absorbers will begin to make the room sound unnatural.  To avoid this, a better solution is to use diffusion to spread the reflection around rather than absorbing it.  With diffusion, the waves are still reflected, but instead of a clear reflection of parallel (in-sync) waves, the reflections are spread out in different directions so that there is no direct reflection back toward the listener.  Diffusers can be purchased, but it is much easier (and aesthetically acceptable) to use objects in the room as diffusers.  Sculptures, furniture, wall mounted knick-knack shelves and bookshelves (with randomly spaced book groupings as opposed to completely filled) make excellent diffusers.

Return

Standing Waves

Standing waves are a much more difficult problem.  These are waves which bounce back and forth between parallel walls.  They set up a resonance such that the bass wave appears to be "frozen" in space, and you can walk into an area of a crest which will sound boomy, or you can walk into an area of a rarefaction (sag in the wave) which will sound as though the bass disappears.  Standing waves occur at a frequency equal to twice the distance between the walls (a trip from one wall to the other and then back) divided into 1130 ft/sec. (the speed of sound).  We can try to build rooms to specific ratios, build non- parallel walls, and partially break rooms say with a staircase or partial wall offset from the middle of the room.  However, studies are revealing that so-called ideal room ratios are not much better at controlling the problem than other ratios except for the notorious square room to be avoided at all costs.  It usually isn't possible or aesthetically pleasing to build non-parallel walls.  Equalizers aren't much help as other frequencies are usually affected too much, unless it's a parametric equalizer.  Some audiophiles will place their speakers at third points and the listening chair at the other third point in the room.  While this isn't possible in most rooms, try to keep the listening position away from the rear wall by at least a few feet.  One way to help alleviate room problems is to use satellite/subwoofer systems.  The main speakers (satellites) can be placed away from walls while the subwoofer can be optimally located for bass and control of standing waves.  Try to keep the crossover as low as possible.  40 Hz to 60 Hz is ideal with 80 Hz still being a reasonable compromise especially if crossed over with high slopes (18 dB to 24 dB/octave).  As a side note, subwoofer placement is easier if you place the sub at the listening position (in the listening chair).  Now crawl around the room and listen for the most optimum bass - a compromise of smoothness and deepness.  Various types of bass traps can help (see below).

Return

Room Gain & Corner Loading

Many people still have a misunderstanding of these phenomenon.  Corner loading is actually related to baffle step diffraction in that bass energy is redistributed.   Whereas in baffle step, the bass rolls off by 6 dB because it "wraps around" the speaker enclosure instead of reflecting forward with the treble energy, room gain and corner loading in particular increase the bass because it is being "funneled" into a smaller angle of dispersion.  Picture a speaker out in the open away from any surfaces (walls, ceiling or floor).  Assume the driver is capable of radiating 360 at all frequencies.  Technically, this driver will have a certain SPL at 1 watt.  Because it is on a baffle, at some point as the frequency increases (roughly the frequency whose wavelength is equal to the baffle width) it will start radiating the rear energy forward, because it is "reflecting" off the baffle.  Let me rephrase that - 180 of the radiated sound energy is now being radiated forward and in-phase with the other 180.  The SPL increases by 6 dB as the frequency gets higher.  This is baffle step.  Actually, the SPL of a loudspeaker assumes sound is radiated forward into half-space.  Therefore, to be more accurate, it is better to say that sound decreases 6 dB with decreasing frequency as bass radiates into a full 360.  When sound is radiating 360 it is said to be radiating into full (spherical) space (4π steradians).  When you place this speaker against a boundary, say the floor, you are now radiating into a hemisphere or half space (2π steradians) and the SPL increases by 6 dB.  When this speaker is placed at the junction of the floor/wall, it is radiating into a quarter sphere or quarter space (π steradians) and the SPL increases by 6 dB.  When you place the speaker in a corner, it is now playing into one eighth space (π/2 steradians) and again, the SPL increases by 6 dB.  This means a theoretical increase of bass on the order of 12 dB (from half space to quarter space, then to one eighth space) if the speaker is placed in a corner.  In reality, rooms have doorways and windows and not totally rigid walls and speakers that are generally a few feet away from the walls.  All these factors allow for some loss of bass so that corner loading usually contributes around 7 to 9 dB of gain in the bass frequencies beginning around 50 Hz or so for small rooms down to maybe 20 Hz or so for large rooms.

Technically, true room gain (a 12 dB/octave rise as frequency is decreased) is a pressurization effect that kicks in at 565/L Hz, where L is the longest room dimension (actually the lowest standing wave).  It's effects are probably not developed to any great extent in most rooms (except for cars), but this and corner loading act together to extend bass response for about an octave.  When speaking about room gain, I think most people are talking about both effects combined.

Room gain (including corner loading) is generally beneficial.  It allows us to have truly deep bass by using speakers with a modest F3.  For subs, if you design a sealed sub that has an F3 near 30 Hz, you will have in room response to below 20 Hz.  If you design a speaker to have very low F3 (say 20 Hz), then room gain and corner loading will kick in to make your sub boomy at around 30 or 40 Hz.  Some may like the boomy effect, but it is not accurate.  It is best to take into account room gain and corner loading if you design for deep bass.  Don't be overly preoccupied with formulas for affected frequencies.  Just be aware of these effects and choose an appropriate design.  Generally we are talking about frequencies of around 30 to 50 Hz, so I tend to use the following rules of thumb:


Desired F3
Sealed Speakers
Vented Speakers
Above 40 Hz
Design for a Qts=0.7
Design for a flat QB3 alignment
Below 30 Hz
Design for a Qts<0.6
Design for an EBS alignment

Between 30 Hz and 40 Hz gets trickier, but either alternate should work fine.  EBS would be a good choice here.  By the way, the room gain/corner loading effects want to rise at 12 dB/octave (even though it will only amount to about 9 dB total), which is the rate of bass roll-off in sealed enclosures, so sealed enclosures with an F3 near the frequency where the gain kicks in will generally yield the smoothest bass with the drooping response of an EBS (extended bass shelf) alignment coming in a close second.

Return

Subwoofer Placement

Another topic worth mentioning is speaker placement.  While nothing will cover all rooms, there are some general rules that should at least be good starting points.  Generally, speakers should be placed away from walls for best imaging, yet closer to the walls for deeper bass output.  This is definitely a compromise, and many speakers may also have specific placement recommendations, so I will concentrate on subwoofers.  As mentioned previously, satellite/subwoofer systems overcome these compromises to some extent because the satellites can be placed on stands away from the walls for best imaging, and the sub can be placed at a location which gives optimum bass.  And, again, as previously noted, the sub can be placed at the listening position (in the listening chair), and you can crawl around the room and listen for the most optimum bass.  If there was one general statement about subwoofer placement, it would be that subs generally sound best located in or near a solid corner of the room - one with no nearby windows or doors or other openings.  We want the corner to reinforce the bass.  If you have two subs, it is still generally best to set both in one corner for optimum output as long as they are crossed low enough (less than 80 Hz).  Some recommend stereo placement or non symmetrical placement to smooth the bass and to have stereo bass.  I agree with the stereo bass if the crossover is relatively high.  Some say they can hear the difference with stereo bass even with low frequencies, however, I disagree with this.  Every study I can think of (which, admittedly, is not very many) have shown that low bass is non directional and indeterminate unless the distances are comparable to the wavelengths being reproduced.  I think what most people hear is either higher frequencies of the initial transient (reproduced by the satellites), distortion of their stereo subwoofers giving away their placement or the effects of a room null which is overcome by separating the subs.  I think the last choice is the most likely scenario.  When switching between stereo subs and one mono sub, the location of peaks and nulls in the room changes because the location of the bass source changes.  However, separating the subs will generally result in a more ragged bass response rather than a smoother response as is claimed - not always, but usually.  But all rooms are different, so experiment. 

Return

Optimum Dimensions

There have been many optimum dimensions recommended for rooms.  As mentioned above, studies are showing that optimum dimensions are not the cure all for controlling acoustic problems, though the square room should be avoided (8' W x 8' L x 8' H).  However, as with anything else DIY, every little bit helps. Here are three recommendations gleaned from various sources: 


Ratio #1
Ratio #2 Ratio #3
Height
H
H
H
Width
1.14xH 1.28xH
1.60xH
Length
1.39xH
1.54xH
2.33xH

For normal 8' ceilings:


Ratio #1 Ratio #2 Ratio #3
Height
8'
8'
8'
Width 9'
10.25'
12.8'
Length 11'
12.3'
18.64'

For 10' ceilings:


Ratio #1 Ratio #2 Ratio #3
Height 10'
10'
10'
Width 11.4'
12.8'
16'
Length 13.9'
15.4'
23.3'

Ratio #3 seems to be one of the better ratios.  I have often seen people with floor dimensions of 16'x20' or 16'x24'.  I have also seen people recommend using different prime numbers as recommended ratios.


Return

Homemade Traps and Diffusers

(Soon!)

Return

SoundProofing
 
Sound Transmission

Sound transmission occurs mainly through direct acoustical transmission or through mechanical transmission.  Acoustical transmission is basically the direct sound wave.  If there is an unbroken path through the air to your ears, you will hear it.  Even though you close a door, sound still travels under the door or around the edges through the air.  Mechanical transmission occurs when the sound wave in the air meets a solid structure and vibrates the structure.  This occurs at walls.  The sound vibrates the wall which then vibrates the studs, which then vibrates the outer wall.   The outer wall will act like a transducer and transmit the sound.  Following are some home construction tips to help control sound transmission, but first, some definitions.

STC (Sound Transmission Class) quantifies a material's effectiveness at blocking the transmission of sound.  Expressed in dB, it generally applies to hard materials.  The higher the number, the better it blocks sound.  Note that 3 dB is barely perceptible, but 10 dB is twice as loud (or half as loud as the case may be).  This means an STC 50 wall is four times quieter than an STC 30 wall.  This is the most common way walls are rated.

NRC (Noise Reduction Coefficient) tells us how much airborne sound a material absorbs. Expressed as a decimal, it generally applies to soft materials.  The higher the number, the better the material is at absorbing sound.

Normal conversation can be heard and understood through a wall of STC 25 - basic wood stud wall, no insulation or caulking.
Loud talking not heard through a wall of STC 50 - achievable with reasonable care and special construction techniques.
Loud shouting can be heard, but not understood through a wall of STC 60 - very difficult to achieve, but possible with care in installation.

Recommended wall STCs:
bedroom to bedroom - STC 48
bedroom to adjacent kitchen - STC 52 (STC 58 would be optimum).

Mass and dead air space are the most important things for stopping the transmission of sound from one place to another. 

Note that breaking dead air space into (more) smaller spaces may actually make the noise transmission worse.  In walls, the transmission loss depends on mass (and stiffness) of the (outer) surfaces and on the thickness of the airspace between them.  Mass and dead air are your friends!!!

Return

Mechanical Sound Transmission Construction Tips

The best method of mechanical isolation is to build two walls adjacent to each other with a dead air space between them, and no cutouts (electrical boxes, etc...) in the wall.  Period.  However, since this is impractical for most home owners, two other methods are nearly as good (essentially the same).  One method involves building walls using 2x4's on 6" top and bottom plates.  The other method uses resilient channel under the drywall to break the mechanical connection to the stud.  Briefly I'd like to mention that using double or triple drywall layers or thicker drywall (increasing wall mass) can also help control sound, but the following methods offer the most protection from sound transmission.

Staggered studs  Staggered Studs

Staggered stud construction requires building 2x4 walls on 2x6 top and bottom plates (Click picture above).  Essentially, two walls are built instead of one, but sharing the same 6" wide top and bottom plate.  Each wall's studs are offset by 8".  This breaks the connection between the two walls.  Even though they share the same plates, there's no real connection of the vibrating surfaces (the walls are pinned at the top and bottom which is not as critical a connection).  In the picture, the tan studs connect to drywall on one side of the wall, and the blue studs connect to the drywall on the other side, but neither set of studs touches the other's drywall.  The primary drawback is with respect to windows and doors.  Normal walls are about 4 1/4" wide (3 1/2" stud plus two sheets of 3/8" drywall).  The 6" top and bottom plates increase the thickness to about 6 1/4".  This precludes normal door jambs and window jambs.  Basically an approximate 2" strip would need to be added to the jamb before the casing (trim) is applied.  However, bare walls (ones with no windows or doors or other openings) should present no problem.  There's just the cost of the second set of studs.  For new construction, this is an excellent way of soundproofing a particular room (say a home theater room) and isolating it from the rest of the house.

Resilient Channel  Resilient Channel

Resilient channel (click picture above) is another way to control sound transmission - by breaking the sound path from drywall to stud.  It acts like a shock absorber.  It is a metal strip attached to the studs with the "nailer tab" side down.  This is important!  The small "nailer tab" should be down so that the weight of the wall "floats" away from the studs.  The drywall is then screwed to the "wide" flange of the strip rather than directly to the studs.  This greatly reduces sound transmission.  Resilient channel only needs to be applied to one side of the wall.  Be careful not to screw drywall where there's a stud.  Otherwise the screw may go into the stud and rigidly attach the drywall to it.  You may use screws to attach the channel to the stud.  Just make sure not to accidentally screw the drywall through the channel to the stud.  Again, the main drawback is the extra wall thickness that may affect window and door jambs.  Resilient channel will add about 1/2" to the wall thickness.  Also, do not place it closer than about 6" to the floor or ceiling.  At the floor, you could nail a thin strip of drywall to the studs so that accidental kicks won't crush the wall here or leave a hole.  Or better yet, something like Celotex (the black fiberboard you see on houses as they're constructed) cut into strips.  Remember to caulk the small gap between the floor and the bottom of the drywall.  I'm not sure about the best way to attach baseboards - perhaps with glue, or else don't nail it too often.  Or perhaps you could use vinyl cove base.  Leave about a 1/4" gap at intersecting walls, and caulk it.  I'm sure drywall mud and tape is fine, but better safe than sorry.  Then tape and mud it as normal.  Perhaps do the same thing at the ceiling/wall juncture.  Seal the gap between the floor and bottom of the drywall with flexible caulking.  When finished, the walls should flex slightly when pushed.

Resilient channel can also be applied to ceilings.  Make sure all flanges point in the same direction.  Personally, I have my home theater area directly below my bedroom.  I had planned on using insulation and 1/4" drywall on the ceiling, and then adding resilient channel, and then hanging 1/2" drywall off the resilient channel.  But as I found out, I may want to use resilient channel directly on the joists, with maybe 5/8" drywall (maybe even double 5/8", 1/2" or 3/8" drywall layers).  Apparently the most likely reason you don't want to add resilient channel on top of drywall is because the screws holding the new drywall will touch the underlying layer and transmit vibration to it.

Wall Configurations
STCs of various types of walls:

Here is an interesting link:
STCratings.com
I notice some of their numbers are slightly lower than mine.  Perhaps they didn't take caulking into account.

The following table attempts to show how much soundproofing is achieved with various techniques:

Technique
STC
Caulking 5
Insulation 3-4*
Double Drywall
2-3
Metal Studs
10-13
Resilient Channel 7-13
Staggered Studs 12-13
Two Walls 20-22
* - Attributes a bit more STC (up to 8) when walls are already isolated well (read, "when used with resilient channel, staggered studs or double walls")
Note that resilient channel adds closer to 12 or 13 dB to the STC if insulation is used, but perhaps closer to 7 dB when it is not.
Also, caulking implies that ALL gaps and cracks in the wall are sealed including the bottom of the wall, outlet boxes and switch boxes including where the wires enter.

Good combinations would be:


Resilient channel can also be applied to ceilings.  Personally, I have my home theater area directly below my bedroom.  I had planned on using insulation and 1/4" drywall on the ceiling, and then adding resilient channel, and then hanging 1/2" drywall off the resilient channel.  But as I found out, I may want to use resilient channel directly on the joists, with maybe 5/8" drywall (maybe even double 5/8", 1/2" or 3/8" drywall layers).  Of course I could use 1/4" drywall, then use furrowing strips (or maybe even Celotex).  I could then attach the resilient channel to the furrowing strips.  Hmmm, that's an interesting idea.  I may consider it.

Return

Acoustical Sound Transmission Construction Tips

To stop airborne acoustical transmission, picture the room as a giant "fish tank" and that you are attempting to seal all leaks.  Doors, windows, outlets and vents are prime suspects.  Use caulking to seal around electrical outlets and switches.  Also seal the openings where the wires come in.  Turn off the breaker to those outlets and switches while you seal them.  Another thing to be careful of is to not let outlets on opposite sides of the wall share the same wall cavity space.  And seal the holes in the studs where the wire passes from one box to another (or any other holes in the studs)!  Note that this primarily applies to "single" walls with drywall mounted directly to the studs.  If you are using resilient channel, staggered studs or double walls, you already have large gaps. :-)  Use heavy, solid core wood doors.  Remember, mass is a way to fight sound transmission.  Use weather-stripping around the edges of the door, and a sweep seal underneath though HVAC concerns will need to be addressed.  Better yet, buy a door system designed to prevent sound transmission.  It will be built similar to an exterior door with a threshold seal.  Heck, an exterior type door would probably work assuming it is solid core.  Remember, an air return in the room is essential if you totally seal it.  Insulation in the walls will also help, though not nearly as much as resilient channel or staggered stud construction.  Insulation adds about 3 to 6 dB.  Note that "special acoustic fiberglass" insulation may not be much better than plain old pink Owens-Corning.  Multiple glazed windows will help with sound intrusion from outside the home.  Make sure the glass panes are as far apart as possible.

IMPORTANT!!!  Noise leaks through the weakest links in walls - through (or around gaps in) a door or outlet. Don't spend money or effort improving walls until all these weak links are controlled!!!  Other tips:


Return

HVAC Construction Tips

Heating and cooling ducts present there own problems.  If you totally seal a room, there has to be a way to circulate air.  Therefore the room needs its own air return.  Better yet, use a totally separate heating/cooling system to the home theater room :-)  Metal ductwork is notorious for transmitting sound.  If possible, use flexible ducts.  You may want to keep the soundproofed room on its own trunk lines as other rooms branched off of the same trunk line may transmit sound to or from the sound controlled room.  Or at least only attach rooms on the same trunk line that are not critical to the sound controlled room.  If privacy is the issue with the sound controlled room, only attach rooms on the same trunk line that are not "public" where other workers or the public might hear.  If it is a home theater and you wish to keep loud movie sound out of the rest of the house, don't share lines with bedrooms as it will interfere with sleeping.  As with electrical outlets, be aware that ductwork for adjacent rooms don't share the same wall space.  Air returns for adjacent rooms sharing the same wall cavity allow direct transfer of sound.  There are so-called "duct mufflers" to trap sound from traveling through ducts, but they are rather expensive.  They appear to be roughly 2' or 3' insulated "boxes" that install somewhere along the room's air duct in order to trap sound similar to a muffler on a car.


Return

HOME

TOP OF PAGE
 

Last Updated 08/20/11