Old thread, but... whatever.
The bottom line question is this: Why are "integral enclosures" essential to in-wall speakers?Background:
A speaker driver moves forward and backward to generate air pressure in front of, and behind, the driver. These pressurizations and rarefactions of the air then propagate as soundwaves. The front side of the driver generates forward propagating waves; the back side of the driver generates rearward propagating waves. At high frequencies, the soundwaves propagate primarily forward/backward, almost "beaming" these frequencies. However, as frequencies drop, the soundwaves spread out their dispersion, and at bass frequencies the dispersion is virtually omnidirectional. At low frequencies, the front wave from the driver can "wrap around" the driver and virtually cancel the back wave. A driver in free air, (with no enclosure around it), will sound like it's all treble, with almost no bass. This is because the bass waves are cancelled by this "wrap-around" phenomenon. A "baffle" is needed to separate the front wave from the back wave so they can propagate independently and without interference.
An "infinite baffle" is an acoustically non-porous divider that is infinitely wide and infinitely tall. It completely isolates the front wave from the rear wave. Each wave can propagate into it's own space free of interference from the opposing wave. They'll propagate 180 degrees out of phase with each other, but it won't matter because they'll be in completely separate spaces. Of course, nothing can be infinitely wide or tall, but as long as the baffle is larger than 1 wavelength of the lowest frequency of interest, it will effectively block the interference of the front wave and the back wave.
A sealed "enclosure" also separates the front wave from the back wave and allows the front wave to propagate without interference from the back wave. In a sealed enclosure, the air molecules excited by the front wave are isolated from the air molecules excited by the back wave, at least to the extent that the enclosure material is non-porous to sound transmission. However, the inside of the sealed enclosure adds a "spring" to the driver. As the driver moves inwards, it builds positive pressure inside the enclosure pushing back on the driver. As the driver moves outwards, pressure inside the box becomes negative and it "pulls" the driver back inwards. These positive and negative pressures inside the box "load" the driver, and these pressures need to be overcome with extra amplifier power and driver control.
(I am not going to address ported enclosures here because no one that I'm aware of is building ported in-wall speakers. AFAIK, all in-wall designs are either infinite baffle or sealed designs. Suffice it to say that ported/vented designs are a special case between sealed and infinite baffle.) End Background
So, getting back to the original question: Why are integral enclosures essential to in-wall speakers?
First, and most importantly, a single layer of 1/2" drywall is porous to sound. Please study the following graphs:
We see that the Sound Transmission Class, (STC) of drywall is only "fair" at midrange frequencies, but as the frequency falls, so does the STC. At low frequencies, the STC is quite low. This means sound will "bleed" into adjacent spaces. It also means that the front wave and the back wave will "bleed through" the baffle drywall to interfere with each other. This will cause some cancellation. At frequencies below 100 Hz or so, the response won't be as bad as the "free air" response, but it won't be much better than that either. If the wall is made of dual layers of 5/8" drywall the performance improves significantly, but hardly anyone does that for in-wall speakers. It's done often for sound isolation, but not to improve the sound quality for in-wall speakers.
The other significant issue is the "loading" of the drivers. Different drivers perform differently with different "acoustic loads." The Theil-Small, (TS), parameters determine how well a given driver will perform with different loads. http://en.wikipedia.org/wiki/Thiele/Small
A speaker designer should be accounting for the specific TS parameters of the specific drivers he/she uses in his/her design. That is virtually impossible to do without knowing the "load" on the drivers. With an unenclosed in-wall it is impossible to know the load on the drivers. The designer can't know the volume of the enclosure or how well it's sealed or damped. The "enclosure" could be an interior wall using 1/2" drywall attached to wood studs with drywall screws and with no insulation. The "chamber" could be a floor to ceiling space, wide open, with no insulation, between 2 studs. Or, the "enclosure" could be in an exterior, insulated wall, using 5/8" drywall in the "cripple" space above a window, with the drywall "glued and screwed" to the studs. Those 2 "enclosures" would present very different, volumes, dampening and sealing, which will mean very different backloads to the drivers. How an in-wall speaker designer can design a speaker that accounts for these different installation possibilities is beyond me.
Other important issues are sound "flanking," sound transmission to A/C duct work, and wall vibrations. Only audible wall vibrations will affect in-room sound quality, but the other two certainly affect user satisfaction.
OTOH, an in-wall speaker with an "integral enclosure" will allow all the above issues to be accounted for in the design. The designer can predict the volume, dampening and sealing of the enclosure and can compensate for all those things with driver choice. In addition, with this level of predictability, other issues can be addressed with crossover design without regard to things like baffle step compensation or diffraction compensation. Finally, there should be no sound "flanking," no transmission to A/C ductwork and no wall vibrations.
Bottom line... an in-wall speaker with an integral enclosure has the *potential* to be a better speaker than an open-back in-wall. Of course, the final result will be a function of the skill of the speakers designer AND the speaker installer. It's certainly possible that a skilled and experienced installer could optimize an unenclosed in-wall, and get an excellent result. However, if I was a speaker designer, I would not want to put the excellence of my design into the hands of some unknown installer. I would prefer to make the design choices myself and take the installer and the installation out of the equation, as least as much as possible.