Building in-wall speakers that sound as good as floor standing ones - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 13 Old 02-26-2006, 04:51 PM - Thread Starter
Newbie
 
madmanaenewman's Avatar
 
Join Date: Dec 2005
Posts: 8
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Although the following post deals more with speaker selection and placement than with building speakers, what I am proposing effectively means that I will have to build at least some of my own speakers to accomplish what I desire. Hence it’s inclusion in this forum.
If I’ve done my research correctly, the following information is accurate. However, I present it as a question because it is based purely on reading-type research and deductive reasoning by a newbie, me. Here goes.
The center speaker is the most important one and should ideally be located with the drivers mounted vertically and the tweeter at ear level when seated. Of course, the center of the screen should ideally be at eye level, which basically puts the speaker right where the screen is. The only way to achieve this involves forward projection onto an acoustically transparent screen, the speaker being located directly behind the screen. Indeed, this is exactly how movie theaters “center†the dialogue on their screens. So logically, those seeking sound perfection in their home theaters have only one option … an in-wall speaker behind a screen. It is a perfect solution … or is it? What other considerations apply?
All three front speakers should be fairly directional and pointed directly at the viewers. The reason for off axis fade (if that’s the right terminology) is that we do not want the sound reflecting off of the side and back walls. This would lose the directionality of the front speakers, and it would interfere with the directionality of the surround speakers. In effect, it would “advertise†to your ear that you are in a room watching a movie, not in the middle of a battle field as the picture on the screen would have you believe. And when the action pans from left to right, you want the associated sound to move with it. This need for directional speakers is good because all drivers are mounted co-planar, which is exactly what a flush-mounted in-wall speaker requires. By contrast, both dipolar and bipolar speakers would need to “push†the screen away from the wall to allow for the multiple mounting planes (thereby wasting valuable room space).
However, it seems generally accepted that in-wall speakers seriously compromise the sound quality. Why is far from clear. I have read what I can on the subject and the only conclusion I can derive is that the area is at best poorly understood even amongst those with a passion for this sort of thing. For example, the A/V salesman I talked to said it was because the ports need lots of space behind them to function, i.e., to allow for free air movement so the drivers meet with less resistance and will therefore be louder and clearer at all frequencies. I have a hard time believing that this explains why a speaker sounds better 2’ from a wall than 6†from a wall. Besides, not all speakers are ported, and sealed speakers are also compromised when placed too close to surrounding objects.
A better explanation, I think, is that sound “bends†sharply and is seriously distorted when it meets the edges of the speaker box, which is why the edges on many high end speakers are seriously rounded. But even with the rounded edges, the bent sound will travel to the side of the speaker. If it meets other objects it will bounce around and soon be directed back at the listener. The waves of the reflected sound may cancel the waves of the direct sound at certain frequencies, and may heighten the sound at other frequencies. Bringing the speakers away from the walls or any other reflective surface would clearly minimize this problem. Great. That explains why it is foolish to place a floor standing speaker in a book-shelf and why book-shelf speakers should be placed flush with the edge of the shelf. It also partially explains why book-shelf speakers do not sound as good as floor standing speakers. But it does nothing to explain why an in-wall speaker suffers. In a flush-mounted system, there are no corners for the sound to bend around. Providing it is mounted on a flat wall well away from other objects, there are no other objects for the sound to bounce off of with unpredictable results.
Another theory is simply that in-wall speakers do not have optimum box volumes and may not even have optimally designed drivers in order to fit into a 2x4 wall cavity. Fair enough, but what if the speaker is mounted through the wall instead of in the wall? Perhaps the other side of the wall is in a closet, or maybe you use the protruding box as a shelf in the adjoining room. You can then build the box to sensible dimensions and shape and use good components. Is there still a problem?
Maybe. One person writes that “the fundamental flaw with most in-wall speakers is that they are installed up high, we listen to them from down below. The wave interference between the woofer and tweeter that conventional crossovers permit will cause deep peaks and dips in the response- resulting in a hollow nasal sound.†He goes on to mention that the new “Joseph Audio Insider in-wall speaker has a patented Infinite Slope crossover so it sounds clear everywhereâ€. More research on these high-end speakers suggests that they sound as good as comparably priced floor standing speakers, suggesting that my post to this point must have merit. In other words, if you can identify the why and eliminate the underlying causes, there is no logical reason why an in-wall speaker can not sound just as good as a floor standing one. Note that you presumably do not need an infinite slope crossover providing the speakers are mounted at ear-level, at least if the author of those remarks is correct in what he says.
And now for my theory. I believe that most in-wall speakers are of poor quality. I also believe that floor-standing speakers do not so much have to be placed well away from walls as they have to be placed in locations that will not lead to problems with reflected sound. For practical purposes, this amounts the same thing, but if one were to recess a floor standing speaker completely into a wall with an opening exactly equal to the speakers dimensions, then that would accomplish the desired result. In fact, as a primary source of wave distortion is caused by the sound bending around the sharp corners of a speaker’s box, eliminating those corners would eliminate the distortion. In theory, a properly built directional flush mounted in wall speaker should actually out-perform a floor standing model, all other things being equal.
Of course, the subject is way too complicated for me and perhaps for anyone to fully understand. A floor standing speaker has the advantage of being able to move it slightly to the left etc until it sounds best, while you simply have to live with wherever you mount your in-wall speaker. Moreover, you can demo various floor standing speakers until you find one that sound good to you in your room, but you will have to live with whatever you decide if you try my approach. Also, I am not sure how much sound comes out of the back of a sealed directional speaker, though I do not see why the sound could not be dealt with using sound-proofing materials around the box. Still, I am guessing that very few people have actually tried what I am suggesting, and, in terms of room layout, it seems ideal. The left and right speakers should, of course, be equal in distance to the viewer, but either you make the walls accordingly or you use floor standing models for those speakers, placing them on either side of the screen.
As for the surrounds, my understanding is that dipolar is the way to go. They have the advantage of being vague in their directionality (giving the allusion of being in a much larger room or even outside) and yet have sufficient directionality to be heard when it is necessary. But that is a whole other topic.
So here is the question. I’ve done my research (just ask my frustrated wife) and little is written that is exactly on topic. Does what I write jive with reality, or have I just fallen into my own line of b*** s***?
madmanaenewman is offline  
Sponsored Links
Advertisement
 
post #2 of 13 Old 02-27-2006, 09:03 PM
AVS Special Member
 
DanFrancis's Avatar
 
Join Date: Aug 2002
Location: Aurora, IL, USA
Posts: 1,845
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 9 Post(s)
Liked: 23
No, you're doing ok with your reasoning- I personally disagree with dipole surrounds, but that's insignificant. As far as in-wall vs freestanding: the major difference between the performance of the two is that the ability to move freestanding speakers in both the x and z axis allows freestanding speakers to image more 3-dimensionally than an in-wall.

If you were to take a pair of high-performance in-walls vs identical freestanding speakers (same drivers, same crossover design, both optimized for their respective placement in the room): you would find that tonally they are identical, the soundstage depth is identical- but with the in-walls it goes from the wall-in, and with the fs it's in front of you, away from the wall. Freestanding speakers also sound more enveloping because they can be aimed to work with room reflections as opposed to an in-wall that cannot be "toed-in".

You're also very correct in stating that MOST in-walls are of poor quality, however that is changing quite rapidly. Companies like Triad, JBL Synthesis, CAT, and now Bay Audio have shown that just because a speaker is located in-wall doesn't mean that it can't be a very high-performance loudspeaker. Although there's a love-it or hate-it connotation to the product CAT builds, they have set the bar high through their on-site engineering program. Until recently, they have been the only manufacturer to actually optimize their loudspeakers for the environment they are in (extensive measurement, crossover modification, and remeasurement). Bay Audio has recently adopted a similar program as well- on their custom theater product line.

Any DIY speaker builder can tell you that performance is wrought through much trial and error, and many reach their goals without the luxury of engineering degrees and the advanced measurement equipment that the manufacturers use. The idea that any speaker will sound the "same" regardless of the room it's in is absolute fallacy, you've touched on that yourself. A perfect example to that effect is Wilson Audio: when a dealer "vowels" the room, they end-up placing the speakers in different locations in every room with relation to the 2 nearest vertical boundaries- a form of optimization. JBL uses extensive equalization in their Synthesis product. CAT uses onsite engineering (actually the engineers that design the drivers, enclosures, and crossovers for the very speakers you purchase). Triad, although they do not do on-site work was really the first manufacturer to say: "our in-walls are the exact same speakers as our freestanding product...just in a not-so-pretty-box" .

I have to thank you, this is perhaps the most thought-provoking thread I've read in a month. I expect this will become a somewhat hotly-debated topic, as I'm sure some of the things I've stated- as well as some of the things you've stated go against the grain of more conventional-speaker-indoctrination.

Dan

Dan Francis
Head of Sales US
C'SEED Entertainment Systems GMBH
www.cseed.tv
df@cseed.tv
DanFrancis is offline  
post #3 of 13 Old 03-01-2006, 08:01 AM
AVS Special Member
 
M NEWMAN's Avatar
 
Join Date: Nov 2000
Location: dallas,tx. usa
Posts: 3,718
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Thought you might like a link to my last project which was specifically high end inwalls. Feel free to add your project to this thread, as I was hoping to get a single thread source going for this topic so it could be beneficial to others wanting this kind of project.
http://www.avsforum.com/avs-vb/showt...ht=DIY+INWALLS

PS: you got a cool last name, man... :D

Mike
M NEWMAN is offline  
post #4 of 13 Old 03-01-2006, 02:54 PM
Advanced Member
 
Kingdaddy's Avatar
 
Join Date: May 2003
Location: Fort Worth Texas
Posts: 877
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Problem with in-walls is the vocal region, they will never sound as natural as free standing speakers can. To prove this idea have a friend move around the room talking (facing you) while you set in the listening position, then ask your friend to put his head close to the wall and listen to the change in the voice, its not pretty.
Kingdaddy is offline  
post #5 of 13 Old 03-01-2006, 03:17 PM
AVS Special Member
 
M NEWMAN's Avatar
 
Join Date: Nov 2000
Location: dallas,tx. usa
Posts: 3,718
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Welllllllllll...that's a bit of an exaggeration isn't it King? :D I mean, after all, we don't normally turn our speakers around and aim them off the wall (with the obvious exception of Bose...which explains why they're so disliked). What you're really alluding to is baffle step compensation, or in other words, how much midbass reinforcement you get being part of the wall, compared to floor standers. I've found that well done inwalls tend to sound just as good in the midrange area. The main thing that inwalls struggle with is low end extension in sealed (or vented) back box enclosures, and imaging soundstage. I will concede that imaging depth and soundstage accuracy can be a problem with typical inwalls (most are only "satisfactory" and cheap inwalls just don't image). This is one of the reasons I chose a psuedo line array for my inwall design in the link above - gives a nice big, wide and deep soundstage, and the speakers "disappear" much like good floorstanders. ;)

Mike
M NEWMAN is offline  
post #6 of 13 Old 03-01-2006, 03:29 PM
Advanced Member
 
Kingdaddy's Avatar
 
Join Date: May 2003
Location: Fort Worth Texas
Posts: 877
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by M NEWMAN
Welllllllllll...that's a bit of an exaggeration isn't it King? :D I mean, after all, we don't normally turn our speakers around and aim them off the wall (with the obvious exception of Bose...which explains why they're so disliked). What you're really alluding to is baffle step compensation, or in other words, how much midbass reinforcement you get being part of the wall, compared to floor standers. I've found that well done inwalls tend to sound just as good in the midrange area. The main thing that inwalls struggle with is low end extension in sealed (or vented) back box enclosures, and imaging soundstage. I will concede that imaging depth and soundstage accuracy can be a problem with typical inwalls (most are only "satisfactory" and cheap inwalls just don't image). This is one of the reasons I chose a psuedo line array for my inwall design in the link above - gives a nice big, wide and deep soundstage, and the speakers "disappear" much like good floorstanders. ;)
You completely misunderstand my post. I mean to place you head as close to the wall as possible Facing the listener while speaking, if you could actually get your head in the wall so only your mouth was flush with the wall the effect would be even more pronounced, I did use the word Facing in my original post.

Try it and see what I mean, the reflection of the mid frequencies changes the tonal character of the voice drastically, no way around it. There is no baffle step for in-walls BTW.
Kingdaddy is offline  
post #7 of 13 Old 03-01-2006, 04:03 PM
AVS Special Member
 
M NEWMAN's Avatar
 
Join Date: Nov 2000
Location: dallas,tx. usa
Posts: 3,718
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Kingdaddy
You completely misunderstand my post. I mean to place you head as close to the wall as possible Facing the listener while speaking, if you could actually get your head in the wall so only your mouth was flush with the wall the effect would be even more pronounced, I did use the word Facing in my original post.
Ok, I misunderstood your implication.

Quote:
Originally Posted by Kingdaddy
Try it and see what I mean, the reflection of the mid frequencies changes the tonal character of the voice drastically, no way around it.
I know...see my comment about midbass (and more specifically, lower midrange) re-inforcement from the wall placement. I'm agreeing with you in part here, but wording it a little differently.


Quote:
Originally Posted by Kingdaddy
There is no baffle step for in-walls BTW.
That's correct, but I meant BSC for floor standing, whereas there isn't baffle step needed for inwalls (I guess I wasn't being clear there). The full re-inforcement of the lower midrange (plus a little extra in some cases) is all already there. Alternatively, in most floor standing designs, you actually have to put in an attenuation (Lpad) circuit to back off the mids and highs (or use slightly less efficient mid/high frequency drivers), to balance the response out with the woofer response.

In the inwall designs that suffer from a little too much midbass/lower midrange re-inforcement, coloring the mids, a simple notch or tank filter in the offending range can easily tame it - basically, the reverse of BSC in floor standers.

Mike
M NEWMAN is offline  
post #8 of 13 Old 03-01-2006, 04:13 PM
Member
 
cobbpa's Avatar
 
Join Date: Jan 2005
Posts: 20
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
M Newman: haven't read all through your thread, but looks slick. Very much what someone in my family may be doing eventually.

As King mentioned, there is no BSC factored into the xovers for inwall speakers (assuming woofers & tweets are of similar efficiency) , so theoretically the overall system can be more efficient. Just a generalization, it depends on speaker selection, but in most cases.
The advantage of no edge diffraction is definitely a plus with these types of speakers, too. Diffraction can cause lobing, so the lack thereof is an obvious plus. One problem that does arise, though, (in my family member's case it will be in the surround speakers) is the possibility of having to fit a speaker's box into very shallow depth. Even using 2 x 6's for framing doesn't allow a deep box at all, especially if the baffle is double thick, as it should be to prevent resonating. The shallow box problems mostly deal with backwaves & other types of resonance. They can be faught with polyfill, bracing, & damping, but to what extent--? I'm not sure.

I don't quite understand the mouth-in-wall argument. It may sound different, but probably just because that's a wall the sound isn't reflecting off of, as it previously was? I dunno, maybe I'm misunderstanding. I'm going to read more through that other thread, looks pretty informative.
cobbpa is offline  
post #9 of 13 Old 03-01-2006, 04:17 PM
AVS Special Member
 
M NEWMAN's Avatar
 
Join Date: Nov 2000
Location: dallas,tx. usa
Posts: 3,718
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
A couple more really outstanding examples of manufacturer made inwalls to check out are Snell James - both are exceptional in tonal balance, dynamics and imaging. They also both use proprietary integrated back boxes & xovers designed specifically for this design too....see a pattern?.... ;)

Mike
M NEWMAN is offline  
post #10 of 13 Old 03-01-2006, 04:21 PM
AVS Special Member
 
M NEWMAN's Avatar
 
Join Date: Nov 2000
Location: dallas,tx. usa
Posts: 3,718
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by cobbpa
M Newman: haven't read all through your thread, but looks slick. Very much what someone in my family may be doing eventually.
Thanks! Hope you enjoy the project info.

Quote:
Originally Posted by cobbpa
As King mentioned, there is no BSC factored into the xovers for inwall speakers (assuming woofers & tweets are of similar efficiency) , so theoretically the overall system can be more efficient.
Yea, you must've been typing your post when I was clarifying what I meant about BSC to King, but see above for what I was talking about. ;)

Mike
M NEWMAN is offline  
post #11 of 13 Old 03-01-2006, 10:28 PM - Thread Starter
Newbie
 
madmanaenewman's Avatar
 
Join Date: Dec 2005
Posts: 8
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I guess its about time I thank everyone for their wonderful responses to my query. Most of the comments I understand, and a few essentially serve as research assignments. I am definitely going to be reading the pre-existing post on quality DIY in-wall speakers ... not sure why I didn't find it in the first place. I believe I will eventually be adding my own project to that post, but it will be a slow process.

Two observations might prove thought provoking. The first comes from now distant memories from my high school and university physics courses in which the study of waves (such as sound waves) provided a significant amount of content. I remembered making waves in water and observing what happened under a variety of situations, such as being directed through a narrow opening, passing by an edge, reflecting off another surface, interacting with other waves, and the like, and then expressing the behavior mathematically. Fascinating stuff, really. At any rate, if memory and logic serves me well, a wave pattern that originates at a small surface with a large surface directly behind it with adjoining surfaces at right angles to these, such as when a speaker is placed up against a wall, is going to behave far differently from a wave pattern that originates at a large smooth surface, such as a speaker perfectly recessed into a wall. The former is going to be riddled with skewed and intersecting waves, while the latter is going to be relatively free of such chaos. Indeed, the wall-recessed speaker should theoretically produce a slightly truer pattern than a floor-standing speaker placed away from the wall, all else being equal. I wish I could go back and try some more experiments, or at least talk to a physics buff. Bottom line, both placing a speaker well away from a wall and perfectly flush-mounting a speaker should both result in similar sound reproduction, while placing a speaker against a wall is bound to lead to disagreeable distortions. Of course, its more complicated than that and ultimately the ear is the best judge.
The second observation is that anything we do is a compromise. Mounting in the wall just behind the screen loses the flexibility of "playing" with the position, but it gains the advantage of perfect theoretical placement. Moreover, I will not need to make the box look pretty, allowing more of the budget for better drivers and x-overs. Of course, no-one will see the awesome front speakers I build, but for me its more about the sound than the box. I am still undecided about the surrounds ... maybe I can show off my woodworking skills with them.

Any way, my basic query is answered well, and I thank you once again.
madmanaenewman is offline  
post #12 of 13 Old 03-02-2006, 07:27 AM
AVS Special Member
 
M NEWMAN's Avatar
 
Join Date: Nov 2000
Location: dallas,tx. usa
Posts: 3,718
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by madmanaenewman

if memory and logic serves me well, a wave pattern that originates at a small surface with a large surface directly behind it with adjoining surfaces at right angles to these, such as when a speaker is placed up against a wall, is going to behave far differently from a wave pattern that originates at a large smooth surface, such as a speaker perfectly recessed into a wall.
Yes, its a difference in boundry re-inforcement. However, let me caution you that boundry reinforcement isn't necessarily the only factor in how the speaker sounds when used as an inwall. Here's 3 quick factors:

1) A floor standing model has typically been optimized for the lack of bass and midbass reinforcement from their small baffle dimensions, so sticking it flush with a wall would yield excessively "muddy" or bassy sound.

2) Drywall has a natural resonance in the midbass and lower midrange frequency band, increasing the likelyhood of "muddy" sound - this is partially what Kingdaddy and I were alluding to. A separate enclosure, somewhat "decoupled" from the wall (not using the drywall for support), will dramatically reduce this problem. This is why I used this design topology in my inwalls.

3) Infinite baffle mounting inwalls (no backboxes - mounted with dogleg clamps like 90% of the inwalls on the market) yields no control over the bass response and also are extremely subject to the aforementioned resonance problems. This is why most inwalls don't reproduce a very good soundstage, because you can
point right to 'em when they're playing. They also lack rigidity in the front baffle, creating even MORE midrange resonance problems - the baffle is stiffer, raising the natural resonance into the upper midrange, but not stiff enough to quell resonance like they need to. You'll notice that the really good inwalls that can rival a floorstander's performance typically have integrated enclosures made from wood, metal or at least ABS plastic, which yield improved results in a predictably lessening amounts, respectively.

Mike
M NEWMAN is offline  
post #13 of 13 Old 03-02-2006, 01:57 PM
Advanced Member
 
hawkeye3.1's Avatar
 
Join Date: Feb 2005
Location: Southern NH
Posts: 632
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Just to throw this out there, I built low profile MDF enclosures for my inwalls (ended up being 5" thick) and made onwalls out of them. May not be the most elegant solution but I am quite satisfied with them. As for dialog lift, I have a Yamaha Rcvr with Presence channels fed to a decent set of satellites mounted at the top corners of the screen. Very effective in centering voice audio. Some .02 for you to consider.
hawkeye3.1 is offline  
Closed Thread DIY Speakers and Subs

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off