Maximizing effects of acoustic diffusers

[Boomzilla]

I own two 2x4 foot x 8 inch-deep acoustic diffusers made of plywood. They are from ATS acoustics if I remember right. They are "quadratic" diffusers, intended to diffuse multiple frequencies.

I initially placed them behind the listening couch (two feet behind the listening position, in front of the back wall). I was not noticing a difference. I replaced them with absorbers, that seemed to help some.

I moved the diffusers to the first-reflection points on the side walls. Again, no difference (the side walls are about 13 feet to either side of the listening position). I think that the drapes, furniture, etc. combined with the distance (inverse-square law) made the diffusers surperfluous.

As I was shuffling furniture, I chanced to place the diffusers at the front corners of the listening couch, and at right angles to the couch and instantly the center image (in particular) became more focused. I then tried the diffusers at right angles to the front wall both inboard and outboard of the speakers. Again, improvements in the center image.

I eventually ended up with the diffusers parallel to the front wall and behind the speakers. Just as good an effect and without any intrusion by the diffusers into the room.

And, ever unable to leave well-enough alone, I'm considering trying the diffusers on the ceiling. My room is very live with hard floor, sheetrock walls, and a nine foot sheetrock ceiling. Were I planning to stay in this house for more than another year or two, I'd carpet the floor and install acoustic tiles on the ceiling. But there's no reason to invest in anything just now that I can't get my investment back on when I sell the house.

So it seems to me that with a rug on the floor and diffusers at the first reflection point off the ceiling, I might maximize the effectiveness of the diffusers while having a "removable" and temporary addition to the living room. Is this a sound theory (no pun intended)? Would I do better putting absorbers on the ceiling and the diffusers elsewhere?

Advice appreciated.

Thanks - Boomzilla

[ttocs]

Sept 17, 2021 16:51:19 GMT -5 Boomzilla said:

And, ever unable to leave well-enough alone, . . . . .

Thanks - Boomzilla

Some of us just can't. Curiosity ensues. But it's all good fun.

[marcl]

You can't have too many bass traps. And you can't have too many diffusers!

[DYohn]

acousticgeometry.com/how-sound-diffusers-work/

[Boomzilla]

Sept 17, 2021 18:40:23 GMT -5 DYohn said:

acousticgeometry.com/how-sound-diffusers-work/

Thanks, but I found and digested that exact article before posting here. Would a first-reflection ceiling point be different from a first-reflection side-wall point?

Due to the peculiarities of my room the biggest untreated surface IS the ceiling. But would an absorber or a diffuser be the most effective there (and why)?

Thanks - Boom

[DYohn]

Sept 17, 2021 21:29:44 GMT -5 Boomzilla said:

Sept 17, 2021 18:40:23 GMT -5 DYohn said:

acousticgeometry.com/how-sound-diffusers-work/

Thanks, but I found and digested that exact article before posting here. Would a first-reflection ceiling point be different from a first-reflection side-wall point?

Due to the peculiarities of my room the biggest untreated surface IS the ceiling. But would an absorber or a diffuser be the most effective there (and why)?

Thanks - Boom

The ceiling is closer to the listening point, yes? Diffusers work across a broader frequency spectrum. I believe I recommended hanging from your ceiling a couple years ago.

[Boomzilla]

Thanks, DYohn. While I have your attention, may I also enquire about port plugging? As someone with extensive experience with car audio, I suspect you’re no stranger to Misters Thiel & Small…

My speakers are what is commonly described as 2.5-ways. There are two woofers, the bottommost of which is tuned to begin rolling off slowly above 250Hz. The topmost woofer also produces sound down to the cabinet’s roll off (-6dB @ 30 Hz.) but is “optimized for the midrange.”

Both cones have their own internal enclosure and are individually ported. I’m using an external 12dB / octave high pass filter and a sub, so neither driver is required to operate at any significant excursion. On some material, the speakers can sound as if vocalists are slightly recessed in the soundstage. I understand that this can be slightly helped in some speakers by partially or completely obstructing the ports. The trade-off is that lower bass response is lost (not a concern since I’m using a sub).

What’s your take on port manipulation?

Thanks - Boom

[novisnick]

Sept 17, 2021 21:57:20 GMT -5 DYohn said:

Sept 17, 2021 21:29:44 GMT -5 Boomzilla said:

Thanks, but I found and digested that exact article before posting here. Would a first-reflection ceiling point be different from a first-reflection side-wall point?

Due to the peculiarities of my room the biggest untreated surface IS the ceiling. But would an absorber or a diffuser be the most effective there (and why)?

Thanks - Boom

The ceiling is closer to the listening point, yes? Diffusers work across a broader frequency spectrum. I believe I recommended hanging from your ceiling a couple years ago.

Quick question if you don’t mind me asking. Aren’t diffusers made to be used with specific problems and frequencies? Please forgive my question if its too rudimentary. 

[novisnick]

Sept 17, 2021 17:48:27 GMT -5 marcl said:

You can't have too many bass traps. And you can't have too many diffusers!


View Attachment

An incredible mix you have there! Did you build your diffusers or spend a boatload of cash for them? Love to build some like i did for my absorbers if a purchase would sink my boat! LOL

[Boomzilla]

novisnick -

Diffusers can be tuned to a specific frequency range (these are typically hemispherical), or can be designed to affect a wide frequency range (typically made with slats of different widths, and called "quadratic" diffusers).

Hemispherical diffuser:

Quadratic diffusers:

[marcl]

Sept 17, 2021 23:09:40 GMT -5 novisnick said:

Sept 17, 2021 17:48:27 GMT -5 marcl said:

You can't have too many bass traps. And you can't have too many diffusers!


View Attachment

An incredible mix you have there! Did you build your diffusers or spend a boatload of cash for them? Love to build some like i did for my absorbers if a purchase would sink my boat! LOL

I thought about building them but was working at the time and I don't have any DIY space, so yeah they were expensive but I did manage to find some B-stock and sale prices.  30 Vicoustic Multifuser DC2 (five boxes of 6).  Not in this photo, there are six white in a rectangle on the ceiling over and slightly behind the MLP and two each side left and right and slightly behind the MLP.  Some of the gray ones I got a deal on white and painted them with water-based paint that matched perfectly.  There are also four 1D GIK GridFusors (~7") center of the rear wall.  All the diffusers are EPS, no wood or hard plastic, though the Vicoustic have a smoother surface and denser composition than the GIK. 

I had experimented originally with some thin (3"?) 1D diffusers on the front wall and was sufficiently convinced that they helped but were not enough (but I have a couple around the room mounted over bass traps to reduce mid-high absorption). There was a slight but measurable lower frequency absorption with the Vicoustic when I first used them on the front wall, so I expect they help a tiny bit with nulls between 100-300Hz.  I tried the 3" and 7" overhead before I found the white Vicoustic.  Overall I found the Vicoustic had a (very subjectively) better and more uniform effect, less obvious in higher frequencies. 

I had the left and right side diffusers more forward exactly next to the MLP, and I liked them there because they made the wall disappear ... but I had to move the surrounds forward and there wasn't room, so I moved the diffusers back a couple feet.  The interesting consequence was an obvious change in right side imaging which I tracked down with measurements to be due to right side wall reflection.  So I added absorption on the right wall in front of the surrounds and that fixed it.

As for the choice of 2D "skyline" type vs the large wooden 1D ... for my small-ish room I thought the more random scatter off the ceiling and other surfaces the better, and I'll admit to liking the aesthetics. 

[DYohn]

Sept 17, 2021 22:20:31 GMT -5 Boomzilla said:

Thanks, DYohn. While I have your attention, may I also enquire about port plugging? As someone with extensive experience with car audio, I suspect you’re no stranger to Misters Thiel & Small…

My speakers are what is commonly described as 2.5-ways. There are two woofers, the bottommost of which is tuned to begin rolling off slowly above 250Hz. The topmost woofer also produces sound down to the cabinet’s roll off (-6dB @ 30 Hz.) but is “optimized for the midrange.”

Both cones have their own internal enclosure and are individually ported. I’m using an external 12dB / octave high pass filter and a sub, so neither driver is required to operate at any significant excursion. On some material, the speakers can sound as if vocalists are slightly recessed in the soundstage. I understand that this can be slightly helped in some speakers by partially or completely obstructing the ports. The trade-off is that lower bass response is lost (not a concern since I’m using a sub).

What’s your take on port manipulation?

Thanks - Boom

T/S parameters are not just for car audio, of course. re: port plugging. What you accomplish by doing this is you are changing the enclosure alignment from what was designed, which changes several system parameters including system Q factors, FR and power handling.
1. You convert the enclosure from bass-reflex to "sealed" infinite baffle. The enclosure will likely be too large to consider it operating as acoustic suspension. This may cause bass output to sound more "tight" and will help prevent over-excursion - although it might also have no impact whatsoever depending on the size of the enclosure VS what would be optimal sealed size.
2. You change the system FR and likely limit net in-room low FR output. The port functions as a resonator to emphasize frequencies around the port tuning frequency, commonly to boost the lowers woofer output range and extend bass response in room. Sometimes "plugging a port" ends up sounding like the system has lost "authority" or "punch." Other times it sounds like those perceptual descriptions are improved. It all depends on the size of the enclosure and the specific driver involved.
3. You change system Q, which will lower efficiency, affect total power handling, change overall FR, and change low frequency roll-off from -12db/oct to -6db/oct. This means the system, especially if the woofer is designed to function in a multi-way system, may no longer seamlessly blend with the other drivers. Or, again, it may have absolutely no impact.
4. You change in room group-delay performance, likely improving delay as sealed system exhibit about half the GD as ported systems. This can improve the perceived "overhang" or reverberation effects in room depending on the pass-band of the woofer. The lower frequency the higher the GD.

Bottom line, unless a system is designed to be changed from sealed to ported and back, you really have no clue what the impact of the various changes might be. But it does no harm to try it and see. Do what sounds best in your application. Use a rolled-up sock.

[DYohn]

Sept 17, 2021 23:00:23 GMT -5 novisnick said:

Sept 17, 2021 21:57:20 GMT -5 DYohn said:

The ceiling is closer to the listening point, yes? Diffusers work across a broader frequency spectrum. I believe I recommended hanging from your ceiling a couple years ago.

Quick question if you don’t mind me asking. Aren’t diffusers made to be used with specific problems and frequencies? Please forgive my question if its too rudimentary. 

Yes, they can be. Or they can be broad-band systems. It all depends on how they are made.

[Boomzilla]

Sept 18, 2021 9:35:49 GMT -5 DYohn said:

T/S parameters are not just for car audio, of course. re: port plugging. What you accomplish by doing this is you are changing the enclosure alignment from what was designed, which changes several system parameters including system Q factors, FR and power handling.
1. You convert the enclosure from bass-reflex to "sealed" infinite baffle. The enclosure will likely be too large to consider it operating as acoustic suspension. This may cause bass output to sound more "tight" and will help prevent over-excursion - although it might also have no impact whatsoever depending on the size of the enclosure VS what would be optimal sealed size.
2. You change the system FR and likely limit net in-room low FR output. The port functions as a resonator to emphasize frequencies around the port tuning frequency, commonly to boost the lowers woofer output range and extend bass response in room. Sometimes "plugging a port" ends up sounding like the system has lost "authority" or "punch." Other times it sounds like those perceptual descriptions are improved. It all depends on the size of the enclosure and the specific driver involved.
3. You change system Q, which will lower efficiency, affect total power handling, change overall FR, and change low frequency roll-off from -12db/oct to -6db/oct. This means the system, especially if the woofer is designed to function in a multi-way system, may no longer seamlessly blend with the other drivers. Or, again, it may have absolutely no impact.
4. You change in room group-delay performance, likely improving delay as sealed system exhibit about half the GD as ported systems. This can improve the perceived "overhang" or reverberation effects in room depending on the pass-band of the woofer. The lower frequency the higher the GD.

Bottom line, unless a system is designed to be changed from sealed to ported and back, you really have no clue what the impact of the various changes might be. But it does no harm to try it and see. Do what sounds best in your application. Use a rolled-up sock.

Thanks kindly DYohn - I've contacted B&W for their advice. I believe that this speaker DID come with factory port plugs, but I don't have their literature about when to use the plugs (or why). I may experiment in the meantime.

[DYohn]

Boomzilla, here are some quick models intended just to show the most gross response change due to FR from plugging a port. This is not intended to be a "good" design, by the way, just intended to illustrate how low frequency output changes in plugging a port.

This is a random woofer from my library in a random 100-liter enclosure, ported and sealed. Note how the predicted FR above 100Hz is basically the same. It's only in the lower frequencies that the change is notable.

Vented VS closed box models:

[Boomzilla]

Yes - I found an online manual for these speakers. They were originally sold with "foam bungs" for the ports. The manual states that the bungs can be used to adjust bass response (as you've noted), but don't mention any effect at all on higher frequencies. Since I don't use the speakers' bass at all, I don't expect the port plugs to make any significant or audible differences.

Thanks again for the feedback.

[DYohn]

Sept 18, 2021 10:07:15 GMT -5 Boomzilla said:

Yes - I found an online manual for these speakers. They were originally sold with "foam bungs" for the ports. The manual states that the bungs can be used to adjust bass response (as you've noted), but don't mention any effect at all on higher frequencies. Since I don't use the speakers' bass at all, I don't expect the port plugs to make any significant or audible differences.

Thanks again for the feedback.

I suspect it will make an instantly audible difference. Try it and see.

[DYohn]

Sept 17, 2021 22:20:31 GMT -5 Boomzilla said:

Thanks, DYohn. While I have your attention, may I also enquire about port plugging? As someone with extensive experience with car audio, I suspect you’re no stranger to Misters Thiel & Small…

My speakers are what is commonly described as 2.5-ways. There are two woofers, the bottommost of which is tuned to begin rolling off slowly above 250Hz. The topmost woofer also produces sound down to the cabinet’s roll off (-6dB @ 30 Hz.) but is “optimized for the midrange.”

Both cones have their own internal enclosure and are individually ported. I’m using an external 12dB / octave high pass filter and a sub, so neither driver is required to operate at any significant excursion. On some material, the speakers can sound as if vocalists are slightly recessed in the soundstage. I understand that this can be slightly helped in some speakers by partially or completely obstructing the ports. The trade-off is that lower bass response is lost (not a concern since I’m using a sub).

What’s your take on port manipulation?

Thanks - Boom

I just read this post more carefully and I have a couple of comments.

First, you describe vocalists as sounding "recessed." This is because your Xover is set entirely too high. The vocal range is well below 250Hz, so reduce this to, say, 100-120Hz. Or even to 60-80Hz since these speakers seem to have pretty good bass response.

Second, saying things like losing the LF response is not an issue because you are using a sub is not accurate. A properly integrated subwoofer reinforces the sub-bass frequency band below 100Hz and should not be used as part of the primary response band. It is supposed to simply reinforce what the main speakers have trouble reproducing. Also, saying that you are not using the bass from your mains is also inaccurate, as even a "brick wall" crossover operates with a slope and the speakers output below the Xover frequency is simply reduced in volume, not eliminated. If your Xover is -12db/oct set to 250Hz, that means if we assume 250Hz is the zero reference, the sound will be -12db @ 125 Hz. Not eliminated, just lower in volume.

Here's a FR spectrum of typical male and female voices. You can see there is significant content below 250Hz that you want your mains to provide as the main source, not your sub.

[novisnick]

Sept 18, 2021 12:07:05 GMT -5 DYohn said:

Sept 17, 2021 22:20:31 GMT -5 Boomzilla said:

Thanks, DYohn . While I have your attention, may I also enquire about port plugging? As someone with extensive experience with car audio, I suspect you’re no stranger to Misters Thiel & Small…

My speakers are what is commonly described as 2.5-ways. There are two woofers, the bottommost of which is tuned to begin rolling off slowly above 250Hz. The topmost woofer also produces sound down to the cabinet’s roll off (-6dB @ 30 Hz.) but is “optimized for the midrange.”

Both cones have their own internal enclosure and are individually ported. I’m using an external 12dB / octave high pass filter and a sub, so neither driver is required to operate at any significant excursion. On some material, the speakers can sound as if vocalists are slightly recessed in the soundstage. I understand that this can be slightly helped in some speakers by partially or completely obstructing the ports. The trade-off is that lower bass response is lost (not a concern since I’m using a sub).

What’s your take on port manipulation?

Thanks - Boom

I just read this post more carefully and I have a couple of comments.

First, you describe vocalists as sounding "recessed." This is because your Xover is set entirely too high. The vocal range is well below 250Hz, so reduce this to, say, 100-120Hz. Or even to 60-80Hz since these speakers seem to have pretty good bass response.

Second, saying things like losing the LF response is not an issue because you are using a sub is not accurate. A properly integrated subwoofer reinforces the sub-bass frequency band below 100Hz and should not be used as part of the primary response band. It is supposed to simply reinforce what the main speakers have trouble reproducing. Also, saying that you are not using the bass from your mains is also inaccurate, as even a "brick wall" crossover operates with a slope and the speakers output below the Xover frequency is simply reduced in volume, not eliminated. If your Xover is -12db/oct set to 250Hz, that means if we assume 250Hz is the zero reference, the sound will be -12db @ 125 Hz. Not eliminated, just lower in volume.

Here's a FR spectrum of typical male and female voices. You can see there is significant content below 250Hz that you want your mains to provide as the main source, not your sub.

View Attachment

Great information DYohn , thanks so much!

[Boomzilla]

Sept 18, 2021 12:07:05 GMT -5 DYohn said:

I just read this post more carefully and I have a couple of comments.

First, you describe vocalists as sounding "recessed." This is because your Xover is set entirely too high. The vocal range is well below 250Hz, so reduce this to, say, 100-120Hz. Or even to 60-80Hz since these speakers seem to have pretty good bass response.

Second, saying things like losing the LF response is not an issue because you are using a sub is not accurate. A properly integrated subwoofer reinforces the sub-bass frequency band below 100Hz and should not be used as part of the primary response band. It is supposed to simply reinforce what the main speakers have trouble reproducing. Also, saying that you are not using the bass from your mains is also inaccurate, as even a "brick wall" crossover operates with a slope and the speakers output below the Xover frequency is simply reduced in volume, not eliminated. If your Xover is -12db/oct set to 250Hz, that means if we assume 250Hz is the zero reference, the sound will be -12db @ 125 Hz. Not eliminated, just lower in volume.

Here's a FR spectrum of typical male and female voices. You can see there is significant content below 250Hz that you want your mains to provide as the main source, not your sub.

View Attachment

Sorry DYohn - I wasn't clear enough. The two woofers in the cabinet were designed by B&W so that both produce the frequency range from the tweeter crossover (maybe 1,500Hz.?) down to the speaker's roll-off (30Hz. @ -6dB). But one of the two woofers is set by the internal crossover to shelve down a bit at 250Hz. while the other is set to shelve up a bit. So only one of the two woofers is (primarily) active from 250 to 1,500Hz.

The upstream electronic crossover I'm using in my setup sends all info below 90Hz. to the subwoofer, and all above 90 to the satellites.

From what I understand you're recommending, I should discard the subwoofer crossover + send full-range signals to the satellites + blend the sub in below the normal satellites' roll off by using the plate amp on the sub. Is this correct?

In theory, there are some disadvantages to doing what you recommend:

1. The satellite amplifier has to work significantly harder because it now has to reproduce a significant amount of bass that it previously didn't.

2. Intermodulation distortion of the satellite speakers increases exponentially because of the greater excursion required of the bass/midrange drivers (this would be less were the system a true 3-way)

3. Unless the phase is set right for the slopes involved, there can be significant cancellation at the frequency where the satellites begin to roll off and the subwoofer blends in

4. Unless the subwoofer's plate amplifier uses a steep low pass filter (minimum of second or third order), the sub (with significantly more distortion than the satellites) will continue to sing along above its crossover point, muddying the bass of the satellites. Since I'm using an inexpensive subwoofer, and the manufacturer refuses to disclose the order of their plate amplifier's filter, this is of particular concern.

5. Any equalization applied in the bass will now affect the satellite speakers as well. With the crossover active, sub equalization could be applied only after the crossover, and could not possibly affect the satellites.

Comments?