What are your tips on bass in small rooms?

[Jeff B.]

OK, this is how you measure to find your transition frequency. Just mic different room locations and it's the region where the frequency response breaks in the low frequencies.

This graph is showing the Schroeder frequency transition of the room. You can also find this point mathematically as well.

[Jeff B.]

Bill. You led me astray, dude. You got me thinking there was such a thing as cabin gain in small rooms then Earl Geddes said that just ain't the truth. Why are you spreading misinformation?

Wow, dude, relax. Geddes isn't automatically right here. Rooms are lossy, but most still provide room gain. I can measure 11 dB of gain below 20 Hz in my living room. What Bill has been telling you is correct. He didn't mislead you at all. There are two different transitions in the environment of room acoustics. First we transition from free-field to below the Schroeder frequency where the room begins to take control through modal distribution. Then at the frequency whose half-wavelength is longer than the longest dimension of the room we transition to pressurization, this is when the gain begins to occurs. Theoretically it rises at 12 dB/ oct, but due to leakage it will be much less than this. Still it is not uncommon to see 8-12 dB of total gain depending on the room.

[andy19191]

Andy, can you clarify? I thought the typical approach would be something like using REW and a measurement mic. Are you saying that to do THIS right would take forever, or something else?

Measuring a response doesn't help guide you towards good solutions when there are a vast number of options. A vast number of simulations on a computer however might. I have three tasks: the first is to control the response below about 80 Hz using a set of small subs, the second to design and place passive room treatment where it will do most good in the range 80 Hz to the Schroeder frequency and the third to control the directivity of the main speakers within the room to minimise early reflections, maximise later transverse reflections, create a reasonably constant decay time, create a smooth fall in average response with frequency, etc...

I currently have the drivers for 4 small subs but will likely need 8. Simulating where to place them and how to vary the gain and time delay for each sub to obtain a flat response around the listening location with a constant decay time is a significant but fairly straightforward optimisation task. If the locations were fixed then one could use measured transfer functions but if you want to efficiently include all reasonable sub locations then simulation is preferable.

Sorting the locations for effective room treatment above 80 Hz is more challenging and some measurement will be needed to check estimates of absorption coefficients for curtains, furniture as well as the room treatment devices. I suspect the room already has enough absorption at higher frequencies and so plastering the walls with room treatment would not be a good move. More diffusion is likely needed because, unlike my previous room, there isn't much to move the sound out of the axial direction and into the transverse direction. Should be an interesting task.

The speaker directivity and interaction with the room is a longer term set of simulations. I suspect the mid/tweeter coaxial drivers I bought before moving house will have too wide a beamwidth but intend to use them initially to confirm and gather information before moving on to a replacement. Planning the next speaker before completing the current speaker...

[a4eaudio]

I currently have the drivers for 4 small subs but will likely need 8.

Andy, thank you for the thoughtful response! Just curious, but what are "small" subs" in your situation? I think of 8 inch subs as small subs, but I know they can be smaller and I know people who want to load up on 18" Ultimax subs would think 12" was small LOL.

[billfitzmaurice]

I currently have the drivers for 4 small subs but will likely need 8.

Maybe, maybe not. The most improvement over a single sub comes with using two, with less significant improvement coming with a third, less still with a fourth. More than four tend to have only a slight effect. If you think you need more than four because they're too small to provide enough output with four that's a different story.

[Jeff B.]

This is from the reference section of Toole's book Sound Reproduction, 3rd Edition. Now no more nonsense from the Bro-Audio Dude crowd. If your name isn't listed in the reference material of an audio textbook your opinions contradicting those who are doesn't count.

If you believe this, and believe that for the rest of us here our knowledge and experience doesn't count, then why did you bother to ask the question? My name is referenced in some textbooks, but seriously, does that even matter? It shouldn't matter at all. It's the explanation that should matter.

[andy19191]

Maybe, maybe not. The most improvement over a single sub comes with using two, with less significant improvement coming with a third, less still with a fourth. More than four tend to have only a slight effect. If you think you need more than four because they're too small to provide enough output with four that's a different story.

The subs are being primarily used for room control not bass extension. That is, they are distributed and setup to absorb almost as much sound as they create in order to reduce the decay time, fill in the large dips and reduce the large peaks in the room response. The room is significantly asymmetric and so the normal symmetric arrangement of subs in a rectangular room that knocks out various modes does not apply raising the required number. In addition, the two obvious locations for the main speakers (one on wall and one in corner) places the listener in the big axial dip halfway between the front and back walls. The room arrangement doesn't allow a 1/3rd or 2/3rd seating position. This would seem to dictate a plane-wave type arrangement of a source wall and a sink wall but the walls are different sizes with the smaller being mainly doors precluding the desired evenly spaced arrangement. None of the normal sub arrangements work which makes an interesting challenge but does inevitably force up the number of distributed sources to get a reasonably even response. I don't want to go above 8 for hardware cost reasons and to avoid the living room looking like a studio so that is the limit I am working to.

Although I have little interest in low frequency extension it is likely to be reasonable given that below the frequency of the lowest axial mode the subs can transition to all working together as sources since the reason for distributing the sources has gone and long decays times are not going to be a problem listening to exploding planets and other sustained very low frequency booms/noises/notes.

[andy19191]

Andy, thank you for the thoughtful response! Just curious, but what are "small" subs" in your situation? I think of 8 inch subs as small subs, but I know they can be smaller and I know people who want to load up on 18" Ultimax subs would think 12" was small LOL.

I am currently using 8" drivers in sealed cabinets because they can fit on bookshelves. I have some doubts because they will be set to absorb a lot of sound and when you do this (e.g. cardioid and open baffle) it requires substantially more displacement and cone area to reach standard levels. If I am struggling with distortion at standard levels then porting and dual force cancelling arrangements would be options while still being able to placed on bookshelfs but it will push up the cost per sub in both price and space. Alternatively some larger subs could be used on the floor, on top of bookshelves/units or even attached to the wall depending on how the set works together since the subs in some locations will be driven more than others.

[bradley.s]

If Earl said that he's wrong. Cabin Gain is well documented. For instance:
https://legacy.data-bass.com/data?page=content&id=2

I stopped reading anything from Earl years ago, when it became apparent to me that his opinions were anything but objective.

No, Bill. You linked to an article on *vehicle* audio. Geddes didn't just say it offhandedly, he put it on a power point presentation slide. His PhD disseratation at Penn State was small room low frequency acoustics. He published peer reviewed papers on the subject that were cited and became consensus. Read the powerpoint.

[bradley.s]

If you believe this, and believe that for the rest of us here our knowledge and experience doesn't count, then why did you bother to ask the question? My name is referenced in some textbooks, but seriously, does that even matter? It shouldn't matter at all. It's the explanation that should matter.

I asked the question because I didn't know. I received answers that contradicted the established science. Now, should I accept an answer that *contradicts* established science on the basis, "Pffft, I have experience"?

Yes, peer reviewed research and citations matter. Was your name referenced or was your research cited as consensus? What was the topic? Who was the author of the textbook? All these things matter because it establishes the level of expertise in the subject under discussion. Especially when one offers an opinion that contradicts the consensus.

[a4eaudio]

Yes, peer reviewed research and citations matter.

Well, now your changing the rules. Your previous post said "If your name isn't listed in the reference material of an audio textbook". A textbook is NOT peer reviewed research. If you have time to read a lot of scientific research, then that might be a more effective approach than asking a question on a DIY forum.

The point being made by several people is that your approach seems misguided...you ask a question and now you are telling everyone what you think the answer should be and arguing with experts. Again, in post #55 you are trying to tell Bill to read something...Bill is trying to help you. He provided a link relevant to your thread. If you aren't sure you believe it, say why and ask a question. Bill did not ask any questions or start the thread, he didn't express any interest in reading a PPT about Geddes and in fact stated that he quit paying attention to him long ago.

[djg]

"Peer reviewed"

More Computer-Generated Nonsense Papers Pulled From Science Journals - The Atlantic

The Failure of Peer Review

[billfitzmaurice]

No, Bill. You linked to an article on *vehicle* audio.

Audio is audio. Cabin gain is more obvious in a car because the room in question is particularly small. Jeff B's explanation was spot on.

Like others I've also come to the conclusion that there's no point in responding to any of your questions, since you don't actually want to learn anything, you just want to argue. Worse, to argue about that which you have no fundamental understanding of. Lesson learned on my part.

[johnk...]

I really don't have any interest in this thread but with regards to room gain. Room gain is based on the concept that below the room fundamental the room acts as a lumped parameter and the room pressure goes like PV^k = constant. The volume is the volume of the room plus the change in volume due to driver area and displacement, V = Vr + Sd * X sin(wt). So P = const/ V^k) and the pressure in the room will have a sine variation. Since for constant SPL driver displacement increases at 1/f^2, room gain increase SPL by 12dB/oct as the frequency decreases below the room resonance. Now since the pressure is assumed equal everywhere in the room it must be the same on all the walls. Thus any leakage is due to the difference in pressure on the bounding surfaces and the pressure outside the room. The room must "breathe". This would cause the room gain to be below 12db/oct, but for no room gain the pressure would have to remain constant and equal to the outside pressure. That is impossible because for air to leak out of the room through whatever leakage there is, there must be a pressure difference.


Now consider room modes. Consider the room fundamental. The room fundamental has pressure antinodes on the bounding surfaces. Thus, the pressure on those surfaces will vary as sin(wt). If those surfaces are so leaky that they can not sustain some degree of room gain, then it follows that they could not sustain the pressure variations associated with the room mode either and the mode could not be supported.

Thus, if modes can be supported it follows that room gain must also be supported. The one factor that influences both is time. At higher frequencies, there is less time for leakage to occur. The conclusion is that while room gain must exist if low frequency modes are supported, the strength of the room gain, as well as the modes, will weaken as the frequency decreases in a leaky room.

Oh, and for the record, some of my work is referenced in one of Doug Self's books.

No back to Ferrari Chat where the only acoustics that matter is the sound of your exhaust when the car is driven through a tunnel at wide open throttle, at red line, near 130db.

[Jeff B.]

I asked the question because I didn't know. I received answers that contradicted the established science. Now, should I accept an answer that *contradicts* established science on the basis, "Pffft, I have experience"?

Yes, peer reviewed research and citations matter. Was your name referenced or was your research cited as consensus? What was the topic? Who was the author of the textbook? All these things matter because it establishes the level of expertise in the subject under discussion. Especially when one offers an opinion that contradicts the consensus.

Geddes was basing his position that there is no room gain on the premise that there was enough leakage to counter the pressurization, but as John K already stated, that’s very unlikely. It is really not that difficult measure the low frequency gain in a room, which usually begins below 40 Hz or so.

It’s funny, in all my years in speaker design I have never once heard that the consensus was that there was no room gain. Never. I have heard Earl say it, but he was the only one. The funnier part is how easy it is to verify. There really should be no debate on this one.