DIY sub fixed to the wall&floor and firing with an angle?

[Arcsabre]

Hi all, I am currently designing a DIY subwoofer for a rectangular theatre room which is to be built and fixed to the floor and against the back wall. It will be located the midpoint on the virtual floorplan view.

I've done some research and Floyd Toole's paper inspired me of a symmetric dual-sub setup, with Sub1 placed at the midpoint of the front wall, and Sub2 at the midpoint of the backwall. Both are sealed.

However as the front wall of my theatre room is actually french doors I can't have Sub1 built the same way as Sub2. I have to tweak the idea slightly.

I am now considering to build Sub1 at the midpoint of the front wall but just below the ceiling instead of on the floor. so the 2 subs will still be symmetric but the symcenter is in the middle of the room in 3D, instead of the center of the floor.

Is there going to be some issue if I have Sub 1 firing towards the audience's face but tilted downwards (like 25° or so) while Sub 2 firing towards the back of the audience but tilted upwards?

I've never seen such design that's why I'm just a bit concerned.

I'm aware of bass being omnidirectional and I'm not really worried about that part. Just what do you guys think about this idea in terms of any potential phase cancellation/wall reflection? I'm going to have my subs fixed to the walls and possibly the ceiling and floor.

In this case I should be able to get relatively clean bass due to non-existence of reflected bass from the back wall right? I see Floyd Toole's paper in which he tested subs against the walls and some dual setup (and quad sub setups) could actually provide excellent result, contrary to the common understanding that "subs should be as far from the corner & walls as possible".

The size of the sub in the picture 1 below is only indicative and I am planning to build relatively wide enclosures so the size of the boxes isn't going to be too small. the size I'm currently thinking about is h330 d270 w600 (height and depth are shown in the picture) a relatively wide shape. I can certainly go wider (not taller or deeper) but I'm slightly worried about matching the damping factor of the enclosure to the driver. Currently I yield a Qtc of 0.4688 using a 15" subwoofer driver from Lavoce (see picture 2). A lot people recommended a Qtc of 0.707. If I go wider the Qtc of this DIY sub would drop below 0.4688 and getting even lower. So I don't know if this is something that's limiting me from going any wider.

Would be great if any one knowledgeable could shed some light 🐸 thanks all ba(by)ss

 

Pic1

Pic2

 

[kipman725]

Placement looks fine, the main thing is to have more subs and place them in a diverse set of locations around the room, they don't care which direction they point.  Real rooms differ from the research (in your case you have a large lossy door on one wall) so its not necessarily true the best locations would have been mid wall anyway.

Box volume isn't very important (as long as your using EQ) as what produces the sound is the cone moving in and out a larger box is just requiring less power to reach a given SPL when operated below Fc.  

[peniku8]

3 hours ago, kipman725 said:

Placement looks fine, the main thing is to have more subs and place them in a diverse set of locations around the room, they don't care which direction they point.  Real rooms differ from the research (in your case you have a large lossy door on one wall) so its not necessarily true the best locations would have been mid wall anyway.

Box volume isn't very important (as long as your using EQ) as what produces the sound is the cone moving in and out a larger box is just requiring less power to reach a given SPL when operated below Fc.  

I think he is planning to run a double bass array, which works best if everything is perfectly symmetric. I'm not sure how well this particular arrangement would work, but I guess it wouldn't hurt to try.

[Arcsabre]

16 hours ago, peniku8 said:

I think he is planning to run a double bass array, which works best if everything is perfectly symmetric. I'm not sure how well this particular arrangement would work, but I guess it wouldn't hurt to try.

That's right and I'm just more concerned about whether this setup is going to work similarly as the "front-wall-midpoint" and "back-wall-midpoint" dual sub method which has been proven to work well. I will run test with measurement mic and REW after installation but I probably have to fix them to their position once installed (especially the ceiling one). I am able to adjust the phase and EQ afterwards but not relocate them so just wanna see if there's any obvious issue that can only be solved by moving them to a different place.

[Arcsabre]

If someone has tried this setup and found issue I might just go for a single 21" DIY sub in the middle of the backwall like something Ricci designed...

[kipman725]

Ah ok I thought it might be a double bass setup but was thrown off track a bit.  I see a few issues:

1) double bass setup should launch a plane wave which relies on the point sources on the opposed walls been within 1/4 wavelength of each other and there to be point sources within 1/4 WL of the walls.  At say a crossover 80Hz this is 1m, however real world this can probably be pushed up a bit:

https://www.avsforum.com/threads/double-bass-array-dba-the-modern-bass-concept.837744/

So definitely in your configuration there is no plane wave due to the geometry.

2)   Another issue I see is that your room has unequal absorption on each wall due to the french doors.

Even considering this I still think it would be better to run two subs rather than one even if running them as double bass array doesn't work due the smoothing effect on room responses:

https://data-bass.com/#/articles/5cb5fb285389a80004c7e58a?_k=ymaw45

[SME]

@kipman725: I'm pretty sure he's not talking about a DBA.

Hi @arcsabre, I think I know what paper you are referring to.  It's also covered here: https://www.audioholics.com/room-acoustics/optimum-locations-for-subwoofers-in-rectangular-rooms

The study is a theoretical investigation of standing wave room resonance behavior, and the opposing mid-wall placement recommendation arises as the optimal placement for two subs (run together as one, with no differential delay in the signal going to each) that minimizes seat-to-seat variation of frequency response in the room interior  I want to highlight a number of problems with taking that conclusion from the study and running with it.

First, the study is theoretical and does not consider absorption and losses in the walls, floor, and ceiling.  Unless you're inside a bunker, those losses are not trivial .  The lack of symmetry due to different kinds of construction on different walls is one thing.  If your walls aren't particularly rigid, you may not have much of a problem with standing waves anyway.  That's because standing waves require multiple successive reflections to build up.

Second, mid-wall placements typically have poor output efficiency and suffer more from boundary interference effects from the adjacent walls.  My own experience is that these problems can be severe.  Depending on your room, such interference problems may predominate over standing waves for much of the response.  In general, subs are most efficient in a corner because the corner allows more pressure to build near the cone, which improves energy transfer.  This efficiency boost is very important because it reduces the amount of equipment and cost required to reach a particular output goal.  Also because the sound energy spreads through a smaller section of the unit sphere ("pi/8 space" vs "pi/4 space", the directivity is effectively higher, meaning you get high SPL for the same power output too.  Even the paper notes that mid-wall placements tend to have poor output vs. corner placements, but they are focused on minimizing seat-to-seat variation regardless of cost or compromise.

Third, the study treats the subwoofer system completely independently of the mains speakers.  The trouble here is that the subs do eventually cross over with the mains, and that crossover region is quite broad even with fairly steep high order.  For an 80 Hz crossover 4th order LR, it might be 60-110 Hz, which is nearly an octave, and covers frequencies that are crucial for enjoying all bass content.  IMO, a good crossover with the mains is likely to be *more important* than avoiding standing waves at lower frequencies, especially if your focus is music.

My suggestion if you are building two subs is to put one in each front corner of the room.  This puts them relatively close to the mains, so they will be more likely to combine  with the front channels at multiple room locations.  These placements will also cancel out some of the standing waves along the width of the room, if that's a problem for you.  It will not cancel out standing waves across the length of the room.  This is most likely to be a problem for in the very middle of the room (length-wise) or a little further back, and if it's possible to move those seats even a couple feet, it might make a big difference.  If you can build 4 subs, then you can put one in each corner.  Though this might make integrating with the front mains channels a bit trickier---or not.  A lot depends on specifics.

[SME]

I also want to add that the article and sound people in general take "frequency response" measurements captured in a room way too literally.  The study seeks to minimize "mean spatial variation" while ignoring the crucial question of whether that should even be the objective.  My experience suggests otherwise.

The first problem is that bass perception isn't merely a function of hearing through the ears.  Bone conduction and mechano-reception (via nerves throughout your skin and body) likely play major roles too, and they are able to pick up signals from a wider variety of locations and via different transmission mechanisms.  Second, the brain possesses highly advanced cognitive processing that is well adapted to listening in small rooms and inferring the nature of the sound source independently from the "acoustical context".

Floyd Toole and Harman have provided ample evidence that, at least for mid and high frequencies, the listener is capable of "hearing through the acoustics" to a great extent in order to ascertain the original sound source, i.e. the speaker or sub.  They've shown that anechoic chamber measurements, on and off axis, correlate better with blinded listener preference than an in-room measurement (taken "literally").  I believe this remains true for bass, except that as you go lower, the speaker drivers interact with more than just the cabinet.  They begin to interact with nearby boundaries as well.  These boundaries effectively become part of the speaker, and their influence alters the sound of the speaker, just as different cabinet and baffle shapes do.  Go low enough, and the speaker/sub will even interact with the modal /standing wave resonances in the room.

Nevertheless, the Harman people seem to ignore these observations and instead argue that optimizing in-room frequency response is the best course for bass, despite it being sub-optimal for mids and highs.  They believe that essentially all the usual rules "go out the window" below the "Schroeder transition frequency", and that below this point, standing wave behavior dominates the room response.  I disagree with them on all these counts.

First, Schroeder transition frequency is a theoretical construct that assumes large rooms with diffuse reverberant fields, not small residential listening rooms.  They describe this transition as typically happening around 500 Hz, probably because that's the rough point below which in-room frequency response measurements start to look a lot "messier" with peaks and dips.  In reality, this is simply a consequence of lower directivity, meaning more and stronger reflections are contributing to the measured response than for higher frequencies.

Second, I don't believe most rooms exhibit standing waves until much lower in frequency, and that behavior may not even predominate for subwoofer frequencies. unless you are in a room with stiff walls.

Third, the presence of standing waves does not necessarily preclude the listener's ability to "hear-through" the local acoustic effects.  Where standing waves are involved, a null in SPL response coincides with a peak in particle velocity level (PVL) response.  It's possible that listeners may be able to perceive aspects of PVL independently from SPL, perhaps through indirect mechanisms like hair or clothing movement.

And fourth, I've tried optimizing for minimum seat-to-seat variation and then flattest possible response in my own room using practically unlimited custom DSP capability.  The result looked beautiful in the REW plots, but the sound quality was far from ideal.  This was most evident when I auditioned one of Harman's own speakers, the (Revel Salon 2), playing (in another room) with no EQ whatsoever.  Its musical bass quality blew mine out of the water.  For my purposes, this essentially disproved the claim that flat in-room frequency response is ideal, and it inspired me to develop a novel and more clever approach to assess the sound of the speaker independent of the localized acoustics.  My current method seems to work just as  well at 25 Hz as it does at (say) 1500 Hz, and so I'm not aware of any sort of transition frequency where the rules of perception change.  The only thing that changes is how *complete* anechoic/ground plane measurements are for describing sound quality, where for bass those measurements are still relevant but room effects become at least as important.

Anyway, sorry I can't give simple answers, but that's the nature of the subject!  I'd suggest you blame your brain, but actually if it weren't for our brains conspiring to make everything sound better than it really does, none of us would probably have this hobby. 

[kipman725]

I have run between one and 4 subs at home and much prefer 4 subs, they are mostly corner placement.  I take a frequency response measurement at each seat of my sofa and vector average the results.  After I have dialed in the sub crossover I fit the response to a target curve below 500Hz as my main speakers are close in to the corners and designed to be flat in free space  so this has to be compensated for and there are some large peaks in the lower bass that need to be cut to avoid the whole thing been boomy.  Works well but I wish there was a more scientific method.  SME when you say flattest possible response do you mean an actually flat response? as from my understanding a constant directivity speaker in a domestic room which has a flat free space response should have a tilted downwards response in room (and will sound natural in this condition) so equalizing to flat in room will make a speaker sound overly bright.

[Arcsabre]

47 minutes ago, kipman725 said:

I have run between one and 4 subs at home and much prefer 4 subs, they are mostly corner placement.  I take a frequency response measurement at each seat of my sofa and vector average the results.  After I have dialed in the sub crossover I fit the response to a target curve below 500Hz as my main speakers are close in to the corners and designed to be flat in free space  so this has to be compensated for and there are some large peaks in the lower bass that need to be cut to avoid the whole thing been boomy.  Works well but I wish there was a more scientific method.  SME when you say flattest possible response do you mean an actually flat response? as from my understanding a constant directivity speaker in a domestic room which has a flat free space response should have a tilted downwards response in room (and will sound natural in this condition) so equalizing to flat in room will make a speaker sound overly bright.

What you said about the flat response is correct. If a speaker's in-room response is EQ'ed to be flat (or downward-flat) that literally means Direct Sound + all Reflection = fixed number across all frequencies, and since almost all speakers has non-linear off-axis SPL, trying to get flat in-room response like this would mean wobbly output across all frequencies. Of course this wouldn't sound good.

In my layman's understanding the flat response and downward-flat in-room response thing simply means that's people's average preference, if there's a pair of speakers that performs consistently on-axis and off-axis. Statistics averages out different speakers' frequency response (both on-axis and off-axis) so what we are seeing from those study tells us something about all tested speakers as a whole, and simply applying it without careful tweaking would almost guarantee generating weird sound, for any particular pair of speakers.

So if we have a pair of speaker which measures flat in an Anechoic chamber, both on-axis and off-axis, and output narrows linearly as frequency goes up - would measure flat downwards in a normal room, and sound excellent to most people. Of course no speakers are this good yet.

Sounds like we're in the same situation but I guess I'm just asking in advance before going out to the workshop and start cutting MDFs 🤤

[Arcsabre]

5 hours ago, SME said:

@kipman725: I'm pretty sure he's not talking about a DBA.

Hi @arcsabre, I think I know what paper you are referring to.  It's also covered here: https://www.audioholics.com/room-acoustics/optimum-locations-for-subwoofers-in-rectangular-rooms

The study is a theoretical investigation of standing wave room resonance behavior, and the opposing mid-wall placement recommendation arises as the optimal placement for two subs (run together as one, with no differential delay in the signal going to each) that minimizes seat-to-seat variation of frequency response in the room interior  I want to highlight a number of problems with taking that conclusion from the study and running with it.

First, the study is theoretical and does not consider absorption and losses in the walls, floor, and ceiling.  Unless you're inside a bunker, those losses are not trivial .  The lack of symmetry due to different kinds of construction on different walls is one thing.  If your walls aren't particularly rigid, you may not have much of a problem with standing waves anyway.  That's because standing waves require multiple successive reflections to build up.

Second, mid-wall placements typically have poor output efficiency and suffer more from boundary interference effects from the adjacent walls.  My own experience is that these problems can be severe.  Depending on your room, such interference problems may predominate over standing waves for much of the response.  In general, subs are most efficient in a corner because the corner allows more pressure to build near the cone, which improves energy transfer.  This efficiency boost is very important because it reduces the amount of equipment and cost required to reach a particular output goal.  Also because the sound energy spreads through a smaller section of the unit sphere ("pi/8 space" vs "pi/4 space", the directivity is effectively higher, meaning you get high SPL for the same power output too.  Even the paper notes that mid-wall placements tend to have poor output vs. corner placements, but they are focused on minimizing seat-to-seat variation regardless of cost or compromise.

Third, the study treats the subwoofer system completely independently of the mains speakers.  The trouble here is that the subs do eventually cross over with the mains, and that crossover region is quite broad even with fairly steep high order.  For an 80 Hz crossover 4th order LR, it might be 60-110 Hz, which is nearly an octave, and covers frequencies that are crucial for enjoying all bass content.  IMO, a good crossover with the mains is likely to be *more important* than avoiding standing waves at lower frequencies, especially if your focus is music.

My suggestion if you are building two subs is to put one in each front corner of the room.  This puts them relatively close to the mains, so they will be more likely to combine  with the front channels at multiple room locations.  These placements will also cancel out some of the standing waves along the width of the room, if that's a problem for you.  It will not cancel out standing waves across the length of the room.  This is most likely to be a problem for in the very middle of the room (length-wise) or a little further back, and if it's possible to move those seats even a couple feet, it might make a big difference.  If you can build 4 subs, then you can put one in each corner.  Though this might make integrating with the front mains channels a bit trickier---or not.  A lot depends on specifics.

Thanks mate, there's so much useful knowledge in your response...I need to spend more time considering every point you made. And my view is somehow similar to yours, only that the room does not allow me for such setup. Please see the attached picture. Meaning of different color: Yellow-Lights, irrelevant here. Blue-Absorption Purple-Diffusion DarkBlue-Bass Trap. the bottom of this plan is the french doors and that opens up throughout the whole wall (left wall to right wall all the way) thus subwoofers cannot be put there...

After reading your response I do agree with you on the priority of bass design - room mode shouldn't be a major concern, especially when using dual or 4 subs with DSP. However throughout years of using subwoofers from big brands - SVS and Rel especially - I do find putting subs close to the corner do sound muddy and boomy and I had no luck fixing that using DSP in the past. If room modes and phase problems emerging from reflections adding onto/cancelling each other are not that important, how can we make subs sounds good in the corner? (like how they should perform on paper)I mean putting a reasonably sized merchandise subwoofer right in the corner shouldn't blur the bass on paper - but it does in reality. I'm not 100% sure what the issue is.

In my case I have 4 units of Dayton UM15-22 subwoofer driver. The amp I've prepared are not subwoofer's plate amp but Hypex NC500 monoblocks. I'm using one for each subwoofer and these amps can output 700W clean power stably, for each channel. I've got really inefficient floorstanders Dynaudio C5 (sensitivity=83db) and these monoblocks drive them easily ------ only need -30db from preamp for really loud symphonies I compared with my friend's Dynaco and Krell, and they cannot get much bass out from the C5. so with the Hypex amps power isn't really an issue here.

After reading your inspiring response I think there are three ways I could use the drivers in this room and they are:

1. 2 sealed subs, with one driver on each enclosure. Both subs on the floor, one located at mid-backwall, one located at mid-frontwall. But as there's a french door in front the front sub can only get so close to it (there will be a gap at least) and my impression is this is a kinda embarrassing distance from the wall, not far enough to avoid reflection issues but also not close enough to solve them in another way.

2. 2 sealed subs, with one driver on each enclosure. Both subs on the floor, located below the 2 bass traps (see the floorplan). fit snugly against both walls of the corner.

3. 4 subs mount directly onto very thick MDF - such as a 3" thick one made of 3 sheets of 1" with green glue between them. 4 MDF board mount onto the midpoint of each wall, for side walls ------ 1 meter from the floor and 1 meter from the ceiling. For backwall ------ floor. For front ------ Above french doors. Use pure amp power to push them hard enough so they work to desired SPL without enclosure.Kinda like IB subs but I don't have that much space in the wall.

 

Which way would you recommend? Much appreciated (go back to read your response again🏋️‍♂️)

 

[SME]

On 10/15/2020 at 5:54 AM, kipman725 said:

I have run between one and 4 subs at home and much prefer 4 subs, they are mostly corner placement.  I take a frequency response measurement at each seat of my sofa and vector average the results.  After I have dialed in the sub crossover I fit the response to a target curve below 500Hz as my main speakers are close in to the corners and designed to be flat in free space  so this has to be compensated for and there are some large peaks in the lower bass that need to be cut to avoid the whole thing been boomy.  Works well but I wish there was a more scientific method.  SME when you say flattest possible response do you mean an actually flat response? as from my understanding a constant directivity speaker in a domestic room which has a flat free space response should have a tilted downwards response in room (and will sound natural in this condition) so equalizing to flat in room will make a speaker sound overly bright.

Good question!  Originally, yes.  Although first I should clarify that the shape of the spatially averaged response depends quite a bit on what measurements you include in the average.  IIRC, I was flat in the MLP where there was a kind of mid-bass "power ally" effect, but had rise toward the bottom in the outer seats.  I'm pretty sure I went back and made broad-shape tweaks to get the sound to "my liking".  And by "flat", I only meant for the sub anyway.  I totally agree that flat in-room FR will almost always sound too thin and bright.

What you're doing for your room is essentially what Harman et.al. recommend these days, and seems to be pretty standard for "state-of-the-art" DIY system as well.  These days, I believe Harman regards their latest target curve to be "secret sauce"---essentially proprietary.   That's just bizarre to me.  I mean, it's probably only subtly different from previous target curves they've recommended which are publicly known.  But more importantly, a target curve is, at best, a fuzzy indication of where things should end up.  The X-curve in cinema is essentially the same (e.g. +/- 3 dB tolerances, which are huge for SQ purposes, and pretty much gave the projectionist the lee-way to set the EQ to whatever sounds best for him/her, until we replaced them with robots that don't "hear" like humans do).  The same system will "measure" very different, even in the same room, depending on where choose to measure.  Does the average include more or fewer off-axis seats?  Are the extra rows included?  How close are the seats to the speakers, relative to the room size?  Those things and have a huge impact on where your broad response shapes will fall, so the target curve doesn't end up being useful IMO.

Another thing is that the content itself is not all roughly +/- 2 dB or something, broadly.  Instead, there are huge *differences* in spectral balance, not just between programs, but between tracks and/or segment of the same program.  In movies for example, a popular technique seems to be "mixing big", which among other things seems to using a huge "smile" EQ curve to up the impact of "big" on-screen events.  These differences largely overwhelm target curve differences.  What matters far more is that trying to fit in-room response (even with spatial averaging) to a smooth curve doesn't really lead to optimal results.  It might clean up some of the most obvious modal/room resonances, but it tends to make a mess of the innate/anechoic FR of the source and may actually make transients sound worse than without the EQ processing.  I don't really know, but I have often experienced more satisfying bass from speakers without EQ than with some kind of "room correction".

[SME]

On 10/15/2020 at 7:27 AM, Arcsabre said:

Thanks mate, there's so much useful knowledge in your response...I need to spend more time considering every point you made. And my view is somehow similar to yours, only that the room does not allow me for such setup. Please see the attached picture. Meaning of different color: Yellow-Lights, irrelevant here. Blue-Absorption Purple-Diffusion DarkBlue-Bass Trap. the bottom of this plan is the french doors and that opens up throughout the whole wall (left wall to right wall all the way) thus subwoofers cannot be put there...s

After reading your response I do agree with you on the priority of bass design - room mode shouldn't be a major concern, especially when using dual or 4 subs with DSP. However throughout years of using subwoofers from big brands - SVS and Rel especially - I do find putting subs close to the corner do sound muddy and boomy and I had no luck fixing that using DSP in the past. If room modes and phase problems emerging from reflections adding onto/cancelling each other are not that important, how can we make subs sounds good in the corner? (like how they should perform on paper)I mean putting a reasonably sized merchandise subwoofer right in the corner shouldn't blur the bass on paper - but it does in reality. I'm not 100% sure what the isue is.🏋️‍♂️

Is the room drawn to scale?  It looks like a small room, and unfortunately it looks like your seats will fall right in the lengthwise null.  I didn't mean to say that modal/standing-wave resonances aren't important.  They usually do come into play in small rooms.  It's just that other considerations are also important, and that study largely ignored those things.

How response affects how the bass "sounds" is anything but simple.  Putting a sub in a corner without other sub(s) in opposing corners to "cancel out" the standing waves does in fact increase the coupling of the sub to the room's standing wave resonances compared to placements away from those corners.  The advantage, IMO, is that such resonances can be attenuated using EQ.  Placement away from walls tends to cause suck-outs that aren't easily repaired using EQ, and these response issues may be harder to discern from measurements as well.  Either way, EQing a system for the best sound is much easier said than done.

It's hard to say why you were not successful applying DSP to "fix" problems heard when a sub is placed in a corner.  My first suggestion (1) here is to not obsess too much over what your in-room "frequency response" shows.   It's not "wrong", it's just not all that consistent with what you actually hear.  Instead it offers hints.   Second (2), always look at measurements from multiple locations.  Features that appear in several locations within the room are more likely to be audible.  Spatial averaging can help somewhat with selecting for these, but rule (1) still applies.  Third (3), use EQ judiciously.  You're trying to EQ out problematic features, after you've identified them, not make a curve look prettier.  The data you get from measuring is part of a map, not the territory.  Fourth (4), listen to the result and use your ears to make the final judgment. A prettier curve is no guarantee of better sound.

So for example, you might look at your measurements and notice a narrow peak at the same frequency that appears at most locations.  Then you might choose a Q or bandwidth for your filter, based on the shape that appears in a spatial average.  You might further select a gain that's something like half of the spatially averaged peak, listen to evaluate the result, and then iterate on the gain until the character of the resonance can no longer be heard but not so much that the sound becomes lifeless in that frequency area.

Apart from the obvious resonances, you can experiment with broad shape adjustments to improve octave-to-octave balance.  This might be best done by ear, unless you are correcting some aspect of the sub that you know about from outdoor measurements or something.  For example, you can add a Linkwitz transform to alter a sealed sub's roll-off/Qts to different values, albeit with very substantial effects on *power* and *excursion* requirements.  If you know roughly where your subs exhibit an "inductance hump", then a broad EQ dip might be helpful here.

To the extent that you are able to use EQ to reduce or eliminate audible problems, you can make significant improvements to your sound, but you will likely find that other problems still linger.  The consequences of those problems are varied and complicated.  Your changes might lead to much improved clarity, but you might lose a lot of "slam".  Or maybe you get more impact but a "colder, harder" sound.  Almost every good sound system gets hand-tweaked at the end of the process, so don't hesitate to experiment, especially with broader scale EQ adjustments.

On 10/15/2020 at 7:27 AM, Arcsabre said:

In my case I have 4 units of Dayton UM15-22 subwoofer driver. The amp I've prepared are not subwoofer's plate amp but Hypex NC500 monoblocks. I'm using one for each subwoofer and these amps can output 700W clean power stably, for each channel. I've got really inefficient floorstanders Dynaudio C5 (sensitivity=83db) and these monoblocks drive them easily ------ only need -30db from preamp for really loud symphonies I compared with my friend's Dynaco and Krell, and they cannot get much bass out from the C5. so with the Hypex amps power isn't really an issue here.🏋️‍♂️)

I think you are confusing *gain* with *power capability*.  The volume setting on your pre-amp says nothing about how much headroom you have in your amps.  It's possible that your "really loud symphonies" are actually clipping and distorting on some of the bigger hits.  It can be hard to tell unless you are able to A/B the difference unless you're really overloading things badly.  With that said, 83 dB is very insensitive.  For a "reference" home theater experience, where you are sitting ~10 feet away, you'll probably want speakers  closer to 93 dB sensitive.  Mind you, a lot depends on details --- are we talking 2.83V or 1W?  What do the speaker's impedance and FR look like?  IMO, speaker sensitivity (assuming use of subs) is most important in the 120-700ish Hz range where you're most likely to see big suckouts due to baffle and boundary problems, especially where they coincide.  Either way, I don't think 83 dB will not cut it unless you plan to listen at lower volume.

The other thing you need to keep in mind is that each +3 dB "costs you" twice as much power.  So while 700W sounds impressive, it's only 8.5 dB higher than 100W.  You could easily see that by putting the sub in a corner vs. mid-wall.  Or put another way, moving the sub to gain 6 dB would be similar to *upgrading* your 700W amp to 2800W, *and* you'd have less distortion.

On 10/15/2020 at 7:27 AM, Arcsabre said:

After reading your inspiring response I think there are three ways I could use the drivers in this room and they are:

1. 2 sealed subs, with one driver on each enclosure. Both subs on the floor, one located at mid-backwall, one located at mid-frontwall. But as there's a french door in front the front sub can only get so close to it (there will be a gap at least) and my impression is this is a kinda embarrassing distance from the wall, not far enough to avoid reflection issues but also not close enough to solve them in another way.

2. 2 sealed subs, with one driver on each enclosure. Both subs on the floor, located below the 2 bass traps (see the floorplan). fit snugly against both walls of the corner.

🏋️‍♂️3. 4 subs mount directly onto very thick MDF - such as a 3" thick one made of 3 sheets of 1" with green glue between them. 4 MDF board mount onto the midpoint of each wall, for side walls ------ 1 meter from the floor and 1 meter from the ceiling. For backwall ------ floor. For front ------ Above french doors. Use pure amp power to push them hard enough so they work to desired SPL without enclosure.Kinda like IB subs but I don't have that much space in the wall.

 

Do you have any way to take measurements before you design the room?  That would help a lot.

Number 3 sounds sketchy.  They are essentially IB subs if their back-ends fire into open space that is isolated from the room.  However, I have no idea what your giant MDF contraptions are going to do mechanically, and you could end up losing a lot of efficiency among other serious problems that way.  It's definitely best for the sub drivers to be mounted rigidly.

I lean towards your option 2.,putting the subs in the corners under the bass traps.  If you are sitting more toward the back of the room, then the subs will be closer to you than the front wall, so hopefully that lengthwise null with be a fair ways in front of you.  The opposing corners placement will cancel out the major width-wise standing wave.

Is there any reason why both (1) and (2) only involve using 2 subs?  Why not use all 4?  Why not do (2) and also put two subs at the front of the hall, at the midpoint and on the ceiling if need be. 

 

[kipman725]

I would go for corner placement for the reasons SME states but if you could do 1+2 this would be better.   Regarding eq/room correction the best curve tends to be genre/album specific due to the lack of standardization in the frequency response of control rooms so quite often you have a situation where one EQ sounds better with one genre but worse with another.   

[SME]

6 hours ago, kipman725 said:

I would go for corner placement for the reasons SME states but if you could do 1+2 this would be better.   Regarding eq/room correction the best curve tends to be genre/album specific due to the lack of standardization in the frequency response of control rooms so quite often you have a situation where one EQ sounds better with one genre but worse with another.   

I'd argue the situation is a bit more complicated than described here.

First, I agree that there's a lot of variation between content with regard to bass balance---much more so than for higher frequencies.  I'm not sure how much correlation there is in this regard within of specific genres, but I'd certainly expect to see some correlation of content produced put out by the same mixer(s) and/or mastering engineers.

Second, the subjective spectral balance depends not just on the very broad aspects of the shape, but also on finer aspects and on how these different aspects interact.  Where peaks at different scales coincide, there is often enhanced emphasis.  Start with the spectral characteristics of the original content, before any significant EQ is applied in the production process.  The spectrum depends entirely on what the sound is.  The sustained part of continuous sounds exhibit very narrow spectral characteristics.  The transient sounds and attack/release of continuous sounds exhibit broader spectral characteristics.  (Even this description is over-simplified)  Then the mixer and/or mastering engineer applies various EQ, usually of a broader nature when boosting.  Then *you* play the track at home on a system with its own characteristics (some broad, some fine), which may have been altered via your own EQ (or auto EQ) settings.

The end result of all of the above can be rather hard to predict.  One track might seem to have wild bass in one room or one configuration but seem feeble in another.  However my experience suggests that the more neutral the playback system is, the less apparent variation there is between content.  One hypothesis I have is that the brain is capable of compensating for "EQ flaws", but only to a point.  The processing resources are limited.  So if you can minimize EQ flaws in the production process, the resulting soundtrack will likely "translate" well to a variety of flawed playback systems.  And if you can minimize EQ flaws in the playback system, you can substantially increase enjoyment of a wide variety of "flawed" content at home.

Likewise, a very "dry" recording of bass is likely to sound pretty clean when played back in a typical small listening room, albeit with the listening room's acoustic signature imposed on it.  A recording of bass in a room with a strong small-room acoustic signature will also sound pretty clean when played back on a system with no LF reflections or one that is tightly optimized in a way that erases or neutralizes its acoustic signature, and the recorded acoustics will be reproduced faithfully.  However, the same recording will likely have poorer intelligibility when played on a more typical small-room system.  Perhaps the brain struggles to parse LF sound from acoustic effects when the acoustics of multiple rooms impose on the sound.

One other point of mention is that almost all mastered content is processed with harmonic enhancement plugins that I believe have a "neutralizing effect" on the overall sound and likely suppress the negative effect of any EQ flaws.  They may also make those EQ flaws more audible to the engineer, making it more likely that they'll be corrected before the track is finalized.  I don't know for sure as this is not my area of expertise, but they definitely have an important impact on how the end result sounds.

[Ricci]

On 10/27/2020 at 5:01 PM, SME said:

I'd argue the situation is a bit more complicated than described here.

First, I agree that there's a lot of variation between content with regard to bass balance---much more so than for higher frequencies.  I'm not sure how much correlation there is in this regard within of specific genres, but I'd certainly expect to see some correlation of content produced put out by the same mixer(s) and/or mastering engineers.

It is complicated.

The key word here is subjective. 

IMHO there really isn't a wrong way. All that matters is if you like the way it sounds. I'm not necessarily talking about just bass balance or response shape with these comments BTW. We've all heard people wax poetic about the goal being to be transported to the live performance. This is an impossible fools errand IMO. The live performance from who's perspective? Front row, stage left, right, dead center? Most recordings these days are created with a totally fabricated sense of space and "venue" by the recording engineer anyway. A lot of the content is totally computer created. You could say that you want to recreate what the recording or mastering engineer heard but this will be very difficult too. Most studio control rooms are as variable as our own systems. I've been in my fair share of them and very few of them are totally flat on axis, with controlled directivity. The amount of room reverb and the room size vary a LOT too. Mixing balance, quality and artistic expression is all over the map on top of that. If you personally like more bass, less treble, a mountain of mids, whatever...Do it. 

[SME]

On 10/29/2020 at 3:16 PM, Ricci said:

It is complicated.The key word here is subjective. 

There's a lot of unpack here, and I  mostly agree with you but can't resist the opportunity to say more.  I also want to expand on the idea of subjectivity and describe at least two distinct forms of it.  First there is aesthetic subjectivity, which has everything to do with our preferences for music types, artists, etc.  The creative process is all about subjectivity, and there's no general theory that will suffice to explain or predict this kind of subjectivity.  On the other hand, there is the kind of subjective preference that tends to correlate amongst a universal audience.  More likely than not, this has to do with the *sound quality*, which is distinct and likely possible to describe objectively, even if a complete explanation is not yet available.  Then there's a third kind of subjectivity that arises when one must choose between several "flawed" options.  We might all agree that a particular bass resonance and a particular treble resonance are offensive, but we might disagree strongly as to which is *more* offensive.

IMHO there really isn't a wrong way. All that matters is if you like the way it sounds. I'm not necessarily talking about just bass balance or response shape with these comments BTW,

 

I think at the end of the process, this is absolutely true, and no matter what other advice one is given from experts or "the data" or whatever, one's subjective preference should always rule.  And this applies for both kinds of subjective preference described above.   I might also note that in a live performance setting, the two domains of subjectivity blur together.  The FOH engineer is working to optimize the sound both in terms of objective quality and in terms of aesthetic  goals at the same time.

We've all heard people wax poetic about the goal being to be transported to the live performance. This is an impossible fools errand IMO. The live performance from who's perspective? Front row, stage left, right, dead center? Most recordings these days are created with a totally fabricated sense of space and "venue" by the recording engineer anyway. A lot of the content is totally computer created.

I *think* I know what you're trying to say, but as written I must strongly disagree.  I know in the audiophile world there is a kind of obsessive pursuit of certain ideals in recordings and reproductions:  realism, authenticity, and purity.  And I agree that this is not what audio as an art-form is about at least 99% of the time.  That's because we like making things sound better when we can, and we will almost always *embellish* things when we can.

Some thousands of years ago, humans realized that acoustic spaces could be constructed for the purpose of dramatically altering and enhancing the quality of sound produced within them.  Later we built elaborate concert halls for our performers.  Then we had the recording revolution, and later digital technologies.  No longer were we limited by concert hall acoustics but could explore sound in a wide variety of real and simulated spaces.  So this only expanded the creative possibilities.

And that's why I very much disagree with what you literally wrote.  "Transporting the listener" to the live performance *to the sound-scape created by the artist* is ... absolutely not a fools errand!  If it was, then it'd also be a fools errand for the artist to go through the trouble of deciding things like "who's perspective?" or what kind of "fabricated sense of space..." to create.  That is part of the creation and contributes to the aesthetic value of the product.

You could say that you want to recreate what the recording or mastering engineer heard but this will be very difficult too. Most studio control rooms are as variable as our own systems.  I've been in my fair share of them and very few of them are totally flat on axis, with controlled directivity. The amount of room reverb and the room size vary a LOT too.

Trying to recreate what the recording or mastering engineer heard---now *that's* a fools errand.  I doubt it's possible to achieve sufficient uniformity between systems, but if it were, we wouldn't care what the recording or mastering engineer "heard" anyway.  That's because what's actually important is *what the artists/engineers actually DID*.  Creative actions arise from artistic intent, and the consequences of these actions are objectively "recorded" in the final soundtrack.  Even though the artists/engineers might have made slightly different mixing choices, were they to monitor their work on more accurate speakers, the end result is still predominantly a product of choices made without needing to rely on guidance from the monitors.  I want to add some jungle ambience to my mix.  If I'm on Yamahas, I'll end up mixing end frogs.  If I'm on JBLs, I'll mix in birds instead.  Yeah right, that's totally absurd!  Furthermore, the objective sound quality of a production is by no means limited by the sound quality the engineers heard while creating it.  Not even close!  Relatedly, I strongly disagree with Floyd Toole's "Circle of Confusion" analogy.  There is no "Circle" because the relationship is completely asymmetric.  Engineers don't EQ the inverse of their monitors.

The real problem I see with less-than-ideal audio reproduction is not that the different reproductions are inconsistent but that in one way or another the reproductions all obscure and/or distort the objective truth---the ultimate product of artistic intent---that is "recorded" onto the soundtrack.  The issue is not really about whether the bass sounds precisely as loud as the artist wanted it to sound and more about whether every possible tonal, timbrel, and temporal nuance of the bass instrument and the performer's command of it is conveyed to the listener.

Mixing balance, quality and artistic expression is all over the map on top of that. If you personally like more bass, less treble, a mountain of mids, whatever...Do it. 

Yes, But!  If the reproduction system is capable of ideally neutral reproduction,such that the objectively subjective aspects of sound quality are essentially perfect, then how much does "more bass", "less treble", etc. actually matter?

One of the most surprising things I've experienced in my work is that as I seem to improve the finer-scale accuracy of my sound, the broad-scale frequency balance differences between content (between program as well as within the same program) become much less exaggerated and more uniform.  This is very fascinating because many of us who study audio assume that what we perceive should correlate fairly closely related to the physical characteristics of the sound.  However, what I'm consistently finding is that what I actually perceive has much more to do with *informational characteristics* of the sound.  To my brain, a broad-scale response imbalance is essentially a nuisance to be adapted to in the process of extracting as much useful information as possible from my sensory inputs, and it seems that the less my brain has to concentrate on adapting to other issues in the reproduction, the more readily my brain "flattens" the broad differences.

And to be clear, the broad differences don't vanish entirely, they just manifest a lot less dramatically.  If these insights are ultimately confirmed to be true, they open up a lot of fascinating questions.  Like, how does SPL play into all of this?  If frequency response optimization causes relative level difference between bass and treble to matter much less to perception, then can people get satisfying bass using fairly menial sub capabilities?  And what does a steeper broad-curve with more bass contribute to the sound?  More physicality?  How much does that aspect increase per dB?  A lot of bass-heads probably habitually turn the sub up until right before they hear "too much bass", likely related to the amount of mud becoming intolerable.  But what if the sound never reaches that point?  What if a whole wall full of subs aren't enough to muddy the sound (at reasonable mid-high frequency listening levels) even with their headroom maxed out?  Will the bass-heads be dissatisfied because they are able to clearly hear all the mids and highs no matter what they do?

[kipman725]

Regarding the mud issue sometimes in a high headroom system at high SPL (and you don't want to increase the SPL further) putting in a compressor with a medium knee that engages during the loudest passages to increase the harmonic content of the bass is worthwhile as the expectation is that the sub bass system will be limiting (intentionally or not) during those passages.  Without the limiting it can sound a bit weird.

Its all really easy to do with modern DSP!

 

[SME]

16 hours ago, kipman725 said:

Regarding the mud issue sometimes in a high headroom system at high SPL (and you don't want to increase the SPL further) putting in a compressor with a medium knee that engages during the loudest passages to increase the harmonic content of the bass is worthwhile as the expectation is that the sub bass system will be limiting (intentionally or not) during those passages.  Without the limiting it can sound a bit weird.

Its all really easy to do with modern DSP!

Is your suggestion specific to live concert music presentations?  I'm wondering because I believe mastered music usually has lots of tricks like that applied already.  Of course, most mastered music doesn't much rely on "headroom" either. 

Also, my system has high headroom, and when I turn music up to high SPL, I don't usually notice mud that wasn't there before.  If anything, more SPL usually helps a little bit.  I reckon that mud is the cause of that audiophile trope: "those speakers really open up when played louder."  Indeed!  Or maybe the sound is just muddy.  Likewise: "those speakers sound very good when played quietly" is probably just a nice way of saying that they are thin and become irritating at high SPL.  Audiophilese. 

[kipman725]

I meant in a DJ playback situation to increase the mud intentionally but only during the most bass heavy parts, using a long attack time.  I guess in audiophile terms to add 'warmth'. 

well live I would always have a compressor on the bass drum mic and possibly, depending on mic technique on vocal mics.

Good speakers you should always want to turn it up, only wanting to listen quietly indicates faults.

[SME]

On 11/9/2020 at 4:06 AM, kipman725 said:

I meant in a DJ playback situation to increase the mud intentionally

Thanks for the clarification!  DJing is an interesting case because you're using content that was mastered but you're also essentially doing a live show.  And to make it more bizarre, it's like you're working with a different band every time you switch songs.