Improving acoustics in the concrete bunker

Thank you for the long writeup!

When we moved in, I tried convincing the better half that the sofa should be further into the room, and she could have some table with flowers etc behind it, but it did not fly. Maybe I should do another round.. I know it would be a better position, also for stereo imaging, the distance between speakers is short compared to the distance to listening position. And being able to add some absorption behind the sofa would be nice too.

The 'foot rest' I mentioned before is actually a separate part not in the pictures. It's like a free standing cube of maybe 80x80x45cm. I can't really remove anything from the sofa as shown, it's made of two similar sized pieces. I think that can be seen on the pictures where I have it apart and flipped over.

I have bad experiences with glass/rockwool, experiencing some asthmatic symptoms from the dust, working in spaces with this stuff when I was working as an electrician during summers while studying, so I'm not prepared to take the risk of using it indoors. I have used in sealed speaker boxes sometimes, but I prefer not to use it/touch it at all.
I have been looking at insulation made from sheep wool, like this: https://www.isoleringsbutiken.se/isolering/farullsisolering/isolena-optimal-100mm

They are maybe 1h drive away, so I could drive there and buy some of that.
I have not found any data on those, but they appear very 'fluffy', so I guess I would need a thin layer of 'rigid' high flow resistivity material on the surface to improve the LF absorption? I have used sheeps wool in speakers before, and it has been very good for that.
It seems your suggestion is the other way around, use rigid rockwool absorber with a fluffy layer of polyester around it, but as I understand the paper you linked to, the other way around should work too?

It does not seem promising to get more of the 'recycletherm' so it seems I need to find some alternative.
The wool-based is the only alternative I have found that I would be prepared to use indoors so far. It does not appear very good for wall absorbers though, put in a frame with some fabric around it, it will most likely look like a pillow, pushing on the fabric..
 
Sink sub? Inverted with some advanced filters to 'suck up' the modes of the room?

Yes, I have a sealed 10" sub for the TV at the front wall to the right, there is not enough space to put it behind the TV bench as is, but I could possibly make some room if needed. It's a bit stupid because the sub has the plate amp on the side, so it takes up more space in that direction with cables sticking out etc. There was too much bracing and stuff, so it was easiest to put the plate amp on the side -'Dough', quoting Homer Simpson.. :)
I don't know what happens with ground loops if the sub is connected both to the AVR and Flex Eight. Should be doable with one source connected to right input and the other to left in principle though.
I kind of wanted to avoid mixing in subs etc, it's a mess setting up, moving something requires a lot of tuning again etc. I tried it in the past and got tired of it..

I was hoping to do something simple to improve the room, but it seems there is no way to do that.. Massive absorbers or multiple subs with a lot of cables and tweaking/tuning etc. I don't like any of it :)
I would rather have an absorber behind the sofa than a sub, no need for cables or tweaking parameters to the end of days..
What I found before when tweaking multiple subs in the past was that I could make it look nice in the measurements, but still I did not like the sound of drums etc. I just gave up after that. But, then again, I might have been looking for the wrong things in the measurements.
 
Last edited:
multiple subs
Even 24 subs in DBA aren't "multiple subs")). And basic "tweaking" DBA\SSS is simple. Theoreticaly wireless connection may be used; I live with 10+ meters cable)) https://www.diyaudio.com/community/threads/my-double-bass-array-with-low-q-subs.392998/
I don't listen to Metallica, but the drums in for example https://templumanimamorti.bandcamp.com/album/--2 sound good
In Norway StigErik has serious DBA.
DBA\SSS advertising https://www.researchgate.net/public...reference_of_modal_control_in_listening_rooms

 
Last edited:
Oh, I don't listen that much to Metallica either, mostly as background in the garage :) Their old records (that I like) have poor sound quality. I would say the common factor for music I listen to is that there is usually drums, bass and guitar, rock, blues, metal, jazz.. I used to play some guitar, and have spent too much time close to drum sets in small rooms (turning up the guitar amp to hear it), which was no favor to my hearing. I try to avoid loud music these days to preserve my hearing as much as possible. Your music link was blocked here at work.

What is DBA/SSS?
 
Uncompressed sheep wool has a low air flow resistivity but so far I haven’t found any measured figures and probably will not. -After all it is not a standardized material like glass wool or rock wool. Just because sheep wool is a natural product it doesn’t necessarily make it a good material for some people. -There are many who are allergic to cats, dogs, horses etc and react to various proteins in their hair. If you go for sheep wool I would make sure it has been washed and cleaned. Sheep wool, hemp and hay can have smell which is another thing to keep in mind. I wouldn’t like my living room to smeel like a barn. Man made insulation used to smell bad, less bad nowadays as urea and formaldehyde as binder is used less than before.

When / if you compare air flow resistivity between different materials, one thing to keep in mind is that even though they may have the same figure for it, it doesn’t mean they are really equivalent in effectiveness as an absorber. Wood wool mixed with cement (Träullit in Swedish) has been around for decades as absorbers in gymnasium, swimming halls, corridors etc. The wood wool is very large in size compared to glass wool fibre. So a sound wave passing through say 10 cm thickness mean there will be many more collisions for air molecules <=> fibre in the glass wool. Each collision means a transformation of energy into heat = absorbtion. (In oher words, glass wool and rockwool is more effective as an absorber material for the same thickness compared to other material with fibre which are larger in size. 1 liter of ”fine fibre” has a much larger total area reacting on the sound wave than 1 liter of ”large fibre”.)

For choice of best material, much depends on the air flow resistivity in combination with thickness and frequency of interest. Check out that Brandt-link I inserted in previous post. You can also play with http://www.acousticmodelling.com/ If you build a desk / shallow box behind the couch, try out http://www.acousticmodelling.com/multi.php with some slotted wood panel in front. I would increase ”Number of absorbers to be modelled” from 2 to 4 in the program.

I enclose a pdf on the paper ”Subjective Preference of Modal Control Methods in Listening Rooms ” which Flaesh mentioned. (I have a DBA in my room and it it works well as explained in the paper. In small rooms I would advocate DBA or SSS as it saves a lot of room volume if room modes are bad. There is a cost yes, but available room area / volume carries a cost too. Pick your poison ... ;))
 

Attachments

  • Perception_Modal_Control (3).pdf
    351.4 KB · Views: 105
  • Like
Reactions: 1 user
I read the paper now, and I think I get the idea of anti-phase in the back of the room. However, In my case I'm sitting just by the rear wall, and the study has the listening position in the middle of the room? Any delayed opposite phase sub would just kill the bass sitting next to it? Even if I move the couch a bit, it will probably be less than a meter into the room. I will not be able to have the sofa placed close to the center of the room, no matter what..

Sitting next to the rear wall, the peaks and nulls from the rear wall reflections will be higher up in frequency, and I'm thinking that the standing wave ringing along the length of the room heard at this listening position will be due to the reflection bouncing back from the front wall? So, removing the first reflection from the rear wall, would be removing the bass at the listening position? Removing the bounce from the front wall (double delay) should improve LF decay and FR in a listening position close to the rear wall?
I'm thinking main speakers could be used as is (full range) and the sub could be run in opposite phase with a delay equal to two room lengths, and gain adjusted to 'whatever works'? This would be easy to do as is in my case.. if the Flex Eight can do such long delays..

@Adhoc1:
DBA? Short for what? Same paper?
The sheep wool is supposedly washed and treated in some way to prevent insects etc. Could probably be 'compacted' a bit to make it more dense, but maybe it's still no good. The hemp-based is not performing well enough I think.
I do have some hay and cat allergy too.. No easy choices.. Cost is a factor too, the sheep wool stuff is already a lot more expensive than glass/rockwool, and if I was to use a lot of basotect ... $$$!
Could you explain or link to an example picture of the type of slotted absorber you recommend? Most slotted/perforated seem to be some kind of Helmholz resonators.
 
I'm starting to think that maybe doing the 'general areas' like side reflection points and front wall would be easiest and cheapest just hanging some rugs with some distance to the wall, or over a wooden frame, as mentioned earlier in the thread? I'm thinking if it's fairly heavy, thick and has a soft surface, it should work fairly well?

I have been looking at alternative insulation as absorbers, but there is no data. There is some flax-based insulation, some wood-fiber based, but there is no data.. they only mention that the flax-based has good sound insulation properties.
So, it seems it's easy to spend a lot of money and time building something, and end up with something that is useless/ineffective.
 
Insulation material made from wood fibre are made in Norway and Finland. I believe the Hunton brand (Norway) can be bought in Sweden. Link to some technical data: https://hunton.se/wp-content/uploads/sites/17/2019/04/Ytelseserklæring_Hunton-Nativo-Trefiberisolasjon-Plater_se.pdf As often for thermal insulation it says NPD for gas flow resistivity / "luftflödesmotstånd" ... (NPD = No Performance Determind.)
If you google "Säkerhetsdatablad for Hunton Nativo, you will see there are no hasardous ingredients but it is mentioned it does have a smell.

(I used the glass wool URSA 37 from from Byggmax. It has very little smell when unwrapped and the smell went away quickly. I f you are concerned about fibres flying around, it could be a problem only during cutting and installation. Wrap it up with that dacron non woven material with 3 kPa x s / m² ("täckjackfoder") and there will be no fibres flying around in your room later on. You could also use some thin plastic film, that would not influence absorbtion of sound waves in the bass and mids which have a high energy. Air molecules hitting a light weight film / surface will make it vibrate and the air molecules on the other side will start to vibrate with the same frequency and energy be transformed into heat in the insulation -The reason why light weight walls passes through bass waves more easily than heavy and stiff concrete.)

DBA = Double Bass Array is the very same concept as C.A.B.S. = Controlled Acoustic Bass system. With more than 1 bass source and using opposite phase and walls + suitable time delay between them, you "kill" the build up of (some) resonant bass modes in the room. DBA would "kill" at the first bounce from the first source hitting the back wall. It may be harder to adjust with mains playing full low frequency + a delayed sub (corresponding to 2 x room length) at the front wall. If you can, why not try.

A DBA (or subs placed at different positions in the room) does not "remove" and make bass inaudible at listening position, it removes or lessens the build up of resonant modes. So, the bass will be more even in the room, not extremely strong and loud at some places and extremely weak at other.

I will not advice on a specific slat absorber. A suitable one depends on your room and how it "behaves". As you write, a slat asborber = a Helmholtz one has a sealed box. It could look like in this link: https://www.bauhaus.se/akustikpanel-bath-deluxe-silentium-obehandlad-ek-21x600x2420mm (That one is quite useless for low mode frequencies though as the the gaps are large compared to the slats. You need to go down well below 10% for openness. Play with that program in previous post.)
 
Last edited:
In my case I'm sitting just by the rear wall
:
moving the listening position forward is highly desirable
I read on a local forum about some setups with a sofa against the back wall, but I did not listen to them and do not believe in a good enough audible result.
My first system was 2+2 DBA in a 4.2x4.8 room with seats about 1.2m from the back wall and that's good, rear subs was inaudible. Someday I will add measurements in few points to my tread https://www.diyaudio.com/community/threads/my-double-bass-array-with-low-q-subs.392998/
fibre which are larger in size
IIRC somewhere at GS was link to research of GFR of polyesters with different fiber thickness; thicker fibers material had higher GFR at equal bulk density of materials. I would like to know the thickness of the CIB fibers)). IIRC Gernot wrote that CIB is not exactly the same as normal PET.

DBA at ASR, https://www.audiosciencereview.com/...-low-q-subs-quasi-plain-wave-radiation.24396/ and https://www.audiosciencereview.com/...of-the-double-bass-array-configuration.37943/
 
DBA = Double Bass Array is the very same concept as C.A.B.S. = Controlled Acoustic Bass system. With more than 1 bass source and using opposite phase and walls + suitable time delay between them, you "kill" the build up of (some) resonant bass modes in the room. DBA would "kill" at the first bounce from the first source hitting the back wall. It may be harder to adjust with mains playing full low frequency + a delayed sub (corresponding to 2 x room length) at the front wall. If you can, why not try.
I did a small experiment this evening until I ran out of batteries in the mic preamp. So far, it looks to make some sense in measurements. Evens up frequency response and shortened the decay around 35Hz, which is the standing wave between front and rear wall.
I was playing around with delays, and ended up with much shorter delay than I expected somewhere around 20ms seems to work, but I was expecting abt 30ms with the room abt 5,5m long. Playing around with a few ms back and forth had some positive effects in some areas, and negative in other.. I was trying to tune it for maximum effect when the battery ran out, so now I put on some music instead, and I had to turn back the gain on the sub. Also played around with LP for the sub, and it seems just 1st order starting at 30Hz blends in best, Tried LR12 & LR24, but the steeper filters made it sound boomy, I'm guessing it's phase integration problems to the main speakers since I tuned the delay with only the 1st order filter. I also have a shelving filter to further compensate for the low end drop off of the sealed box.
I have now dialed it down to a level where it blends in better, and seems to improve the sound when it is unmuted. I'm not 100% I like it though, will need to give it some listening with different music I think.

I attached some of the REW measurements, maybe somebody can tell me more about what to look for :) I mostly looked at FR and spectrogram, where it was mostly the long tail around 35Hz I concentrated on.
While listening and writing, I have already tweaked the 1st order LP and gain, so the change with the SW on is just quite subtle now. I can still feel the 10" cone move a bit at moderate level though, so it's doing something :)
 

Attachments

  • inverted sub test.mdat
    1.2 MB · Views: 63
:

I read on a local forum about some setups with a sofa against the back wall, but I did not listen to them and do not believe in a good enough audible result.
My first system was 2+2 DBA in a 4.2x4.8 room with seats about 1.2m from the back wall and that's good, rear subs was inaudible. Someday I will add measurements in few points to my tread https://www.diyaudio.com/community/threads/my-double-bass-array-with-low-q-subs.392998/
I could move the sofa forward a bit, enough to get some absorber or sub behind it, but it will never be in the middle of the room, not even I would like having it there :)
Possibly if measured from the windows to my ears I could stretch it to 1m.
 
it will never be in the middle of the room
It doesn't matter and isn't needed. Imagine a travelling wave. In reality, it will be less than ideal due to furniture and openings. And you need some (the more drivers, the less) distance from the transduser to form a wave. Please read my links)), there are illustrations.
You may play with REW RooSim and see how it works

p. s.: ideal case:
1671047323035.png
 
Last edited:
You can also check how things and ringing turns out with the sub in opposite phase versus in phase with the same time delay applied. I looked quickly at the tab “RT60 Decay” and your measurement no 1-frontar and the other measurements with frontar + inverted sub with various time delays. It seems “funny” that around 31,25 Hz the T60 varies so much, between 1,036 seconds for measurement 10 and up to above 2 seconds for the other measurements. Below are pictures from no 1 and no 2. If you move the cursor from low to higher frequencies in the lower diagram, you can see how Model T60-figures changes depending on frequency at the top.

(In a frequency diagram ist is common to show 50 dB difference between top and bottom. So I changed it to max 110 and lowest 60 dB)

In the Waterfall tab and Decay one, one can notice your room modes with long tails in the Waterfall at around 31,4 Hz and low decrease in dB over time in the Decay tab.

When I am checking if moving the couch / speakers / toe in of speakers makes a difference for early unwanted and strong reflections, I use the tab Filtered IR and set the time scale to minus 0,001 to 20 ms. In mixing rooms for music, a common goal is no stronger reflections than -20 dB within the first 20 ms. -To hear what is actually on the recording without the room audibly influencing much. That is not easy to accomplish in a small living room, so one may have to settle for -10 dB within 10 ms or so. (Some people enjoy quite strong early reflections though, as the soundstage may seem wider, a personal choice.)

Edit the 1st picture is measurement 1, then measurement 2 and then measurement 1 in Decay. I see you need to open them in a new window to read the small text and figures..
 

Attachments

  • Skärmbild_20221214_210252.png
    Skärmbild_20221214_210252.png
    168 KB · Views: 57
  • Skärmbild_20221214_210612.png
    Skärmbild_20221214_210612.png
    166 KB · Views: 53
  • Skärmbild_20221214_221208.png
    Skärmbild_20221214_221208.png
    157.9 KB · Views: 54
Last edited:
Missed this. Delay must be room lenght (may be ajusted slightly), rear level slightly less than front.
IME first modes and second lenght suppressed effectively.
View attachment 1119653
I'm just using the main speakers and connected the sub at the front wall, so delay should be double the room length I think. It seems to work to some extent, it does improve the frequency response, and the decay to some extent.

I'm planning to do some more testing and measurements later today. I think one reason for the RT60 variations could be low SPL levels while measuring, and varying background noise.

I also found that changing the spectrogram settings to wavelet and 1/24 resolution gives a lot more detailed picture of the decay.
1671090842636.png
 
Last edited:
I think one reason for the RT60 variations could be low SPL levels while measuring, and varying background noise.
Yes. This can be verified. For example
1671103064722.png

-6 (110-115 dB at LP) is compressed and distorted and #1 is near wall.
the spectrogram settings to wavelet and 1/24 resolution gives a lot more detailed picture
Yes. And lower than 1/6 resolution may cause artefucks in peak energy time. Smooth 1/6 peak energy time graph (in combination with other parameters) seems like a legitimate goal, correlated with the audible impression
just using the main speakers and connected the sub at the front wall, so delay should be double the room length I think
Clear. It doesn't work well even if amplitude and phase responce' are identical. I've read about hobbyists adding an inverted and delayed LF signal electrically to speakers and called this "DBA"
 
I did some more measurements with the sub at the front wall, but no really conclusive results. Seems phase alignment is an issue. Trying to run it in phase should boost the standing waves I think, but it seems different delays are needed to 'align it' depending on the frequency. I have not bothered trying to measure phase/group delays at the relative positions/speakers and try to puzzle that together some way, I would most likely fail and be in a bad mood afterwards.
Subjectively I think there is some improvement to be had since I can even out the frequency response in the listening position, but at the same time the bass sounds a bit 'muddy', I'm thinking it has something to do with the delayed signal mixing with the direct signal.

I also tried moving the sofa, but there is not so much to gain within the limits I have. I tried moving it forward 15 and 30cm, and the 15cm move seems a bit better compared to the original position in the 100Hz range. #0cm forward introduces new problem, and the overall result seems worse. I base this on looking at the RT60/clarity curves and FR. With the sofa moved 15cms, my ears are now about 1m from the windows behind me.

I also measured with the door to the bedroom (front right) closed, but this created another standing wave around 26Hz, possibly the 15Hzresonance was 're-tuned' to 26 when closing the door. I changed the sweep to start at 20Hz, so I did not see if something happened below 20Hz. There was also more problems around 100Hz with the door closed.

In general, looking a the reverb in the room, it's a mess, or should I say way too much reverb. like 200ms and up.. Lowest in the treble and slowly rising until it goes totally crazy in the bass.

I have now moved the sub to the rear window behind the LP, and just been able to get a cable to reach the sub, hanging in the air across the living room.. will try some measurements on that next..