Very bad room acoustics (40hz resonance, bad reverbation time), what would you recommend?

[Keith_W]

How do you see the RT60 is not a reliable indicator?

RT60 = "Time for a reverberant field to decay by 60dB after the early decay time". "Reverberant field" and "60dB decay" do not make sense in small rooms.

First, the 60dB decay. The typical noise floor in a listening room is about 40dB. This means you need to measure in excess of 100dB to obtain a 60dB decay. Not a problem for a pro audio speaker, but typical domestic speakers have trouble. So we typically measure 20dB decay or 30dB decay and extrapolate it to 60dB (T20 and T30 respectively).

Now "reverberant field". All frequencies form room modes. Long wavelengths form fewer, and they create well defined patterns. Shorter wavelengths form hundreds of them. The definition of "reverberant field" is "thousands of overlapping room modes" that is the same SPL everywhere it is measured in the room. In a concert hall or basketball stadium, the RT60 remains the same no matter where it is measured. That's not true of your listening room - it is too small to form reverberant fields in all but the very shortest of wavelengths. You can observe it in your two RT60 measurements - see how they are different to each other. So it is more accurate to call it RDT - "Reflection Decay Time".

I don't know how to interpret REW results, I guess that in that small room I should go for a sound absorbers rather than diffusers? What would you recommend to put as a absorber on the rear wall? I'm thinking of 8 60 cm x 60 cm modules for absorbtion, maybe the top 2 of them on a sides with build-in scatterer? I'm affraid that with added curtains and carpet I'll absorb too much of highs.

Regardless of what I said, we can still observe a general trend that your RT60 is too high. You should try to bring it down by about 50ms. This can be accomplished with furniture alone (as per Toole) but if not, I would suggest some absorbers behind the listening position since that is the closest wall to the listening area.

You are quite right to worry about creating spectral distortion with inappropriate treatment . It looks to me as if your problem frequencies are from about 300Hz upwards. Absorbers need to be 1/8 as thick as the lowest frequency you want to attenuate. 300Hz has a wavelength of 1.14m, so the absorbers should be about 14cm thick. As you may be starting to realise, 14cm is already very thick and it would have to be unbelievably thick if you wish to absorb lower than this. I would start with 2 panels and add more if you need it.

As for the low freqs, once again - forget Helmholtz resonators. DSP is by far the superior tool.

 

[atomek]

You are quite right to worry about creating spectral distortion with inappropriate treatment . It looks to me as if your problem frequencies are from about 300Hz upwards. Absorbers need to be 1/8 as thick as the lowest frequency you want to attenuate. 300Hz has a wavelength of 1.14m, so the absorbers should be about 14cm thick. As you may be starting to realise, 14cm is already very thick and it would have to be unbelievably thick if you wish to absorb lower than this. I would start with 2 panels and add more if you need it.

What do you think about this plan?

1) I'll start with courtains to cover the left side windows and take care of first reflections from the left and dampening mid-highs. According to ChatGPT 2 x 2m x 3m would add ~ 14 Sabins of absorbtion, works in 200-5khz band and should reduce RT60 ~ 0.15 s in this region and it would make biggest overall change to acoustics (too bad I can't tread right side, but maybe one absorbtion panel at first reflection spot I can fit)

2) Measure. Depending on the results (not to overdampen mids/highs), decide on the dampening on back wall. I think must have is at least two 60cm x 60cm panels (like Gik 242) behind listener position. But maybe in addidtion 4 such panels in higher row (with built-in scatterer to reduce dampening) and maybe two scatterer panels on the side of the middle two?

3) Optional - behind left column, 4 monster bass traps (70-120hz) with range limiter to focus more on bass ( https://gikacoustics.net/product/gik-acoustics-monster-bass-trap-flexrange-technology/ ) -

4) Optional - to reduce mid/high reflections from the floor, to extend carpet from 6m^2 to 12m^2 and put a 10mm felt underlay under it.

 

[TurtlePaul]

Did you try Kieth’s original advice yet? As much as it sounds like you want to shop for things, two fixes will address 95% of the issue very cheaply. Turn your sub down by 7 - 10 dB and then setting a single miniDSP filter (39 hz, - 10 dB, Q=6, just eyeballing it).

 

[atomek]

Did you try Kieth’s original advice yet? As much as it sounds like you want to shop for things, two fixes will address 95% of the issue very cheaply. Turn your sub down by 7 - 10 dB and then setting a single miniDSP filter (39 hz, - 8 dB, Q=6, just eyeballing it).

I don't have sub, the sound is from the speakers. I've tried miniDSP but eventually i've disconnected it - I didn't like the changes it added to the sound, even when I limited it to 300hz. I really need to try the bass traps (I'll probably go with membrane traps in the corners), even 2-4 dB may help.

 

[TurtlePaul]

I don't have sub, the sound is from the speakers. I've tried miniDSP but eventually i've disconnected it - I didn't like the changes it added to the sound, even when I limited it to 300hz. I really need to try the bass traps (I'll probably go with membrane traps in the corners), even 2-4 dB may help.

It is a 20 dB problem, so you are asking the absorbers to capture 99% of the energy, which is just not feasible. Can you pull the speakers out several feet into the room?

Also, miniDSP doesn’t have a sound, the filters do.

 

[atomek]

It is a 20 dB problem, so you are asking the absorbers to capture 99% of the energy, which is just not feasible. Can you pull the speakers out several feet into the room?

Also, miniDSP doesn’t have a sound, the filters do.

I know I won't be able to capture the 99% energy thats why I realistically aim for maybe 6dB - it will be about 40% quieter, so it will be perceived. miniDSP filters change the sound, as I said I've tried couple of times with calibration, I was never satisfied with the result. Yes the bass was gone, but also music become dull, unendgaging (and that was with all kind of music even when the bass wasn't a problem before, vocals etc). For me it just doesn't work. Maybe with initial correction with bass traps I'll go back to the miniDSP, but right now the 40hz is overhelming and maybe even unfeasible for EQ correction.

 

[olieb]

I know I won't be able to capture the 99% energy thats why I realistically aim for maybe 6dB - it will be about 40% quieter, so it will be perceived.

But with EQ you could easily pull down 6dB or 12dB or even 20dB in bass, no problem.
When you are talking about attempts with MiniDSP are you talking about PEQ filter dialled in manually or about some automated solution with DIRAC?
There should be no change in sound in mids if you dial in filters at 40Hz or 70-100Hz, aside from less bas of course.
The unevenness in mids (too much energy around 800Hz and to little around 1200Hz) should probably not be "corrected" by EQ/DRC but by correction room acoustics.

 

[Keith_W]

All right, let us think a bit about why you have that massive peak at 40Hz. Your room dimensions are 4m x 4.5m. The first room mode always occurs where the dimension of the room equals a half wavelength - so in your case, 8m and 9m. This corresponds to 42.9Hz and 38.1Hz respectively. In other words, the reason why you have a massive peak at 40Hz is because you have two dimensions which are roughly equal.

You asked about that GIK Monster bass trap. According to the link, it is 7.3" thick (18.5cm). I said earlier that the thickness of the foam absorber will attenuate 1/8 of the wavelength, so in this case the wavelength is 148.3cm - so that absorber will work down to 232Hz. If you want it to absorb 40Hz, it needs to be 1m (3ft) thick. Not to mention, it is very likely that (like all foam absorbers) it will attenuate short wavelengths much more effectively than long wavelengths, so it will make your room very dead sounding. Now, it's your room and of course you can do whatever you want with it. I don't think it's a very good idea.

Re: your experience with MiniDSP. I have said this so many times that ASR people will think I sound like a broken record: the no. 1 cause of bad sounding DSP is user error. To get a good correction, you need proper measurements and you need to know what to correct and leave alone. This is not easy for an absolute beginner, which is why we are here to help. I am not surprised that most people try DSP, get bruised by the experience, and don't want to try it again. Of course bad DSP will kill the sound. It's not the MiniDSP that is to blame for bad sound, it is the arcane nature of DSP itself.

I don't know what source you are using - some of them are DSP capable. For e.g. if you are using a PC as your source, you could fix that problem for free using equipment that you already own and using free software tools.

 

[raindance]

I purchased four of the GIK soffit bass traps and installed them in the corners in my previous space. In that room I had a similar issue, with two modes combining to give a huge 46 Hz peak. They made the peak decay faster but did not attenuate the magnitude in any way. They did, however. improve the mid-bass and upper bass quality.

My solution ended up using SBIR to intentionally cancel this frequency; I placed the speakers (no sub) at the distance from the front wall that cancelled 46Hz for the seating position. Unfortunately this ends up being quite far into the room for that frequency.

In my new home the acoustic qualities of the space I am stuck with are too complex for such a solution (open plan) so I use a Denon AV receiver with Audyssey and I use the app on an iPad to optimize it. I only allow it to EQ up to 500Hz on the left/right mains. It sounds pretty darn good with excellent bass quality as well as quantity. I also tried a mini DSP, in the past, (the first version, not the so-called HD) and disliked it because it had no headroom and couldn't handle 2 volt sources without clipping.

 

[atomek]

I purchased four of the GIK soffit bass traps and installed them in the corners in my previous space. In that room I had a similar issue, with two modes combining to give a huge 46 Hz peak. They made the peak decay faster but did not attenuate the magnitude in any way. They did, however. improve the mid-bass and upper bass quality.

My solution ended up using SBIR to intentionally cancel this frequency; I placed the speakers (no sub) at the distance from the front wall that cancelled 46Hz for the seating position. Unfortunately this ends up being quite far into the room for that frequency.

In my new home the acoustic qualities of the space I am stuck with are too complex for such a solution (open plan) so I use a Denon AV receiver with Audyssey and I use the app on an iPad to optimize it. I only allow it to EQ up to 500Hz on the left/right mains. It sounds pretty darn good with excellent bass quality as well as quantity. I also tried a mini DSP, in the past, (the first version, not the so-called HD) and disliked it because it had no headroom and couldn't handle 2 volt sources without clipping.

SBIR would be ideal but in my case 1/4 of 40hz is 2.15m, it would have to place speakers in the middle of the room But great to hear the traps helped, I was also advised by Gik to buy soffits but they are quite large so i'm more looking into Monster Bass Trap with range limiter to focus more on lows. And for 40hz the scopus T40 tuned traps, but 4 of them could be not sufficient. I have miniDSP flex digital version (i've put it between streamer and DAC), but I wasn't happy with results.

 

[atomek]

You asked about that GIK Monster bass trap. According to the link, it is 7.3" thick (18.5cm). I said earlier that the thickness of the foam absorber will attenuate 1/8 of the wavelength, so in this case the wavelength is 148.3cm - so that absorber will work down to 232Hz. If you want it to absorb 40Hz, it needs to be 1m (3ft) thick. Not to mention, it is very likely that (like all foam absorbers) it will attenuate short wavelengths much more effectively than long wavelengths, so it will make your room very dead sounding. Now, it's your room and of course you can do whatever you want with it. I don't think it's a very good idea.

Re: your experience with MiniDSP. I have said this so many times that ASR people will think I sound like a broken record: the no. 1 cause of bad sounding DSP is user error. To get a good correction, you need proper measurements and you need to know what to correct and leave alone. This is not easy for an absolute beginner, which is why we are here to help. I am not surprised that most people try DSP, get bruised by the experience, and don't want to try it again. Of course bad DSP will kill the sound. It's not the MiniDSP that is to blame for bad sound, it is the arcane nature of DSP itself.

I don't know what source you are using - some of them are DSP capable. For e.g. if you are using a PC as your source, you could fix that problem for free using equipment that you already own and using free software tools.

The monster bass trap I plan to put behind left spekaer (where there is no high pressure of 40hz), to reduce overall bass in 70-120hz range. For the 40hz problem I'm thinking of Scopus T40 tuned membrane traps (and put them behind right speaker where the 40hz peak is present).

As for the MiniDSP I'll give it a try once more, i've used Dirac with a limited band of correction (to bass but don't remember exactly what range). Maybe this time I'll try with manual filters. I went couple of times through measurment process, with mic on boon and tried to be precise with locations, but the final result was always disapointing to me.

 

[atomek]

But with EQ you could easily pull down 6dB or 12dB or even 20dB in bass, no problem.
When you are talking about attempts with MiniDSP are you talking about PEQ filter dialled in manually or about some automated solution with DIRAC?
There should be no change in sound in mids if you dial in filters at 40Hz or 70-100Hz, aside from less bas of course.
The unevenness in mids (too much energy around 800Hz and to little around 1200Hz) should probably not be "corrected" by EQ/DRC but by correction room acoustics.

I've used dirac calibration, I've limited the range but don't remeber exactly, i'll give it a try to limit it even more just to 40-70hz. Maybe I should do the manual filter instead of the DIRAC? Can you recommend a guide how to set it on MiniDSP flex based on the REW measurments?

 

[Oboreal]

What kind of music do you listen to? If it’s not mostly electronic or hip hop, I’d start by testing a high-pass filter at 45 Hz. Keep in mind that the low E on an electric bass is 40 Hz... so there’s not necessarily a need to chase sub-bass at all costs in a room where it will never sound clean—and where it will also eat up headroom from the speakers.

Once that's done, what I would do is install a stretch ceiling and a stretch wall with 20 cm of rock wool behind my rear wall (30 if you can). If it absorbs too much in the highs, add some wooden slats in front. With that, you'll have addressed the vast majority of problems in the low mids—which is personally where I’d start. After that, if you want to go further, you can, but the cost might quickly outweigh the benefit, especially considering the room has dimensions that are far from ideal for a listening space.

"The idea of membrane or Helmholtz bass traps is obviously a good one too, but when there are major issues in a part of the spectrum that isn’t essential for most music, in my opinion it’s better to first make sure everything else is spot-on, and then decide whether it’s really worth trying to treat that zone—where it will be extremely difficult to get good results anyway.

 

[olieb]

Can you recommend a guide how to set it on MiniDSP flex based on the REW measurments?

I would keep it simple and use the procedure that MiniDSP describes.

REW AutoEQ step by step

miniDSP is a leading manufacturer of Digital Audio Signal Processors for the HomeTheater, Hifi, headphone and Automotive market. Join our large community of Audiophiles, Engineers and DIYers using our innovative products.

www.minidsp.com

You might start with a single measurement to see how it works, but it probably is better to do additional measurements left, right, front, back and average all these measurements (as Minidsp explains).

You can use the automatic matching "Match response to target" in "Filter Tasks" and correct the created filters manually after this
or
do the whole thing manually (Chapter 3 of the procedure) - this would be my choice. This way you have full control.

You can start with a single filter at 40Hz, then maybe two more for the bump 70-100Hz.

If that does not give an improvement then something is seriously off.
If it does, you can still try more sophisticated things if you want.
If you get fancy you might try those dashed spots (but in a second round

Dirac probably "corrected" the range where I put the orange arrows and that WILL change the sound and not necessarily for the better. (I would try placement or room treatment here, maybe it is from speaker directivity then it is as it is)

.

 

[boxerfan88]

A little late to the party. Here are my thoughts.

• deal with M1 using PEQ, knock it down about by about 10-15dB.
  • I doubt absorbers are gonna help in any meaningful way.
  • 40Hz you'll need ginormous bass traps
• the blue arrows show that area decaying quite fast then another ridge appears at the bottom
  • this usually indicates something is rattling/resonating in the room (ceiling panels, cupboard doors, etc...)
  • that 75Hz area is a little elevated, so PEQ is your friend, knock it back a little.
• that 650Hz wide-Q bump, is probably caused by front-wall
  • one can use a wide-Q PEQ to shape it

Hope it helps...

 

[raindance]

SBIR would be ideal but in my case 1/4 of 40hz is 2.15m, it would have to place speakers in the middle of the room But great to hear the traps helped, I was also advised by Gik to buy soffits but they are quite large so i'm more looking into Monster Bass Trap with range limiter to focus more on lows. And for 40hz the scopus T40 tuned traps, but 4 of them could be not sufficient. I have miniDSP flex digital version (i've put it between streamer and DAC), but I wasn't happy with results.

I was lucky enough to have a long room that allowed the SBIR idea.

 

[hedrick]

I got rid of a similar resonance by moving one speaker one foot away from the wall. I was shocked.

 

[Titurel]

All right, let us think a bit about why you have that massive peak at 40Hz. Your room dimensions are 4m x 4.5m. The first room mode always occurs where the dimension of the room equals a half wavelength - so in your case, 8m and 9m. This corresponds to 42.9Hz and 38.1Hz respectively. In other words, the reason why you have a massive peak at 40Hz is because you have two dimensions which are roughly equal.

You asked about that GIK Monster bass trap. According to the link, it is 7.3" thick (18.5cm). I said earlier that the thickness of the foam absorber will attenuate 1/8 of the wavelength, so in this case the wavelength is 148.3cm - so that absorber will work down to 232Hz. If you want it to absorb 40Hz, it needs to be 1m (3ft) thick. Not to mention, it is very likely that (like all foam absorbers) it will attenuate short wavelengths much more effectively than long wavelengths, so it will make your room very dead sounding. Now, it's your room and of course you can do whatever you want with it. I don't think it's a very good idea.

Re: your experience with MiniDSP. I have said this so many times that ASR people will think I sound like a broken record: the no. 1 cause of bad sounding DSP is user error. To get a good correction, you need proper measurements and you need to know what to correct and leave alone. This is not easy for an absolute beginner, which is why we are here to help. I am not surprised that most people try DSP, get bruised by the experience, and don't want to try it again. Of course bad DSP will kill the sound. It's not the MiniDSP that is to blame for bad sound, it is the arcane nature of DSP itself.

I don't know what source you are using - some of them are DSP capable. For e.g. if you are using a PC as your source, you could fix that problem for free using equipment that you already own and using free software tools.

I had (have) a similar problem for the same reason, length and width modes were within 4 or 5 Hz of each other . I got good service from DSP and better service from a combination of membrane traps (tuned near the offending modes) plus DSP. Membrane traps are not as tuning critical as Helmholz traps, and they are relatively compact, making effective placement fairly easy. Four relatively small traps calmed down the bass hump in my room to the point where EQ solutions were better and easier to achieve.

An observation about EQ; I spent a year performing in a stage show with significant sound reinforcement. Temperature and humidity drift over the course of the week (according to our "sound guy,") required regular tweaks to the EQ. From on stage, you really can tell the difference matinee audiences makes on the sound, as the patrons skew older and more heavily dressed. (This is also true in non-amplified venues.) Not only do white glove dampen the sound of applause, the fur coats soak up an awful lot of sound.

 

[dougi]

By the looks of the room photo, you are using a Lyngdorf Millennium amp? That has built in DSP by the looks of it https://lyngdorf.steinwaylyngdorf.c...07/Lyngdorf-Millenium-MK-IV-Owners-Manual.pdf, if you can get the software and RS232 port/emulation on your PC working. Or track down a matching DPA-1 and use it's RoomPerfect EQ!

 

[Peterinvan]

Attempting to fix bass problems with room treatment is not easy because of the size of the absorbers required.

Small Room Subwoofer...

In my small den, I have 10 DIY acoustic panels made with Rockwool. My recliner chair is just inside a seven foot triangle, and the bass was just too boomy.

I got rid of the bass problem by removing my SVS1000 sub entirely.

The KEF LS50 Wireless on filled and spiked stands work just fine on their own