Acoustic treatment my room

[funkytech]

Hi everyone,


I’m looking for some advice on how to improve the acoustic treatment in my small mixing/control room.


First, a quick disclaimer: this is a rented apartment. I may stay here for a few years, but I cannot do any permanent construction work. In particular, I cannot open or rebuild the drywall ceiling. The height to the drywall ceiling is 2.85 m, and there seems to be some kind of cavity above it. So I had to work within some practical limitations.


My goal is not to build a perfect professional studio, but to make the room reliable enough to finish and judge my mixes with reasonable confidence. I mostly work in a nearfield setup.


Room / setup


The room is irregularly shaped. The front wall is 2.90 m wide, the rear wall section is 3.00 m wide, and the total left-side length is about 5.25 m. Ceiling height to drywall is 2.85 m. There are also 12.5 cm structural columns protruding in some corners.


I use Focal Alpha 65 Evo monitors with front ports. After many placement tests, the current monitor position seems to be the best compromise in terms of frequency response. The speakers are placed on stands; the stand height is 76 cm from the floor. Listening distance is approximately nearfield, around 1.4–1.5 m.


Unfortunately, I do not yet have a proper measurement microphone. All current measurements were made with an Aston Spirit condenser microphone through my audio interface. I understand that this is not ideal, especially for absolute high-frequency accuracy, but I assume it should still show useful trends in the low and low-mid range. Please correct me if this assumption is wrong.

Current treatment


The room already has quite a lot of treatment, maybe even too much in some areas.

Current treatment includes:

• Full-height corner bass traps made from triangular mineral wool chunks.
• A large mobile absorber on the left side, built from four double-layer mineral wool panels.
• On the right side, at the first reflection area, there is currently only a single layer of mineral wool treatment.
• A large rear absorber panel behind the listening position, about 2.6 m wide.
• Ceiling cloud above the listening position.
• Additional broadband panels and curtains around the front/window area.
• One far corner is still untreated because of the room geometry and practical limitations.

I have attached a room plan with the current acoustic treatment layout.


Main problem


The biggest issues I see in the measurements are around 105 Hz and 165 Hz. These problems seem to affect the right monitor more than the left monitor, although both speakers have their own problems.

The 165 Hz issue also changes noticeably when I change the measurement height, which makes me suspect some floor/ceiling or vertical-mode interaction.


There is also a strong low-frequency issue around 36–40 Hz, with long decay. I suspect this may be related to a room mode and possibly the drywall ceiling/cavity behaving like a resonant surface. Since I cannot modify the ceiling structure, I am currently trying to control this with monitor EQ / correction rather than construction.

Questions


I would really appreciate advice on the following:

1. Based on this room shape and treatment layout, what would you try next to improve the 105 Hz and 165 Hz problems?
2. Could the asymmetry between the left and right side treatment explain why the right monitor behaves worse around these frequencies?
3. Is it worth adding more treatment to the untreated far corner, or would that probably not affect the 105/165 Hz issues much?
4. Since the 165 Hz dip changes with microphone height, should I focus more on speaker/listener height rather than adding more wall treatment?
5. Given that this is a rented room and I cannot modify the drywall ceiling, what would be the most practical way to deal with the 36–40 Hz decay?
6. How reliable are these measurements likely to be when made with an Aston Spirit instead of a calibrated measurement microphone?

Any advice would be appreciated. I understand the room is far from ideal, but I’m trying to get it to a practical, workable point rather than perfection.


Thanks!

 

[Cote Dazur]

Any advice would be appreciated

Welcome to ASR and congratulation on your well detailed post. Help is on its way.

 

[Keith_W]

Welcome to ASR.

Firstly, a bit of Googling tells me that your microphone can be switched between omni, cardioid, and figure of 8 modes. I hope that those measurements were taken in "omni" mode, with the mic pointing at the ceiling.

Lack of a proper microphone with cal file means that certain measurements are uninterpretable. Your mic looks like it has a huge capsule, which means that short wavelengths will suffer from diffraction. Also there is no cal file. What this means is that we ignore general trends in the frequency response, and especially anything above 12-15kHz. For example:

This upward tilting frequency response is very undesirable and it would sound pretty awful. You need to make sure that this is a measurement artefact and not real. For now we'll ignore it.

Since you are concerned about some spikes on your spectrogram, let's look at them.

Here we can see several spikes. 20Hz, 35Hz, 60Hz, 125Hz, and 165Hz. Now switch to the Waterfall view. Click on the 3D button, then rotate the waterfall until you can see the ridges clearly. You have to train your eyes a bit to recognize what is what on the waterfall.

If you see something like this, it is random noise. It is pulsing in SPL a little bit, maybe it's a lawnmower or something.

If you see this, it is a tone. The sound drops down, and then stays constant. We can see it a bit clearer in the RT60 Decay view. Choose "L+R No Panel", then click RT60 Decay, then click "Generate". Move your mouse to the same spot on the graph and you will see this:

So there is a 60Hz tone in your measurement. Maybe it's a ground loop. Or a computer fan.

Same thing at 125Hz. And since you are worried about that 165Hz spike, once you examine it, you will see it's the same.

You don't fix tones with room treatment. You fix it by finding the source of that tone and getting rid of it. The easiest way is to look around your room, look for sources of noise, and stick your ear next to it. Then you can decide if you want to do something about it. Maybe chuck it in a soundproof box. Maybe chuck it out the window. These things usually get masked by music anyway. Just because you see it on your mic does not mean you have to do something about it. Personally I would just ignore it.

Please follow the link in my signature, and it will take you to my Google Drive. There are 2 documents in there you should read: my REW eBook, and "Acoustical Measurement Standards for Listening Rooms" by Nyal Mellor. What you need to do is to check how closely your room approaches the acoustic standard. Since it's a studio, the standard is more stringent.

IMO your room already looks very close to the standard. Just needs a proper measurement with a proper mic to confirm, and that's it.

 

[janbth]

…
The biggest issues I see in the measurements are around 105 Hz and 165 Hz. These problems seem to affect the right monitor more than the left monitor, although both speakers have their own problems.
View attachment 553341

The 165 Hz issue also changes noticeably when I change the measurement height, which makes me suspect some floor/ceiling or vertical-mode interaction.


There is also a strong low-frequency issue around 36–40 Hz, with long decay. I suspect this may be related to a room mode and possibly the drywall ceiling/cavity behaving like a resonant surface. Since I cannot modify the ceiling structure, I am currently trying to control this with monitor EQ / correction rather than construction.

I think you suffer from room modes. The solution to that is not acoustic treatment, but moving speakers and/or listening position, getting a sub or two, and possibly applying EQ.

First I have some recommendation for posting plots:

• Use the camera icon in the upper left corner to produce the plot.
• X axis should show 20 - 20k Hz. OK in your plot above.
• Y axis should show a span of about 50 dB. Not OK in your plot, which is very compressed and shows negative values - very strange!
• Use Var smoothing which applies the right amount of smoothing across the frequency range.

The dips at 105 Hz and 165 Hz are probable due to room modes. Either the speakers or the listening position is in a null of the corresponding modes. Dips are not always audible. Are these dips audible? If so, you may need to move speakers or listening position, probably the latter.

The peak at 36 Hz is a room mode.

To find the room modes in your room, ask your favorite AI to compute them. A suggested prompt may be something like: «What are the lowest lying room modes in a room of dimensions W x D x H = 2.90 x 5.25 x 2.85 m?» The AI will apply the known formulas and compute the modes. Note that the computed frequencies are never completely exact, so you may need to apply some guessing to identify them from your real-world measurements.

Claude tells me that a room of your dimensions have modes at 32.7 Hz (0,1,0), 98 Hz (0,3,0), and 177.4 Hz (3,0,0). Quite close to your peak and two dips, I would say.

If what I say doesn’t make sense, ask your AI to draw some diagrams.

 

[funkytech]

Hello @Keith_W and thank you for your time!

Welcome to ASR.

Firstly, a bit of Googling tells me that your microphone can be switched between omni, cardioid, and figure of 8 modes. I hope that those measurements were taken in "omni" mode, with the mic pointing at the ceiling.

Unfortunately, no. I’ll give it a go, but it would be better to buy a measurement microphone after all.

Lack of a proper microphone with cal file means that certain measurements are uninterpretable. Your mic looks like it has a huge capsule, which means that short wavelengths will suffer from diffraction. Also there is no cal file. What this means is that we ignore general trends in the frequency response, and especially anything above 12-15kHz. For example:

View attachment 553384

This upward tilting frequency response is very undesirable and it would sound pretty awful. You need to make sure that this is a measurement artefact and not real. For now we'll ignore it.

Yes, I realise that, and besides, my microphone boosts the mid-range, so I tend to be a bit sceptical about the whole mid-frequency range as well.

Since you are concerned about some spikes on your spectrogram, let's look at them.

View attachment 553385

Here we can see several spikes. 20Hz, 35Hz, 60Hz, 125Hz, and 165Hz. Now switch to the Waterfall view. Click on the 3D button, then rotate the waterfall until you can see the ridges clearly. You have to train your eyes a bit to recognize what is what on the waterfall.

Here is I have some confuse about the measure, left and right channelas are different, but sum of the channels show me some nulls in 100hz and 165hz and I mostly think its mods of the room because I cannot find the reflection surface...

View attachment 553387

If you see something like this, it is random noise. It is pulsing in SPL a little bit, maybe it's a lawnmower or something.

View attachment 553388

If you see this, it is a tone. The sound drops down, and then stays constant. We can see it a bit clearer in the RT60 Decay view. Choose "L+R No Panel", then click RT60 Decay, then click "Generate". Move your mouse to the same spot on the graph and you will see this:

View attachment 553389

Oh thank you I did not know I can do this... I will check.

So there is a 60Hz tone in your measurement. Maybe it's a ground loop. Or a computer fan.

View attachment 553391

Yeah, I thought about it, but here is in Thailand we have 50hz in the electricity chain.

Same thing at 125Hz. And since you are worried about that 165Hz spike, once you examine it, you will see it's the same.

You don't fix tones with room treatment. You fix it by finding the source of that tone and getting rid of it. The easiest way is to look around your room, look for sources of noise, and stick your ear next to it. Then you can decide if you want to do something about it. Maybe chuck it in a soundproof box. Maybe chuck it out the window. These things usually get masked by music anyway. Just because you see it on your mic does not mean you have to do something about it. Personally I would just ignore it.

As I mentioned before I have a drywall cell...

Please follow the link in my signature, and it will take you to my Google Drive. There are 2 documents in there you should read: my REW eBook, and "Acoustical Measurement Standards for Listening Rooms" by Nyal Mellor. What you need to do is to check how closely your room approaches the acoustic standard. Since it's a studio, the standard is more stringent.

Yes, I found REW eBook, and fell asleep on page 45 last night ))

IMO your room already looks very close to the standard. Just needs a proper measurement with a proper mic to confirm, and that's it.

I was surprised to hear you say that. Thank you, I’ll do just that. It's my first control room. I just want to understand what I’m doing right now, to make sure it makes sense.

 

[funkytech]

I think you suffer from room modes. The solution to that is not acoustic treatment, but moving speakers and/or listening position, getting a sub or two, and possibly applying EQ.

First I have some recommendation for posting plots:

• Use the camera icon in the upper left corner to produce the plot.
• X axis should show 20 - 20k Hz. OK in your plot above.
• Y axis should show a span of about 50 dB. Not OK in your plot, which is very compressed and shows negative values - very strange!
• Use Var smoothing which applies the right amount of smoothing across the frequency range.

The dips at 105 Hz and 165 Hz are probable due to room modes. Either the speakers or the listening position is in a null of the corresponding modes. Dips are not always audible. Are these dips audible? If so, you may need to move speakers or listening position, probably the latter.

The peak at 36 Hz is a room mode.

To find the room modes in your room, ask your favorite AI to compute them. A suggested prompt may be something like: «What are the lowest lying room modes in a room of dimensions W x D x H = 2.90 x 5.25 x 2.85 m?» The AI will apply the known formulas and compute the modes. Note that the computed frequencies are never completely exact, so you may need to apply some guessing to identify them from your real-world measurements.

Claude tells me that a room of your dimensions have modes at 32.7 Hz (0,1,0), 98 Hz (0,3,0), and 177.4 Hz (3,0,0). Quite close to your peak and two dips, I would say.

If what I say doesn’t make sense, ask your AI to draw some diagrams.

Thank you for the clarification regarding the graphs. I’ll take a look and see what’s going on.


Regarding your assumption about the room modes, yes, I completely agree. I definitely took all of that into account, and they are most likely the cause. At least one of the measurements I provided, “NoPanels,” shows significantly deeper nulls because I had moved the left acoustic panel to a different position. Perhaps the panels simply aren’t thick enough to be effective at those frequencies.


Ultimately, though, these nulls are clearly audible when I play sine waves at the corresponding frequencies. Otherwise, I probably wouldn’t be so concerned about them. The current speaker placement also seems to be the best compromise I’ve found across the entire frequency spectrum.


That said, I’ll make the final decision once I get a proper measurement microphone and can take more reliable measurements.


The subwoofer solution is interesting. I haven’t really considered going in that direction yet, mainly because I’m a little concerned about my drywall ceiling. I’m worried that the additional low-frequency energy might make it resonate even more.

 

[SigbergAudio]

Here is left + right from the mdat file in proper format for those who don't want to take the time to download the mdat file.

And 20-200hz:

There's a pretty big peak at 35-50hz, the only reasonable way to fix this is with EQ.

I would second @Keith_W that it's a bit problematic that we don't know if this is accurate, so you should probably get a calibrated microphone.

So much elevation from 1khz and up is almost certainly not the case, as we know that the Focal Alpha 65 Evo does not measure like that, unless you have done something strange with EQ. We need to find out if this measurement is accurate or not. Is there any EQ applied at all at this moment?

Finally, I second @Keith_W again that from an acoustic treatment perspective, the decay already looks very good and relatively balanced. So with regards to your plan for acoustical treatment, I would throw it away, and spend a little bit of money on a calibrated microphone instead, and also apply EQ to the bottom end.

TL;DR: You don't need more treatment in this room, and the bass traps will also likely be a waste of money.

 

[Keith_W]

Yes, I found REW eBook, and fell asleep on page 45 last night ))

You did well. As you can see, I recently revised it. My own writing sent me to sleep earlier than that.

 

[Flaesh]

I agree with what was said above - treatment is sufficient, a measuring microphone is needed, something (EQ, if you look at things realistically) needs to be done about the 37 Hz peak. BTW calculated is 33 Hz for 5.25 m.
As for the sub, it depends. It can be applied both better and worse. It is unlikely (?) that there will be any problems with the ceiling.
My wooden floor structure resonates :

, you can look at your ceiling in the same way.

mixing

Are any of your mixes available anywhere?

 

[829]

TL;DR: You don't need more treatment in this room, and the bass traps will also likely be a waste of money.

The thing with bass trapping is, you cannot overdo it. When you add more bass traps @funkytech, you'll get what is shown in Figure 8 and modal ringing will significantly reduce. Of course, remaining peaks can and should be reduced through room-EQ.

But in your case it's better to use bass traps that really mainly work in the bass region, opposite to broadband absorbers. You probably wouldn't want to absorb more of the higher frequency spectrum through additional bass traps. And you have many suitable positions like behind the speakers and especially all (possible) 90° corners, also alongside the room left! I attached a performance graph of a suitable product (corner trap) profile. Of course effectiveness will increase and also reach down to even lower frequencies the bigger the area of bass absorption. One could also use bass traps which scatter higher frequencies where better suited.

Did you verify the early reflection points via help of a friend with mirror on the ceiling? The ceiling cloud looks placed behind your head position too much. If your work space looks as clean as on the room layout, you could also add some reasonably thick absorbent covers on the table to reduce these reflections.

[Is there a reason why you accept the rear wall reflections to return untreated (absorbed or scattered)? It seems you are still in the transition zone of roughly 3m distance, where comb filtering is likely to occur at the MLP from these reflections.]

Edit: Seems I read over "A large rear absorber panel behind the listening position, about 2.6 m wide." cause it's not on the room layout. Where exactly is that panel positioned?

 

[SigbergAudio]

The thing with bass trapping is, you cannot overdo it. When you add more bass traps @funkytech, you'll get what is shown in Figure 8 and modal ringing will significantly reduce. Of course, remaining peaks can and should be reduced through room-EQ.

My point wasn't really that the bass traps wouldn't work, more about how much effect they will give (not so much until your whole room is full of bass traps) compared to cost and incovenience. With a bit of added EQ, the room is already better than many professional studios.

 

[829]

My point wasn't really that the bass traps wouldn't work, more about how much effect they will give (not so much until your whole room is full of bass traps) compared to cost and incovenience. With a bit of added EQ, the room is already better than many professional studios.

The effect will be significant but yes, the better the bigger the area which is treated with bass traps properly. It doesn't need to be expensive if a DIY approach would be feasible for the OP. But also buying additional bass traps costs less than multiple subwoofers + Dirac ART licence, which works well but I'm sure cannot fully substitute bass trapping.

Room-EQ is quite ineffective at reducing modal ringing; and also boosting nulls too much will result in further problems. So the better approach would be to treat the room as much as personally possible and then use room-EQ wisely.

 

[funkytech]

Here is left + right from the mdat file in proper format for those who don't want to take the time to download the mdat file.

View attachment 553566

And 20-200hz:
View attachment 553567

There's a pretty big peak at 35-50hz, the only reasonable way to fix this is with EQ.

I would second @Keith_W that it's a bit problematic that we don't know if this is accurate, so you should probably get a calibrated microphone.

So much elevation from 1khz and up is almost certainly not the case, as we know that the Focal Alpha 65 Evo does not measure like that, unless you have done something strange with EQ. We need to find out if this measurement is accurate or not. Is there any EQ applied at all at this moment?

NO EQ. I guess maybe the aston spirit give some elevation in cardioid mode... but as we discuss here I need a right microphone to reduce measure artefacts.

Finally, I second @Keith_W again that from an acoustic treatment perspective, the decay already looks very good and relatively balanced. So with regards to your plan for acoustical treatment, I would throw it away, and spend a little bit of money on a calibrated microphone instead, and also apply EQ to the bottom end.

TL;DR: You don't need more treatment in this room, and the bass traps will also likely be a waste of money.

 

[funkytech]

I agree with what was said above - treatment is sufficient, a measuring microphone is needed, something (EQ, if you look at things realistically) needs to be done about the 37 Hz peak. BTW calculated is 33 Hz for 5.25 m.
As for the sub, it depends. It can be applied both better and worse. It is unlikely (?) that there will be any problems with the ceiling.
My wooden floor structure resonates :
View attachment 553600
, you can look at your ceiling in the same way.

Are any of your mixes available anywhere?

Yeah, you can find it by my nickname, but only 2026 release were make in this room and only the last one after the acoustic treatment I did.

 

[funkytech]

The thing with bass trapping is, you cannot overdo it. When you add more bass traps @funkytech, you'll get what is shown in Figure 8 and modal ringing will significantly reduce. Of course, remaining peaks can and should be reduced through room-EQ.

Thank you I will have a look.

But in your case it's better to use bass traps that really mainly work in the bass region, opposite to broadband absorbers. You probably wouldn't want to absorb more of the higher frequency spectrum through additional bass traps. And you have many suitable positions like behind the speakers and especially all (possible) 90° corners, also alongside the room left! I attached a performance graph of a suitable product (corner trap) profile. Of course effectiveness will increase and also reach down to even lower frequencies the bigger the area of bass absorption. One could also use bass traps which scatter higher frequencies where better suited.

View attachment 553606

I think maybe better to add a photo.

So I have some superchanks in the corners of the front wall. I made them out of triangular pieces of mineral wool, from floor to ceiling. The only thing was, I couldn't find anything denser than 40 kg/m³ at the local market.

Did you verify the early reflection points via help of a friend with mirror on the ceiling? The ceiling cloud looks placed behind your head position too much. If your work space looks as clean as on the room layout, you could also add some reasonably thick absorbent covers on the table to reduce these reflections.

I'm almost certain of the locations of the early reflections from the right and left walls, except for the density of the rock wool. But not about the ceiling cloud, and especcaily density of it, because I made it from the felt panels.

[Is there a reason why you accept the rear wall reflections to return untreated (absorbed or scattered)? It seems you are still in the transition zone of roughly 3m distance, where comb filtering is likely to occur at the MLP from these reflections.]

Edit: Seems I read over "A large rear absorber panel behind the listening position, about 2.6 m wide." cause it's not on the room layout. Where exactly is that panel positioned?

this is the back wall behind me:

 

[funkytech]

My point wasn't really that the bass traps wouldn't work, more about how much effect they will give (not so much until your whole room is full of bass traps) compared to cost and incovenience. With a bit of added EQ, the room is already better than many professional studios.

I'm just trying to figure out what's going on first, so I can decide what to do.

 

[funkytech]

The effect will be significant but yes, the better the bigger the area which is treated with bass traps properly. It doesn't need to be expensive if a DIY approach would be feasible for the OP. But also buying additional bass traps costs less than multiple subwoofers + Dirac ART licence, which works well but I'm sure cannot fully substitute bass trapping.

Room-EQ is quite ineffective at reducing modal ringing; and also boosting nulls too much will result in further problems. So the better approach would be to treat the room as much as personally possible and then use room-EQ wisely.

Yeah, this is the way I'm going for now. And thank you for one more article .

One of the next steps I'm considering right now is to build heavy stands with a sand to replace these rickety metal ones. I've definitely done everything I could with them to dampen their natural resonances, but still...

 

[funkytech]

Thank you all so much for your time and attention to my research. I do everything myself without any formal training behind me; I have to figure everything out from scratch. Thank God (sic!) I studied physics in school.

 

[829]

1. Could the asymmetry between the left and right side treatment explain why the right monitor behaves worse around these frequencies?

Hard to say with the information provided but it could be. *Just saw the photos you posted.

Since the distance from the L speaker to the L wall seems to be the same as R speaker to the R wall, and underlying wall material is the same(?), then certainly it is advisable to maintain symmetry also in room acoustic treatment.

So I have some superchanks in the corners of the front wall. I made them out of triangular pieces of mineral wool, from floor to ceiling. The only thing was, I couldn't find anything denser than 40 kg/m³ at the local market.

The key elememt for bass absorption is the right, rather low airflow resistance [Pa*s/m²] and of course >> thickness of chosen material. Unfortunately only in German, but maybe quite usable with automated translation, this article gives a good overview what to expect when building porous absorbers. Addition @funkytech: Another suitable build.

One of the next steps I'm considering right now is to build heavy stands with a sand to replace these rickety metal ones. I've definitely done everything I could with them to dampen their natural resonances, but still...

I think that can wait.

 

[patoulol]

welcome in ASR @funkytech

the fact that the 165 Hz dip shifts quite a lot with mic height is interesting, so I ll definitely check that properly once I have the calibrated mic


the 105 Hz behaviour is less clear to me, so I agree it’s better to wait for the measurements and check it properly in REW, including the modal simulator and measurements at different positions.


tor the LR asymmetry, I also suspected the very different amount of treatment on each side could be playing a role. I’ll probably wait before adding anything to the last corner and see what the measurements show first.


ror the 36 40 Hz region, I agree it makes more sense to look at sub positionning and EQ once I have proper measurements rather than adding more absorption blindly