How to get started with room treatments for the home

[radix]

@Sancus and @sigbergaudio thank you for the example spectrograms and RT60s. Having some examples like that I think has helped me understand how to read those charts. I'll take a few measurements today for my L, C, and R and see how it comes out.

 

[radix]

Reading the REW manual section on Energy-Time Curves, it describes a few techniques to help locate what wall locations contribute to side reflections, like the string technique.

 

[radix]

Here are the LCR measurements with my UMIK-2 in the main listening position. My Macbook connected to the Denon 6700 via HDMI 2. The midi was set to 8-channel, 16-bit, 48 kHz (this was a mistake, I meant to use 24-bit, but it was left this way when I was troubleshooting the connection). The UMIK also uses 48 kHz sample rate. REW was setup to use the Denon-AVR sound device and I verified the 8-channel mappings for LCR, LFE, SL, SR were correct.

The spectrograms seem to have decent control under 200 msec, except maybe in the 100-300 Hz range. But there is not much of a red band -- i don't know if that's good or bad (I suspect bad).

The RT60s show a couple spikes in the 100 - 300 hz range for the LR, and a big dip for the center around 180 Hz.

The SPL shows a pretty big drop for the center around 180 Hz, which seems to correspond to a "peak energy time" spike in the spectrogram and a big dip the in th RT60. The left has a big dip around 360 Hz.

I'm not sure why the LFE shows about +6 dB over what I would think of as "smooth" -- I'll need to review the Audyssey app PEQ settings.

I could not figure out how to get the RT60 chart to display the time to distance conversion on my mac.

These are the corresponding RT60 plots

And this is the combined (1/12 smoothed) SPL. The LFE is a Sigberg Inkognito 10 on Preset #1. It is against the front wall on floor stands. There is a second sub (an older klipsch) on the side wall, on the Denon sub 2 output, but it did not seem to be used in the LFE test.

 

[abdo123]

Here are the LCR measurements with my UMIK-2 in the main listening position. My Macbook connected to the Denon 6700 via HDMI 2. The midi was set to 8-channel, 16-bit, 48 kHz (this was a mistake, I meant to use 24-bit, but it was left this way when I was troubleshooting the connection). The UMIK also uses 48 kHz sample rate. REW was setup to use the Denon-AVR sound device and I verified the 8-channel mappings for LCR, LFE, SL, SR were correct.

The spectrograms seem to have decent control under 200 msec, except maybe in the 100-300 Hz range. But there is not much of a red band -- i don't know if that's good or bad (I suspect bad).

The RT60s show a couple spikes in the 100 - 300 hz range for the LR, and a big dip for the center around 180 Hz.

The SPL shows a pretty big drop for the center around 180 Hz, which seems to correspond to a "peak energy time" spike in the spectrogram and a big dip the in th RT60. The left has a big dip around 360 Hz.

I'm not sure why the LFE shows about +6 dB over what I would think of as "smooth" -- I'll need to review the Audyssey app PEQ settings.

I could not figure out how to get the RT60 chart to display the time to distance conversion on my mac.

View attachment 173465

View attachment 173466

View attachment 173467

These are the corresponding RT60 plots

View attachment 173468


View attachment 173469

View attachment 173470

And this is the combined (1/12 smoothed) SPL. The LFE is a Sigberg Inkognito 10 on Preset #1. It is against the front wall on floor stands. There is a second sub (an older klipsch) on the side wall, on the Denon sub 2 output, but it did not seem to be used in the LFE test.

View attachment 173471

Your most problematic areas are in the 100Hz to 300 Hz region. It looks like your room sticks to 0.2s decay time above that, which is where you want to be.

The foolproof method to remedy the situation is simply to cover up as much corners of the room with broadband bass traps.

GIK Acoustics Tri-Trap Corner Bass Trap – GIK Acoustics Europe

gikacoustics.eu

 

[sigbergaudio]

It would be interesting to see a combined measurement of subwoofer + left and right channel, because this is what you typically actually hear, and we won't know how the sub and speakers integrate until we see that.

I don't think this looks bad at all.

 

[Snarfie]

I have a room 3.4m wide, 5.4m long, and 2.5m tall with a 7.2 home theater setup. Is there a good post or link about how to get started with room treatments. I'm especially interested in if there's a process to determine what's the biggest ROI for treatments. For example, should I start out with corner bass traps or do wall reflections or rear speaker treatments or something else first?

I've seen a reference to, Speaker Boundary Interference Response (SBIR), but what is the best way to use that information? My speaker positions are pretty fixed based on where they can be mounted and where the TV is and such.

I'm also interested in what would be good brands of room treatments for the home that work well, look ok for a living space, but don't break the bank.

I do have a UMIK and can use REW. The AVR is a Denon 6700, and I can use the Audyssey app for PEQ.

Thanks,
Marc

I did a combination of room correction software, (mathaudio room eq) a few absorption panels an for the lateral (multipoint verticale) reflections a thick carpet (2x3 meters) worked in my situation outstanding.

 

[radix]

It would be interesting to see a combined measurement of subwoofer + left and right channel, because this is what you typically actually hear, and we won't know how the sub and speakers integrate until we see that.

I don't think this looks bad at all.

I was trying to figure out how to make L+sub work with the Denon 6700 over HDMI. When I switched the mac MIDI setup to 2-channel, it did not seem to play the sub. There must be some magic combination of MIDI settings plus the Denon "music" processing setting.

I think when I used Airplay and had the Denon set to "multi-channel stereo" for music, it used the subs and rears. I'll try that next for a combined reading.

 

[radix]

I did a combination of room correction software, (mathaudio room eq) a few absorption panels an for the lateral (multipoint verticale) reflections a thick carpet (2x3 meters) worked in my situation outstanding.

I was thinking of finding the valley (e.g. I have a 180 Hz dip in L), then setting the REW tone generator (e.g. for 180 Hz) and looking at the real-time response in REW while I walk around with a thick folded wool blanket. If I can find the spot/spots, then I can build/buy something. I kind of think it would be fun to build a 2" or 3" frame for rockwool or carpet cushion then decorate it with jute.

 

[sigbergaudio]

I was trying to figure out how to make L+sub work with the Denon 6700 over HDMI. When I switched the mac MIDI setup to 2-channel, it did not seem to play the sub. There must be some magic combination of MIDI settings plus the Denon "music" processing setting.

I think when I used Airplay and had the Denon set to "multi-channel stereo" for music, it used the subs and rears. I'll try that next for a combined reading.

It should probably play with subs if you choose "Stereo direct" or "2-channel stereo", and if it doesn't, it's likely a setting that needs to change. To begin with your main speakers need to be set to "small". You may also have a specific setting in there somewhere for "2 channel playback" that allows you to choose whether the sub should be active. I have a Marantz which is more or less the same thing but I don't think it looks exactly the same.

 

[Trdat]

I am comparing my measurements to the ones above and mine also look okay, technically its the only way I can learn to interpret measurements by seeing what consitutes as good or bad. But, my waterfall has some really long decay times, I think I have a ceiling SBIR issue and possibly front wall. Things do get better for bass when the measurement is taken for both left and right but some issues still persist.

Unfrotunately I couldn't get the spectrogram to look like the ones above.

Frequency response is 1/24 smoothing

And these are all my left speaker. A DSP triamped speaker.

 

[sigbergaudio]

@Trdat You're recording at a relatively high level too, so then it will take longer to hit the noise floor. Try to record at 75-80dB as opposed to 90. But yes, seems to be some room nodes in the bass.

Your response looks somewhat uneven. Try to set the y-axis on the frequency response (second to last graph) to 45-105dB (now you have -70 to 180, so anything will look flat).

 

[Trdat]

@Trdat You're recording at a relatively high level too, so then it will take longer to hit the noise floor. Try to record at 75-80dB as opposed to 90. But yes, seems to be some room nodes in the bass.

Your response looks somewhat uneven. Try to set the y-axis on the frequency response (second to last graph) to 45-105dB (now you have -70 to 180, so anything
will look flat).

Oh geez, now I see it. This is still with smoothing 1/24

I also added a left right waterfall. I am presuming, left right together works okay to understand about room modes(is it same as room nodes?) as all subs will be playing at same time.

Can you advise what frequency is problematic or what am I looking at? I think the long decay rates are the issue right? I can do some research and with a SBIR calulator and room mode calculator figure out where to put absorber and if i need specific targeted absorbtion.

To be honest, my room is heavily treated and have 4 subwoofers with two placed in the opposite polarity of a height mode and a lenght mode. So either the density of my rockwook is causing problems or it just to difficult to alleviate the long decay rates in a small room.

 

[sigbergaudio]

@Trdat There seem to be a number of issues here.

Now if you post another response curve of just 20-200hz it will be easier to look at just the bass part.
And possibly a full waterfall from 20hz-20khz

Do you have any DSP/EQ in the bass at all? If not it looks like you should seriously consider it.

@radix I'm not sure if you expected this thread to be just about your room, and all of this is very off-topic?

 

[Trdat]

@Trdat There seem to be a number of issues here.

Now if you post another response curve of just 20-200hz it will be easier to look at just the bass part.
Do you have any DSP/EQ in the bass at all? If not it looks like you should seriously consider it.

@radix I'm not sure if you expected this thread to be just about your room, and all of this is very off-topic?

I have DSP, the system is a Triamp with digital filters/DSP. The room is also very heavily treated with a large sofa and a 10cm thick ceiling absorber panel plus I have 4 subs with two on opposite polairites of the length and height axial modes(which I can't tell if they made any difference) so I was hoping that the issues were little. But it seems not...

I thought the RT60 looked okay.

The below are left and right 20-200 waterfall and frequency response no smoothing.

I also added the right side Filtered IR for good measure, no smoothing.

Any help would be apprecaited.

 

[sigbergaudio]

Well, things often look worse than they sound, so if you're happy with the sound you shouldn't go crazy about the graphs. But in general I would expect 20-100hz to look smoother with four subs and DSP. At what frequency are the subwoofers crossed over?

Your waterfall is a bit hard to determine since your volume is a bit high, and also the window is a bit short, so if you increase that to 500ms so we can see when it falls off below the floor, it would be easier to determine. It's extremely hard to fix anything below 100hz with room treatment, so you probably have to live with what you have. But it should be possible to get a smoother curve.

The subs also roll off very early in the bottom, is that intentional?

 

[Trdat]

Well, things often look worse than they sound, so if you're happy with the sound you shouldn't go crazy about the graphs. But in general I would expect 20-100hz to look smoother with four subs and DSP. At what frequency are the subwoofers crossed over?

Your waterfall is a bit hard to determine since your volume is a bit high, and also the window is a bit short, so if you increase that to 500ms so we can see when it falls off below the floor, it would be easier to determine. It's extremely hard to fix anything below 100hz with room treatment, so you probably have to live with what you have. But it should be possible to get a smoother curve.

The subs also roll off very early in the bottom, is that intentional?

Sound isn't bad, bass is very tight and love the dynamics of the horn. Always keen to learn how I can make things better. I understand below 100hz is difficult but I have some SBIR issue around 200hz I think that should be fixable.

I have always also expected smoother bass as well, but each time I attempt a new technique I don't get much difference.

This particular filter is crossed at 150hz, its for techno so I can't localize the subs but I have other filters crossed at 100hz.

I will measure again with 80db, but i measured loud as I listen loud but if it's messing up the measurement than I take your word for it.

No, I actually roll off slowly to help with group delay.

Hopefully the below graph helps, I wasn't sure if i was increasing the window ms or time range ms to 500 so I increased both.

 

[sigbergaudio]

Okay, yes seem like there's quite a bit of reverb in the low end.

 

[radix]

@radix I'm not sure if you expected this thread to be just about your room, and all of this is very off-topic?

Any exchange that helps understand how to measure and even out a room is good for me!

 

[radix]

Okay, yes seem like there's quite a bit of reverb in the low end.

How do you see this? Is it because the 20-200 waterfall is smushing up against the 265 msec limit? The RT60 EDT looked over 500 msec too.

Do you think that is reflections or poor timing of the 4 subs or is there not a likely cause?

Thanks,
Marc

 

[sigbergaudio]

How do you see this? Is it because the 20-200 waterfall is smushing up against the 265 msec limit? The RT60 EDT looked over 500 msec too.

Do you think that is reflections or poor timing of the 4 subs or is there not a likely cause?

Thanks,
Marc

It's smushing against the 489ms limit in the last waterfall he posted too. Difficult to speculate on cause in an unknown room with four subs.

The RT60 actually doesn't look much different from my own room down in the bass.

And he says it sounds great, so maybe there's no problem? It can be hard to identify a problem from measurements if there's no known problem to begin with. But measurements can often help find a solution to a known problem.

It's a bit like building speakers. We listen, then when something doesn't sound right, measurements help pinpoint it. But we can't just blindly measure our way to good speakers without listening.