Corner Bass Traps - Always beneficial?

[Bruce B]

---What is the percentage of movies recorded with real sound content in the 18 Hz area?

It's pretty small... though there are a few movies that have real explosions, thundersorms, volcanic activity and such.

 

[NorthStar]

---Yup, I had already a pretty good idea of that Bruce. :b

Methinks that a clean 25-50 Hz low bass sound in your room is sufficient for the vast majority of films' impact.

 

[microstrip]

Micro, would you mind posting the RT60 graph from these results as well?

Here it goes. BTW, can an helpful mind explain what is EDT in REW RT60 plots?

The figure captions are not preserved by REW capture function: dark green Topt, cyan EDT dark blue T20, magenta T30

 

[microstrip]

It looks like a room.

Seriously, the decay times aren't terrible except for the lowest two modes. Fortunately, most music has little if any energy at 35 Hz, and probably nothing at 18 Hz. If this room also serves as a theater, the extra oomph at 18 Hz could be seen as a plus!

--Ethan

The room sounds very good - very good dept, excellent imaging and good tone, except that bass has little impact - it sounds somewhat soft. I have tried several speakers and positions and all share this problem. It is why I was considering a low bass trap or resonator. Should I focus in the 35 Hz and forget the 19 Hz?

 

[NorthStar]

---Hey that's a pretty cool graph that last one just above.

* Looks like an ocean's reflection.

 

[Rutgar]

Micro, Can you tell me how you managed to set the time scale to display within a 1 second scale?

Edit: Okay, I believe I figured it out. However, mine is not as 'pretty' as yours. I almost want to hang yours on the wall!

 

[Bruce B]

Micro, Can you tell me how you managed to set the time scale to display within a 1 second scale?

Edit: Okay, I believe I figured it out. However, mine is not as 'pretty' as yours. I almost want to hang yours on the wall!

View attachment 4651

Is this the graph where the ideal is for everything to be within 0.2 - 0.4?

 

[Rutgar]

Is this the graph where the ideal is for everything to be within 0.2 - 0.4?

If I'm understanding your question Bruce, no. This is the actual RT60 of my room. Which I think shows is a little over damped (sounds that way too). I believe some of this is due to all of the Echo Busters and RPG Bad Panels I have straddling corners, in an effort to have some semblance of bass trapping.

 

[amirm]

Micro, Can you tell me how you managed to set the time scale to display within a 1 second scale?

Edit: Okay, I believe I figured it out. However, mine is not as 'pretty' as yours. I almost want to hang yours on the wall!

View attachment 4651

For small room acoustics (i.e. all the rooms we are talking about) there is no reason to search for any specific shape or flat response. The technical reason is that a "Reverberant field" will not be generated in such small space. In English , we are looking for a gross measure of "late reflections" in the room. Late reflections determine how live or dead a room is. A room that is too dead is uncomfortable for humans to live in as we are used to reflections existing in everyday life so removal of them to high degree feels unnatural and uncomfortable. Too live of a room interferes with clarify of music and dialog. To use RT60 to measure the extent of late reflections, you want to look at just one frequency: 500 Hz. In that region as Bruce mentioned, you want to strive for a number between 0.3 and 0.5. Values below 0.3 indicate the room is becoming too dead, above 0.5 means too live. The former means you have too much general absorption and the latter, too little.

Late reflections are impacted by anything in the room from acoustic products to furnishings. If you have a well furnished room, you may not need any absorption to decay them as this step by step measurement from Dr. Toole shows:

Note how the survey of 600 homes by J Bradley showed that normal living rooms fortuitously fall in the target range. We may not know about audio but we do know what sounds good . In a dedicated listening space that is devoid of furnishings otherwise, then addition acoustic products would be necessary to get the reverberation times low.

BTW for this purpose you want to just look at one graph. I usually pick T30. Answering the earlier question of what these graphs are, REW uses regression analysis to estimate RT60. RT60 as you may know is the time it takes for the reverberations to decay 60 db. If you click on Impulse Response (IR) likely you will find that the decay is not that long. So REW uses shorter term decays to estimate "what may have been" if the graph kept going lower and lower (again, using regression analysis). The selection of the graph segment as the predictor of the rest is sort of encoded in the name of the graph. "EDT" uses the values between 0 and -10 db, T20 -5 and -25 db (i.e. 20 db differential) and 30 -5 and -35 db (i.e. 30 db differential). You will probably notice that the T20 and T30 are similar since the shape of the graph is probably established by the first decay chunk so no reason to use both. EDT is supposed to be the early decay time. I don't know of a use for that in small rooms and it is not as good as a predictor for the shape of the graph. So no need to check that.

So to summarize: look at T30 @500 Hz and keep the value between 0.3 and .5. Do not worry about small changes there or worrying about all the values being the same. This is a gross measure and not a accurate target to hit at due to limitations of small room acoustics mentioned earlier.

Assuming EDT is the lowest graph in light blue and hence can be ignored, then your room is already in target area. I would not be adding more absorptive material as that would start to push you toward the dead region. If you need more to say, absorb bass frequencies, I would look to remove some others.

 

[Rutgar]

Thanks for the in depth explanation Amir. So to simplify, If I look at the T30 line (purple), at 500hz, the desirable range is between 300 and 400 mSec (which in this case, it is)?

The EDT (Teal) appears to be most erratic.

 

[microstrip]

For small room acoustics (i.e. all the rooms we are talking about) there is no reason to search for any specific shape or flat response. The technical reason is that a "Reverberant field" will not be generated in such small space. In English , we are looking for a gross measure of "late reflections" in the room. Late reflections determine how live or dead a room is. A room that is too dead is uncomfortable for humans to live in as we are used to reflections existing in everyday life so removal of them to high degree feels unnatural and uncomfortable. Too live of a room interferes with clarify of music and dialog. To use RT60 to measure the extent of late reflections, you want to look at just one frequency: 500 Hz. In that region as Bruce mentioned, you want to strive for a number between 0.3 and 0.5. Values below 0.3 indicate the room is becoming too dead, above 0.5 means too live. The former means you have too much general absorption and the latter, too little.

Late reflections are impacted by anything in the room from acoustic products to furnishings. If you have a well furnished room, you may not need any absorption to decay them as this step by step measurement from Dr. Toole shows:
(...)
Note how the survey of 600 homes by J Bradley showed that normal living rooms fortuitously fall in the target range. We may not know about audio but we do know what sounds good . In a dedicated listening space that is devoid of furnishings otherwise, then addition acoustic products would be necessary to get the reverberation times low.

BTW for this purpose you want to just look at one graph. I usually pick T30. Answering the earlier question of what these graphs are, REW uses regression analysis to estimate RT60. RT60 as you may know is the time it takes for the reverberations to decay 60 db. If you click on Impulse Response (IR) likely you will find that the decay is not that long. So REW uses shorter term decays to estimate "what may have been" if the graph kept going lower and lower (again, using regression analysis). The selection of the graph segment as the predictor of the rest is sort of encoded in the name of the graph. "EDT" uses the values between 0 and -10 db, T20 -5 and -25 db (i.e. 20 db differential) and 30 -5 and -35 db (i.e. 30 db differential). You will probably notice that the T20 and T30 are similar since the shape of the graph is probably established by the first decay chunk so no reason to use both. EDT is supposed to be the early decay time. I don't know of a use for that in small rooms and it is not as good as a predictor for the shape of the graph. So no need to check that.

So to summarize: look at T30 @500 Hz and keep the value between 0.3 and .5. Do not worry about small changes there or worrying about all the values being the same. This is a gross measure and not a accurate target to hit at due to limitations of small room acoustics mentioned earlier.

Assuming EDT is the lowest graph in light blue and hence can be ignored, then your room is already in target area. I would not be adding more absorptive material as that would start to push you toward the dead region. If you need more to say, absorb bass frequencies, I would look to remove some others.

Amir,

Nice to see we share the same book! When reading it sometime ago I also stopped at this caption - the reference to masonry construction and no low frequency absorption was just my case, as my room has thick stone walls. What I am looking is for the proper "custom built low frequency absorbers". And yes, I hate seeing acoustic panels - all acoustic treatments are and must be hidden.

Thanks for explaining the different RT60 parameters - I had never seen EDT before and I am still looking for what it Topt - the preferred parameter of the REW creator.

 

[AudioExplorations]

In a room where there are only 1, 2, or 3 problematic room modes, what are the merits of treating those with tuned traps vs. using more broadband low frequency absorbers (such as super chunks at the corners)?

 

[amirm]

Thanks for explaining the different RT60 parameters - I had never seen EDT before and I am still looking for what it Topt - the preferred parameter of the REW creator.

Topt is an attempt to apply a better curve fit than the other measures by using them together as the decay slope may imply. For the purposes of of using this as a gross measure in our small rooms, that kind of specificity -- even if better -- is not really relevant. We are looking at pretty wide latitude here.

 

[amirm]

Thanks for the in depth explanation Amir. So to simplify, If I look at the T30 line (purple), at 500hz, the desirable range is between 300 and 400 mSec (which in this case, it is)?

That's right.

 

[Ethan Winer]

The room sounds very good - very good dept, excellent imaging and good tone, except that bass has little impact - it sounds somewhat soft.

If the bass is weak, then the more important problem is the two nulls between 100 and 200 Hz. What happens between 200 and 300 Hz? Nulls even as high as 300 Hz can affect fullness.

--Ethan

 

[Rutgar]

If the bass is weak, then the more important problem is the two nulls between 100 and 200 Hz. What happens between 200 and 300 Hz? Nulls even as high as 300 Hz can affect fullness.

--Ethan

I have pretty much the same issue as Micro. No real impact. And as the response of my room shows, I have some problems in the range you describe Ethan.

 

[NorthStar]

If the bass is weak, then the more important problem is the two nulls between 100 and 200 Hz. What happens between 200 and 300 Hz? Nulls even as high as 300 Hz can affect fullness.

--Ethan

-----Thanx Ethan, great to know! :b

 

[microstrip]

If the bass is weak, then the more important problem is the two nulls between 100 and 200 Hz. What happens between 200 and 300 Hz? Nulls even as high as 300 Hz can affect fullness.

--Ethan

Just to clarify - the bass is not weak, it lacks attack and defintion. Exaggerating a bit I would say as in an old fashioned bass reflex speaker - more quantity than quality. IMHO, fullness does not seem to be the key issue. I know some of you hate this wording , but bass is slow, not fast enough.

Also the nulls between 100 and 300 Hz depend on small changes in position of the microphone - averaging 10 readings in a 2 feet square around the seating place makes them disappear. But the peaks at around 19 and 35 Hz are stable.

 

[NorthStar]

---Micro, then perhaps you need better loudspeakers, with better controlled bass. :b

...And some fine-tuning with a top-notch digital Room EQ. ...Like one included with a quality sub, or two. ...Or from a separate box (TacT, Trinnov, Audyssey, JBL Synthesis, Real EQ, ...).

 

[Ethan Winer]

I know some of you hate this wording

Since you showed graphs, all is forgiven. It may well be that your problem is the peak at 35 Hz, though that seems kinda low to affect much in the grand scheme of things. But you know that my stock answer is always to add more bass traps.

--Ethan