room treatment coupled with digital room correction
capoeira
Thread Starter
#1
29th August 2011
Old 29th August 2011
  #1
Gear nut
 
capoeira's Avatar
 

Thread Starter
room treatment coupled with digital room correction

I want to use digital room correction to use minimal physical acoustic treatment.
I guess I should start making acoustic treatment, but just to see wat is possible digitally I created a filter in REW-Software and ran it through a convolver.
These are automaticly created filters, I guess I can take care digitally of that 129Hz and 140Hz issues creating some filters manually.

I want to start with the first reflections from the walls, monitors are near the wall (tweeters are +- 40 and 45 cm from wall). Looking at the graphs, what kind of material should i use?

yellow is without filter, violett with filter:
Attached Thumbnails
room treatment coupled with digital room correction-unfiltered-48.jpg   room treatment coupled with digital room correction-filtered-48.jpg   room treatment coupled with digital room correction-waterfall-unfiltered.jpg   room treatment coupled with digital room correction-waterfall-filtered.jpg  

Last edited by capoeira; 29th August 2011 at 01:10 AM.. Reason: added waterfalls
#2
29th August 2011
Old 29th August 2011
  #2
Lives for gear
 
John White's Avatar
 

Stickies

Lots of great info in the stickies to get you started.

Also this may be relevant for you:
RealTraps - Audyssey Room EQ

A well written article about the use of EQ to control peaks and nulls.
Quote
1
#3
29th August 2011
Old 29th August 2011
  #3
Gear addict
 

You are clued into an interesting direction. Here is the KEY (and it's been known for 30+ years as fact and will be fact for evermore): you cannot EQ a reflection. Do this, go to the impulse response tab in REW and convert the Imp Resp to Minimum Phase. Now go to the Group Delay tab. For simplicity, you can only EQ where the curve is minimally deviated from the "0" line. If (just for example) you EQ say 100Hz where it may show a 75ms deviation, all you will do is distort the waveform in the process of creating a flatter pressure amplitude response...not good at all for mixing.

DRC is most effective in acoustically well controlled rooms.

Use the ETC tab to track down early reflections, add corner traps (and I get you said minimal treatments...this is minimal) and then go at the PEQ filters.

What you are "poking" at can help, you just need to be cautious.

best of luck!
#4
29th August 2011
Old 29th August 2011
  #4
Lives for gear
 
boggy's Avatar
 

I agree with Jeffrey.

Also, capoeira, did you notice that your waterfall graphs still display room resonances even after "compensation" is applied?

This means that your room will continue to "sing" at some frequencies even after musical note is finished... (39Hz, 42Hz, 60Hz, 81Hz...)

EDIT: Music isn't only a test tone without beginning and end. Resonances greatly distort room acoustic transient response...

Fairly flat frequency response isn't everything that we need to get a decent listening experience in room.

Only (serious) "physical acoustic treatment" can deal with room resonances. After you "calm down" this room resonances in your room, you can use DRC, if you like, for some "cosmetic" purposes, but not necessarily.

Also, some cancellation issues you can improve with careful listener and loudspeaker placement. (I think about dips and peaks not directly related to room resonances, but related to room boundaries and loudspeaker/microphone position.




#5
29th August 2011
Old 29th August 2011
  #5
Gear Guru
 
Glenn Kuras's Avatar
 

Quote:
This means that your room will continue to "sing" at some frequencies even after musical note is finished... (39Hz, 42Hz, 60Hz, 81Hz...)

EDIT: Music isn't only a test tone without beginning and end. Resonances greatly distort room acoustic transient response...
Just to give this a bit of layman's terms it is called "One note bass". I am sure you have been in a room that when the bass player is playing it all sounds like one note (just a muddy mess). That is where the "sing" that boggy is talking about. As the one frequency is still singing it covers the other (basic theory). If the room is ringing (or singing) that bad you will never be able to judge the low end. Watch the following video (I recommend headphones) and note how clear the low end becomes.

GIK Acoustics. Treated and untreated listening room video.

Now with all of that said what you need is bass trapping and the other acoustics to control things like early reflections and so on. If you get the room to certain point you could try the correction program, but I have found that most rooms don't even need it and or it just messes things up.
#6
29th August 2011
Old 29th August 2011
  #6
Lives for gear
 
boggy's Avatar
 

I agree, and most of "light" treated rooms has problems with "one bass note ringing".
I'll try to explain this:
Even if note is not exactly the same frequency of room mode, beginning of higher energy low end tone from speaker (Heaviside step function) induce transient process when room respond with its own "notes" whatever played from speaker (bass notes has usually higher energy in music, and they can "provoke" low end room modes much easily).

Quote:
Originally Posted by Glenn Kuras View Post
.......
............but I have found that most rooms don't even need it and or it just messes things up.
I found same for "automatic" room corrections (REW is some example), but....

Instead of this, at already digital crossover of our SM-3, in Pressed Lizard studio, I tried to do some cosmetic improvements of frequency response, with practical success (not only better measurement results).

I added filters manually, instead of automatic algorithms, with continuous measurements, listening tests and cooperation with mixing engineer that works in this studio.

I added filters in similar way like I equalize anechoic loudspeaker response (loudspeaker design), only difference that this measurements aren't anechoic and I have continuous mixing engineer's help.

All filters that don't give any listening improvement we throw away, even if there was measured situation that filter is obviously needed.

Not all measured anomalies has the same importance for human listener (not microphone) and that was possibly a biggest problem with DRC algorithms after impossibility to improve room resonances and non-minimal phase room behaviour.
capoeira
Thread Starter
#7
29th August 2011
Old 29th August 2011
  #7
Gear nut
 
capoeira's Avatar
 

Thread Starter
Quote:
Originally Posted by boggy View Post

Also, capoeira, did you notice that your waterfall graphs still display room resonances even after "compensation" is applied?
Yes, a noticed it. Actually ringing even got worse, since I have to add more gain now to have same loudness.

I was hoping to be able to bypass basstraps
see now that there is no way to do so.
problem is in the corners it would be very difficult to install them. But I found some DIY-traps I can install between walls and ceiling. That would be a solution.

I will make a 3d-model of my room today and post it here afterwards.

And I will studie more in the stickies.


Quote:
Originally Posted by Jeffrey Hedback View Post
Do this, go to the impulse response tab in REW and convert the Imp Resp to Minimum Phase. Now go to the Group Delay tab. For simplicity, you can only EQ where the curve is minimally deviated from the "0" line.
well, means I can do nothing look at the terrible graph:
(no need to respond....I know, basstraps)


Quote:
Originally Posted by Glenn Kuras View Post
Watch the following video (I recommend headphones) and note how clear the low end becomes.

GIK Acoustics. Treated and untreated listening room video.
yeap, thats quite impressive
Attached Thumbnails
room treatment coupled with digital room correction-right-excess-group-delay.jpg  
#8
29th August 2011
Old 29th August 2011
  #8
Lives for gear
 
boggy's Avatar
 

Quote:
Originally Posted by capoeira View Post
......Actually ringing even got worse, since I have to add more gain now to have same loudness.
Correct!
Quote:
Originally Posted by capoeira View Post
I was hoping to be able to bypass basstraps
Sorry, no other way.
Quote:
Originally Posted by capoeira View Post
see now that there is no way to do so.
problem is in the corners it would be very difficult to install them. But I found some DIY-traps I can install between walls and ceiling. That would be a solution.

I will make a 3d-model of my room today and post it here afterwards.
That will be nice.
Quote:
Originally Posted by capoeira View Post
...........
yeap, thats quite impressive
After any step in wideband conventional acoustic treatment, of your walls/ceiling, your excess group delay will be more and more smoother.
capoeira
Thread Starter
#9
1st September 2011
Old 1st September 2011
  #9
Gear nut
 
capoeira's Avatar
 

Thread Starter
so, this is my room in a picture (sorry for the bad cam)
its 4105x2502x2484(height):
Attached Thumbnails
room treatment coupled with digital room correction-20110819_2008.jpg  
capoeira
Thread Starter
#10
1st September 2011
Old 1st September 2011
  #10
Gear nut
 
capoeira's Avatar
 

Thread Starter
There is no way to use the corners between the walls so I am thinking of "trapping" the corners between wall and celing:
Attached Images
File Type: jpg studio oeste.jpg (42.1 KB, 115 views) File Type: jpg studio oeste traps.jpg (42.9 KB, 114 views) File Type: jpg studio leste.jpg (32.5 KB, 108 views) File Type: jpg studio leste traps.jpg (34.8 KB, 125 views)
capoeira
Thread Starter
#11
1st September 2011
Old 1st September 2011
  #11
Gear nut
 
capoeira's Avatar
 

Thread Starter
(why don't the grafics appear?)

There is one thing I'm in doubt of:
I can't "trap" directly to the window. there is a "thing" (lack of vocabulary) in front of it. Would a bass-trap make sense here, since it's not a "real" corner?
Attached Thumbnails
room treatment coupled with digital room correction-20110819_2010.jpg  
Attached Images
File Type: jpg studio detale.jpg (58.0 KB, 74 views) File Type: jpg studio detale traps.jpg (48.1 KB, 87 views)
capoeira
Thread Starter
#12
1st September 2011
Old 1st September 2011
  #12
Gear nut
 
capoeira's Avatar
 

Thread Starter
ignore the spaces between wall/ceiling and trap...it's because of limitations of the software i used. The spaces simulate the thickness.

My plan is to use rigid fibreglass without frames of 60cmx10cm.
The spaces behind the boards must be airtight don't they?
How can I seal them? It has to be a easy and nondestructive to be removed solution.

So what do you guys think?
capoeira
Thread Starter
#13
2nd September 2011
Old 2nd September 2011
  #13
Gear nut
 
capoeira's Avatar
 

Thread Starter
well, I see I have to think more practical. According to all the pictures I've seen, the space behind the traps don't need to be airtight. Since the fibreglass boards here generaly meassure 120cmx60cm I am thinking of making 7 traps and use them like this (still I have this doubt about the "false corner"):

With a thickness of 10cm this will treat the lows down to what frequency? Will a design like this be any sufficiant? I know, only testing can say, but from your experiences......

Thanks
Attached Thumbnails
room treatment coupled with digital room correction-studio-air.jpg   room treatment coupled with digital room correction-studio-inside.jpg  
#14
2nd September 2011
Old 2nd September 2011
  #14
Gear Guru
 
Glenn Kuras's Avatar
 

Quote:
Originally Posted by capoeira View Post
(why don't the grafics appear?)

There is one thing I'm in doubt of:
I can't "trap" directly to the window. there is a "thing" (lack of vocabulary) in front of it. Would a bass-trap make sense here, since it's not a "real" corner?
That is going to cause a problem. The only way to deal with it is fiberglass between the "thing" and the window. But I think you are going to have to skip that area for treatment.

Quote:
well, I see I have to think more practical. According to all the pictures I've seen, the space behind the traps don't need to be airtight. Since the fibreglass boards here generaly meassure 120cmx60cm I am thinking of making 7 traps and use them like this (still I have this doubt about the "false corner"):

With a thickness of 10cm this will treat the lows down to what frequency? Will a design like this be any sufficiant? I know, only testing can say, but from your experiences......
Right they do not need to be air tight.
How Bass Traps Work. Acoustic Panels and Bass Traps.

If you are going to go through all of that then maybe think about building a soffit design. Something more like this.
capoeira
Thread Starter
#15
2nd September 2011
Old 2nd September 2011
  #15
Gear nut
 
capoeira's Avatar
 

Thread Starter
Quote:
Originally Posted by Glenn Kuras View Post
That is going to cause a problem. The only way to deal with it is fiberglass between the "thing" and the window. But I think you are going to have to skip that area for treatment.
yea, thought of this....in theory closing the gap would be very nice but in practice it will be difficult to achieve, besides darkining the room a bit.
But what about the trap before the "thing"? does this make any sense?



Quote:
Originally Posted by Glenn Kuras View Post
maybe think about building a soffit design.
thinking practicly this would give me a square of 30cmx30cm. While this would result in more volume behind the trap (900cm ³ per m vs 779cm ³ per m) and the range of distance between wall and trap would be the same (0cm to 42,43cm) the triangle design would have balanced range of distances while the square design would have a high emphasis at 30cm distance wouldn't it? What does this mean in practice? Wouldn't the triangle design be more effective at the lower lows?
#16
2nd September 2011
Old 2nd September 2011
  #16
Gear Guru
 
Glenn Kuras's Avatar
 

Quote:
yea, thought of this....in theory closing the gap would be very nice but in practice it will be difficult to achieve, besides darkining the room a bit.
But what about the trap before the "thing"? does this make any sense?
Not really as that is not a true corner. Don't get me wrong it would not hurt but not like covering a real corner.

Quote:
thinking practicly this would give me a square of 30cmx30cm. While this would result in more volume behind the trap (900cm ³ per m vs 779cm ³ per m) and the range of distance between wall and trap would be the same (0cm to 42,43cm) the triangle design would have balanced range of distances while the square design would have a high emphasis at 30cm distance wouldn't it? What does this mean in practice? Wouldn't the triangle design be more effective at the lower lows?
No the soffit is like 2 triangles put together. So it will be much more effective.
capoeira
Thread Starter
#17
2nd September 2011
Old 2nd September 2011
  #17
Gear nut
 
capoeira's Avatar
 

Thread Starter
Quote:
Originally Posted by Glenn Kuras View Post
No the soffit is like 2 triangles put together. So it will be much more effective.
but in my practical case the two triangles together will have the same size as the the big triangle.

I know you are pro, but I don't understand the explanation.

As i would cut the boards in the middle of the longer side it would gave me this dimensions. See how the two triangles are exactly the same (2 triangles of 30x30x42,4):
Attached Thumbnails
room treatment coupled with digital room correction-same-triangles.jpg  
#18
2nd September 2011
Old 2nd September 2011
  #18
Gear Guru
 
Glenn Kuras's Avatar
 

Quote:
Originally Posted by capoeira View Post
but in my practical case the two triangles together will have the same size as the the big triangle.

I know you are pro, but I don't understand the explanation.

As i would cut the boards in the middle of the longer side it would gave me this dimensions. See how the two triangles are exactly the same (2 triangles of 30x30x42,4):
Sorry I am thinking of making triangles that are 17"x17"x24"(face) vs 17" square. If you make the triangle bigger then I guess that would be about the same.
capoeira
Thread Starter
#19
2nd September 2011
Old 2nd September 2011
  #19
Gear nut
 
capoeira's Avatar
 

Thread Starter
Quote:
Originally Posted by Glenn Kuras View Post
Sorry I am thinking of making triangles that are 17"x17"x24"(face) vs 17" square. If you make the triangle bigger then I guess that would be about the same.
ok, I guessed it

I have to go by the dimensions of the board.
But I am thinking, perhaps the square design would be more beautiful, at least in the tri corners where two of them will come together.
I will draw this, too

thanks
capoeira
Thread Starter
#20
3rd September 2011
Old 3rd September 2011
  #20
Gear nut
 
capoeira's Avatar
 

Thread Starter
looks more ugly in the model but I guess in the real world it would look "smaller" and therefore pertubate less. Good thing of this design is that I have more mass in the tri-corners don't I?
Attached Thumbnails
room treatment coupled with digital room correction-sala-studio-3-air.jpg   room treatment coupled with digital room correction-sala-studio-3.jpg  
#21
4th September 2011
Old 4th September 2011
  #21
Lives for gear
 
boggy's Avatar
 

Quote:
Originally Posted by capoeira View Post
looks more ugly in the model but I guess in the real world it would look "smaller" ......
It's not possible to build a decent acoustical treatment in an existing room which can be without volume and literally invisible. Even using 30-50% of room volume for acoustical treatment is ok. Especially in small rooms where we can't dampen lowest room modes absolutely even if we use all room volume for acoustical treatment.
Acoustics is art of making compromises...
capoeira
Thread Starter
#22
5th September 2011
Old 5th September 2011
  #22
Gear nut
 
capoeira's Avatar
 

Thread Starter
Quote:
Originally Posted by boggy View Post
It's not possible to build a decent acoustical treatment in an existing room which can be without volume and literally invisible. Even using 30-50% of room volume for acoustical treatment is ok. Especially in small rooms where we can't dampen lowest room modes absolutely even if we use all room volume for acoustical treatment.
Acoustics is art of making compromises...
yea, I know....but we should try to make the least ugly solution, don't we....e

What's your opinion, between my two options? Could one expect any performance diferences beween these two:

#23
5th September 2011
Old 5th September 2011
  #23
Lives for gear
 
boggy's Avatar
 

Quote:
Originally Posted by capoeira View Post
yea, I know....but we should try to make the least ugly solution, don't we....e
I understand.
Quote:
Originally Posted by capoeira View Post
What's your opinion, between my two options? Could one expect any performance diferences beween these two:
.....
Use one which looks better and/or will be easiest to you.
#24
6th September 2011
Old 6th September 2011
  #24
Gear Guru
 
Glenn Kuras's Avatar
 

Quote:
Originally Posted by capoeira View Post
ok, I guessed it

I have to go by the dimensions of the board.
But I am thinking, perhaps the square design would be more beautiful, at least in the tri corners where two of them will come together.
I will draw this, too

thanks
Yes it is going to look much more pro that way.
capoeira
Thread Starter
#25
15th September 2011
Old 15th September 2011
  #25
Gear nut
 
capoeira's Avatar
 

Thread Starter
what a pain in the arse to get the boards here.

probably I have to opt for the mineral-wool. So I need to use a frame, which will make the soffit-design complicated.
capoeira
Thread Starter
#26
21st September 2011
Old 21st September 2011
  #26
Gear nut
 
capoeira's Avatar
 

Thread Starter
very tired.

after 2 weeks of investigation I found out that my only option is rockwool-panels
Problem is: I found 2 types, both are 32kg/m³ only !!!!
Question: does it make sense to make traps with panels of 32kg/m³ or is it a waist of money? I have to say that those panels are expensive here

my options are:
Google Tradutor
Google Tradutor

interesting though is that the first option (32kg/m³ - composed of long insulating mineral fibers (rock), intertwined and bonded with special resins.) seams to have better low-freq absorbtion than the regular 64kg/m³ :
Google Tradutor


Do you guys think, the first option is an option for me?
#27
23rd September 2011
Old 23rd September 2011
  #27
Gear interested
 

Just curious, why is there a gap between the soffit at about half the length of your room? Are you just making more surface area for increased absorption?
capoeira
Thread Starter
#28
23rd September 2011
Old 23rd September 2011
  #28
Gear nut
 
capoeira's Avatar
 

Thread Starter
Quote:
Originally Posted by Oatbunches View Post
Just curious, why is there a gap between the soffit at about half the length of your room? Are you just making more surface area for increased absorption?
no, it's because of the length of the boards (1.20m) All those 6 traps are of that lenght I just put them like this to have them more distubiuted over the room.

but I won't be able to do the soffit-design anyways.
I will buy 32kg/m³ rockwool as it is my only option here. so, almost impossible to do a soffit I guess. I'm going to do 5 or 6 6"-traps with the 45°-design.

Today I'll make measurements in all 5 (4 ceiling/wall and 1 wall/wall) corners where I can place them and than make a new "plan"
capoeira
Thread Starter
#29
19th November 2011
Old 19th November 2011
  #29
Gear nut
 
capoeira's Avatar
 

Thread Starter
I cant believe it took me 11 weeks to make my first basic treatment.
And yes, I'm playing around again with DRC, even if most here say not to do. e.
This is my first try, with my now a bit treated room, a big open window at the front wall, which helps a lot with the modes, too. My sidepanels aren't in a defenit position yet, as I still don't understand that reflection at 1ms.
I would like you experts and semi-experts to have a look at the graphs and tell me why it's not recomended to use DRC.
I will only show the right speaker for now.
One thing I will prove now, is that DRC can have effect on ringin, yes sir. Not directly, but indirectly, and near the corner. As the LFs are mouch louder in this setup, DRC will attunate them, and at the same overall loudness ringing is less. I offseted the graphs to match at + or - 4KHz, so loudness should be comparable. Look:
Attached Thumbnails
room treatment coupled with digital room correction-right-water-30-200.jpg   room treatment coupled with digital room correction-right-drc-water-30-200.jpg   room treatment coupled with digital room correction-right-water-full.jpg   room treatment coupled with digital room correction-right-drc-water-full.jpg   room treatment coupled with digital room correction-right-spec.jpg  

room treatment coupled with digital room correction-right-drc-spec.jpg   room treatment coupled with digital room correction-right.jpg   room treatment coupled with digital room correction-right-drc.jpg   room treatment coupled with digital room correction-right-etc.jpg   room treatment coupled with digital room correction-right-drc-etc.jpg  

room treatment coupled with digital room correction-right-decay.jpg   room treatment coupled with digital room correction-right-drc-decay.jpg   room treatment coupled with digital room correction-right-t30.jpg   room treatment coupled with digital room correction-right-drc-t30.jpg  
#30
19th November 2011
Old 19th November 2011
  #30
Lives for gear
 
John White's Avatar
 

Ringing

WOW. Very flat- A great use of the system. I did notice the time graph is set up a bit differently on the waterfall with and w/o the DNR so it's hard for me to get a comparable reading. I think it was 600ms for w/o and 300ms with. Probably not a big deal, but good to set up equally for comparison sake.

Are the 2 significant notches related to SBIR?

Great thread. Thanks-
New Reply Submit Thread to Facebook Facebook  Submit Thread to Twitter Twitter  Submit Thread to LinkedIn LinkedIn  Submit Thread to Google+ Google+ 
 
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
 
Register FAQ Search Today's Posts Mark Forums Read

SEO by vBSEO ©2011, Crawlability, Inc.