Room Treatment Advice

[sound_cycle]

Hi,

So this big a$$ 4K sheet of glass now sits between my speakers (Leeco SX3 55").

To get that in place had to change my arrangement / placement. I did not want the telly on the wall, so had to get suitable furniture. While at it decided to raise the speaker height to get it perfectly ear level (about 1.5 cm up)

All of which meant finding the time and energy to measure my room again etc.

Decided to respostion speakers. So it was back to the audiobeat method. I erased my earlier marks after taking pics. Ended up very close to where I was earlier ( within about ./- 5mm).

Sampling for Dirac was very quick this time. Measure distances with string to get the mic perfectly in the sweet spot. The chair arrangement in DLCT. Once the main measurement (the central one is done) the sequence for the remaining 8 is not important (I used to follow the diagram earlier). All just as well because of fireworks going off randomly and neighbours progeny rolling gas cylinders and screeching their lungs out. On previous attempts I had found the defaults suggested by Dirac to be better than my target curves, this time I had found some interesting info a short while back in https://www.minidsp.com/support/community-powered-tutorials/263-custom-dirac-live-target-curves and Possible to rolloff treble with Dirac?.

So set up bravely with my custom target curves.

So that's done. I don't have to throw the telly out

Sounds quite a bit nicer than it was (I will know if i really like after listening to it for a week or so OR the son makes a face). Voices are nicer, airier. And a bit bassier. (and the impulse looks nicer than in http://www.hifivision.com/av-enhancers-room-acoustics/53080-room-treatment-advice-8.html#post664777 (TBPH I would like to know how to interpret it correctly)

ciao
gr

 

[jls001]

GR, need to pick your brain on how to correctly use REW with an UMIK1 mic. So far all my attempts at measurements have produced funny looking room response graphs. I also tried DRC Designer and used the default corrections generated after measuring in DRC Designer. I don't like any of the corrections when plugged into fb2k. None of them sound right as all of them seem to take away too much amplitude. Plus the bass boom is unbearable. These are with stock flat curve response.

Later I tailored the response curve for about -10 to -6 dB attenuation between 40-120 Hz and things sounded better, though not acceptable. Extending the attenuation all the way to 300 Hz improved things further, but results were still not acceptable. At normal listening levels (or a bit higher) my bass is controlled sans the corrections.

I don't know what I'm doing wrong:sad:

Default corrections in a friend's room sounds very nice, BTW.

PS: I didn't try Dirac.

PPS: your impulse response looks surreally sharp. A real life response ought to have some negative overshoot and exponential decay, no?

P3S: I use a diffuser on the front wall behind the audio rack, and one old pillow each at first reflection points.

 

[sound_cycle]

Joshua . Sent pm

 

[sound_cycle]

Spent a week trying to rack my brains, read up on the 'net, watch videos etc.

I can't say I figured out much but hopefully I get sound that is decent.

I have only figured out very basic stuff with REW leaving the heavy lifting to Dirac. Nothing is fixed or finalized. WIP, so suggestions welcome

My goal was to improve imaging and SQ in my room. The strategy was to identify and stomp out (absorb) early reflections. I plan getting chunky bass traps in the top four corners but that is in material procured and nothing else done stage. Having got to use Dirac to squash the bass issues

unable to load image use the links to view please

1) The room and the rig
2) The Before Waterfall
3) The impulse after I got a reflection free zone setup as best as I could manage (Rockwool + some Basotec because I am lazy to cut small panels). There is some stuff that is sticking its head above the -20dB at ~ 10 ms, that is a door which I cannot do anything about just now
4) The post treatment waterfall plot
5) Dirac measurement. Dirac added a short delay and a small gain to get over that asymmetry
6) Dirac Impulse plot
7) As it looks

I've got foam diffusers in the front to cover the TV, I did not succeed in remote desktopping with them in front and left that for another day.

I think that I have a reasonably imaging and SQ.

@drkrack might be coming over sometime this weekend, will be interested in what he thinks. The last time he came over it was screwdriver-in- ear bad

ciao
gr

 

[elangoas]

unable to load image use the links to view please

Please use external image hosting sites like imgur (if you wish to) and share the BBC code here.. So the graphs are visible on your post..

There seems to be a huge null between 80 - 100Hz (per waterfall).. Hope you have tried re positioning speakers for any improvement in response..

There also seems to be huge peaks below 300Hz.. At 50, 100Hz and a few more.. So i guess, you would need to build some thick traps.. Where is your listening position in the room w.r.t room dimensions?..

Having got to use Dirac to squash the bass issues

By how much is the Dirac able to reduce the peaks?..

 

[sound_cycle]

There seems to be a huge null between 80 - 100Hz (per waterfall).. Hope you have tried re positioning speakers for any improvement in response..

There also seems to be huge peaks below 300Hz.. At 50, 100Hz and a few more.. So i guess, you would need to build some thick traps.. Where is your listening position in the room w.r.t room dimensions?..

I can't do much about that null. It is very clear in the Dirac measurement. Bass is not my issue at all.

Moving either the speakers or the seat sadly will do nothing.

I have verified this with the RTA feature in REW (it's really cool that one, I don't have to rely on the son anymore ) I got the generator to get me pink noise in the 40 to 400 Hz range and tried variations of position - of speaker and/or LP. Even the extreme of walking around with the mic a bit. The issue is cancellation from the front wall. The bass energy gets bounced by the wall behind the speaker back into the room, and cancels it.

http://www.customaudiodesigns.co.uk/soffit-mounting-control-room-monitors.htm explains it

The frequency response dip caused by a reflection.
Generally, in most rooms, free standing speakers are surrounded by walls/ceiling/floor that will generate reflections. These surfaces act as acoustical mirrors to the speaker's radiation, enhancing or cancelling the direct sound, depending on the phase difference between the reflection and the direct sound where you are sitting.

Generally, at lower frequencies, the reflected sound from the wall behind the speakers interferes with the direct sound. This is often in phase with the direct sound and causes deep peaks and troughs in the frequency response. These effects, depending on the relative amplitude of the direct vs reflected sound, can be between 6dB-20dB!!

The frequency of the first dip can be calculated by measuring the distance of the speaker to the wall. Divide 340 (the speed of sound in metres) by four times the speaker distance and you will have the dip frequency. Obviously further cancellations will occur at increasing harmonic frequencies.

340 / 1.8 x 4 = 47.2 and what we are seeing is its harmonic at ~ 94 Hz.

A padded cell is not possible so I'll live with it. I don't see how to avoid it.

I did not use REW after Dirac was in use, so I am not able to graph it for you.

i guess, you would need to build some thick traps.. Where is your listening position in the room w.r.t room dimensions?..

TBD : as I wrote above I plan getting chunky bass traps in the top four corners but that is in material procured and nothing else done stage.

As for my room dimensions, 24 x 10.5 x 10 feet

I've used the cardas dipole formula for getting the speaker into place

Results
The following distances should be used for speaker setup:
Dipole speaker (center front) to a side wall: 35.88 " Dipole speaker (center front) to rear wall: 74.16"

I have my listening position .38 of room length from the back wall ~ 9'

Here's a graph that shows how it is with absorption on the wall behind the speakers https://photos.app.goo.gl/6TQekAGEjfaiAiyM9

But that cannot be done, there are three doors behind the left speakers. FML C'est la vie game over

ciao
gr

 

[elangoas]

I can't do much about that null. It is very clear in the Dirac measurement. Bass is not my issue at all.

I see some 15 dB peaks, below 100Hz.. But if Dirac can manage that, then fine..

Moving either the speakers or the seat sadly will do nothing.

I have verified this with the RTA feature in REW (it's really cool that one, I don't have to rely on the son anymore ) I got the generator to get me pink noise in the 40 to 400 Hz range and tried variations of position - of speaker and/or LP. Even the extreme of walking around with the mic a bit. The issue is cancellation from the front wall. The bass energy gets bounced by the wall behind the speaker back into the room, and cancels it.

http://www.customaudiodesigns.co.uk/soffit-mounting-control-room-monitors.htm explains it

Ok.. Got it..

340 / 1.8 x 4 = 47.2 and what we are seeing is its harmonic at ~ 94 Hz.

As for my room dimensions, 24 x 10.5 x 10 feet

I used the room mode calculator with your room dimensions..

Length mode peaks at 24, 47, 71 & 94Hz in your room.. One the measurement graph (Dirac), i see 50Hz peak, 70Hz and few after 100hz..

Width mode peaks at 54, 108, 161, 215Hz..

Height mode peaks at 57, 113, 170, 226Hz..

But, as i mentioned earlier, if you are able to tame the peaks from Dirac measurements then fine..

A padded cell is not possible so I'll live with it. I don't see how to avoid it.

Understand that..

I did not use REW after Dirac was in use, so I am not able to graph it for you.

Not a problem.. Just checked if you had..

TBD : as I wrote above I plan getting chunky bass traps in the top four corners but that is in material procured and nothing else done stage.

Hope to see your bass traps build sooner..

As for my room dimensions, 24 x 10.5 x 10 feet

I've used the cardas dipole formula for getting the speaker into place

Results
The following distances should be used for speaker setup:
Dipole speaker (center front) to a side wall: 35.88 " Dipole speaker (center front) to rear wall: 74.16"

I have my listening position .38 of room length from the back wall ~ 9'

Here's a graph that shows how it is with absorption on the wall behind the speakers https://photos.app.goo.gl/6TQekAGEjfaiAiyM9

But that cannot be done, there are three doors behind the left speakers. FML C'est la vie game over

Is the speaker distance calculation, specific to your dipole speakers..

 

[sound_cycle]

But, as i mentioned earlier, if you are able to tame the peaks from Dirac measurements then fine..
<snip>
Is the speaker distance calculation, specific to your dipole speakers..

What are the alternatives you suggest ?

I have no possibility (no funds to change house , no know how or DIY ability to get tuned membrane traps and no space if I did get past that)

My major tools (cheap and easy) is absorption.

Consider now that 94 Hz problem. Let's say the speed of sound is 340 m/s. The wavelength is 3.6170212765957 m.

To absorb that I would need my trap to be a quarter wavelength .9 m. So let's say that I build a trap that is ~ 4 feet across and ceiling high. (It means blocking off all the inner rooms in my house by the way ) What about the room mode that caused it, 94 was the harmonic, that fundamental is twice as large.

There is as I see it no option to DRC. And therefore bass is not my issue. I will squash it. What I lose through attenuation I gain in tonality. I'm curious though why you state " But if Dirac can manage that, then fine" is there any reason to doubt DRC or a specific software ?

The cardas dipole formula for getting the speaker into place applies to dipoles ofc Look on the cardas site you will find other calculators and formulae for other situations and types of speakers.

ciao
gr

 

[elangoas]

What are the alternatives you suggest ?

I have no possibility (no funds to change house , no know how or DIY ability to get tuned membrane traps and no space if I did get past that)

My major tools (cheap and easy) is absorption.

Consider now that 94 Hz problem. Let's say the speed of sound is 340 m/s. The wavelength is 3.6170212765957 m.

To absorb that I would need my trap to be a quarter wavelength .9 m. So let's say that I build a trap that is ~ 4 feet across and ceiling high. (It means blocking off all the inner rooms in my house by the way )

Per the quarter wavelength calculation, the trap has to be 4ft thick.. Ofcourse 4ft traps in a living room isn't possible and wouldn't look good.. But a 2 ~ 3 ft thick trap might be able to help to an extent, somewhere where the frontwall meets the ceiling..(Entire width of the room)..

What about the room mode that caused it, 94 was the harmonic, that fundamental is twice as large.

The 1st order length mode is 24Hz.. Since the floorstanders isn't capable of producing, i guess it isn't showing up in response plot.. Where as the other consecutive modes show up..

There is as I see it no option to DRC. And therefore bass is not my issue. I will squash it.

Ok.. So you mean to say, since you are not able to tame the peaks, therefore bass is not a problem in your room?..

What I lose through attenuation I gain in tonality. I'm curious though why you state " But if Dirac can manage that, then fine" is there any reason to doubt DRC or a specific software ?

Nope.. Not doubting the DRC.. I have read that Audyssey has the capability of handling peaks of upto 20dB.. "In that relation, was referring if Dirac had the same capability, then fine"..

The cardas dipole formula for getting the speaker into place applies to dipoles ofc Look on the cardas site you will find other calculators and formulae for other situations and types of speakers.

I fairly recollect one the FM shared the cardas room placement long back when i had FS in my room (without sub).. I did place the FS accordingly in the room.. While i didn't know why i had to place it there in the room then, now i fairly believe that the positioning was for getting a smoother bass response in the room ( It was close to 1/4th and 3/4 room width placement- to counter width room modes)..

 

[sound_cycle]

Ok.. So you mean to say, since you are not able to tame the peaks, therefore bass is not a problem in your room?..

NO ! (in caps)I have no problems or issues with the bass in my room (no boominess or one note bass.) I am not attempting to use absorption for bass. I am doing that in DRC . I have BBA primarily for creating a reflection free zone and reducing reverberation/ ringing.. BTW adding random thickness of bass is not going to address the issue of my null. I am trying to figure out options.

I have planars. I measure them to ~ 32 hz. I don't therefore excite the lowest of the room modes.

ciao
gr

 

[Hari Iyer]

IME, if you require room treatment, then it should be more to do with the speakers and other gears than directly the room. If the room is good enough for normal family / friends conversations then it should be good for music too. Its difficult to tune the room by REW and even more difficult to interpret the graph / minimum phase / excess phase / group delay / decay / RT60 etc. Better results can still be obtained by trusting our ears and getting the basics correct.

 

[sound_cycle]

If the room is good enough for normal family / friends conversations then it should be good for music too. <snip>Better results can still be obtained by trusting our ears and getting the basics correct.

I think I have that. It is enjoyable to listen to music in, just concluded a listening sessions with a couple of FM. So I've gotten that bit right. I think that I should be able to improve on it, and those improvements are within the realm of realization with my current equipment.

ciao
gr

 

[elangoas]

NO ! (in caps)I have no problems or issues with the bass in my room (no boominess or one note bass.) I am not attempting to use absorption for bass. I am doing that in DRC .

Ok.. Now i understand.. After using Dirac, you wouldn't have issues with bass..

I have planars. I measure them to ~ 32 hz. I don't therefore excite the lowest of the room modes.

So, lucky..

IME, if you require room treatment, then it should be more to do with the speakers and other gears than directly the room.

Do you mean to say expensive speakers wouldn't have issues with bass in the room?.. What are the other gears you are referring to?..

Its difficult to tune the room by REW and even more difficult to interpret the graph / minimum phase / excess phase / group delay / decay / RT60 etc.

It takes some amount of reading, digest the terms and then approach solution..

I see, you emphasis on the word "phase" more often w.r.t room treatment..Though am naive to this and some terms, would appreciate some light on this topic..

Better results can still be obtained by trusting our ears and getting the basics correct.

What if the ears can't differentiate between 80Hz and a 40Hz?.. How would you know what the problem (bass) is in the room without measurement?..

 

[sound_cycle]

Length mode peaks at 24, 47, 71 & 94Hz in your room<snip>Width mode peaks at 54, 108, 161, 215Hz. Height mode peaks at 57, 113, 170, 226Hz

Possible solution ? http://teresaudio.com/haven/traps/traps.html; https://sonicscoop.com/2017/02/16/diy-acoustics-building-tube-traps/2/

another is here https://www.gearslutz.com/board/bas...locity-based-vs-pressure-based-absorbers.html

maybe you can have your cake and eat it. Broadband porous in Corners, floor to ceiling. Pressure in the middle of the walls tuned to that axial.

An interesting end goal to set would be the The BBC rule of thumb for an "acceptable" room to "No third octave decay time should differ from it's neighbours by more than 10% over a spectrum of probably 100-4K" before DRC ?

ciao
gr


.

 

[elangoas]

An interesting end goal to set would be the The BBC rule of thumb for an "acceptable" room to "No third octave decay time should differ from it's neighbours by more than 10% over a spectrum of probably 100-4K" before DRC ?

I recollect the same quote, which you had mentioned on the other thread, and i was hoping you would fairly aim for this in your room..

Possible solution ?

Most of these treatments in the above links shared (or) in general, have no specifics to the density of the material used.. They just say if it is fiberglass / rockwool and the DIY method to mount them..

From fair amount of reading, i think the lowest density of material (Air flows more in to the panel), maximum thickness that you can afford to do in your room and more surface area should help reduce the peaks..

another is here

If you are tentative to do the traps in your room, how have you planned?..

 

[sud98]

Since its a living room it comes with its pros and cons. I have a dedicated room and has a different set of issues.

For starters, if you can move the existing traps to the point of first reflection rather than above the sofa, and see if there is a difference.

Also, is there a possibility of a rug in front of the speakers?

Lastly, how far back is the back wall from your listening position?

 

[sound_cycle]

I recollect the same quote, which you had mentioned on the other thread, and i was hoping you would fairly aim for this in your room..

Most of these treatments in the above links shared (or) in general, have no specifics to the density of the material used.. They just say if it is fiberglass / rockwool and the DIY method to mount them..

From fair amount of reading, i think the lowest density of material (Air flows more in to the panel), maximum thickness that you can afford to do in your room and more surface area should help reduce the peaks..

If you are tentative to do the traps in your room, how have you planned?..

I am in no hurry. Would like to figure out and cut/ make holes in walls once.

(I have just spent on undoing my first approach - done for a different set of speakers/ configuration/ arrangement - so any putty painting or reconfiguring treatment, or in other words the period of regret is six years)

I already have 8 assembled ready to fix 4 x 2 panels. I have 4 4 sqft ones as well. I already have them placed (temporary nothing is bolted or stuck on yet) and can hear their influence (ofc with this approach I can only place on the floor/ furniture)

Provisionally I think that tube traps to cut the lowest of my length/ width modes, (probably placed in the corner behind either speaker, they are velocity absorbers after all ) is what think will be my most productive option. If I can cut this modal frequency I will be able to reduce the amount of attenuation that Dirac has to put in and get slightly higher listening levels (which I miss occasionally)

For starters, if you can move the existing traps to the point of first reflection rather than above the sofa, and see if there is a difference.

Also, is there a possibility of a rug in front of the speakers?

Lastly, how far back is the back wall from your listening position?

dimensions and positions upthread a few posts

Yes, as you suggest I have got the first and second reflection points on the sides and back* of the room. And I figured out positions using the mirror method but also with the ETC / REW method. So I have the positions figured out

Magnepans are supposed to have a figure of 8 dispersion pattern (when viewed from above). Also I cannot have a rug/ carpet in my room

My issue right now seems to be tacking the fundamental modes with trapping if that is possible

ciao
gr

* one side is a door to the master bedroom, so cannot have Rockwool there, instead I just leave the door open and hope for the best

 

[Hari Iyer]

Do you mean to say expensive speakers wouldn't have issues with bass in the room?.. What are the other gears you are referring to?..

I never said that, what i said was do not directly suspect the room and start treating the room. The problem could be elsewhere - viz.amplifier, IC, speaker cables and speakers. Once you have ruled out everything finally comes the room treatment.

If you are 80% satisfied with the setup, then only balance 20% can be achieved only by understanding your speaker and proper placement of the speakers in the room. Check if some other speakers sounds ok in the room.

It takes some amount of reading, digest the terms and then approach solution..

I think in addition to the reading, practical approach of measuring and listening and correlating the findings are more easy than just reading and digesting.

I see, you emphasis on the word "phase" more often w.r.t room treatment..Though am naive to this and some terms, would appreciate some light on this topic..

In an ideal speaker, the SPL phase will vary from say around 20Hz to 20KHz from +180 deg to -180 deg and 0 deg will be around 1KHz. The minimum phase of such a system will vary from +180deg and around 300Hz onward till around 15KHz stay at 0 deg without any phase shift inbetween. The gated response of such a system will show the excess phase at exactly 0 deg throughout.

If in the above speaker you happen to add a subwoofer to take care of freq below 80Hz, but this subwoofer is not a minimum phase design, then it will interfere with the low frequency of the full range driver and cancel out the bass respsonse. Hence its very important to know the minimum phase response of your front speaker to integrate it effectively with the subwoofer. If your front speakers are having a minimum phase of around 40 deg, then it will be wise to have your subwoofer phase control also set to 40 deg so that cancellation of common frequency is minimum at the sweet spot.

What if the ears can't differentiate between 80Hz and a 40Hz?.. How would you know what the problem (bass) is in the room without measurement?..

Ears can easily diffrentiate beween 80Hz and 40Hz. You can try playing a 40Hz and a 80Hz single sine wave and check out for yourself. When you check the FR curve you need to check at which frequency there is a room gain and correlate with the larger room dimension. Most standing wave frequency can be attributed to the larger room wavelength (atleast in my room). Checking spectogram can also be useful here to check the frequency where the room dictates the boom.

 

[Hari Iyer]

@sound_cycle
The impulse response and waterfall plot that you have displayed does not show proper scaling. You will need to check impulse at 1sec interval to make out the speaker response to an impulse.

Also the waterfall plots are to be adjusted for rise time and delay and should be in CSD mode for analsysis.

You have mentioned that bass in your speaker cancel out because of the rare wall. Can you let me know how far is your speaker from the rear wall and how far from the left and right wall are your OB speakers?

I can suggest you some placement which you can try after i hear from you.

Regards, Hari.

 

[sound_cycle]

@sound_cycle
The impulse response and waterfall plot that you have displayed does not show proper scaling. You will need to check impulse at 1sec interval to make out the speaker response to an impulse.

Also the waterfall plots are to be adjusted for rise time and delay and should be in CSD mode for analsysis.

You have mentioned that bass in your speaker cancel out because of the rare wall. Can you let me know how far is your speaker from the rear wall and how far from the left and right wall are your OB speakers?

I can suggest you some placement which you can try after i hear from you.

Room dimensions, 24 x 10.5 x 10 feet

calculated from dipole calculator
Dipole speaker (center front) to a side wall: 35.88 " actual is different (I have to re measure), I could not have them that far out from the side walls in my room, it is the max practicable at the moment
Dipole speaker (center front) to rear wall: 74.16"

I have my listening position ~.38 of room length from the back wall ~ 9'

I have cross checked the "eared" best position for bass with REW/ RTA (and also around the listening position). The null is there before correction. An exercise to be completed is to verify at the wall behind the speakers. At the moment I am convinced that it is not something that I can solve with positioning. But conviction does not mean a closed mind

will see if i can figure out and improve the graphs

ciao
gr