REW measurements. Is it possible to fix 100-1000 Hz range with treatment?

[Tupisac]

I've attached the mdat file.

I have a desktop system in small room (roughly 3x3 m). Two JBL LSR 305 speakers and LSR 310S sub, fed from MOTU M2 interface via t.racks DSP Mini. Setup is asymmetric (right speaker 30 cm from a corner, left speaker in the middle of a wall). I have a GIK panel in the corner next to right speaker (90 x 60 x 10 classic bass trap).

Here is a draft:

Frequency response of full system after crossover and delay:

Individual responses (no xo):

MMM measurement:

Waterfall of full system with 80 Hz xo and delays:

I wonder what might be the cause of the mess between 100 and 1000 Hz.

Is it desk?

Is it ceiling/floor?

Is it back/front wall?

Or all of the above?

 

[Keith_W]

I wonder what might be the cause of the mess between 100 and 1000 Hz.

Small room --> high Schroder frequency --> high transition zone.

The Schroder frequency can be calculated with this formula: Fs = 2000 * sqrt (RT60/V).

From your REW measurements, your RT60 is about 0.2sec. And you did not supply your room height, I am assuming it is the standard 2.4m. So room volume is 21.6 m^3. Plugging that into the formula, the Schroder freq is 192Hz, meaning the transition zone ends at 770Hz or so.

If you look at my signature, you will see a link. Follow that link and you will find a document called "Acoustic Targets for Listening Rooms". Compare your measurements with those targets. If you need help interpreting those measurements, there is also a free eBook that I wrote that will guide you.

That's a beautiful illustration of your room, BTW. What software did you use?

 

[Tupisac]

That's a beautiful illustration of your room, BTW. What software did you use?

Thanks . I use old version of SketchUp (Pro 2022).

Height is 2,6 m. According to AMROC calculations Schroder frequency is 211,5 Hz, although it doesn't take slanted ceiling into account. More details here: https://www.audiosciencereview.com/...r-2-1-desk-setup-placement-suggestions.72538/

If you look at my signature, you will see a link. Follow that link and you will find a document called "Acoustic Targets for Listening Rooms".

Can't find it . Is it on the linked google drive?

 

[kyuu]

Hmm. The pdf titled "acoustic_measurement_standards" is a paper called "Acoustical Measurement Standards for Stereo Listening Rooms" by Nyal Mellor, which is the same fellow he mentions in note at the beginning of the REW eBook (but calls the paper something else). Either that's the one he meant, or it seems to be missing.

 

[Keith_W]

Hmm. The pdf titled "acoustic_measurement_standards" is a paper called "Acoustical Measurement Standards for Stereo Listening Rooms" by Nyal Mellor, which is the same fellow he mentions in note at the beginning of the REW eBook (but calls the paper something else). Either that's the one he meant, or it seems to be missing.

That's the one.

 

[Tupisac]

That's the one.

I'm slowly chewing through it. I've also glanced through other pdfs. Lots of great info to digest. Thank you

Meanwhile, would you mind looking at my mdat file? Any remarks will be highly appreciated. I've tried doing the measurements by the book, with loopback timing reference for each channel. Mic is an ECM8000 (uncalibrated).

I've also drafted a treatment plan and I wonder if it makes sense in my situation:

Green one is the 10 cm bass trap from GIK that I already have. (60x90 cm)

Yellow ones are planned purchases. So five additional 120x90x10 panels (two on the slanted ceiling), one cloud (120x90x20) and three bass traps (30x30x100).

Pink ones are optional small panels to fit around the door area. I wonder if they are even worth considering since they are fairly small (25 cm width, 10 cm deep). But it is technically a back wall and a point of first reflection.

 

[Keith_W]

Your RT60 is already very low, the last thing you need is more absorbers. You will turn your room into an unpleasant space with no ambience. What you need to do is focus on your ETC, you will see there are a few specular reflections. You need to identify them and fix them. You can calculate the time-lag of the reflections using the ETC which will give you the path-length difference, or the simpler way is to guess where the reflection is coming from and slap some acoustic foam on there and see if it goes away. Don't forget that you are likely to get quite a few reflections from your desk as well. If you need a clearer explanation on how the ETC works, take a look at "MAC Talk Treble" from Slide 40 onwards.

If your RT60 is already low, you could consider diffusers instead of absorbers.

 

[Redacted]

If your office/mancave looks anything like the renders in real life, my wife might give you a call to ask you to advise me!

 

[ZolaIII]

Flip the phase for sub and right speaker (180°) hit the room fundamental (first peak about 53 Hz by hand PEQ in REW so that Q in lines), use L speaker for that (apply it to all on MiniDSP) and mesure again. RT times and back to front simply can't be very good with such placement or in very small room.

 

[Tupisac]

If your office/mancave looks anything like the renders in real life, my wife might give you a call to ask you to advise me!

Well... It kinda does when it's freshly cleaned . Usually it contains piles of stuff on the couch, desk and sometimes even floor. The Kallax is also full of colorful random mess on the verge of spillage which lowers the WAF significantly.

Here is the photo of my desk from a year ago when I did my last cable management attempt:

And yes, speakers were upside down but currently they are back to normal orientation. The sub and drawers arrangement also changed and the port is firing in the correct direction.

 

[Tupisac]

What you need to do is focus on your ETC

Erm... I'm ashamed to admit it but yours look nothing like mine and I don't even know where to start... All my REW knowledge comes from random tutorials.

 

[Tupisac]

If your RT60 is already low, you could consider diffusers instead of absorbers.

What about those popular absorbers with diffusers on the front?

 

[Redacted]

Well... It kinda does when it's freshly cleaned . Usually it contains piles of stuff on the couch, desk and sometimes even floor. The Kallax is also full of colorful random mess on the verge of spillage which lowers the WAF significantly.

Here is the photo of my desk from a year ago when I did my last cable management attempt:

View attachment 549608

And yes, speakers were upside down but currently they are back to normal orientation. The sub and drawers arrangement also changed and the port is firing in the correct direction.

Complete rubbish - you need help from @Ropeburn to get some piles of random cables scattered about and I can show you how to use paint pots, white spirit bottles, telescopic ladders, random cordless power tools and more, artistically arranged on the floor for the full effect. The women, they don't like climbing over them, Mr Mainwaring, Sir.

 

[Tupisac]

paint pots, white spirit bottles,

Lol. Did I mention that my other hobby involves wood and traditional resins?

 

[Tupisac]

Flip the phase for sub and right speaker (180°) hit the room fundamental (first peak about 53 Hz by hand PEQ in REW so that Q in lines), use L speaker for that (apply it to all on MiniDSP) and mesure again. RT times and back to front simply can't be very good with such placement or in very small room.

Just to be sure - you want me to kill the 53 Hz peak on the left speaker and invert sub and right speaker, right?

But the speakers are high passed at 80 which lowers the peak significantly. Should I flatten what remains, like pictured below?

 

[Tupisac]

Flip the phase for sub and right speaker (180°) hit the room fundamental (first peak about 53 Hz by hand PEQ in REW so that Q in lines), use L speaker for that (apply it to all on MiniDSP) and mesure again. RT times and back to front simply can't be very good with such placement or in very small room.

Ok, interesting. So I did some calculations in REW. Chopped the peak from left speaker as pictured in previous post. Summed it with inverted right and then with inverted sub. Here is the predicted result (green) against the last measurement (orange). Crossover at 80 with best looking delay results set in alignment tool.

Will report back after next measuring session.

EDIT: So I've clicked few boxes in DSP. It sounds really weird. Like with some fake surround effect engaged. Definitely interesting but I'm not sure it will translate well in the long run. Might make the sound of my guitar impossible to replicate

 

[Tupisac]

Getting back into the acoustic treatment plan - here is my current reasoning in the form of a picture:

 

[ZolaIII]

Ok, interesting. So I did some calculations in REW. Chopped the peak from left speaker as pictured in previous post. Summed it with inverted right and then with inverted sub. Here is the predicted result (green) against the last measurement (orange). Crossover at 80 with best looking delay results set in alignment tool.

View attachment 549634
Will report back after next measuring session.

EDIT: So I've clicked few boxes in DSP. It sounds really weird. Like with some fake surround effect engaged. Definitely interesting but I'm not sure it will translate well in the long run. Might make the sound of my guitar impossible to replicate

It's room fundamental (the first peak) so applied it's correction as on left speaker to all MiniDSP outputs (put it to both input chenels banks side). It's not just first peak it's it and it's harmonics (following two peeks and downs). Phase shift whose obvious and with it it sums better especially in transition area. Anyway with both in plece (on Mini) when you doo measurement all left to do is iron out the peeks up to shoulder frequency (with PEQ's and per chenel now they sum fine). Do that and for curtuasy I will talk with you about room treatment all do I am against it until you get a longer one and with possibility of right placement. In such it's not very justifiable. We are in this too long not to be straight.

 

[Tupisac]

It's room fundamental (the first peak) so applied it's correction as on left speaker to all MiniDSP outputs (put it to both input chenels banks side).

I've got the t.racks DSP Mini. It can apply PEQ only on it's output channels which turned out is a whole another can of worms.

Do that and for curtuasy I will talk with you about room treatment all do I am against it until you get a longer one and with possibility of right placement. In such it's not very justifiable. We are in this too long not to be straight.

I really appreciate the honesty. Maybe it's time to get out of this rabbit hole...

 

[ZolaIII]

I've got the t.racks DSP Mini. It can apply PEQ only on it's output channels which turned out is a whole another can of worms.


I really appreciate the honesty. Maybe it's time to get out of this rabbit hole...

Come on now you are almost there and without dime spend. It doesn't matter put it to first output PEQ bank on each chenel you are using (L, R & Sub).