• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Room measurements with KEF R3

Marcin

Member
Joined
Jan 12, 2022
Messages
49
Likes
20
Location
Poland
Hi everyone,

This is my first attempt to room measurement. My goal is to acoustically optimize the room.

I have KEF R3 loudspeakers set up in my living room, standing on speaker stands that are 75cm high. The dimensions of the room are 520cm x 505 x 253 (L x W x H) and that would be 17.06ft x 16,57 x 8.3. The room is treated like a common room, without any special acoustic absorbers etc. Just some furniture, carpet (2x3m), curtains, two sofas and some flowers here and there. Floor is a wooden oak parquet, walls made of brick (right wall significantly thinner than others).
Room.png

The speakers are set up:
- 90 cm measuring from front wall to the speaker cone (60 cm to back of speaker cabinet)
- about 125 cm from side wall to center of the speaker cone.
- 250 cm apart from each other

Listening position is around 155 cm from the back wall, more or less in the center (this is where I set the mic).

I am attaching the measurements I made using Umik-1 microphone.

This is L + R channel, center position:
LR CENTER.png


This is L+R using RMS average:
LR CENTER RMS AVERAGE.png


This is L+R using Vector average:
LR CENTER VECTOR AVERAGE.png


Spectrogram:
SPECTROGRAM.png


Waterfall:
WATERFALL.png


I apologize if I forgot to provide a crucial graph. Thus I'm also attaching all my measurements from different positions (center, bit on the left, right, up, down etc.) and additionally with foams stuffed in BR port (half foams and full foams). I hope the descriptions are clear. https://www.mediafire.com/file/j5onbqfenvpq3yd/REW.zip/file

I would extremely appreciate your help with making my response as linear as possible.
 
Last edited:

Jdunk54nl

Addicted to Fun and Learning
Joined
Aug 5, 2020
Messages
968
Likes
1,048
Location
Arizona
From my perspective, everything above 150hz looks good and I probably wouldn't mess with it. I might make a couple changes and A/B the changes and see which one I liked better, but it would have to be some quick switching to probably here any differences.

Maybe a wide q adjustment from 150hz-600hz just to bring that down a little more in line with 600hz+ (or a shelf filter at 600hz to bring all that up a couple db). Otherwise it MAY sound a little "dull" with the highs being a little lower. Or maybe some might describe that as "muddy"as the 150hz to 600hz is a little higher. I hate trying to use "descriptors" for audio. I like just saying these frequencies are x different than these frequencies.

None of it is bad by any means and I would probably A/B/C the different options and see if I like it.

You can see the couple different trend lines I drew below. If you follow the orange, the highs are a little low, if you follow the red, the midrange is a little high. If you follow the blue, it is a little steep.

LR CENTER RMS AVERAGE.png




Now where you do need some fixing is about 150hz and below. Do you have DSP? That peak at ~35hz needs to either be brought down or the 40hz-150hz needs to be brought up (if it can and it is not caused by a room mode.

Do you have subwoofers or just the r3's? If you have subs, it looks like your r3's would be a perfect match for an 80hz crossover as they would be about 6db down naturally by that point and you could cross to subs there.
 
Last edited:
OP
M

Marcin

Member
Joined
Jan 12, 2022
Messages
49
Likes
20
Location
Poland
Thank you for your response.
I don't have a DSP. I thought about creating filters in REW, adding them to APO Eq and then try it out in my laptop connected to CXN V2 streamer. However I'm not really sure how to set Cutoff and Target level in REW EQ.

As for a sub I still have a REL T7x at home, but I will be returning it shortly, as it messed up with low frequencies for my taste.
 

Snarfie

Major Contributor
Forum Donor
Joined
Apr 30, 2018
Messages
1,168
Likes
925
Location
Netherlands
Have a try with Mathaudio room eq incombination with foobar2000 for free.
Regarding functionality it's REW an APO in one handy box an running in minuts.
 

Jdunk54nl

Addicted to Fun and Learning
Joined
Aug 5, 2020
Messages
968
Likes
1,048
Location
Arizona
Well, without a dsp, not much can be done. You could add a frequency specific acoustic panel for the 150hz-600hz region to tame that frequency range down. You would have to find a panel specific for those frequencies though.

You would then have to place it in different spots, take measurements and see what placement reduces the 150hz-600hz region by about 2db-3db. Once you have that tamed just the right amount, that is where you would want to permanently place that panel. However, based on your picture, a panel probably isn't going to be able to be placed appropriately with the sofa on one side and window on the other. Bass traps are HUGE and not worth it IMO. A dsp works much better. You could try some thicker curtains on the window and moving the couch (if leather couch, try covering with a thick blanket and see if that helps). I am a big fan of using sofa's, curtains, and rugs as room treatment.

The REL probably didn't sound great due to not having dsp to properly integrate it. Just adding a sub at random will usually just cause issues. You need some way to integrate mains to subs properly. A mini dsp 2x4hd would work well for you. It appears you definitely have some room modes at 35hz and 65hz. The sub probably excited those even more and made it sound terrible and "boomy".
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
i would only use one PEQ filter to bring the peak at 35Hz down, how far down (gain) would be up to preference.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
The sound must be a bit sub-bass anemic (for my taste) but there is nothing (menaingful) to do but buy a subwoofer because i don't think there is enough headroom to boost things up with something as small as the R3.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
One piece of advice, before you buy anything make sure that you actually have a problem and you know for sure that the product you will buy will fix that problem. Too many sales man will recommend their 'one solution fits all' products that may or may not in particular fix your particular problem.

Personally i think you would greatly benefit from corner bass traps to help the region between 100Hz and 600Hz. However the measurements do not show anything horrible really, i could definitely live with that.

I would not bother with any panels that are not at least 20 cm thick.
 

Jdunk54nl

Addicted to Fun and Learning
Joined
Aug 5, 2020
Messages
968
Likes
1,048
Location
Arizona
If you want to try APO Eq, try this filter.

Frequency 250hz
Q = 0.55
Gain = -3db (Might need something a little less here)

Turn it on and off as you listen to a known song and see if it works.

For the 35hz problem

Frequency 35hz
Q = 2.5
Gain = -5db

You might have to adjust the center frequency some. Without being able to use the cursor to get the exact center frequency, it is hard to guess what the exact frequency should be. These should be pretty close though.

To get the q, I took center frequency / bandwidth to impact. So for the 35hz, I took the bandwidth of 29hz-43hz or 14hz width. Then 35hz/14hz = 2.5.
Same thing for the 250hz. But I took bandwidth of 150hz-600hz or 450hz width. So 250hz/450hz = 0.55.
 

Jdunk54nl

Addicted to Fun and Learning
Joined
Aug 5, 2020
Messages
968
Likes
1,048
Location
Arizona
This is surely an effective way to take the warmth out of music :p, lets see if it will be useful for their particular situation.
It will take the slight muddy/boxy dwarf out ;)
Copied from Rational Acoustics.
7 Bad System Dwarfs.png
 

Jdunk54nl

Addicted to Fun and Learning
Joined
Aug 5, 2020
Messages
968
Likes
1,048
Location
Arizona
This is surely an effective way to take the warmth out of music :p, lets see if it will be useful for their particular situation.
I do agree though, their response is VERY good as is. I would have to A/B this through quite a few songs to see which one I would personally like more. I have no clue which one would be better, but if you are after a "more perfect" response, that would be needed.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
It will take the slight muddy/boxy dwarf out ;)

attachment.php
With headphones I always appreciated a flat mid-bass instead of the V Harman shape, makes the sound fuller and warmer. I have a lot of experience with The filter you recommended, but every room and person is different so who knows.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
Just some of my thoughts...

There is room for improvement beyond what DSP can offer. What's often missing here is an analysis of how the in-room time domain performance can be improved. If planning to acoustically optimize the measured listening position(s) further, at least aim to improve the clarity (C50) below 500Hz with acoustic treatment. The magnitude response below 100Hz may be improved perhaps by experimenting with positional adjustments -- also consider specialized diaphragmatic panels -- but, generally those are rather quite large and expensive, and require much more careful planning. Above that, acoustic absorption panels as thick as 4-6 inches will definitely help.

1642010315142.png

Try aiming for 5-10 dB

WAVE RIGHT.png 1642011832093.png
 

Attachments

  • @MARCIN averaged.mdat.zip
    3.2 MB · Views: 64
OP
M

Marcin

Member
Joined
Jan 12, 2022
Messages
49
Likes
20
Location
Poland
Today I made measurements with Dirac Live Trial version.
This is the outcome before correction:
2022-01-13 13_54_43-Dirac Live.png


and after:
2022-01-13 13_55_16-Dirac Live.png


I didn't have much time to do proper listening, so I'm not publishing any thoughts yet.

Huge thanks to all of you for the input. I will be checking those suggestions as well.

As for the acoustic treatment - I don't have anything on the back wall, it's totally "naked"... Would it be a good idea to hang some panels there and which frequencies would potentially benefit?
 

sharock

Active Member
Joined
Feb 9, 2021
Messages
259
Likes
276
Can you re-measure with REW after correcting with Dirac? In my (very limited) experience, Dirac is rather optimistic about what it can achieve. The measured results are usually not quite as good.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
@Marcin Try this target curve in Dirac instead.

Save it in a .targetcurve text file under C:\Users\*your username*\Dirac\targets and select it during the filter creation step

NAME
Unnamed
DEVICENAME

BREAKPOINTS
40 8.6
80 7.6
100.7 3.6
115.8 3.3
133.0 3
152.9 2.7
175.7 2.4
201.9 2.1
232.1 1.9
266.7 1.6
306.5 1.4
352.3 1.2
404.8 1
465.3 0.9
534.7 0.7
614.5 0.6
706.2 0.4
811.6 0.3
932.8 0.2
1072.0 0
1232.0 -0.1
1415.8 -0.2
1627.1 -0.3
1870.0 -0.4
2149.1 -0.4
2469.9 -0.5
2838.5 -0.6
3262.1 -0.7
3749.0 -0.8
4308.6 -0.8
4951.6 -0.9
5690.7 -1
6540.0 -1
7516.1 -1.1
8637.9 -1.1
9927.1 -1.2
11408.7 -1.2
13111.5 -1.3
15068.4 -1.4
17317.4 -1.4
18801.1 -1.4
LOWLIMITHZ
19.5
HIGHLIMITHZ
21797.2
 

Dj7675

Major Contributor
Forum Donor
Joined
Jan 12, 2019
Messages
2,116
Likes
2,781
Today I made measurements with Dirac Live Trial version.
This is the outcome before correction:View attachment 178742

and after:
View attachment 178743

I didn't have much time to do proper listening, so I'm not publishing any thoughts yet.

Huge thanks to all of you for the input. I will be checking those suggestions as well.

As for the acoustic treatment - I don't have anything on the back wall, it's totally "naked"... Would it be a good idea to hang some panels there and which frequencies would potentially benefit?
It will be interesting to hear your thoughts after listening to some of your favorite songs with/without correction. That is quite a bit of ”fixing” a really good speaker. I would experiment with different room curves, not correcting above different frequencies unless you find something lacking due to personal taste (too bright etc). Depending on the type of music/content you listen too, the biggest improvement would be the addition of a sub to handle the output under 80-100hz.
 
Top Bottom