• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Need some help for my room correction

Max

Member
Forum Donor
Joined
Mar 8, 2019
Messages
68
Likes
50
Location
France
#1
Hi everyone :)

If you don't mind, I guess I need your help to push one step forward my ongoing room correction. I'm not sure being on the right sub-forum, so if so, just tell me ;).

At this step,
- I've made multiple measurements around my listening point (MMM, via REW)
- I've synchronized these measures in the time domain (to avoid any time shift, via REW)
- I've added the signals to get the average measure (via REW)
- I've used the EQ tool of REW to fit a Harman-like target and to get a first global correction
- I've tuned this correction with the Paragraphic Gain EQ via Rephase.
(NB : I avoided to over-tune by using Q<8 and I avoided to fill the down gaps).
(NB: I've tried to tune the phase, too, with the Paragraphic Phase EQ, but the phase is too messy and I've never succeeded to get good results, so I give up)
- Exporting the impulse file (65536 taps, the right sample rate) and uploading into the convolver
- Final test: the listening test. Until now, I find the correction weird: it sounds bad, not natural at all :rolleyes:

That makes me think I overdid the correction. So, I'm going back to the basis and I want to start the correction again from the beginning :)

Could you advise me about the right way to correct the following (average) measures? Which waveband in priority? In which way?

Thanks for your help :)
 

Attachments

Last edited:

Hipper

Addicted to Fun and Learning
Joined
Jun 16, 2019
Messages
564
Likes
462
Location
Herts., England
#2
Firstly this is the right sub-forum.

I don't have any knowledge of the Rephase or exactly what it is you have done. I've used REW but to create filters to use in a Behringer DEQ2496 equaliser after using lots of room treatment and taking care with positioning of speakers and chair.

I can offer the following advise. Firstly the biggest issue in solving the problems of the interaction with the room and your speakers is the bass region, say 0-300Hz. This area can get so damaged that it hides the mids and highs. Sort it out and these higher frequencies can be heard as they should.

I've always worked only with frequency response (FR) graphs, never phase or decay times, but I've noticed that as you make the FR smoother, so phase response also improves. I found decay times improve quite dramatically with room treatment (some say with EQ too but I don't know). Decay times make a big difference to the sound, phase I'm not so sure.

Your measurements show major bass issues - below 200Hz it's way to high. Concentrate on getting this down to the same levels as the rest of the graph (around 80dB) and trying to get this as smooth as possible.

Have a listen. You may find the bass a bit anemic compared to the past but you should get used to it. The main point is that you should now hear all bass notes clearly. If you want more bass then do something like the Harman curve - give it a downward slope from 40-500Hz of 3 to 6dB say.

If you find you need to adjust the frequencies above 300Hz treat them like using tone controls, with a wide Q.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
11,395
Likes
11,848
Location
Riverview FL
#3
Could you advise me about the right way to correct the following (average) measures?
No.

My way is flat across the audio band.

Get rid of the bass hump, as advised above, would be the first thing I'd do.

That leaves the sound nicely balanced for 80dBSPL average and 100dBSPL peaks in my opinion, critical listening levels, the flat area of the old Fletcher Munson curves.

I don't "miss" bass at lower levels, and it isn't too much when louder.
 
OP
Max

Max

Member
Forum Donor
Joined
Mar 8, 2019
Messages
68
Likes
50
Location
France
Thread Starter #4
I can offer the following advise. Firstly the biggest issue in solving the problems of the interaction with the room and your speakers is the bass region, say 0-300Hz. This area can get so damaged that it hides the mids and highs. Sort it out and these higher frequencies can be heard as they should.
[...]
Your measurements show major bass issues - below 200Hz it's way to high. Concentrate on getting this down to the same levels as the rest of the graph (around 80dB) and trying to get this as smooth as possible.

Have a listen. You may find the bass a bit anemic compared to the past but you should get used to it. The main point is that you should now hear all bass notes clearly. If you want more bass then do something like the Harman curve - give it a downward slope from 40-500Hz of 3 to 6dB say.
Great, thanks for the advice, I'll try that :).
And to start with, I guess I will first only tune the amplitude through an equalizer to hear and better understand what you describe about low/higher frequencies hearing.
 
OP
Max

Max

Member
Forum Donor
Joined
Mar 8, 2019
Messages
68
Likes
50
Location
France
Thread Starter #5
I've always worked only with frequency response (FR) graphs, never phase or decay times, but I've noticed that as you make the FR smoother, so phase response also improves. I found decay times improve quite dramatically with room treatment (some say with EQ too but I don't know). Decay times make a big difference to the sound, phase I'm not so sure.
.
When you say "room treatment", you're talking about a (physical) treatment to avoid reflections in the room? With absorbing/scattering media like carpets, etc.?
 
OP
Max

Max

Member
Forum Donor
Joined
Mar 8, 2019
Messages
68
Likes
50
Location
France
Thread Starter #6
My way is flat across the audio band.

Get rid of the bass hump, as advised above, would be the first thing I'd do.

That leaves the sound nicely balanced for 80dBSPL average and 100dBSPL peaks in my opinion, critical listening levels, the flat area of the old Fletcher Munson curves.

I don't "miss" bass at lower levels, and it isn't too much when louder.
Great, thanks for the feedback. I'll try it first, too. Thanks a lot :)
 

Webninja

Active Member
Forum Donor
Joined
Oct 8, 2018
Messages
274
Likes
271
Location
Los Angeles
#7
I use REW to verify changes done in the Audyssey app, so no pinpoint adjustments, just trying to get closer to the Harman curve.

One of the best tips I read was to only “play” with frequencies under ~500Hz I’ve only made it worse when I EQ over that. I’ve tried applying Audyssey to 300Hz and under and 1k and under, but 500 Hz worked the best.

That way you let your speakers sounds the way they were designed for the majority of the frequencies. Also if you’re looking at other people’s REW graphs try to match the scales, I usually see 1/6.
 

raindance

Senior Member
Joined
Sep 25, 2019
Messages
486
Likes
430
#8
Is that really 20dB of bass boost or are the speakers very close to the wall (and corner) causing room boost? There is no amp in the world that could handle 20dB of bass boost and retain enough headroom to play music. You need to work on sensible room placement and then EQ out the peaks. No boosts. See how that sounds before concerning yourself with a house curve.

Also, if you are using Audyssey, disable dynamic EQ before measuring.
 
OP
Max

Max

Member
Forum Donor
Joined
Mar 8, 2019
Messages
68
Likes
50
Location
France
Thread Starter #10
Is that really 20dB of bass boost or are the speakers very close to the wall (and corner) causing room boost?
No bass boost, actually. The speakers are indeed close the wall and one of them is near the corner.

There is no amp in the world that could handle 20dB of bass boost and retain enough headroom to play music. You need to work on sensible room placement and then EQ out the peaks. No boosts. See how that sounds before concerning yourself with a house curve.
Unfortunately, I have currently no real other choice. My hifi setup is in the only place available (small appartment). So, I try digital room correction instead :).
 

raindance

Senior Member
Joined
Sep 25, 2019
Messages
486
Likes
430
#12
No bass boost, actually. The speakers are indeed close the wall and one of them is near the corner.



Unfortunately, I have currently no real other choice. My hifi setup is in the only place available (small appartment). So, I try digital room correction instead :).
No problem, just EQ the peaks down and let's see how it goes. The SBIR dip is high enough in frequency that you can absorb it with 4" thick acoustic absorbers.
 
OP
Max

Max

Member
Forum Donor
Joined
Mar 8, 2019
Messages
68
Likes
50
Location
France
Thread Starter #13
You have a bad case of SBIR. Visible in sharp dips in 150-300 Hz.

Boosting that will sound really bad.

Feel free to check this thread for advice on how to deal with it

https://www.audiosciencereview.com/...dip-between-200hz-and-300hz-in-my-room.22274/
No problem, just EQ the peaks down and let's see how it goes. The SBIR dip is high enough in frequency that you can absorb it with 4" thick acoustic absorbers.
Thanks abdo and raindance, I didn't know SBIR. It makes me remember that I have some foam addon (given with the speakers) to fill more or less the acoustic vent. I made tests with and without (and measured it) but I guess I looked at the wrong waveband: I gueesed it will attenuate the strong bass bump and not the SBIR area (and it had no impact on the bass bump's behavior). Do you think it could be used for this purpose?
 

alex-z

Active Member
Joined
Feb 19, 2021
Messages
130
Likes
190
Location
Canada
#14
Pretty much anything under 300Hz is a consequence of the room itself or speaker placement. Be sparing with the EQ there, as it often makes problems worse.

If your speakers are close to the walls, put absorption behind them. A 4" mineral wool panel spaced 2" off the wall should improve things significantly.

500-900Hz region looks like a region you would want to EQ. You can run an impedance sweep to check if that variation is caused by cabinet resonances.

That peak just after 42Hz appears to be the port tune of your speakers, perhaps coinciding with a room mode? I would knock off 5dB there.

Your entire bass region seems to be running at least 5dB too high. I assume you aren't using a sub because of the sharp rolloff after 40Hz. I strongly recommend adding a sub and crossing at 80Hz. That will have multiple benefits, such as lower frequency extension, less distortion, and proper level matching via the amp gain, instead of using EQ.
 
Last edited:

abdo123

Major Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
2,579
Likes
2,235
Location
Brussels, Belgium
#15
Thanks abdo and raindance, I didn't know SBIR. It makes me remember that I have some foam addon (given with the speakers) to fill more or less the acoustic vent. I made tests with and without (and measured it) but I guess I looked at the wrong waveband: I gueesed it will attenuate the strong bass bump and not the SBIR area (and it had no impact on the bass bump's behavior). Do you think it could be used for this purpose?
what you need is foam or something that is not dense at least 10-15 cm thick surrounding all boundaries of the speaker that are not in touch with the wall, the thicker the better.

even a hollow/empty 'flexible' wooden box could show improvements.
 
Last edited:

Hipper

Addicted to Fun and Learning
Joined
Jun 16, 2019
Messages
564
Likes
462
Location
Herts., England
#16
When you say "room treatment", you're talking about a (physical) treatment to avoid reflections in the room? With absorbing/scattering media like carpets, etc.?
My room treatment is lots of bass traps and some smaller panels. I use Soffit Bass Traps, 244 and 242 panels from GIK:

https://gikacoustics.co.uk/product-category/bass-traps/

I have a dedicated room though and am not concerned with looks.

Like Ray Dunzl, I aim for a flat response and am happy with that. But there are no rigid rules. Aim for flat and try for a bit, then if not happy adjust for your taste.

I always used 'no smoothing' when looking at REW FR graphs for bass. You can see mine and other efforts here:

https://www.audiosciencereview.com/...in-room-measurements.13540/page-3#post-411614
 
Last edited:

raindance

Senior Member
Joined
Sep 25, 2019
Messages
486
Likes
430
#17
Thanks abdo and raindance, I didn't know SBIR. It makes me remember that I have some foam addon (given with the speakers) to fill more or less the acoustic vent. I made tests with and without (and measured it) but I guess I looked at the wrong waveband: I gueesed it will attenuate the strong bass bump and not the SBIR area (and it had no impact on the bass bump's behavior). Do you think it could be used for this purpose?
SBIR is physics. It's determined by the distance that the speakers are from the front wall. Plugging the ports may help a little with the port tuning freq.
 

raindance

Senior Member
Joined
Sep 25, 2019
Messages
486
Likes
430
#18
Pretty much anything under 300Hz is a consequence of the room itself or speaker placement. Be sparing with the EQ there, as it often makes problems worse.

If your speakers are close to the walls, put absorption behind them. A 4" mineral wool panel spaced 2" off the wall should improve things significantly.

500-900Hz region looks like a region you would want to EQ. You can run an impedance sweep to check if that variation is caused by cabinet resonances.

That peak just after 42Hz appears to be the port tune of your speakers, perhaps coinciding with a room mode? I would knock off 5dB there.

Your entire bass region seems to be running at least 5dB too high. I assume you aren't using a sub because of the sharp rolloff after 40Hz. I strongly recommend adding a sub and crossing at 80Hz. That will have multiple benefits, such as lower frequency extension, less distortion, and proper level matching via the amp gain, instead of using EQ.
Are we looking at the same graphs? I saw 20dB elevated low frequencies in a clump which indicates proximity to boundaries. Or did I miss the "after" graphs?
 
OP
Max

Max

Member
Forum Donor
Joined
Mar 8, 2019
Messages
68
Likes
50
Location
France
Thread Starter #19
Your entire bass region seems to be running at least 5dB too high. I assume you aren't using a sub because of the sharp rolloff after 40Hz. I strongly recommend adding a sub and crossing at 80Hz. That will have multiple benefits, such as lower frequency extension, less distortion, and proper level matching via the amp gain, instead of using EQ.
Indeed, no subwoofer used. I don't have any for the moment but I had wondered about it a while ago, so why not in the future and maybe in a near future. So, you recommend to add a sub and cross at 80Hz. Why not adding a sub without crossing? I thought there was some benefits to multiply source of bass frequencies in the room (to reduce the resonances related to the room).
 
OP
Max

Max

Member
Forum Donor
Joined
Mar 8, 2019
Messages
68
Likes
50
Location
France
Thread Starter #20
what you need is foam or something that is not dense at least 10-15 cm thick surrounding all boundaries of the speaker that are not in touch with the wall, the thicker the better.

even a hollow/empty 'flexible' wooden box could show improvements.
I think it will be a bit hard to make so (WAF - Picture of the current positionning of my right speaker - the left one is on the other side of the piano).
But it is very interesting and I enjoy to learn about that. To apply as soon as I live in a bigger place :)
 

Attachments

Top Bottom