• 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.

Room EQ, do's and dont's

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,476
Likes
1,133
Location
Zg, Cro
Alternatively, I could try center 72 Hz, 1/7 octave each side (Q=5). That would cover 65 to 80 Hz. Then add my current center 64, Q=1.4 filter on top of it, +6 dB each, to flatten the entire range 45 to 90 Hz and bring it up to level. But that still would be +12 dB of boost, which demands 16x more power from the amp in this frequency range, and consequent phase distortion from a steep filter, and a -12 dB overall level reduction to avoid digital clipping.
Filters with Q up to 10 are perfectly ok in the LF region as long as your measurement is precie enough to jsutify their usage.

Filters with Q up to 5 are certainly not "steep" filters. From what I know human ear is fogriving to phase variations with LF, less with MF and much less with HF.

If your filters have max gain of 12dB and you reduce overall gain in convolution engine for the same ammount (and you shuold, to avoid clipping) your amp and your speakers won't have any additional stress.

P.S. I still suggest you try it with 70Hz and Q=3 and see what will happen. Nothing to loose.. :)
 
Joined
Feb 5, 2019
Messages
96
Likes
55
I'll try this and post results.
If all these mitigating factors are true, the worst thing that could happen is due to the -12 dB gain reduction I run out of gain, volume knob cranked nearly wide open for listening. Some high dynamic range recordings already require a -6 dB setting on the volume knob!

NOTE: I think the amp will indeed have more stress. If I listen to the same music at the same volume level, I will be turning up the volume 12 dB louder to have the same volume through most of the frequency range. And this will demand the amp produce 16x more power in the 45 to 90 Hz range. But I'm not too worried about it, it has plenty of power (Adcom 5800, 450 wpc into this 4ohm load).
 
Last edited:

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,476
Likes
1,133
Location
Zg, Cro
I'll try this and post results.
If all these mitigating factors are true, the worst thing that could happen is due to the -12 dB gain reduction I run out of gain, volume knob cranked nearly wide open for listening. Some high dynamic range recordings already require a -6 dB setting on the volume knob!
Looking forward to it - good luck! :)

Most of the folks are pleasantly surprised after they make EQ. And yes, that is pretty much the only trade off.
 
Joined
Feb 5, 2019
Messages
96
Likes
55
Adding +6 dB @ 72 Hz, Q=5 in addition to the +6 dB @ 64 Hz Q=1.4, (overall -12 dB in level) gives only a +2 dB improvement. Whatever I EQ, the room is soaking up nearly all of the energy. Also, measuring the L only and R only individually each gives the exact same FR curves.
Conclusion: this 72 Hz null is definitely a room mode and I'm going to need room treatment to weaken it.

So: what's the best recommended treatment for narrow room modes like this? I would guess bass traps; I've read they can be tuned more narrowly than tube traps, which I already use and have broader spectrum effect.
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
2,624
Likes
1,718
Adding +6 dB @ 72 Hz, Q=5 in addition to the +6 dB @ 64 Hz Q=1.4, (overall -12 dB in level) gives only a +2 dB improvement. Whatever I EQ, the room is soaking up nearly all of the energy. Also, measuring the L only and R only individually each gives the exact same FR curves.
Conclusion: this 72 Hz null is definitely a room mode and I'm going to need room treatment to weaken it.

So: what's the best recommended treatment for narrow room modes like this? I would guess bass traps; I've read they can be tuned more narrowly than tube traps, which I already use and have broader spectrum effect.
Yep, I think your reasoning is sound here. (Although technically it's not a room mode - it's a node ;) )

Anyway, to just take one last step to verify your hypothesis, could you tell us what your room dimensions are and what the distances are from each speaker (approximate acoustic centre of the woofer) to each adjacent wall.

It's conceivable that this dip is actually caused by interference from the first reflection points rather than (or as well as) a room node per se. If this were the case, changing speaker placement would be the first port of call.
 
Joined
Feb 5, 2019
Messages
96
Likes
55
I thought it was a room mode (https://en.wikipedia.org/wiki/Room_modes). Is a mode the peak and a node the trough?

I've moved the speakers around and measured each position. The 72 Hz null doesn't change, but other parts of the FR do change, and the current position gives the flattest response at the listener position without EQ.

No woofer, these are Mag 3.6/R. Here's the REW room sim screenshot. It shows a boost around 70 Hz and a cut around 100, which is the exact opposite of what I actually have. Am I reading the black line backward?
REW-room-sim.png
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
2,624
Likes
1,718
I thought it was a room mode (https://en.wikipedia.org/wiki/Room_modes). Is a mode the peak and a node the trough?

I've moved the speakers around and measured each position. The 72 Hz null doesn't change, but other parts of the FR do change, and the current position gives the flattest response at the listener position without EQ.

No woofer, these are Mag 3.6/R. Here's the REW room sim screenshot. It shows a boost around 70 Hz and a cut around 100, which is the exact opposite of what I actually have. Am I reading the black line backward?
View attachment 23163
Ok yes, that indicates that the dip is not a product of room placement/early reflections.

I'd missed earlier in the thread when you said these were large dipole speakers. I wonder, could that dip be a property of the speakers themselves?

It's very difficult to measure speakers like these so I'm not sure we can glean much from Stereophile's measurements, and a single on-axis measurement is not very helpful, but there does seem to be a dip in the bass response at a similar frequency measured there.

So I think there's a distinct possibility that the front and rear waves are creating a cancellation in that frequency range that won't be easily corrected.

Perhaps others here with more experience with planar speakers can offer a better-informed opinion.

And perhaps you could also try taking some nearer-field measurements to see if they give a better idea what's going on?
 

Attachments

DonH56

Major Contributor
Technical Expert
Patreon Donor
Joined
Mar 15, 2016
Messages
2,477
Likes
2,334
Location
Monument, CO
70 Hz is either a room mode or more likely the back wve from the Maggies canceling the front wave. The back wave hits the wall behind, inverts, and returns so when it hits the speaker again it can cancel the front wave. That is one of the reasons I have usually damped the wall behind and used a sub so they can be a reasonable distance to the wall behind in my usually modest-sized rooms.

The wavelength of a 70 Hz signal is about 16.1 feet. A quarter-wavelength difference causes cancellation so if they are about 4' from the wall behind you will have a null at 70 Hz.
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
2,624
Likes
1,718
The wavelength of a 70 Hz signal is about 16.1 feet. A quarter-wavelength difference causes cancellation so if they are about 4' from the wall behind you will have a null at 70 Hz.
The curious thing here is that according to @MRC01 the dip is there regardless where in the room the speaker is positioned.
 

DonH56

Major Contributor
Technical Expert
Patreon Donor
Joined
Mar 15, 2016
Messages
2,477
Likes
2,334
Location
Monument, CO
That would point to a room node (null). What if he measures at a different position in the room? Room modes at LF are somewhat insensitive to speaker positioning.
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
2,624
Likes
1,718
That would point to a room node (null). What if he measures at a different position in the room? Room modes at LF are somewhat insensitive to speaker positioning.
Yep, I think that’s the most likely explanation too.

@MRC01 perhaps try a different room mode calculator to rule out an issue with the one in REW. Or is your room an unusual shape/construction perhaps, enough so as to confound the modelling software?
 
Joined
Feb 5, 2019
Messages
96
Likes
55
The speakers are 39" from the front wall, which should cancel 87 Hz. That portion of the FR curve is OK. The room has openings along the front corners which should weaken its modes and the back wave reflection. The back corners have big tube traps (2' diameter, floor to ceiling). The back wall has 4-layer thick RPG acoustic foam.

I'm going to make some new measurements. One near-field to confirm the 72 Hz dip goes away. Others from other locations in the room to confirm whether the dip is location-sensitive; I expect it will be.

If all this confirms that this is a room null, as I expect it will, what is the best room treatment? I'm looking for something tuned for 72 Hz with Q=5. Bass trap?
 
Joined
Feb 5, 2019
Messages
96
Likes
55
70 Hz is either a room mode or more likely the back wve from the Maggies canceling the front wave. The back wave hits the wall behind, inverts, and returns so when it hits the speaker again it can cancel the front wave. That is one of the reasons I have usually damped the wall behind and used a sub so they can be a reasonable distance to the wall behind in my usually modest-sized rooms.
...
That's true. Measured near-field, the 3.6/R has big a rise in FR around 50 Hz, which suggests they're designed to be 5-6' from the front wall so the reflected back wave cancels this to give flat response. That requires a big room. If the room isn't big enough for that they require alternate approaches like damping the front wall.
 
Joined
Feb 5, 2019
Messages
96
Likes
55
I just realized it's possible my room is boosting 72 Hz just like the simulator predicts. But my listener position happens to be located between the peak pressure ridges in the room, so I'm hearing the trough rather than the peak. If so, moving the listener position a few feet could eliminate this null, or turn the 72 Hz trough into a peak.
If that's how it works, I'll smack myself in the head now for missing the obvious.
PS: a high-Q bass absorber tuned to 72 Hz would still be a great solution for this room.
 
Last edited:
Joined
Feb 5, 2019
Messages
96
Likes
55
In case it's helpful for others, here's a followup of my plan:

1. Any EQ settings more than +6 dB make the bass sound loose and bloated. And, it also shows in measurements with high distortion and ringing. I'd rather have taut clean attenuated bass, then big bloated bass. The bass is still OK with taut linear response from 35-50 and from 100 on up. I'll fix that 50-100 suckout octave with room treatment not EQ.

2. The 2 best remedies for high Q bass attenuation seem to be GIK Scopus tuned membranes, and the RealTraps MegaTrap. Their effective ranges are 55-100 Hz and 70-150 Hz respectively. This neatly fits my problems at 74 and 100 Hz.

I couldn't decide which, so I'm starting with a couple of each. If anyone wants to build your own, Google "Tim Farrant BuzzAudio limp mass bass absorber".

If the room were not already treated, I would take a completely different approach. But I already have a well-damped room that sounds and measures well, but for these nodes at 74 and 100 Hz. (The room was FAR worse before I started!) I don't want the room to be more dead, more live, or any other changes. Just a surgical change at these particular frequencies. So a couple of tuned treatments seems to be the right scalpel.
 
Joined
Feb 5, 2019
Messages
96
Likes
55
PS: I was wondering why my speakers didn't show the normal SBIR frequency effects given their distance to the front wall (the wall behind the speakers, in front of the listener). Thinking further, I believe the formula must be different for dipoles than for conventional speakers.

Normally, the SBIR null frequency is 1/4 wavelength from the front wall. That is, the back-wave bounces off the front wall so it covers the distance twice, so 1/4 wavelength puts it a half wavelength delayed behind the front-wave which is 180* out of phase. But this assumes radiating spherically with the back-wave having the same polarity as the front wave.

The back-wave from a dipole has inverted polarity from the front-wave, so the null should happen when the back-wave has traveled a full wavelength, which is when the speaker-wall distance is 1/2 wavelength. In fact, a dipole should BOOST the 1/4 wavelength frequency because after it bounces & returns it's covered 1/2 wavelength which, because its polarity is inverted, aligns it with the front-wave.

If true, it would help explain some of my measurements.
 
Joined
May 30, 2018
Messages
75
Likes
54
I'll play - here's a work in progress. Need some room correction but not terrible terrible:

c1 eq.jpg


Major room suck outs at around 70Hz, 150 and 11K, a stubborn peak at 1100Hz. I can do something about 330Hz but still a work in progress. This is from about 12' back from mains. Not exactly the sweet spot as I'm trying for pretty even coverage for the room as it's a work space. 3 way active system (miniDSP 4x10) with 2 subs.
 

Attachments

Last edited:

Music1969

Active Member
Joined
Feb 19, 2018
Messages
160
Likes
39
Apologies if I've missed it recently but has @amirm shared his current room response curve?

I assume it'd look pretty good! Using Dirac Live?
 
Joined
Feb 5, 2019
Messages
96
Likes
55
I'll play - here's a work in progress. Need some room correction but not terrible terrible:
...
Major room suck outs at around 70Hz, 150 and 11K, a stubborn peak at 1100Hz. I can do something about 330Hz but still a work in progress. This is from about 12' back from mains. Not exactly the sweet spot as I'm trying for pretty even coverage for the room as it's a work space. 3 way active system (miniDSP 4x10) with 2 subs.
What are the room dimensions/shape?
What have you done so far with room arrangement and treatment?
Do the peaks & troughs you measure correspond to expected room modes, SBIR or LBIR distances?
What does your spectral decay look like -- do you have ringing at these peaks & troughs?
PS: the reason for these questions. Each frequency response peak or dip can be caused by different things. First identify what's causing it, then you know how to fix it. Always try to identify & eliminate the problem at its root cause, before resorting to EQ as a last resort.
 
Last edited:
Top Bottom