Subwoofer Room Mode Treatment

Teerhand

Registered
Thread Starter
Joined
Jan 21, 2019
Messages
7
So i have been tinkering with REW for a couple of Months now, with devastating Results in the Beginning, because i don´t know much about Eqíng at all.
My first Targets in REW were set much too low, resulting in no Bass at all (No Bass, no Problem with Modes though :)
Then later on i managed to get nice sounding Bass but i realized after a couple of days, that i couldn´t hear the Bassist in certain Songs, which made me reconsider Eq´ing as a Noob alltogether.

So i hope you Guys can give me some tips on Measuring and applying EQ Methods.

I use an uncalibrated Samson condensator Mic, in the hope this doesn´t matter so much below 150hz
1) I use an external SPL Meter to get 85db on my Speakers (Adam A5x) and 93 on the Sub (Krk 10s) using the speaker calibration signal for the speakers and the sub signal for the sub (calibrating the subs volume is tricky because of the modes i guess)
2) I measure 23-150hz ; -17db; 256k; 8Sweeps; and all Speakers (because of Phase, right? at least that was my idea)
3) I open EQ with 1/24 smoothing and try to Eq just a little tiny bit, meaning i try to catch the bumps at 33 and 130 hz, by adjusting the target level pretty high. and a Match Range of 29-140hz without boost and withought narrow Filters.


This sounds pretty good, but heck what do I know basically?

Apart from this, i wasn´t able to apply boost to smooth out the dips in frequency response. I´m aware of the complications boost introduces, but i couldn´t get REW to try to flatten out the dips even when i set the highest boost values available.

So what can i improve? Is what i did catastrophic in your opinion etc. as i said, it does sound better now, but i had that boomy subwoofer signature for a while now, and before that i didn´t have a sub
 

Attachments

  • 76db24 smooth.txt
    680 bytes · Views: 6
  • Mär 11 11_50_40.mdat
    84.3 KB · Views: 9
Last edited:

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,212
As far as the EQ goes, your target level is too high - try 75 dB. You should also enable narrow filters, room resonances are narrow peaks and need narrow filters to counter them. It is fine to allow individual boost, that helps REW make sure the cuts are restricted to the range of the peaks, but better avoiding overall boost.

Measuring with single sweeps is fine, there's no need to use multiple sweeps.

There are arguments for running sub and both mains for a low frequency, principally an assumption that much bass is mixed monophonic - that can be a big assumption though, worth measuring the sub on its own and the combinations of left+sub and right+sub.

The main tool for addressing the dips is to change the positions of the speakers and where you sit and whatever acoustic treatment you can accommodate.
 

Matthew J Poes

AV Addict
Joined
Oct 18, 2017
Messages
1,903
Or multiple subs. By leveraging multiple Lf sources throughout a room you can more evenly energize the modes and reduce the chance to a dip in any one position.
 

Teerhand

Registered
Thread Starter
Joined
Jan 21, 2019
Messages
7
Hi John, thanks for replying! I have some questions to clarify what you wrote, if you dont mind.

Is applying smoothing the right thing to do? is 1/24th ok?

It is fine to allow individual boost

So what value do I adress to individual max boost then? i´m a bit shy, working with this, allthough my sub should have enough headroom. like 5 or 6 db? Overall max boost and Flatness Target stays at 0?


worth measuring the sub on its own and the combinations of left+sub and right+sub

I will do that, but how do i go from there? the automatic EQ can only be applied to one Measurement, right? Or can/should I merge all three Measurements to apply EQ?
I really don´t feel confident, to come up with my own EQ Settings after the terrible results with my first tries.

cheers,

Thomas
 

Teerhand

Registered
Thread Starter
Joined
Jan 21, 2019
Messages
7
Or multiple subs

Hi Matthew,

my gear is in a 3x3 m Listening Room, and while with such a bad architecture two subs would probably be my best bet, i really want to keep it simple, though, and doors and patio with big windows really doesn´t allow a lot of placements neither.

I moved the sub right next to my listening position, which really helped a lot. the measurement may still look bad to you, but that 60hz dip is nothing against what i had to work with before

>Edit: the Room is 4m x 4,5m so its not really square
 
Last edited:

Matthew J Poes

AV Addict
Joined
Oct 18, 2017
Messages
1,903
Placing the sub right next to your listening position does seem to make for much better response. I do not know this with certainty, but suspect that it increases the ratio of direct to reflected sound at lower frequencies. Normally there is far more reflected sound at low frequencies, but this method may shift that ratio.

Without multiple subs are significant bass traps there are not a lot of ways to combat dips in the response. Eq is really only effective against peaks or low q dips. Moving the speakers and subs can only shift things around. That doesn’t mean Johns advice is wrong, changing placement can shift things for a more favorable overall response in your listening room. It just can’t do as much as the other approaches.

High Q dips are not so audible anyway, so if you are happy with this, be happy and don’t worry about it. If you hear something specific you want to address we are here to help.
 

Teerhand

Registered
Thread Starter
Joined
Jan 21, 2019
Messages
7
changing placement can shift things for a more favorable overall response in your listening room

your replies gave me some motivation, to reconsider subwoofer placement, and i moved my desk as well. so i did this till i had only peaks and no dips below 100hz (i really don´t trust the mic enough to go much beyond that, and the power of the sub poses the biggest problem anyway)
With the new placement i get a pretty flat response and REW proposes only three EQ adjustments ( 32.53 Hz Gain -10.00 dB Q 7.680 ;64.33 Hz Gain -9.40 dB Q 8.574; 74.12 Hz Gain -6.90 dB Q 15.843)
so i´m glad i moved the sub again.

btw, did you do the presentation on REW? That was pretty awesome, helped me a lot!

cheers
 

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,212
Is applying smoothing the right thing to do? is 1/24th ok?
Low frequency measurements are best viewed unsmoothed.

So what value do I adress to individual max boost then? i´m a bit shy, working with this, allthough my sub should have enough headroom. like 5 or 6 db? Overall max boost and Flatness Target stays at 0?
Allowing individual boost does not use up any headroom. The overall max boost setting controls the overall boost of the filter set, if that is set to zero REW ensures there is no overall boost regardless of whether individual filters have boost. Filters with boost that overlap filters with cut can do their job without creating any overall boost. I typically use 9 dB for the individual max boost.

I will do that, but how do i go from there? the automatic EQ can only be applied to one Measurement, right? Or can/should I merge all three Measurements to apply EQ?
I really don´t feel confident, to come up with my own EQ Settings after the terrible results with my first tries.
A separate measurement of the sub lets you see what it is contributing, you might choose to only EQ the sub. You might also find that changing sub phase helps remove a dip that is seen in sub + main measurements. You should look at the results of all 3 measurements with any EQ you apply to see what tradeoffs you are making.
 

Teerhand

Registered
Thread Starter
Joined
Jan 21, 2019
Messages
7
Filters with boost that overlap filters with cut can do their job without creating any overall boost

Ah ok, i didn´t know that, but i moved my whole Desk around the room and measured the best response (also switched phase back to 0) and i found a position for the sub where no dips occured and REW EQ only proposed three Filters, sounds pretty amazing now. I seem to have caught only the elevated resonances, because an on/off comparison sounds pretty similar but takes away the boomyness...love it.
Also i only eq´ed up to 100hz, because i don´t really trust that samson mic. at some point i´ll get a umik 1 and give a bigger spectrum eq a try...for now i´m very happy though.

you´re awesome man, thanks
 

Matthew J Poes

AV Addict
Joined
Oct 18, 2017
Messages
1,903
your replies gave me some motivation, to reconsider subwoofer placement, and i moved my desk as well. so i did this till i had only peaks and no dips below 100hz (i really don´t trust the mic enough to go much beyond that, and the power of the sub poses the biggest problem anyway)
With the new placement i get a pretty flat response and REW proposes only three EQ adjustments ( 32.53 Hz Gain -10.00 dB Q 7.680 ;64.33 Hz Gain -9.40 dB Q 8.574; 74.12 Hz Gain -6.90 dB Q 15.843)
so i´m glad i moved the sub again.

btw, did you do the presentation on REW? That was pretty awesome, helped me a lot!

cheers

Yes that was me. Thanks. Glad it helped.

John answered a lot of your technical questions. Glad this worked out.

Maybe you said but what mic are you using?
 

Matthew J Poes

AV Addict
Joined
Oct 18, 2017
Messages
1,903
Oh I see.

Yeah that has a non-flat response and a cardioid pattern. If I were you, I would consider investing in a measurement mic. A usb measurement mic can be had for cheap.

@John Mulcahy might better know what happens at low frequencies. Cardioid mics are directional, with a null toward the back and sides of the rear of the mic. It becomes forward biased in what it picks up. However, at low frequencies and pointed up, im not sure if it makes such a big difference. The bigger concern is that it seems like the response is sloping downward. It might need a compensation curve as it looks about 5dB down at 20hz.
 

Teerhand

Registered
Thread Starter
Joined
Jan 21, 2019
Messages
7
However, at low frequencies and pointed up, im not sure if it makes such a big difference

Yes, that is what i was counting on, also my sub (krk10s v1) doesn´t do much below 30 hz really, but that downward slope starts earlier as it seems, according to their response curve. If i would assume a 2,5 db deviation between 100 and 30hz i could live with that, until i decide i want to delve deeper into eq´ing full range...i was contemplating buying a umik 1 for that anyway...its about 110€ here in Germany. But at the moment i´m quite happy (i´ve got to say, that i liked the A5x without a sub out of the box, and the only room treatment is a diffusor on the back wall, well and furniture :)
 
Top Bottom