Crossover, speaker alignment, room equalization - what do I start with?

Lid9497

Member
Thread Starter
Joined
May 23, 2025
Posts
16
More  
Preamp, Processor or Receiver
Sound Blaster X4
Main Amp
Fosi Audio V3-48V
Front Speakers
Denon SC-M41
Subwoofers
SVS PB-1000 Pro
Hello wonderful community,

So I have 2.1 system and basically nothing set up, no crossover, no room correction, nothing and I don't know what should be the first thing to fix or do: speaker alignment, or crossover, or first equalize mains and sub separately...what is your first step in REW after connecting and positioning all the speakers? I will first try to fix mains and go from there, but I think maybe speaker alignment should be a first step since I have subwoofer very far away from me.

Thanks :)
 
Welcome to the Forum!

I know this stuff can be intimidating. First, set up the crossover between the mains and sub, time-align the speakers to the sub, then do the EQ and fine-tuning.

Regards,
Wayne
 
Your speakers frequency response is 45 Hz to 40 kHz, according to Denon...
If your system is primarily for stereo music listening you might set your sub/bass crossover at 45Hz to ~55Hz...
If your system is primarily for AudioVideo you might set your sub/bass crossover at ~80 Hz as suggested by both Dolby and THX...
 
Your speakers frequency response is 45 Hz to 40 kHz, according to Denon...
If your system is primarily for stereo music listening you might set your sub/bass crossover at 45Hz to ~55Hz...
If your system is primarily for AudioVideo you might set your sub/bass crossover at ~80 Hz as suggested by both Dolby and THX...
Thanks for the tips, it is audio,video, games, I was using my old subwoofer from Genius 5.1 sound system and I had 80Hz on subwoofer and 65Hz on mains, seemed to work well and boosted some nulls in 60Hz range that subwoofer could not deliver to my LP, tomorrow I will begin experimenting with sub placement and measuring, really excited :D have to work today :(
 
So new subwoofer arrived, I have tested my entire room, changing listening positions, sub placement, all while measuring each position. Interestingly, I got best response when I face subwoofer cone directly into the corner of my room diagonally across from me. Any other positions had extreme nulls either in crossover region or low-end (below 35Hz). So this is best I think I can get out of this room.

I measured what I could think of, best response was at 80Hz/24db slope sub facing corner. Crossover in SVS app is set to 80Hz/24db slope, and I decided to not put any crossover on my mains. Using frequency generator and going by 1Hz, I determined they can produce sound loud enough and without distortion from around 55Hz, so I applied simple graphics equalizer on L+R speakers and put -50db on all frequencies below 53Hz or so, 55Hz and above mains get full signal, all of this is applied in final measurement "80Hz 24db eq.mdat" which also include my attempt at equalizing.

But I am having little trouble with those dips. Let's say I have -10db at 50-60Hz from 75db to 65db. All other frequencies are above 75db. How do I set the equalizer to drop everyhing to 65db (where 50-60Hz is) so I don't have to boost anything? It always takes the dips with entire curve.

Is it also possible to determine how the room needs to be treated, or even whether it could be, from the measurements?

I have included all my measurments.Thanks for reading and tips :)
 

Attachments

Last edited:
Are using a windoz PC or Mac as your main platform to drive this Sound Blaster X4? What software do you use to apply a "simple graphics equalizer"? Is there some reason you aren't using a Parametric EQ? You might read about the differences, as some of your issues may be better solved with PEQ...

It would be better to measure each L and R speaker separately...

As for those dips, they may be mostly room modes... You can use the Room Sim function of REW that will estimate where your room modes would occur... Or https://amcoustics.com/tools/amroc... Are you low passing your new sub around where you are rolling off your mains?

Bass traps are crucial for absorbing low-frequency sound waves that cause room modes, especially in corners. Consider using thick, broadband bass traps or tuned traps for specific frequencies... Wide-band absorption panels strategically placed can help reduce reflections and mitigate modal behavior, particularly in the mid and high frequencies... This is a good place to start understanding Bass trap placement... http://arqen.com/bass-traps-101/placement-guide/ You can build your own or get some from https://www.gikacoustics.com/ or many others...
 
Last edited:
Are using a windoz PC or Mac as your main platform to drive this Sound Blaster X4? What software do you use to apply a "simple graphics equalizer"? Is there some reason you aren't using a Parametric EQ? You might read about the differences, as some of your issues may be better solved with PEQ...

It would be better to measure each L and R speaker separately...

As for those dips, they may be mostly room modes... You can use the Room Sim function of REW that will estimate where your room modes would occur... Or Are you low passing your new sub around where you are rolling off your mains?

Bass traps are crucial for absorbing low-frequency sound waves that cause room modes, especially in corners. Consider using thick, broadband bass traps or tuned traps for specific frequencies... Wide-band absorption panels strategically placed can help reduce reflections and mitigate modal behavior, particularly in the mid and high frequencies... This is a good place to start understanding Bass trap placement...You can build your own or get some from or many others...
Yes, those are room modes. I am on Windows using EQApo with Peace ParametricEQ. I am applying GEQ on L+R and PEQ from REW AutoEQ on All channels, system-wide. I am downmixing from 5.1 to 2.1 using Peace command window, with these values:
Copy: L=0.33*L+0.33*C+0.33*RL+0.134*RR
Copy: R=0.33*R+0.33*C+0.33*RR+0.134*RL
Copy: SUB=0.35*R+0.35*L+0.17*SUB

I am not using any high-pass on mains, they go full range from 55Hz, subwoofer is low-passing 80Hz/24db through SVS app.
Aren't bass traps for really low frequencies? Will full-range bass traps treat the room modes that I deal with? and I guess I would have to cover the corner from floor to ceiling.


Room modes:
140.97E1
241.22E1
358.26A#1
466.42C2
578.29D#2
 
I think something like the GIK Soffit Traps might take care of some of that... You would have to contact GIK to get a better idea given what they think... They give free advise...

I'm not an expert on EQApo or PeaceParametricEQ... I just think it may be that the GEQ you are cutting with may be adding two wide of a cut that may be stepping on the PEQ... Even 32 Graphic Sliders does not give you much resolution down low where you are needing it...
Something else to think about would be Convolution... Doesn't APO or Peace do Convolution? And there are couple of methods that AVNirvana members have cooked up using REW to create FIR filters... :dizzy:

I was having trouble understanding your mixing commands, noticed that you did not end up with 100%, so I ran it through an AI to explain what you are doing... The AI thinks your down mixing commands may lead to clipping and/or distortion...

And since this is all being mixed down to 2.1 you might try moving the low pass on you sub down to around where your mains start to roll off...
 
Last edited:
Below the room transition frequency or Schroeder frequency, typically between around 100 and 250 Hz, the room acts as a resonator, and above the Schroeder frequency, as a reflector and diffusor of sound...

This master class in Digital Room Correction is a worthwhile watch...
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
 
Getting back to your mixdown commands... After some more reading and asking an AI questions... I think things should add up to 100%...
for instance "Copy: L=0.33*L+0.33*C+0.33*RL+0.134*RR" the 0.33+0.33+0.33+0.134= 1.124... I think you really want this to add up to 1.00... And the extra 0.124 might cause clipping and/or distortion... This for the left and right channels... As for the Bass command:
"Copy: SUB=0.35*R+0.35*L+0.17*SUB"... My take is that you are not using enough of the original 5.1 bass signal and adding too much left and right to that... I would have thought something more like 0.25 left and 0.25 right with 0.5 original bass... I could be way off in my take on all of this...

P.S As an analyst this is intriguing...
 
Last edited:
Below the room transition frequency or Schroeder frequency, typically between around 100 and 250 Hz, the room acts as a resonator, and above the Schroeder frequency, as a reflector and diffusor of sound...

This master class in Digital Room Correction is a worthwhile watch...
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
Thanks for the video, that looks great, I will check it out Friday :)
Getting back to your mixdown commands... After some more reading and asking an AI questions... I think things should add up to 100%...
for instance "Copy: L=0.33*L+0.33*C+0.33*RL+0.134*RR" the 0.33+0.33+0.33+0.134= 1.124... I think you really want this to add up to 1.00... And the extra 0.124 might cause clipping and/or distortion... This for the left and right channels... As for the Bass command:
"Copy: SUB=0.35*R+0.35*L+0.17*SUB"... My take is that you are not using enough of the original 5.1 bass signal and adding too much left and right to that... I would have thought something more like 0.25 left and 0.25 right with 0.5 original bass... I could be way off in my take on all of this...

P.S As an analyst this is intriguing...
It should add up to 1.0, but I am using digital negative pre-amplification on all channels, so there is literally zero chance of clipping, but yes that was my mistake, never meant for it to exceed 1.0

Since I am using my 5.1 sound card with only subwoofer and two mains (true 2.1 system with separate RCAs for Left, Right, Subwoofer and complete control over each channel (I also tried measuring and EQing Left and Right separately, but it didn't produce very good results)), I need to do downmixing. So naturally, when playing stereo, if no bass management is active, it only plays sound through main speakers, because of course, it's 2.0 audio track, makes sense. But if I used only SUB=0.35*R+0.35*L, this is an absolute command, meaning the ONLY L and R channels get sent to the subwoofer, I have to specify that I also want .1 channel in the subwoofer. I tested some movies with lots of sound effects in LFE, and discovered that LFE is much louder, so with command you suggested "0.25 left and 0.25 right with 0.5 original bass" this would mean extremely loud LFE audio track and extremely weak bass in stereo, so I adjusted subwoofer so that is sounds like I want in stereo, and then started slowly adding up LFE into the mix until I was satisfied with the loudness of the audio effects.
 
Hi, back with an update.

I could find only one good spot where mains sound good without insane nulls in 200-300Hz region, so I had to plan sub placement according to that. But there was a problem: sub on right side - big 60-100Hz nulls, sub on the left side: big 20-50Hz nulls. That was a sign that I will most likely need dual subwoofers, and I was right. I decided to use my old Genius 8 inch subwoofer and place it diagonally to SVS, my current setup:
roomeqwizard_kOoF1AXqgg.png

and went from this:
svs only.jpg


to this:
genius+svs.jpg


combined with mains:
genius+svs+mains.jpg


I was blown away by the difference. Unfortunately, after EQ, when I raise volume of the subs by the same amount, it doesn't sound right, probably because the Genius sub brings overall quality down. So I decided to approach it differently.
I only need the Genius sub to cover 60-120ish Hz region, 58Hz and lower sounds great on the SVS alone. So I hard limited the Genius sub to play only 60Hz and higher to avoid it destroying lower bass produced by SVS. Result is the measurement you see above. After EQ, I first set my SVS to desired volume, and then adjusted Genius volume fill in the missing frequecy range. I am also not sending LFE signal to Genius, only SVS will receive it.

And it actually sounds great, definitely better than with SVS only after some testing. I will absolutely get second SVS afer I save up some money. The Genius on it's own sounds good, I was happy with it but it could do only 29Hz, I guess playing together, it clearly shows how much better the SVS is. As of right now, I am quite happy with the sound, can't wait for second SVS :) I also included my measurements.

Update: Did some more listening and tuning, one nice effect I noticed is it made bass, idk, "quicker"? Like some bass-heavy electronic music, with the dual sub, the beat seems to arrive and dissipate way quicker than with a single sub, I mainly noticed when playing some classical music with lots of bass, especially cellos sound amazing now.
 

Attachments

Last edited:
Nice bit of work @Lid9497 :T You might take another pass at pushing some of those peaks down by a few db with EQ... 40Hz, 77Hz and 140Hzish stand out... :justdontknow:
 
Nice bit of work @Lid9497 :T You might take another pass at pushing some of those peaks down by a few db with EQ... 40Hz, 77Hz and 140Hzish stand out... :justdontknow:
Forgot to mention this is all before EQ, here is EQd response.
EQ genius+svs+mains.jpg
 
Now that is pretty impressive... And you've come a long way in just 2 months...:bigsmile: Good job @Lid9497 :T

Are you going to leave it flat or are you going to add a house curve? B&K or Harmon style to taste...
 
Last edited:
Now that is pretty impressive... And you've come a long way in just 2 months...:bigsmile: Good job @Lid9497 :T

Are you going to leave it flat or are you going to add a house curve? B&K or Harmon style to taste...
I tried several house curves, but they never add enough bass, or I am just using it wrong. Anyways, I just increase subwoofer volume through app until it sounds strong enough. I measure with SVS at -32db and after EQ I usually set it to -11db, I could probably use like +20db harmann, but I figured it's best to increase volume on the subwoofer itself to avoid digital clipping.
 
REW will let you bake in a house curve into the EQ process... So you could try a harmon curve that boosts about +5 to +7 db on the low side and taper off across the entire frequency range down to a -3 to -5 db at 20kHz... And then make your final gain adjustments to your SVS like you are currently doing... I think there is a low shelf and high shelf that could also be used in the process...
 
REW will let you bake in a house curve into the EQ process... So you could try a harmon curve that boosts about +5 to +7 db on the low side and taper off across the entire frequency range down to a -3 to -5 db at 20kHz... And then make your final gain adjustments to your SVS like you are currently doing... I think there is a low shelf and high shelf that could also be used in the process...
I was thinking about most effective/efficient way to apply PEQ. As of now, I am applying PEQ to all channels because I am measuring all speakers together. But what about this: I measure, equalize and apply PEQ to each channel separately, making them as flat as possible on their own, and than do last measurement with PEQ on all channels in place as a single system, and apply that PEQ to all channels. What is your way/methodology of applying EQ?
 
It is better to measure each speaker individually... You can sum them if need be using REWs maths... The quick answer to my method is, I create FIR filters using REW and rePhase for my left, right, and sub, leaving my center, surround sides, soruound backs alone... I am mostly focused on stereo with a sub... My AVR runs my center and surrounds which I EQ to taste in the AVR for a total of 7.1...

When in stereo mode I add to my B&H house curve convolution with FIR fillers an additional layer of old skool EQ to taste... You can see my kit in my more screen...
 
Last edited:
Back
Top