EQ Limit for Cutting - Subwoofer FR

[CK.]

Hello guys,

So I am trying to manually EQ my sub using REW, and although the aspect of boosting has been extensively discussed I have a question about how much to cut using EQ PEQ filters!

Attached is the FR of my 10" sealed sub with the gain knob at 45% and the sub volume in my avr at -4 (limits are -12/+12).

In order to get decent extension down to 20Hz I thought of setting my EQ limit in REW to about 75dB.
This results in EQ gain of -20 dB or more! (-25 at the modal frequency of 48Hz)

Is that ok taking into account that my sub gain knob is at 45%?

Many thanks!!

 

[staticV3]

From the graph it looks like a ~14dB Peak.

Yes, it's totally fine to cut by this much. You're just compensating for unwanted room gain, basically.

Though to be sure, I would do a Moving Microphone Method frequency response measurement around the main listening position, as sweeps with static microphone position tend to exaggerate room modes a bit.

 

[CK.]

Thank you @staticV3 for your feedback!

One more thing if I may, suppose you are after a 6db harman curve, would you evaluate this using both L plus R playing or with seperate measurements.

The spl difference of the sub between both channels vs seperate channels is obviously different after the cross over point so how would you approach this?

Thanks in advance

 

[staticV3]

The spl difference of the sub between both channels vs seperate channels is obviously different after the cross over point so how would you approach this?

Hmm not sure I understand.

Regardless of whether you feed just one channel or both, the Mains<->Sub balance should stay the same.

 

[CK.]

The subwoofer is a single channel but the l/r when playing together would be summed when in phase at least below the schroeder. So the responce with sub+R would be different than sub+R+L right? The slipe of the house curve would be different at thr bass region above xo point right? Am I making sense?

 

[OCA]

Do remember you will measure LFE channel with a 10dB boost relative to other channels under normal circumstances in REW. You should either drop LFE volume by 10dB or use REW's relative measurement parameter to see actual sub/satellite levels.

 

[sigbergaudio]

The subwoofer is a single channel but the l/r when playing together would be summed when in phase at least below the schroeder. So the responce with sub+R would be different than sub+R+L right? The slipe of the house curve would be different at thr bass region above xo point right? Am I making sense?

The house curve target would be with both L+R playing.

 

[staticV3]

The subwoofer is a single channel but the l/r when playing together would be summed when in phase at least below the schroeder. So the responce with sub+R would be different than sub+R+L right? The slipe of the house curve would be different at thr bass region above xo point right? Am I making sense?

The subwoofer should be configured so that it plays 6dB louder when bass is present in both channels vs just in one, therefore preserving the Sub<->Mains balance no matter if you measure one channel or both.

 

[CK.]

Do remember you will measure LFE channel with a 10dB boost relative to other channels under normal circumstances in REW. You should either drop LFE volume by 10dB or use REW's relative measurement parameter to see actual sub/satellite levels.

Hm but I measuring using rew sweeps for front channels but having channels disconnected in order to get the sub only (using a high avr cross over). Isnt that ok?

 

[OCA]

Hm but I measuring using rew sweeps for front channels but having channels disconnected in order to get the sub only (using a high avr cross over). Isnt that ok?

That should be fine in terms of sub volume but you can directly measure LFE from REW with LFE -10dB parameter (this is defo there in REW Beta, not sure if it's there in the current non Beta version) if your sub is connected to the AVR's sub output. If you change the AVR bass mode to LFE+Main and set bass extraction lpf to 250Hz, you will get your sub's actual full response.

 

[CK.]

The subwoofer should be configured so that it plays 6dB louder when bass is present in both channels vs just in one, therefore preserving the Sub<->Mains balance no matter if you measure one channel or both.

Uhm yeah good point I got confused thinking that when playing both channels only affects fs above cross over point…..

My last question would be….each channel has a different responce below say 80hz and even if you eq the peaks for each channel seperately and apply the cross over slope at 80hz, each channel will still affect the 20-80hz responce differenlty mainly due to dips. Is that a valid concern? Should you manually introduce dips in the reponce of the L and R channels in order to match exactly below the cross over?

 

[CK.]

That should be fine in terms of sub volume but you can directly measure LFE from REW with LFE -10dB parameter (this is defo there in REW Beta, not sure if it's there in the current non Beta version) if your sub is connected to the AVR's sub output. If you change the AVR bass mode to LFE+Main and set bass extraction lpf to 250Hz, you will get your sub's actual full response.

Hm but the lfe-10db responce vs the redirected bass responce from the sub with a high cross over should be the same right? Is there a benefit for measuring specifically the lfe channel?

 

[sigbergaudio]

Uhm yeah good point I got confused thinking that when playing both channels only affects fs above cross over point…..

My last question would be….each channel has a different responce below say 80hz and even if you eq the peaks for each channel seperately and apply the cross over slope at 80hz, each channel will still affect the 20-80hz responce differenlty mainly due to dips. Is that a valid concern? Should you manually introduce dips in the reponce of the L and R channels in order to match exactly below the cross over?

If you introduce dips to match the response, the dips will in total be even worse, so that's not a good strategy. Left and Right response will not be perfectly similar. Narrow dips are often not that easy to hear either, so try not to worry about them too much. Measuring your room is a slippery slope to OCD.

 

[OCA]

Uhm yeah good point I got confused thinking that when playing both channels only affects fs above cross over point…..

My last question would be….each channel has a different responce below say 80hz and even if you eq the peaks for each channel seperately and apply the cross over slope at 80hz, each channel will still affect the 20-80hz responce differenlty mainly due to dips. Is that a valid concern? Should you manually introduce dips in the reponce of the L and R channels in order to match exactly below the cross over?

What AVR are you using?

 

[CK.]

If you introduce dips to match the response, the dips will in total be even worse, so that's not a good strategy. Left and Right response will not be perfectly similar. Narrow dips are often not that easy to hear either, so try not to worry about them too much. Measuring your room is a slippery slope to OCD.

Yeah I get it I always keep psychoacoustic smoothing handy i order to preserve my sanity However, what I dont get is that below the crossover frequency you would want to only have the sub playing i order to have a consistent reposnce. With the 12 db slope of the crossover I find that the main channels affect that sub responce….
Isnt more reasonable to apply a non linear slope to really reduce the influence of the other channels to the sub??

 

[CK.]

What AVR are you using?

Denon 3700h

What AVR are you using?

denon 3700h

 

[staticV3]

My last question would be….each channel has a different responce below say 80hz

If you high-pass the mains and let the sub handle bass response, then the response below 80Hz should be identical between each channel.

 

[sigbergaudio]

Yeah I get it I always keep psychoacoustic smoothing handy i order to preserve my sanity However, what I dont get is that below the crossover frequency you would want to only have the sub playing i order to have a consistent reposnce. With the 12 db slope of the crossover I find that the main channels affect that sub responce….
Isnt more reasonable to apply a non linear slope to really reduce the influence of the other channels to the sub??

The crossover will never be a brick wall, so you will get some response from the speakers as you seem to have noticed already. What you can do is choose a higher crossover, for instance 100 or even 120hz, to get less energy from the speakers in the lower frequencies.

 

[OCA]

Denon 3700h

denon 3700h

Then just run a standard Audyssey calibration with the Editor app and use the A1 Evo script. Your sub and mains (and all other speakers) will be optimally aligned and crossed over.

 

[CK.]

If you high-pass the mains and let the sub handle bass response, then the response below 80Hz should be identical between each channel.

it should but its not because HPF is still allowing the front channels to contribute to some degree. I will give it another try though.

My approach is to measure all channels and subs full range without any cross-overs (or very high) and EQ them "flat" (below schroeder) using cuts only for peaks mainly. Then use REW alignment tool to find the sub delay for an 80Hz cross-over.
Then increase the volume of the sub +6db and maybe apply a harman like boost in the bass of each speaker below 200Hz.

Does this sound like a correct approach?