Help with subwoofers cut off

[Davide]

I forgot to specify that the crossover is set to 118hz.
Doing some tests, the frequency was found to have the best consistency in the critical area between 125 and 300Hz.
Also post the relevant spectrogram.
Unfortunately the decay in that area is faster than the rest, and in fact it is precisely where with my ear I feel the greatest deficiency. The male vocals and instrumental bass lack a bit of body.
Do you think I could do anything to improve the situation?

 

[ppataki]

I forgot to specify that the crossover is set to 118hz.
Doing some tests, the frequency was found to have the best consistency in the critical area between 125 and 300Hz.
Also post the relevant spectrogram.
Unfortunately the decay in that area is faster than the rest, and in fact it is precisely where with my ear I feel the greatest deficiency. The male vocals and instrumental bass lack a bit of body.
Do you think I could do anything to improve the situation?View attachment 317286

Have you tried a low shelf filter or a simple boosting bell filter?

 

[sigbergaudio]

I forgot to specify that the crossover is set to 118hz.
Doing some tests, the frequency was found to have the best consistency in the critical area between 125 and 300Hz.
Also post the relevant spectrogram.
Unfortunately the decay in that area is faster than the rest, and in fact it is precisely where with my ear I feel the greatest deficiency. The male vocals and instrumental bass lack a bit of body.
Do you think I could do anything to improve the situation?

Could be interesting to see the response with maybe 1/12th smoothing.

You could try experimenting with speaker placement to see if a different placement will improve the 125-300hz situation without messing up something else.
You could also try a shelf filter on the speakers as suggested above, perhaps +2dB starting at 300hz and see what happens to the curve (and how it sounds).

 

[ozzy9832001]

Could be a lot of things. I would first suspect that as the xover slop goes down the interference from the sub is cancelling out with the mains until about 350hz or so. I don't know where your MLP is in relation to the speakers or any boundaries, but the 150hz-400hz are common areas for SBIR if you are nearfield listening and speakers are close to the front wall (or sidewalls for that matter). Simple solution would be to either boost that region in question or shelf everything else down a few dB to bring everything down to its level.

 

[Davide]

This is the latest attempt:

Aside from it looks good, it sounds very very good now!

The strange thing is to obtain this result I had to manipulate Dirac's target a bit. In fact, a straight line does not correspond to a straight measurement...

However, do you say that group delay at 45hz is audible?

However, it's a shame that Dirac doesn't allow you to choose the length of the filter in order to choose between latency and low GD. Audiolense and Acourate are much more versatile in this.

 

[ozzy9832001]

This is the latest attempt:

View attachment 317895

However, do you say that group delay at 45hz is audible?

Across the entire measurement you have some time domain issues. The room is probably fairly lively. It's especially prevalent from 600hz and below.

I don't know how accurate REW is with the time domain below say 100-150hz. The room is going to contribute to a lot of the imbalance that you see.

As far as audible, it depends...do you listen to a lot of music that has those frequencies represented? How about movies?

I'd say you're at the point where you can start thinking about acoustic treatment if you really want to take it any further.

 

[Davide]

I mainly listen to music with a lot of bass, while I almost never watch films.

I'm currently renting so it doesn't make sense for me to spend on acoustic treatment.

However, I am anxiously awaiting the release of Dirac ART. I assume it can do something about this.

 

[sigbergaudio]

I think this looks good frequency wise, and not terrible in the time domain either for a normal domestic space.

 

[ozzy9832001]

I mainly listen to music with a lot of bass, while I almost never watch films.

I'm currently renting so it doesn't make sense for me to spend on acoustic treatment.

However, I am anxiously awaiting the release of Dirac ART. I assume it can do something about this.

ART uses other speakers to create cancellations (or that's how I interpreted it).

I listen to mostly music on my system as well -- more metal genres than anything else and while they certainly have their fair share of bass, it rarely, if ever goes below 50-55hz.

The only issue with the 45hz cause possible be created something down the line at it's harmonics, but I don't see anything to support that in any of your measurements.

 

[sigbergaudio]

It would be interesting to see a measurement with Dirac on/off in the same graph to see what has been done.

 

[Davide]

It would be interesting to see a measurement with Dirac on/off in the same graph to see what has been done.

Sure. As soon as I have time I will post all the comparison measurements.

 

[Davide]

Sure. As soon as I have time I will post all the comparison measurements.

EDIT. Note that measurement with Dirac ON/OFF will not be significant below 100hz, due to the Bass Control not redirecting the signal to the subs when OFF (subs remain silent).

 

[Davide]

Here we are with the measurements.

First the various frequency responses with and without Dirac, and also with and without grids (I also included the subs alone).

Then we have the spectrograms of the speakers and subs again without Dirac.
NOTE: GD at 45hz is half than with Dirac ON.

Finally another interesting measurement of Dirac filters (different from those posted previously).
NOTE: for some strange reason, even though the crossover is now set to 70hz, it seems to occur at 45hz, precisely where there is greater GD.....

 

[Davide]

ART uses other speakers to create cancellations (or that's how I interpreted it).

Post 1
Post 2

Here the effectiveness of ART in linearizing the GD at low frequencies, compared to BC, was shown.

I therefore have high expectations regarding ART.

 

[Palmspar]

I use the same drivers in a 60 liter dual opposed subwoofer with a lot of wool, 1 driver sees 35 liter or so.
And a Hypex FA501 plate amp with dsp, the sub sounds verry good tuned @22hz+-3db in Hypex filter design.
The Scan Speaks are great drivers!

 

[Davide]

I use the same drivers in a 60 liter dual opposed subwoofer with a lot of wool, 1 driver sees 35 liter or so.
And a Hypex FA501 plate amp with dsp, the sub sounds verry good tuned @22hz+-3db in Hypex filter design.
The Scan Speaks are great drivers!

I also initially thought of a dual opposed configuration, but then it seemed to me that I could achieve greater consistency in the room with two different subs controlled by Dirac (also because I already had an NC502MP with 2 channels).
In your case, 35 liters per driver is very little, almost half of the recommended. But if your frequency response is good I don't see any problems. Maybe just low efficiency.
Anyway, I think these ScanSpeak drivers are very good for the price (I paid €240 each). Furthermore, in terms of TS parameters I hadn't found anything better for this project.

However, the other day I put my ear right in front of the driver and I noticed that one channel had a very slight noise, at a rather high frequency, like 1khz. Sounds like SMPS noise (coil whine). The driver responds well up to almost 2khz and I have no analog low pass filter, so maybe some upstream noise is captured, or maybe it is the amp noise.
Nothing abnormal, at 40 cm I can't feel it already...

 

[Palmspar]

Green line in REW is the Scan Speak and red the SVS SB2000 subwoofer with the mic +- 2cm away from the cone.
When i put my small box in speakerboxlite its not bad for use in a smaller room, and when the box is way bigger the driver exeeds xmax already with 150 watt or so.
You have plenty of watts, my amp has 400 watt @ 2 ohm load for the 2 drivers.
For a 0.707 Qtc box the driver needs 39.1 liter, but its a design goal small boxes need big amps to play loud.