Using four bass guitar drivers to build a dual opposed subwoofer project

[312elements]

A little bit more progress today:

View attachment 393173

View attachment 393174

Both the vibrodamping and the felt treatment are done

And a little knock-knock for fun:

New video by Peter Pataki

photos.app.goo.gl

I have built quite many speakers already but this sounds pretty remarkable (no bracing whatsoever yet!)

are you doing melamine foam again too?

 

[ppataki]

are you doing melamine foam again too?

Yes I will

 

[ppataki]

Just wanted to share this idea for bracing - I think it works really well (as far as the knock-knock test goes) but any comments are more than welcome

I think it is called a wire-strainer in English but apparently it has other functions too

 

[Old_School_Brad]

I love your ingenuity. -Following this one.

 

[ppataki]

A few updates in the meantime

The paint job is done
I have used a sponge brush hence the kind of brushed look (which is very divisive, half of my family does not really fancy it...)

I have mounted the front speakers:

And placed the cabinet to its planned position (remember that this is actually a table for my wife! it has 4 subs as a bonus only...)

The high-gloss beauty fronts will be added later, next week or the week after

By next week I shall finish cabling and mounting everything else then I will test the cabinet for airtightness and then measurements will follow
Have a nice weekend everybody!

 

[ppataki]

And the melamine foam:

 

[ppataki]

The cabinet is now fully done, technically speaking
It is also airtight, just tested it
I am really happy so far, next week I will publish some measurements

 

[ppataki]

And now let's see some measurements!

Note: these measurements were NOT taken in the final position of the speaker (in the living room next to the armchair) but in my home office upstairs where I have this sub already set up
I have just replaced that sub with the new one and took measurements in the main listening position (after res-measuring and re-setting the delay and gain values) + I also did a round of optimization with Dirac Live 3.10.3
Measurements in the final position downstairs with my main sub in the front integrated will follow probably next month only since I need to buy an amp first to drive this new sub there

Frequency response at MLP (approx. 75cm from the fronts and approx. 3 meters from the sub in the back) using 1/12 smoothing
Delay, gain, polarity set and crossover in place (linear phase, 24dB/oct. at 80Hz)

Blue is without Dirac Live, red is with Dirac Live - note the room mode related dips (not really audible in my opinion)

The same using Psy smoothing:

Distortion:

I would say even with Dirac this is pretty good

GD:

That peak between 80 and 90Hz is there with only the sub playing too so it is room related and not crossover related

Wavelet with Dirac:

I consider this to be fine for a system with a sub

Step response with Dirac:
Note the minor pre-echo due to the linear phase crossover - nothing to be worried about though

Subjectively speaking: it sounds wonderful!
Really, nothing to complain about; extreme fast and precise lows, as one should expect from a sealed cabinet
The difference I hear vs. the other sub I have here is that with this one the bass is less 'meaty' - it sounds less 'full' but it sounds 'faster' and more 'precise' - I am not really sure why but that's what I hear. I love both but this is closer to my taste when it comes to a sub.

Next step is that once I get the amp (Sabaj A30A) I will bring the speaker downstairs to its final location, perform the integration with the main sub and publish the measurements here
And of course the beauty fronts are yet to be done too

Final note: I was really delighted to use my new UMIK-2 microphone here! Making delay measurements was a walk in the park - I was done in literally a few minutes (performed 3 measurements and got the same delay value up to 2 millisec. decimals)

 

[tmuikku]

Hi, did you add some fabric over the sponges? I mean, the melamine foam is highly abrasive, so if there is even a remote chanse some of it ends up in the voice coil gap there is potential for disaster. This might be unnecessary concern, but a concern nevertheless so wanted to express

 

[ppataki]

Hi, did you add some fabric over the sponges? I mean, the melamine foam is highly abrasive, so if there is even a remote chanse some of it ends up in the voice coil gap there is potential for disaster. This might be unnecessary concern, but a concern nevertheless so wanted to express

Thanks for raising this - actually I did not consider this risk to be high on the list. The sponges can't move in the cabinet since they are strained.

 

[ppataki]

It took a bit longer than expected but final design is now there:

And the final location:

Nexth month I will get the amp (Sabaj A30A) and will integrate the new sub into the existing setup (there is already a sub in the front; this is in the back, approx. 3 meters from the MLP)

I will of course post all the measurements

 

[ppataki]

Happy to report that there is a spinoff to this project!
One of my friends will more or less replicate this idea, please see it here:

DIY 2.2 living room project

One of my friends came over a few weeks ago to listen to my new DOS sub (that was placed in my home office system for testing purposes before being moved to my living room) He loved it so much that he decided to build the same for his living room :) Before venturing into this project we used...

www.audiosciencereview.com

 

[ivanj]

Just wanted to share this idea for bracing - I think it works really well (as far as the knock-knock test goes) but any comments are more than welcome

View attachment 393314

I think it is called a wire-strainer in English but apparently it has other functions too

It might be called a 'turn-buckle' if it is what I think it is> How do you have it attached to the cabinet, and what is the function? I take it to add stiffness?

 

[ppataki]

It might be called a 'turn-buckle' if it is what I think it is> How do you have it attached to the cabinet, and what is the function? I take it to add stiffness?

It actually 'attaches' itself
As you keep turning it, it becomes longer so it strains itself into the cabinet, providing additional bracing/stiffness
The knock test shows the result really well

 

[ppataki]

Finally I had some time to play around with integrating this sub with my front sub and with my main speakers (not the spheres but a pair of heavily modified Adam Audio T5V because those are my current experimental project)

I did volume matching then I measured all the delays and then I just applied a 24dB/octave linear phase symmetrical crossover at 80Hz

It did kind of a nice job, this is the measurement at the MLP with zero additional DSP correction:

Frequency response + Phase

GD

Step

Wavelet

In my opinion these graphs look pretty OK

Then I did a 9-point Dirac Live optimization (with some additional manual correction)

Actually I was quite happy with the looks of it!
(you can see all the other graphs in the attached mdat file for your scrutiny)

But then when listening to it it sounded muffled and 'out of place'....
I have managed to get a pretty good sound at the end of the day by applying different low shelf filters for each sub individually (different Q, freq and gain)

But I was not quite happy with it still.....

I have decided to take a plunge into playing around with minimum phase crossover settings
I have tried 18, 24, 36 and 48dB/octave settings. I saw that the 36dB symmetrical setting looked the best (actually nicer FR and phase vs. with linear phase but much uglier Step response and Wavelet)
Then I started combining slopes and found that combining a 24dB low-pass with a 36dB high-pass yielded the absolute best results!

Frequency response vs linear phase 24dB:

Phase

GD

Here the LP version looks nicer (naturally)

Wavelet also looks a bit worse with MP (again, naturally)

After applying Dirac and a few manual corrections this is what I got:

A little bit elevated GD but a lot of those huge peaks are gone

And the peak energy delay is kept under 20ms

Overall I am really happy now! It sounds more natural and you literally cannot tell that there is a sub (let alone two) in the room
Please see the mdat file here

 

[ppataki]

Just a quick update:
I have found a better spot for the subs in the back - I was performing measurements at multiple positions and found out that this one yields the best response (there was not much of a difference in terms of frequency response, but huge differences in terms of the temporal graphs)

I also have replaced my full range drivers with these coaxial drivers
For the time being in temporary test cabinets but next months I will move them to the sphere cabinets and will slightly redesign the stands too