Another 8" dual opposed subwoofer project

[ppataki]

After finishing my new DIY system in my home office/bedroom I have decided to also add (build) a subwoofer to it

I have really huge dips in the frequency response due to room modes; see the before/after DSP correction curve below:

When running REW simulations for adding a sub to the system, the resulting curve looks a bit better (highlighted):

So driven by curiosity I have decided to add a 8" dual opposed subwoofer (DOS) to the system

I will be using two Lavoce SSF082.00L 8-ohm drivers for the project (wired in parallel to get 4-ohm)

The system will look like this:
- current mains (Pluvia 11-based spheres) driven by a Topping PA7 Plus amp and an SMSL D-6 DAC
- I will add an RME Digiface USB digital-digital converter in the signal chain and connect the above devices to it
- Plus I will connect another D-6 DAC and a Fosi V3 Mono amp to the Digiface too (to drive the DOS)

Like this I will have fully discrete channels so I could apply DSP with maximum flexibility

I will post the DOS cabinet details later today, stay tuned!

 

[staticV3]

I have really huge dips in the frequency response due to room modes; see the before/after DSP correction curve below:

Are you aware that flat in-room response is a flawed target?
https://youtu.be/_tnWB8Rl0Ms

Edit:
That is, for a typical far field system. Your DIY system OTOH appears to be near field, my apologies.

 

[ppataki]

Are you aware that flat in-room response is a flawed target?
https://youtu.be/_tnWB8Rl0Ms

Edit:
That is, for a typical far field system. Your DIY system OTOH appears to be near field, my apologies.

No worries @staticV3
By the way this is not my target anyway, if you read the post that I linked above you will see that what I do is that I correct to flat and then I add totally custom shelf filters to 'voice' the lows and the highs to taste
This is what I do with all my systems and it works really well

 

[ppataki]

Let's see the cabinet details now

Size will be 300x260x350 mm
Cabinet material is not yet decided but it will be either MDF or plywood or bamboo
The cabinet will be painted matte white and once done, I will add a 'beauty' front panel on both ends using Ikea magic

The speaker drivers shall arrive in 5-6 weeks' time - I might start the cabinet build sooner though

I will post updates as I progress

 

[ppataki]

Some simulation curves for those that are fond of these

Approx. 250W will be the amplifier-limited maximum power output at 4 ohm
At 24Hz at 1 meter that will yield an SPL of 95dB (+ room gain)

Here the cone excursion will be at about 65% of the xmax of the driver so I would not expect much distortion

Group delay:

To me these all look really fine!

 

[ppataki]

The cabinet is in progress:

Finally we used pine wood panel with a thickness of 18mm
I know that this might not be the most optimal material but there was a bigger panel just lying around in the workshop, so.....

Next steps will follow as soon as the drivers arrive - best case scenario is mid-July, worst case scenario is end of August but I keep my fingers crossed

 

[ppataki]

The drivers will arrive either next Friday or the following Monday.
I will post an update probably in two weeks' time
Stay tuned!

 

[ppataki]

Today the drivers have finally arrived!

Both of them work fine, just tested them

Next week I will finish the cabinet (probably first without the 'beauty front panels'; those will come a bit later)
Will post pictures and measurements too!

 

[ppataki]

The speaker is 90% done, see the photos about the progress below:

I have done some tests to see if the cabinet is airtight and to hear the motor noise - happy to confirm that it is all pretty fine!

Next steps:
- the Fosi V3 Mono driving the sub shall arrive today
- I will integrate the sub, probably crossed at 80Hz with 24dB/octave slope (after proper volume matching and delay setting)
- I will post measurements (will be using Dirac Live 3 for DRC)
- the piano black beauty finish for the two front sides are yet to be done but I have the panel already (from Ikea, of course )

 

[ppataki]

I just had some time now to perform some measurements and finish the integration of the sub
All the below measurements were done at MLP (about 80 cm from the speakers) without Dirac


Red is the L+R without the sub; blue is the sub - totally separately, just to show the overlay
As you can see using a sub successfully eliminated that huge dip around 54Hz - smiley face!

Now let's see the FR after setting the correct delay and applying a min-phase crossover at 80Hz, 24dB/oct.

Looks great! (relatively speaking, compared to the totally untreated room)

Distortion

To me it looks good; I would be curious to see this after Dirac.....

Now let's see an interesting exercise that I always do when adding a sub to a system: let's see what happens if I change the crossover from min-phase (MP) to linear phase (LP)

This is how the FR is impacted: blue is MP, red is LP

LP seems to be a few dB shy...

But let's see the other curves

Phase:

450° less rotation at the low-end!

Spectrogram - you see an animation starting with MP then switching to LP

With MP the energy delay in the low end is about 5-6ms which is really great already but with LP it goes down to ZERO!

And pre-echo is negligible (totally inaudible)

So far so good! I am totally satisfied
Next step is to apply Dirac - then I will be able to provide some subjective comments too
Stay tuned!

 

[ppataki]

Let's see the measurements now with Dirac Live 3.10.3
I have made only one measurement, at the MLP - later on we will see what happens when taking multiple measurements

FR and phase look really great:

GD looks fine until 30Hz but that's the rolloff so I guess it is fine

Distortion looks fine, except for the peak at around 60Hz but that is due to a dip in the FR curve

Spectrogram looks fantastic:

Overall I am very happy, I have reached my goal to reduce distortion and even out the FR curve as much as possible with this room/placement combination


Now here comes the catch: I have made a 9-point measurement with Dirac too and verified it in the MLP

The FR looks really horrible (though phase looks fine)

Distortion is somewhat lower on the low end but higher where the FR dips were not corrected

GD is somewhat reduced too:

What I find inexplicable is that the 9-point measurement sounds way better than the 1-point measurement despite the fact that the FR looks really horrible...
I would love to hear some ideas around that; I am a bit shocked to be honest (and on the other hand very happy since the system really sounds great!)

MDAT file attached for reference

 

[ozzy9832001]

Let's see the measurements now with Dirac Live 3.10.3
I have made only one measurement, at the MLP - later on we will see what happens when taking multiple measurements

FR and phase look really great:
View attachment 382503


GD looks fine until 30Hz but that's the rolloff so I guess it is fine

View attachment 382504


Distortion looks fine, except for the peak at around 60Hz but that is due to a dip in the FR curve

View attachment 382505

Spectrogram looks fantastic:

View attachment 382506


Overall I am very happy, I have reached my goal to reduce distortion and even out the FR curve as much as possible with this room/placement combination


Now here comes the catch: I have made a 9-point measurement with Dirac too and verified it in the MLP

The FR looks really horrible (though phase looks fine)

View attachment 382508


Distortion is somewhat lower on the low end but higher where the FR dips were not corrected

View attachment 382511


GD is somewhat reduced too:




What I find inexplicable is that the 9-point measurement sounds way better than the 1-point measurement despite the fact that the FR looks really horrible...
I would love to hear some ideas around that; I am a bit shocked to be honest (and on the other hand very happy since the system really sounds great!)

MDAT file attached for reference

9 points has more information to draw a better conclusion from.

Single point measurement doesn't account for the width of a persons head. 2 measurements would probably be better more in line with a persons ears. Maybe space them out 3" or so, front to back or however you move your head typically.

Or use MMM technique, but only good if you are staying within the area your head will go; otherwise data could skew the results.

When building a final eq, I'd recommend taking several measurements per speaker over an area and then avg them. Helps to uncover strange anomalies or reflections. Once that's done, then I'll build an eq filter. I try to do them manually first.

 

[ppataki]

Forgot to include the final picture of how the system looks like:

Note: it is not the very final state since the high gloss piano black beauty fronts are still to be done and I will do something about cable management too

 

[ppataki]

And with this, I consider this project fully finished!

I am totally satisfied - the project has managed to reach its goals, I am more than impressed with the sound
If I take into account that these drivers cost 85 EUR each.....an incredible price/value ratio that we see here indeed

The final chain looks like this:
Source is a PC --> connected to an RME Digiface USB --> connected to two SMSL D-6 DACs via optical --> one of them connected to a Topping PA7 Plus via XLR driving the mains (the Pluvia 11-based spheres), the other is connected to a Fosi V3 Mono via XLR driving this sub

Using Jriver as player and DSP: crossover is thEQorange VST plugin (linear phase), DRC is Dirac Live 3.10.3 (sub is integrated so Dirac 'sees' the system as stereo)
Final 'room curve' is created by using custom low and high shelf filters (Dirac is set to fully flat)

 

[MCH]

connected to a Fosi V3 Mono via XLR driving this sub

It looks like a worthy project!
Question for you. There is a bit of controversy about the Fosi monos inverting phase on the XLR input or not (seems to be confirmed on the RCA). I can imagine you have seen if it does or not. Any comments?

 

[ppataki]

It looks like a worthy project!

Thank you!

I can imagine you have seen if it does or not. Any comments?

To be honest I don't know - I have not checked it; when I did the crossover setup I have inverted the sub since that way the sum of the signal was way better

 

[ppataki]

I thought I would share my latest adventure with this sub project:

I was experimenting with various placements in the room and I managed to find a spot (at the opposite end of the room) where the response is truly remarkable!

See the measurements (with Dirac, using just 1 point)

There are dips indeed but between 17Hz (the new -3dB point) and 100Hz the response became nearly flat! This has never been achieved before in this room....

Especially taking into account the distortion curve:

Huge improvement vs the previous placement

And as mentioned above the low-end -3dB point got moved from 28Hz to 17Hz.....I have never heard anything like this in this room before; really jaw-dropping experience
I was worried that I will be able to locate the sub being behind me at the far-end of the room but with correct time alignment this is a non-issue, all the bass sounds are coming from the 'stage' in front of me
I think this exercise shows that moving your sub in the room can make a huge difference!

 

[312elements]

@ppataki your projects always make me smile. I’m just about ready to drop the drivers in my in the speaker cabinets which took much longer to finish than anticipated. Coat of wax this weekend and I’ll be good to go. Placed the order for my udio8 today and it should be here Friday. I’ll figure out what the host is going to be next week once the drivers are installed and wired up.

 

[surroundman]

Amazing projects. Keep up the incredible work!

 

[Somafunk]

I think this exercise shows that moving your sub in the room can make a huge difference!

Yep , it’s surprising just how much of a difference it can make. (But after a few years on here it is not surprising at all)

My ideal sub placement in my room was unfortunately in front of the log burner so not ideal, so I bought two of them as it was easier to find suitable placement, both dual opposed sealed Dynaudio 18s subs