Four subs for smoother bass

[greg2putt]

Very insightful Jonathan. I'll now put in my plans the need to pick up a 2nd Descent I, that hopefully happens in the next year as the new generation subs come out. Yes, pick it up at 1/2 price like the Stage pricing.

 

[JonFo]

[Note : it's been more than two years since I started this thread, just updating with latest findings]

So I finally moved the mini-media room to the small room I mentioned earlier. It is the worst possible shape, roughly a cube, 11.5'x11.5'x9', in terms of low-frequency acoustics, a nightmare.

This is why I wanted to see what fur subs could do, and in the loft, it worked well enough, way beyond my expectations. Well now, I just placed the four subs in the corners (offset 4" or so from the boundaries), wired it all up and broke out the measurement rig.

Wow, that's about all I could say for about 5 minutes. Especially once I saw the graphs. Please note: there is no EQ used in the chain during these plots. None.

So here's the baseline frequency response of the left-front speaker plus all four subs operational:

[Edit 12/25 - These plots are all messed up, as sb6 noted, I had a problem with the calibration of my new soundcard for REW, so of course the low end looks good when you mix in the original signal along with the mic . I've added new posts later in the thread, so feel free to ignore these and skip ahead]



Pretty smooth and extended for such small subs, but then, it's a small room.

Now, here's the waterfall plot, again, almost cubic room with no room treatments (well, a rug on the back wall, but that's for HF):



Note how fast the decay is in most of the range. Except for the little tail at 50Hz, it's great.

Now the Spectrogram chart, again, showing rapid decay and extremely even response, one of the best I've ever seen:



Pretty amazing results for roughly $600 in subs strategically placed in a small room.

 

[Peter Hogan]

As I mentioned in my other thread on the mini media-room system, I'm experimenting with using four subs to smooth out the bass in the room.

Well, even though I've yet to move what gear I do have to the smaller room, I decided to experiment and just place the four subs into the loft system and grab some measurements.

So I carted all four Dynamo's to the loft, good thing they are not too heavy

The loft is a medium-sized room, 20' long by about 14'wide with the 'left' side open to the stairwell, and the right open to an adjoining office.
It is also far from an ideal audio setup, so it presents some challenges.

The receiver I use is the Onkyo NR-818 with Audyssey XT32 room correction (lowest cost receiver I know of with XT32).

Mains are some old Mirage M-290 Bookshelf speakers with matching Mirage center and rear wall mounted surrounds.

Room is untreated stuco, so very reflective at all frequencies.


I brought up my measurement rig to the loft so I could tune the basic setup prior to running the room correction.

So that's the context, here come the tests and measurements.

Has your Onkyo experienced the speaker/network/USB chip failure yet? Mine is currently in transit for warranty repair of it...it appears they all fail within a couple years, and Onkyo has extended the warranty to cover it... http://www.onkyousa.com/Support/service_info.php

 

[sb6]

[Note : it's been more than two years since I started this thread, just updating with latest findings]

So I finally moved the mini-media room to the small room I mentioned earlier. It is the worst possible shape, roughly a cube, 11.5'x11.5'x9', in terms of low-frequency acoustics, a nightmare.

This is why I wanted to see what fur subs could do, and in the loft, it worked well enough, way beyond my expectations. Well now, I just placed the four subs in the corners (offset 4" or so from the boundaries), wired it all up and broke out the measurement rig.

Wow, that's about all I could say for about 5 minutes. Especially once I saw the graphs. Please note: there is no EQ used in the chain during these plots. None.

So here's the baseline frequency response of the left-front speaker plus all four subs operational:

View attachment 18354

Pretty smooth and extended for such small subs, but then, it's a small room.

Now, here's the waterfall plot, again, almost cubic room with no room treatments (well, a rug on the back wall, but that's for HF):

View attachment 18355

Note how fast the decay is in most of the range. Except for the little tail at 50Hz, it's great.

Now the Spectrogram chart, again, showing rapid decay and extremely even response, one of the best I've ever seen:

View attachment 18356

Pretty amazing results for roughly $600 in subs strategically placed in a small room.

Congrats, that's an unbelievably smooth response. Not to doubt you, but from 50Hz and below is dead flat. Are you certain that the sound card and mic are properly calibrated? Also, there appears to be a nasty room mode at ~60Hz(if you check dims in a room calc), I'm amazed it doesn't even show up. Can I ask what level you took your measurements at?

 

[Brad225]

JonFo,

My experiences are a match for yours.

I added 2 more Depth i subs for a total of 4 earlier this year. Making adjustments by ear resulted in amazingly improved sound throughout the entire frequency range.
An electrical engineer friend that rebuilt my interfaces on my CLS's had come over to listen to my system after adding my new front end and he pulled out his phone with an analyzer apt on it.
I realize the phones don't have the best mics on them but it showed a flat trace + or - about 3-4db. By this I mean less than 10db total movement. This is using only the adjustments on the subs and the treatments that exist in the room.

He said that was as flat a trace he had seen on the many systems he checked.

There is an entire thread on WBT forum by a DR. / Professor about using 1 good sub and 3 much less expensive subs. If I find it again I will post a link.

 

[sb6]

JonFo,

My experiences are a match for yours.

I added 2 more Depth i subs for a total of 4 earlier this year. Making adjustments by ear resulted in amazingly improved sound throughout the entire frequency range.
An electrical engineer friend that rebuilt my interfaces on my CLS's had come over to listen to my system after adding my new front end and he pulled out his phone with an analyzer apt on it.
I realize the phones don't have the best mics on them but it showed a flat trace + or - about 7db. This is using only the adjustments on the subs and the treatments that exist in the room.

He said that was as flat a trace he had seen on the many systems he checked.

There is an entire thread on WBT forum by a DR. / Professor about using 1 good sub and 3 much less expensive subs. If I find it again I will post a link.

+/- 7dB I get, +/- 0dB below 50Hz with no smoothing is better than mfr specs. No doubt more subs = smoother response set up properly, but the rule of thumb is diminishing returns >2 subs, so I'm amazed at the results above. IMO - I'd be concerned about running 1 high quality sub and 3 sub-par subs (pun unintended) whereby quality would become an issue since more subs doesn't automagically make a cheap sub accurate, fast, etc, etc...

 

[Brad225]

sb6, the difference in the addition of the second 2 subs was truly amazing in comparison.

 

[Speedskater]

Some recent activity in the:
Multiple Small Subs - Geddes Approach
thread:

http://www.diyaudio.com/forums/subw...mall-subs-geddes-approach-88.html#post4507217

 

[sb6]

BTW, MLs specs for the Dynamo 300 are as follows: System Frequency Response: 32–150 Hz ± 3 dB. Anechoic in the LFE mode. Something is awry JonFo's measurements to achieve flat bass below 30Hz.... the graph in the old room looks to be measured correctly dropping off at ~30Hz sharply.

 

[Brad225]

Some recent activity in the:
Multiple Small Subs - Geddes Approach
thread:

http://www.diyaudio.com/forums/subw...mall-subs-geddes-approach-88.html#post4507217

Hey Speed, Dr Geddes was the person I was referring to earlier, thanks.

 

[MPS]

Congrats, that's an unbelievably smooth response. Not to doubt you, but from 50Hz and below is dead flat. Are you certain that the sound card and mic are properly calibrated? Also, there appears to be a nasty room mode at ~60Hz(if you check dims in a room calc), I'm amazed it doesn't even show up. Can I ask what level you took your measurements at?

I agree, something could be wrong with the measurement.
Normally no calibration is required if quality sound card and microphone are used. Lots of measurements around with faulty calibration thus wrong measurement results.

SPL with mic calibration shown, sound card calibration quite unnecessary.

Spectrogram,note the 50Hz line frequency measurement anomaly .

 

[JonFo]

Has your Onkyo experienced the speaker/network/USB chip failure yet? Mine is currently in transit for warranty repair of it...it appears they all fail within a couple years, and Onkyo has extended the warranty to cover it... http://www.onkyousa.com/Support/service_info.php

Nope, I sure hope it does not, because so far, it's a nice unit.

Sorry to hear yours had to go in for repairs, I hope you got it back by now.

 

[JonFo]

Congrats, that's an unbelievably smooth response. Not to doubt you, but from 50Hz and below is dead flat. Are you certain that the sound card and mic are properly calibrated? Also, there appears to be a nasty room mode at ~60Hz(if you check dims in a room calc), I'm amazed it doesn't even show up. Can I ask what level you took your measurements at?

Ding, Ding, Ding! sb6 wins the prize!

The sound card (a Tascam US-366 USB audio interface) was indeed incorrectly calibrated. I can't believe I messed up that badly, but this was the first time I used it, as my old interface was Firewire based, and that seems to have gone the way of the Dodo.

The measurements that follow were all taken at a reference of 75dB SPL.

 

[JonFo]

JonFo,

My experiences are a match for yours.

I added 2 more Depth i subs for a total of 4 earlier this year. Making adjustments by ear resulted in amazingly improved sound throughout the entire frequency range.
An electrical engineer friend that rebuilt my interfaces on my CLS's had come over to listen to my system after adding my new front end and he pulled out his phone with an analyzer apt on it.
I realize the phones don't have the best mics on them but it showed a flat trace + or - about 3-4db. By this I mean less than 10db total movement. This is using only the adjustments on the subs and the treatments that exist in the room.

He said that was as flat a trace he had seen on the many systems he checked.

There is an entire thread on WBT forum by a DR. / Professor about using 1 good sub and 3 much less expensive subs. If I find it again I will post a link.

Brad, that's great to hear that your setup was improved now that you have the 4. There is still a bit more improvement to be had with well-applied DRC.

While combining cheaper subs works to an extent, I find that keeping them identical means the cancellations are smoother if they are matched.

 

[JonFo]

OK, so after yet another battle with the measurement system, I finally snagged good readings. So here they go.

First, the before DRC; not too bad. The worst is the room-boundary induced 50Hz resonance. If I measure a few feet out into the room, that's gone. The MLP is less than a foot away from the back wall, so unless I put a massive bass trap there, the 50 Hz node is a fixture.



And the corresponding Waterfall plot



The combined output of the 4 puny subs enables some reasonable output below 40Hz, but then it drops precipitously as expected for such small drivers.

 

[JonFo]

And now for the post Room Correction (Audyssey XT32 in this case)

Much smother in the 60 to 200Hz range. Energy is better distributed, with a bit more in the under 50Hz range thanks to a bit of boost:



And the corresponding waterfall:

 

[sb6]

And now for the post Room Correction (Audyssey XT32 in this case)

Much smother in the 60 to 200Hz range. Energy is better distributed, with a bit more in the under 50Hz range thanks to a bit of boost:

View attachment 18373

And the corresponding waterfall:

View attachment 18374

Great to see you got it all figured out, Jon Fo. The new measurements look reasonable and very good. Running a room mode calc you've got some nasties (as all of us do) at 50Hz, 80hz and 98Hz but it looks like you successfully tamed them quite well.