15-18" Powered Subwoofer Pair Build

[gnarly]

For home use there’s no need to do that.

That's an opinion based entirely on your personal preferences, isn't it?

You’ll get plenty of output out of a smaller closed box, which offers the additional benefits of pressurizing the room lower and avoiding vent resonances.

For me, no, I'll not get sufficient output out of smaller closed boxes.
And I need to be added to the list of folks who say vent resonances are a non-issue, if a sub is being used as a sub and crossed sufficiently low. (120Hz 4th order etc.)

Basically small signal modeling is pointless today. All you really need to concern yourself with is power and excursion. EQ will set the frequency response.

I fully agree large signal performance is what counts. But are you aware how far small signal simulation has come in predicting large signal performance?
Have you used Hornresp for subs? Excursion and power are modeled quite well imo. I've done a fair amount of large signal sub testing outdoors and have found good correlation with Hornresp.

A good rule of thumb if you don’t want to get into setting limiters is: enter the driver and your amp power (or the biggest amp you might use) into Unibox or whatever, and find the smallest closed box that will push the driver to rated xmax.

Won't that strategy apply clipped waves to the sub?
That is, unless you have a clip limiter built into the amp (which unfortunately, most home gear does not, afaict)

I'm happy setting limiters, I want all the transient head that is available! Heck, I've even built frequency dependent limiters in programable DSPs to get every clean mm excursion.

Subs are easy, especially when you’re dealing with this kind of extreme driver.

For reference, I recently had to replace an Aurasound NS18 that started to make some noise after ~15 years of service in three different cabinets (pour out a glass). I used a BMS driver in the I think ~85L closed box, though I think 850 not 862. After EQ there’s no difference I can detect. (The other 2 subs use Aurasound drivers.) The system still extends down to 16Hz with a smooth rolloff below that. It is still smooth to ~700Hz, well beyond

I agree, subs are easy. Easiest DIY of all.
You use a sub up to 700Hz?
Yikes, I never cross higher than 120Hz, and i use a 16th order low pass. (linear phase)
Glad you're happy with the amount of subs and extension you have...
I could still use a little more of each.....and I'm running four of the 862's (ported)
Back to personal preferences

 

[sarumbear]

Yikes, I never cross higher than 120Hz, and i use a 16th order low pass. (linear phase)

How on earth you achieve that!? 96dB/oct!

 

[sarumbear]

But correct drivers are not cheap, because it takes a lot to do the engineering right and make them. There are two ways to get good drivers, one is to do as much engineering as purifi, the other one is servo or klippel system. All other ways are close but cannot compete with those two mentioned above. Besides linear Xmax, the other benefits of servo are getting rid of the effect of aging on the driver, driver tolerance, heat compression, limit excursion in a range and able to compromise some part of the driver to get lower price. Those are even better benefit than linear Xmax. I would say servo subs are the real high end subs if the servo works.

Why is servo not used often then?

 

[abdo123]

A pair of any of the drivers i previously listed will match it, or even best it for a fraction of its price.
You could even do two dual opposed subwoofers (4 drivers) if you lack the space, which would certainly best it everywhere.

Dual opposed do not save space, a subwoofer with two drivers needs twice as much air as a subwoofer with only one driver.

 

[Farenheit]

¿Dónde te encuentras?

Si estás en Europa, tendrás que recurrir a marcas profesionales, como B&C y 18sound.

En los EE. UU., tendrá muchas más opciones.

+ Beyma

 

[ryanosaur]

Why is servo not used often then?

I'll take Outdated Technology for 400, Alex.

 

[morpheusX]

Dual opposed do not save space, a subwoofer with two drivers needs twice as much air as a subwoofer with only one driver.

I think we are both right

For example, here on data-bass.com:
- Single B&C 21SW152 in a 23" x 23" x 20" box.
- Dual opposed B&C 21SW152 in a 24" x 24" x 24" box.

The dual opposed box is not optimal and you will need more power, but its just a tiny bit bigger than the single one.

In some situations, it might be a good option.

 

[sarumbear]

I think we are both right

For example, here on data-bass.com:
- Single B&C 21SW152 in a 23" x 23" x 20" box.
- Dual opposed B&C 21SW152 in a 24" x 24" x 24" box.

The dual opposed box is not optimal and you will need more power, but its just a tiny bit bigger than the single one.

In some situations, it might be a good option.

For the same FR, two drivers will require twice larger box. Not to mention that the volumes you listed above are 30% different.

 

[gnarly]

How on earth you achieve that!? 96dB/oct!

FIR. I thought saying the low pass is linear phase would be enough to make FIR usage clear.

If fixed delay is not a problem, it's a pure joy to be able to choose any order xover, from first order to brick wall, without suffering any phase rotation.

 

[bigjacko]

Why is servo not used often then?

I actually don't know.

 

[jhaider]

That's an opinion based entirely on your personal preferences, isn't it?

I would say personal experience, not personal preference. The distinction is, it's not about what I might or might not want, but what is actually necessary to accomplish a goal.

I suppose one caveat would be is I think about fidelity first, fidelity last, and fidelity at every single point in between. In that respect personal preference is operative. If one's more just about boom boom boom boom boom...yes, she will not find my experience applicable.

Let's be real about the goal for a minute. Even if you take the oft cited "cinema reference" levels as the gold standard - which is counter to guidance* and will make you deaf if done regularly, but we're assuming arguendo here - the cinema spec is 112dB for the subs down to 31.5 Hz**. If you look at data-bass a single closed box 18N862 will get you beyond that - at two meters away in free space outdoors. In a small room, against two or more boundaries, one can anticipate greater output. In multiples, there's enough headroom to ensure the sub amps are all loafing all the time.

If you go by the real spec for small rooms, that goes down to 108dB. A JL Audio E112 can do that - again at two meters away in free space outdoors. In a room a JL Audio E110 or the old Peerless XXLS-based SVS sealed 12 would likely suffice. Three of either of those certainly would.

*For a typical large domestic room (1500-5000 ft^3) "reference" level is 78dB, not 85dB. See this thread: Are you getting the "THX Reference Level" in the listening position? Do we really need such an output level in our home-theater rooms?

**Dolby's actual LFE spec is, per https://professional.dolby.com/siteassets/cinema-products---documents/dolby-atmos-specifications.pdf,

3 Screen Subwoofer
3.1 Sound Pressure Level: +10 dB (Compared to Center Loudspeaker)
The Low-Frequency Effects channel subwoofer must have a flat response over the range of 31.5 to 120 Hz. When compared with a full-range screen channel, the subwoofer channel must be capable of producing +10 dB of in-band gain (for example, as viewed on a real-time analyzer).
3.2 Frequency Response: 31.5–120 Hz, ±3 dB

Somebody's invariably going to whine that program has content below 31.5 Hz. That is besides the point. Some of it is likely in error (have you seen a mastering studio with an AVS level of subwoofer excess?), but certainly some is intentional as well. The spec is such that there's no expectation that the subwoofers playing it will be flat down to those frequencies. And if you want deep low bass, it makes more sense to closed box subs than subs that die below a tuned frequency.

For me, no, I'll not get sufficient output out of smaller closed boxes.

That says one of two things for me:
1) You set up your system in a space the size of the Musikverein, that was turned into a padded cell with "treatments," with some PA line arrays playing above the subs, or
2) There are core issues in setup or calibration that cost a lot of headroom throughout the subs' operating bandwidth.

And I need to be added to the list of folks who say vent resonances are a non-issue, if a sub is being used as a sub and crossed sufficiently low. (120Hz 4th order etc.)

One thing of value Erin and Amir have provided with their Klippel scans is much more granular behavior about how ports behave.

Let me put it another way. If a speaker had a midwoofer with an out-of-band breakup, wouldn't you expect them to notch it out, even if it's well above the passband?

I fully agree large signal performance is what counts. But are you aware how far small signal simulation has come in predicting large signal performance?

No, because it's not necessary to bother with futzing with all that shit for home use, with the powerful drivers and amplifiers we have readily available.

Won't that strategy apply clipped waves to the sub?

How so, even at eardrum-rupturing levels?

You use a sub up to 700Hz?

Re-read my post.

I could still use a little more of each.....and I'm running four of the 862's (ported)
Back to personal preferences

Work on setting them up better. Or close the ports.

 

[Adi777]

I-Pal

B&C woofers?
As for the servo, I've read a lot of very positive feedback about the Rythmik G25HP. Four G25HP in my cinema room, sounds nice, but I think Rythmik not send subwoofers for Poland.

 

[sarumbear]

FIR. I thought saying the low pass is linear phase would be enough to make FIR usage clear.

If fixed delay is not a problem, it's a pure joy to be able to choose any order xover, from first order to brick wall, without suffering any phase rotation.

I am sorry. I wrongly assumed it is a subwoofer unit with some sort of a plate amplifier and I do not know any such unit with a DSP on it. May I then assume that you are using an external digital DSP device to work as a crossover?

 

[gnarly]

I am sorry. I wrongly assumed it is a subwoofer unit with some sort of a plate amplifier and I do not know any such unit with a DSP on it. May I then assume that you are using an external digital DSP device to work as a crossover?

No problem, sorry i was too terse for easier understanding.
Yes, i use external DSP. The platform i absolutely love is QSC's Q-Sys.

 

[sarumbear]

No problem, sorry i was too terse for easier understanding.

Yes, i use external DSP. The platform i absolutely love is QSC's Q-Sys.

I wasn't aware that there is Q-Sys device that is suitable for home use. What model do you use?

 

[gnarly]

I would say personal experience, not personal preference. The distinction is, it's not about what I might or might not want, but what is actually necessary to accomplish a goal.

Well, here's the thing...."actually necessary to accomplish a goal" is "your goal", isn't it?

Personally, i feel it's extremely difficult to separate experience from preference, after mental rationalization defines 'my experience' as optimum (for 'my preference').
Experience and preference are umbilically linked I think. A lot of chicken and egg going on....

Anyway, i try not to claim my audio experiences or preferences have universal fit for the next guy, except on very clear cut technical items.
And i heavily discount the experiences/preferences of others who feel theirs have an universal fit.

I suppose one caveat would be is I think about fidelity first, fidelity last, and fidelity at every single point in between. In that respect personal preference is operative. If one's more just about boom boom boom boom boom...yes, she will not find my experience applicable.

Of course. I define fidelity as an acoustic transfer function vs signal that stays linear at highest desired SPL, including peak transients.
A less flowery definition that most use, but works for me.

Let's be real about the goal for a minute. Even if you take the oft cited "cinema reference" levels as the gold standard - which is counter to guidance* and will make you deaf if done regularly, but we're assuming arguendo here -
the cinema spec is 112dB for the subs down to 31.5 Hz**.

I'm sorry, i don't care for arguendo at all....i say screw debate...I'm here to learn and share.
Besides, I know what i like and all the recommended commercial specs on earth aren't going to change that.

If you look at data-bass a single closed box 18N862 will get you beyond that - at two meters away in free space outdoors. In a small room, against two or more boundaries, one can anticipate greater output. In multiples, there's enough headroom to ensure the sub amps are all loafing all the time.

I wish. Granted my room is larger than most, and I no doubt enjoy occasional blasts louder than most.
So I still see my sub amp, a PL380 with 2500W per ch @ 4ohm per channel clip at times, driving two of the BMS 18"s per channel.

Somebody's invariably going to whine that program has content below 31.5 Hz. That is besides the point. Some of it is likely in error (have you seen a mastering studio with an AVS level of subwoofer excess?), but certainly some is intentional as well. The spec is such that there's no expectation that the subwoofers playing it will be flat down to those frequencies. And if you want deep low bass, it makes more sense to closed box subs than subs that die below a tuned frequency.

For the sake of fidelity, there's more to it than just how low in frequency a sub reaches. ime/imo.
A benefit of a sub reaching lower than musical content is that it flattens phase to a lower frequency as well, which of course pushes group delay / phase rotation lower.
31.5Hz changes relative timing in a positive way, with a sub that reaches linearly below 31.5.

That says one of two things for me:
1) You set up your system in a space the size of the Musikverein, that was turned into a padded cell with "treatments," with some PA line arrays playing above the subs, or
2) There are core issues in setup or calibration that cost a lot of headroom throughout the subs' operating bandwidth.

With a primary goal of getting along, I'll refrain comment on those

Work on setting them up better. Or close the ports.

Funny!
Here's mag and phase from my dual opposed 18" reflex build. Taken outdoors ground plane.

Here's distortion at 125dB @ 1m

I've already posted CEA pass rates for the build in this thread.

What would you suggest I do to set them up better ???
Apart from porting them lower lol...

 

[gnarly]

I wasn't aware that there is Q-Sys device that is suitable for home use. What model do you use?

The Core 110f is undoubtedly the most widely used model for home use.
It has USB I/O and 8x8 bal/unbal analog channels. Along with another 8ch that can be individually set either as input or output.

Larger cores, like the current 510i ($$$) , or recently discontinued 250i and 500i that can be found on ebay ($) are also used, but a lot less seldom.
They need plug-in cards for I/O.
But have much more processing power than the 110f, if your are heavily into FIR.
For moderate FIR, say 8ch of 4k taps per channel, the 110f is totally fine, and show up on ebay quite often.

 

[sarumbear]

The Core 110f is undoubtedly the most widely used model for home use.
It has USB I/O and 8x8 bal/unbal analog channels. Along with another 8ch that can be individually set either as input or output.

Larger cores, like the current 510i ($$$) , or recently discontinued 250i and 500i that can be found on ebay ($) are also used, but a lot less seldom.
They need plug-in cards for I/O.
But have much more processing power than the 110f, if your are heavily into FIR.
For moderate FIR, say 8ch of 4k taps per channel, the 110f is totally fine, and show up on ebay quite often.

Thank you. It’s a beast! It must be fun to wire it

For a 2-way active stereo crossover do you think it offers more filter flexibility than a miniDSP?

 

[azzy_mazzy]

a subwoofer with two drivers needs twice as much air as a subwoofer with only one driver.

not always, if you use two drivers in Isobaric design you can cut the volume needed in half and supposedly decrease the distortion BUT you lose 6db in sensitivity.

 

[sarumbear]

Isobaric is a special case which was outside of the scope that was being discussed.