There is a recent article on the topic of room gain published in audioXpress by René Christensen, Simulation Techniques: Room Gain. The article includes some in-room measurements.
That 101dB SPL is for one woofer at a time. With six drivers working simultaneously, that should produce about a maximum of 116dB SPL, somewhat dependent on the effects of mutual coupling between the drivers.
What is needed is Vb=68.4 litres per driver. If an enclosure has two woofers fitted (e.g., to complement the front stereo speakers), and we want the same response as for one driver, then the enclosure volume needs to double to get the enclosure compliance correct for that configuration.
The triangle configuration makes it relatively easy to fit the subwoofers snugly into the corner locations. For a 2-driver subwoofer configuration, it would be helpful to use a design where the drivers are working in-phase but are mechanically horizontally opposed. This will serve to cancel some of the vibrations that might otherwise be transmitted.
I see no reason why it shouldn't. Just keep in mind that it's 68.4 litres per driver per enclosure. If going the sealed route, then enclosure filling will serve to increase the effective enclosure volume for a given physical volume.
Yes. The Q value used on the PEQ adjusts the width of action of the peak (or dip) that is being introduced.
You could look at the offerings to be found here. They include high-pass and low-pass filtering and EQ. With 4 subwoofers, the costs will begin to add up.
The Rolls SX95 uses only 2nd-order low-pass filters on the subwoofer signal. It also has no time delay or EQ facilities. That's a bit of a limitation in terms of getting the subwoofer to blend in with the main loudspeakers.
If you are going to use the Crown XLS1002, which is a stereo amplifier, then you will need two of them. One for the front pair of subwoofers and one for the rear pair of subwoofers. To provide the EQ and whatever filtering is necessary, you will need to consider using a pair of miniDSP units, one per Crown XLS1002. These will allow you to time-align the subwoofer outputs at the listening position.
The Xmax=7.0mm excursion limit shouldn't be exceeded if you wish to maintain the nominally linear operation of the drivers. Exceeding that limit will add distortion, and the drivers can be damaged if Xmax is exceeded in a big way. A little bit of over-excursion probably won't do any lasting damage, but it's best avoided from a sound-quality perspective.
We essentially lose 2.5dB of maximum SPL output capability above the cut-off frequency (in the passband), as we are driving the woofer into its Xmax region at a lower nominal power level.
The Q=2.0 value just means that the PEQ introduces a sharper peak than does a Q=1.0 setting. The plot below shows two different +5dB PEQ settings applied at 28Hz. The wider one corresponds to Q=1.0, while the narrower one corresponds to Q=2.0.
I don't think anything like that is going to occur here.
In the sealed configuration, we are getting 101dB SPL in the passband per driver. We have three drivers per left/right stereo side. With appropriate amplification, that configuration will produce 110dB SPL maximum of clean response, essentially down to 30Hz. That's quite loud, considering that many domestic speakers have an efficiency of around 87dB/W, and when driven by 100W will produce 107dB. We are comfortably beyond that, all the way down to around 30Hz.
Do you think that you might wish to exceed that sort of SPL output capability?
I did note that the Crown XLS1002 implements the PureBand Crossover System, which provides Linkwitz–Riley 24dB/octave filters allowing you to choose a cut-off frequency point between 30Hz and 3kHz on standard 1/12th-octave centers. Three filter types are available: Low Pass, High Pass and Band Pass. In the subwoofer application, the Low Pass option is the most useful one. However, there is no PEQ available.
It seems that using a miniDSP, or equivalent, for filtering and PEQ duties is the way to go if you are going to be making use of your available amplifiers.
An In Room response in and around 20hz is more than enough for music and easy with 6 12” drivers.
The SDA is a low Q driver……I don’t see any advantage to running them sealed with only 7mm of xmax……I think you’ll want/need the benefit of the port output.
If you were to take my advise and do the corner builds, each corner offers more than enough internal volume for one driver per corner/enclosure. You could build 2 of them slightly larger to get enough for 2 drivers…….5 cuft would do it.
Typically I’d advocate for sealed as well…..better impulse response……..but the SDA is xmax limited and has a very low Fs for its size and very efficient. This would make for an awesome horn loaded sub but to get to 20hz, they’d have to be huge.
@mayhem13, I chewed on this last night thinking about the info you and @witwald provided. I-hear-you about the Xmax of the SDA, and I am inclined to opt for your approach of ported. My mind likes the idea of sealed, but I have to work with the reality of the specs of the driver selected. I know I may sound fickle, but this is a process that I think through and enjoy pondering. Once it's done, a lot of the building fun will be gone....
I do think I need a better understanding of the PEQ aspects. I watched a Youtuber: Audioholics' video last night that shed light on my short knowledge about this aspect of bass management.....
I'm off to work but ill reach out when I get back in the research seat. Thanks again.
Have a Great Day!!!!!
I do think I need a better understanding of the PEQ aspects. I watched a Youtuber: Audioholics' video last night that shed light on my short knowledge about this aspect of bass management.....
I'm off to work but ill reach out when I get back in the research seat. Thanks again.
Have a Great Day!!!!!
I watched a video by a youtuber audioholics last night. He explained a lot on this subject; mentioning the need for a DSP manager for multiple subs. I think he uses Audyssey. This system is a bit out of my $$$ threshold but I will look into options for the miniDSP this evening.
I'm off to work now, but I do appreciate you getting back this morning..
Have a Great Day!
With 6 subs ported and if you select the tuning frequency correctly, you won’t need any EQ……….and that’s a really good thing and the whole purpose of a multi sub array. Wire one of the six out of phase for an even smoother response.
A note on EQ……some understanding……it’s a phase operation whether it linear or non linear…..and no matter how you slice it, phase adjustments leave behind artifacts. When in the studio, engineers will select the microphone that best captures the response of the instrument or voice so as NOT to have to apply EQ as it is well understood and easily apparent that shaping a signal sounds quite different from a native signal.
Keep in mind that using an AV receiver with Audyssey assumes that the subwoofers are powered. To be able to independently control 4 subwoofers, then a mid-range AV receiver is required, such as the Denon AVC-X3800H 9.4 channel receiver.
Don't forget that EQ changes the amplitude response and not just the phase response. The amplitude changes will create a very audible effect, which is likely to be the dominant effect.
And, of course, EQ adjustments leave behind amplitude artifacts too, which are generally very audible and the reason for applying the EQ in the first place.
Although we are getting into the realm of somewhat subjective evaluation of microphones, many microphones have a built-in EQ that forms their natural response, which, together with their sound capture directivity response pattern as a function of frequency, influences their sonic signatures.
As alluded to above, merely shaping a sound signal of one microphone to match that of another microphone does not take into account each microphone's directivity pattern. Is it any wonder, then, that the resultant sounds might be different?
I'm less than convinced that that would be the case.
Below is a simulation of two 12" drivers in a 142-litre (5.0 cubic foot) vented enclosure tuned to Fb=29.8Hz. The −3dB cut-off frequency is F3=34.1Hz, and with the rapid roll-off below F3, we are −15dB at 20Hz. How might an in-room response of −3dB at 20Hz be achieved in practice?
If the enclosure volume is doubled to Vb=284 litres (10.0 cubic feet) and the box is tuned to Fb=20Hz, the following response is obtained. Here we have F6=21.2Hz, which might give some reasonable scope for achieving an in-room 20Hz response at −3dB. Is this a viable option?
I didn't see it written how leakage was taken into account. The measurement stops at 5Hz and it's not clear that other factors aren't involved.
Hmm, F3 =0.28*0.44^-1.4 = 31.81 Hz
Don't know how to rearrange the formula to find a ~0.61 Qts' is required, then using this calculator to find 2.8 ohm to get a ~30 Hz F3, though factoring in a bit of heat rise and 2.5 ohm should do it based on published specs, but using a trim pot can quickly dial it in easy enough and of course calc a new net Vb: 20*Vas*Qts'^3.3, Fb = 0.42*Fs*Qts'^-1.4
The Margolis and Small (1981) formula is for a "natural flat alignment", and is:
F3 = 0.28*Fsb*Qts^(-1.4)
For the driver in question, this gives F3 = 0.28*24.2*0.33^(-1.4) = 32.0Hz.
VituixCAD produces a result of F3 = 34.6Hz for a tuning that is similar to a natural flat alignment.
F3 = 0.28*Fsb*Qts^(-1.4)
For the driver in question, this gives F3 = 0.28*24.2*0.33^(-1.4) = 32.0Hz.
VituixCAD produces a result of F3 = 34.6Hz for a tuning that is similar to a natural flat alignment.
Your sims are Anechoic. OP says his space is 12ft x18ft……….even assuming a very leaky space, there’s going to be a solid 6db of gain at 20hz. You can also tune them as an EBS to lower the cutoff……which is what I suggested based on the gain structure of his space. Building these into the corners as mentioned delivers significant boundary reinforcement from the side walls and if placed near the floor, additional ground plane boost. Down fire the last two stand alone enclosures and place them closer to the LP.
Since my contention and those from most professional acousticians is that it’s ALL modal below 80hz, excite as many modes as possible……..6 drivers do this well…..and if the OP tunes them all slightly different (2-3db) even more modes are shifted.
Hey guys, Good Morning!!! I had a longer work-day than expected yesterday so I haven't had a moment to digest your discussion point. I can say that from first skim-through, I find a great deal of positive options from both @mayhrm13 and you @witwald. Data is math and math is usually king.
When I get a minute I will study all that you guys have put forth.... What "I"can add, is that you both are polite and respectful as you discuss your thoughts and expressing your understandings; I appreciate it!
As the OP, I'm learning on the fly. And as I mentioned earlier, this is a fun project. Off-the-cuff, I'm thinking of how can I build a first version starting from @mayhems' concept, being that it seem the least cost involved, and if it work to my satisfaction, well good deal! And if this plan leaves something to be desired, I can roll-back to the plan that incorporates the DSP and EQ. BTW, I found an offering to possibly suit this roll, but it cost $250-$350. This is not a lot, but it pushes the project overhead up a bit; I don't really care, ultimately I want to land at the best sound possible. I'll leave a link of the DSP devise in case anyone wants to take a look and comment.
Have a Great day, and be Safe! 🙂 I'll be in touch soon. Thanks.
Mini DSP
https://deercreekaudio.com/products...VVm8ob8cMKzjmLksS27_64QWvS77eR3UaAg5JEALw_wcB