Couple of questions regarding subwoofers

[mk1981]

Since I'm going to renovate, I want to go big. Maybe even six subwoofers. I realized it's fairly affordable to simply order pre-cut MDF panels if it's just a cube. All I need are proper speakers to go into the enclosures.
And here's the thing: As much as those Dayton Ultimax speakers are enticing, I don't think I want to spend that much, at least not from the get go. If anything, maybe I'll decide on the size first and upgrade to Dayton later on as funds allow, buying a couple every year or two.

So I have a couple of questions regarding how to best approach equipping the room, especially with regards to subwoofers.

The room is 4.25 m (14') wide, 9.2 m (30') long and 2.45 m (8') tall. 39 m² (420 sq.ft.) surface area, 95.7 m³ (3383 cu.ft.) volume. Well outside of Bolt-area, I'm afraid. In the future, the room may get smaller if I decide to renovate and add a room for work from home.

At this time, however, roughly one half of the room will be for TV/theater and will have a 5.x.4 configuration. Hard to place extra two speakers for 7.x.4 since one would go into a doorway, and the other in front of a window door.
The other half of the room will be dining area.

I want sealed subwoofers. I'm big on music, less so on movies. I know ported subwoofers are great for using less power, but they won't go as low in a room that small, and another reason is that cycle delay in a ported subwoofer is fairly great (like 20 cycles vs. 6-7 in a sealed sub). If I understand that correct, it's half a second delay for 40 Hz, so it really seems like a lot. I'm happy to be proven wrong.

1. I was thinking of four 15" subwoofers, now I'm thinking of six 12" or six 15". Sure, bigger is better, fell free to tell me to get eight 18" subwoofers, but I hope you won't and you can tell me what the reasonable maximum is.

2. Based on T-S parameters, there's an optimum box size for a speaker. Is that optimum size the smallest one needed, or is it optimum and one shouldn't go up or down? Suppose I build a box for a 12" speaker but size it for a 15" one. Will that work if I decide to upgrade to 15" in the future?

3. What about dual-opposed speaker configurations? Do they make sense in DIY subwoofers to take advantage of force canceling, or is it better to go for a larger subwoofer speaker in the first place? If building a dual-opposed subwoofer, do I simlpy take the recommended box size and double the volume?

4. Suppose I get six subwoofers. How do I place them in the room? For four, I was thinking two on the front wall, two in the surround speaker position. For six, what if I add them on the far wall?
Sounds reasonable, but what happens when I want to listen to music in the dining area? Much of it plays from afar, but the subwoofers in the dining area would be much closer. To compensate, the front speakers would have a delay of roughly 22 milliseconds compared to rear subwoofers. When a sound starts, the rear subwoofers play first, 5 milliseconds later the sound reaches people in the dining area, front speakers (and subs) begin playing another 17 ms later, and sound from them reaches the listeners in the dining area after 22 ms, for a total delay of almost 40 ms. Seems like a lot. Am I overthinking this?

5. If I get six subwoofers, is there a mini DSP that has 6 or 8 outputs and won't break the bank, or should I simply get two miniDSP 2x4?

 

[MediumRare]

Sounds to me like you’re preparing to film your homemade version of Ben Hur just because you can. You’re going to lose out on a lot of technology as well. I would buy 4 REL or SVS 12" and declare victory. Sonically, you won’t improve on that. IMO of course!

 

[amail]

As a real life result, you'll need to remember that nothing you design/calculate will work as intended on paper. Reality/room has its own rules. The best you can do is design your enclosures as per manufacturer's guidelines. When ready, acoustically measure the room's sound (mic, software) and tweak the whole system via the DSP with the real results to get closer to your intended aim.

Generally speaking, you may want to consider the fact that multiple large woofers have their own set of interference problems in a room, particluarly if they are spread in different locations. Due to their omidirectionality, not much is really gained by placing them in different places. Companies such as ML and others usually cluster, as much as possible, multiple woofers in one place (possibly stacked) to attain, again as much as possible, a single sound source spot.

 

[Umikuser]

It all depends on the room obviously, but imho, spreading multiple subwoofers evenly against and along the same wall as your main front speakers is generally beneficial compared to a single one. If you move them around and measure the results at your listening position you can achieve very good results. The more subwoofers you have, the easier it is to get good results, the returns are diminishing though.

Spreading multiple subwoofers around the room is something you can experiment with if your room is very small and your options are very limited (ie it is dang near impossible to place them along/against the wall where your front speakers are).

Related:
Keep in mind, the room itself do have a huge influence on the result, almost more so than any subwoofer. Treating acoustical issues at lower frequencies usually requires physically large measures, while they might not be all that expensive and able to build from cheap materials. It is still something to consider doing since the results will be considerable.

 

[amail]

Not quite clear to me how spreading woofers around the room will improve things, apart from arrival times, which in themselves add a new set of problems. Acoustically measure your room when done and you will be surprised by the "jungle" of interference, so best to minimise variables as much as possible. Not to worry, there will still remain enough of them to tackle and smooth out. If you could only "see" the hundreds of waves skipping and jumping around your room, some creating peaks at certain freq., others creating (close to) nulls, you'll probably think twice or more about the whole project. Keep it simple as much as possible.

 

[Umikuser]

For clarification. I don't suggest that anyone spread "subwoofers around the room", I did however suggest that multiple subwoofers behind the main front speakers along/against the wall is beneficial.
Spacing them equally along said wall is a good starting point, but it may not end up being the ideal placement after measuring.

 

[Todd Anderson]

That’s a decent sized room… four sealed 15” subs would be more than enough to have impactful bass with lots of headroom.
Placement-wise, you can download REW and play around with some models. Quarter points of your front and rear walls might work.

Honestly, I can’t imagine 6 subs paying bigger dividends

 

[mk1981]

I read a couple of articles in the past detailing subwoofer placement. I'm attaching them. I had downloaded two and was able to find the other two.

One is the highly regarded paper from 2006 by Welti and Devantier of Harman. Second is a presentation by Welti showing the same results.

Another one I found was a PDF by Lynx audio, though I think it's more applicable to open-air stage than home. I'm attaching it nevertheless.

And there's a fourth link: https://sunmaiblog.wordpress.com/2010/09/23/the-placement-of-one-or-several-subwoofers/
This one advises for putting multiple subs along the front wall as well.

 

[mk1981]

After adding shipping and handling, customs, the price of Dayton speakers about doubles for an item that has effectively no warranty since it involves paying return shipping. I could smuggle them here (in the sense of trying to avoid paying import tax and tariff), but if caught, it's a rather big problem, including risk of losing my job. There are some importers here which take the burden of warranty and who already paid import costs, but after adding their fees on top of the speaker price, the price is more than double.

So I was thinking if there were some speakers available locally which could provide a similar effect with more bang per buck.

I couldn't find anything remotely approaching Ultimax in terms of performance, and I know you get what you pay for, but there are some speakers which are suspiciously similar in specs but come at very different price levels. I suspect they're made in the same factory and depending on how budget the brand they slap on it is, the price is lower or higher.

I'm attaching two pictures for a typical speaker. I can get twenty for the price of one Ultimax and what I'm thinking is that since this speaker handles 200 watts (100 watts RMS), it should be possible to use four in parallel (2 Ω), quad-opposed (or stacked dual-opposed) from a single 800 W amplifier and get a pretty good frequency response.

It's hard to read, but the top charts' x-axis is frequency (log 10, starting at 10 Hz), y-axis is impedance in Ω, starting at 0, horizontal lines are 20 Ω apart, small ticks are 2 Ω apart. Bottom chart is SPL frequency response. X-axis starts at 20 Hz, y-axis starts at 0, horizontal lines are 30 dB apart, small ticks are 3 dB apart. The impedance and bottom chart is SPL frequency response.
20-200 Hz SPL is 80-93 dB (the table states -6 dB extension is from 20 Hz).
The second one is just parameters of a box.

Similar data for 15" Ultimax shows that resonant frequency is 19.5 Hz rather than 29.8 Hz, power handling and Xmax are much higher, but here's the thing: If one Dayton's Xmax is 19 mm, and this one's is 5 mm, then four of these cheap speakers sweep the same volume of air along Xmax as a single Dayton, so it's not that bad.

What am I missing?

 

[Umikuser]

Had to look up the Dayton Ultimax data and from what I can tell, it is a completely different speaker. Very little in common with the Dibeisi one. Apart from the cheap one being made with cheaper materials, it more importantly also has completely different TS parameters, confirming that it has very little in common with the Dayton. I am sure you already knew that though

For the sake of argument, let's assume that you can buy 20 cheap woofers for the price of one good one, will it net you something useful? Hmm, it will most likely move a lot more air. But being that cheap, it is likely to have issues with distortion, lower frequency behaviour, quality control etc. It looks like its intended use is for car audio. I would probably look for another alternative. Car audio speakers are used in a completely different environment, the car itself with the doors closed and windows rolled tend to boost the lower frequencies A LOT. That never happens in a home environment, for that reason alone I would stay away from anything marketed as car audio unless you can do the proper math and do some measurements to confirm that the TS-parameters are actually what the manufacturer claim they are.

I am not familiar enough with the current woofers or I might have a recommendation for you. Perhaps someone else can help out?

 

[amail]

Since you've decided to go with multi-multiple woofers, use the golden rule regarding cheap or cheaper drive units: use them about 1/4-1/5th of their rated max travel. Good expensive units tend to display good linear performance at or near their max travel, cheap units won't. Deficiencies in materials, manufacturing tolerances, basket, voice coils (diameter, length, wire type, etc), cone breakups in various modes are all more pronounces in cheaper units. Using them a fraction of their max performance will also minimise these deficiencies and keep them linear. I've tried it in the past with 4 cheaper units, 2 stacked per side and the THD was quite acceptable with a rather tight clean sound.

Also related:
Most of the graphs shown in your uploads show almost exclusively simplified direct sound paths. This is not how sound behaves in a room; you, and everybody else, will have untold numbers of reflected waves from wall, furniture and what not, at various freq. which are difficult to depict in graphs. Software like REV and similar will however give you and overall picture of what's happening at various points in the room, corresponding to mic placement, which will largely include the cumulation of all thess phenomena.

 

[amail]

To Umikuse reply:

***Everything*** else being equal, more drive units will physically travel less than one. How much less? Double the units (cone area) - half the travel. Double it again - half it again. Etc. So in a room where you want a certain sound pressure, more drive units will travel less for the same acoustic output.

 

[bearr48]

I think you should buy and read Dr. Floyd Toole's book, "Sound Reproduction". It goes into great detail about speaker/room interaction, and the possibility of canceling room modes by careful placement of subwoofers. And REW is a great tool for learning what's really going on in a listening room.

 

[bearr48]

I think you should buy and read Dr. Floyd Toole's book, "Sound Reproduction". It goes into great detail about speaker/room interaction, and the possibility of canceling room modes by careful placement of subwoofers. And REW is a great tool for learning what's really going on in a listening room.

 

[bearr48]

Sorry, my computer is acting weird.

 

[mk1981]

Okay, I see what you mean. I played with this closed box simulator here: http://www.loudspeakerdatabase.com/Simulators/ClosedBox
and I can see that the results would be terrible.

I'm still looking for a good and affordable speaker, and found a couple of candidates from a different company that manufactures here in Poland:

W.30.200.8.MC

Diameter - 305 mm Power - 200 W Impedance - 8 Ω SPL - 91 dB Fs - 29 Hz Qts - 0,56 Vas - 166 L Cone material - Modified Cellulose Recommended enclosure - 60 L closed

stx.pl

W.32.500.8.MC

Diameter - 332 mm Power - 500 W Impedance - 8 Ω SPL - 92 dB Fs - 25 Hz Qts - 0,30 Vas - 155 L Cone material - Modified Cellulose Recommended enclosure - Bass - Reflex

stx.pl

W.46.1400.8.MCX

PA Subwoofer Driver Diameter - 465 mm Power - 1400 W Voice coil diameter - 100 mm Impedance - 8 Ω SPL - 98 dB Fs - 30 Hz Qts - 0,27 Vas - 269 L Cone material - Mo

stx.pl

(The last one is marketed as a PA speaker, as evidenced by its power handling, but T-S parameters indicate it would work great as a subwoofer driver)

I do have a couple of questions regarding sealed vs. ported.

Some speakers are recommended to be used in a sealed box, others are recommended to be used ported.
Is it only because lower power speakers need bass reflex to provide any sort of reasonable output, or is there another reason for that?

The Subwoofer Conundrum

The Subwoofer Conundrum - to use one or two subs, subwoofer placement, fast bass, slow bass, standing waves, Xmax and more

sound-au.com

This paragraph there:

5 - Soundwave Vs. Pressure Mode
One claim that is very popular is that you cannot reproduce bass in a small space. While this is certainly true of soundwave propagation, it is completely false if pressure mode is excited within the space. (...) Most subwoofers ultimately rely on pressure mode to obtain the lowest frequencies in typical rooms, and it is notable that vented systems in particular are unable to excite the pressure mode properly in many rooms.

One of the reasons for this could be that there is a vent that allows the pressure to equalise. In theory, this is meant to be a resonant system, where the back wave of the loudspeaker is inverted in phase and augments the main cone wavefront. That the principle works is demonstrated by many large systems in auditoria, theatres and even outdoor venues, but all of these are large spaces where soundwave propagation is dominant. (...)

Do vented subwoofers make sense, then? Is there a difference between a vented box and one with passive radiators? The pressure won't equalize, but since the principle behind them is exactly the same (use the pressure front generated by the rear of the loudspeaker to excite a moving mass — air in a vented box, passive radiator cone mass in a passive radiator box), will they or will they not work better in a room?

 

[Umikuser]

The very short version. The TS parameters describes a drivers physical+electrical behaviour/properties when in free air.
If put in any sort of enclosure, the behaviour/properties will change drastically.
TS parameters dictates what enclosure is best suitable for any perticular driver, vented/sealed etc.

IMHO, what type of enclosure you choose should be decided primarily by the driver you use. It is possible to achieve good results with both closed and vented enclosures (assuming obviously that it is a good match for the woofer).
Consider tho, a sealed enclosure means simpler math, easier to build with predictable results, but generally larger in size. A vented enclosure is more difficult to get right, harder to build, but if you get it right it can be smaller and achieve very good result.

You could also say that an enclosure will change the mechanical parts of the TS parameters

PS. You actually also manipulate the electrical TS parameters of the driver with things like negative impedance and other electronical wizardry (like Yamahas Ace-Bass).
HOW that is done is above my pay grade though, physics still apply tho, so it comes at a price. Power usage/sound pressure etc. :-) DS.

 

[Umikuser]

If it were me I would pick a couple of different drivers, run some simulations for enclosures and most likely end up with a couple of alternatives to choose from.

 

[amail]

For a given woofer that can be used in either a ported or sealed box, the volume of the sealed box would generally be smaller. By quite a margin. Personally I prefer sealed boxes as, among many other factors, the rollof frequency is milder and smoother.

 

[Umikuser]

*doh* My bad, amail is indeed correct, a sealed enclosure for a given driver is generally smaller than a bass reflex. Not sure what I was thinking writing the opposite :-)
It is also true that a sealed box has a less sharp roll off than a bass reflex. But a bass reflex enclosure tends to go deeper in frequency than a sealed one, this might be important to you and should show in any simulations you do, it is a game of compromises.

 

[amail]

Just a qwick headup on the 3 units you posted. None of them are "outstanding", but of the three I would take the W.32.500.8.MC. None of them have adequate coil diameter (not even the 18"), but life is compromises. Luckily they are vented magnets. You won't find concave caps at this price range, so convex it must be. I hope the surround cone suspension is rubber, difficult to say from the photo and the specs. say nothing about it. "Modified cellulose" usually means some type of plastic/polypropylene material. I guess a good bet would be to use two such woofers per cabinet (I've seen and heard 4 units used on each facet of a large cabinet but they can't then be placed against a wall), placed back to back on left-right sides of the cabinet. Better still if you can connect the back of the two magnets with a strong, tensed metal rod. Placing them back to back and rod-connected will somewhat cancel vibrations and resonance of the two opposite units/panels/baffles; placing the units facing forward on one baffle is way worse. Total power handling should be no problem. If you can, direct drive them with two quality class D amps instead of passive xovers and elec. xover or better still DSP. Power amped will necessitate power supplies (possible inside the boxes) and mains connection. Placing the woofers along the top room wall and on the floor will further increase bass output (though not in the cleanest way). Note:- two 8 ohm woofers in one box, connected in parallel, will become a 4ohm speaker system, but that's not an issue with quality class D amps. Don not connect in series (16ohm; if one woofer goes, both go). Woofers connected in phase.

Of woofers:
Every detail of the unit's build is important and ccontributes/detracts from the final result (everything else being equal. The basket and its rigidity, does so. The spider material does so, the convex/concave caps do so. The diameter of the coil and the type of wire used do so. The cone material - which is always a compromise between contradicting attributes - do so. Most common woofer cones are some sort of plastic, paper, metal (aluminium) or aramid/kevlar. Some have good damping properties and are rigid but have high mass; some are low mass but not well damped; etc. The sound timbre also vary from type to type. Of all traits, the stored energy is one of the worst since the higher it is, the less ability of the driver to start/stop quickly. Materials also have micro-porous/molecules-stored residual energy, the higher value of which causes the cone to ring. The smaller coil diameter will mechanically cause the apex of the cone to start/drive faster than the perimeter thus to cause flexing of the cone. For a given cone diameter and its concavity height, there is a sweet spot diameter of its spider. Unfortunately, for a 12" example that you show, a 2" spider is not that spot. You may remember the Kef B-139 (and later others) which attempted to minimise cone flexing and breakup modes by adding a flat face sealing the entire cone, and break the symmetry of a round unit into an oval shape. Unfortunately, adding that flat face also added mass. There are no free meals in physics.

There are many more parameters well beyond the scope here. The TS data is a good start, but inside the cabinet the driver then does "whatever the hell it pleases", as the saying goes. There is no other way but to build, test, read room acoustics, fix, test again, and so on. The shape of the enclosure also has a marked influence on the result, starting from an "ideal" sphere (difficult to build), to a cylindrical one, to an oval or rounded corners box, etc. One way of combining some of the bass reflex with a sealed cabinet properties is a transmission line (1/4 back wave-length). Of the three I built, none let me down. There's a lot to be liked about a well deigned TL, plus you get rigiid, well-braced panels "for free". For me it's a battle between sealed and TL in terms of audio quality. Some say sealed boxes are simpler to build, but when you consider good inner bracing, etc. it's not much so. Currently, the speakers I build are exclusively self-driven (class D) and DSP.

Apologies for the lengthy reply.

 

[mk1981]

Don't apologize, it's a great reply.

I agree none of them is outstanding, which is why I'm asking if they're very far off from a renowned driver like the Ultimax line. I guess they're fine for the price.

I was exactly thinking that it would be reasonable to use two per enclosure and connect them in parallel. Why do you say I should use two class D amplifiers, though? My thought was to use one per two woofers connected in parallel to one amplifier.

I was also thinking of three drivers per enclosure arranged in a hexagonal base box (or a truncated triangle), and the only thing that's putting me off from this is difficulty in working the materials for a box like this (a cube is easy, any rectangular box is easy, any oblique angle makes it way more difficult).

I did think of four drivers, and the reason I'm wary is that I don't have enough room to place them away from the walls.

 

[amail]

>I agree none of them is outstanding, which is why I'm asking if they're very far off from a renowned driver like the Ultimax line. I guess they're fine for the price.

Keep those cheap drivers from their max excursion and max perforformance as far as possible, and you are likely to be okish. Achieving that means using multiple drivers.

> Why do you say I should use two class D amplifiers, though? My thought was to use one per two woofers connected in parallel to one amplifier.

You can use just one stereo class D instead of two mono ones. The reason is that xovering them BEFORE the power amp is much much better than horrible passive xovers POST amplification. But if you have a good spare power amp lying about, you could use that for the woofers.

> I was also thinking of three drivers per enclosure

How would you connect them?

> arranged in a hexagonal base box (or a truncated triangle), and the only thing that's putting me off from this is difficulty in working the materials for a box like this (a cube is easy, any rectangular box is easy, any oblique angle makes it way more difficult).

- True.

I did think of four drivers, and the reason I'm wary is that I don't have enough room to place them away from the walls.

- You can still use 4 units AND also place the woofer against the wall, providing that the 4th woofer fires downwards, about 3-4" (c. 7-10cm) from the floor and you don't have thick, or hopefully any, carpets.... :-)

 

[Umikuser]

Having a single driver per enclosure and more enclosures nets you more subwoofers, which is beneficial (as per your links above).
A cube-ish shaped box is about as simple as it gets to put together (braces can be as simple as "a stick" cut to length glued between in the inner wall. You can have the parts pre-cut and just glue/screw them together. Only way to make it simpler is asking someone else to do it for you. The only tricky thing about it is making the holes, but that is the same for all types of enclosure.

The issue with enclosures vibrating is very much exaggerated in a subwoofer scenario. As long as the enclosure is decently rigid, it is orders of magniture lower than the actual output of the subwoofer. Rattling windows or furniture will probably annoy you more ;-)

Issues with cone break up or cone shape is also a lot less of an issue for a subwoofer. At higher frequencies however it becomes very relevant. And most importantly, at the price range you are looking at, the materials used, tolerances etc won't be good. So more subwoofers diven less hard will most likely get you better performance.

A lot less things can go wrong with a closed enclosure. Both construction and doing the math

 

[mk1981]

I don't have an amplifier yet, I was thinking of getting a Behringer NX3000 (https://www.behringer.com/product.html?modelCode=P0CHS) which is able to feed 2×900 watts into a 4 Ω load, so if I get two, I can use them for four separate units.

More than four doesn't make sense because it's then a pain to time align everything, requires an extra mini DSP and so on.

Since the drivers are very affordable, I thought I would get eight, make four enclosures with two drivers each.

Since one driver is nominally 8 Ω, two of them in parallel will be 4 Ω and will be a good load for an amplifier, I think.

Or is that overkill, and simply getting four drivers and putting them in four enclosures is more than enough?