Do we NEED subwoofers with full-range speakers, and how many?

I would answer 'yes' to the subject question. I have 2 subs in addition to a pair of Geddes Abbeys. One is located between the mains ~1/3 of the distance along the long wall (that wall is ~29 ft long and my mains are ~14 ft apart); the other is near a rear corner. The subs are custom 4th order bandpass designs with ports on top so they radiate approximately in the midplane of the room. The room is around 29x15 with a cathedral ceiling and is furnished as a normal living room with 2 couches and thick oriental rug. The front sub is pushed a bit at higher frequencies to fill in a trough I had at ~125Hz. This might have been audible coming from the rear sub (near the main seating area), but blends seamlessly as it is. The subs and mains "overlap" in the sense that there is no crossover (the mains play full range, although their response tails off below 100Hz due to their sealed design, and the bandpass subs are acoustically limited). I used a fair amount of equalization on the subs to get a nice response curve at the main listening position. I'm sure I could have gotten away with less, though, if I had used a third sub. I also added equalization on the mains in the 125-250 Hz range since the output of my subs falls off pretty rapidly over 100Hz. Phase adjustment didn't do much for me so I left them both in phase with the mains. I implemented a house curve so that the response increases gradually from 125 to 20Hz (up by about 5 db at the low end). I have played with this for a while and find that it sounds better for most music - as well as providing a greater sense of drama for movies, which is important for me since my system doubles for HT use. The in-room bass response is quite smooth and uniform (within a few db) across the main seating area (a couch) and I am very happy with the sound.

The velodyne 1812 looks great. I would love to have four in a room with the ML CLX Art. It would only set back back $60k(same price as a new MB 350 4matic). The velodyne however does not mix and match different six cones. It goes all the way up to 120hz. As you mihgt suspect the 10"woofer handles the upper bass and the 12" handles the lower. Here is a mini review:
The massive Velodyne Signature 1812 subwoofer tips the scales at almost 400 pounds. The driver complement, just as massive as the aluminum and wood cabinet, includes 12" and 18" drivers mated to two 1250W amplifiers. The system is actually the only two-way subwoofer I’m aware of (the 12-incher covering the upper bass while the big 15" driver handles the bottom-most frequencies), but that’s not the only unique feature. The Signature 1812 incorporates Velodyne’s Digital Drive room-correction system. This computer program analyzes the listening room and digitally corrects the Signature 1812 for flat response at the listening position. I’ve been impressed with the Digital Drive’s user interface, as I’ve stood in awe of its slick readout at trade shows. Hopefully this is an indication of the system’s at-home usefulness. The price of the Velodyne Signature 1812 is $15,000. (Street price of $9.222.00 http://www.elegantaudiovideo.com/pr..._medium=CPC&utm_content=&utm_campaign=1000562)

Brian, My subs are as follows, 1) 18" Maelsrom X driver in a 8 cu/ft sealed enclosure powered with 2400 watts crossed at 72Hz, 2) 15" Dayton Titanic Mark III driver in a 6.75 cu/ft sealed enclosure powered with 950 watts crossed at 72Hz, 3) 12" Peerless in a 3 cu/ft ported enclosure powered with 550 watts crossed at 70Hz and delayed 5.82 mS.

I dispute the need for the subs to all be towards the front of the room, I have one up front an two behind me. They are however low passed at 35 hz.

I agree most vigorously that you need to get the timing right.

For some reason (prior to a few comments in this thread) I seemed to be an only one on that point. I know earle will say 'there is no study or evidence to back that up' as we have disagreed on this many time before on various fora.

I find it absolutely critical that the impulse from all the subs align exactly with the impulse from the mains (which are not high passed in my case, they run full range as normal). It drives me absolutely insane if that condition is not met.

Like others here I use rather sophisticated gear to achieve this, in my case the DEQX, others have mentioned tact etc.

I have been trying to 'work out' just why I strenuously insist on correct timing when it seems most others get excellent results with earls multisub approach. I have no idea if this has any validity at all?? but I suspect that it may not matter in most systems that don't use a deqx (or tact???) because those systems may not be totally clean in the time or group delay area??

Dunno, but if all the other frequencies arrive willy nilly then what is a few more in the bass?? That kind of thought. In any case it seems to be very true that the advantages of smoother bass frequency response well outweighs any disadvantage of time delay? Or is it one of those 'you don't know till you hear it' things?

A week or so ago I had a bunch of guys here for our annual shindig. One of them (who's opinion I value very highly) preferred the system without the subs. Which was interesting for reasons I was already aware of, so we pulled out the mic and redid the time measurement (one of the things he wanted me to teach him to do anyway, so that was good actually...a bit hard to teach him via email!). I already knew this, but the subs were advanced from the mains so we redid the adjustment of the timing.

He completely changed his opinion about the subs being in use and now far preferred the system with the subs in. The point?? I can only remember one figure at this stage in time, but I do recall we adjusted on set of subs by 1.7 ms, about 1 3/4 feet. (the other would have been slightly different, as one is in front and the others behind remember).

1 3/4 feet is 'nothing' in subwoofer land supposedly, and completely ignored as having any bearing in the geddes multisub setup. Yet my friend could hear it which helped me 'know' I am not crazy in this regard. Always good to get that stamp on your forehead that says 'not insane' haha.

Put another way, I have found it is as critical as getting the mains equal distance from the LP as covered in another thread here. So much so that I can sit in my chair think to myself 'urgh, the bass is wrong' only to find my daughter has moved the chair eight inches or so.

Luckily I only listen to two channel!

To the original question, no I guess it is not essential that we have subs, many many are extremely happy with a two way setup for example. But I agree with an earlier comment that the bass is the foundation, and am very happy with what the subs bring to the sound. I am not talking low bass per se, indeed I limit mine to 25 hz (I don't like being shaken, I don't find shaking a musical experience but rather an amusement park experience..others may vary).

I have mentioned elsewhere that bass is a funny animal, it has effects not limited to the bass region. This will sound crazy, but I feel, that bass is some sort of carrier wave for the rest of the spectrum. What I mean by that is that I INCREASE the sense of bass envelopment, the rest of the spectrum is carried by that, so any increase in bass envelopment brings with it an increased envelopment everywhere...sounds now come form above, the side, below, behind. The complete immersion in sound is facilitated by bass.

very odd stuff, but MOST enjoyable.

It is interesting you say that. Perhaps I'll experiment some more with phase alignment.

It depends in my case. I have a dedicated 2 channel rig and an HT set-up that does both. For 2 channel as long as the speakers have a solid 25hz in room I would say no. For HT absolutely no matter how low the mains go if only for the LFE channel. I have 3 subs now a L/R and LFE. I am building a forth to give me a pair to handle the LFE in my HT set-up. I want a bit more headroom on more demanding LFE track movies. An example would be the cannons in Master and Commander.

Rob

Terryj

It is interesting that you have a sub in the back. I am no sure such would provide good fidelity without the use of something like the DEQX or the Tact.
Group delay is one of these parameters whose impact on music reproduction has not yet brought any consensus... I am not aware of any studies that shows the efffect of group delay ay frequencies below 500 Hz... If we were to consider the 3 subs of this e3xample as a system , what would be he group its group delay. I would surmise in the vicinity of 2~3 ms .. Would that audible, reliably under blind conditions? I sincerely doubt it. I could be wrong. I do however, understand the concern.
I have not found the reproduction wanting in my experiments rather the contrary...

Rob3606

The addition of subs, emphasis on subS, is to smooth the response in addition to extending it lower... My former speakers went down happily under 25 Hz with authority and room shuddering ease ... it is the fact that in the fact that in most normal rooms the modes are predominant and you endup with a very jagged response with peaks of up to 30 dB and nulls at various frequencies in the lower region of Audio... Room Acoustics treatment mitigates the problem, EQ somewhat ... What seems to work to tame room modes is the application of multiple LF sources... A reading of some works cited in this thread would give you the rational behind the use of multiple subwoofers ...

FrantzM said:

Terryj

It is interesting that you have a sub in the back. I am no sure such would provide good fidelity without the use of something like the DEQX or the Tact.
Group delay is one of these parameters whose impact on music reproduction has not yet brought any consensus... I am not aware of any studies that shows the efffect of group delay ay frequencies below 500 Hz... If we were to consider the 3 subs of this e3xample as a system , what would be he group its group delay. I would surmise in the vicinity of 2~3 ms .. Would that audible, reliably under blind conditions? I sincerely doubt it. I could be wrong. I do however, understand the concern.
I have not found the reproduction wanting in my experiments rather the contrary...

Click to expand...

hiya frantz, look don't 'hold' me tyo the group delay thing ok? as I said, it's just wondering aloud because 'I'm not supposed to be able to tell' if you follow, So WHY can I tell?? The deqx naturally handles group delay across the entire spectrum (not 'just' time aligning drivers) so maybe it becomes more important that the bass is too???

Actually, it is one sub at the front (18 maelstrom) and two at the back (four * 10 XLS peerless each). In any case, they are low passed very low (35hz) so maybe if I crossed them at say 80 then it could be problematic??

Hey, I'm a great believer in blind tests!! I have accidently done that BTW, my daughter was playing the system for a friend recently and she happened to move the chair which I didn't notice she had done, and when I sat down I was annoyed that the bass had 'somehow gone out of calibration' or whatever the correct term might be...that's when I glanced down and saw the chair had moved back a foot or so (ie a double whammy, it put me closer to the rears and further from the front)..the point is that for whatever reason I did notice quite quickly.

There is something else regarding this aspect as well, the effect the timing has on the FR itself can be quite large. I really should post some graphs to illustrate this point, after all the entire reason for multiple subs is to even out the FR! Any dips and bumps left over after the rpocess is normally 'fixed' with eq, but if some of the bumps and dips are TIME based rather than frequency based, then just how much is handled by eq???

In that case it is a bit like using a hammer rather than a screwdriver if you follow.

In any case, having them properly time aligned cannot actually hurt, and is free gain in output (FR wise) so if you can then why not?? No point in having all that amplifier and sub power to waste it by not time aligning them...tho the graphs would help illustrate this point.

terryj said:

hiya frantz, look don't 'hold' me tyo the group delay thing ok? as I said, it's just wondering aloud because 'I'm not supposed to be able to tell' if you follow, So WHY can I tell?? The deqx naturally handles group delay across the entire spectrum (not 'just' time aligning drivers) so maybe it becomes more important that the bass is too???

Actually, it is one sub at the front (18 maelstrom) and two at the back (four * 10 XLS peerless each). In any case, they are low passed very low (35hz) so maybe if I crossed them at say 80 then it could be problematic??

Hey, I'm a great believer in blind tests!! I have accidently done that BTW, my daughter was playing the system for a friend recently and she happened to move the chair which I didn't notice she had done, and when I sat down I was annoyed that the bass had 'somehow gone out of calibration' or whatever the correct term might be...that's when I glanced down and saw the chair had moved back a foot or so (ie a double whammy, it put me closer to the rears and further from the front)..the point is that for whatever reason I did notice quite quickly.

There is something else regarding this aspect as well, the effect the timing has on the FR itself can be quite large. I really should post some graphs to illustrate this point, after all the entire reason for multiple subs is to even out the FR! Any dips and bumps left over after the rpocess is normally 'fixed' with eq, but if some of the bumps and dips are TIME based rather than frequency based, then just how much is handled by eq???

In that case it is a bit like using a hammer rather than a screwdriver if you follow.

In any case, having them properly time aligned cannot actually hurt, and is free gain in output (FR wise) so if you can then why not?? No point in having all that amplifier and sub power to waste it by not time aligning them...tho the graphs would help illustrate this point.

Click to expand...

With wavelengths on the order of 10-50 ft what is the true significance of time alignment? The reflections off of the various walls in a small room will mix everything up at the listening positon anyway after less than one cycle. It takes more time than that for the ear/brain to register a specific tone. The ear can pick up differences of 100 ns or so in arrival times at midrange frequencies (or at least that is my subjective experience), but at bass frequencies it is a whole different ball game.

The effects you heard may simply be due to changes in the modal peaks and nulls at a specific position in the room with different time alignment of the sources.

Terry, we see eye to eye for sure on this topic, I also have 3 subs with one being 6 feet behind the LP which is a 6 foot difference from the 2 front subs (12') so I delay the rear sub 5.82mS. I think the last paragraph in your last post (#49) pretty much says it all, if you have the ability to do it with your equipment why wouldn't you?

Microstrip

I wrote a reply that somehow got lost in the ether … Will try it again here

You can use an external crossover. It has the potential to provide a better solution as the crossover in most powered subwoofer is limited in functionality. What needs to be understood abut this solution is that the mains whatever they are not crossed-over… IOW the full signal is sent to the mains as usual. If the mains are full range it actually is a better deal in this scheme. You are introducing more LF radiators/source and that is all good... So the mains should be left alone.
Part of the signal is sent to the subs… There is NO STEREO subs here what is sent to the subs is mono anyhow most subs can sum the bass from two channels.
The audiophile reflex would be to acquire the “best” crossover available … well it is not necessary to go there. This crossover will be operating only in the bass and what goes to the mains does not go through it, not at all. Some audiophiles call this arrangement “Augmentation”.

There exist on the Market some very good crossovers, Marchand, makes some, Lab-Gruppen makes these too. I am not sure you need any of these products. If you use TacT or DeQX you already have units that can take of this to the maximum but if you don’t …well no despair.
The unit I would recommend is one you may have seen mentioned quite a bit throughout this thread. The El Cheapo Behringer DXC 2496. It is almost overkill for the purpose, is virtually transparent in the range involved and inexpensive, very inexpensive, down right cheap by Audiophile Standards. It costs in the US the vast sum of $300 likely delivered. The unit can take 3 inputs and provides 6 (XLR) outputs. Each output can have its own crossover point slope, level and delay/phase. It can do much more but let’s limit ourselves to these for now. It can be controlled by a PC / RS-232. So it is likely you can have different configurations stored as a file that you can recall anytime … It is a professional unit and the level in the PRO world are higher than in the Residential. This can be compensated in the unit itself and if that is not sufficient there are countless of adapters from XLR to RCA, one of these available, I believe from Radio Shack. The slope of the crossovers can be set from 6 dB/octave to 48 dB/Octave and with Bessel, Butterworth and Linkwitz-Riley characteristics. You can set the delay thus matching the wave launch of the different subs to match arrival times. It takes a fair a mount of time to master this unit. Rather steep learning curve. You don’t need to master it really, once you can set Crossover point level and delay you are done on this side. You do need to understand how it works though to extract the maximum from it.

I know this is almost a heresy to talk about such products in the context of audiophile forums. This unit is pretty serious and has surprisingly good parts inside. There is a vibrant cottage industry centered on modifying it for audiophile purposes. The Emerald Physics CS series of speakers use it as part and parcel of the products, it is the crossover the speakers use (They also offer the option to use a crossover from a company named Spatial). These speakers have recently received a TAS Golden Ear award … I have NOT heard these but just to let you know the Behringer is not as lowly as its prices would seem to indicate in term of audiophile reflexes … Out of the box I believe it is good for Bass only and superior to any processor in any commercial subwoofer you want to dream of at any price (emphasis is on commercial sub)

You really nailed it there Frantz, that is exactly the set-up I use and I know this is taboo for audiophiles too but I use Behringer pro amplifiers to power my subs too, they work great and the cost is ridiculously low for the whole set-up. A DCX-2496 and 3 EP-4000 amplifiers for the total sum US $1250 delivered to the door.

Terry, what type and size is the enclosure you use with the Maelstrom?

Dan

dougsmith said:

With wavelengths on the order of 10-50 ft what is the true significance of time alignment? The reflections off of the various walls in a small room will mix everything up at the listening positon anyway after less than one cycle. It takes more time than that for the ear/brain to register a specific tone. The ear can pick up differences of 100 ns or so in arrival times at midrange frequencies (or at least that is my subjective experience), but at bass frequencies it is a whole different ball game.

The effects you heard may simply be due to changes in the modal peaks and nulls at a specific position in the room with different time alignment of the sources.

Click to expand...

yeah I know doug, it seems that I should not be able to pick it, as earle has mentioned many times to me 'there is no evidence' and your logic does seems impeccable, yet there ya go. That was why I was so happy to get that stamp on my forehead saying 'not insane' (it was too small a stamp to add 'at least on this one topic' haha).

And I completely get the idea of the amount of time it takes to recognise the frequency..I mean it has to at least (by definition) almost a cycle?? Praps we only need half a cycle but yeah, it takes time.

You could be right, maybe it is some change in FR (if that was what you meant)...well I always go and flatten the FR in the bass natch, so tho they will not be identical we are not talking more than a db difference between iterations. But what I 'hear' does not feel to me to be related to FR as such, but it is the THUMP (if you will). If they are out of time you get the distinct feel of THUMP thump rather than the single event.

Your point raised does back up the fleeting point I mentioned just before, that the effect of timing between the sources can be surprisingly large in the FR domain. Most would then go to the next step and EQ the response to get as flat (or the desired shape) in the bass. So let's say that we cannot hear what I think I can hear, yet on a theoretical level the waves from the bass sources would be in phase in the real event, so it makes sense (if you can) to do it in the reproduction.

Now this is the point that comes out of your possibility...THAT we then eq the result (if we use eq) it makes logical sense to eq the correct thing, ie the resultant of all the bass sources being in phase (ie time). As things in phase generally (always??) add, then that is 'free' bass output, it is what we are paying for with drivers, cabinets and amplification is it not?? so it makes logical sense to maximise all that even if it cannot be heard any other way.

My advice is don't sweat it if you cannot hear it, it's simply that it drives ME insane when they are not aligned. And I can, so I do. Who knows, maybe it IS all in my head, but at least it does not hurt to align them and gives me peace of mind.

I really should test this blind, it would not be hard to do...just for curiosities sake I guess. And I will post some graphs that illustrate the interaction between FR and time, at the very least I feel we should equalise the correct thing.

DWR said:

You really nailed it there Frantz, that is exactly the set-up I use and I know this is taboo for audiophiles too but I use Behringer pro amplifiers to power my subs too, they work great and the cost is ridiculously low for the whole set-up. A DCX-2496 and 3 EP-4000 amplifiers for the total sum US $1250 delivered to the door.

Terry, what type and size is the enclosure you use with the Maelstrom?

Dan

Click to expand...

Hi dan, first I'd just like to address something frantz mentioned. The dcx can be a PITA to work out and master, using the front panel will drive you nuts (and forget the manual, it presupposes a great deal of knowledge..it is after all meant for pros).

However, using the download and PC that frantz mentioned (a free download from the website btw) well I found that a breeze, and very intuitive. So I'd say most would very quickly master the unit that way...you CAN just download it and play with it without any unit connected to it, just to get the drift.

Dan, funny story with the mealstrom. Well to answer the question,it is sealed and in a box around 200 l at a guess. Hmm, because this MAY be of help to others at some stage I might bore the pants of you all and tell you the story...even tho it will make me look like a fool. Still, as long as we learn from our foolishness eh?

A mate has lent me the driver, I've had it for a few months now just sitting in the shed. Every year we have a big annual shindig here, and in a nutshell the two rear subs were not cutting the mustard. I listen very loudly at times, and have very capanle mains (they use 18 PHL as the bass units) so the subs need to be able to 'keep up'.

These didn't, they clicked and clacked and struggled at times. So a week before the gtg I bit the bullet and quickly whipped up a box for the maelstrom, and to make it easy for myself I cannabilized the old boxes I used to use for my mains, luckily I had been lazy cleaning up and they were still sitting in the shed...cut the front off one, the back off the other and join the two together. (they already had a cutout that would take an 18 remember).

Hence the uncertainty when I answered about the volume..it was literally something I threw together in a day or two.

Now for the lesson I learned. The same mate also lent me a pro amp, well a couple actually!! and I hooked up the 18 to that amp, no sound. grr. So I quickly hooked the sub back with the amp I ran the two trear subs off, a digital amp with around 200 watts per channel. (only use one channel of course)

Man, the maelstrom clicked and clacked and bottomed out!! Bloody hell??!! WTF!!??

Haha, then I remembered that the pro amp had a knob on the front...'I wonder if they are gain knobs"?? Yep, sure enough, and they were turned down.

See what I mean about looking an idiot?

Hook the 18 up to that, NO clacking, clanking, bottoming out! The lesson?? Insufficient amp power gave me sounds that I thought were mechanical noises.

I would have sworn on a stack of bibles that was the case, it wasn't, it was the AMP that was not cutting it. NOW I am wondering if the struggling I heard with the rears was ALSO amplifier caused! Aftert all, four ten inch drivers (with 12.5 mm excursion) should be the equivalent of a 21 inch driver, so it sounds hopeful as a theory.

None of this may be news to others here, it was new to me and I now know more than I did a few weeks ago.

FWIW.

FrantzM said:

The El Cheapo Behringer DXC 2496. It is almost overkill for the purpose, is virtually transparent in the range involved and inexpensive, very inexpensive, down right cheap by Audiophile Standards.

Click to expand...

Frantz,
Thanks - you are just saying what I wanted to listen! Yesterday I downloaded the specs of the DCX2496 and was just looking for a supplier.
Can you tell me if I can configure it as three channels low pass with different phase each?

Terry, Yep thats a good one and I think most of us here can come up with a story very close to that, I do some things sometimes that just leave me scratching my head and shaking my head at myself in disbelief hahaha. Now on the other hand you made me feel good because I have learned to navigate the DCX-2496 real well now, it took a while but I've got it now. Also yes not enough amp power can do more damage sometimes to drivers than too much power to a point. I use 2 15" Daytons with 18" passive radiators in sonotubes for my LFE subs which are separate from the 3 used with the Abbeys. They are tuned to 17Hz and use 950 watts each, I used them with no low frequecy limiting on them and they were fine until I played the new remake or A Nightmare on Elm Street there are passages in that movie that were bottoming the drivers out, so now I have a subsonic filter set at 17Hz on those.

Yes Microstrip you can set the phase for 3 channels separately for each channel, you can actually do it for up to 6 channels.

microstrip said:

Frantz,
Thanks - you are just saying what I wanted to listen! Yesterday I downloaded the specs of the DCX2496 and was just looking for a supplier.
Can you tell me if I can configure it as three channels low pass with different phase each?

Click to expand...

The answer is yes; and it is also yes for individual delays and slopes and crossover types and crossover frequencies and level for each of the 6 outputs

dougsmith said:

With wavelengths on the order of 10-50 ft what is the true significance of time alignment?

Click to expand...

That sentence basically summarizes the debate we have gotten ourselves into!

Where would one look (books, AES papers, websites) to understand the psychoacoustic models you are referring to? I don't know if there has been any conclusive studies done into the audibility of time delay / group delay / phase distortion at low frequencies