[C.Popovich]

Hey guys, new to playing with OB speakers in general.

I am curious if anyone has experimented with using OB subs, particularly in a W frame, in the extreme near field to augment a traditional full range system? 

I'm toying with a behind-the-couch nearfield sub design using quite a few low distortion subwoofer drivers. 

These would be time aligned, equalized, and crossed over very low; likely attenuating output above 35-40 hz.

I'm probably coming at this from a different angle than many of you here, the open baffle appeals to me for transient purposes but in the extreme near field it seems like a way to get a driver to behave similarly to being in an infinite baffle; if sitting about a foot or so away from the cone is close enough to escape the cancellation issues, this could be a nice skirt around some traditional acoustics problems without loading the room to a ridiculous level. 

Thanks for your thoughts,
C.

[JohnR]

Not used with box speakers, but the nearfield dipole sub is totally worth it. You don't "escape the cancellation issues" but it's still a lot more output and (importantly) less mode excitation. A foot might actually be too close though, if you hear driver or aerodynamic noise it will ruin the effect.

Give it a go 

[Anonamemouse]

Have a peak here: http://daudio.nl/ontwerp/

It's in Dutch (Google translate is your friend), but if you email them they will answer your questions.

[Bumpy]

Have a peak here: http://daudio.nl/ontwerp/

It's in Dutch (Google translate is your friend), but if you email them they will answer your questions.

 Does anyone know what midrange driver they are using in that?

[JohnR]

C-Quenze. Judging by size, the 6.5" unit.

[gab]

Just hit the "English" tab at the top right for english

Appears they are using the woofers SL developed for the LX521:

http://seas.no/index.php?option=com_content&view=article&id=398:d1004-04-l26ro4y&catid=94:design-by-seas-woofers&Itemid=506

So this looks like a Linkwitz copy cat approach to me (woofers, active crossover they even call it an ASP, W-frame, etc.). But 17500,- euro / pair is wacky to me compared to what an LX521 can be built for. Oh well.....

www.linkwitzlab.com

Not sure what tweeter they are using

gab

[JohnR]

Mundorf AMT25D1.1. I don't see it as being particularly "copycat"... but do wonder why this discussion is not about nearfield OB subs 

[Bumpy]

Mundorf AMT25D1.1. I don't see it as being particularly "copycat"... but do wonder why this discussion is not about nearfield OB subs 

You are absolutely right. I hate it when others hijack my threads and here I am doing it. So sorry.

[C.Popovich]

No worries on the topic migration -it happens!

So, I procured a pile of inexpensive and low distortion woofers.  By a pile, I mean a dozen of the JBL CS1214's.  It's a pretty good (albiet ugly) driver.  Quick basic specs:
Fs 28hz
Qts .54
12.1mm Xmax (more like 15-18 range from some testing over on another forum)

Amplifier will be a Crown 2502, there will be two subs of 6 drivers each wired in a series/parallel arrangement to present two ~2.6 Ohm loads.  When all is said and done each driver will get the better part of 200 watts, which should be more than enough to get some noise out of them.   

I have a few issues that I have to work through still, and contingency plans.  I'll start with worst case; if this doesn't work out, I can just make some sealed enclosures and (if necessary) pick up another amp.  We all know that will work, especially in the near field, but that's not why I'm here in the OB section.  Armed with the knowledge that I can come up with a fun and impressive result with simple sealed boxes and some good eq (Minidsp/DIRAC), I'm having fun digging into the open baffle world.

I like Nelson's take on the slot loading; not only does it hide the hideously ugly driver faces, but you get force balancing (important with a dozen of the damn things doing work) and the idea that loading the driver can drop Fs and all the fun things that come from that sound AWESOME for near field listening.  I've heard some arguments about far field slot loading not being worth the hassle, but again, that doesn't apply here since I'll be seated 1-2' from the slot.

The first major issue I'll be working through is placement; current setup mandates that the couch be against the wall, or close.  That means this sub will be between the couch and wall, and there will be a high value placed on how shallow I can make it.  I'm aiming for 14-16" all in.  I'm simply not sure how this is going to work out using OB so close to the rear wall.  If the sub were a traditional boxed design, I'd count it as a plus and eq to taste.  What little objective measurements I've seen so far seem to support that I'll still be ok, but it's in the back of my mind as something to test before I get too far.  I'll test by making a simple U shaped enclosure from an existing sealed box (i.e. chop off the back) and measure before and after, mic in same location.  I picked up a cheap carpeted dual 12 subwoofer box for a whopping $11 so that will be the sacrificial lamb there. 

Secondly with current design I'll have two of them behind me; they can be pretty close to one another but I'm not sure if I will want them essentially acting as one baffled unit, or have a couple of feet between them.  If the latter, I don't want to inadvertently design myself into a null.  I'm guessing closer together will be more advantageous, but again, this will be easy to determine with some sweeps once it's all said and done.   

Finally I am unsure of how narrow to make the slot.  Yes, yes, I know, between 1/4 and 1/3 Sd (which is 85.4 sq in per woofer) is the standard, but I've seen some experiments with decreasing the volume in the slot and it paid positive dividends -significant bump in ~20hz output.  What to do here.... except TEST!  I'm an engineer by trade so I love the science, but nothing trumps test results. 

So order of operations:

1:  Pull couch out from wall.  Stick el-cheapo sub box (sealed) back there.  Perform sweeps.
2:  Chop back of el-cheap-o box off to make also-el-cheap-o U-Shaped OB.  Perform sweeps.  Move things around.  Hopefully learn.
3:  Build test box with some capability of adjusting slot width.  Test for impedance peaks, compare to freq response at listening positions.  I foresee this taking me a while.
4:  Based on all the above and some leaps of faith, start the real build.   

I'll take pics along the way. 

Chris

[matevana]

So order of operations:

1:  Pull couch out from wall.  Stick el-cheapo sub box (sealed) back there.  Perform sweeps.
2:  Chop back of el-cheap-o box off to make also-el-cheap-o U-Shaped OB.  Perform sweeps.  Move things around.  Hopefully learn.
3:  Build test box with some capability of adjusting slot width.  Test for impedance peaks, compare to freq response at listening positions.  I foresee this taking me a while.
4:  Based on all the above and some leaps of faith, start the real build.   

I'll take pics along the way. 

Chris

You might think about reversing the order of 2 & 3.  You can start by cutting a smallish slot in the back wall and then measuring.  Then just increase the size of the slot and continue to measure. Ultimately you will arrive at #2, but might save some time and effort.

[C.Popovich]

I'm not sure how equivalent that would be, but it sure is an easy thing to play around with and test -since the back of that speaker box is coming off either way, it makes sense to capitalize on the opportunity.  Good idea, thanks!

[JohnR]

To test the effect of near-wall placement on a dipole it would be better to use a dipole... How close to the wall are you thinking?

[gregfisk]

The rule of thumb is to keep an OB speaker 3' from the wall. Many people have used OB speakers in nearfield with good results.

You may want to touch base with the good people who hang out on the GR Research Circle right here on AC. There has been a lot of discussion through the years on much of what you are discussing here.

Greg

[C.Popovich]

Hey guys, thanks for the input!  I'll wander over into the GR research circle in a bit and see what I can learn.

I'm going to back up a notch and talk about my goals.  I am solidly in the category of basshead and that's likely going to seem excessive to some (most?) of you guys here, particularly when it's understood that I do not use the system for any home theater usage AND, when no one is looking, I often use a subharmonic synthesizer (mostly on older recordings).  I've spent a lot of time learning about target frequency response, studying the B&K work done in the 70's through Harmon's target curves of current times.  Attached is a graph of some standards (B&K, DIRAC, HARMON) and 2 of my system curves, the computer system is a little lighter in the bottom end due to me sitting 2.5' from the subwoofers and the limits of a dual 10" sealed system (even in nearfield).  It is easy for me to overdrive them during "spirited" listening sessions.   These curves are often tweaked, but #5 has been a standard for me for a while.  It's not perfect but I enjoy it on a wide range of music.

So you see, I'm going to need significant displacement capability down low to reach my goals, and I'm not even going to tell you my SPL goals, suffice to say "A" weighted it's loud, "C" weighted it's wandering towards excessive.  Again, I realize this makes me something of an outsider here, but I'm not married to any particular concept as much as I'm enjoying looking at/learning possibilities. 

[C.Popovich]

Yesterday (July 4th) I was mostly preoccupied with an unwelcome plumbing issue at the house, but after that was squared away I did a little experimentation with my beater-subwoofer.

Subwoofers:  Infinity 1262's (qty 2)
Alignment: Sealed, ~1.25' per driver
Amplifier: Crown 2502, no crossover used for these tests

As previously mentioned, I picked up the box for $11 so my plan was to chop the back of it off and directly compare sealed to what would be a poorly designed U shaped OB.  Someone mentioned trying small slots cut in the box, but this ended up making it behave like a poorly designed vented loudspeaker (expected).  I didn't bother to save the frequency response graphs there.

Below is a chart, sealed vs. open baffle, both 18" and 24" from the front face of the box, microphone approx seated ear height, box approx 2' from rear wall.  I used 1/6th octave smoothing

A few thoughts from the first part of my crude experiment, in no particular order:
-surprisingly similar output at a given signal level above 30hz.
-OB really does result in far smoother in-room response.  I didn't measure, but it was obvious walking around the room there were far fewer suckouts and peaks.  This doesn't matter to me nearfield but it's interesting none the less.
-Driver excursion is far higher for a given SPL, which is partly, but not totally, mitigated by the fact that driver excursion is also higher for every given watt.  Ultimate SPL capability will go down (obviously).
-The box the infinities came in was appropriately sized for a vehicle, with cabin gain, but far too small for home use without a ton of boost/very low crossover -so don't get too caught up in the freq response as much as the change in freq response.
-I now understand what some of you guys mean when you say you don't hear the box, you hear the driver.  This is a double edged sword, the infinities are a great value but clearly run into some trouble down low at significant excursion.  It makes me glad I have a dozen of the JBL's, I hope it's enough.
-The box I chopped the back off of is about 30" wide and 14" tall, so not a huge baffle profile.  I understand the limitations of this test, but it was still worth doing.

More to come soon.....

[JohnR]

The red/purple curves are with the back cut off?

[C.Popovich]

Yep!  Drops like a rock below 30hz, and excursion is definitely very high.  Seems these OB are not good bottom dwellers, but there may be hope in the nearfield yet. 

Chris

[JohnR]

Seems these OB are not good bottom dwellers

Right. Wrong choice of "enclosure" if that is the goal

I'd suggest making a W-frame out of cheap material to get more meaningful measurements. Or even better, make a folded baffle (like the Pass one) using 4 or 6 of them. In the latter case you don't even need to put a bottom or top on it 

[C.Popovich]

Update:  Canned the project.  System is impractical for infrasonic reproduction and extremely inefficient.  Put the dozen drivers in small sealed boxes, and Linkwitz Transformed them down to near 10hz.  FWIW I was wrong on crossover points; ended up going much lower than I thought I'd have to, 30hz @ 12 db/oct feels and measures well for the extreme near field and is not detectable as coming from behind you.  The more shallow roll-off was necessary not from a sound perspective but from a tactile perspective; these things have a LOT of punch and it feels odd if the couch doesn't shake at all on some frequencies then starts violently shaking a few notes below.  The shallow roll-off made that transition very smooth, natural, and pleasing vs. an extremely gimmicky/distracting feeling.  Near field tactile response is fantastic.

The OB idea was worth a shot either way, might work better in a bigger room with more space behind the listening position but just wasn't going to happen here.  Not a loss though, very happy with the sealed response and with a dozen of them a few thousand watts I can pretty much play anything I want at any volume I want.  That's a lot of heat dissipation capability!  I'd consider OB again for midbass, not likely for anything much below 40hz or so.  I have some extreme skepticism of some OB reports of frequency and SPL capability... maybe in a perfect room, and I'll leave it at that. 

Attached is pic of early on eq work avg freq response (of total system, not just the dozen JBL's) at listening position.  Headroom for days.  Will part your hair and give you a minor religious experience.

[JohnR]

System is impractical for infrasonic reproduction

Of course... I'm wondering why you wanted to use OB for this application... perhaps you are mixing up open baffle with infinite baffle?