Considering a small OB expirement

[p0lar]

Hi all,

After listening to a pair of OB speakers (in a relatively large baffle, I admit) I'm considering an experiment with a relatively small OB design and was hoping to get a few ideas on what pitfalls I should expect aside from what I've already read about on the linkwitz site, amongst a few other threads on this forum.

I'm considering the Tang Band W4-657SH 4" Aluminum Driver assisted on the top end by an Aurasound NT1-204-8D 3/4" Titanium Dome Tweeter, probably crossed somewhere in the 4.5 KHz range.

I've done some research on OB, the primary part of which seems to indicate that the low end response will suffer. In the name of futility since I'm interested in the OB concept, and am willing to sacrifice some raw materials in the name of learning, how ugly is this going to end up being since I'm using relatively low Qts drivers with low Sd? I planned on adding a small subwoofer and the primary objective is more of a home theatre application. I figure if I really mess this up, I could probably convert the system into a small set of speakers for the bedroom or computer, or possibly even find a small set of enclosures to turn them into HT satellites.

I suppose my biggest concern is that I'll have to run the subwoofer at a much higher cutoff point and be limited by localization of the subwoofer with respect to the placement of the speakers. Will it be possible to create a baffle of approximately 8"x12" with 15 degree 'wings' of approximately 4" that is going to allow this driver to reach down to 120 Hz or is this entirely unrealistic? All my research seems to indicate that anything under an 8" driver is probably a pipe dream, but at the price of the components (<$35 plus crossover materials or pre-built?) I suppose I could flub the baffle and build a small-enclosure afterwards at only the time and expense of the baffle materials.

Your thoughts are most welcome prior to my foray into futility!

P.S. I've not dabbled in crossover design (yet?) but am electronically competent and could build nearly any of the circuits I've read about. At the moment, I'm considering a pre-made PE crossover but have my reservations.

[DougL]

Being a cheapskate myself, I tried an OB with 2 NSB, a Dayton ND20 and 2 generic buyout guitar woofers on an 18" by 46" baffle.

The NSB bottomed at a volume that we could still talk if the crossover was under about 200 hz.

I think that small driver and small baffle = limited volume capabilities.

Good luck with whatever you decide on.

Doug

[badman]

Your design won't work, to put it plainly.

You can't do a small open baffle with small drivers for any real purpose except nearfield listening. The baffle rolloff is going to be all the way through the midrange, you'll have basically a 2 way mid-tweeter.

You can add a sealed or vented woofer to match below, probably a 7" or thereabouts (though I like big woofers) and have a 3 way.

I'd go back to the drawing board, mate. Open baffles want to be big. Big drivers, big baffles.

[p0lar]

Originally posted by badman
Your design won't work, to put it plainly.

You can't do a small open baffle with small drivers for any real purpose except nearfield listening. The baffle rolloff is going to be all the way through the midrange, you'll have basically a 2 way mid-tweeter.

You can add a sealed or vented woofer to match below, probably a 7" or thereabouts (though I like big woofers) and have a 3 way.

I'd go back to the drawing board, mate. Open baffles want to be big. Big drivers, big baffles.

Yeah, this seems to be the consensus based on what I've read. I don't mind being wrong as long as I learn something; but, to know beforehand and still do it anyway.. well, that defines futile.

By my calculations, if I flub this, I'll essentially need some tiny enclosure that is approximately 0.15 ft.^3 with a 1.5"x4.5" port tuned to 75Hz, which will probably resonate like crazy. (I'm thinking computer multimedia systems here...)

Anyway, I'll try to stay on topic with the OB idea, despite how crazy it seems or the certainty of the failed experiment. Thanks for your input!

[mashaffer]

You can probably make this work but you will need helper woofers. In a small room smallish drivers can work if they have a high Qts but low Qts small driver needs to be treated as a wide range midrange.

I would be inclined to skip the tweeter and take care of the bottom end. That TB doesn't really need much help on the top end.

You would have to make the baffle about 14 to 15 inches across but something like this http://www.partsexpress.com/pe/showd...number=290-330 is pretty cheap and would give you an affordable look see. If you like then you can upgrade.

I did similar with a 10". For a picture of the raw baffle see the bottom of this thread.

http://www.hawthorneaudio.com/forums...p=14423#p14423

These have now become knick knack table speakers which I use daily. I am working on active crossover design right now but it works surprisingly well using the tone controls of the old receiver that is driving the woofers as a crossover.

mike

[gainphile]

I did many experimentation with "Small OB" to the point of obsession. Tried multiple variations like hybrid, etc. and it was a complete failure. There is simply no bass, and while listening to pure midrange is good, it's not so nice after awhile



My conclusion was there must be a minimum of 10" woofer which is equalised actively. This will give "decent" bass.

Add another 10" and you get good bass.

Also do a search on "Manzanita OB". It uses a tweeter and 12" woofer and looks small (ish). It's also passive.

[holdent]

I've been looking at the same thing and am going to use Martin J. King's U-frame idea here and pair it with a fullrange driver. The U-frame is only 17.5" x 17.5" x 8.25" (WxHxD). Mount a baffle above the U-frame for a fullrange driver and it should work out well.

Instead of using the Eminence Alpha 15A's MJK suggests, I have some cheap Goldwood GW-215s that I'm going to try. The driver is also 15" but has a higher Qts. In an earlier OB project I paired these drivers with HiVi B3Ns and was very happy with the results so I'll probably try B3Ns as well again. (Coincidently I also have a pair of Alphas I can try).

One concern I have is the crossover. I have an Behringer CX3400 (analogue) active crossover that only allows 4th-order Linkwitz-Riley filters. In MJK's example he used a 2-order Linkwitz-Riley low-pass filter which helped flatten the rising response of the U-frame. When using the CX3400 I have to set the low pass point lower to reduce the rising response problem. But now in modelling the the design I'm getting a dip in the freq. resp. at the crossover point because the B3N in a smaller baffle is rolling off early.

Other than increasing the baffle size for the B3N I wonder if anyone has any other suggestions to try? I'd like to avoid using the passive 2nd-order filter due to the cost of the components required (particularly the large inductors).

[p0lar]

Originally posted by gainphile
I did many experimentation with "Small OB" to the point of obsession. Tried multiple variations like hybrid, etc. and it was a complete failure. There is simply no bass, and while listening to pure midrange is good, it's not so nice after awhile

My conclusion was there must be a minimum of 10" woofer which is equalised actively. This will give "decent" bass.

Add another 10" and you get good bass.

Also do a search on "Manzanita OB". It uses a tweeter and 12" woofer and looks small (ish). It's also passive.

I actually went over your site with a fine-toothed comb the other night while concocting the idea - thanks for your insight!

It looks like that .15 ft.^3 enclosure is looking nicer by the post... and my driver selection is beyond futile.

If I stick with that selection of drivers, I might still give it a shot. Materials are cheap and I'd still like to go down in flames on a good effort. Unfortunately, a 10" or 12" 'speaker on a board' would get categorically denied based on WAF alone.

[gainphile]

Why not use 5.6" and 2" aura instead, and if you don't like it convert them into the Plutos

[Saurav]

I'd like to avoid using the passive 2nd-order filter due to the cost of the components required (particularly the large inductors).

You could try a passive line-level filter:

http://www.t-linespeakers.org/tech/f...ssiveHLxo.html

If you don't have a sub, you could also try using the high and low outputs of the 3400 to give you independent control of the two rolloff points. But I suspect you'll need an overlap, and I'm not sure the 3400 can do that (i.e. a low-mid XO that's higher than the mid-high XO).