1st BFM build: 2 X Omni 15 TallBoy in 2-12 configuration

[aeolos]

Thanks 88h88! Your encouragement means a lot, no fooling. As I keep saying, a good community has assembled around Bill's work.

Of course Bill's plans, and his cutting jig too, get much of the credit. Nevertheless the builder has at least to try to cut things to the proper size, at the correct angle, and so on. Learned much about how to do that during this build thus far.

My wife and daughter have made clear that they are not interested in hearing this cabinet work at ordinary operating levels so I have not yet had a chance to crank it. The most powerful amp I've got at the moment is 60 watts RMS at 8Ω if memory serves, so won't be able really to crank it anyway, but will be able to run it much louder than in the bedroom (45 watts RMS at 8Ω max there), where the cab has run at maybe 15 watts max, for perhaps 30 seconds.

This cab is I guess nominally 4Ω, so perhaps there will be a few more watts available. If all goes well, I will drag the box outside tonight and run a few tests using AudioTool, then report the results here. This will not be the half-space test yet, have to go up to the front field for that, but should get more accurate results than in a bedroom.

Best
--aeolos

[aeolos]

Got clobbered with work of all kinds, so testing and completion of the one nearly complete Omni 15 Tallboy got pushed back for a while, and still are far from the top of the stack. But there are a few new data points (no photographs yet, but these will be forthcoming).

Early testing results:

1) the Selenium D220Ti horn driver is reversed in phase from the 3012LFs and Alpha 6 CBMRA.

2) the L-pads are connected backwards , so "minimum" is maximum and vice versa.

3) the 3012LFs leak like madness between the factory gasket and the baffle.

a) a 10 Hz - 100 Hz sweep of 30 seconds' duration using the AudioTool generator running on a
Samsung Galaxy Note II, with amp output voltage of about 10V, results in clearly audible chuffing,
particularly at or below 40 Hz. "A mighty rushing wind" indeed .

4) under these conditions, output SPL at 1 meter, driven with pink noise at 2.0 average volts, is 103dB.

a) 2.0 V at 4 Ω is 1 watt if my understanding is correct.

b) the Dayton Audio IMM-6 test microphone appears to be damaged, or something: at SPLs above
about 81 dB, a sinusoidal wave form displays in AudioTool running on a Samsung Galaxy Note II
with an inverted peak, sort of a bactrian-camel-double-hump effect. The mic arrived in a dented
box, and the little screen over the diaphragm appears to have been displaced, so before
conducting any more testing I will seek a replacement.

Even with all these errors and problems, the cab yields promising results. Will correct the known issues and re-test, reporting the results here.

More soon, and all the best
--aeolos

[aeolos]

Got a chance to devote a few cycles to working on Box 1 of 2 of the Omni 15 Tallboy in the 2-12" Configuration Project. And what a project it has been, wow. The product is starting to come together at last.

Fixed most of the errors and omissions noted in the previous post. This cabinet most definitely develops significant air pressure when driven at say 10V at 40 Hz: to prove that, just stick your hand into 1 of the 2 ports created by the restrictor plate. Quite the huff and puff, and while it may not blow the house down, it will sure enough rattle the walls and the things hanging on the walls. And that's with an amp of about 50 watts per channel. Can't wait to hear what it will do when adequately powered.

The Selenium D220Ti compression driver in use for high frequencies is wired "backwards," by the way, like many others. When signal hot is connected to the tab with the red dot, the driver is in reverse phase from the 3012LFs and the Alpha6A-CBMR. Having now tried it both ways, I like it better with all drivers in phase.

Does anyone have suggestions for the torque to be applied to the mounting bolts for the 3012LFs? I've been reluctant to apply too much gusto, but would sure like to get them to stop leaking completely. Right now leakage has been greatly reduced, but not eliminated. But the reduction has been enough to allow me to hear the HC23-25 horn leaking a little bit, so that's the next candidate for another layer of gasket tape. The rest of the box seems tight, at least to my ear. Maybe I'll get some of that tar-like stuff, have been relying on neoprene gasket tape to this point.

At this point the cab sounds very good, but I think there is improvement still to be made, primarily through fixing the remaining leaks. Once that has been done, and I've gotten the replacement IMM-6 mic, I'll re-run the tests and report the results here.

As promised, if anyone is still interested, here are some photographs of the inside of Box 1 of 2.

Omni 15 Tallboy in the 2-12" configuration at altitude :

Omni15TallboyOpenBackInOrbit1.jpg

OK OK here's just the box:

UntitledOmni15TallboyInteriorFull1.jpg

And here is the low-pass filter connected to the woofers:

Omni15TallboyLowPassFilterWithWoofers1.jpg

And here is the midrange bandpass filter with the midrange, kinda hard to see the filter board:

Omni15TallboyBandPassFilterWithMidrange1.jpg

The high pass filter:

Omni15TallboyHighPassFilter1.jpg

And for completeness, the next post will feature the kindly plain inside of the removable back access panel.

All the best
--aeolos

[aeolos]

Here's the inside of the back access panel, for completeness:

Omni15TallboyBackAccessPanel1.jpg

Crude, but one hopes effective .

Question: is this box overpadded?

Thanks, and all the best
--aeolos

[aeolos]

Here's the "tar-like substance" I finally used to fully seal the drivers and horn where they meet their mounting boards:

speakerseal1.jpg

This is nasty sticky stuff but it creates a great seal. Here's what it looks like before the driver is mounted:

speakersealondrivergasket1.jpg

When the driver is mounted the material squishes out, and care must be taken to make sure it doesn't interfere with the surround.

Both 3012LFs, the Alpha6A-CBMR, and the Selenium HC23-25 horn got this treatment, and there is no detectable leakage at all around the drivers now, nor anywhere else around the cabinet.

At low frequencies (40-50 Hz) and middlin' power levels (10 - 12V), a rushing noise is still audible if one's ear is close to the LF horn. This seems to be port noise: the sound of the air "turning the corner" between the interior of the cabinet and the horn, and there is a lot of air moving in that neighborhood. The restrictor plate no doubt contributes to this noise by making 2 small ports instead of 1 big one, and by reducing the overall port area. Soon I'll take the plate out to see what difference that makes.

The sound may also be made louder by the foam padding attached to the removable back panel, and someday maybe I'll try removing some of that foam to find out. But not today. At 3 - 4 meters from the cabinet, when the cabinet is reproducing music instead of a pure sine wave, the noise is inaudible. It might be worthwhile to rout that edge of the baffle board to a bullnose as well.

Next: first photographs of the completed Box 1 of 2.

Best
--aeolos

[aeolos]

Box 1 of 2 is complete!

101 pounds (45.9 kg) of robust gear. It's always good when the product meets or exceeds the vision for it. I think we can safely say that this compact speaker system outperforms a legacy JBL 4663 (based on a Perkins bin), at significantly lower weight, and in a significantly smaller space, than the JBL.

And it looks exactly like what I had in mind from before the beginning, when I hadn't yet heard of Bill Fitzmaurice or this community.

Now the following pictures were taken inside in low light conditions and are not particularly good, but they show the current state of the project. It's cold here, at least to me, and a few minutes spent trying to get photographs outdoors convinced me to move inside. The grille is black Mellotone cloth.

box1of2complete1.jpg

box1of2complete2.jpg

box1of2complete3.jpg

box1of2complete4.jpg

box1of2complete5.jpg

That's my old P-bass, serial number 635833. Still have the original hardware, but here you see it with a Badass II bridge, Schaller machines, and a brass nut. Back in the day it was a serviceable workingman's bass guitar.

The replacement test microphone has arrived, although it's too darn cold to test outdoors. Results absolutely as soon as I can generate them, but again, this box sounds good. I like it .

Best
--aeolos

[aeolos]

And until the test results are available, the current State of the Project (with lots of lens distortion ):

bothboxes1.jpg

That's Box 2 of 2 on the left .

When funds and time are once again available, Box 2 will be completed, with all the benefit of the learning from finishing Box 1. Should go a lot faster.

And now off to the doc for a nice shot of steroids in the neck. Fell on the escalator at a trade show and now my left hand is partially paralyzed, as a result of nerve compression in the neck. Hard to play when your left hand won't open. Swell. Hope I get my hand back. Think good thoughts y'all, if you will.

Thanks all
--aeolos

[88h88]

Hope your hand gets better quicksharp if only so you can create another beautiful cab. The finished one looks better than a lot of professional manufacturers can put out...

[Tom Smit]

Good looking cab! Kudos.
Hope that healing will take place, soon.

[Bill Fitzmaurice]

At 3 - 4 meters from the cabinet, when the cabinet is reproducing music instead of a pure sine wave, the noise is inaudible.

When designing a cab one must consider port noise, but one must also consider the program material, which is never sine waves.

It might be worthwhile to rout that edge of the baffle board to a bullnose as well.

Only if the bullnose radius is roughly equal to or more than 1/4 wavelength. With 12mm construction that radius is 6mm, 1/4 wavelength at about 13.5kHz.

[aeolos]

88h88 and Tom, I really appreciate your good words and wishes. Hope the shot does some good: it did knock the pain down some, but the hand still doesn't work properly. The docs said that the relative success of the shot can only be evaluated after 2 weeks.

But that hand works well enough to allow continued work on Box 2 of 2, and there is plenty of plain old wood cutting, sanding, screwing, and gluing to be done before any more electronic parts are required. Everything necessary for construction is in hand (ahem) except for the Formica and some of the aluminum extrusion, so work will continue as time is available... AFTER testing of Box 1 of 2 is complete. Can't use a fretboard, but can use a table saw, with extreme care and full attention. Which are requirements anyway.

Bill, thanks for your expertise, and for getting back to me about the port noise. The thought was that the noise might be caused by vortex shedding as air moves back and forth across the edge of the baffle board, and that smoothing the path might help. Your comments sent me looking for more information about vortex shedding and Karman streets and all that sort of thing. Still reading and thinking, but you are certainly right about the relatively high frequency (13.5 KHz) of the noise arising from a rigid 6 mm radius half cylinder. The noise I'm hearing seems to be random, more or less uniformly distributed in frequency across a range from perhaps 200 Hz to 8 KHz, and is audible only when the signal is a low-frequency sine wave and the listener is close to the horn. As you point out, a sine wave is very boring program material and is rarely encountered in practice. To me the Omni 15 Tallboy is a surprisingly effective design, given its apparent simplicity. Would you call the LF section a tapped horn, or something else?

Have worked out an agreement with the wife and daughter that they will go out Saturday morning and do some roller skating, so I can finally do some testing with (I hope) an un-busted microphone and sufficient amplifier power. Obtained a Behringer (yes, I know) iNUKE NU6000DSP that I hope will, used with proper precautions, allow me to drive this cab at its designed input power.

This will be the first time that I have ever used a 6,000 watt power amp . If I am reading it correctly, the Behringer Web brochure states that at "1/8 Rated Power" this unit draws 620 watts of input power at 120 VAC. So to the linear thinker, that would mean it draws 8 * 620 W at full power, or 4,960 watts. This then gives us a current draw of 41.3 A at 120 VAC. The brochure says that the required electrical service is 25 A at 120 VAC. These are certainly interesting numbers. Does this unit have a Mr. Fusion in it or something? Where does the free power come from? Or am I being too simple-minded with this analysis?

But it's cheap and has gotten many good customer reviews around the Web, along with some really bad reviews to be sure. I don't think I have really heard this cab work yet, given the max available amplifier power until now has been perhaps 60 watts.

Thanks once again for tolerating my verbosity, and all the best, with test results coming Real Soon Now if all goes well
--aeolos

[Bill Fitzmaurice]

To me the Omni 15 Tallboy is a surprisingly effective design, given its apparent simplicity. Would you call the LF section a tapped horn, or something else?

Where its path length is 1/4 wavelength or more the flared port functions as a horn. But since the port is short horn loading of the woofer only occurs above 200Hz, and mainly allows the woofer sensitivity to match up with that of the midhorn at the crossover. It doesn't work as a horn at all for the woofer rear wave, as passage of anything above roughly 200Hz through the flared port is prevented by the damping. If that wasn't done then the midrange frequencies passing through the port would mix with those from the front wave at various degrees of phase, and that would cause massive peaks and dips in the response.

[Grant Bunter]

This will be the first time that I have ever used a 6,000 watt power amp . If I am reading it correctly, the Behringer Web brochure states that at "1/8 Rated Power" this unit draws 620 watts of input power at 120 VAC. So to the linear thinker, that would mean it draws 8 * 620 W at full power, or 4,960 watts. This then gives us a current draw of 41.3 A at 120 VAC. The brochure says that the required electrical service is 25 A at 120 VAC. These are certainly interesting numbers. Does this unit have a Mr. Fusion in it or something? Where does the free power come from? Or am I being too simple-minded with this analysis?

My understanding is that the switch mode power supply is the answer.
Altering Power Factor correction in SMPS's increases efficiency and output.

A bit of further research suggests the upper limit mandated draw is 80% of the receptacle, eg 12A in a 15A circuit.

I thought this was an interesting read:
http://us.tdk-lambda.com/ftp/other/pfc_ ... pplies.pdf

Smoke and mirrors:
Behringer claims that ouptut of the inuke 6000 is 2 x 3000W/4ohms, hence the 6000

I'm pretty sure that's 3000W as peak, not average RMS...

[Bill Fitzmaurice]

Smoke and mirrors:
Behringer claims that ouptut of the inuke 6000 is 2 x 3000W/4ohms, hence the 6000
I'm pretty sure that's 3000W as peak, not average RMS...

With sufficient capacitor storage capacity you can get 10kW or more from a 15A wall outlet.

But not for long.
Continuous output is what really matters. You'll never find that spec from Behringer, but to be fair, not from many others either.

[aeolos]

Thanks Grant and Bill, for helping me better understand how Behringer can claim that the INUKE 6000 DSP can produce 6000 watts.

And Bill, thanks especially for your explanation of the dynamics of the Omni 15 Tallboy cabinet. I would be most grateful to learn more about the thinking that went into the design. If memory serves, you have written elsewhere about "signal shaping" being one of the design drivers.

Now that I have that Behringer amp out of the box and have used it briefly, it gives me the subjective impression of being close in power to an old-school Class A stereo power amp of perhaps 300-400 watts RMS into 8 ohms, 20Hz-20KHz, both channels driven, continuous operation. But that impression is based on brief use. It's a good strong amp, assuming that it doesn't crap out quickly, but 6000 watts just ain't in it, unless you think "instantaneous peak power" is a useful metric.

Here's a link to a .jpg of the annotated SPL chart generated today, from data acquired Saturday afternoon:


Sorry for the weird ads at the bottom, but that site is the first free, anonymous, and working picture-posting site found today. No doubt the link will be broken shortly, but at least some people will be able to see it, for a short while.

These results were obtained with the restrictor plate installed. Next step is to take the restrictor plate out and re-measure. It's getting easier to run this measurement procedure as time goes on and I learn the drill.

Here's a link to the SPL chart, with Bill's published Omni 15 curve overlaid on the curve obtained from Box 1 of 2:


Please note that the placement and scaling of Bill's curve are best guesses.

And here is a screen shot of the Behringer DSP-controller software showing the hasty-hack filter settings that allow Box 1 of 2 to rock and roll with the sound I am seeking (not connected to the amp here, but these are the settings):

inukeEQ1.jpg

Will be refining these settings as time goes on, but these are in the ballpark. "Meaty, beaty, big, and bouncy."

As Bill has noted, most people who build Omni 15's are looking for big bottom end. Me too.

More shortly, but wanted to finally pay off on the repeated promise to post test results.

All the best
--aeolos