[JohninCR]

It's been over 2yrs since I built my last OB line arrays.  In the meantime I've been chasing dynamic OB bass output in as small a form as possible, but none of the folded forms got me where I wanted in my narrow 12' x 21' room.   I finally decided to go back to arrays, and these babies would leave a pair of Jamo's stuck in park.

They consist of 5 15's per side.  The center driver is a coax, the Hawthorne SI.  The next 2 drivers are Warrior 15's that I picked up at http://www.musicsupplycenter.com for the whopping price of $20/ea. The top and bottom are low bass drivers, the Hawthorne Augie.

I have them mounted on doubled up 18mm plywood with just enough side supports for rigidity, which I have placed against the side walls giving me as much boundary reinforcement as possible and to keep them out of the way.  The idea was to ensure dynamic bass for any type of music I throw at them by using a floor to ceiling bass array.  I roll the Augies off very low.  Then I let the Warriors run up higher on a 1st order filter.  Then I roll off the top and bottom off for the coax woofer leaving the tweeter in the center for the top end.

I was trying to obtain array behavior starting at the bottom with a gradual transition to point source for high frequency.  Not only did it work, but the results surprized me, even with, at this point, a guesstimated 4-way active cross.  Not only do I have way more bass than I need, but these 18.5"x86" monsters absolutely disappear.  In addition, I'm getting the sonic signature unique to arrays.  That is instead of bringing the performers into your room like a good point source setup can do, you are transported to the venue of the live performance with the sense of greater size and space, but in this case the sharp imaging and placements in the soundstage of a point source speaker are retained.  It's the best of both worlds with the only compromise compared to a full range array being the sweet spot is only slightly enlarged over a point source, but in my setup I don't need the "sweet area" instead of a sweet spot for two.

I put a couple of rubber feet on the bottom and they're held upright by bungie cords attached to a rafter a few inches above.  This decoupled connection keeps the woofers from transferring energy into the structure, and it give each speaker unit a resonant frequency of less than 1hz.  This combines with the woofers just idling along to very little baffle vibration.

I'm one happy camper, and I know the best is yet to come as I get things measured and really dial in the XO, but they're sounding great already.  aa

[Bob in St. Louis]

 



After all that, I need a


You kick ass John   

Bob

[Brad]

Very nice John.
No substitute for cubic inches......

[JohninCR]

Thanks guys.  While I can't argue with "there's no replacement for displacement", I'm using far less
displacement with this configuration than I have with other alignments that didn't get me to where I
wanted.  The drivers a just barely moving, so the +6db I'm picking up due to array behavior plus at
least another +9 or 10db from boundary reinforcement are huge benefits.

[scorpion]

Bungle in the Jungle, at last !   

The Warriors are sold out now after our recommendations, they I suppose were the bargain of the year.
What happens if you disconnect the upper two Warriors ?

/Erling

[opnly bafld]

Bungle in the Jungle, at last !   

The Warriors are sold out now after our recommendations, they I suppose were the bargain of the year.
What happens if you disconnect the upper two Warriors ?

/Erling

The first driver on the top and bottom are Hawthorne Augies, and I believe John is crossing these at lower frequencies than the other two bass drivers.

Lin

[playntheblues]

Hey John, I have a question for you or anyone that is more knowledgeable than me.  Most conventional line arrays have a row (top to bottom) of mid drivers and a row of tweeters or electrostatic panels.  With this in mind.......would a line array of 5 silver iris's work?  Most standard arrays use approximately 6 inch drivers and can get down to the mid twenty's or thirty's (htz).  I understand the whole MTM thing but if you use a traditional line array as an example you would think 5 Silver Iris's would work?  Your thoughts are greatly appreciated, Guy

[JoshK]

Guy,  the problem with that setup is the 31"+ center to center distance between successive tweeters in the Irises would create massive comb filtering which would cause all sorts of peakiness in the critical presence region. 

What John is doing is an extended MTM and getting a quasi line array like bass loading due to shallow slope xo's.  I don't know what slopes Dunlavy uses, but the concept is a bit of a take on that.   Of course he is using his walls to extend bass response and his are OB.

John, have you tried using thick wool felt (the industrial kind) for diffraction control?  You might try it, particularly on the side joining the wall, to control diffraction and thus widening your sweet spot.  There is quite a bit discussed on this topic on htguide's mission possible.  Mark's speaker page has some measurements.  Avalon acoustics does this integrated into their grills. 

[playntheblues]

Got it! thanks Josh

[scorpion]

Lin,

Yes, I see my mistake, but experimenting a bit and disconnecting two of the woofers at a time to see what happens
to the soundpicture could be a worthwile excercise. Say the top and bottom ones, and the two top ones.

/Erling

[Davey]

Josh,

The Dunlavy systems all used true, first-order acoustic crossovers with incorporated driver physical offsets and symmetrical layout.  All of them very nice designs.

This particular system by John isn't even a quasi-line-array.  There is no line-array aspect to it at all from his description.  The drivers are all mounted in-a-line with each other though.  I guess that makes it a "line-array" in John's world. 

Cheers,

Davey.

[playntheblues]

'I guess that makes it a "line-array" in John's world."  You need to put a smiley behind that or else it doesn't sound nice

[JohninCR]

Erling,
With the Augie amp off, I lose some bass, but I need to measure to see how it affects
array behavior.  It definitely changes the venue perception and sense of a large space,
bringing things back toward room size.  Despite the steep rolloff of the Augies, there is
still some higher frequency content having an effect.  I tried 2 woofers below the coax
before and it wasn't anything like this, which is more similar to the array sound without
the larger than life imaging.  The arrays I built before always gave the front row center
stage feel, but some vocals, eg Barry White, sounded like giants with the other instruments
unnaturally distant on the soundstage.  These are more like 10th row center stage with a
more natural image and placement.

Guy,
Yes 5 of the woofers with a tweeter only on the center one would work fine.  That tweet can
keep up just fine since it's sensitivity is 104db and can handle more power than your ears could
withstand at any reasonable distance.  Keep in mind that you'd need to make it a 4 way to avoid
the ill effects that Josh mentioned as much as possible.  I used 16.5" center spacing.  Closer would
be theoretically better as long as you don't have high ceilings, because you want the true infinite
length array behavior in the bass region.  If you research arrays, theory assumes the center of a
driver is the single point source of sound which isn't the case.  Arrays of 15's are in widespread use
for pro sound.

Josh,
Thanks for the felt idea.  I was going to purchase some carpet anyway and will buy extra felt padding.
I didn't flush mount the drivers and felt will be a perfect easy solution, even though diffraction isn't
an issue with large drivers as Rudolph proved for us in the NoBaffle thread.

Just so you know, there's no "quasi" about this as a bass array.  In the bass region, with such small
spacing compared to wavelength and a floor to ceiling line length it's an array.  It seems that I lucked
out once again with speaker building and pulled off a smooth transition from a line array for bass to a
point source in the treble.  I think the natural early rolloff of the Warriors compared to the extended coax
woofer helped to keep the XO simple by permitting a 1st order filter on the Warriors.  Measurements will
be interesting and really test how well my ears are trained.  Thank goodness for the ease and flexibility
of Thuneau's computer based filter program, so I don't have to learn the art of XOing from scratch.

[JohninCR]

Davey,

You need to do your research before spouting off.  This is the 6th OB line array I've
built, and have done quite a bit of research.  This is the first one designed for array
behavior in the bass region, and I purposefully wanted point source behavior above 1khz.
I'm not familiar with Dunlavy's systems, so I have no basis for comparison, but I'm sure
they're nice.  Arrays are great and as different sounding from normal speakers, as OB's
are different from boxes.  I seem to have pulled off a hybrid that exhibits some of the
advantages of both.

I have a bunch of small drivers, so I may play around with a narrow add-on to extend
array behavior up through the critical range.  Those are fun too, but with my narrow room
I don't really need the extra sweet spot width that full range arrays provide.

[Davey]

John,

Sorry to spout off.  I'm not a "line-array" expert like you, but I have read this document and a some of the references on the last page:

http://www.audiodiycentral.com/resource/pdf/nflawp.pdf

(We are talking about "line-arrays" for normal listening rooms and not gymnasiums, correct?)

It's my interpretation of this material that true "line-array" performance down to low frequencies is not possible in a typical domestic environment.  The wavelengths involved, complexity of the room and room modes are such that the optimal cylindrical wavefront associated with line-arrays at higher frequencies completely falls apart when the frequencies are much lower.  If the wavefront is not cylindrical with controlled vertical directivity do you really have a "line-array?"

Cheers,

Davey.

[JLM]

Davey,

Thanks for the link.  Wow at 2,000 Hz and nearfield in a setup like John's I'm gathering that he's recommending 10 tweeters at 4 inch centers per channel!  And realistic bass limitations in my 13 ft x 21 ft x 8 ft carpeted room of about 250 Hz (lower with wall vicinity and room size factors).  Again using the 2,000 Hz cutoff wavelength for woofer spacing would dictate (21) 4 inch woofers per channel for nearfield conditions in John's setup!


John,

Gee how could the bass not be adequate with ten 15 inch woofers?    

Dunlavy is out of business, but promoted wide wall setup with WMTMW vertical driver arrangements.

[JohninCR]

Davey,
While that's typically true, what is required in our acoustically small rooms is a floor to ceiling array
for the bass, so with the floor and ceiling boundaries the line appears to be of infinite length.  When
I haul mine to the beach and use them outside I'll have to boost the bottom because they won't
exhibit array behavior for any meaningful distance for frequencies below 100hz.  Dr. Griffin's paper
is a great resource, and I'm not contradicting anything he's saying there.  In fact, I'm hoping to get
his opinion on the approach I've used with these to transition from line array to point source behavior
in the opposite direction than is typical.  He's fairly active in the DIYaudio forum, so I maybe he'll see
my post there.


JLM,
4" ctc spacing for tweeters is way too far apart to avoid the comb filter effects, but as long as you have
a good number of drivers, comb filtering isn't much more than an early roll-off of the higher frequencies.

[AJinFLA]

It seems that I lucked out once again with speaker building and pulled off a smooth transition from a line array for bass to a point source in the treble.

Why don't you measure it to see how lucky you keep getting?
I didn't have any such luck with my coax/woofers. Farfield measurements looked a lot like the nearfield, with top to bottom propagation losses (and greater room influence). If your seated position sound is spectrally balanced, then the nearfield would need to be tilted upwards as you transition from line to point source behavior.
Suprise us John. Show us how easily your ears can hear line source behavior by confirming the measurements of the system. We may all learn something here.

cheers,

AJ

[JohninCR]

AJ,

Back at the time I meant to tell you that your arrays were too short to obtain much array
behavior.  Sorry it slipped my mind.

Please let me know what exactly it is that you're doubting.  I didn't say I could hear line
source behavior, although I have heard the effects quite clearly with other arrays.  I guess
I could turn one sideways to listen past the end of the array and verify array behavior with
my ears.  The science allows me to take for granted that these function as arrays
in the lower frequencies.  I do know that the transition is at a much lower frequency than
other single tweeter arrays I've built just based on the XO points.  My statement about
retaining some of the sonic attributes of arrays is a subjective analysis not likely proveable
through measurement.

Come up with something good, please.  A challenge may be just the thing for me to stop
procrastinating and turn into a measuring machine.  I have a lot of different things I want
to measure with a lot of different alignments

[AJinFLA]

I actually like what you have done, since that's pretty much the format I have preferred. I just wanted to add to Davey's comments about array behavior at LF in small acoustic spaces. I believe you have a point source, albeit a rather large one, depending on XO frequency/driver correlation. If I were to work with the same drivers, thats pretty much exactly how I would configure them.
Drag them outside to remove room influence and do some nearfield & farfield measurements. You'll see what I have already. The line would have to be rather large to establish a cylindrical front at LF.

cheers,

AJ