JBell PA Top

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I've collected JBell's statements on his PA top that looks very promising, gathered from his excellent TH sub thread

my CAD skill sucks so anyone want to draw plan for this, thank you very much.
======================================================
JBell said:
I can't remember if this was no eq, or if I had applied a db or 2, I know I have VERY limited eq applied in the dcx2496.

I originally went to drastically tame the 8khz piezo spike, but in the end, I didn't completely eliminate it. Leaving it let the system seem 'crisper' even though I know it should have been eq'd out. Also notice, the complete lack of the 500hz 'boxy' sound that you get with a 10" long mid-horn. It was breezy enough that day (from memory) that I was only looking at 350hz on up.

I wasn't using smaart for this (didn't have it on my laptop at the time) so this is Spectralab, creative usb mp3+, art preamp, and superlux ecm999 reference mic. Measurement in the stands, about 10 yards to the side of the press box, about 15 yards to the front of it.

Oh, and tops integrating with the TH... well, let's just say a big grin happens anytime I listen to this system

that's about all I can remember from a year back.
137069d1247245046-jbells-set-four-tapped-horns-outsidepaola.jpg


137106d1247280180-jbells-set-four-tapped-horns-picture2-101b.jpg


Did some playing around with hornresp, and it appears that the dual flare box I'd built awhile back, still seems to come out pretty good in 4pi space. I've learned the hard way, to ALWAYS use 4pi when designing a top. Also as you stack, the response evens out, and is not near this lumpy.

Anyway, if you are interested, your beta 10's would do very well in this style cabinet. Notice the bump at 250hz, instead of 500hz... If you want more details, I can provide them. (or maybe don will be kind enough to draw this up in cad, like he did for the TH.)

What I have simmed is a dual driver cabinet, 24" cube. If you stack 4 cabinets high, that'll be 8 drivers, and 132db@1M, and with some nearfield line array propagation, well, these should carry pretty well. If you stack 6 cabinets high with all 12 of your drivers, that'll get you 135. That's rock concert level a long ways from the stacks.

Attached is the hornresp sims, and a pic of a single driver test cabinet. I started cutting out ply tonight to make some dual driver cabinets for my park pa gig later in the summer.

cabinet details:
Dual 10" top.
build 24" cube shell.
horn path, two flares
13" long x height, 20degree at horn throat, 25degree at other end
8" long 45 degree at front of cabinet, square against joining 25degree angle
Where above two boards meet is approx 6" from front and 6" from side.
horn throat/driver mounting baffle, is 12" by height, with 3.5" wide x 8" hole, driver centered over hole.
Note, rounding top and bottom of driver opening slightly, instead of square hole sounds better.

Ports can be either deep in horn (even on driver baffle) or toward front of horn. If ports are 'back in horn' it lowers cabinet tuning.

Best sound with subs+tops is with ports tuned to 80-95hz with tops crossed at 100hz.
137145d1247368418-jbells-set-four-tapped-horns-top.jpg

Here's a pic of one I was playing with, using the port deep in the horn throat.

Yes, if you use 3.5K piezo's you get a big hole in response, right where you don't want it. If you use 1800hz piezo's, you don't.

138030d1248968293-jbells-set-four-tapped-horns-picture-143a.jpg


I know this design is kinda lacking at 100-150hz, but with the overlap between the subs/tops at 100, this seems to integrate well with the tops coming on strong at 200. Also, if you run your port tuning up to 100hz, you'll minimize your excursion in that range and you can push a little harder
 
I did search last night but didn't see an "official" HS12 thread, but rather some tidbits of info in some threads showing various pictures of progress, nothing like official build plans etc unless I missed it.

I also see a pic of some up top with a piezo array and the ones I found only talked about CD and horns.
 
HS12 INFO

There was no HS12 thread that I'm aware of, just tidbits here and there.
The Hs12 did use a CD and horn. I've gathered up some bits of info I found in no particular order and placed some pics and drawings at the bottom of this reply.

Per Jbell
"I don't really have a thread on any of my tops. They are just dime-a-dozen kind of design... They key on these is to find the right driver to make a horn/ported cabinet work. Once you find the driver, you have to play with hornresp for awhile to get a nice flat response in 4pi. (tops don't sit on the ground, so response curves tested that way are bogus in my mind)

Here's the best pic to show how my tops are built. The horn 'plates' are 13" long, and 45 degree angles on each end. It was designed to use partsexpress 15x22 metal speaker grills. I just bolt a 12" dayton waveguide to the metal grill and use a 16ohm selenium crossed at 2k. It uses an eminence beta 12a-2, and ports are tuned to 95hz. (sorry, not a fan of the cheap eminence compression driver....)

22x15 grill
try delta pro 12a
selenium 220ti-16 80w/40w
Goldwood GM-500PB
better-
Dayton Audio PA310-8
Selenium D2500Ti-Nd-16

I've had a couple questions about the HS12 tops that I've been using with the SS15 subs. The name HS12 is 'half sheet' 12" driver tops, as I can get a pair of them (barely) out of a single sheet of ply.

Attached are a couple pics, and any enterprising person can from these, reproduce my tops. I personally use eminence beta 12a-2's with selenium 16ohm 220i compression drivers. I use a simple 12db/oct 1900hz crossover with high C=2.6uf, L=2.68mh and low C=5.24, L=1.34mh. (at least that's the values the crossover calculator gave me.) I used parts express components that were as close to that as possible, the high capacitor needs to be a poly, and the high inductor needs to be air core, electrolytic and iron core is ok for woofer. If you are going to stack, the goldwood gm-500pb horn works, and for single cabinets I've used the dayton h12rw 12" conical waveguide. In both cases I just bolt the horn to the metal grill. The PE square hole grill is very strong and is actually what I designed this cabinet around.

Ports on the sides create feedback. Ports deep in the horn sound funky...
The baffle with the small 'fill' plates is easier/faster
The cabinet size is designed for a PE metal grill / and horn lens bolts to grill. (only use the square hole grill, round hole PE grill is too flimsy)
The horn is a low pass, and the eminence typical 'rising response' counteracts that. (horn low pass keeps you from having to eq out the midrange 'honk')
Any cabinet not 'midbass heavy' in 2pi sounds weak in 4pi (on stands) so yes, this is kinda a 'midbass heavy' design. You can see how even in single cabinet, this design is strong to 100hz.
A selenium 220ti-16 80w/40w sounds MUCH better than an ads1001
16ohm selenium matches up to the 8 ohm beta12a-2 almost perfectly, no resistors needed in crossover.
VERY LITTLE eq needed when used with the dayton 12" waveguide, and a nice open, non honky sound comes from compression driver.
1.9khz crossover means no 'fillers' needed, no need to get 12" to run up to 3.5k (selenium can go down to 1.5khz if needed)
95hz port tuning means this cabinet is good xmax down to 60hz if you 'have' to run no subs, and is strong down through 100hz. (which is important for male vocals)

So you're running a single Beta 12A-2 with the Selenium? I'm curious, because the Selenium is listed as 109dB / 2.83V (which is 0.5W into 16O), whereas the Eminence sims as 103dB in your horn.

Is it no problem for the amp to run a speaker with 8O woofer and 16O tweeter?

I read earlier you were using the Goldwood GM-500PB, now you mention a Dayton... is this one better?

This looks like an interesting top, if it can hit 123dB at 100W and can be built for under 200 bucks.
__________________
After that, I'd go selenium and goldwood if you are going to stack otops, or dayton waveguide if you are not. Both the goldwood and dayton benefit from a triangle shaped 'reflector' of sorts vertically in the middle of the horn to avoid a HF 'hotspot' when directly on axis.



You are right on the money on the selenium at 109 for the 8ohm, go to 16ohm and you are -6db..... perfect match for a 103db woofer section. Impedance swings wildly within frequency range of a speaker -- amp has no issues with it. I would still consider this a nominal 8 ohm cabinet.

Either horn works well, it's a matter of if you want to stack them or not. If you run the goldwood turned 90 degrees from 'normal' they do seem to integrate a little better than a stack of cabinets using the dayton waveguide. I also like the sound of the dayton waveguide a bit more. It has less overall gain, and a less 'honky' sound than the goldwood. If you use the goldwood, you actually need to eq it down just a bit, It's a bit on the bright side straight on. It's really a good 'long throw' option.

So horn choice is up to how you are going to use them. For me personally, I use the dayton in most of my applications. (where I'm running a single cabinet per side and 1 or 2 SS15 subs)

SS15 is good for 63v (500watts) and HS12 with the beta12a-2 top is good for 37v(180watts)
GC note... try delta pro 12a for high output

Originally Posted by k parker
Hi there i have been looking at this and other treads for a while now,i have fall in lov the SS15 [i have pic of SS15]and HS12, i have suggestion for the HS12 were the components is concern the Selenium D2500Ti has a better sound than D220Ti and what about the B&C ME10 H90+V60 horn it seems to have very good dispersion and the Kappa lite 3012HO for a over all higher output enclosure (just my two cents)

Hey, thanks for the thoughts. The hs12 components were chosen primarily for cost reasons, and yes there are many good upgrades that work well in that cabinet.

The dayton pa310 driver is a great upgrade and provides more 80-100hz that the beta12a-2, and gives basically the same power handling as the 3012ho at half the price.

Yes the neo 2500ti sounds better in the typical 40x90 horn than the 220ti, but the low expansion dayton waveguide seems to mitigate that. The 220ti/dayton is a reasonably nice sound.

This being said, my 'upgraded' drivers that I like for the HS12 are the pa310/2500ti. (keeping in mind the 2500ti is a bit louder than the 220ti)

If you want to make these cabinets lighter DL2512/2500ti puts it on about a 6.5lb diet.
SS15 is good for 63v (500watts) and HS12 with the beta12a-2 top is good for 37v(180watts) Your kappa+psd tops (if you build them) would be good for about 49v(300watts)
Ben, you are on the money on the design. Yes, I started with an OT12 style cabinet and fixed the things I didn't like.

Ports on the sides create feedback. Ports deep in the horn sound funky...
The baffle with the small 'fill' plates is easier/faster
The cabinet size is designed for a PE metal grill / and horn lens bolts to grill. (only use the square hole grill, round hole PE grill is too flimsy)
The horn is a low pass, and the eminence typical 'rising response' counteracts that. (horn low pass keeps you from having to eq out the midrange 'honk')
Any cabinet not 'midbass heavy' in 2pi sounds weak in 4pi (on stands) so yes, this is kinda a 'midbass heavy' design. You can see how even in single cabinet, this design is strong to 100hz.
A selenium 220 sounds MUCH better than an ads1001
16ohm selenium matches up to the 8 ohm beta12a-2 almost perfectly, no resistors needed in crossover.
VERY LITTLE eq needed when used with the dayton 12" waveguide, and a nice open, non honky sound comes from compression driver.
1.9khz crossover means no 'fillers' needed, no need to get 12" to run up to 3.5k (selenium can go down to 1.5khz if needed)
95hz port tuning means this cabinet is good xmax down to 60hz if you 'have' to run no subs, and is strong down through 100hz. (which is important for male vocals)
 

Attachments

  • HS12_1-1.png
    HS12_1-1.png
    72.3 KB · Views: 1,790
  • hs12 (1).JPG
    hs12 (1).JPG
    48.4 KB · Views: 593
  • hs12notes.jpg
    hs12notes.jpg
    138.8 KB · Views: 678
  • Picture 164a.JPG
    Picture 164a.JPG
    61.5 KB · Views: 613
I made the back plate removable... I made 2 versions... one just like the measurements specify and one that was 24 wide by 16 deep.. Woofer I used was from ebay.... Pro Audio 12" Pair DJ Replacement Woofer Speakers 1000W PP121 | eBay and they sound very decent... I have no way to measure results and you can't push them too much but other than that very very pleased with the sound.

I am waiting for my order from Parts Express, I ordered the dayton 12" suggested above...
 
According to jbell's post: "Ports deep in the horn sound funky"
Those ports seem fairly deep in there. Did you experiment with port placement?

Also, have you experimented with compression driver horn placement? One of its purposes is to prevent beaming of the woofer by mounting it in the center. I notice on your 2nd (from the bottom) enclosure you've got it mounted on the top. Is this due to mounting difficulty in the middle? Or something else?
 
I just placed them in the center of the panels, I read that but I guess I didn't think they were deep in there..

The top ones sound like the bottom one, so yes I am still experimenting...

These things get loud in that configuration, I had an officer show up in my home while testing them because he heard them from my back yard all the way to the street... it was fairly warm so I am almost sure he had his windows up... Also the speakers were parallel to the street... (Not facing the back of the house)

I think they sound good as they are, of course the best sounding one is the one with the bigger waveguide...

can beaming be measured?

I am still trying to learn the REW software and what all those things mean.
 
I just placed them in the center of the panels, I read that but I guess I didn't think they were deep in there..

can beaming be measured?

I am still trying to learn the REW software and what all those things mean.

Yeah, the ports are intended to be 2-3" from the front.

I assume beaming will show up as being heavy in some frequencies head-on and less so off axis.

REW gives me fits on my Mac (or I'm not doing it properly). Gonna give it a shot on the Windows machine soon.

On another note, I moved the horn & grill up to the very front, but the back on the back instead of inside, and there's still only 0.75" clearance at the woofer magnet. I wonder if the cabinet's depth measurement from jbell needs updating.
 
Beaming? It's only the directional curve at various frequencies.

For a small cabinet place it on a record player turntable, place a protractor on top and do a frequency response curve every five degreed or so. Then turn it on its side and do it all over again. While you're at it, change the pink noise generator for a fixed frequency oscillator, and do a series of polar curves like that.

For a big loudspeaker extend your protractor larger than the cabinet, find a railway shunting yard which shuts down at night, with a railway turntable…:D

Then repeat for two stacked cabinets, preferably with the mic at different heights.
 
Yeah, the ports are intended to be 2-3" from the front.

I assume beaming will show up as being heavy in some frequencies head-on and less so off axis.

REW gives me fits on my Mac (or I'm not doing it properly). Gonna give it a shot on the Windows machine soon.

On another note, I moved the horn & grill up to the very front, but the back on the back instead of inside, and there's still only 0.75" clearance at the woofer magnet. I wonder if the cabinet's depth measurement from jbell needs updating.


I did what I could with the information found online...

Well I am using REW with a Hackintosh (Mountain Lion on an HP) and I am still learning REW... I've only used it one day...

Yes I found that out also, I moved it up .5" for my first build, my second two are completely different dimensions... But they still come out of one 1/2 sheet, keeping with the name... I moved the speaker baffle up and left made the 45 degrees about 1.5 or 2" smaller....
 
Beaming? It's only the directional curve at various frequencies.

For a small cabinet place it on a record player turntable, place a protractor on top and do a frequency response curve every five degreed or so. Then turn it on its side and do it all over again. While you're at it, change the pink noise generator for a fixed frequency oscillator, and do a series of polar curves like that.

For a big loudspeaker extend your protractor larger than the cabinet, find a railway shunting yard which shuts down at night, with a railway turntable…:D

Then repeat for two stacked cabinets, preferably with the mic at different heights.

I will try that...
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.