Front and rear coaxial 8"s (from JBL 328CT) on a sideless enclosure as shown- I'm currently running it as a front/back dipole (coaxes out of phase front and rear) but it's designed to allow for a "dipole/bipole" configuration where cancellation will be more like a pair of traditional open baffles firing front and rear. They are set up on top of subs rather than on the floor (as shown).
Dipole cancellation's pretty straightforward, either the "inside" of drivers cancel and the front and rear have an effective baffle width of the baffle width plus 1" (Added waveguide effective depth) plus 7.5" (1/2 the total system depth between baffles)
When run as "bipole dipole" effective baffle width is simply Baffle Width + .75 (1/2 of the 1.5" thick baffle) +1" (Added Waveguide effective depth)
I haven't had a chance to measure in the current setup, but I'm currently crossing at 140hz, which is probably an overly aggressive high-pass (could probably be lower), but I'd love to hear any thoughts people have on such a setup.
The coaxes are normally ceiling mounted with a shallow waveguide, my waveguide is somewhat improved, in that it closely tracks the cone profile and is slightly larger, deeper, and smoother than the JBL ceiling mount metal waveguide. This is important as the horn leverages the cone as the waveguide so that transition must be as smooth as possible.
I'd love to hear any thoughts- this is an experimenter system, regardless of the (rather ridiculous) effort required to create the waveguide. Naturally I have my own ideas for what might work in this highly configurable system, but I figure this is the best place for such conversations.
Last edited by a moderator:
Here's the response sim from the edge based on the modified dimensions for the setup with the numbers as-run (drivers out of electrical phase). -3@ 100, and it's probably flatter than the "idealized" setup due to all the randomizing elements, but if this is accurate from a peak perspective, a simple notch at 400 could let me run a 100hz XO very comfortably.
I believe those are back-to-back dipoles, which can be used to form a "second-order gradient" (cardioid at low frequencies) system. See:
Subwoofer positioning, orientation and calibration for large-scale sound reinforcement (PDF Download Available)
Second Order Gradients
Subwoofer positioning, orientation and calibration for large-scale sound reinforcement (PDF Download Available)
Second Order Gradients
Thanks Charlie- I think that because of the symmetry, cardioid operation is not the case on them as-set, but could be achieved with some additional manipulation. I'll plan to take some measurements over the weekend and see what exactly is happening- the 100-500hz range is where it's critical for me to see what's going on- it's in heavy cancellation below 100 and the baffle acts much like an infinite baffle above.
I have been doing some research on the JBL 328CT, after seeing a speaker kit using it pop up:
Bookshelf Speaker Kit v1.0
– Canary Yellow
Like he says in the description, there is very little information about people using these in speaker design, so I was surprised to see your unique speaker pop up! I was thinking this could perhaps be an interesting driver for open-baffle use and you seem to have somewhat done that. It doesn't seem like I can find much of youe impressions on the speaker but if you still have them and have been listening to them, what are your thoughts? The 12 inch version also seems very interesting!
Did you ever try running this in a cardioid configuration?
Bookshelf Speaker Kit v1.0
– Canary Yellow
Like he says in the description, there is very little information about people using these in speaker design, so I was surprised to see your unique speaker pop up! I was thinking this could perhaps be an interesting driver for open-baffle use and you seem to have somewhat done that. It doesn't seem like I can find much of youe impressions on the speaker but if you still have them and have been listening to them, what are your thoughts? The 12 inch version also seems very interesting!
Did you ever try running this in a cardioid configuration?
Been thinking about this design for a long time though with a wider baffle. The thought is to run the drivers bipole and line the inside baffle faces with thick foam, enough to eliminate standing waves but not so thick as to impede bass exiting the sides. Fill the remaining space between foam with acoustic absorbent stuffing tuned to suppress high frequency backwave energy exiting the sides down to the frequency at which the drivers start becoming directional. My guess is the result would be an approximate figure 8 power response from the bass frequencies and up.
My results have been mixed. When the tweeter on these gives up... it gives up. I've separated the panels, and am using them as acoustic sources, the individual panels as OBs do a pretty decent job of creating a directional device from 3-500 up, but obviously need bass help. Right now I'm using them as "wings"- horn extenders. I'll be hooking up the drivers to the width and height channels from my pre-pro.
I’ve tried doing this before (along with bipole front-to-back), while the dipole version was better the freq. range well above the start of cancellation always seemed off. This wasn’t because of reflections but rather (I think) time-delay at higher freq.s because of the amount of physical off-set between the 2 sides relative to wavelength.
May the critiques begin. The top is a JBL 2374 1.5" horn, with BMS 4595HE coax on it, from a bit over 600 on up. Will probably try 800 sooner for a better balance to the large EAW MR102 horns. They're loaded with B&C 12MH32, from 150-600. Beneath that are the subs, 4x12" in 2 cabs, old NHT systems that fit the space nicely. Simple sealed alignment with a 50hz knee, but it couples to the room fine and needs no boost to hit 20. Just brute force by spreading the sources, but they are stereo and run to 150. The coaxes on OBs are tangentially aligned to the midbass horns, the idea simply being to limit the sidewall impact on the left in the asymmetrical basement room, while the right is wide open.
Behind the listening seat is a short gap to the rear wall, only about 2', but it's heavily treated with diffusor/absorbers. Just individual chairs for the moment. You can see those and the ceiling height surrounds in the linked panorama.
Ideally I'd rebuild everything but I move things around a lot- it's more like audio legos.
Behind the listening seat is a short gap to the rear wall, only about 2', but it's heavily treated with diffusor/absorbers. Just individual chairs for the moment. You can see those and the ceiling height surrounds in the linked panorama.
Ideally I'd rebuild everything but I move things around a lot- it's more like audio legos.