or Connect
AVS › AVS Forum › Audio › DIY Speakers and Subs › 6th order bandpass hybrid passive radiator surround speaker
New Posts  All Forums:Forum Nav:

6th order bandpass hybrid passive radiator surround speaker

post #1 of 3
Thread Starter 
Yep, it's a mouthfull. But, things are going pretty well so far. A little background first...

My theater (see sig) will eventually be a 7.1 surround setup, with 3 of my unity horns (see sig) for the mains (plus low end woofer boxes, yet to be built), and full range surround speakers. This build pertains to the "4 and 5" channels, or the front surrounds.

Goals and constraints:

1) must be built into the walls with as low a profile as possible.

2) must reach 110 db+ at 1 meter before reaching excursion limits.

3) must maintain a reasonably flat response to at least 10k at listening positions, which are up to 45 degrees off axis from these speakers.

4) will be controlled by miniDSP crossover/EQ

5) must look cool!

So, I came up with an MTM + passive radiator design, which will have ported front plates covering the two woofers. These ported woofers plus the passive radiator will create a dual bass reflex alignment. Right now, it looks pretty boring, but once I have it built into the walls with curved baffle sides and drywall shaping, I think it will command a double take.

Originally, I had planned on a 3 way design, but I decided that I didn't want the expense of another decent driver, nor the added amp channel that this active crossover design would require. That leaves two way variants. At first, I thought that I could get away with a ported low end to the Ushers, but the allure of a passive radiator and no port "mid bleed through" won me over. So, off to WinISD. I chose the Dayton 12" passive radiator for this build:

http://www.parts-express.com/pe/show...number=295-496

I already had 4 Usher 8137 drivers (2 for each speaker), so I chose to use them. Zaph's data shows that they have very low distortion to around 1k hz.

http://www.zaphaudio.com/8test/compare.html

This, coupled with their relatively low profile and highish xmax (5mm) fit the bill for me. However, to take full advantage of these drivers' distortion profile, I need to cross no higher than 1k hz. Hmm... off to Zaph's site again. This time, I stole his Dayton 8" waveguide plus SB acoustics SB29DC idea from his blog:

http://www.zaphaudio.com/blog.html see Feb 5, 2010

This hopefully will provide both the low crossover point as well as the constant directivity of high end response that I need. I will shape the response curve of the horn loaded tweeter with the miniDSP. Here are pics of what I have so far: pic #1 shows the ground plane layout, and pic #2 shows the two "leaves" of the box apart.

John
LL
LL
post #2 of 3
Thread Starter 
Graphs #1 and #2 show the ground plane measurement of the woofers/passive radiator combo with no box stuffing, taken at around 1 meter with a radio shack SPL meter, C-weighted (SPL not calibrated). You can see the big dip at 114hz, which persisted off axis... panic time! Is this a "reverse null" generated by woofer-passive radiator phase cancellation? Then it hit me: that notch looks kind of like a passive radiator notch that one normally sees, which is associated with its resonance. Except that the resonance of this PR in this box should be WAY lower than this, according to WinISD. Hmm...

Graph #3 shows the close mic responses (an inch or two from each woofer cone and the PR) combined with the system response. Notice that the dip persists at all sites. So, I thought maybe it's caused by a box resonance or an internal standing wave, since I would expect differing peaks/nulls at the various mic locations if this problem were indeed phase related... time to add some stuffing.

Graphs #4 and 5 show what happens to that null with the progressive addition of stuffing... problem solved! As you can see, the low end rolloff gets more and more gentle as stuffing is added, corresponding to decreasing box Q. Also, note how the Q of the peak decreases in the "half stuffed" box compared to that seen in the box with no stuffing.

More to come, stay tuned...

John
LL
LL
LL
LL
LL
post #3 of 3
Thread Starter 
Update: swapped out the round waveguide for a Seos-12/DNA 360 combo... much better all around. Thanks Erich!



Here are the graphs:

1) the Seos-12/DNA 360, taken at varying degrees vertically off axis from ground plane. That peak-dip combo likely comes from the cavities created by the two woofers and doesn't represent the smooth FR that the Seos-12/DNA 360 normally provide in a more traditional box. So, please don't use these FR plots as reference if you're considering buying from Erich! I should be able to EQ these out.

2) Usher 8137s in an MTM arrangement. Listening window is fairly tight, no good beyond 20 degrees vertical off axis. Since my seating will stay within this range, no problem.

3) Combined MTM plots. Looks like crossover point will be in the 900s after some tweaking.






John
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: DIY Speakers and Subs
AVS › AVS Forum › Audio › DIY Speakers and Subs › 6th order bandpass hybrid passive radiator surround speaker