SEOS-12 Synergy / Unity

BradH

Member
2019-07-13 10:35 pm
I've read a lot of Synergy threads but I still don't know what it takes to make one from scratch.

Aside from a pair of SEOS 12s, I have laying around:
Faital Pro 3FE25
Dayton DE250 clones (mylar)
Seas P21RE
blowout PE 6.5" polycones
3" Aurasounds, and a few other randoms.

I'm fine with buying new drivers. Keeping it on a budget would be nice. Cost of BB ply, CNC, and XO parts will get me if the drivers don't. (I live in an apartment, no saws or routers)

Here are my requirements / constraints:

-I don't need high output. I listen at low volume levels.

-Passive XO. I have miniDSP HDMI but I want to use the speakers as part of a home theater. So one channel each.

-Sealed bass cab. I have two Seas P21RE that deliver as much bass as I need. I'd like to use them. Was thinking of using miniDSP just to match the woofer and synergy levels while still having them occupy just one channel via passive XO.

-I use planar tweeters currently and really like them. I like what Patrick Bateman is doing with getting HF extension in a Unity. I'd like to have that, but if the cost of admission today is just going with the DE250s I have then I'm fine with that.

-It's gotta be easy on the ears. I find a lot of speakers, especially those with metal drivers, to be fatiguing and make me want to turn it off.

I have a umik for measuring.

I have seen Bill Waslo have good success doing this with the 15" SEOS. But I haven't seen it done with the 12".

Thanks
Brad
 
An idea that I've had kicking around my head, is to make a phase plug for the Tympany NE19 that would adapt it to one inch horns and waveguides.

The current phase plug turns a 3/4" dome tweeter into a 16mm compression driver. But I think if I made it bigger, the same concept should work fine in a one inch format.

LgU84Or.jpg


Here's a BMS 4550 on an 18Sound XT1086. Note that it gets really beamy above 13,500Hz, because the horn throat is one inch wide. (13,500Hz is one inch long.)

WdmF4uc.jpg


Here's one of my DIY compression drivers on "Celilo II", an asymmetrical waveguide with a 16mm throat. (21,250Hz is 16mm long.)

yu6eBtC.jpg

Here's the exact same tweeter, on an ATH4 waveguide. You can see that the high frequency response becomes unpredictable. Sometimes you get lucky and it just works, but sometimes you get a dip, or really bad performance off axis.

TLDR: DIY compression drivers seem to be a compelling alternative to domes and compression drivers. Basically the best of both worlds : the low frequency output that domes have and compression drivers lack, along with high frequency performance that crushes every one inch tweeter out there.

Now if I'd just finish the project so I can actually *listen* to the darn thing...

Data from The Preference for Direct Radiators and Celilo : A Small, Affordable Two-Way Unity Horn and DIY Compression Drivers.
 
In regards to the SEOS 12, I personally don't see a lot of good reasons to go with a three way Unity horn any longer. Theoretically, hi-fi midranges and midbasses shouldn't work on a Unity horn. But in reality, they seem to work fine. YMMV, but I've had no issues getting surprisingly large midranges and midbasses to work. I've gone as large as eight inches, I think Art Welter used something like twelves or fifteens.

The only real drawback is that the midrange taps need to be kinda enormous.

But that opens another can of worms, which is that there's two ways to place the midranges or midbasses:

method 1: Put the midrange taps about 2.5" - 3.5" from the throat. This is what's done in the "real" Unity horns. The advantage of going this route is that you get some real horn loading on the midranges, it raises your output about 6-10dB. On the downside, you need a big horn. An SH-50 loads the midranges down to about 200Hz, but it's something like 28" wide.

method 2: Just treat the speaker like an array. For instance, you can place the mids REALLY far out, if you treat the whole thing like an array. With a crossover point of 1350Hz and half-wavelength spacing, the midrange taps would be spaced 6.75" apart.

The tricky part with method 2, is that I'm unsure if there's a middle ground, because I've never tried to find out. IE, what if you placed the mids somewhere between "really far apart" and "really close together? I fear that if you did that, you might get a reflection off of the throat, which would create a notch in the response. When the midranges are really far apart, like the large black waveguide below, there's basically not radiating into the throat very much.

zKzJnHj.jpg