DCAudioDIY.com

OK, I'm gonna post the parts list/crossover schematic/baffle plans in this thread. It may take me a few days to take the measurements on the baffles as-built, but we'll get there. This design is copyright 2015 Roscoe Primrose. Feel free to built it for personal use, commercial use prohibited...

So, here's the parts list....

Part numbers are Parts Express #s

Quantities listed are for a pair of speakers/crossovers.

2 - Goldwood GW-1858 18" Pro Woofer Part # 290-386
2 - Tang Band W8-1808 8" Neodymium Full Range Driver Part # 264-894
4 - 100uF 100V Electrolytic Non-Polarized Crossover Capacitor Part # 027-360
4 - Dayton Audio DMPC-8.2 8.2uF 250V Polypropylene Capacitor Part # 027-426
4 - Jantzen 15mH 15 AWG P-Core Inductor Part # 255-128

You can use whatever binding posts you want, quantity needed will depend on whether you build the crossover onto the speakers themselves, use an external crossover or bi-amp, I'll leave that decision up to the builder. You may also want to use better caps in the crossover, replacing the 100uf NP electrolytics may be an improvement, the total value of the caps needs to be around 108uF.

Roscoe

_________________
I can explain it to you, but I can’t understand it for you.

Replacing the 100-ufd caps with polypropylene or polycarbonate caps will most definitely be an improvement.

_________________
Walt

Looking forward to seeing the transfer function that you are using with these drivers (especially in an Open Baffle environment) -- from the cap/coil mix, sounds like a 12db/octave Linkwitz/Riley crossover, somewhere below 100Hz.
Of course, there is a third camp (perhaps numbering 'just' one), a camp that advocates that you "chuck out" the passive components and use an active crossover which allows one to do many things; One important thing is to compensate for open-baffle cancellation.
Personal opinion, but nothing betters a direct coupling of the driver to the amp, driven by a well crafted active crossover: the proof is in the pudding -- not a single room at the show even came close to the musicality of an LX521, though the Quad '56 with the >12Khz "super-tweeter" was getting there.

\\Charles

_________________
"We are all inventors, each sailing out on a voyage of discovery, guided each by a private chart, of which there is no duplicate. The world is all gates, all opportunities."
Ralph Waldo Emerson

OK, so it's been way too long, but I've finally got the rest of the info one would need to build these together.... The baffle plans are based on a design I found on Joseph Esmilla's website, which doesn't seem to be there anymore. However, it's been archived, so the original information that inspired me to come up with these can be found here: http://jelabsarch.blogspot.com/search/l ... n%20Baffle A good basic explanation of open baffles can be found here: http://jelabsarch.blogspot.com/search/l ... n%20Baffle

My baffles are the same size & basic design as the ones Joseph published, with some added mass loading and some handles added as they get heavy....


Here's a couple pics and a drawing:





If you're going to build these, and don't plan on doing a lot of experimentation, I'd just build the main panel as a single piece, I made mine modular to make it easy to swap out drivers of different sizes/mount patterns. The bracing on the back might still be a good idea, as it'll stiffen up the panel a bit. I didn't add the pockets on the back for the lead shot to the drawing, but they're constructed from 3/4" plywood like the rest of the design, with inside dimensions of 3"x5.5"x8.5". Use the best void-free 3/4" plywood you can find. If your woodworking skills are better than mine, you might consider using biscuit joints to mount the sides and the top baffle so there's no screw holes to deal with on the front. Another option would be to glue/screw everything together, and then smooth out the panel and apply a veneer. As shown in the picture, the top panel is mounted flush with the top on the baffle, that's not really shown in the drawing.

At this point, I don't plan on doing any refinements to the passive crossover, as I'm getting much better results bi-amping. Even the lowly Behringer CX2310 (Parts Express 248-664, $80) sounds better than the passive crossover, and it's definitely a huge step up from the CX2300 it replaced in Behringer's lineup. B-amping also give you the option of trying different combinations of drivers (which I'm planning to do shortly), having to match driver sensitivities severely limits your driver choices.

Roscoe

_________________
I can explain it to you, but I can’t understand it for you.

Have you ever tried dipole with high output impedance amplifier?

_________________
SamB

Yeah, I built a J-FET current source amp a while back, worked quite well with low QTS drivers on an OB. The GW-1858 has a QTS just over 1, so it's happy on an OB with a low impedance amp...

Roscoe

_________________
I can explain it to you, but I can’t understand it for you.

Looks nice Roscoe. I have a set of NIB 6" Pioneer PIM6 that I would like to mate with a 12" or 15" woofer in an OB at some point.

After listening to these for a while, and using some test tones, it's become apparent that there's a bit of a suck-out between about 110Hz & 225Hz due to the close proximity of the full-range driver to the top of the baffle. Sometimes solutions come in very simple packages:




The added piece of plywood is 10"x24", centered over the full-range driver, with 8.5" above the top of the original baffle. This moves the baffle-induced LF roll-off of the full-range driver down by more than an octave, and is a huge improvement. I'm currently working on new baffles that will be slightly larger, with the full-range driver offset to the side and lower on the baffle to create the same effect.

More soon...

Roscoe

_________________
I can explain it to you, but I can’t understand it for you.

here is an article about tamming the rising top in on TB full range driver
http://www.transcendentsound.com/Full_Range.html

That's for the 1772, which has a significantly hotter high-end than the 1808s that I'm using....

Roscoe

_________________
I can explain it to you, but I can’t understand it for you.