OB + BR Feastrex / Eton new build - HELP!

[dpaws]

Evenin' all...

I've had to put my Feastrex BVR build on hold for now but seeing as the tools were out... Inspired by the modular Focal speakers, the speaker is built in seperate modules allowing me to play with different configurations. At some stage I'm sure a rear firing tweeter or a seperate front firing super tweeter module will follow. All timber is 28mm Baltic ply.

http://cid-a332bc3aefc89f6b.skydrive...rexproject.jpg

At present the BR base houses an Eton 11-581/50 Hex ala Duetta enclosure albeit with an 8' sloped baffle. Once glued together I intend to stuff the port until I achieve a quasi-aperiodic alignment to negate the need for a Zobel.

The xover will be my active Pass XVR1 follwed by a newly aquired pair of Deltec D50S power amps (Graafs retired at present, it's too hot in Italy to listen in the same room!)

The top baffle will hold the D9nf driver in an OB arrangement. At present the baffle dimensions are 456mm wide x 330mm tall. Edges are square but I intend to route a 38mm radius edge around top and sides.

The wide baffe is to avoid a BSC circuit, giving the facility to roll the driver off ~250Hz. However, if after listening I find that ~500Hz sounds better then I'll slim the baffle accordingly.

Help! A question for the gurus.. I understand that the Feastrex should be centrally mounted in relation to width but where would it be best height wise? Do I need 330mm or more in height or would it be advantageous to minimise this dimension? Should I consider consider a more complex shape for the top and sides? BTW the BR baffle below is ~750mm tall x 456 wide.

All thoughts / comments most welcome and appreciated.......

PS A non commercial aside - I have to give credit to Dewalt. I'm an average guy with little to no woodworking experience. However, using their electric tools and in particular the guide rail and biscuit joining kit I've achieved <+/- 1mm and 0.5' tolerances - the fit is amazing!!

[dpaws]

I forgot to mention - the other "speaker" is the D9nf mounted in a redundant pine shelf.... to be retired shortly! It was only meant to be a jig for running in purposes but the female vocals etc do sound spooky!

[seanzozo]

The driver should be "centrally located" ONLY if you intent to operate it with active equalization on a narrow baffle, ala linkwitz. Centrally locating it gives an even polar response, but makes the frequency response quite wonky.

If you want to run a wide baffle with simple XO you need deterimiine baffle width to correspond with your XO anf to offset the driver using proper simulation (like MJKs software or Basta). There will be wonky polar response and the product might need wall treatment to get rid of room response in the treble where the driver is not behaving like a dipole.

If I were not studying for the bar, I would run some sims for you.

Sean

[dpaws]

Originally posted by seanzozo
If I were not studying for the bar, I would run some sims for you.

Sean

Thanks for the thought and the advice. I wasn't sure how a wide range driver with its wizzer cone would sim. Unfortunately there's no (complete) idiots guide to the softwares, it's rubbish in rubbish out so I guess I'd best wait until last orders... ;o)

The rear of the baffle can have wings if it helps although I've read somewhere that >260Hz they're redundant... it's all so confusing!

[Nelson Pass]

I run my D9nf's with .9 m width, but I've had some success with
.6 m and less. First off, if you are constricted on width, try adding
height, as it helps. I recommend 1.5 m height for the top edge, with
the driver mounted at a height of about .9 m.

The other thing you can do is add a bit of side baffling going back
from the inner edge, and then mounting the drivers a bit toward that
edge on the front surface. This gives you a little more surface area
and helps to break up the "half-wave" front-to-back cancellation that
bothers smaller baffles.

Try a few things, see what works. Since you have an XVR1, you will
have plenty of opportunity to play around.

[dpaws]

Many thanks for input Nelson - yes, having the XVR1 does instill some confidence!

I'm looking to cross over >250Hz, a little higher than your own 2 way project (~100Hz if I recall) so I'm not sure that such a large size is quite as critical. You appear to have mounted your driver ~ centrally and then commented of the uneven treble response... what has confused me is that with the field coil variant you suggested a more even response >3kHz, in contradiction to any baffle induced response distortion.

I suppose I really need to know the raw driver response to identify any driver induced peaks and troughs before trying to superimpose a baffle response - otherwise I assume I run the risk of accidentally summing rather than cancelling any variation?

[Nelson Pass]

In a large baffle, the treble response is pretty much that of the driver.

I wouldn't worry about it at this point - I run my D9nf's without eq,
and you can trim the sound quite a bit by angling the baffles in your
room.

I think your real issues will be in getting the crossover just right.

You can start with a sim and take it from there. If you don't mind
some trial and error, you'll probably get what you want.

[dpaws]

Guys... would someone mind running a sim for me?

The driver is the Feastrex D9nf. The baffle size overall is 1086mm tall x 476mm wide. (The 476mm is correct, not 456mm as mentioned before)

Feastrex centre postioned at coordinates x = 259mm y = 900mm

Very much obliged....

[dpaws]

Just thinking aloud.... what would be the effect / predicted response a wide rage driver was mounted in the centre of an acoustic panel like the Deflex?

http://www.deflex.co.uk/catalog/index.php?cPath=46_51

I need to get out more.....

[dpaws]

Interesting...

http://www.audiotweaks.com/tweaks/tweak_284.htm

Has anyone here experimented with different materials and surface properties and then acoustically evaluated the result?