2 way Crossover Help

If i were you i would buy a minidsp and just adjust it on the fly... and youd have to be nuts to cross over the woofers at 9000hz, i would say anything higher than 4000hz is pushing it.

agree, with measurement and variable component values, you should be able to get quite close...sarah palin is a nice person.

is this what you are building?

http://www.youtube.com/watch?v=hjScTaqCcas

Nice driver choice.

I'd use 24 tweets, though. Vertical sweet spot is dicatated by the tweeter line length. They need to be aimed well.

I wouldn't cross higher than 4kHz, lower is preferable.

If you will not be optimizing the crossover with detailed measurements and a program like PCD (Passive Crossover Designer), I would invest in a receiver with Audyssey MultEQ XT or some other hi resolution automatic EQ system.

I use SLAs, and Audyssey makes them sound like more refined speakers than their generic crossover can.

JSS

Due to 4" driver spacing (center to center), and to widen the angle of the crossover nulls.

Probably the hardest part of speaker design is the crossover. The BFM designs use more generic crossover topologies, because the are low cost systems, and driver choices can change. BFM designs offer some of the best bang for buck in the DIY audio world. But the crossovers are not what I would consider optimized, and Audyssey will help with this, but it is more of a band-aid than a cure.

If you are gonna spend that much on drivers, You may want to spend some $$ on measurement equipment if you are gonna tinker with optimizing a crossover.

You will likely be very pleased with the generic crossover and especially once Audyssey has done its thing; I am. I of course want to replace my LCR at some point with more optimized speakers, but that will take time and $$.

JSS

"I am really trying to learn here so bear w/ me. Why does driver spacing dictate crossover freq choice? "

that depends on what you are doing. 1/4 wave couples...

More info:

www.diy-audio.narod.ru/litr/nflawp.pdf

Good paper. More good reading:

http://www.audioartistry.com/products_CBT.htm

Click on the links. Lots of good info.

IIRC, you can fit three of those tweets per 4" driver. Use as many as you can tightly spaced to equal the woofer line.

To do crossover work, active crossovers would be easiest. You would need more amplifier channels and an electronic crossover like a MiniDSP.

For passive crossover work, you would need a measurement mic (preferably calibrated), soundcard, Room EQ Wizard or HolmImpulse software, and Passive Crossover Designer software. Here at AVS there aren't too many passive crossover designing folks, but at the PartsExpress forum, you can get some help.

Good reads on crossovers:

http://sound.westhost.com/articles.htm

http://sound.westhost.com/lr-passive.htm

http://www.pispeakers.com/Speaker_Crossover.pdf

http://www.pispeakers.com/Speaker_Crossover_Lab.pdf

http://www.pispeakers.com/Pi_Speakers_Info.pdf

Bottom line is that you can get most of the performance out of the design with the generic cross and Audyssey. If you really want to optimize the crossover, you may want to shade the line and possibly curve it like a CBT.

JSS

Remember not to get too wrapped up in theory. The ultimate goal is to build a speaker that sounds good. I use measurement equipment and theory guidelines to get started but it does not dictate the end result. My ears have the final say.

The main purpose of the crossover is to shape each driver's output so that they blend properly to make a total output that somewhat accurately reproduces the material and is pleasing to the ear. The crossover frequencies, padding and slopes will be determined by each driver's output.

That said, if your woofer's output begins to reduce in the 4 khz range then you'll need to choose a tweeter that has output in that range and high pass it about that point.

You should be able to run a lower crossover on the tweets since each one will only be seeing 1/24th the power a single tweet would see (for a line of 24).

The reason for the crossover a 4k is not only center/center distance and crossover nulls, but the 4" driver becomes more directional as freq is increased.

With MiniDSP, Audyssey will still function. MiniDSP will alow you to tinker as much as you would like without having to solder anything. You can then build a passive XO based on what works best with the DSP.

JSS

Looking at those response curves, I can see the problem. That 4" driver is not really a woofer. It's really a full range driver. What will you be using for bass since the output of those 4'' drivers begin to drop off at 150 hz?

Have you considered using a different low frequency driver?

a couple more reads on line arrays.

http://www.roger-russell.com/ids/ids.htm

http://www.trueaudio.com/array/index.htm

if i were going to do one, i'd probably go with a full range driver, such as this one:

http://www.parts-express.com/pedocs/...ifications.pdf

"not literally touching as I had planned - thoughts?"

as long as the center of each driver is less than 1/4 of the wavelengths that they are producing, you should get pretty good coupling.

if you were to get creative with a straight line array and digital delay and levels, you might be able to simulate the curved array with a vertical stack of speakers. ah, i see that is what jbl is doing with their cbt series, but they do it passively...now that is slick.

have you visited: http://www.xlrtechs.com/dbkeele.com/CBT.php

some good stuff there.

The theory behind the TLAH is that the transmission line will help out the response below 150Hz. It will not act like a sealed box, but I do not know how to best model the TLAH short of using Akabak, which I am not good at.

Bill would be the best person to ask concerning how much low end the TLAH will provide.

The CBT36 crosses over lower than you think to the tweets, letting the woofs get away with a slightly larger C-C spacing. ideally C-C spacing should always be minimized in array designs.

THT with the Dayton HF driver can be crossed over at 150Hz with no problem.

Ask Bill if you should change the dimensions of the TLAH with he drivers you have.

Cross Spectrum Labs offers a calibrated EMM-6 mic.

JSS

Drivers are rear mounted...flanges likely touch on the back side of the baffle.

crossover frequency is often calculated using this formula:
SOS (speed of sound = 13500 ips) divided by CTC spacing...this can give you a starting point for crossover development.