Tweeter XO

[Ilianh]

I am working on a speaker using the d'appolito configuration with the Illusion audio t25an tweeter and 2 cambridge 4inchers.

The tweeter's center is at exatly 10 cm from each of the mids center, that makes a cutoff point of around 3500 hz.

I would prefer using a first order crossover on all drivers, I'm a little worried about cutting such a small tweeter with a first order 3500hz. The speaker is not meant to play too loud, its only around 90db\w\m

http://www.peerlessaudio.com/products/drivers/tweeters/main3.asp?pname=6

http://www.solen.ca/pics/illusion/t25an.jpg

what do you guys think about the idea, will I burn the tweeter?

 

[leadbelly]

I can't speak from experience, but just looking at the posted specs, yes I think 1st order at 3500 Hz is too risky.

 

[kelticwizard]

Ilianh:

My understanding is that third order or higher crossovers are recommended for d'Appolito configurations.

Not because of tweeter power handling, but concerns about lobing over a broad band. Drivers in first order crossovers really interact with each other over a band of several octaves.

At 3500 Hz, I would think that the crossover components would be inexpensive enough for you to consider third order or above.

 

[Dave Jones]

You might want to consider also the amount of excursion you'll have on the tweeters with that low crossover point and a first order crossover. (Where I'm coming from: Newbie who's been reading a lot.)

 

[GM]

Butt the 4"ers together, with the tweeter plate trimmed as much as possible to fit beside them, ergo mirror image baffles are required if components aren't centered. Considering these are for moderate SPLs, this should allow a high enough XO point to get close enough to the recommended -12dB/Fs minimum. WRT room setup, orient them so the tweeters are on the outside to gain a little time delay.

GM

 

[Salas]

90 degrees

Hello,

If you dont have 90 degrees phase difference between tw and midbasses, ie a 1st or 3rd order acoustic symmetric slope then there is no Dappolito. You just have 2 midbasses and a tweeter in the middle elsewise. No lobe symmetry & less vertical cross point dispersion ripple can be achieved.
Also watch out for too much horizontal energy. You may need to cut about 1 to 1.5 dB of mids (octave broad) to get a natural cohesive result when driving a real room.

Good luck

 

[Ilianh]

Thanks guys for the great input.

Well here I basicly have two choices, first of all I don't know why but most of the sites I visited opted for a first order Xo saying it sounds beter, even though I strongly beleive in what Kelticwizard said about drivers interacting with each other.

I'd either keep the 3500 xo point and make a 18db/octave slope or make a 5000hz xo point with a 6db/octave slope (the 4"'s can take it preety well up to that" (thats a 7cm distance between mid's and tweet's center)

On this project I have a bigger budget than my other speaker projects, I made the prototype box, tp play with xo's, polyfill amounts and so on, and I have everything I need for the 2 finals.

But I dont think I wanna make 2 xo's one 18db slope the other one 6db in order to see wich one is best.....

 

[Ilianh]

I've deeply thought about making a third order XO on this prototype wich has 10 cm between mids and tweeter, and another prototype baffle with a 7cm distance / 5000hz / 6db xo and to compare both, but then that won't be too accurate as its only one speaker, in order to have an idea about stereo imaging, what this config is all about, I'd need two of both.

That implys making 3 other prototype boxes and 3 other xo's.... and so on... I don't think I'll have the time/money to do that and then do 2 final boxes (wich have a heavily complicated design, the prototypes are just a simple baffle with the needed dimensions..).


So any shared experience from people that have already been through this would be greatly appreciated.

Btw, where can I find the original papers by Joseph DAppolito?

 

[Ilianh]

heres a pic.

 

[Bose(o)]

I have used a first order Butterworth@2400Hz. with the T25AN tweeter. I used that tweeter for an install in a buddy's car, sounds beautiful, warm and rich, unfortunately the inherent sounds of a metal dome attached themselves to the sound. I wonder how the softdome version sounds?

 

[Ilianh]

Oh.... really? arnt you scared that it'll blow? I'm scared for its life at 3500hz with a 6db slope.... I'd never even think about a first order at 2400hz, and it isnt getting to xmax?

Btw, I wanted to specify that all the modeling was done using Martin J Kings's wonderful Ported box worksheet. Results are so accurate...Its scary

Also, I think its harder to have a good sounding third order than to make a good sounding first order....
any ideas? guys?

 

[Ilianh]

Well... ok... as noone seems to be able to help.... I'll have to do the try'n'hear thing.... anyway.....



Thanks to all.

 

[planet10]

Re: 90 degrees

Butt the 4"ers together, with the tweeter plate trimmed as much as possible to fit beside them, ergo mirror image baffles are required if components aren't centered

I'll 2nd that comment... essentially how i've always done MTMs (1st one in 1975 with clear Harbeth poly 8" and probably a philips tweeter)

If you dont have 90 degrees phase difference between tw and midbasses, ie a 1st or 3rd order acoustic symmetric slope then there is no Dappolito

Joe might disagree with you... i remember seeing the question posited to him and he said he is using 4th order now...

dave

 

[Ilianh]

Interesting,
Thanks for the reply,

And what about series filters? I thought maybe about a second order series xo with a lower Zeta than butterworth, like 0.7 for instance, in order to make the slope stronger, but I guess that might create some bumps around the xo point.

I've read all the posts on series xo's and everything on the web, every single website contradicts many other ones on many points.... wich one to beleive?

Anyone tested the calculator on Argos loudspeakers website? for 2 way series xo? and whats that thing about needing to have the lowest dcr on the coils?