Active xover help!

[led_zeppelin]

Yeah... i'm building an active pa system ofr my a level project, consisting of a pair of mid hi's (some old peaveys i have) and some subs (modified rog's X1 (with a 15inch kappa)), all powered by 4 lynx poweramps from acd with 4 625va transfo's 45-0-45. MMMMMMmmmmm....

In these boxes will be a PCM2409 (not sure if thats the part number) usb soundcard with spdif, and active crossovers. and microcontroller temp control of fans...

Basically i'm a bit unsure of the crossover... i've designed this one, but i haven't got any decent simulation software, not to mention dont know what i'm doing!
basically i want it to cross at around 150hz (if anyone knows a freq that would sound better i'd gladly use that) and also, i would like it to be variable... as in a pot to change the xover point.... Pls could i have your advice/opinions... All help as usual is much appreciated!!

Cheers

Matt

edit... and in the pic, why are the two out's out of phase?

[led_zeppelin]

Basically, i could do with an idea of the freq to volume graph? to 'see' the crossover and ideas as how to make it variable? cus there's many components id need to change at once i'd think?

[Konrad]

http://www.microchip.com/stellent/id...0&filterID=442

[jackinnj]

Originally posted by led_zeppelin
Basically, i could do with an idea of the freq to volume graph? to 'see' the crossover and ideas as how to make it variable? cus there's many components id need to change at once i'd think?

You can use Texas Instruments "Filter Pro" software -- freeware and it works quite well. It can be downloaded from their website, just plug FilterPro into their search engine.

[demons_wing]

try the ESP website man...rod has a comprehensive article on active filters...also the headwize site is a good read

[Swedish Chef]

Basically these are two cascaded Sallen-Key active stages to end up with a 4th order filter + buffer with variable gain for each channel. However, with the component values given these will NOT give a standard Butterworth (thus an overall Linkwitz-Riley) response for each stage. Look at the Texas Intstruments website under FilterPro for appropriate component values.

If these were indeed L-R filters, the two outputs would be in phase at the crossover frequency.

Having a variable crossover frequency with this circuit is quite a mess as you would need a 4-gang pot (8-gang for stereo).

You could simulate this circuit with PSpice Student which AFAIK could be downloaded for free.

/Magnus

[cunningham]

First of all, a capacitor will create a phase shift. I would say that just one 40dB/dec filter would be sufficient. May save parts.
You could always add a series capacitor to one filter to create a phase shift in order to make the two have the same phase shift. Seems fairly simple to me.

However, the phase shift may be different depending on the frequency so the bandwidth couldn't be to large.

[megacef]

Hi,

Elektor has published an article with the entire plans for a 2 Way linear phase crossover.

Here is the link for a correct copy :

http://users.otenet.gr/~athsam/2way_...near_phase.htm

Hope it helps !

Regards.

Moncef Christian Mentagui

[Swedish Chef]

Elektor has published an article with the entire plans for a 2 Way linear phase crossover.

Yes, that is the standard way of building a 4th order L-R filter using 2 cascaded 2nd order Butterworth stages. Works very well. I'm using it at home crossed @ 2kHz between Fane Studio 12B midbass/midrange woofers and B&C ME45 + JBL2405 horns.

/M