custom xovers

[ameuba10 10+ year member]

im interested in making my own crossovers. has anyone here made their own? any tips or advice would be appreciated. and if you have pics of your work or schematics could you please post them up for me. i am studying electronics engineering and i understand alot of it but i am curious to see what everyone else has done. thanks

Click to read more...

 

[JimJ 5,000+ posts]

Unless these are for home speakers, there really isn't a point.

 

[ameuba10 10+ year member]

why do you say that?

 

[JimJ 5,000+ posts]

In a vehicle, electronic crossovers provide much more flexibility, especially because you really don't know where the optimal crossover points/slopes may be...

If this is just a project for the learning experience, then that's fine //content.invisioncic.com/y282845/emoticons/smile.gif.1ebc41e1811405b213edfc4622c41e27.gif

 

[gunz4me 10+ year member]

Building a passive crossover network with more than a 6 dB per octave slope isn't worth it for the automotive environment. You need to know way too many factors to design a 24 dB per octave passive network that it is much easier to just do like JimJ said and go with an electronic crossover and an extra amp.

If you want to drive yourself bonkers trying to design your own passive network feel free to knock yourself out. Just be warned that it isn't as easy as you think because you have to worry about things such as on axis versus off axis as well as phasing in addition to getting your crossover points right for the impedance of your speakers.

 

[DidUHearThat? 10+ year member]

There's a tutorial on the PartsExpress website that goes into great detail on designing and building crossovers. But yes, you have to be more than a little crazy to do it. There are a lot of non-sensical parameters, too much trial and error, and an active crossover is almost always superior anyway.

As an electronics exercise I can understand, but not very practical. There are designs, schematics and instructions for buiding one that someone else designed or going off the deep end and trying your own design.

Building one from a design is reasonable, designing your own puts you into that geek category where you will never have a girlfriend.

 

[IDSkoT 5,000+ posts]

word.

 

[thch 10+ year member]

im interested in making my own crossovers. has anyone here made their own? any tips or advice would be appreciated. and if you have pics of your work or schematics could you please post them up for me. i am studying electronics engineering and i understand alot of it but i am curious to see what everyone else has done. thanks

places like sound.westhost.com also have some interesting tips. somewhat in the projects, but moreso in the articles section.

the following may be of help:

the main choices are for the cutoff, final slope, and Q.

slope, for the filters of interest, will consist of 1st and 2nd order stages. and for audio, will typically fall into the 1st (-6B/oct) to 4th (-24dB/oct) range.

each 2nd order stage has an associtated Q, which ranges from

the choice of Q is based on some goal. the common goals are:

butterworth -- maximally flat bandwidth. the stages are set to provide the flattest response in the passband. This is often the default for filters.

bessel -- maximally linear phase. the stage are set to provide a near constant delay in the passband. for some reason, it is sometimes claimed to be "critically damped" by some. This is not true, though it is closer to critical damping then butterworth.

chebychev -- allowing ripple in the passband (type 1) or stopband (type 2), the transition band between the pass and stopbands is minimized.

elliptic -- allow ripple in both pass and stopband, the transition band is minimized.

chebychev type 2 and elliptic based designs require what are essentially band-stop or "notch" element to achieve their goals, and this ends up requiring more complex circuits. chebychev type1 and elliptic both have more severe group delay spikes.

one thing of note is that selected capacitor values will be higher for low frequency or low impedance designs. the selected values for inductors will be higher for low frequency or high impedance designs. as such, its less expensive to built a crossover for a mid-tweeter then it is for a midbass or subwoofer.

one of the things I did when getting into electronics was to build active filters.