2nd order linkwitz-riley crossover

[ZLyzen]

I'm designing a pair of small 2-way bookshelf speakers that will be used for standalone and as part of a 2.1 system.

Anyways, my question is pertaining to the crossever that will deal with the midbass and tweeter in the 2 way speaker itself.

For a 2nd order linkwitz-riley crossover there are some phase issues..

this is the response if I cross at 3000Hz, where the -3dB point is 3000Hz for both drivers

However if I make the -3dB point of the tweeter 3500Hz and the -3dB point of the midbass 2500Hz or in a sense "pull them apart" on the graph their combined response improves in theory.

Are there any experienced veterans who can give me a hand in deciding what will actually work the best? I'm pretty set on this 2nd order configuration because:

Its popular, well known, and reliable
plus its only 2nd order so not too many parts, but 1st order isn't steep enough for my drivers

So should they have the same frequency and suffer canceling each other due to phase, or should I "pull apart" their frequencies, will that actually work to combat the negative effects of the phase?

 

[godfrey]

With a Linkwitz-Riley crossover, both outputs are -6dB (not -3dB) at the crossover frequency. If there is a notch in the combined output, that is because one of the drivers is connected with the polarity reversed. If you change the polarity of one of the drivers, it should sum correctly.

Aside from that, it sounds like you're trying to get the electrical outputs to sum correctly. To get the acoustic output from the drivers to sum correctly, you need to take into account the driver's frequency responses.

 

[AllenB]

...along with the driver's acoustic phase 😉

 

[ZLyzen]

okay, I get almost a perfectly flat sum when i reverse the polarity of the tweeter, any other advice besides having that flipped and crossing them at the same frequency?

 

[AndrewT]

As advised earlier, you need to get the acoustic crossover correct for the acoustic response to sound right.

The electrical responses of a theoretical pair of filters feeding resistors is not the answer.

 

[AudioEngineer]

I'm designing a pair of small 2-way bookshelf speakers that will be used for standalone and as part of a 2.1 system.

Anyways, my question is pertaining to the crossever that will deal with the midbass and tweeter in the 2 way speaker itself.

For a 2nd order linkwitz-riley crossover there are some phase issues..

this is the response if I cross at 3000Hz, where the -3dB point is 3000Hz for both drivers

However if I make the -3dB point of the tweeter 3500Hz and the -3dB point of the midbass 2500Hz or in a sense "pull them apart" on the graph their combined response improves in theory.

Are there any experienced veterans who can give me a hand in deciding what will actually work the best? I'm pretty set on this 2nd order configuration because:

Its popular, well known, and reliable
plus its only 2nd order so not too many parts, but 1st order isn't steep enough for my drivers

So should they have the same frequency and suffer canceling each other due to phase, or should I "pull apart" their frequencies, will that actually work to combat the negative effects of the phase?

For a linkwitz riley you should be looking at the -6db points and not the -3db points.

 

[MarcelvdG]

Like godfrey already wrote more than seven years ago...

 

[bentoronto]

Hardly worth fussing over. Do it on the back of an envelope.

The (in)tolerance of the components swamps any fine-points of the analysis. All a crap-shoot until you test in your room and then fine-tune. (Make sure your tweeters have an L-pad.)

Do you have the means of measuring/matching your coils and capacitors?

B.