Hi, I've got a question about crossover design.
If the 2nd order crossover has the 180 degree phase shift (+90 degree from hi-pass and -90 degree from low-pass, the sum is 180 degree phase shift) so, the tweeter in 2nd order crossover must be wired 'reverse' polarity.
But If I have only hi-pass filter (without lo-pass filter) which I'll call it 1.5th order. Should the tweeter still be wired 'reverse' polarity ?
Thank you.
If the 2nd order crossover has the 180 degree phase shift (+90 degree from hi-pass and -90 degree from low-pass, the sum is 180 degree phase shift) so, the tweeter in 2nd order crossover must be wired 'reverse' polarity.
But If I have only hi-pass filter (without lo-pass filter) which I'll call it 1.5th order. Should the tweeter still be wired 'reverse' polarity ?
Thank you.
You are correct in your first post but ONLY if you also ignore the drivers acoustic roll offs and the horizontal spacing between drivers. Your analogy only applies to "text book" crossovers and not the real world. Each driver has it's own natural rolloff anywhere between 6 to 18 dB per octave then you have the horizontal acoustic difference which may add another 90 to 180 degrees or anything in between depending upon the crossover frequency. All must be taken into account and the actual crossover adjusted which isn't possible just by guessing at it.
Please read the Sticky at the top of this forum Introduction to designing crossovers without measurement
Please read the Sticky at the top of this forum Introduction to designing crossovers without measurement
Last edited:
Noticed ur post...
Your question revolves around two areas.
First what is happening in the area where the drivers outputs overlap, commonly known as the crossover region.
Second is what is happening to the overall summation of the drivers output WRT the original inputted signal. For example, music, or better still, an impulse. The impulse is often used in FFT measurement systems, and to look at "impulse response".
Within reason, in my view it is more important to get the HF section and the LF section to sum when looking at the impulse than to get a perfectly flat response in the xover region.
Serendipity and physics tend to team up to make the absolutely best impulse response also result in very good results in the xover region, but one can get one without the other...
Btw, I think your analysis of the amount of phase shift between the pass band and the stop band of the 2nd order filter is not quite right, although it is possible that i am confused on that subject too.
There is no 1.5th order filter in standard analog designs...afaik.
_-_-
Your question revolves around two areas.
First what is happening in the area where the drivers outputs overlap, commonly known as the crossover region.
Second is what is happening to the overall summation of the drivers output WRT the original inputted signal. For example, music, or better still, an impulse. The impulse is often used in FFT measurement systems, and to look at "impulse response".
Within reason, in my view it is more important to get the HF section and the LF section to sum when looking at the impulse than to get a perfectly flat response in the xover region.
Serendipity and physics tend to team up to make the absolutely best impulse response also result in very good results in the xover region, but one can get one without the other...
Btw, I think your analysis of the amount of phase shift between the pass band and the stop band of the 2nd order filter is not quite right, although it is possible that i am confused on that subject too.
There is no 1.5th order filter in standard analog designs...afaik.
_-_-
I would think it would depend mainly on the natural roll off characteristics of the woofer (I'm assuming you're talking about not using a low pass filter on the woofer). Also, physical spacing between sound sources should also be considered, as pointed out by DavidL. From what little I know, you'd normally only use a woofer without a low pass filter if its inherent roll-off is in a desired location and is smooth enough to do so, and that's pretty rare these days. One of the only good examples that comes to my mind at the moment is the 10" Seas A26RE4, intended to be a modern version of the woofer used in the old Dynaco A25. I'm sure there are some others out there, poly cones and whatnot with smoother roll offs. As for 1.5th order phase shift, I'm not sure where you're getting that from.
Last edited:
I wouldn't attempt to guess phase without doing a sim of the whole setup, since the bass and treble drivers will usually have acoustic centres offset by a few centimeters.
Learn to sim a speaker:
Downloads
Visaton provide plenty of kit examples and sample Boxsim files which you can import too:
HiFi-/Surround Kits
Boxsim Projektdatenbank 2 Wege
Learn to sim a speaker:
Downloads
Visaton provide plenty of kit examples and sample Boxsim files which you can import too:
HiFi-/Surround Kits
Boxsim Projektdatenbank 2 Wege
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.