Hello, I've found somewhere a description of a crossover design in which there was: 'This crossover is asymmetrical 4th order Linkwitz-Riley at 1750 Hz. The slightly steeper tweeter slope along with a slightly shallow woofer slope helps to align the listening axis with the tweeter.'
I'm such a newbie in crossover design and I'd like to go into detail with on-axis and off-axis effects of various typologies. What exactly does that phrase mean?
I'm such a newbie in crossover design and I'd like to go into detail with on-axis and off-axis effects of various typologies. What exactly does that phrase mean?
Any link to the full description? By def LR4 is symmetrical, so not sure what they are describing.
Hi,
By varying the slope of the filter, the group delay of filter is changed, so the tweeter is 'retarded/delayed' wrt the woofer. This way both drivers are 'time aligned'( point of emmision in vertical axis is now on same axis for both drivers): the lobe will point in the tweeter axis, not up or down.
That said assymetrical LR doesn't exist... so it shouldn't be called like that imho. In other words, they had a target of LR, it didn't work because of physical limitation of drivers and custom taylored an 'electrical answer' ( they could have answered this by angling the front of loudspeaker if the delay needed to compensate was short enough). So it's abusive to call it like that to me.
They could have used a regular LR profile and inserted an allpass cell to compensate but either they didn't know about it, either it increased cost too much.
http://www.rane.com/pdf/ranenotes/Linkwitz_Riley_Crossovers_Primer.pdf
By varying the slope of the filter, the group delay of filter is changed, so the tweeter is 'retarded/delayed' wrt the woofer. This way both drivers are 'time aligned'( point of emmision in vertical axis is now on same axis for both drivers): the lobe will point in the tweeter axis, not up or down.
That said assymetrical LR doesn't exist... so it shouldn't be called like that imho. In other words, they had a target of LR, it didn't work because of physical limitation of drivers and custom taylored an 'electrical answer' ( they could have answered this by angling the front of loudspeaker if the delay needed to compensate was short enough). So it's abusive to call it like that to me.
They could have used a regular LR profile and inserted an allpass cell to compensate but either they didn't know about it, either it increased cost too much.
http://www.rane.com/pdf/ranenotes/Linkwitz_Riley_Crossovers_Primer.pdf
Last edited:
This is not what i understand Bill: he says 2pole on woofer. For tweeter yes it could be acoustical 4pole. Edit: take a look at transfer function and individual drivers response ( which seems to include filters to me). But i may be wrong.
Andreaemme, Zaph is a good designer so he knows what he is doing. I suppose this is to keep costs down and probably for other details regarding tweeter protection.
Anyway what matter is the acoustical results as outcome, not the electrical profile used to achieve the target. Asymmetrical filter are often used with passive coaxial driver's filters as they help to compensate the offset between woofer and tweeter.
They don't have name on them but are as valid as named ones as long as they meets the acoustical target ( as the one which have 'name' are 'special case' along an infinity of possible answer, not dissimilar to the situation with bass reflex alignements).
Try this one ( it's the same content but pdf is more readable imho, it's available at the end of article).
https://www.ranecommercial.com/kb_article.php?article=2134
Andreaemme, Zaph is a good designer so he knows what he is doing. I suppose this is to keep costs down and probably for other details regarding tweeter protection.
Anyway what matter is the acoustical results as outcome, not the electrical profile used to achieve the target. Asymmetrical filter are often used with passive coaxial driver's filters as they help to compensate the offset between woofer and tweeter.
They don't have name on them but are as valid as named ones as long as they meets the acoustical target ( as the one which have 'name' are 'special case' along an infinity of possible answer, not dissimilar to the situation with bass reflex alignements).
Try this one ( it's the same content but pdf is more readable imho, it's available at the end of article).
https://www.ranecommercial.com/kb_article.php?article=2134
Last edited:
Yes it is nice. There is another very interesting one about bessel's filter. Worth a read ( study 🙂 )too.
Yes, asymmetrical crossovers are quite common. Often B2 and B3 bit a 3rd/4th works as well.
To call it LR-4 is not quite right. Maybe it can achieve an effective LR-4 ACOUSTIC, which would be more correct.
Yes, Zaph ( John) does this and gets very good phase coherence, but he also has a tendency to cross over WAY too low as in the SR-71, and rely on the natural roll-off of the drivers. This pushes them way too hard and the result is much higher distortion from the tweeter in the midrange as he is pushing it @ 1700 when those tweeters are much happier around 2200.
To call it LR-4 is not quite right. Maybe it can achieve an effective LR-4 ACOUSTIC, which would be more correct.
Yes, Zaph ( John) does this and gets very good phase coherence, but he also has a tendency to cross over WAY too low as in the SR-71, and rely on the natural roll-off of the drivers. This pushes them way too hard and the result is much higher distortion from the tweeter in the midrange as he is pushing it @ 1700 when those tweeters are much happier around 2200.