general/basic crossover questions

[tech.knockout]

Firstly, what happens when part of the low pass slope is in a drivers breakup region?(say about -6db down the slope the breakup begins)

What happens, or what to avoid, when you crossover using different order acoustic slopes?

Generally what (bad)things happen if the crossover acoustic slopes dont approximate any particular alignment , butterworth, LR, etc?


I find it strange people seem to talk about these alignments like they have a choice.
However in reality it seems that with drivers' frequency response imperfections and variations(on axis and off axis) getting acoustic slopes that has much resemblance to these theoretical ideals seems difficult.

Take for example the "Thor speaker" using seas excel bass drivers. Its low pass filter is supposedly 4th order, but im pretty sure its not the case when listening 30 off axis, the drivers' 30 off axis curve already starts deviating(from the on axis curve) before the crossover point.

[sreten]

Hi,

the Thor's low pass is electrically 1st order + 2nd order notch
filter which may end up approximating 4th order acoustic when
combined with the drivers response.

The final acoustic order, alignment, whatever is what counts.

I'd say the most important thing to avoid is not knowing
what you are doing and convincing yourself that you do.

Your others questions are far too complex to answer simply.

/sreten.

[Salas]

Originally posted by tech.knockout
Firstly, what happens when part of the low pass slope is in a drivers breakup region?(say about -6db down the slope the breakup begins)

What happens, or what to avoid, when you crossover using different order acoustic slopes?

Generally what (bad)things happen if the crossover acoustic slopes dont approximate any particular alignment , butterworth, LR, etc?


I find it strange people seem to talk about these alignments like they have a choice.
However in reality it seems that with drivers' frequency response imperfections and variations(on axis and off axis) getting acoustic slopes that has much resemblance to these theoretical ideals seems difficult.

Take for example the "Thor speaker" using seas excel bass drivers. Its low pass filter is supposedly 4th order, but im pretty sure its not the case when listening 30 off axis, the drivers' 30 off axis curve already starts deviating(from the on axis curve) before the crossover point.

1. When in breakup region, the outcome is 'dirtier' than optimal for a given driver compliment. More distortion, less definition. Also generally you are already away from good off axis dispersion and power response suffers in cross region. Heavier sounding outcome bcs of recessed mids, more heavy if enough bsc is combined.

2. Lobe is non symmetrical, find best listening height by measurement, avoid typical stands, make custom ones with proper height &/or tilt. Tweeter under woofer baffle placement is a classic British school means of sending the lobe to seated listener's height.

3. People talk out of reading textbooks, popular missunderstandings prevail. Following typical alingments is attainable nontheless if crossover frequency and drivers bandwidth is enough and engineer's prowess is cool. You end up with a complex and costly crossover though that compromises some detail.
Your example suffers from off axis cancellations lower than cross region bcs the centre to centre distance of midbasses is long enough to get em out of phase enough in those wavelengths and angle. It suffers the typical forward and cold sound of LR24 MTM. Power response is weak under crossover and on crossover. Direct sound prevails on crossover, so weakend mids room power there and lower, vs healthy upper mids power response higher, give glassy feel.