In-depth Analysis Of Speaker Crossover Design(I)

- Nov 06, 2020-

In fact, if you want to have a deeper understanding of the role and debugging of the speaker crossover, in addition to having a certain subjective evaluation ability and understanding the sound of the reference level, it is better to have a good understanding of the acoustic characteristics of the speaker itself and the psychology of people in the room. acoustics.


The first question is the balance between on-axis frequency response and off-axis frequency response


Near the crossover point, if the directivity coefficient DI of the tweeter and the woofer is different, it is likely that if the axial frequency response curve is leveled, the off-axis frequency response curve is not standard. At this time, it is usually not necessarily the optimal solution, and it may sound stained. Take the ATC SCM7 I evaluated some time ago as an example,

It can be seen that around 4kHz, the directivity appears abnormal, changing from strong to weak. If the axial frequency response curve is forced to be leveled at this time, it will cause too much off-axis frequency response curve, and the subjective sense of hearing may overflow around 4kHz. (Of course, SCM7 is indeed a little less than 4k.)

From a quantitative point of view, the off-axis frequency response affects about 60~70% of human subjective sound quality perception. Of course, this is also related to the specific placement of the speakers.

For example, when the speaker is placed, the sound center is not directly facing the listening position, so the direct sound is not an axial frequency response.


In any case, there is no doubt that the impact of off-axis frequency response on the sound quality of speakers cannot be ignored in many cases. In many cases, the axial frequency response is "sacrificed" at the crossover point or nearby frequency bands, and the off-axis frequency response is better, but the actual hearing will be better. The off-axis frequency response here refers in general, and the frequency response of the specific angle and direction needs to be determined as appropriate. And specifically how to choose between the axial frequency response curve and the off-axis frequency response curve, because sometimes it is not possible to take it into account at the same time, it needs to be selected, and subjective evaluation is required. If you want to adjust to the best, the actual hearing will prevail.


Of course, the above-mentioned sense of hearing and frequency divider adjustment usually occur when the frequency response and directivity of the tweeter and woofer do not match near the crossover point. Or to put it in a more layman way, the two units are not well matched, so they must be done. However, bad unit matching happens from time to time. DIY is the most common example. If two units are not matched well, the effect may not be good even if you spend a lot of money. This phenomenon may also occur in products designed by manufacturers, especially when the units are not completely designed by themselves, which may be more common.


But if the performance near the crossover point is considered at the beginning of the design, it is usually not necessary to consider this problem, or the tuning workload for this problem is not large. Waveguide and horn are the most common structures to control directivity. Of course, ordinary dome units may also find woofers corresponding to directivity. However, usually the absolute value of directivity DI will be smaller and beamwidth will be larger.

The reason for the sound near the crossover and the "uncertainty" that some people understand is not that the frequency response curve is useless, and in many cases it is not the group delay introduced by the crossover, but that the whole problem is not clearly seen.