Joachim,
Except for a coaxial speaker I don't think you can get the phase or time alignment exact except for one point in space as Earl just said. As soon as you add the separation in the vertical plane as Earl does with a horn over a woofer you only have one place in time and space that everything can be correct. I'm sure you know that, all the talk of getting things perfect is a nice dream but can never truly be accomplished over the entire audio bandwidth.
Except for a coaxial speaker I don't think you can get the phase or time alignment exact except for one point in space as Earl just said. As soon as you add the separation in the vertical plane as Earl does with a horn over a woofer you only have one place in time and space that everything can be correct. I'm sure you know that, all the talk of getting things perfect is a nice dream but can never truly be accomplished over the entire audio bandwidth.
Trying not to be limited by one specific aspect, I am in the process of trying things out one by one to see how technology adds up. Quite interestingly, each technology aspect makes sense. If you implement them in the right order, every improvement expected comes out the way it should.
I generally would try to stick with full range drivers over the widest audio range possible for the reason of alignment. Using more drivers is generally necessary when you aim for higher SPL capability, thus the tradeoff.
Full range driver design is difficult to get right. My instinct is that possibly somewhere around 6.5" is the limit. If it can be done, that would really be the ultimate for normal home listening in average apartments. Hope to get it done no later than next year.
Coaxial drivers are not ideal due to the interaction between the two drivers. But that is a tradeoff if you want to go beyond a certain driver size.
I generally would try to stick with full range drivers over the widest audio range possible for the reason of alignment. Using more drivers is generally necessary when you aim for higher SPL capability, thus the tradeoff.
Full range driver design is difficult to get right. My instinct is that possibly somewhere around 6.5" is the limit. If it can be done, that would really be the ultimate for normal home listening in average apartments. Hope to get it done no later than next year.
Coaxial drivers are not ideal due to the interaction between the two drivers. But that is a tradeoff if you want to go beyond a certain driver size.
Last edited:
I agree that the transient response can only be made perfect on one axis with the usual driver arrangements. But it doesn't necessarily have to be worse off-axis compared to a speaker that is not transient perfect (maybe we should define a better term for that, basically we are talking of a speaker that is minimum phase) at all.
It depends on the crossover used for that purpose. FIR and subtractive-delay for instance don't behave the same way.
But it is also on-axis where the tranisent response matters the most, if we assume that it matters at all.
Regards
Charles
It depends on the crossover used for that purpose. FIR and subtractive-delay for instance don't behave the same way.
But it is also on-axis where the tranisent response matters the most, if we assume that it matters at all.
Regards
Charles
You see that is the problem. I don't assume that things matter just because they are. I worry about those things that I know, from valid studies, make a difference. There is no study that shows that transient-perfect is audible. Lots of studies show that poor frequency response and poor polar response are quite audible. Get those right and then maybe transient-perfect is worth worrying about, but I see it as another issue that is a non-issue. The ear cannot be phase perfect because of the way it works. It is not an FFT.
Generally, if the system phase is linear phase, the sound field image depth is better preserved. But the prerequisite is that the CSD needs to drop pretty fast and relatively even in spectrum drop.
The benefit if having driver response follow the power response is that when you compensate for the frequency response properly, you also provide the benefit of leading the phase as well. This could be why Earl feels his speakers are perfect.
The benefit if having driver response follow the power response is that when you compensate for the frequency response properly, you also provide the benefit of leading the phase as well. This could be why Earl feels his speakers are perfect.
That statement is not logically correct. Power response is phase blind so what does it have to do with phase?
The reason that you want the driver response to follow the power response is then they both get corrected simultaneously with EQ. It has nothing to do with phase.
some more recent food for thought from JW:
Loudspeaker Technology Part 2: The Time Domain and Human Hearing - The Broadcast Bridge - Connecting IT to Broadcast
prof. Wieslaw Woszczyk also on importance of the time domain in loudspeakers:
An internationally recognized audio researcher and educator with leading expertise in emerging technology trends in audio explains the TD-M1 and a new time-domain theory by Prof. Wieslaw Woszczyk. | ECLIPSE Home Audio Systems
Loudspeaker Technology Part 2: The Time Domain and Human Hearing - The Broadcast Bridge - Connecting IT to Broadcast
prof. Wieslaw Woszczyk also on importance of the time domain in loudspeakers:
An internationally recognized audio researcher and educator with leading expertise in emerging technology trends in audio explains the TD-M1 and a new time-domain theory by Prof. Wieslaw Woszczyk. | ECLIPSE Home Audio Systems
Last edited:
It is never possible to dampen resonances entirely without effecting the forced response. This is the part that gets tricky. Therefore, during measurements it is necessary to look at both forces and natural response to help make the design balance. Of course distortion needs to be looked at too depending on material used.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.