Hello. This is my first real question on any forum of any kind, ever.
My question comes out of my recent completion of a pair of Discovery-4 speakers designed by Troels Gravesen. The speaker is a 4-way with 2 crossovers. The top crossover (@2500 Hz) serves a mid-woofer and the tweeter. The crossovers are second order and so the design calls for the mid-woofer to have its leads reversed on the mid-woofer to put it 180º out from the rest of the drivers. That's all relatively clear to me, though this is the first speaker I've ever built and have only the most rudimentary understanding of cross-overs. My question is this, "though the mid-woofer is hooked up electrically 'backwards' to the tweeter, is this simply correcting for (reversing) a signal from the cross-over that is itself out of phase with the tweeter?" I have run a test with an acoustic speaker phase tester and all the other drivers are in negative phase while acoustically the mid-woofer is positive. I would have thought that if the prescribed lead reversal was correcting for an out of phase signal from the cross-over, then the mid-woofer would be acoustically aligned in its timing with the rest of the drivers. So, if the mid-woofer is not only wired "backwards" but also is intended to be acoustically "backwards" that would say that I'm wrong about my assumptions regarding this all being a correction for an 180º signal with respect to the tweeter. In short, while the mid-woofer is wired in reverse, should all the drivers produce an acoustic wave that is in phase?
The second related question is this. If the mid-woofer IS intended to be out of phase with the rest of the drivers, then which way should I listen to these speakers, with the mid-woofer firing in the positive or with the other drivers firing in the positive? My Museatex Pa6i preamp has a phase control where I can easily choose.
Long-winded but I hope my question is clear enough.
My question comes out of my recent completion of a pair of Discovery-4 speakers designed by Troels Gravesen. The speaker is a 4-way with 2 crossovers. The top crossover (@2500 Hz) serves a mid-woofer and the tweeter. The crossovers are second order and so the design calls for the mid-woofer to have its leads reversed on the mid-woofer to put it 180º out from the rest of the drivers. That's all relatively clear to me, though this is the first speaker I've ever built and have only the most rudimentary understanding of cross-overs. My question is this, "though the mid-woofer is hooked up electrically 'backwards' to the tweeter, is this simply correcting for (reversing) a signal from the cross-over that is itself out of phase with the tweeter?" I have run a test with an acoustic speaker phase tester and all the other drivers are in negative phase while acoustically the mid-woofer is positive. I would have thought that if the prescribed lead reversal was correcting for an out of phase signal from the cross-over, then the mid-woofer would be acoustically aligned in its timing with the rest of the drivers. So, if the mid-woofer is not only wired "backwards" but also is intended to be acoustically "backwards" that would say that I'm wrong about my assumptions regarding this all being a correction for an 180º signal with respect to the tweeter. In short, while the mid-woofer is wired in reverse, should all the drivers produce an acoustic wave that is in phase?
The second related question is this. If the mid-woofer IS intended to be out of phase with the rest of the drivers, then which way should I listen to these speakers, with the mid-woofer firing in the positive or with the other drivers firing in the positive? My Museatex Pa6i preamp has a phase control where I can easily choose.
Long-winded but I hope my question is clear enough.
Polarity and phase are two things, but they can be used interchangeably in certain circumstances. Swapping polarity can be seen as shifting phase by 180 degrees. It's not the same as running the driver backwards. Imagine a phase plot, and the trace shifts by 180 degrees but is otherwise the same.. Yes, it is to do with the crossover.
1 - The phase response will vary over the driver's frequency range. How the frequency response behaves near the crossover points
depends on a number of factors.
2 - You can connect the entire speaker system either way, since there's no standard for absolute polarity in recordings.
If you hear a consistent difference that you prefer in one polarity, then use that one anyway.
depends on a number of factors.
2 - You can connect the entire speaker system either way, since there's no standard for absolute polarity in recordings.
If you hear a consistent difference that you prefer in one polarity, then use that one anyway.
Thank you for the replies Gentlemen! So it sounds like I shouldn't be alarmed that the polarity tester shows the mid-woofer opposite polarity to the rest of the drivers, acoustically. Mr. Gravesen DOES require the mid-woofer to be wired reversed, I was just having trouble trusting my wiring when the acoustic polarity was also reversed.
No, don't be alarmed. Ordinary phase shifts can add up to the amount of a change in polarity.
Sure the polarity of multi-way speaker is something that can be used to achieve the best result - one of the tools in the crossover designer tool box.
Normally the designer choose the woofer (20Hz up) with the "correct" polarity. All other speakers (lest's say 200Hz up) will have the convenient polarity set to achive the best acoustical result, which is the final result based on the phase changes at the following domains:
-Electrical: crossover changes phases according to frequencies
-Geometrical: position of each speaker to the final listener contributes to phase changes
-Acoustic: speakers changes phase according to frequencies
-Polarity: changes 180deg in the electrical domain
Example: you may have a crossover changing +90deg in a certain frequency for one speaker and changing -90deg for the other speaker. So, even if the polarity is the same for the 2 speakers, they will be electrically out of phase. If you change the polarity in one of them, you put them back in phase for that specifc frequency. It's a complex matter but nowadays much simpler resolved by simulation using actual speaker measurements.
Polarity would be always the same for each speaker, if there were no phase changes in crossover, geometry and speaker acoustic response.
In the real world, as I said, polarity is a decision of the speaker designer.
Normally the designer choose the woofer (20Hz up) with the "correct" polarity. All other speakers (lest's say 200Hz up) will have the convenient polarity set to achive the best acoustical result, which is the final result based on the phase changes at the following domains:
-Electrical: crossover changes phases according to frequencies
-Geometrical: position of each speaker to the final listener contributes to phase changes
-Acoustic: speakers changes phase according to frequencies
-Polarity: changes 180deg in the electrical domain
Example: you may have a crossover changing +90deg in a certain frequency for one speaker and changing -90deg for the other speaker. So, even if the polarity is the same for the 2 speakers, they will be electrically out of phase. If you change the polarity in one of them, you put them back in phase for that specifc frequency. It's a complex matter but nowadays much simpler resolved by simulation using actual speaker measurements.
Polarity would be always the same for each speaker, if there were no phase changes in crossover, geometry and speaker acoustic response.
In the real world, as I said, polarity is a decision of the speaker designer.
Thank you for your replies, all. I'm only slightly further along in understanding the effect of cross-overs on phase, etc. but I at least know that I needn't be concerned that my polarity tester sees that mid-woofer as reversed polarity. Really, that's what i needed. So again, thanks.