Home |
Search |
Today's Posts |
#81
|
|||
|
|||
Speakers testing
|
#82
|
|||
|
|||
Speakers testing
|
#84
|
|||
|
|||
Speakers testing
(Bob-Stanton) wrote in message . com...
(Svante) wrote in message om... Errh.. I mean, I don't have the book, but it seems as if the book might refer to a scientific paper (from 1983) or similar. My guess was that it was published in Journal of Audio Engineering Society (JAES), but that is of course a guess. Books often have a reference list and I wondered if you could look there to see where this 1983 paper was published? I found it. :-) Speaker Builder, published by Audio Amateur Press P.O.box 494, Peterborough, NH, 03458-0494 Telephone: 800 -524-9464 "Back issues are available." "A High Powered Satellite Speaker", by Joseph D'Appolito, Issue 4/1984 Actually, no, that's not "it." The original paper is, as mentioned, Joseph D'Appolito, "A Geometrical Approach to Eliminating Lobing Error in Multiway Loudspeakers," presented at the 74th Convention of the Audio Engineering Society, New York, 1983 October, Convention Preprint #2000. This is available by hardcopy from the AES. Try: http://www.aes.org/publications/preprints/search.html and enter 2000 as the preprint number. |
#85
|
|||
|
|||
Speakers testing
(Bob-Stanton) wrote in message . com...
(Svante) wrote in message om... Errh.. I mean, I don't have the book, but it seems as if the book might refer to a scientific paper (from 1983) or similar. My guess was that it was published in Journal of Audio Engineering Society (JAES), but that is of course a guess. Books often have a reference list and I wondered if you could look there to see where this 1983 paper was published? I found it. :-) Speaker Builder, published by Audio Amateur Press P.O.box 494, Peterborough, NH, 03458-0494 Telephone: 800 -524-9464 "Back issues are available." "A High Powered Satellite Speaker", by Joseph D'Appolito, Issue 4/1984 Actually, no, that's not "it." The original paper is, as mentioned, Joseph D'Appolito, "A Geometrical Approach to Eliminating Lobing Error in Multiway Loudspeakers," presented at the 74th Convention of the Audio Engineering Society, New York, 1983 October, Convention Preprint #2000. This is available by hardcopy from the AES. Try: http://www.aes.org/publications/preprints/search.html and enter 2000 as the preprint number. |
#86
|
|||
|
|||
Speakers testing
(Stewart Pinkerton) wrote in message ...
On 18 Dec 2003 09:57:24 -0800, (Svante) wrote: "Tony Pearce" wrote in message . au... It's a vertical array with twin bass/mids and a third-order crossover, and it's noted for a very smooth vertical dispersion pattern around the crossover frequency. Many such 'WTW' designs are around today, but not all are true d'Appolito arrays. Many are so-called '2.5 way' designs, which are conventional single bass/mid designs at the tweeter crossover point. Invented by audio legend Joe d'Apollito, hence the name. What would a '2.5 way' speaker be? I mean, where is the .5? |
#87
|
|||
|
|||
Speakers testing
(Stewart Pinkerton) wrote in message ...
On 18 Dec 2003 09:57:24 -0800, (Svante) wrote: "Tony Pearce" wrote in message . au... It's a vertical array with twin bass/mids and a third-order crossover, and it's noted for a very smooth vertical dispersion pattern around the crossover frequency. Many such 'WTW' designs are around today, but not all are true d'Appolito arrays. Many are so-called '2.5 way' designs, which are conventional single bass/mid designs at the tweeter crossover point. Invented by audio legend Joe d'Apollito, hence the name. What would a '2.5 way' speaker be? I mean, where is the .5? |
#88
|
|||
|
|||
Speakers testing
(Stewart Pinkerton) wrote in message ...
On 18 Dec 2003 09:57:24 -0800, (Svante) wrote: "Tony Pearce" wrote in message . au... It's a vertical array with twin bass/mids and a third-order crossover, and it's noted for a very smooth vertical dispersion pattern around the crossover frequency. Many such 'WTW' designs are around today, but not all are true d'Appolito arrays. Many are so-called '2.5 way' designs, which are conventional single bass/mid designs at the tweeter crossover point. Invented by audio legend Joe d'Apollito, hence the name. What would a '2.5 way' speaker be? I mean, where is the .5? |
#89
|
|||
|
|||
Speakers testing
"Svante" wrote in message om... (Stewart Pinkerton) wrote in message ... On 18 Dec 2003 09:57:24 -0800, (Svante) wrote: It's a vertical array with twin bass/mids and a third-order crossover, and it's noted for a very smooth vertical dispersion pattern around the crossover frequency. Many such 'WTW' designs are around today, but not all are true d'Appolito arrays. Many are so-called '2.5 way' designs, which are conventional single bass/mid designs at the tweeter crossover point. Invented by audio legend Joe d'Apollito, hence the name. What would a '2.5 way' speaker be? I mean, where is the .5? I guess he means it's half way between a 2 way and a 3 way design (2-1/2), but really it's a 3 way with 2 crossover points. The difference is that the "mid-range" overlaps with the woofer at low frequencies. A real D'Appolito WTW design would be 2 way of course with only one x-over point. And a D'Appolito 3 way is WMTMW. TonyP. |
#90
|
|||
|
|||
Speakers testing
"Svante" wrote in message om... (Stewart Pinkerton) wrote in message ... On 18 Dec 2003 09:57:24 -0800, (Svante) wrote: It's a vertical array with twin bass/mids and a third-order crossover, and it's noted for a very smooth vertical dispersion pattern around the crossover frequency. Many such 'WTW' designs are around today, but not all are true d'Appolito arrays. Many are so-called '2.5 way' designs, which are conventional single bass/mid designs at the tweeter crossover point. Invented by audio legend Joe d'Apollito, hence the name. What would a '2.5 way' speaker be? I mean, where is the .5? I guess he means it's half way between a 2 way and a 3 way design (2-1/2), but really it's a 3 way with 2 crossover points. The difference is that the "mid-range" overlaps with the woofer at low frequencies. A real D'Appolito WTW design would be 2 way of course with only one x-over point. And a D'Appolito 3 way is WMTMW. TonyP. |
#91
|
|||
|
|||
Speakers testing
"Svante" wrote in message om... (Stewart Pinkerton) wrote in message ... On 18 Dec 2003 09:57:24 -0800, (Svante) wrote: It's a vertical array with twin bass/mids and a third-order crossover, and it's noted for a very smooth vertical dispersion pattern around the crossover frequency. Many such 'WTW' designs are around today, but not all are true d'Appolito arrays. Many are so-called '2.5 way' designs, which are conventional single bass/mid designs at the tweeter crossover point. Invented by audio legend Joe d'Apollito, hence the name. What would a '2.5 way' speaker be? I mean, where is the .5? I guess he means it's half way between a 2 way and a 3 way design (2-1/2), but really it's a 3 way with 2 crossover points. The difference is that the "mid-range" overlaps with the woofer at low frequencies. A real D'Appolito WTW design would be 2 way of course with only one x-over point. And a D'Appolito 3 way is WMTMW. TonyP. |
#92
|
|||
|
|||
Speakers testing
"Tony Pearce" wrote in message . au...
"Svante" wrote in message om... (Stewart Pinkerton) wrote in message ... On 18 Dec 2003 09:57:24 -0800, (Svante) wrote: It's a vertical array with twin bass/mids and a third-order crossover, and it's noted for a very smooth vertical dispersion pattern around the crossover frequency. Many such 'WTW' designs are around today, but not all are true d'Appolito arrays. Many are so-called '2.5 way' designs, which are conventional single bass/mid designs at the tweeter crossover point. Invented by audio legend Joe d'Apollito, hence the name. What would a '2.5 way' speaker be? I mean, where is the .5? I guess he means it's half way between a 2 way and a 3 way design (2-1/2), but really it's a 3 way with 2 crossover points. The difference is that the "mid-range" overlaps with the woofer at low frequencies. A real D'Appolito WTW design would be 2 way of course with only one x-over point. And a D'Appolito 3 way is WMTMW. TonyP. So, 2.5 way would have two speakers playing bass, of which one also manages the midrange? Have you ever seen how such a crossover filter is designed? |
#93
|
|||
|
|||
Speakers testing
"Tony Pearce" wrote in message . au...
"Svante" wrote in message om... (Stewart Pinkerton) wrote in message ... On 18 Dec 2003 09:57:24 -0800, (Svante) wrote: It's a vertical array with twin bass/mids and a third-order crossover, and it's noted for a very smooth vertical dispersion pattern around the crossover frequency. Many such 'WTW' designs are around today, but not all are true d'Appolito arrays. Many are so-called '2.5 way' designs, which are conventional single bass/mid designs at the tweeter crossover point. Invented by audio legend Joe d'Apollito, hence the name. What would a '2.5 way' speaker be? I mean, where is the .5? I guess he means it's half way between a 2 way and a 3 way design (2-1/2), but really it's a 3 way with 2 crossover points. The difference is that the "mid-range" overlaps with the woofer at low frequencies. A real D'Appolito WTW design would be 2 way of course with only one x-over point. And a D'Appolito 3 way is WMTMW. TonyP. So, 2.5 way would have two speakers playing bass, of which one also manages the midrange? Have you ever seen how such a crossover filter is designed? |
#94
|
|||
|
|||
Speakers testing
"Tony Pearce" wrote in message . au...
"Svante" wrote in message om... (Stewart Pinkerton) wrote in message ... On 18 Dec 2003 09:57:24 -0800, (Svante) wrote: It's a vertical array with twin bass/mids and a third-order crossover, and it's noted for a very smooth vertical dispersion pattern around the crossover frequency. Many such 'WTW' designs are around today, but not all are true d'Appolito arrays. Many are so-called '2.5 way' designs, which are conventional single bass/mid designs at the tweeter crossover point. Invented by audio legend Joe d'Apollito, hence the name. What would a '2.5 way' speaker be? I mean, where is the .5? I guess he means it's half way between a 2 way and a 3 way design (2-1/2), but really it's a 3 way with 2 crossover points. The difference is that the "mid-range" overlaps with the woofer at low frequencies. A real D'Appolito WTW design would be 2 way of course with only one x-over point. And a D'Appolito 3 way is WMTMW. TonyP. So, 2.5 way would have two speakers playing bass, of which one also manages the midrange? Have you ever seen how such a crossover filter is designed? |
#95
|
|||
|
|||
Speakers testing
"Svante" wrote in message om... So, 2.5 way would have two speakers playing bass, of which one also manages the midrange? Have you ever seen how such a crossover filter is designed? It's just two separate low pass filters, one for the bass and another for the bass/mid, and a high pass for the tweeter. Nothing special required, other than to match levels. TonyP. |
#96
|
|||
|
|||
Speakers testing
"Svante" wrote in message om... So, 2.5 way would have two speakers playing bass, of which one also manages the midrange? Have you ever seen how such a crossover filter is designed? It's just two separate low pass filters, one for the bass and another for the bass/mid, and a high pass for the tweeter. Nothing special required, other than to match levels. TonyP. |
#97
|
|||
|
|||
Speakers testing
"Svante" wrote in message om... So, 2.5 way would have two speakers playing bass, of which one also manages the midrange? Have you ever seen how such a crossover filter is designed? It's just two separate low pass filters, one for the bass and another for the bass/mid, and a high pass for the tweeter. Nothing special required, other than to match levels. TonyP. |
#98
|
|||
|
|||
Speakers testing
"Tony Pearce" wrote in message . au...
"Svante" wrote in message om... So, 2.5 way would have two speakers playing bass, of which one also manages the midrange? Have you ever seen how such a crossover filter is designed? It's just two separate low pass filters, one for the bass and another for the bass/mid, and a high pass for the tweeter. Nothing special required, other than to match levels. That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. |
#99
|
|||
|
|||
Speakers testing
"Tony Pearce" wrote in message . au...
"Svante" wrote in message om... So, 2.5 way would have two speakers playing bass, of which one also manages the midrange? Have you ever seen how such a crossover filter is designed? It's just two separate low pass filters, one for the bass and another for the bass/mid, and a high pass for the tweeter. Nothing special required, other than to match levels. That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. |
#100
|
|||
|
|||
Speakers testing
"Tony Pearce" wrote in message . au...
"Svante" wrote in message om... So, 2.5 way would have two speakers playing bass, of which one also manages the midrange? Have you ever seen how such a crossover filter is designed? It's just two separate low pass filters, one for the bass and another for the bass/mid, and a high pass for the tweeter. Nothing special required, other than to match levels. That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. |
#101
|
|||
|
|||
Speakers testing
|
#102
|
|||
|
|||
Speakers testing
|
#103
|
|||
|
|||
Speakers testing
|
#104
|
|||
|
|||
Speakers testing
"Svante" wrote in message om... (Dick Pierce) wrote in message om... "Tony Pearce" wrote in message . au... It's just two separate low pass filters, one for the bass and another for the bass/mid, and a high pass for the tweeter. Nothing special required, other than to match levels. That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. Sure. Would not those approaches give problems with the frequency response and electrical impedance? Below the lower crossover frequency we would have two drivers playing = + 6dB and also an impedance of half the electrical impedance of one driver. Yeah, but that's what I meant about matching levels. Most speaker manufacturers are not worried about maintaining a particular electrical impedance. Have a look at an impedance curve for any ported speaker for example. The important thing is a flat SPL vs frequency curve (or whatever they deem satisfactory) and a load that is not *too* difficult to drive (although some fail on this one too :-) The increase in SPL from two drivers is obviously what they are after to compensate for LF roll off and/or power limitations. TonyP. |
#105
|
|||
|
|||
Speakers testing
"Svante" wrote in message om... (Dick Pierce) wrote in message om... "Tony Pearce" wrote in message . au... It's just two separate low pass filters, one for the bass and another for the bass/mid, and a high pass for the tweeter. Nothing special required, other than to match levels. That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. Sure. Would not those approaches give problems with the frequency response and electrical impedance? Below the lower crossover frequency we would have two drivers playing = + 6dB and also an impedance of half the electrical impedance of one driver. Yeah, but that's what I meant about matching levels. Most speaker manufacturers are not worried about maintaining a particular electrical impedance. Have a look at an impedance curve for any ported speaker for example. The important thing is a flat SPL vs frequency curve (or whatever they deem satisfactory) and a load that is not *too* difficult to drive (although some fail on this one too :-) The increase in SPL from two drivers is obviously what they are after to compensate for LF roll off and/or power limitations. TonyP. |
#106
|
|||
|
|||
Speakers testing
"Svante" wrote in message om... (Dick Pierce) wrote in message om... "Tony Pearce" wrote in message . au... It's just two separate low pass filters, one for the bass and another for the bass/mid, and a high pass for the tweeter. Nothing special required, other than to match levels. That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. Sure. Would not those approaches give problems with the frequency response and electrical impedance? Below the lower crossover frequency we would have two drivers playing = + 6dB and also an impedance of half the electrical impedance of one driver. Yeah, but that's what I meant about matching levels. Most speaker manufacturers are not worried about maintaining a particular electrical impedance. Have a look at an impedance curve for any ported speaker for example. The important thing is a flat SPL vs frequency curve (or whatever they deem satisfactory) and a load that is not *too* difficult to drive (although some fail on this one too :-) The increase in SPL from two drivers is obviously what they are after to compensate for LF roll off and/or power limitations. TonyP. |
#107
|
|||
|
|||
Speakers testing
(Svante) wrote in message . com...
(Dick Pierce) wrote in message om... That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. Would not those approaches give problems with the frequency response and electrical impedance? Below the lower crossover frequency we would have two drivers playing = + 6dB and also an impedance of half the electrical impedance of one driver. First, we would you tthink the electrical impedance issue is an important one? Remember that we're driving our speakers from pretty good approximations of voltage sources, and we are designing our speakers (hopefully) to have flat frequency when so driven. As to the frequency response problem, consider one scenario where this method actually solves a problem. Depending upon the baffle dimensions, below a certain frequency, one loses about 6 dB of axial output due to what's called the baffle diffraction effect. This is a result of the dispersion going from hemispherical at higher frequencies to spherical at lower frequencies. You get, essentially, a step in the response where below this frequency your output is lower than above. Fine, so you arrange your "2 1/2" way crossover such that both drivers are operating below that frequency, and only one is working above. You can, with careful design, thus arrantge so that the step in output of the speaker compensates for the roughly opposite step in response due to diffraction loss. Now, if you arrange the crossover such that the narrower band driver and its low-pass element is in parallel with the wider band driver, indeed you WILL get a lower impedance at lower frequencies, but, again, pick your points right (i.e., design your system as a system, a novel concept to many speaker "designers"), and you deal with the fact that at these lower frequencies, the overall impedance of the driovers are higher anyway, due to the increase impedance around resonance. The impedance "problem" is thus not that much of a problem. If you're worried about impedance, use drivers of an appropriate impedance to work in the design to begin with. Also, one can arrange system such the the drivers are driven in series, with a pass element (like a capacitor) around one of them that provides the low-pass function for that driver. |
#108
|
|||
|
|||
Speakers testing
(Svante) wrote in message . com...
(Dick Pierce) wrote in message om... That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. Would not those approaches give problems with the frequency response and electrical impedance? Below the lower crossover frequency we would have two drivers playing = + 6dB and also an impedance of half the electrical impedance of one driver. First, we would you tthink the electrical impedance issue is an important one? Remember that we're driving our speakers from pretty good approximations of voltage sources, and we are designing our speakers (hopefully) to have flat frequency when so driven. As to the frequency response problem, consider one scenario where this method actually solves a problem. Depending upon the baffle dimensions, below a certain frequency, one loses about 6 dB of axial output due to what's called the baffle diffraction effect. This is a result of the dispersion going from hemispherical at higher frequencies to spherical at lower frequencies. You get, essentially, a step in the response where below this frequency your output is lower than above. Fine, so you arrange your "2 1/2" way crossover such that both drivers are operating below that frequency, and only one is working above. You can, with careful design, thus arrantge so that the step in output of the speaker compensates for the roughly opposite step in response due to diffraction loss. Now, if you arrange the crossover such that the narrower band driver and its low-pass element is in parallel with the wider band driver, indeed you WILL get a lower impedance at lower frequencies, but, again, pick your points right (i.e., design your system as a system, a novel concept to many speaker "designers"), and you deal with the fact that at these lower frequencies, the overall impedance of the driovers are higher anyway, due to the increase impedance around resonance. The impedance "problem" is thus not that much of a problem. If you're worried about impedance, use drivers of an appropriate impedance to work in the design to begin with. Also, one can arrange system such the the drivers are driven in series, with a pass element (like a capacitor) around one of them that provides the low-pass function for that driver. |
#109
|
|||
|
|||
Speakers testing
(Svante) wrote in message . com...
(Dick Pierce) wrote in message om... That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. Would not those approaches give problems with the frequency response and electrical impedance? Below the lower crossover frequency we would have two drivers playing = + 6dB and also an impedance of half the electrical impedance of one driver. First, we would you tthink the electrical impedance issue is an important one? Remember that we're driving our speakers from pretty good approximations of voltage sources, and we are designing our speakers (hopefully) to have flat frequency when so driven. As to the frequency response problem, consider one scenario where this method actually solves a problem. Depending upon the baffle dimensions, below a certain frequency, one loses about 6 dB of axial output due to what's called the baffle diffraction effect. This is a result of the dispersion going from hemispherical at higher frequencies to spherical at lower frequencies. You get, essentially, a step in the response where below this frequency your output is lower than above. Fine, so you arrange your "2 1/2" way crossover such that both drivers are operating below that frequency, and only one is working above. You can, with careful design, thus arrantge so that the step in output of the speaker compensates for the roughly opposite step in response due to diffraction loss. Now, if you arrange the crossover such that the narrower band driver and its low-pass element is in parallel with the wider band driver, indeed you WILL get a lower impedance at lower frequencies, but, again, pick your points right (i.e., design your system as a system, a novel concept to many speaker "designers"), and you deal with the fact that at these lower frequencies, the overall impedance of the driovers are higher anyway, due to the increase impedance around resonance. The impedance "problem" is thus not that much of a problem. If you're worried about impedance, use drivers of an appropriate impedance to work in the design to begin with. Also, one can arrange system such the the drivers are driven in series, with a pass element (like a capacitor) around one of them that provides the low-pass function for that driver. |
#110
|
|||
|
|||
Speakers testing
In article ,
Svante wrote: It's just two separate low pass filters, one for the bass and another for the bass/mid That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. Would not those approaches give problems with the frequency response and electrical impedance? Nope. Below the lower crossover frequency we would have two drivers playing = + 6dB In an anechoic environment on-axis response increases 6dB once the sound waves are short enough. It's a gradual function having about 1dB of effect for wavelengths 3X baffle width and the full 6dB once they're 1/5th width. This is compensated for by either a low frequency boost as in the .5 way speaker or high frequency cut. In-room the number is probably more like 3-4.5dB, although the bass driver can be padded down to compensate. and also an impedance of half the electrical impedance of one driver. The amp won't care. -- a href="http://www.poohsticks.org/drew/"Home Page/a Life is a terminal sexually transmitted disease. |
#111
|
|||
|
|||
Speakers testing
In article ,
Svante wrote: It's just two separate low pass filters, one for the bass and another for the bass/mid That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. Would not those approaches give problems with the frequency response and electrical impedance? Nope. Below the lower crossover frequency we would have two drivers playing = + 6dB In an anechoic environment on-axis response increases 6dB once the sound waves are short enough. It's a gradual function having about 1dB of effect for wavelengths 3X baffle width and the full 6dB once they're 1/5th width. This is compensated for by either a low frequency boost as in the .5 way speaker or high frequency cut. In-room the number is probably more like 3-4.5dB, although the bass driver can be padded down to compensate. and also an impedance of half the electrical impedance of one driver. The amp won't care. -- a href="http://www.poohsticks.org/drew/"Home Page/a Life is a terminal sexually transmitted disease. |
#112
|
|||
|
|||
Speakers testing
In article ,
Svante wrote: It's just two separate low pass filters, one for the bass and another for the bass/mid That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. Would not those approaches give problems with the frequency response and electrical impedance? Nope. Below the lower crossover frequency we would have two drivers playing = + 6dB In an anechoic environment on-axis response increases 6dB once the sound waves are short enough. It's a gradual function having about 1dB of effect for wavelengths 3X baffle width and the full 6dB once they're 1/5th width. This is compensated for by either a low frequency boost as in the .5 way speaker or high frequency cut. In-room the number is probably more like 3-4.5dB, although the bass driver can be padded down to compensate. and also an impedance of half the electrical impedance of one driver. The amp won't care. -- a href="http://www.poohsticks.org/drew/"Home Page/a Life is a terminal sexually transmitted disease. |
#113
|
|||
|
|||
Speakers testing
(Dick Pierce) wrote in message om...
(Svante) wrote in message . com... (Dick Pierce) wrote in message om... That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. Would not those approaches give problems with the frequency response and electrical impedance? Below the lower crossover frequency we would have two drivers playing = + 6dB and also an impedance of half the electrical impedance of one driver. First, we would you tthink the electrical impedance issue is an important one? Remember that we're driving our speakers from pretty good approximations of voltage sources, and we are designing our speakers (hopefully) to have flat frequency when so driven. Yes, sure, but two 4 ohm speakers in parallel would perhaps give problems with the current driving capabilities. So, I guess 4 ohm speakers are out of the question, which perhaps is no big deal, I would agree to that. As to the frequency response problem, consider one scenario where this method actually solves a problem. Depending upon the baffle dimensions, below a certain frequency, one loses about 6 dB of axial output due to what's called the baffle diffraction effect. This is a result of the dispersion going from hemispherical at higher frequencies to spherical at lower frequencies. You get, essentially, a step in the response where below this frequency your output is lower than above. Like you and the other posters pointed out this appears like a good way to USE the effect, rather than having to compensate for it. Smart! One would have to select the crossover frequency carefully of course. I forgot about the diffraction effect, or possibly saw it as a separate issue, which it need not be. Fine, so you arrange your "2 1/2" way crossover such that both drivers are operating below that frequency, and only one is working above. You can, with careful design, thus arrantge so that the step in output of the speaker compensates for the roughly opposite step in response due to diffraction loss. Now, if you arrange the crossover such that the narrower band driver and its low-pass element is in parallel with the wider band driver, indeed you WILL get a lower impedance at lower frequencies, but, again, pick your points right (i.e., design your system as a system, a novel concept to many speaker "designers"), :-) and you deal with the fact that at these lower frequencies, the overall impedance of the driovers are higher anyway, due to the increase impedance around resonance. The impedance "problem" is thus not that much of a problem. Agree, but it is a *potential* problem that has to be dealt with. If you're worried about impedance, use drivers of an appropriate impedance to work in the design to begin with. Also, one can arrange system such the the drivers are driven in series, with a pass element (like a capacitor) around one of them that provides the low-pass function for that driver. |
#114
|
|||
|
|||
Speakers testing
(Dick Pierce) wrote in message om...
(Svante) wrote in message . com... (Dick Pierce) wrote in message om... That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. Would not those approaches give problems with the frequency response and electrical impedance? Below the lower crossover frequency we would have two drivers playing = + 6dB and also an impedance of half the electrical impedance of one driver. First, we would you tthink the electrical impedance issue is an important one? Remember that we're driving our speakers from pretty good approximations of voltage sources, and we are designing our speakers (hopefully) to have flat frequency when so driven. Yes, sure, but two 4 ohm speakers in parallel would perhaps give problems with the current driving capabilities. So, I guess 4 ohm speakers are out of the question, which perhaps is no big deal, I would agree to that. As to the frequency response problem, consider one scenario where this method actually solves a problem. Depending upon the baffle dimensions, below a certain frequency, one loses about 6 dB of axial output due to what's called the baffle diffraction effect. This is a result of the dispersion going from hemispherical at higher frequencies to spherical at lower frequencies. You get, essentially, a step in the response where below this frequency your output is lower than above. Like you and the other posters pointed out this appears like a good way to USE the effect, rather than having to compensate for it. Smart! One would have to select the crossover frequency carefully of course. I forgot about the diffraction effect, or possibly saw it as a separate issue, which it need not be. Fine, so you arrange your "2 1/2" way crossover such that both drivers are operating below that frequency, and only one is working above. You can, with careful design, thus arrantge so that the step in output of the speaker compensates for the roughly opposite step in response due to diffraction loss. Now, if you arrange the crossover such that the narrower band driver and its low-pass element is in parallel with the wider band driver, indeed you WILL get a lower impedance at lower frequencies, but, again, pick your points right (i.e., design your system as a system, a novel concept to many speaker "designers"), :-) and you deal with the fact that at these lower frequencies, the overall impedance of the driovers are higher anyway, due to the increase impedance around resonance. The impedance "problem" is thus not that much of a problem. Agree, but it is a *potential* problem that has to be dealt with. If you're worried about impedance, use drivers of an appropriate impedance to work in the design to begin with. Also, one can arrange system such the the drivers are driven in series, with a pass element (like a capacitor) around one of them that provides the low-pass function for that driver. |
#115
|
|||
|
|||
Speakers testing
(Dick Pierce) wrote in message om...
(Svante) wrote in message . com... (Dick Pierce) wrote in message om... That's one way to do it. Another is to cascade two low-pass filters. The first, say a second order low pass at, oh 2500 Hz, drivers the main mid-bass driver and the input to a second low-pass filter set to something lower, like 200 Hz. Would not those approaches give problems with the frequency response and electrical impedance? Below the lower crossover frequency we would have two drivers playing = + 6dB and also an impedance of half the electrical impedance of one driver. First, we would you tthink the electrical impedance issue is an important one? Remember that we're driving our speakers from pretty good approximations of voltage sources, and we are designing our speakers (hopefully) to have flat frequency when so driven. Yes, sure, but two 4 ohm speakers in parallel would perhaps give problems with the current driving capabilities. So, I guess 4 ohm speakers are out of the question, which perhaps is no big deal, I would agree to that. As to the frequency response problem, consider one scenario where this method actually solves a problem. Depending upon the baffle dimensions, below a certain frequency, one loses about 6 dB of axial output due to what's called the baffle diffraction effect. This is a result of the dispersion going from hemispherical at higher frequencies to spherical at lower frequencies. You get, essentially, a step in the response where below this frequency your output is lower than above. Like you and the other posters pointed out this appears like a good way to USE the effect, rather than having to compensate for it. Smart! One would have to select the crossover frequency carefully of course. I forgot about the diffraction effect, or possibly saw it as a separate issue, which it need not be. Fine, so you arrange your "2 1/2" way crossover such that both drivers are operating below that frequency, and only one is working above. You can, with careful design, thus arrantge so that the step in output of the speaker compensates for the roughly opposite step in response due to diffraction loss. Now, if you arrange the crossover such that the narrower band driver and its low-pass element is in parallel with the wider band driver, indeed you WILL get a lower impedance at lower frequencies, but, again, pick your points right (i.e., design your system as a system, a novel concept to many speaker "designers"), :-) and you deal with the fact that at these lower frequencies, the overall impedance of the driovers are higher anyway, due to the increase impedance around resonance. The impedance "problem" is thus not that much of a problem. Agree, but it is a *potential* problem that has to be dealt with. If you're worried about impedance, use drivers of an appropriate impedance to work in the design to begin with. Also, one can arrange system such the the drivers are driven in series, with a pass element (like a capacitor) around one of them that provides the low-pass function for that driver. |
#116
|
|||
|
|||
Speakers testing
"Svante" wrote in message om... Yes, sure, but two 4 ohm speakers in parallel would perhaps give problems with the current driving capabilities. So, I guess 4 ohm speakers are out of the question, which perhaps is no big deal, I would agree to that. Actually two by 4 ohm drivers in parallel (2 ohms nominal) are not that uncommon unfortunately. The manufactuers preferring to advertise a higher sensitivity (for 2.83V input), rather than worry whether your amp can drive them or not. There is still a myth that drivers should never be conected in series, even among speaker designers. As Dick points out, you can even connect drivers in a "2.5 way" system in series if you want. TonyP. |
#117
|
|||
|
|||
Speakers testing
"Svante" wrote in message om... Yes, sure, but two 4 ohm speakers in parallel would perhaps give problems with the current driving capabilities. So, I guess 4 ohm speakers are out of the question, which perhaps is no big deal, I would agree to that. Actually two by 4 ohm drivers in parallel (2 ohms nominal) are not that uncommon unfortunately. The manufactuers preferring to advertise a higher sensitivity (for 2.83V input), rather than worry whether your amp can drive them or not. There is still a myth that drivers should never be conected in series, even among speaker designers. As Dick points out, you can even connect drivers in a "2.5 way" system in series if you want. TonyP. |
#118
|
|||
|
|||
Speakers testing
"Svante" wrote in message om... Yes, sure, but two 4 ohm speakers in parallel would perhaps give problems with the current driving capabilities. So, I guess 4 ohm speakers are out of the question, which perhaps is no big deal, I would agree to that. Actually two by 4 ohm drivers in parallel (2 ohms nominal) are not that uncommon unfortunately. The manufactuers preferring to advertise a higher sensitivity (for 2.83V input), rather than worry whether your amp can drive them or not. There is still a myth that drivers should never be conected in series, even among speaker designers. As Dick points out, you can even connect drivers in a "2.5 way" system in series if you want. TonyP. |
#119
|
|||
|
|||
Speakers testing
"Stewart Pinkerton" wrote in message ... On 18 Dec 2003 09:57:24 -0800, (Svante) wrote: "Tony Pearce" wrote in message . au... You still miss the point I was making. The tweeter frequencies can be easily measured using MLS gating systems. The problem is with multi - woofer systems eg. D'Appolito designs commonly available these days. Near field measurements are made on one woofer, then MLS measurements are made for high frequencies. If the room height is small as it usually is, then errors will occur when combining near field measurements at frequencies lower than can be gated. Hmm, in that case I still fail to understand what you mean. Could you enlighten me on what a D'Appolito design is (I must confess I don't know). It's a vertical array with twin bass/mids and a third-order crossover, and it's noted for a very smooth vertical dispersion pattern around the crossover frequency. Many such 'WTW' designs are around today, but not all are true d'Appolito arrays. Many are so-called '2.5 way' designs, which are conventional single bass/mid designs at the tweeter crossover point. Invented by audio legend Joe d'Apollito, hence the name. A minor correction. Mr. d'Appolito does not claim he invented the MTm or if you prefer, the WTW design. This was his comment in an interview I read, I'm sorry, but I can't recall who he said was the first. There is no doubt that he popularized it. I recall in another article that he now prefers implementing these designs with 4th order xovers. If I can find the source of these atrributions I'll post it. Most likely SB. Are you speaking about a closed box or bass-reflex. Can be either, as the key element is the crossover region from bass/mids to tweeter, although most commercial designs seem to be sealed. I realise that the time between direct sound and the first reflection will determine the lowest frequency that can be used when gating an impulse response, but I fail to see why this would make combination of this and near-field measurement hard. Do you mean that it would force us to use the near-field measurement at a too high frequency? I believe that's what he's getting at. -- Stewart Pinkerton | Music is Art - Audio is Engineering |
#120
|
|||
|
|||
Speakers testing
"Stewart Pinkerton" wrote in message ... On 18 Dec 2003 09:57:24 -0800, (Svante) wrote: "Tony Pearce" wrote in message . au... You still miss the point I was making. The tweeter frequencies can be easily measured using MLS gating systems. The problem is with multi - woofer systems eg. D'Appolito designs commonly available these days. Near field measurements are made on one woofer, then MLS measurements are made for high frequencies. If the room height is small as it usually is, then errors will occur when combining near field measurements at frequencies lower than can be gated. Hmm, in that case I still fail to understand what you mean. Could you enlighten me on what a D'Appolito design is (I must confess I don't know). It's a vertical array with twin bass/mids and a third-order crossover, and it's noted for a very smooth vertical dispersion pattern around the crossover frequency. Many such 'WTW' designs are around today, but not all are true d'Appolito arrays. Many are so-called '2.5 way' designs, which are conventional single bass/mid designs at the tweeter crossover point. Invented by audio legend Joe d'Apollito, hence the name. A minor correction. Mr. d'Appolito does not claim he invented the MTm or if you prefer, the WTW design. This was his comment in an interview I read, I'm sorry, but I can't recall who he said was the first. There is no doubt that he popularized it. I recall in another article that he now prefers implementing these designs with 4th order xovers. If I can find the source of these atrributions I'll post it. Most likely SB. Are you speaking about a closed box or bass-reflex. Can be either, as the key element is the crossover region from bass/mids to tweeter, although most commercial designs seem to be sealed. I realise that the time between direct sound and the first reflection will determine the lowest frequency that can be used when gating an impulse response, but I fail to see why this would make combination of this and near-field measurement hard. Do you mean that it would force us to use the near-field measurement at a too high frequency? I believe that's what he's getting at. -- Stewart Pinkerton | Music is Art - Audio is Engineering |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
rec.audio.car FAQ (Part 4/5) | Car Audio | |||
rec.audio.car FAQ (Part 2/5) | Car Audio | |||
Comments about Blind Testing | High End Audio | |||
P/review of Jupiter Audio Europa speakers pt.1 | Audio Opinions | |||
Are there in-line amplifiers for speakers? | Tech |