Hey guys another stupid question....
when you put a bunch of drivers in a vertical array (aka line source) how does one compensate for or counter comb filtering?
I ask think becuase I have seen a few projects that use 4 or more 3" or 4" drivers like the Jordan 92 or W3 871 in a vertical array.
I was thinking of building a system like this (using a vertical array for fullrange drivers) with a dipole bass at the bottom (the dipole is another thread) for a friend.
thanks.
when you put a bunch of drivers in a vertical array (aka line source) how does one compensate for or counter comb filtering?
I ask think becuase I have seen a few projects that use 4 or more 3" or 4" drivers like the Jordan 92 or W3 871 in a vertical array.
I was thinking of building a system like this (using a vertical array for fullrange drivers) with a dipole bass at the bottom (the dipole is another thread) for a friend.
thanks.
Make sure the XO from Mid to Tweet is above the range of comb
filtering? There is a formula to figure out where combing will take
effect and if you just cross above that you shouldn't experiance
the comb effects.
Scratch That...
You want to use wide range drivers I have no idea at all what
you have to do in that case.
filtering? There is a formula to figure out where combing will take
effect and if you just cross above that you shouldn't experiance
the comb effects.
Scratch That...
You want to use wide range drivers I have no idea at all what
you have to do in that case.
line source
hi navin,
you can always curve your linesource, when the
drivers are spaced as close to each other as possible. if you
work out where your average listening distance is and then
create a simailar radius so you are looking at a concave baffle
hi navin,
you can always curve your linesource, when the
drivers are spaced as close to each other as possible. if you
work out where your average listening distance is and then
create a simailar radius so you are looking at a concave baffle
Re: line source
First the basics ...
You have to place your full-rangers as close together as possible to minimise the combing effect. Even so, if they're any sort of diameter, you're going to get some combing at some frequencies and in some listening positions. With the Nonsuch 4 you have to sit down to listen - stand and you lose some high frequency.
Start by modelling what you're likely to get. Try downloading this spreadsheet freeware from the FRD consortium: Vertical Polar Response Software for Line Arrays. This should tell you what you're likely to get in the way of frequency response at different distances and vertical alignments. You can also model what happens if you wire up your drivers so that the centre ones take more power than the outers, which can be a useful approach.
This is an excellent suggestion and a good approach. However, it does rather demand either designing the speaker for one particular listening room/set-up or designing some sort of variable curve (now that might be interesting
)
I'm playing with switchable ways of getting round the 'don't stand up' thing on the Nonsuch 4s and this is the subject of a separate thread: Low-pass filter network for mini-array. Perhaps it will be applicable to your system.
Hopefully, we'll get some of the line array specialists posting on this thread so we can all learn something.
Good luck and keep us posted.
Hi Navinnavin said:
First the basics ...
You have to place your full-rangers as close together as possible to minimise the combing effect. Even so, if they're any sort of diameter, you're going to get some combing at some frequencies and in some listening positions. With the Nonsuch 4 you have to sit down to listen - stand and you lose some high frequency.
Start by modelling what you're likely to get. Try downloading this spreadsheet freeware from the FRD consortium: Vertical Polar Response Software for Line Arrays. This should tell you what you're likely to get in the way of frequency response at different distances and vertical alignments. You can also model what happens if you wire up your drivers so that the centre ones take more power than the outers, which can be a useful approach.
exurbia said:
This is an excellent suggestion and a good approach. However, it does rather demand either designing the speaker for one particular listening room/set-up or designing some sort of variable curve (now that might be interesting
) I'm playing with switchable ways of getting round the 'don't stand up' thing on the Nonsuch 4s and this is the subject of a separate thread: Low-pass filter network for mini-array. Perhaps it will be applicable to your system.
Hopefully, we'll get some of the line array specialists posting on this thread so we can all learn something.
Good luck and keep us posted.
You could also read Dr. Jim Griffin's white paper on low cost line arrays (i.e. linus array). Dr. Griffin frequents the Madisound board.
Regards,
Dan
Regards,
Dan
I have given considerable thought to a line array for my own use but I am still in a quandry about it's ability to provide quality reproduction.
A friend has built and measured (MLSSA) several non-curved line arrays, the last being of 32 3" drivers. The measured frequency response was quite flat to about 18K which is where the drivers used begin to roll off. When listening to this array I got the feeling that something was not quite right. Even though I could tell that all the frequencies were there in the correct balance the sound seemed lifeless and without ambiance. This may be because the listening was done in a VERY large room - there was a back wall and floor but the side walls and ceiling were 25' away. So, the jury is still out.
As to comb effects:
First of all, I want to say that I believe there are 2 different effects that are labeled "comb." One is related to inter-driver spacing and the other is related to drivers being at different distance from the measurement point (listener). The comb effect that I refer to in the following text is related to different distances from the listener.
Consider a 32 driver array and pick any two drivers whose distance from the microphone are unequal. When measured, the typical comb effect will be evident, with nulls at frequency X and multiples thereof. Now pick any 2 other drivers and the results will be the same except that frequency X will be different. Keep on doing this until you have exhausted all 2 driver permutations. Now, overlay all the comb graphs and you will see that the peaks of one will tend to fill in the nulls of another, producing a relatively flat response curve. Vaying the measuring microphone height makes little difference so you could say that this array has no "sweet spot."
All well and good but the issue I have is that since the signals from different drivers are arriving at your ears at different times, the result is smearing. Many would argue that these arrival time difference are small enough to be inaudible. But, let's say for sake of argument that these varying arrival times are "wrong" and contribute to a subtle degradation in sound quality. If so, the previously mentioned "sweet spot" (which doesn't exist) is NOT sweet at any vertical position because regardless of the listening position you select, it is wrong.
Now, lets curve the array such that the distance between the listening position to each driver is identical. In this case there will be no comb effect. There is however a sweet spot and in context the sweet spot is "right." This sweet spot is small though and moving the listening position up or down results in the same kind of wrongness as the non-curved array. So, at least with a curved array you get one right spot! And I believe that the wrongness resulting from a curved array is less significant that the wrongness of a non-curved array.
There is much more to it than this though. A vertical array will have a transition frequency, below which it will be right and above which it will be wrong. The transition frequency will be higher in a curved array so you could say that if you are slightly out of the sweet spot the wrongness will at least be confined to a narrower band of frequencies. Many might say that this is good but I'm not sure because the transition from right to wrong happens fairly quickly.
Some 40 years ago I heard a speaker system that had a plasma tweeter and dynamic woofer. It was quite evident that the tweeter was right and the woofer wrong - you could hear where the tweeter was taking over. So, there was a transition frequency here as well. Unfortunately, the difference between the rightness and wrongness of the system was so obtrusive as to command your attention, thereby detracting from the presentation. So, in my opinion, the transition from wrong to right should be fairly wide so as to escape a focused notice.
The only solution to the wrongness of an array, as I see it, is to have a single center tweeter taking over below the transition frequency. With a curved array the crossover frequency can be higher than with a non-curved array.
I believe that the tapering of any array, whether it be a power tapering, a frequency tapering or a combination of the two is a non-solution if my previously defined conception of wrongness is correct - the later arrival times from the speakers toward the end of the array.
To me, a line array is simply a way to get an increase in SPL and power handling in the 100Hz to tweeter crossover band. And because of the greater output from a series/parallel connected array, multi-amping seems to be a must.
If this isn't clear, blame my communication skills.
A friend has built and measured (MLSSA) several non-curved line arrays, the last being of 32 3" drivers. The measured frequency response was quite flat to about 18K which is where the drivers used begin to roll off. When listening to this array I got the feeling that something was not quite right. Even though I could tell that all the frequencies were there in the correct balance the sound seemed lifeless and without ambiance. This may be because the listening was done in a VERY large room - there was a back wall and floor but the side walls and ceiling were 25' away. So, the jury is still out.
As to comb effects:
First of all, I want to say that I believe there are 2 different effects that are labeled "comb." One is related to inter-driver spacing and the other is related to drivers being at different distance from the measurement point (listener). The comb effect that I refer to in the following text is related to different distances from the listener.
Consider a 32 driver array and pick any two drivers whose distance from the microphone are unequal. When measured, the typical comb effect will be evident, with nulls at frequency X and multiples thereof. Now pick any 2 other drivers and the results will be the same except that frequency X will be different. Keep on doing this until you have exhausted all 2 driver permutations. Now, overlay all the comb graphs and you will see that the peaks of one will tend to fill in the nulls of another, producing a relatively flat response curve. Vaying the measuring microphone height makes little difference so you could say that this array has no "sweet spot."
All well and good but the issue I have is that since the signals from different drivers are arriving at your ears at different times, the result is smearing. Many would argue that these arrival time difference are small enough to be inaudible. But, let's say for sake of argument that these varying arrival times are "wrong" and contribute to a subtle degradation in sound quality. If so, the previously mentioned "sweet spot" (which doesn't exist) is NOT sweet at any vertical position because regardless of the listening position you select, it is wrong.
Now, lets curve the array such that the distance between the listening position to each driver is identical. In this case there will be no comb effect. There is however a sweet spot and in context the sweet spot is "right." This sweet spot is small though and moving the listening position up or down results in the same kind of wrongness as the non-curved array. So, at least with a curved array you get one right spot! And I believe that the wrongness resulting from a curved array is less significant that the wrongness of a non-curved array.
There is much more to it than this though. A vertical array will have a transition frequency, below which it will be right and above which it will be wrong. The transition frequency will be higher in a curved array so you could say that if you are slightly out of the sweet spot the wrongness will at least be confined to a narrower band of frequencies. Many might say that this is good but I'm not sure because the transition from right to wrong happens fairly quickly.
Some 40 years ago I heard a speaker system that had a plasma tweeter and dynamic woofer. It was quite evident that the tweeter was right and the woofer wrong - you could hear where the tweeter was taking over. So, there was a transition frequency here as well. Unfortunately, the difference between the rightness and wrongness of the system was so obtrusive as to command your attention, thereby detracting from the presentation. So, in my opinion, the transition from wrong to right should be fairly wide so as to escape a focused notice.
The only solution to the wrongness of an array, as I see it, is to have a single center tweeter taking over below the transition frequency. With a curved array the crossover frequency can be higher than with a non-curved array.
I believe that the tapering of any array, whether it be a power tapering, a frequency tapering or a combination of the two is a non-solution if my previously defined conception of wrongness is correct - the later arrival times from the speakers toward the end of the array.
To me, a line array is simply a way to get an increase in SPL and power handling in the 100Hz to tweeter crossover band. And because of the greater output from a series/parallel connected array, multi-amping seems to be a must.
If this isn't clear, blame my communication skills.
Bill Fitzpatrick said:
Interesting posting Bill. I'm interested in this "not quite rightness". Your description of "lifeless" and "without ambience" doesn't sound like the smearing that you associate with arrays.
Of course it's possible that the problem is the 18kHz roll-off but I somehow doubt it. The last time I visited Tim de Paravicini (of EAR-Yoshino) he played me an efficient speaker that didn't produce anything above about 12kHz. It sounded anything but lifeless or without ambience though.
I wonder whether it's a secondary effect of the array, not related to the array per se. What sort of cabinet loading is used for these 3" drivers? I ask because I associate terms like "lifeless" and "without ambience" with an overstuffed enclosure, particularly if the stuffing is near the rear of the drivers.
Often, arrays tend to be tall and very slim enclosures and the drivers don't have much room to breathe. I know from my experience with Bandor full-range drivers that in tall, slim enclosures stuffing is critical. This is one reason why so many array designers use open baffle designs or, like me, go to extraordinary lengths to minimise the requirement for enclosure stuffing.
I've certainly heard arrays, both full and mini that don't suffer from this problem.
7V:
Let me replace the word "lifeless" with "high frequency crispness." It was like a table saw blade that seems sharp but has lost it's edge.
What I heard could have easily been a two part problem - a lack of ambiance due to the far away walls and ceiling + time smearing.
I can't wrap myself around the concept of the drivers needing room to "breathe." In fact, knowing the builder of said array, there is probably only a little fiberglass behind the drivers. Personally I would load it up with something very absorbing to minimize rear reflections from arriving back at the cone - an errant rear wave is just another opportunity for time smearing. Either that or do it "open baffle."
I hope I can get to this project soon. I've been talking about it long enough, now it's time to actually do something.
Let me replace the word "lifeless" with "high frequency crispness." It was like a table saw blade that seems sharp but has lost it's edge.
What I heard could have easily been a two part problem - a lack of ambiance due to the far away walls and ceiling + time smearing.
I can't wrap myself around the concept of the drivers needing room to "breathe." In fact, knowing the builder of said array, there is probably only a little fiberglass behind the drivers. Personally I would load it up with something very absorbing to minimize rear reflections from arriving back at the cone - an errant rear wave is just another opportunity for time smearing. Either that or do it "open baffle."
I hope I can get to this project soon. I've been talking about it long enough, now it's time to actually do something.
To be honest, Bill, the vast majority of my speaker designing over the last 15 years has been with Bandor drive units and I suppose that I know the 50mm units as well as anyone. However, by the same token, I don't have so much experience with the other stuff out there - apart from when I stay with my father-in-law in Germany and the two of us play with his 'vintage paper cones and big boxes' gear,Bill Fitzpatrick said:
I don't know all the 'whys' of the 'whats' that I hear but I think that many (by no means all) Bandor customers come to the same conclusions.
What works for me works for me - each to his own.
However, although you can't wrap yourself around the concept of drivers needing room to 'breathe', I notice that you tell us that the builder of said array probably uses only a little fibreglass behind the drivers. Why do you say that? Is it a sort of "real designers don't need much damping" thing or more of a "he's completely crazy, too" thing? 🙂
Line Array Performance
I wanted to comment on Bill Fitzpatrick’s posting on line arrays. He comments on his friend’s array of 32 3-inch drivers that are apparently mounted on a flat baffle and used full range (up to at least 18 kHz). He finds the sound lifeless and dull with ambiance in a large room (the ceiling and side walls are 25’ away). I would say that most speakers would sound lifeless and without ambiance in that room--it is as if you have a speaker system in an anechoic chamber. A line array contours the sound into a cylindrical column in their near field (if you talking of a 32 x 3” = 96” high array, you would have a near field throughout that room) so all vertical sound energy is radiated perpendicular to the array. Hence, even with a normal height room very little sound would impinge from the ceiling or the floor. Without nearby side walls (for ambiance reinforcement) you effectively have an anechoic like situation with a line array in that large room.
Furthermore, if the 32-driver array is operating (as Bill states) full range up to 18 kHz, then this line violates well known criterion for line array driver spacing. The accepted isophase (creation of a single elongated line source) criterion for center-to-center spacing between drivers in a line array is no more than one wavelength (WL) separation at their highest operating frequency. Above that spacing the directivity of the array starts to degrade and the first cancellation (severe comb lines) occurs at 2 WLs spacing. The speaker’s sensitivity declines as the directivity suffers. For 3-inch drivers (assuming minimal flanges) the one WL spacing would suggest an operating frequency no greater than 4520 Hz and certainly comb lines would show-up at 9040 Hz. Bottom line is that this line array needs to be crossed over to a tweeter line at less than 4500 Hz for best sound.
Bill comments on potential sound smearing because of different paths lengths between the speaker and the listeners. He also suggests that a curved baffle that focuses a line of speakers the same distance from the listener would alleviate the ‘smearing’ issue. My work suggests a properly implement line array (with a straight baffle) that operates in the near field doesn’t have a ‘smearing’ problem. I do notice a sound ‘bloom’ that occurs if the woofer and tweeter lines in a two-way array differ significantly in their lengths. The best solution for the ‘bloom’ issue that I have found is to use a slight power tapering of the array (higher amplitude in the center of the array). The power taper works on the precedence principle (Haas Effect) so that the listener hears sound that is slightly biased toward the center of the array. Bottom line in my opinion a curved baffle isn’t necessary to produce good sound from an array. A flat baffle line array has significantly more latitude for listeners in the room versus the specific focal point or distance of a curved baffle.
His comments on the sweet spot of a line array should be considered as well. A near field line array places the listener in the near field of the source across the majority of the frequency band. Therefore, the vertical sound fall off within the room is only 3 dB per doubling of the distance from the source. Point source speakers exhibit a 6 dB per doubling of distance fall off. Hence, a line array listener will notice less variation in the sound volume from the front to the back of the room as well as from side to side within the room. In conclusion, a near field line array’s sweet spot is larger than with point source speakers. I call it a sweet area. You can observe that you can stand close to one speaker and still hear the opposite speaker! Some listeners are at first concerned by a line array’s broader image and wider sound stage but I think it is a improvement versus head-in-a-vise listening of most point source speakers.
My take on Bill’s concerns about line arrays is that he really needs to hear a properly designed near field line array, that mitigates the comb line issue, and is located in a room that will help recover some ambiance. He’ll enjoy that experience.
I wanted to comment on Bill Fitzpatrick’s posting on line arrays. He comments on his friend’s array of 32 3-inch drivers that are apparently mounted on a flat baffle and used full range (up to at least 18 kHz). He finds the sound lifeless and dull with ambiance in a large room (the ceiling and side walls are 25’ away). I would say that most speakers would sound lifeless and without ambiance in that room--it is as if you have a speaker system in an anechoic chamber. A line array contours the sound into a cylindrical column in their near field (if you talking of a 32 x 3” = 96” high array, you would have a near field throughout that room) so all vertical sound energy is radiated perpendicular to the array. Hence, even with a normal height room very little sound would impinge from the ceiling or the floor. Without nearby side walls (for ambiance reinforcement) you effectively have an anechoic like situation with a line array in that large room.
Furthermore, if the 32-driver array is operating (as Bill states) full range up to 18 kHz, then this line violates well known criterion for line array driver spacing. The accepted isophase (creation of a single elongated line source) criterion for center-to-center spacing between drivers in a line array is no more than one wavelength (WL) separation at their highest operating frequency. Above that spacing the directivity of the array starts to degrade and the first cancellation (severe comb lines) occurs at 2 WLs spacing. The speaker’s sensitivity declines as the directivity suffers. For 3-inch drivers (assuming minimal flanges) the one WL spacing would suggest an operating frequency no greater than 4520 Hz and certainly comb lines would show-up at 9040 Hz. Bottom line is that this line array needs to be crossed over to a tweeter line at less than 4500 Hz for best sound.
Bill comments on potential sound smearing because of different paths lengths between the speaker and the listeners. He also suggests that a curved baffle that focuses a line of speakers the same distance from the listener would alleviate the ‘smearing’ issue. My work suggests a properly implement line array (with a straight baffle) that operates in the near field doesn’t have a ‘smearing’ problem. I do notice a sound ‘bloom’ that occurs if the woofer and tweeter lines in a two-way array differ significantly in their lengths. The best solution for the ‘bloom’ issue that I have found is to use a slight power tapering of the array (higher amplitude in the center of the array). The power taper works on the precedence principle (Haas Effect) so that the listener hears sound that is slightly biased toward the center of the array. Bottom line in my opinion a curved baffle isn’t necessary to produce good sound from an array. A flat baffle line array has significantly more latitude for listeners in the room versus the specific focal point or distance of a curved baffle.
His comments on the sweet spot of a line array should be considered as well. A near field line array places the listener in the near field of the source across the majority of the frequency band. Therefore, the vertical sound fall off within the room is only 3 dB per doubling of the distance from the source. Point source speakers exhibit a 6 dB per doubling of distance fall off. Hence, a line array listener will notice less variation in the sound volume from the front to the back of the room as well as from side to side within the room. In conclusion, a near field line array’s sweet spot is larger than with point source speakers. I call it a sweet area. You can observe that you can stand close to one speaker and still hear the opposite speaker! Some listeners are at first concerned by a line array’s broader image and wider sound stage but I think it is a improvement versus head-in-a-vise listening of most point source speakers.
My take on Bill’s concerns about line arrays is that he really needs to hear a properly designed near field line array, that mitigates the comb line issue, and is located in a room that will help recover some ambiance. He’ll enjoy that experience.
7V:
I am saying that while he is good enough to realize that making a tweeter flush to the baffle has an effect on the response curve, he would not necessarily compare two different stuffings. You might say he is lax.
So here is a good question. There a two usual ways to mount a tweeter. One way is where the front of the faceplate is flush with the baffle and produces response curve A and the other way where the tweeter is simply mounted "on" the baffle producing response curve B. Now curve B will contain some response irregularities due to the diffraction effects caused by the discontinuity around the rim of the faceplate. Now, let's introduce a mid-range which happens to have response irregularities which exactly cancel out the irregularities of the non-flush mounted tweeter providing a total response curve that is flatter that it would be if the tweeter were to be mounted flush. The question is, do we mount the tweeter flush or not? I know that people frequently visiting this forum and building speakers will mount it flush unless they are too lazy to route out a relief. But is it always the right thing to do?
Further, the discontinuity around the rim of a typical non-flush mounted tweeter causes it to essentially re-radiate part of it's output at the lower frequencies about .3 milli-seconds after the initial output, this being the actual cause of the diffraction effect. Does this .3 milli-second delayed re-radiation degrade the quality of the sound? I don't know. Some say that our ear/brain minimizes it so severely that it becomes unimportant. An interesting analogy would be dust on reading glasses. How many particles of dust does it take before we are aware that our glasses need cleaning? 1? 100? 100,000? And, does our vision improve if we clean our glasses BEFORE we are aware that they need it? Pardon the pun but there is more here in sound reproduction than meets the eye. We have one continuim of effects layered upon another and another and another . . . One thing that I believe has never been addressed is how the various layers interact and impact the end result. There a lot of IFs, ANDs, ORs and ELSEes going on that we don't have the first clue about. Then, add on top of all this, the matter of how much error are we WILLING to accept?
I recall a particularly befuddling experience I had 20 years ago when I went to a concert and, with things still relatively fresh in my mind, returned to our stereo store and played what amounted to the same thing - same performers, same compositions. It sounded the same as the live performance to me. Was I suffering from an untrained ear? Did I hear what I WANTED to hear? I don't know. Some years prior to that I took some speakers which I co-designed to a stereo store to see if they wanted to carry them. The guy listening to them said they sounded good and wanted to show me something. He switched in an equalizer and pushed all the frequency controls to the bottom. He then pushed EVERY OTHER control to the top. And, you know what? The speakers still sounded good! Why?
Now, off the soapbox and back to practical matters.
Maybe you, or anybody, could explain to me the "need to breathe" and why one would not want to completely kill the rear wave?
And, that's another concept I can't wrap around - "What works for me works for me - each to his own." Do people grinding a telescope mirror to the precision of an Angstrom so they can get a clearer view say that?
I get clearer vision when I wear my prescription glasses. Do I wear them? No. I don't want to see things clear. I don't want to see the dirt on things, the scratches on the bar top. I don't want to see people's crummy complexions and I don't want to see what poor workmanship there is in the commonplace things. I didn't see the minute chips in the paint on my car until I looked at it with my glasses on. I was miserable. I'm happy with the illusions I can create when things are fuzzy. Ignorance is indeed bliss. Perhaps I have lost my mind my mind my mind . . . .
Jim:
Despite the fact that you got my last name wrong I bear no grudges and intend to fully digest what you say. I hope you will not cause me to start wearing my glasses full time. I was happy before you came along and, god willing, I'll be happy again.
I am saying that while he is good enough to realize that making a tweeter flush to the baffle has an effect on the response curve, he would not necessarily compare two different stuffings. You might say he is lax.
So here is a good question. There a two usual ways to mount a tweeter. One way is where the front of the faceplate is flush with the baffle and produces response curve A and the other way where the tweeter is simply mounted "on" the baffle producing response curve B. Now curve B will contain some response irregularities due to the diffraction effects caused by the discontinuity around the rim of the faceplate. Now, let's introduce a mid-range which happens to have response irregularities which exactly cancel out the irregularities of the non-flush mounted tweeter providing a total response curve that is flatter that it would be if the tweeter were to be mounted flush. The question is, do we mount the tweeter flush or not? I know that people frequently visiting this forum and building speakers will mount it flush unless they are too lazy to route out a relief. But is it always the right thing to do?
Further, the discontinuity around the rim of a typical non-flush mounted tweeter causes it to essentially re-radiate part of it's output at the lower frequencies about .3 milli-seconds after the initial output, this being the actual cause of the diffraction effect. Does this .3 milli-second delayed re-radiation degrade the quality of the sound? I don't know. Some say that our ear/brain minimizes it so severely that it becomes unimportant. An interesting analogy would be dust on reading glasses. How many particles of dust does it take before we are aware that our glasses need cleaning? 1? 100? 100,000? And, does our vision improve if we clean our glasses BEFORE we are aware that they need it? Pardon the pun but there is more here in sound reproduction than meets the eye. We have one continuim of effects layered upon another and another and another . . . One thing that I believe has never been addressed is how the various layers interact and impact the end result. There a lot of IFs, ANDs, ORs and ELSEes going on that we don't have the first clue about. Then, add on top of all this, the matter of how much error are we WILLING to accept?
I recall a particularly befuddling experience I had 20 years ago when I went to a concert and, with things still relatively fresh in my mind, returned to our stereo store and played what amounted to the same thing - same performers, same compositions. It sounded the same as the live performance to me. Was I suffering from an untrained ear? Did I hear what I WANTED to hear? I don't know. Some years prior to that I took some speakers which I co-designed to a stereo store to see if they wanted to carry them. The guy listening to them said they sounded good and wanted to show me something. He switched in an equalizer and pushed all the frequency controls to the bottom. He then pushed EVERY OTHER control to the top. And, you know what? The speakers still sounded good! Why?
Now, off the soapbox and back to practical matters.
Maybe you, or anybody, could explain to me the "need to breathe" and why one would not want to completely kill the rear wave?
And, that's another concept I can't wrap around - "What works for me works for me - each to his own." Do people grinding a telescope mirror to the precision of an Angstrom so they can get a clearer view say that?
I get clearer vision when I wear my prescription glasses. Do I wear them? No. I don't want to see things clear. I don't want to see the dirt on things, the scratches on the bar top. I don't want to see people's crummy complexions and I don't want to see what poor workmanship there is in the commonplace things. I didn't see the minute chips in the paint on my car until I looked at it with my glasses on. I was miserable. I'm happy with the illusions I can create when things are fuzzy. Ignorance is indeed bliss. Perhaps I have lost my mind my mind my mind . . . .
Jim:
Despite the fact that you got my last name wrong I bear no grudges and intend to fully digest what you say. I hope you will not cause me to start wearing my glasses full time. I was happy before you came along and, god willing, I'll be happy again.
Bill, Sorry About the Name
Bill,
Please forgive my screw-up. I composed my message from memory which was not a good thing on which to rely.
Jim
Bill,
Please forgive my screw-up. I composed my message from memory which was not a good thing on which to rely.
Jim
Re: Bill, Sorry About the Name
As the popular expression these days goes, "chill." One of it's meanings is to relax and don't be so serious. Fitzmaurice is fine - it lends and air of dignity. How about "Lord Fitz Maurice." I like that a lot.
Jim Griffin said:
As the popular expression these days goes, "chill." One of it's meanings is to relax and don't be so serious. Fitzmaurice is fine - it lends and air of dignity. How about "Lord Fitz Maurice." I like that a lot.
To get rid of the comb filtering you need to use small FRs very closely spaced. Jim's paper gives the math -- IIRC you need a driver smaller than 1.5" to approach 20k (that means like 144 holes & drivers to wire if you use something like the HiVi 1").
Another good paper on the subject is "The Line-source Loudspeaker and its applications" by Paul H Taylor" (Jan 1963, published in the Mar 64 British Kinematography)
In particlular it has a graphic display of comb-filtering. With 5 sources, this should be of particular interest to 7V, shows what happens when you stand up. Note: the chart is for a wavelength 1.5 the driver separtaion, higer frequencies will have a smaller "open" area.
dave
Another good paper on the subject is "The Line-source Loudspeaker and its applications" by Paul H Taylor" (Jan 1963, published in the Mar 64 British Kinematography)
In particlular it has a graphic display of comb-filtering. With 5 sources, this should be of particular interest to 7V, shows what happens when you stand up. Note: the chart is for a wavelength 1.5 the driver separtaion, higer frequencies will have a smaller "open" area.
dave
Attachments
Drivers have a mass/compliance ratio that defines what it wants to 'feel' for it to behave in a linear fashion, or at least as linear as its design allows. Too much stuffing adds acoustic mass, causing it to be overdamped WRT its specs. If the driver is underdamped to start with, then this can be a way to 'tighten' it up, but with a well done design, even a little bit of stuffing can overdamp it to the point of muted dynamics.
GM
I would agree that this in the bass range where the air in the box will act to control (damp) the drivers motion, and in turn its output. However, in the mid-bass range and up any stuffing, whether a little or a lot, should have little or no effect on the drivers performance, save to damp the reflected soundwaves in the boxGM said:
- Status
- Not open for further replies.