Ravens and Orcas have a special, non-European meaning to anyone that's lived in the Pacific Northwest. They are a living link back to the Old Times, the world of ancient forests (a few of which still stand, if you know where to look) and the rhythms of tribal life, the annual salmon runs, the potlatch, totem poles, the winter rains in the deep dark forest, and the crash of the cold Pacific surf on the rocky beach. There's no European equivalent, exactly, except for the wilder regions of Scotland and Scandinavia, but the tribal resonance isn't there.
That's the odd thing about North America - even though smallpox and the later campaigns of extermination and Bantustans (reservations) had a devastating impact on the native populations, the native people were tenacious and survived. They've left a subtle psychic imprint on the areas where they lived, which you can feel if you stay for a little while. Where I live now are the plains-next-to-the-great-mountains, with high, clear, near-desert air, big-sky country, and the realm of buffalo and prairie dogs, with the sun glinting off the first snows of the season in the mountains. It wasn't until I met a few tribal people and asked about the stories of this area that I finally felt "in place" here. It still feels raw and new, but that's part of the character of a high desert. I like hearing the coyotes howling and owls hooting almost every night, and seeing the red-tailed hawks silently circling in the sky in the daytime.
The Pacific Northwest is all about deep, dark forests that feel very different than European forests, with silent bands of coyotes and wolves, and ravens looking down from the immense trees and cawing down at you with their usual contempt for mere humans. The coyotes and wolves have retreated away from the urban areas, but the ravens are everywhere, not afraid of humans in the slightest. When a raven looks at you - which is just about any time you go outside in the Northwest - you have the oddest feeling they're looking right through you, not at you.
Um, where were we? Oh yes, panomaniac was asking about lowpass filtering. I'm sure most of the readers are aware the transfer function of a driver is the vector sum of the driver itself and all prior electrical filtering, whether active or passive. Thus, if a driver has 1st-order lowpass function, and you add a complementary 2nd-order electrical lowpass (whether active or passive), there's a net 3rd-order rolloff for the whole system.
As long as we describe a single driver (and it's in a minimum-phase region) and a mininum-phase electrical network, the sum of the driver+network remains minimum-phase.
OK, let's a go a little further. Lowpass networks have the property of delaying the signal. An easy way to synthesize a delay network is construct a Nth-order Bessel or Gaussian lowpass filter - these filters are free of the time-domain overshoots of high-order Butterworth of Chebychev filters, and have reasonably flat delays within the passband of the filter.
Since the total system has has uniform delay - as much as we want, limited only by the complexity of the filter network - it moves acoustic center of radiation away from the listener. The total delay is the sum of the delay of the driver + lowpass delay.
We don't have the same option for the tweeter. Nth-order Bessel highpass networks do have overshoots in the time domain, just not as much as Butterworth and Chebychev filters. Short of physically moving the tweeter backward or using digital filtering, there's no easy way to move the tweeter backward. So the only drivers we can move backwards electrically are the woofers, not the tweeters. This has the implication that if the baffle is slanted back too much, we have the option of electrically delaying the woofers, something that's not so easy to do with the tweeter.
Now as for how much mismatch between the 8 and 18-inch driver we can accept, that's a good question. The frequencies are low, the wavelengths large, and the direct-arrival wave cannot be separated from the first room reflections, since we're deep into the room-mode region at the 18-inch lowpass filter frequency. As a percentage, it's kind of ridiculous - 120 Hz is 115 inches long, and we're talking about 5 inches of offset! That's 1/23 of a wavelength - not much, compared to the much more severe polar effects a decade higher, close to the 8-inch crossover frequency.
In practical terms, the baffle slope probably doesn't make any difference at all, except for aiming direct on-axis colorations (and those are persistent and hard to remove) up and away from the listener. The on-axis colorations are both from the driver(s) and standing-wave/diffraction effects of the baffle itself, and tend to concentrate in a narrow beam right on-axis. That's why most loudspeakers sound best at least slightly off-axis. By pointing the whole thing upward, it relaxes the necessity for tricky L/R aiming, where you have to trade image quality against on-axis colorations.
That's the odd thing about North America - even though smallpox and the later campaigns of extermination and Bantustans (reservations) had a devastating impact on the native populations, the native people were tenacious and survived. They've left a subtle psychic imprint on the areas where they lived, which you can feel if you stay for a little while. Where I live now are the plains-next-to-the-great-mountains, with high, clear, near-desert air, big-sky country, and the realm of buffalo and prairie dogs, with the sun glinting off the first snows of the season in the mountains. It wasn't until I met a few tribal people and asked about the stories of this area that I finally felt "in place" here. It still feels raw and new, but that's part of the character of a high desert. I like hearing the coyotes howling and owls hooting almost every night, and seeing the red-tailed hawks silently circling in the sky in the daytime.
The Pacific Northwest is all about deep, dark forests that feel very different than European forests, with silent bands of coyotes and wolves, and ravens looking down from the immense trees and cawing down at you with their usual contempt for mere humans. The coyotes and wolves have retreated away from the urban areas, but the ravens are everywhere, not afraid of humans in the slightest. When a raven looks at you - which is just about any time you go outside in the Northwest - you have the oddest feeling they're looking right through you, not at you.
Um, where were we? Oh yes, panomaniac was asking about lowpass filtering. I'm sure most of the readers are aware the transfer function of a driver is the vector sum of the driver itself and all prior electrical filtering, whether active or passive. Thus, if a driver has 1st-order lowpass function, and you add a complementary 2nd-order electrical lowpass (whether active or passive), there's a net 3rd-order rolloff for the whole system.
As long as we describe a single driver (and it's in a minimum-phase region) and a mininum-phase electrical network, the sum of the driver+network remains minimum-phase.
OK, let's a go a little further. Lowpass networks have the property of delaying the signal. An easy way to synthesize a delay network is construct a Nth-order Bessel or Gaussian lowpass filter - these filters are free of the time-domain overshoots of high-order Butterworth of Chebychev filters, and have reasonably flat delays within the passband of the filter.
Since the total system has has uniform delay - as much as we want, limited only by the complexity of the filter network - it moves acoustic center of radiation away from the listener. The total delay is the sum of the delay of the driver + lowpass delay.
We don't have the same option for the tweeter. Nth-order Bessel highpass networks do have overshoots in the time domain, just not as much as Butterworth and Chebychev filters. Short of physically moving the tweeter backward or using digital filtering, there's no easy way to move the tweeter backward. So the only drivers we can move backwards electrically are the woofers, not the tweeters. This has the implication that if the baffle is slanted back too much, we have the option of electrically delaying the woofers, something that's not so easy to do with the tweeter.
Now as for how much mismatch between the 8 and 18-inch driver we can accept, that's a good question. The frequencies are low, the wavelengths large, and the direct-arrival wave cannot be separated from the first room reflections, since we're deep into the room-mode region at the 18-inch lowpass filter frequency. As a percentage, it's kind of ridiculous - 120 Hz is 115 inches long, and we're talking about 5 inches of offset! That's 1/23 of a wavelength - not much, compared to the much more severe polar effects a decade higher, close to the 8-inch crossover frequency.
In practical terms, the baffle slope probably doesn't make any difference at all, except for aiming direct on-axis colorations (and those are persistent and hard to remove) up and away from the listener. The on-axis colorations are both from the driver(s) and standing-wave/diffraction effects of the baffle itself, and tend to concentrate in a narrow beam right on-axis. That's why most loudspeakers sound best at least slightly off-axis. By pointing the whole thing upward, it relaxes the necessity for tricky L/R aiming, where you have to trade image quality against on-axis colorations.
A possibly quite unwanted effect of the baffle slant is aiming the 18-inch driver at the listener. The bigger drivers have more undesirable on-axis colorations, especially narrow-angle dustcap colorations. I don't think removing the dustcap from a 15 or 18-inch driver is a good idea, since it is very likely part of a carefully-designed VC-cooling system. That's something we want to keep - even we're far away from prosound SPL's, the rapid VC cooling of professional drivers is very desirable, since it shortens the time constant for heating and cooling of the voice coil.
As for listening distances, I think 2.5 meters (100 inches or 8 feet 4 inches) is probably the absolute mininum, for reasons of sonic cohesion. The wavefronts are pretty lumpy the closer you get, enough so they considerably interfere with measurements at less than 2 meters. 1 meter, for example, is useless for evaluating a multiway system, since if it is equalized at that distance, it will be grossly in error at larger distances.
As for listening distances, I think 2.5 meters (100 inches or 8 feet 4 inches) is probably the absolute mininum, for reasons of sonic cohesion. The wavefronts are pretty lumpy the closer you get, enough so they considerably interfere with measurements at less than 2 meters. 1 meter, for example, is useless for evaluating a multiway system, since if it is equalized at that distance, it will be grossly in error at larger distances.
Re: Raven
I simply think that for low frequencies the 18 inch is going to radiate with much more of its cone than the 8 inch for high mids, and the acoustic centre of the 18 inch is going to be located further enough infront from its vc. This will bring acoustically the 18 inch and 8 inch significantly closer than their physical distances.
Because its rough guessing at this stage since not even the final drivers have been chosen, I would not determine the slant now. I would build an experimental baffle when the hands on project starts, and by using 2 sticks at the back and a holding weight at its front low edge, I would vary the slant with the final drivers mounted, final slopes chosen, measure, listen, and then build and paint a secure, final, 'Raven black' baffle. Slanted as proven. Better be safe.
Lynn Olson said:
I simply think that for low frequencies the 18 inch is going to radiate with much more of its cone than the 8 inch for high mids, and the acoustic centre of the 18 inch is going to be located further enough infront from its vc. This will bring acoustically the 18 inch and 8 inch significantly closer than their physical distances.
Because its rough guessing at this stage since not even the final drivers have been chosen, I would not determine the slant now. I would build an experimental baffle when the hands on project starts, and by using 2 sticks at the back and a holding weight at its front low edge, I would vary the slant with the final drivers mounted, final slopes chosen, measure, listen, and then build and paint a secure, final, 'Raven black' baffle. Slanted as proven. Better be safe.
Thinking about it some more, well, if you slant the baffle, the worst crud of the 18" driver- which is no prize at 1.5 kHz - is aimed at the listener. Then again, at least it's lowpass-filtered, with a serious notch filter at the offending frequency if required.
By contrast, listening to the 8-inch driver slightly off-axis will almost certainly be an improvement - more importantly, the on-axis colorations fall both into the desired passband and mid/tweeter crossover range, where they have to be handled a lot more delicately, since it affects inter-driver phase relations.
By contrast, listening to the 8-inch driver slightly off-axis will almost certainly be an improvement - more importantly, the on-axis colorations fall both into the desired passband and mid/tweeter crossover range, where they have to be handled a lot more delicately, since it affects inter-driver phase relations.
Re: Raven
That makes sense. There will be some slant, of that I'm sure, just not how much. If the bass arrives a little early, that's an error I prefer - it's tweeter-ahead-of-bass time distortion I object to more strongly. The system will be slightly biased in this direction, giving a wider choice of listening distances and heights.
salas said:
That makes sense. There will be some slant, of that I'm sure, just not how much. If the bass arrives a little early, that's an error I prefer - it's tweeter-ahead-of-bass time distortion I object to more strongly. The system will be slightly biased in this direction, giving a wider choice of listening distances and heights.
Interesting Links and Thanks
Yes, it's true, I enjoy reading Romy the Cat's "goodsoundclub" site - mostly for the endlessly entertaining invective about the depraved and corrupt high-end industry. Although I'm pretty sure he regards me - along with the rest of the DIY community - as just another "barbarian", I keep finding all sorts of interesting little nuggets.
With a hat tip towards Boston, here's an excellent article by Peter Gutmann on Furtwangler, one of the greatest interpreters of Beethoven that ever lived. I've always wondered why Toscanini was publicized and built up so much during the Fifties in this country, when the RCA recordings are so dry and tedious, while Furtwangler was ignored as much as possible. Turns out there was a lot of politics involved with Furtwangler, Toscanini, and Karajan.
For those of you who like surrealist science-fiction served with a side dish of acerbic social commentary, here's a link to a Russian film, Kin Dza Dza!. Part Two is here, and a fun review (with excerpts) by SadCast is here. Another hat-tip towards Boston for bringing a one-of-a-kind movie to our attention.
Watch for the scene where the woman on the six-wheeled cart adjusts her "stereo", hears the violin for the first time, and her reaction to the sounds she's hearing.
P.S. Another hat-tip to "thomaseliot" for the 18Sound contact with Mr. Previ, which has now started up. I expect good things from this.
Yes, it's true, I enjoy reading Romy the Cat's "goodsoundclub" site - mostly for the endlessly entertaining invective about the depraved and corrupt high-end industry. Although I'm pretty sure he regards me - along with the rest of the DIY community - as just another "barbarian", I keep finding all sorts of interesting little nuggets.
With a hat tip towards Boston, here's an excellent article by Peter Gutmann on Furtwangler, one of the greatest interpreters of Beethoven that ever lived. I've always wondered why Toscanini was publicized and built up so much during the Fifties in this country, when the RCA recordings are so dry and tedious, while Furtwangler was ignored as much as possible. Turns out there was a lot of politics involved with Furtwangler, Toscanini, and Karajan.
For those of you who like surrealist science-fiction served with a side dish of acerbic social commentary, here's a link to a Russian film, Kin Dza Dza!. Part Two is here, and a fun review (with excerpts) by SadCast is here. Another hat-tip towards Boston for bringing a one-of-a-kind movie to our attention.
Watch for the scene where the woman on the six-wheeled cart adjusts her "stereo", hears the violin for the first time, and her reaction to the sounds she's hearing.
P.S. Another hat-tip to "thomaseliot" for the 18Sound contact with Mr. Previ, which has now started up. I expect good things from this.
Actually things can be done like that.DDF said:
Signals can be generated to test for the reflection signature of a room. This data can then be used and subtracted during speaker measurement. Multiple reflections will show in the reflection signature. The advantage of this is that when we do speaker measurement, we can see a more accurate CSD of the DUT. If we just use editing techniques on the cepstral or impulse signal, the CSD/ETC cannot be adequately preserved.
Hi
DDF, Does this also hold true for whizzer cones ?
-----------------------------------------
JohnL, What bass driver is this ?
Are you satisfied with it ?
------------------------------------------
The DCX is a really good ( cheap, versitale and easy to handle ) starting point.
Don't forget to 1.) cut the output muting trannies 2.) change the elecrolytics and 3.) bridge the input ( analog / digital ) relay if you want to easily upgrade sound wise.
The RANE Dragnet family seems to be quite expensive and top notch. EQing is in the range of 0.27 < Q < 96 and the frequency stepping at 1 Hz I was told!!!!
( Edward, thanks for the informations I didn't find on RANE homepage)
-----------------------------------------
Thanks, chrismercurio !
In the meanwhile I am working through the topic of computer based XO's .
This thread started 2005 for example :
http://www.diyaudio.com/forums/showthread.php?postid=709772#post709772
I got my firewire Mackie ONYX 400F ( 8 analog IN / 8 analog OUT ) soundcard working with "CONSOLE" on a 1.4 GHz Centrino notebook with low CPU load, though the demo version of "frequency allocator" spoils everything once some 20 min have past.
-------------------------
RAVEN –
Who wouldn' think of Alfred Hitchcock?
These are quite intelligent animals!
Greetings
Michael
DDF said:
DDF, Does this also hold true for whizzer cones ?
-----------------------------------------
JohnL, What bass driver is this ?
Are you satisfied with it ?
------------------------------------------
Lynn Olson said:
The DCX is a really good ( cheap, versitale and easy to handle ) starting point.
Don't forget to 1.) cut the output muting trannies 2.) change the elecrolytics and 3.) bridge the input ( analog / digital ) relay if you want to easily upgrade sound wise.
The RANE Dragnet family seems to be quite expensive and top notch. EQing is in the range of 0.27 < Q < 96 and the frequency stepping at 1 Hz I was told!!!!
( Edward, thanks for the informations I didn't find on RANE homepage)
-----------------------------------------
chrismercurio said:
Thanks, chrismercurio !
In the meanwhile I am working through the topic of computer based XO's .
This thread started 2005 for example :
http://www.diyaudio.com/forums/showthread.php?postid=709772#post709772
I got my firewire Mackie ONYX 400F ( 8 analog IN / 8 analog OUT ) soundcard working with "CONSOLE" on a 1.4 GHz Centrino notebook with low CPU load, though the demo version of "frequency allocator" spoils everything once some 20 min have past.
-------------------------
RAVEN –
Who wouldn' think of Alfred Hitchcock?
These are quite intelligent animals!
Greetings
Michael
Raven
http://www.orcadesign.com/products/raven/raven.htm
An externally hosted image should be here but it was not working when we last tested it.
http://www.orcadesign.com/products/raven/raven.htm
Hi Lynn I’m from Italy.
A friend of mine (another fanatic like diyAudio title says
) has a great experience with Eighteen Sound Drivers.He obviously has contacts with the company.
If you need information we will be proud to help you.
I’m a fanatic too, no business interest in my proposal!
Ciao
Re: Raven
Correct me if I misread this and your earlier post on aligning the drivers for first arrival but that sounds like what Spica did back in the late 70's. There are a couple of caveats here though. Considering the 2.5k crossover, the group delay of the 2nd order Bessel would be about 0.1 msec so the tweeter would have to be offset about 1.35" behind the acoustic center of the 8". No big deal. Then the 250 Hz crossover would have a GD of 1.0 msec and the 8"/tweeter combo would have to be offset 13.5" behind the 12" AC. Now there's a bit of a problem, no? Lastly, for the 100 Hz x-o the GD is 2.2 msec requiring an offset of almost 33" relative to the 12"/8"/tweeter combination. These offsets are in addition to the offsets of the driver ACs.
The Spica approach was fine for a small 2-way box speaker with crossover in the 2 to 3 k range, but using physical offset to time align 3 or 4 way systems isn't the way to go, IMO. Additionally, with a dipole system if you physically offset the drivers to align the 1st arrivals from the front you make a mess out of the rear response.
My feelinng is if you want time coherence in a dipole the driver AC's must be aligned and then time coherent crossover must be used as I am doing in my ICTA design development.
http://www.musicanddesign.com/icta_cross.html
Lynn Olson said:
Correct me if I misread this and your earlier post on aligning the drivers for first arrival but that sounds like what Spica did back in the late 70's. There are a couple of caveats here though. Considering the 2.5k crossover, the group delay of the 2nd order Bessel would be about 0.1 msec so the tweeter would have to be offset about 1.35" behind the acoustic center of the 8". No big deal. Then the 250 Hz crossover would have a GD of 1.0 msec and the 8"/tweeter combo would have to be offset 13.5" behind the 12" AC. Now there's a bit of a problem, no? Lastly, for the 100 Hz x-o the GD is 2.2 msec requiring an offset of almost 33" relative to the 12"/8"/tweeter combination. These offsets are in addition to the offsets of the driver ACs.
The Spica approach was fine for a small 2-way box speaker with crossover in the 2 to 3 k range, but using physical offset to time align 3 or 4 way systems isn't the way to go, IMO. Additionally, with a dipole system if you physically offset the drivers to align the 1st arrivals from the front you make a mess out of the rear response.
My feelinng is if you want time coherence in a dipole the driver AC's must be aligned and then time coherent crossover must be used as I am doing in my ICTA design development.
http://www.musicanddesign.com/icta_cross.html
john k... said:
I haven't been too forthcoming regarding the exact technique, but that's consciously. Fundamentally there is enough info known in perhaps 2 to 3 measures to be able to solve this.
The optimization would occur in the quefrency domain as it better suits the optimization method.
I appreciate we can’t achieve closure on this discussion when the algorithm details aren't revealed, nor the background techniques. However, I’m surprised that it can be so readily dismissed and critiqued, without this information.