AndrewT said:
It's meant to create a less directional / diffuse effect, closer to the multiple sides used in real cinemas. A lot of people say its not as good as a monopole for 5.1 music though.
I like my dipole surrounds fwiw, but only listen to music in stereo at present.
Rob.
I'm actually hoping that the 5.1 and 7.1 systems will be expanded.
Change from 3 to 5 front channels. So you have inboard and outboard Left and Right, plus the Center.
Then for the side channels, have Left and Right Side Surrounds, but also have each a direct monopole and a dipole.
These could be two drivers per side, each driven with an amp.
But the pair of amps has two inputs. One is for in phase. The other is for out.
So, for an experience like a large atheltic stadium, you drive the side surrounds as monopoles. This is the sound of the crowds all around. It is diffuse, and from multiple sources. But it also can have some higher frequency component.
But for an experience which is largely reflected sound, like sound reflected from a concert of nite club, you drive the surrounds a dipoles. It is even more difuse, and has less high frequency component.
Then, some situations contain elements of both.
So, if 7.1 means a subwoofer, two rear surrounds, two side surrounds, and 3 front channels,
We expand that to 11.1 by adding to more front channels, and two more ways of driving the side surrounds.
Maybe also add a ceiling surround. Fantasia had this. Infinity makes a quadrapole loud speaker. 4 x 4" drivers gently pointing out in four directions, with alternating phase.
Make something like this, 12.1
With digital it is not real problem because tracks don't take up space on the recording medium except for when they are used. And that is only proportionate to the volume they are used at.
Anyway, I'd like to talk about Active Subtractive Crossover topologies.
Change from 3 to 5 front channels. So you have inboard and outboard Left and Right, plus the Center.
Then for the side channels, have Left and Right Side Surrounds, but also have each a direct monopole and a dipole.
These could be two drivers per side, each driven with an amp.
But the pair of amps has two inputs. One is for in phase. The other is for out.
So, for an experience like a large atheltic stadium, you drive the side surrounds as monopoles. This is the sound of the crowds all around. It is diffuse, and from multiple sources. But it also can have some higher frequency component.
But for an experience which is largely reflected sound, like sound reflected from a concert of nite club, you drive the surrounds a dipoles. It is even more difuse, and has less high frequency component.
Then, some situations contain elements of both.
So, if 7.1 means a subwoofer, two rear surrounds, two side surrounds, and 3 front channels,
We expand that to 11.1 by adding to more front channels, and two more ways of driving the side surrounds.
Maybe also add a ceiling surround. Fantasia had this. Infinity makes a quadrapole loud speaker. 4 x 4" drivers gently pointing out in four directions, with alternating phase.
Make something like this, 12.1
With digital it is not real problem because tracks don't take up space on the recording medium except for when they are used. And that is only proportionate to the volume they are used at.
Anyway, I'd like to talk about Active Subtractive Crossover topologies.
Hi Zen,
what software sources do you have access to that encode 11 or 12 channels?
Do you think these will ever reach the domestic market?
what software sources do you have access to that encode 11 or 12 channels?
Do you think these will ever reach the domestic market?
I've set up a thread in digital, Open Source.
I think this should be taken out of the hands of marketeers. When it comes to digital, we don't have to turn into cell phone bimbos. We should take charge of the situation.
Here, I want to talk about active subtractive crossovers.
I think this should be taken out of the hands of marketeers. When it comes to digital, we don't have to turn into cell phone bimbos. We should take charge of the situation.
Here, I want to talk about active subtractive crossovers.
Most of the time what is published for active filters are the non-inverting Salen-Key circuits.
But a few places I remember references to inverting Salen-Key circuits. Anyone have access to any information on this?
You see, using the inverting form gives lower distortion.😎
But a few places I remember references to inverting Salen-Key circuits. Anyone have access to any information on this?
You see, using the inverting form gives lower distortion.😎
Hi Zen,
the Sallen&Key filters whether unity gain or equal value are all non-inverting.
The MFB is an inverting filter with independantly adjustable gain and Q.
the Sallen&Key filters whether unity gain or equal value are all non-inverting.
The MFB is an inverting filter with independantly adjustable gain and Q.
Circuit and dimensioning of Sallen-Key, MFB and state variable can be found here:
http://www.analog.com/Analog_Root/static/techSupport/designTools/interactiveTools/filter/filter.html
Very convenient BTW !!!
Regards
Charles
http://www.analog.com/Analog_Root/static/techSupport/designTools/interactiveTools/filter/filter.html
Very convenient BTW !!!
Regards
Charles
phase, that analog devices page is a simulation program. I want to do things by the mathematics.
But the MFB circuit option of there's is inverting. That might be the same as what I saw for inverting Sallen Key. It probably amounts to the same thing.
This program could come in handy later on. Thanks.
One of the best books, one I'm going to get back, is by Lawrence P. Huelsman.
But the MFB circuit option of there's is inverting. That might be the same as what I saw for inverting Sallen Key. It probably amounts to the same thing.
This program could come in handy later on. Thanks.
One of the best books, one I'm going to get back, is by Lawrence P. Huelsman.
Currently reading:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=5435&highlight=
Also, I read that speakers from Linkwitz Labs are multiamped with active crossovers, something the audiophile community has been slow to accept.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=5435&highlight=
Also, I read that speakers from Linkwitz Labs are multiamped with active crossovers, something the audiophile community has been slow to accept.
One thing I'm looking for, is how do active subtractive crossovers sound?
Most people agree that multiamping sounds better, more open, less intermodulation.
But this approach to crossovers, subtractive, or constant voltage, or flat summing, does it really work well? Does it sound nice?
It would not use a standard filter transfer function like L-R or Butterworth, at least not on both slopes.
So does it sound nice?
Most people agree that multiamping sounds better, more open, less intermodulation.
But this approach to crossovers, subtractive, or constant voltage, or flat summing, does it really work well? Does it sound nice?
It would not use a standard filter transfer function like L-R or Butterworth, at least not on both slopes.
So does it sound nice?
The transient-perfect versions of subtractive crossovers don't have steep slopes around the crossover frequency.
So they are only recommended for robust drivers that are closely spaced. The latter is not very common with sub applications but is feasible of course.
Regards
Charles
P.S. Who is BK electronics ???
So they are only recommended for robust drivers that are closely spaced. The latter is not very common with sub applications but is feasible of course.
Regards
Charles
P.S. Who is BK electronics ???
Substractive crossovers are a big no-no because their outputs have dissimilar phase response thus causing the waves coming out of the drivers to sum in any direction but on-axis. The simulated flatness is actually achieved through boost and cancellation.
Use Butterworth or Linkwitz-Riley style filters with coherent phase response instead. On-axis summing is a great thing, particularly inside roms.
Use Butterworth or Linkwitz-Riley style filters with coherent phase response instead. On-axis summing is a great thing, particularly inside roms.
It is actually vice-versa: Flat on-axis but bad off-axis behaviour. but this is only true if the driver-spacing is large compared to the wavelength at crossover. I two drivers are spaced 25 cm apart and your crossover frequency is 200 Hz for instance then lobing is a minor problem.
For the typical midrange-to-tweeter crossover frequencies better transient-improved topologies are available than subtractive ones.
But if you want good transient performance LR is definitely a NO-NO.
Regards
Charles
Do the math yourself if you want. Any substractive crossover will produce higher output somewhere off-axis in the crossover region.
For example, a first order approach suffers 90 degrees phase mismatch in the crossover region and flatness is achieved through 3dB cancellation on-axis, which prodduces 3dB boost off-axis. A second order approach with Q=1 produces nearly 6dB cancellation on-axis for flat response and thus 6dB boost off-axis, with nearly 120 degrees of phase mismatch.
My first filters were of that kind, but I quickly dropped them due to the disappointing results. Left-Right imaging was particularly non-existent with these filters due to the huge amounts of off-axis stuff beihg trashed by room acoustics and then reaching you and masking clean on-axis radiation.
For example, a first order approach suffers 90 degrees phase mismatch in the crossover region and flatness is achieved through 3dB cancellation on-axis, which prodduces 3dB boost off-axis. A second order approach with Q=1 produces nearly 6dB cancellation on-axis for flat response and thus 6dB boost off-axis, with nearly 120 degrees of phase mismatch.
My first filters were of that kind, but I quickly dropped them due to the disappointing results. Left-Right imaging was particularly non-existent with these filters due to the huge amounts of off-axis stuff beihg trashed by room acoustics and then reaching you and masking clean on-axis radiation.
you have to do the math properly !
Agreed, you will have a hump off- axis but you will also have a hole in another direction. But tihis is only a problem if not properly used.
As I said: Don't use these for tweeters unless you know exactly what you are doing !
Regards
Charles
Agreed, you will have a hump off- axis but you will also have a hole in another direction. But tihis is only a problem if not properly used.
As I said: Don't use these for tweeters unless you know exactly what you are doing !
Regards
Charles
AndrewT, re post 16. Thank you. I remember reading that post. It was part of what inspired me to pose the question about how such sound.
Phase_accurate, re post 53. I follow you. My interest is for drivers that are closely spaced, like a 4 way system plus subwoofer.
So the gentle slopes would be ok. But I still need to know of a source where someone has worked out all the math. I also would like to know how these things sound.
See, crossover thinking is based on the idea that the sound adds up in the air. Likewise, there have always been phasing and flat summation issues. So crossover design has never been pure science.
I want to see the math, and I want to see precedent for what works.
The audiophile community has been too slow to go to multiamping and active crossovers, IMHO.
Commerical sound has gone futher with this.
So the gentle slopes would be ok. But I still need to know of a source where someone has worked out all the math. I also would like to know how these things sound.
See, crossover thinking is based on the idea that the sound adds up in the air. Likewise, there have always been phasing and flat summation issues. So crossover design has never been pure science.
I want to see the math, and I want to see precedent for what works.
The audiophile community has been too slow to go to multiamping and active crossovers, IMHO.
Commerical sound has gone futher with this.
Eva and Phase,
I follow what you are saying. Of course I understand all about close driver spacing, physically, and in frequency ranges.
But unless you two are willing to post a general treatise, we need a reference document, where someone has worked these things out.
Also, the passive crossover schemes are not that great either.
How about some good references for Active and Passive cross over schemes, and how the phases sum up, and what that does to the beam pattern?
Of course subtractive is attractive, but only if it really can be made to work.
I don't see it as a trivial subject, but I have never been able to find any references where it has been analyzed.
Also, seeing circuits for what is used would help!!
I follow what you are saying. Of course I understand all about close driver spacing, physically, and in frequency ranges.
But unless you two are willing to post a general treatise, we need a reference document, where someone has worked these things out.
Also, the passive crossover schemes are not that great either.
How about some good references for Active and Passive cross over schemes, and how the phases sum up, and what that does to the beam pattern?
Of course subtractive is attractive, but only if it really can be made to work.
I don't see it as a trivial subject, but I have never been able to find any references where it has been analyzed.
Also, seeing circuits for what is used would help!!
- Status
- Not open for further replies.