As my last off-topic in this issue, got a little time last night and made some homework:
"Audibility of a CD-Standard A/D/A Loop Inserted into High-Resolution Audio Playback
E. Brad Meyer and David R. Moran "
Here is a short explanation of test arrangements:
http://www.bostonaudiosociety.org/explanation.htm
"One of the authors, using a short repeated section of room tone on the Hartke disc mentioned above, obtained a positive result (15/15) at a gain of only 10 dB above our standard level."
I dont see here much need for conversation. Just wondering why AES wanted to publish it.
"We also conducted a series of tests at a local CD/DVD mastering facility. I do not currently have a detailed equipment list for this venue, but the speakers were very large and capable high-end monitors, approximately 7 feet tall, and the power amps were sufficient to drive the speakers to very high levels without audible distortion."
"Audibility of a CD-Standard A/D/A Loop Inserted into High-Resolution Audio Playback
E. Brad Meyer and David R. Moran "
Here is a short explanation of test arrangements:
http://www.bostonaudiosociety.org/explanation.htm
"One of the authors, using a short repeated section of room tone on the Hartke disc mentioned above, obtained a positive result (15/15) at a gain of only 10 dB above our standard level."
I dont see here much need for conversation. Just wondering why AES wanted to publish it.
"We also conducted a series of tests at a local CD/DVD mastering facility. I do not currently have a detailed equipment list for this venue, but the speakers were very large and capable high-end monitors, approximately 7 feet tall, and the power amps were sufficient to drive the speakers to very high levels without audible distortion."
DougL said:
Who will know until the ultimate, no compromise speaker, amplifier source and recording is discovered? Who will judge that?
Looking through this thread I see most people using cones and domes, LOL, (Elsinore, Ariel, ect) how would they know what's SOTA with their unacceptable compromises?
Interesting discussion.
I use a 'B' for a x-over at 350 HZ. While this is probably far from excellent my system sounds vastlly better bi-amped with the 'B' then it does mono amped without the 'B'
My point... for those thinking about taking the active plunge don't be scared off by discussions like this waiting for and paying for the ultimate x-over.
Carry On gentlemen!
I use a 'B' for a x-over at 350 HZ. While this is probably far from excellent my system sounds vastlly better bi-amped with the 'B' then it does mono amped without the 'B'
My point... for those thinking about taking the active plunge don't be scared off by discussions like this waiting for and paying for the ultimate x-over.
Carry On gentlemen!
Hi
Stepping up the layers from comparison of circuit designs / comparison of symmetrical / unsymmetrical chaining to the comparison of different system approaches I would like to add that an other benefit of digital crossovers lays in the simplification of analogue circuits needed .
As I pointed out earlier, for a 4 way crossover you have to deal with quite a lot of stages. This sounds simple as you might think you just have to duplicate what works on a simple single line stage but actually isn't.
Its much more comparable to the design of a mixing console and this for sure isn't an easy task if you are heading towards top notch sound quality .
With a digital crossover you usually have to deal with only one analogue receiver stage plus anti-aliasing filter for each IN -channel before AD ( if you don't use digital sources anyway ) and one reconstruction filter stage ( I/V sometimes ) plus one analogue cable driver stage for each OUT-channel after DA.
One of biggest advantages of digital crossovers is that you can add any delay needed with a breeze.
------
For Lynn having built his dream amps already I could imagine to use a passive / active mixed approach.
Why not using the Karna as active XO driver stage for the lower crossover frequencies ?
Seriously – IF its not gonna to be a high order 4 way active XO anyway, why not using the amp that feeds the full way driver also as driver stage for an "active" XO circuit for bass and sub ?
This way size and cost of high quality inductors and capacitors for the low crossover frequencies could be kept minimal due to the highish input impedance of the following bass and sub amps .
For those not using the Karna or valve amps a simple output resistor might provide sufficient compensation for the higher out-impedance of transformer amps.
-----
Talking already on the level of system comparison, there is a relatively new approach that is promising though not fully mature.
PC based XO's are the ultimate to merge advanced XO functionality and feed forward compensation of imperfections of circuits, channel matching , drivers and room behaviour carried out by convolution – kind of almost unlimited EQ .
http://www.duffroomcorrection.com/wiki/Main_Page
http://www.acourate.com/
http://www.juicehifi.com/index.html
http://drc-fir.sourceforge.net/doc/drc.html#htoc197
ShinOBIWAN has outlined his experiences in detail giving a proof of concept to say the least
http://www.diyaudio.com/forums/showthread.php?s=&threadid=63078&perpage=50&highlight=&pagenumber=1
An other most interesting project was started by Peufeu considering seriously the impacts of clocking , ground loops and avoiding PC noise in the listening room
http://www.diyaudio.com/forums/showthread.php?s=&threadid=106497
Basically PC XO's can be built around a simple soundcard.
http://www.thuneau.com/allocator.htm
IF you want to have high quality DA conversion though, an internal PCI card might not be the ultimate basis for tweaking and USB cards are not useful at all if you want to use your computer as a highest quality juke box.
Firewire sound cards excel over USB in that Firewire specs allow also for asynchronous data transfer whereas USB only allows for isosynchronous data transfer AFAIK.
The point here is that you MUST locate the clock at the DA unit and NOT at the digital source ( = the PC ) meaning the digital source has to follow the DA clock .
Peufeu uses the term of "clock injection" to give this most important issue in digital chains a name that might be more or less plausible.
To accomplish that, data transfer from PC to soundcard HAS to be asynchronous meaning that the soundcard is telling the PC at what time the next data blocks have to be sent.
Greetings
Michael
miksin said:
Stepping up the layers from comparison of circuit designs / comparison of symmetrical / unsymmetrical chaining to the comparison of different system approaches I would like to add that an other benefit of digital crossovers lays in the simplification of analogue circuits needed .
As I pointed out earlier, for a 4 way crossover you have to deal with quite a lot of stages. This sounds simple as you might think you just have to duplicate what works on a simple single line stage but actually isn't.
Its much more comparable to the design of a mixing console and this for sure isn't an easy task if you are heading towards top notch sound quality .
With a digital crossover you usually have to deal with only one analogue receiver stage plus anti-aliasing filter for each IN -channel before AD ( if you don't use digital sources anyway ) and one reconstruction filter stage ( I/V sometimes ) plus one analogue cable driver stage for each OUT-channel after DA.
One of biggest advantages of digital crossovers is that you can add any delay needed with a breeze.
------
For Lynn having built his dream amps already I could imagine to use a passive / active mixed approach.
Why not using the Karna as active XO driver stage for the lower crossover frequencies ?
Seriously – IF its not gonna to be a high order 4 way active XO anyway, why not using the amp that feeds the full way driver also as driver stage for an "active" XO circuit for bass and sub ?
This way size and cost of high quality inductors and capacitors for the low crossover frequencies could be kept minimal due to the highish input impedance of the following bass and sub amps .
For those not using the Karna or valve amps a simple output resistor might provide sufficient compensation for the higher out-impedance of transformer amps.
-----
Talking already on the level of system comparison, there is a relatively new approach that is promising though not fully mature.
PC based XO's are the ultimate to merge advanced XO functionality and feed forward compensation of imperfections of circuits, channel matching , drivers and room behaviour carried out by convolution – kind of almost unlimited EQ .
http://www.duffroomcorrection.com/wiki/Main_Page
http://www.acourate.com/
http://www.juicehifi.com/index.html
http://drc-fir.sourceforge.net/doc/drc.html#htoc197
ShinOBIWAN has outlined his experiences in detail giving a proof of concept to say the least
http://www.diyaudio.com/forums/showthread.php?s=&threadid=63078&perpage=50&highlight=&pagenumber=1
An other most interesting project was started by Peufeu considering seriously the impacts of clocking , ground loops and avoiding PC noise in the listening room
http://www.diyaudio.com/forums/showthread.php?s=&threadid=106497
Basically PC XO's can be built around a simple soundcard.
http://www.thuneau.com/allocator.htm
IF you want to have high quality DA conversion though, an internal PCI card might not be the ultimate basis for tweaking and USB cards are not useful at all if you want to use your computer as a highest quality juke box.
Firewire sound cards excel over USB in that Firewire specs allow also for asynchronous data transfer whereas USB only allows for isosynchronous data transfer AFAIK.
The point here is that you MUST locate the clock at the DA unit and NOT at the digital source ( = the PC ) meaning the digital source has to follow the DA clock .
Peufeu uses the term of "clock injection" to give this most important issue in digital chains a name that might be more or less plausible.
To accomplish that, data transfer from PC to soundcard HAS to be asynchronous meaning that the soundcard is telling the PC at what time the next data blocks have to be sent.
Greetings
Michael
Magnetar said:
ooking through this thread I see most people using cones and domes, LOL, (Elsinore, Ariel, ect) how would they know what's SOTA with their unacceptable compromises?
I know where you are coming from, but there ARE other considerations.
Once man's acceptable compromise is another man's unacceptable compromise. I think that full-range drivers as a whole are fun. I like having fun. I like listening to all kinds of successful speakers. Let us not limit the technology. Maybe one day you will hear a 'cones and domes' speaker that will hit you for six!
Then I am also aware that there are LOTS of designs using 'cones and domes' that I have no time for. At least I do have some time for full-range, open baffle etc. In fact, I did OB decades ago before it became (rather recently) fashionable. I like to believe that any current design includes lessons learned from those times and DO incorporate them. Like sans box is the best box a la midrange. Measure the midrange fairly nearfield on an OB (open space behind - outdoors), then put ANY box behind it, and you will find comparable excess energy coming out at the front. You have to get the midrange right or else everything else in wrong.
IF I have to omit an open baffle (WAF and all that), then I make sure that I have the near next best thing to OB, like 75 Litres for the midrange to be absorbed into, so it mimics a mini anechoic chamber as much as possible. Maybe not quite as good but MUCH better and a workable solution in the context of what must be achieved by the whole as a concept.
I happen to know that my earlier OB designs were great for the DIY in me, but keep them out of sight of the wife, right? Then the Elsinores? "That's not to bad Darling?" The Ariels... likewise.
We are not just audiophiles, we must have a social life too.
I tell you what the real challenge is, get 'cones and domes' not to sound like cones and domes. That's the one I have taken up as it leads to a designs that have broader acceptability and hence 'commercial'. That is not a dirty word.
Then make sure that is very sensitive with an easy load on the amplifier - so you can even use small and sweet sounding (tube) amps with it. I find that the usual low sensitivity 'cones and domes' requiring BIG amps lack dynamic expressiveness. I like jump factor that real live music gives you. Get that plus low distortion, even temperament, genuine cohesiveness, decent (realistic) image (correct size too)and just sheer sense of naturalness. Hear that slap echo of the rear of the stage disappear into nothing in a realistic soundstage? Seems to go on forever and then dies. Oooh! Then what does it matter what the technology is, provided it is successful. When it all comes together it is the total system, of course, not just the bits on the end of the last wire.
Joe R.
Excellent thread! I'm currently trying to find the right frequencies and slopes for my subwoofer/woofer/tweeter with a couple of Marchand crossovers before I try my hand at a "SOTA" design. Does anyone have an opinion as to what IC does the best (or least offensive) job as a buffer in analog crossovers such as the Marchand? I'm currently using OPA 2134's.
Yes. Excellent thread and the exact topic I'm interested in, right now.
My bit to add,and it can be an important one, of you haven't tried it:
Passive impedance compensation filters. (zobels, etc) I know it's an electronic crossover. Please put them on the drivers anyway. You might be very suprised by what you hear. We are talking maximum fidelity here, are we not? Then put the zobels on the tweeters and the compensation filters on the woofers. This shunts maximum/peak back emf transient impedance variations. It makes a notable improvement.
The greatest change comes when you are using a high feedback amplifier. It keeps the feedback circuit calm and more passive, due to lower load transient variance, and more 'out of the picture', so to speak. There are other reasons as well.........
My bit to add,and it can be an important one, of you haven't tried it:
Passive impedance compensation filters. (zobels, etc) I know it's an electronic crossover. Please put them on the drivers anyway. You might be very suprised by what you hear. We are talking maximum fidelity here, are we not? Then put the zobels on the tweeters and the compensation filters on the woofers. This shunts maximum/peak back emf transient impedance variations. It makes a notable improvement.
The greatest change comes when you are using a high feedback amplifier. It keeps the feedback circuit calm and more passive, due to lower load transient variance, and more 'out of the picture', so to speak. There are other reasons as well.........
Decker said:
A pair of OPA627's on a 'brown dog' adapter. With a bit of class A resistive shunting, maybe. Make sure you have some immediate and close by capacitive power supply buffering, with possibly some polystyrene buffering of such. Use the Cardas solder. This will be the least offensive and most neutral, in my experience-whilst retaining as much of the signal as possible. After that (better), it's 'discrete' buffers. (discrete op-amps)
It's a transient modulation thing, so it affects the whole driver/motor/cable set, at the least. It always has a positive net outcome, in my experience. No matter the amp design/operational considerations. The benefits stack up a bit, when you get into heavy feedback amps. Just my experience. It tends to allow for about 3-6db higher output that is deemed 'comfortable', compared to no passive filters in the system. As you may imagine mass-motional created back emf generated by the system has increased dramatically at about 95-100db output. Peak transient impedances measured at the speaker cable are going through the roof. We are talking (instantaneous) horsepower and higher numbers and levels, here.
If you are using a wholly active crossover system and have residual 'glare' that you are and have been fighting like mad to get rid of, this can be the source, or a part of it. If not the major remaining component. Especially when you crank things up.
If you are using a wholly active crossover system and have residual 'glare' that you are and have been fighting like mad to get rid of, this can be the source, or a part of it. If not the major remaining component. Especially when you crank things up.
Hah. Just about the only amp situation that might suffer. Figures
Then try just 'some' -not that much. Like about 20-25ohms in series with, oh, 0.25uf of styrene or teflon, across the tweeter terminals, and the woofer terminals. Find some limited, low interference compensation that cuts peak back emf numbers, just a hair. Fool around with it a bit.
The first thing you tend to think, is that it is darkening up the sound and cutting output, crushing harmonics, etc. In my experience it tends to cut distortion and is of benefit. Very little damage, if any at all. Some of what we hear in 'spatial cues' is many times falsified and exaggerated micro-transient distortion anyway. I'm not trying to 'compensate' (hah, a pun!) for anything that is done that may be 'untoward' re the signal, as I find it does no damage. And I'm a space and imaging freak. If it was eating micro-detail, I'd notice.
Then try just 'some' -not that much. Like about 20-25ohms in series with, oh, 0.25uf of styrene or teflon, across the tweeter terminals, and the woofer terminals. Find some limited, low interference compensation that cuts peak back emf numbers, just a hair. Fool around with it a bit.
The first thing you tend to think, is that it is darkening up the sound and cutting output, crushing harmonics, etc. In my experience it tends to cut distortion and is of benefit. Very little damage, if any at all. Some of what we hear in 'spatial cues' is many times falsified and exaggerated micro-transient distortion anyway. I'm not trying to 'compensate' (hah, a pun!) for anything that is done that may be 'untoward' re the signal, as I find it does no damage. And I'm a space and imaging freak. If it was eating micro-detail, I'd notice.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.