Agree. But catapult simmed just a crossover with parallel connections, which shouldn't produce an output shift of more than 720 degrees relative to another driver. That the sim comes up with more suggests to me there may be an issue with how phase is wrapped.
I'm not fully following your concern or question. I think Catapult's sim give just what I expected. Of the 720 degrees of a single issolated bandpass we see only half that (360 degrees) since at the cutoff frequency we hand off signal from one unit to the next. His phase plot shows from +180 to - 180 and wraps up or down beyond that, as expected.
This is the interesting as well as arbitrary nature of phase: none of the units gives more than 180 (+-) degrees shift in its passband. They are all truly 0 degrees at their center frequency, yet when we string them all together their effects cascade and the LF end is 3960 degrees in advance of the HF end. This is also shown in the group delay plot which generally rises for LF as well as having a peak at every crossover point steming from the faster phase rotation at crossover (not due to phase wraps which are purely connected with plotting conventions).
David S.
First, a distinction that parallel and series isn't as in parallel and series connected crossover networks. Catapult was drawing what would be parallel blocks for a total model simulation.
For example, if a woofer had a lowpass, a dip circuit, and another lowpass, they would all be 3 series connected or cascaded blocks (plus the woofer as a 4th block) in any simulation you would set up. If the same input goes to a high pass and then a tweeter, those 2 blocks would be series connected, but at the same time both the woofer chain and tweeter chain would be "parallel connected" to represent the full system. (Parallel connected meaning their response combines at the listeners position.)
For the sake of the 12-way system we were talking shorthand by referring to 12 parallel connected bandpasses, that being generaly what the system is.
And the phase response? The phase response comes from the vector sum of the 12 sections. That is, we have to consider both strength (magnitude) and phase of each section to see how they add together. If these are 12 sequential sections with little overlap, then within any driver's range that driver will have the only vector with significant strength and it will contribute the bulk of magnitude and phase of the output. At crossover points then usually 2 adjacent sections will have similar strengths and we will get a phase midway between that of the 2 contributing vectors at some combined strength. If we design a system well, then the units will be nearly in phase at every crossover point, little cancelation will occur and we will essentially be splicing the individual phase curves together. That is how 12 units gave 3960 degrees of phase shift.
Hope that makes sense,
David S.
Holger Barske wrote:
Thats the same thing as I thought, why spend rediculous amounts of time in getting the exact value for a (calculated)xover component. It's about the real response (driver+xover) that matters. No calculated xover component will do the trick in the real world.
Holger, keep up the good work with K&T
Thats the same thing as I thought, why spend rediculous amounts of time in getting the exact value for a (calculated)xover component. It's about the real response (driver+xover) that matters. No calculated xover component will do the trick in the real world.
Holger, keep up the good work with K&T
i put on my thinking hat because this is a very exiting topic. i think that the precision mathematics is especially insightful but i dare to call out a mistake in the assumption.
the solution of the frequency bands covered per speaker is incorrect! the correct number is to have one filter aligned perfectly per decade and subdecade of the psychoacoustic loudness level standards as defined by iso 226. i understand the temptation of making filters align to speakers but this is a flaw in engineering thinking. speakers must always first align to each hearing frequency and that specific ba ds sensitivity area to ensure there is a perfect correlation between audiable hearing and the psychoacoustic analysis of the sound. if this correlation isnt done in the sometimes digital but always analog domain then the mathematics by logic must be incorrect! the tables should be readjusted for this in my opinion important assertion or at the very least a filtering mechanism should be insterted before power is applied to the speakers. when i read up on this standard i realized that up to 40% error in percieved intrafrequency sound levels can be analysed and this is particularly evident in the lower frequency bands.
the standard illustrates these self-correcting auditory filtering mechanisms that exists in the human psyche wonderfully and anyone can clearly see where the importance of psycho-targeted driver selection is an absolutely overlooked discipline in both near field and far field enclosure construction. personally i believe that this is overlooked because the precise alignment of the right systems parameter takes exceptionally long time to calculate and simple log(e) or even parameterized analysis of the audiable frequency band when cross referenced towards the iso226 correction factors is frankly a highly complex solution to implement correctly in the standard passive filtering mechanism and apparatus of the current speaker systems.
for your reference i include two additional pieces of evidence.
the first one is the motto by which great audio engineers must live by: audio is not an ism like perception based disciplines live by; we are not in the field of audio-ism, sound-ism etc. we are in the art of making sound loud, i assert that those who lay claim to the science of loud-ness by necessity also are adhering the principled approach of the art of loud-ness!
my second evidence to support the unifiying relativity theory of -ism vs -ness is in the standard itself, see and critize this excellent evidence based analytics of electing frequency band via referential selection of the frequencies and their relativity in the audiotory spectrum. altough i am not claiming novelty in this i believe it is plain that anyone skilled in the science and the art will conclude the plain necessity for proper filter calculation and those who mistrust or are yet to be convinced; i say that the entire international standardization body known as iso serves as my witness to the thoroughness and factuality of this assertion!
it is friday night here in asia, im going to go and have another beer. happy octoberfest to all my german friends, i am celebrating on your behalf in-absentia!
the solution of the frequency bands covered per speaker is incorrect! the correct number is to have one filter aligned perfectly per decade and subdecade of the psychoacoustic loudness level standards as defined by iso 226. i understand the temptation of making filters align to speakers but this is a flaw in engineering thinking. speakers must always first align to each hearing frequency and that specific ba ds sensitivity area to ensure there is a perfect correlation between audiable hearing and the psychoacoustic analysis of the sound. if this correlation isnt done in the sometimes digital but always analog domain then the mathematics by logic must be incorrect! the tables should be readjusted for this in my opinion important assertion or at the very least a filtering mechanism should be insterted before power is applied to the speakers. when i read up on this standard i realized that up to 40% error in percieved intrafrequency sound levels can be analysed and this is particularly evident in the lower frequency bands.
the standard illustrates these self-correcting auditory filtering mechanisms that exists in the human psyche wonderfully and anyone can clearly see where the importance of psycho-targeted driver selection is an absolutely overlooked discipline in both near field and far field enclosure construction. personally i believe that this is overlooked because the precise alignment of the right systems parameter takes exceptionally long time to calculate and simple log(e) or even parameterized analysis of the audiable frequency band when cross referenced towards the iso226 correction factors is frankly a highly complex solution to implement correctly in the standard passive filtering mechanism and apparatus of the current speaker systems.
for your reference i include two additional pieces of evidence.
the first one is the motto by which great audio engineers must live by: audio is not an ism like perception based disciplines live by; we are not in the field of audio-ism, sound-ism etc. we are in the art of making sound loud, i assert that those who lay claim to the science of loud-ness by necessity also are adhering the principled approach of the art of loud-ness!
my second evidence to support the unifiying relativity theory of -ism vs -ness is in the standard itself, see and critize this excellent evidence based analytics of electing frequency band via referential selection of the frequencies and their relativity in the audiotory spectrum. altough i am not claiming novelty in this i believe it is plain that anyone skilled in the science and the art will conclude the plain necessity for proper filter calculation and those who mistrust or are yet to be convinced; i say that the entire international standardization body known as iso serves as my witness to the thoroughness and factuality of this assertion!
An externally hosted image should be here but it was not working when we last tested it.
it is friday night here in asia, im going to go and have another beer. happy octoberfest to all my german friends, i am celebrating on your behalf in-absentia!
Last edited:
So after all - a week has past and a lot has been thrown into the circle....
yesterday I performed the task to do the measurements "reloaded" and that´s allready the first point to answer:
"Must admit, good way of using up all those odd drivers laying around.
Can't wait to see measurements! Also has me beat how he designed them without measurements in the first place. Can we have a list of drivers please?"
Where throughout this thread do you believe that i shall have stated that all the construction has been performed without measurements ? I can´t find that statement - i only see some people that suggest that.... but the opposite is correct -
throughout every stage the entire project has always been ruled by measurements up to the very point that each single speaker and each single crossover has been precisely measured and adjusted ! To me it seems that you´ve just "dropped" some pages from reading....
the next post is rather poor from arguments but rich with suggestions...
"I will continue to think it's a joke! It looks like my parts bin from around 30 years ago!
Some of those drivers have to be thirty years old! The reason I think it's a joke is those
two piezos nearly at floor level! They're there for the little folk!
"Up till now no physical measurements have been made...."
I did not saythat ! I just said that the measurements from time of construction ( 2005 ) are not availible anymore...
"No, really! You don't say! Very practical way to proceed! I must admit, I would love a view of the crossover, I bet it's all wired straight to the drivers!"
First: The speakersystem has been realized from 2003 to 2005.....
Second: The Speakersystem has been desired right from the beginning - that at least a distance of 5 Meters is between the listener and the speakers and at least 4 meters between the speakersystems themselves -
and last not least that no furniture is between the speakers and the listener ( so drop off from thinking that some furniture is between the horns and the listener ) and - what really is not well to be seen from the front photograph - the horn are not exactly plain to the front but have been mounted with a slight angle upward and to the imaginable middleaxis of the speakerpair.
And its not "practical" that the measurement from 2004 and 2005 have not been archived - but the basics of the entire project truely are reliable to measuremments that have been made.... but it´s always a good argument to issue suspicios thoughts and doubts if your out of reliable arguments....
"If you are approaching this in an ideal way you must be giving each driver less than an octave of spectrum?
Otherwise you must have multiple units covering overlapping frequency bands. I guarantee that approach will lead to a very messy 3 dimensional response. Can you show us polar curves or on and off axis frequency response?"
Within the running week all the rest of the measurements will be published and they all show one thing very clear - there is nothing comparable around.....
"Yikes!! 11 crossovers * 360 degrees = 3960 degrees of phase rotation. The group delay caused by phase rotation is generally inaudible with music and normal speakers but I'd be surprised if this one weren't audible. We're talking about the harmonics of an instrument arriving way before the fundamental."
... well thats a nice sugest but it does not come out true in measurement....
So also a lot of post originated from Phoenix358 are like:
"You really don't understand filter design at all! Inductors can be unwound to get an exact value.
It's not rocket science! Slightly altering capacitor values to stay within preferred values only slightly alters the filter Xover point and does to both the high pass and the low pass. The higher order the x-o the harder it gets but it is still not rocket science."
"Good for you! Let's see some measurements to back up your so called science!
Building without measurements ..... pah! I don't mind you deluding yourself!
I don't care for your attitude!
Don't try and delude the noobs with your nonsense, it's not fair on them.
Your claims so far are nothing but unsubstantiated male bovine droppings."
"If it's a joke it goes too far. I am thinking that he is for real. No one can design properly without measurements, not and come up with a design that beats the entire world and not from a collection of miss- matched drivers that rightly belong in a speaker museum."
"Yes, it's prediction time. My crystal balls are in play. We will never see a response graph!"
So beside the fact that most of his "output" is beyond any other comment just agitation instead of normal communication - your crystal ball is "out of order" and your "style" of posts disqualifies yourself from a serious response.... it´s not my level to argue below the guts...
And a short point to Holger:
In many years of more or less professional speaker design I have learned a thing or two. For example:
Component values with five percents of tolerance are absolutely fine, this is mostly far better then speaker parameter tolerances.
And if i look at the... well, not very high value drivers used here, I doubt that it is possible to build two somewhat identical speakers at all.
First Point: I am not talking about 5% tolerance - I´m talking about hiting the values within less then 0,5% of tolerance because i don´t trust tolerance - I only trust measurements just right down each single component....
And second: Your truely right that the overall tolerance of speakers ist far beyond 5 % - but there are two points that you did not mention that have been explained in the text:
a) there is only very small bandwith that the speaker has to perform within - and that is only in the very area where the speaker is operating within its optimum of performance... and b) each single speaker is at
the end calibrated within just that very small area.....
and another objection was:
"There does not seem to be much 'breathing space' for the 2 drivers <15Hz and 15-30Hz.
If you can pull that one off I'd start thinking about going into business making tiny subwoofers."
That is at the first view an interesting objection - but after all - I´ve done a lots of experiments with the large speakers to validate the real need of air to operate ( air and not casevibration ! ) and it turned out that it´s a good "fistrule" that a speaker is dependent to the air within a cube with sidelengths only at least the diameter of the speaker. Adding more volume does not really have much more influence to the ability of the cone to swing - it only causes the case to swing .....
The walls of this speakersystems are 50mm to 60mm thick beachwood and the true length does not exceed 60 cm. This wood is to thick and to small to vibrate in low frequencies and that was desired to be that way...
I want to generate sound by the speakers and not by resonance of the case.
and the best ( due to the fact that they restrict to simulation instead of adjustment ! ) Objection to my mind
"I'd never actually sim'd something so silly so I gave it a shot. This is the SPL, phase and group delay of 11 parallel electrical bandpass filters using the OP's frequencies. It's an 11-way as he has 2 supertweeters. The lowest bandpass is an LR4-15 highpass and an LR4-30 lowpass. The highest is an LR4-15360 highpass and an LR4-30770 lowpass. The others are spaced an octave apart in between.
Notice how the phase wraps through 360 degrees every octave."
I´m not only MBA in Math and electronic engineering - I´ve been working for years as freelancer with a company designing measurement-instruments and writing the programs for the Microprocessors....
and I´ve seen one thing for sure: programs are dependent to the abilities of the programmer.....
and a lot of simulations do not really carry out the task clean.... those you might get for free in the internet are of "midclass quality".....
"1) Colouration. Unless it is open back the cabinet will reflect the sound around
- eventually it will be heard as a rather serious 'boxy' sound."
The photographs published at the moment show the open steps - but of course prior to the final calibration of the single speakers all the "subcases" had been closed and on the backside of the speakersystems after the calibration has been performed the backwalls ( each 60 mm thick beechwood ! ) had been mounted and the Cases where closed
absolute airtight !
"2) Focus. I'm not usually big on focus, but by spacing your tweeters that far apart
then you have the sound for a cymbal (for instance) arriving from two different places,
causing some very odd imaging.. also carpet will eat the treble. "
maybe possible but not fact - in my livingroom the floor is massive wood ( otherwise it would have been foolish to make the subsonic-speaker looking towards the floor ) and the minimum distance from the speakers is from the beginning of the design to the final location in the room 5 meters away from both speakers. And frequencies above 10 kHz start to get dificult to locate - it´s the same trouble like with the very lower frequencies - it´s a disability of the ears - they can hear the frequencies but
the higher they get, the more difficult it becomes to locate them precisely .... and at the distance of more than 5 meters the distance between the tweeters ( less then 50 cm ) is not important any more...
"3) Complexity. Audio doesn't like complexity, you will hear the 12 crossovers ,
and your ears will be confused by the spacing of the drivers. The best sound these days
comes from single drivers or perhaps 2 drivers in my experience. Remember we hear all
sound via a single ear-drum on each side of our head!!"
That´s so true - but the sound of the drum does not come from 12 speakers but only from maximum of 2 speakers.... the real trouble comes from the fact that one speaker has problems to transmit many different sounds at the same time....
"I think you could improve your design by making a pyramid shape and putting the highest frequencies at the top and working down - in frequency order - to the floor with the bass.
You also need an enclosure for every open back unit - can't recall if you had that."
The Pyramid is a very good solution to avoid resonance and reflection-frequencies like "standing waves" and if the space between the listener and the speaker is less then 5 Meters it is also usefull due to the fact that midrage frequencies and lower highrange frequencies ( between 6500 Hz and 8000 Hz ) are quite well locateble and at the bottom of the pyramid is the needed volume of air to operate the bass-speaker ( at the top of the pyramid that volume would not be availible... )
Another Advantage is the fact that pyramidform enables you to have them less critical by placing them in the corner of a room...
"I would be interested to see a design with a wide midrange and fill in tweeter and bass,
but at least you are building and experimenting which is all good stuff and more than many do, just not sure it will sound how you imagine."
In fact the entire design has always been dependent to several "set limits" like minimum distance between auditors and speakers, a wide range of performance room ( the largest room the speakers have been used was a factory loft with floorarea of 25 meters x 15 meters and the distance to the ceiling 7,5 meters - straight down to the livingroom where i use them usually - with 6,5 meters x 7meters and ceiling at 2,7 meters. The speakers are usualy close to the 6 meter wall and 5 meters apart and at the oposite
side of the room 5 meters away are the seats. And there is no furniture between
seats and speakers.
"An 8th order bandpass (4 high and 4 low) has to swing from 360 degrees positive at the low end to 360 degrees lag at higher frequencies. Generally you'l have half that phase shift at the corner frequencies so: +180 and -180 or a 360 degree swing each section (repeat 11 times = 3960 degrees.).
How you have the sections add is not something Ronin is telling us, or is apparently aware of, but you always get the best response if you can get the phase swing of one section to overlay the next. Did we mention that the drivers have phase shift due to their response and physical depth? Did we mention that textbook crossovers don't acount for the driver as a load?"
Of course I´ve thought about that....
First of all each speaker has been isolated entirely from interaction with other speakers..... so in fact they might be treated like isolated single speakers
So in fact its a group of independent singlespeakers with extrem sharp "cuttoff" of unwanted sharing of frequencies.
Then second each speaker has been calibrated to operate with precise same soundpressure as any of the other speakers
and third the load... thats realy an issue due to the fact that during normal operation the passive crossovers all together drag nearly 16 amp power during normal performance - the amplifier must be able to deliver that current ... I have a MOS-FET Poweramp with 2 x 250 Watt Output ( up to 25 amp current to top of 70 Volts peakvoltage ) and the other alternate poweramp is a tubeamp with 2 x 140 watt ( each channel 6 x EL34 ) that comes up with 18 amp current from the output-transformers
and up to 60 Volt peakvoltage.
If you don´t bear the issue of the load in mind you realy can toast any outputstage into pure smoke...
"Any single cone is theoretically - and of course to some degree only practically as well - able to handle as many frequencies at once as one could imagine. In the same manner as the air is able to transport many frequencies at once and also like the mechanical parts of our ear are capable of handling. I think the OP's imagination doesn't allow for something like this and therefore all practically existing systems from single driver to four- or five-ways in the world won't work - probably not even when standing in front of one and actually listenenig to it."
In basic the thought is right - but in close view it can´t stand up....
The music (i.e. the instruments and vocals ) mix up within the air of a room to many different waves but in fact the ear does not only consist of the membrane in the ear .- but another point is often dropped from the view and thought: The outside foldings of the external ear fullfil an interesting and important duty: they are in a very special way a kind of "frequency-dependat-delay-filter". And the very tiny membrane of the inner ear is realy a very fast acting unit. On the other side the very large membrane
of a speaker is a very very slow unit. As long a the signal is within the amplifier there is not much trouble due to the fact that electricity is realy fast - but in the moment that electrical waves leave the mechanical transducer ( and electrostatic transducers are in that relation very slow and mechanic too... ) the trouble starts up to the fact that mechanic is bound to multiple misfits:
starting from the very first: the delay due to weight and size ....
so up to the point now the first scanned parts of the measurement can be viewed at:
Miscellaneous Page 04
and at the end of the week the rest of the measurement papers should be scanned and published too....
Right Speaker-System:
so that´s at the point a thing to really discuss about....
yesterday I performed the task to do the measurements "reloaded" and that´s allready the first point to answer:
"Must admit, good way of using up all those odd drivers laying around.
Can't wait to see measurements! Also has me beat how he designed them without measurements in the first place. Can we have a list of drivers please?"
Where throughout this thread do you believe that i shall have stated that all the construction has been performed without measurements ? I can´t find that statement - i only see some people that suggest that.... but the opposite is correct -
throughout every stage the entire project has always been ruled by measurements up to the very point that each single speaker and each single crossover has been precisely measured and adjusted ! To me it seems that you´ve just "dropped" some pages from reading....
the next post is rather poor from arguments but rich with suggestions...
"I will continue to think it's a joke! It looks like my parts bin from around 30 years ago!
Some of those drivers have to be thirty years old! The reason I think it's a joke is those
two piezos nearly at floor level! They're there for the little folk!
"Up till now no physical measurements have been made...."
I did not saythat ! I just said that the measurements from time of construction ( 2005 ) are not availible anymore...
"No, really! You don't say! Very practical way to proceed! I must admit, I would love a view of the crossover, I bet it's all wired straight to the drivers!"
First: The speakersystem has been realized from 2003 to 2005.....
Second: The Speakersystem has been desired right from the beginning - that at least a distance of 5 Meters is between the listener and the speakers and at least 4 meters between the speakersystems themselves -
and last not least that no furniture is between the speakers and the listener ( so drop off from thinking that some furniture is between the horns and the listener ) and - what really is not well to be seen from the front photograph - the horn are not exactly plain to the front but have been mounted with a slight angle upward and to the imaginable middleaxis of the speakerpair.
And its not "practical" that the measurement from 2004 and 2005 have not been archived - but the basics of the entire project truely are reliable to measuremments that have been made.... but it´s always a good argument to issue suspicios thoughts and doubts if your out of reliable arguments....
"If you are approaching this in an ideal way you must be giving each driver less than an octave of spectrum?
Otherwise you must have multiple units covering overlapping frequency bands. I guarantee that approach will lead to a very messy 3 dimensional response. Can you show us polar curves or on and off axis frequency response?"
Within the running week all the rest of the measurements will be published and they all show one thing very clear - there is nothing comparable around.....
"Yikes!! 11 crossovers * 360 degrees = 3960 degrees of phase rotation. The group delay caused by phase rotation is generally inaudible with music and normal speakers but I'd be surprised if this one weren't audible. We're talking about the harmonics of an instrument arriving way before the fundamental."
... well thats a nice sugest but it does not come out true in measurement....
So also a lot of post originated from Phoenix358 are like:
"You really don't understand filter design at all! Inductors can be unwound to get an exact value.
It's not rocket science! Slightly altering capacitor values to stay within preferred values only slightly alters the filter Xover point and does to both the high pass and the low pass. The higher order the x-o the harder it gets but it is still not rocket science."
"Good for you! Let's see some measurements to back up your so called science!
Building without measurements ..... pah! I don't mind you deluding yourself!
I don't care for your attitude!
Don't try and delude the noobs with your nonsense, it's not fair on them.
Your claims so far are nothing but unsubstantiated male bovine droppings."
"If it's a joke it goes too far. I am thinking that he is for real. No one can design properly without measurements, not and come up with a design that beats the entire world and not from a collection of miss- matched drivers that rightly belong in a speaker museum."
"Yes, it's prediction time. My crystal balls are in play. We will never see a response graph!"
So beside the fact that most of his "output" is beyond any other comment just agitation instead of normal communication - your crystal ball is "out of order" and your "style" of posts disqualifies yourself from a serious response.... it´s not my level to argue below the guts...
And a short point to Holger:
In many years of more or less professional speaker design I have learned a thing or two. For example:
Component values with five percents of tolerance are absolutely fine, this is mostly far better then speaker parameter tolerances.
And if i look at the... well, not very high value drivers used here, I doubt that it is possible to build two somewhat identical speakers at all.
First Point: I am not talking about 5% tolerance - I´m talking about hiting the values within less then 0,5% of tolerance because i don´t trust tolerance - I only trust measurements just right down each single component....
And second: Your truely right that the overall tolerance of speakers ist far beyond 5 % - but there are two points that you did not mention that have been explained in the text:
a) there is only very small bandwith that the speaker has to perform within - and that is only in the very area where the speaker is operating within its optimum of performance... and b) each single speaker is at
the end calibrated within just that very small area.....
and another objection was:
"There does not seem to be much 'breathing space' for the 2 drivers <15Hz and 15-30Hz.
If you can pull that one off I'd start thinking about going into business making tiny subwoofers."
That is at the first view an interesting objection - but after all - I´ve done a lots of experiments with the large speakers to validate the real need of air to operate ( air and not casevibration ! ) and it turned out that it´s a good "fistrule" that a speaker is dependent to the air within a cube with sidelengths only at least the diameter of the speaker. Adding more volume does not really have much more influence to the ability of the cone to swing - it only causes the case to swing .....
The walls of this speakersystems are 50mm to 60mm thick beachwood and the true length does not exceed 60 cm. This wood is to thick and to small to vibrate in low frequencies and that was desired to be that way...
I want to generate sound by the speakers and not by resonance of the case.
and the best ( due to the fact that they restrict to simulation instead of adjustment ! ) Objection to my mind
"I'd never actually sim'd something so silly so I gave it a shot. This is the SPL, phase and group delay of 11 parallel electrical bandpass filters using the OP's frequencies. It's an 11-way as he has 2 supertweeters. The lowest bandpass is an LR4-15 highpass and an LR4-30 lowpass. The highest is an LR4-15360 highpass and an LR4-30770 lowpass. The others are spaced an octave apart in between.
Notice how the phase wraps through 360 degrees every octave."
I´m not only MBA in Math and electronic engineering - I´ve been working for years as freelancer with a company designing measurement-instruments and writing the programs for the Microprocessors....
and I´ve seen one thing for sure: programs are dependent to the abilities of the programmer.....
and a lot of simulations do not really carry out the task clean.... those you might get for free in the internet are of "midclass quality".....
"1) Colouration. Unless it is open back the cabinet will reflect the sound around
- eventually it will be heard as a rather serious 'boxy' sound."
The photographs published at the moment show the open steps - but of course prior to the final calibration of the single speakers all the "subcases" had been closed and on the backside of the speakersystems after the calibration has been performed the backwalls ( each 60 mm thick beechwood ! ) had been mounted and the Cases where closed
absolute airtight !
"2) Focus. I'm not usually big on focus, but by spacing your tweeters that far apart
then you have the sound for a cymbal (for instance) arriving from two different places,
causing some very odd imaging.. also carpet will eat the treble. "
maybe possible but not fact - in my livingroom the floor is massive wood ( otherwise it would have been foolish to make the subsonic-speaker looking towards the floor ) and the minimum distance from the speakers is from the beginning of the design to the final location in the room 5 meters away from both speakers. And frequencies above 10 kHz start to get dificult to locate - it´s the same trouble like with the very lower frequencies - it´s a disability of the ears - they can hear the frequencies but
the higher they get, the more difficult it becomes to locate them precisely .... and at the distance of more than 5 meters the distance between the tweeters ( less then 50 cm ) is not important any more...
"3) Complexity. Audio doesn't like complexity, you will hear the 12 crossovers ,
and your ears will be confused by the spacing of the drivers. The best sound these days
comes from single drivers or perhaps 2 drivers in my experience. Remember we hear all
sound via a single ear-drum on each side of our head!!"
That´s so true - but the sound of the drum does not come from 12 speakers but only from maximum of 2 speakers.... the real trouble comes from the fact that one speaker has problems to transmit many different sounds at the same time....
"I think you could improve your design by making a pyramid shape and putting the highest frequencies at the top and working down - in frequency order - to the floor with the bass.
You also need an enclosure for every open back unit - can't recall if you had that."
The Pyramid is a very good solution to avoid resonance and reflection-frequencies like "standing waves" and if the space between the listener and the speaker is less then 5 Meters it is also usefull due to the fact that midrage frequencies and lower highrange frequencies ( between 6500 Hz and 8000 Hz ) are quite well locateble and at the bottom of the pyramid is the needed volume of air to operate the bass-speaker ( at the top of the pyramid that volume would not be availible... )
Another Advantage is the fact that pyramidform enables you to have them less critical by placing them in the corner of a room...
"I would be interested to see a design with a wide midrange and fill in tweeter and bass,
but at least you are building and experimenting which is all good stuff and more than many do, just not sure it will sound how you imagine."
In fact the entire design has always been dependent to several "set limits" like minimum distance between auditors and speakers, a wide range of performance room ( the largest room the speakers have been used was a factory loft with floorarea of 25 meters x 15 meters and the distance to the ceiling 7,5 meters - straight down to the livingroom where i use them usually - with 6,5 meters x 7meters and ceiling at 2,7 meters. The speakers are usualy close to the 6 meter wall and 5 meters apart and at the oposite
side of the room 5 meters away are the seats. And there is no furniture between
seats and speakers.
"An 8th order bandpass (4 high and 4 low) has to swing from 360 degrees positive at the low end to 360 degrees lag at higher frequencies. Generally you'l have half that phase shift at the corner frequencies so: +180 and -180 or a 360 degree swing each section (repeat 11 times = 3960 degrees.).
How you have the sections add is not something Ronin is telling us, or is apparently aware of, but you always get the best response if you can get the phase swing of one section to overlay the next. Did we mention that the drivers have phase shift due to their response and physical depth? Did we mention that textbook crossovers don't acount for the driver as a load?"
Of course I´ve thought about that....
First of all each speaker has been isolated entirely from interaction with other speakers..... so in fact they might be treated like isolated single speakers
So in fact its a group of independent singlespeakers with extrem sharp "cuttoff" of unwanted sharing of frequencies.
Then second each speaker has been calibrated to operate with precise same soundpressure as any of the other speakers
and third the load... thats realy an issue due to the fact that during normal operation the passive crossovers all together drag nearly 16 amp power during normal performance - the amplifier must be able to deliver that current ... I have a MOS-FET Poweramp with 2 x 250 Watt Output ( up to 25 amp current to top of 70 Volts peakvoltage ) and the other alternate poweramp is a tubeamp with 2 x 140 watt ( each channel 6 x EL34 ) that comes up with 18 amp current from the output-transformers
and up to 60 Volt peakvoltage.
If you don´t bear the issue of the load in mind you realy can toast any outputstage into pure smoke...
"Any single cone is theoretically - and of course to some degree only practically as well - able to handle as many frequencies at once as one could imagine. In the same manner as the air is able to transport many frequencies at once and also like the mechanical parts of our ear are capable of handling. I think the OP's imagination doesn't allow for something like this and therefore all practically existing systems from single driver to four- or five-ways in the world won't work - probably not even when standing in front of one and actually listenenig to it."
In basic the thought is right - but in close view it can´t stand up....
The music (i.e. the instruments and vocals ) mix up within the air of a room to many different waves but in fact the ear does not only consist of the membrane in the ear .- but another point is often dropped from the view and thought: The outside foldings of the external ear fullfil an interesting and important duty: they are in a very special way a kind of "frequency-dependat-delay-filter". And the very tiny membrane of the inner ear is realy a very fast acting unit. On the other side the very large membrane
of a speaker is a very very slow unit. As long a the signal is within the amplifier there is not much trouble due to the fact that electricity is realy fast - but in the moment that electrical waves leave the mechanical transducer ( and electrostatic transducers are in that relation very slow and mechanic too... ) the trouble starts up to the fact that mechanic is bound to multiple misfits:
starting from the very first: the delay due to weight and size ....
so up to the point now the first scanned parts of the measurement can be viewed at:
Miscellaneous Page 04
and at the end of the week the rest of the measurement papers should be scanned and published too....
Left speaker-system:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
and the polar measurement:
An externally hosted image should be here but it was not working when we last tested it.
Right Speaker-System:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
and the polar measurement:
An externally hosted image should be here but it was not working when we last tested it.
so that´s at the point a thing to really discuss about....
Last edited by a moderator:
So what is the input power for the measurements? 1 watt?
Rob
do you seriously want a reply to that question ? probably the offset function in the setup of the Rhode and Schwarz measurement-equipment is the answer..... ?
please be kind - wait with requests untill all images and paperwork have been scanned and published.... this was just the "first sight" of the most demanded quests.... we spent yesterday seven hours of measuring work and I´m sittin on a bunch of protocols.... give me a chance to view and compress for the web and then publishing them... up to the end of this week the most of that will be done and if you still believe something is missing give it a shot on sunday...... is that possible ?
A kick drum covers from about 70Hz to a few kilo Hertz, that would be 5 or 6 drivers. A bass guitar covers a spectrum from 42Hz (38Hz for a 5 string) to 8kHz or more depending on strings and playing style which would be 6 or 7 drivers.
Last edited:
That´s true - but how many frequencies does the kickdrum issue at one single small time ? remember the frequencies of the instruments are issued sequentialy and not reallly as a parallel task. So in this system the sound is issued exactly from that speaker that is able to perform that very sound at its best and if the frequency changes up or down one of the neighbor-speakers continues the task. And now again a close view: what happens if a kickdrum is played at the same time with the bassguitar and a flute or violine for example ? In this system the different instruments "spread out" and each is reproduced with that speaker that is in the very small frequency area performing best... and what happens in a very close look to a speaker if you use only one two or three transducers ? Have you really thought about what happens - when a speaker is to issue a kickdrum at 80 Hz and a bass guitar at at for example 260 to 180 Hz and an organ at say something around 480 Hz at the same time ? Do you please explain how the cone membrane issues 3 different frequencies at the same time ? Think about it and you´ll start running in the right direction. Do you really believe that the stiff cone issues in one part one frequency and in another part another frequency and in a third area another third frequency ? No it does not ! The speaker starts to make an own "interpretation" of some kind of "Soundmixing" and dependent to the abilities of that speaker it will solve the problem with a more or less disability and compromize to its physical lacks.... and if you continue to view the physical action in detail, you will start to realize the real disadvantages of a System that contains only few speakers... the problem start at the physics of the speaker
Last edited:
But instruments do produce many frequencies in parallel! Otherwise a trumpet playing a c would sound exactly like an oboe playing the same note.
Think slap bass (like Level 42) as an extreme example: The guy hits the E string fretted so the fundamental is somewhere in 50s but the bass guitar will produce all the frequencies up to 10kHz simultaneously.
In a kick drum the beater will produce frequencies in the kHz region while the drum itself will produce the fundamental of around 70Hz and a number of overtones, all happens at the same time.
Even on a Hammond organ you have to take all the draw bars into account which adjust the harmonics and subharmonics which sound in parallel with the fundamental. That is you play an A (440Hz) but what you hear is made up of two subharmonics plus a number of harmonics.
Or even your source, be it vinyl or digital, it will produce one continuous voltage containing all frequencies at the same time.
Think slap bass (like Level 42) as an extreme example: The guy hits the E string fretted so the fundamental is somewhere in 50s but the bass guitar will produce all the frequencies up to 10kHz simultaneously.
In a kick drum the beater will produce frequencies in the kHz region while the drum itself will produce the fundamental of around 70Hz and a number of overtones, all happens at the same time.
Even on a Hammond organ you have to take all the draw bars into account which adjust the harmonics and subharmonics which sound in parallel with the fundamental. That is you play an A (440Hz) but what you hear is made up of two subharmonics plus a number of harmonics.
Or even your source, be it vinyl or digital, it will produce one continuous voltage containing all frequencies at the same time.
I`d have accepted nothing less than such a ruler flat response from a design of this gauge.
Unless the dent in the lower part of the curve (need a better circle) pretty nice even polar response too: all the same curve for 15 Hz, 300 Hz, 800 Hz, 2 kHz, 5 kHz, 9 kHz and 15 kHz, just as expected for a 12-way with drivers randomly placed all over the baffle
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.