Hello. I have been lurking and reading about acoustics for some time now, and have just enough information to thoroughly confuse myself. Hopefully you smart people here can chime in set me straight on a couple of items.
You are seated in the middle of The Bonneville Salt Flats in a comfortable chair in the early evening, sipping a brandy. You notice that the nearest vertical reflective surface is well over 1 mile away in all directions. In front of you, at a convenient distance (28 ft?) and on either side, are full range drivers. They are mounted (magically) without any baffle or hardware, but you have no doubt they are perfectly rigid. The speakers are fed by a system that, you again have no doubt, provides a perfect signal to the contacts on the driver. The system begins to play and you hear Cassandra Wilson's voice.
1) Is this system as close as possible to the ideal setup for accurately reproducing the signal?
2) In other words, is loudspeaker cabinet design simply the attempted solution to the problems:
a)
i) the driver generates sound in two directions +
ii) we usually listen to music around objects that sound would reflect upon =>
iii) we need to find something to do with the sound coming out of one direction of the speaker before it hits against something and the low frequency waves which are typically uniform in our listening environment cancel out
b)
There is usually no way to magically mount a speaker perfectly rigid without baffle or hardware, or even with baffle and hardware
3) Does the system without having to worry about bass cancellation deliver bass in the proper amounts? Or will single driver design always need a way to increase bass? E.g. is a full range driver capable of delivering appropriately leveled bass down to its rated limit? E.g. Is bass-loss with open baffle designs a matter of cancellation and not necessarily a limitation on the driver proper?
4) Does cancellation occur because: say you have a steady wave of 100 cycles / sec. The peak comes out the front, 1/200th of a second later the valley comes out the back. They are the same wave, just shifted 180degrees, so if they meet, they perfectly cancel?
If so 5) Then ideally you could take the wave out the back and "route" it for 1/200th of a second before "mixing" it back with the wave that came out the front, and therefore the two wave would be additive? Right?
If so 6) Then this would only work on the 100Hz waves because all the other frequencies would suffer from some percentage of addition.
If so 7) Then most single driver cabinets are just choosing a main frequency to enhance (with regard to sound that comes out of the port hole) which will enhance that frequency, and fall off on either side with the knowledge that over some frequency (is this 100Hz?) that cancellation becomes less and less of an issue.
8) What would happen if you build a speaker but your "cabinet" was the end of a tube (of the right size) that extended "somewhere else" -- like down into the basement (some 20 yards away physically) where the tube opened up directly into a box which contained 32 layers of Owens-Corning 703?
9) Is great sound out of a full range driver possible with innovation and the breaking of standard cabinet confines?
10) Does any of this have anything to do with the price of tea in China?
TIA.
You are seated in the middle of The Bonneville Salt Flats in a comfortable chair in the early evening, sipping a brandy. You notice that the nearest vertical reflective surface is well over 1 mile away in all directions. In front of you, at a convenient distance (28 ft?) and on either side, are full range drivers. They are mounted (magically) without any baffle or hardware, but you have no doubt they are perfectly rigid. The speakers are fed by a system that, you again have no doubt, provides a perfect signal to the contacts on the driver. The system begins to play and you hear Cassandra Wilson's voice.
1) Is this system as close as possible to the ideal setup for accurately reproducing the signal?
2) In other words, is loudspeaker cabinet design simply the attempted solution to the problems:
a)
i) the driver generates sound in two directions +
ii) we usually listen to music around objects that sound would reflect upon =>
iii) we need to find something to do with the sound coming out of one direction of the speaker before it hits against something and the low frequency waves which are typically uniform in our listening environment cancel out
b)
There is usually no way to magically mount a speaker perfectly rigid without baffle or hardware, or even with baffle and hardware
3) Does the system without having to worry about bass cancellation deliver bass in the proper amounts? Or will single driver design always need a way to increase bass? E.g. is a full range driver capable of delivering appropriately leveled bass down to its rated limit? E.g. Is bass-loss with open baffle designs a matter of cancellation and not necessarily a limitation on the driver proper?
4) Does cancellation occur because: say you have a steady wave of 100 cycles / sec. The peak comes out the front, 1/200th of a second later the valley comes out the back. They are the same wave, just shifted 180degrees, so if they meet, they perfectly cancel?
If so 5) Then ideally you could take the wave out the back and "route" it for 1/200th of a second before "mixing" it back with the wave that came out the front, and therefore the two wave would be additive? Right?
If so 6) Then this would only work on the 100Hz waves because all the other frequencies would suffer from some percentage of addition.
If so 7) Then most single driver cabinets are just choosing a main frequency to enhance (with regard to sound that comes out of the port hole) which will enhance that frequency, and fall off on either side with the knowledge that over some frequency (is this 100Hz?) that cancellation becomes less and less of an issue.
8) What would happen if you build a speaker but your "cabinet" was the end of a tube (of the right size) that extended "somewhere else" -- like down into the basement (some 20 yards away physically) where the tube opened up directly into a box which contained 32 layers of Owens-Corning 703?
9) Is great sound out of a full range driver possible with innovation and the breaking of standard cabinet confines?
10) Does any of this have anything to do with the price of tea in China?
TIA.
Ye Gods. I suspect you might be lucky to get many replies as few people have time to respond to a lengthy barrage of questions, myself included, but FWIW, briefly
1) No. It depends upon the criteria you use to define 'ideal.' As a ROT, there's no such thing.
2a) Who said radiation from both sides of a cone (or whatever) is a 'problem'? For the rest, it depends on the objectives of a design. There is no one answer.
2b)Er, correct. A remarkable observation.
3) Again, criteria and driver dependant. Partly cancellations, but if, say, you have a unit with a high mass-corner frequency, it's going to roll off with a more or less shallow slope regardless unless you support it with a baffle / enclosure. And then we come to things such as excursion -OBs for example are inefficient WRT the amount of mechanical movement a driver has to make (see Linkwitz)
4 &5) Basically yes for the first, more dubious for the second. A single tone, steady-state signal, you might get away with it. Anything else, you'd be in trouble, to say nothing of how you'd route it back to the front anyway.
6) I thought you were talking about a single 100Hz tone? But trying to follow what I think you're implying, basically yes.
7) Assuming a resonant phase inverting box like a BR, correct. You have a strong resonance at tuning, box output will then fall away more or less rapidly depending on the design to either side of this. A BLH is somewhat different of course -you have phase inversion, but a horn & their variations usually will have a wide[er] operating BW, with upper & lower corner frequencies (i.e. the upper & lower frequencies at which it rolls off, with it's BW occuring between the two). The upper corner frequency is necessary to prevent the horn running up too high in frequency, causing two distinct radiating sources which will screw up the imaging, and cause combing issues.
8) You've described one variation of an acoustic transmission line.
9) Depends how you define 'great sound' but it's possible to get very good sound. As for innovation etc., I don't want to sound discouraging, but one of the things I've learned over the past few years is that there is very little indeed that the pioneers of audio didn't do back in the 1920s & '30s when money & resources were being hurled at the subject by major players like Bell Labs. Since then, people have been rediscovering a lot of this (with the exception of a few people who never forgot it in the first place), so we see a situation, to paraphrase GM, where 'what once was old is again new.'
10) Unlikely.
1) No. It depends upon the criteria you use to define 'ideal.' As a ROT, there's no such thing.
2a) Who said radiation from both sides of a cone (or whatever) is a 'problem'? For the rest, it depends on the objectives of a design. There is no one answer.
2b)Er, correct. A remarkable observation.
3) Again, criteria and driver dependant. Partly cancellations, but if, say, you have a unit with a high mass-corner frequency, it's going to roll off with a more or less shallow slope regardless unless you support it with a baffle / enclosure. And then we come to things such as excursion -OBs for example are inefficient WRT the amount of mechanical movement a driver has to make (see Linkwitz)
4 &5) Basically yes for the first, more dubious for the second. A single tone, steady-state signal, you might get away with it. Anything else, you'd be in trouble, to say nothing of how you'd route it back to the front anyway.
6) I thought you were talking about a single 100Hz tone? But trying to follow what I think you're implying, basically yes.
7) Assuming a resonant phase inverting box like a BR, correct. You have a strong resonance at tuning, box output will then fall away more or less rapidly depending on the design to either side of this. A BLH is somewhat different of course -you have phase inversion, but a horn & their variations usually will have a wide[er] operating BW, with upper & lower corner frequencies (i.e. the upper & lower frequencies at which it rolls off, with it's BW occuring between the two). The upper corner frequency is necessary to prevent the horn running up too high in frequency, causing two distinct radiating sources which will screw up the imaging, and cause combing issues.
8) You've described one variation of an acoustic transmission line.
9) Depends how you define 'great sound' but it's possible to get very good sound. As for innovation etc., I don't want to sound discouraging, but one of the things I've learned over the past few years is that there is very little indeed that the pioneers of audio didn't do back in the 1920s & '30s when money & resources were being hurled at the subject by major players like Bell Labs. Since then, people have been rediscovering a lot of this (with the exception of a few people who never forgot it in the first place), so we see a situation, to paraphrase GM, where 'what once was old is again new.'
10) Unlikely.
Hi,
1) No.
It will not sound good given the limited info for lots of reasons.
Analysing reality in terms of perfection, e.g. perfectly rigid drivers,
gets you nowhere because they are not. Ignoring room acoustics
does not really help either, and without them any listening is
always effectively nearfield irrespective of distance.
2) The short answer is No.
3) is Yes, Usually, and Yes and Yes together (only).
4) is Yes but not perfectly.
The rest is a misunderstanding to a degree of ported and
transmission line speakers. Sealed boxes are straight forward.
Vented and transmission lines do not work by phase inversion.
Very big sealed boxes and long stuffed lines = large open baffle.
/sreten.
http://www.rjbaudio.com/Audiofiles/FRDtools.html
http://www.geocities.com/woove99/Spkrbldg/DesigningXO.htm
http://www.zaphaudio.com/
http://www.rjbaudio.com/projects.html
http://www.troelsgravesen.dk/Diy_Loudspeaker_Projects.htm
http://www.linkwitzlab.com/
1) No.
It will not sound good given the limited info for lots of reasons.
Analysing reality in terms of perfection, e.g. perfectly rigid drivers,
gets you nowhere because they are not. Ignoring room acoustics
does not really help either, and without them any listening is
always effectively nearfield irrespective of distance.
2) The short answer is No.
3) is Yes, Usually, and Yes and Yes together (only).
4) is Yes but not perfectly.
The rest is a misunderstanding to a degree of ported and
transmission line speakers. Sealed boxes are straight forward.
Vented and transmission lines do not work by phase inversion.
Very big sealed boxes and long stuffed lines = large open baffle.
/sreten.http://www.rjbaudio.com/Audiofiles/FRDtools.html
http://www.geocities.com/woove99/Spkrbldg/DesigningXO.htm
http://www.zaphaudio.com/
http://www.rjbaudio.com/projects.html
http://www.troelsgravesen.dk/Diy_Loudspeaker_Projects.htm
http://www.linkwitzlab.com/
Wow, thanks for all the great information.
I wanted to follow up:
With regard to the Salt Flats thought experiment, what I had in mind was kind of like the test for Artificial Intelligence, that is to say, getting as close as possible to being able to close your eyes, and having reasonable doubt whether you were listening to a live performance or a recording.
Given that the signal presented to the driver is "perfect", a loudspeaker setup involves many things which corrupt the sound one hears.
i) Given a "perfect" signal is the driver capable of creating the corresponding perfect sound?
ii) What do you do with the out-of-phase signal from the back of the driver?
iii) What do you do with reflecting signals?
So I created the thought experiment because I figured it removed as many of the problems as possible, e.g. the out-of-phase signal out the back of the driver is gone - you never hear it, there is no reflected signal to hear, you are hearing a full-range driver, so there are no cross-over components to color the signal.
I thought this has got to be as close as I can get to the live performance. ???
But you mention that OBs are inefficient with respect to the amount of mechanical movement that the driver has to make.
Are you saying that driver is inefficient and this inefficiency will cause distortion? or are you saying something like the driver is inefficient, and you will need to turn up the volume more than with a speaker in an enclosure but there is not necessarily any distortion introduced because of the inefficiency....
Because these 2 options indicate different paths...
If the IB does not introduce distortion because of inefficiency, then surely this is the way to go....
I had thought that enclosures are "good" solutions to the following:
a) out-of-phase radiation
b) aesthetics
And perfect solutions to
a) physical source placement
But a compromise to
a) Signal "purity"
Whereas an IB would be a near-perfect solution to
a) out-of-phase radiation
b) signal purity
And have significant work to do with regard to
a) aesthetics
b) physical source placement
Steve Deckart talks about OB
http://www.decware.com/paper92.htm
http://www.decware.com/paper93.htm
and
http://www.decware.com/paper94.htm
and I am fascinated by his claims of freeing the driver from the influences and slurring that occur as it is contained in an enclosure.
I am assuming that an IB mounted such that the back of the driver is enclosed in a large enough space effectively acts as an OB (if not, well then this whole line of "logic" is screwed).
His solution to bass cancellation that occurs because the front and back of his driver vent into the same space is to capture signal (which contains the whole pure bass) and amplify it.
So basically this whole line of reasoning is to say that if we take a less portable solution (the enclosure) and replace it with a non-portable solution (IB), can we have it all? the whole enchilada?
I am assuming that you see a lot of IB subs because the placement is not as critical, but you don't see a lot of IB full range drivers because then placement comes into play.
But suppose you have a room where you could mount the full-rangers on an enclosure that protrudes from the wall or ceiling to bring the speakers into proper position? You place one 45 degree ceramic tile behind the speaker to route the out of phase signal "into" the wall or ceiling.
You could even make the enclosures detachable.
So, in my minds eye, if you can live with the aesthetics of the IB full range "solution", you solve all the problems.
Hit me guys, what a) am I missing? b) have I over-simplified?
TIA.
I wanted to follow up:
With regard to the Salt Flats thought experiment, what I had in mind was kind of like the test for Artificial Intelligence, that is to say, getting as close as possible to being able to close your eyes, and having reasonable doubt whether you were listening to a live performance or a recording.
Given that the signal presented to the driver is "perfect", a loudspeaker setup involves many things which corrupt the sound one hears.
i) Given a "perfect" signal is the driver capable of creating the corresponding perfect sound?
ii) What do you do with the out-of-phase signal from the back of the driver?
iii) What do you do with reflecting signals?
So I created the thought experiment because I figured it removed as many of the problems as possible, e.g. the out-of-phase signal out the back of the driver is gone - you never hear it, there is no reflected signal to hear, you are hearing a full-range driver, so there are no cross-over components to color the signal.
I thought this has got to be as close as I can get to the live performance. ???
But you mention that OBs are inefficient with respect to the amount of mechanical movement that the driver has to make.
Are you saying that driver is inefficient and this inefficiency will cause distortion? or are you saying something like the driver is inefficient, and you will need to turn up the volume more than with a speaker in an enclosure but there is not necessarily any distortion introduced because of the inefficiency....
Because these 2 options indicate different paths...
If the IB does not introduce distortion because of inefficiency, then surely this is the way to go....
I had thought that enclosures are "good" solutions to the following:
a) out-of-phase radiation
b) aesthetics
And perfect solutions to
a) physical source placement
But a compromise to
a) Signal "purity"
Whereas an IB would be a near-perfect solution to
a) out-of-phase radiation
b) signal purity
And have significant work to do with regard to
a) aesthetics
b) physical source placement
Steve Deckart talks about OB
http://www.decware.com/paper92.htm
http://www.decware.com/paper93.htm
and
http://www.decware.com/paper94.htm
and I am fascinated by his claims of freeing the driver from the influences and slurring that occur as it is contained in an enclosure.
I am assuming that an IB mounted such that the back of the driver is enclosed in a large enough space effectively acts as an OB (if not, well then this whole line of "logic" is screwed).
His solution to bass cancellation that occurs because the front and back of his driver vent into the same space is to capture signal (which contains the whole pure bass) and amplify it.
So basically this whole line of reasoning is to say that if we take a less portable solution (the enclosure) and replace it with a non-portable solution (IB), can we have it all? the whole enchilada?
I am assuming that you see a lot of IB subs because the placement is not as critical, but you don't see a lot of IB full range drivers because then placement comes into play.
But suppose you have a room where you could mount the full-rangers on an enclosure that protrudes from the wall or ceiling to bring the speakers into proper position? You place one 45 degree ceramic tile behind the speaker to route the out of phase signal "into" the wall or ceiling.
You could even make the enclosures detachable.
So, in my minds eye, if you can live with the aesthetics of the IB full range "solution", you solve all the problems.
Hit me guys, what a) am I missing? b) have I over-simplified?
TIA.
Oh dear God in Heaven, another battery. Piece of advice: ease up. Sloooowww. A couple of things at a time. We don't have time to devote to lengthy missives. Very quickly because I've only a couple of minutes
i) No signal is 'perfect' so this has to be kept in mind.
ii) Whatever you want, providing you design it well
iii) What reflections? Off objects in the room? Short of some room treatments, live with them. Accept the inevitable.
No, because the recording is limited itself.
You're trying to read things into the above that aren't there. OBs are mechanically inefficient. Read up on your Linkwitz. The driver will have to move more for a given SPL than it would in other types of enclosure / cabinet / whatever you want to call it. As there is a limit to linear excursion (Xmax), this will limit ultimate SPLs for a reasonable level of distortion (Xmax = 10%).
You're obsessing over the back-wave. Why is it a problem all of a sudden? Aesthetics = a matter of personal taste. Physical source placement & signal purity = in what context are you talking about? For the next couple of lines, read this paragraph once more.
The mythical 'perfect' IB is analogous to a dipolar souce insofar as if the latter were large enough to prevent you from ever hearing any of the rearward radiation from the driver.
No, you can never have it all. There is always compromise. Period. We live in the real world.
No, IB subs are prefered (usually) due to their transient response. You've got low Fs units with fairly heavy cones, massive linear excursion & high power handling, usually coupled to dedicated plate-amps & frequently with some Eq applied. Not traits FR drivers are famous for. Some can be used effectively in this way, many can't.
Re suddenly thinking you've found a panacea WRT IBs, what about the driver's mass-corner & LF roll-off? Midrange or HF peaking? Its... you get the idea.
'Proper' placement depends on the room, drivers, preferences etc.
i) No signal is 'perfect' so this has to be kept in mind.
ii) Whatever you want, providing you design it well
iii) What reflections? Off objects in the room? Short of some room treatments, live with them. Accept the inevitable.
No, because the recording is limited itself.
You're trying to read things into the above that aren't there. OBs are mechanically inefficient. Read up on your Linkwitz. The driver will have to move more for a given SPL than it would in other types of enclosure / cabinet / whatever you want to call it. As there is a limit to linear excursion (Xmax), this will limit ultimate SPLs for a reasonable level of distortion (Xmax = 10%).
You're obsessing over the back-wave. Why is it a problem all of a sudden? Aesthetics = a matter of personal taste. Physical source placement & signal purity = in what context are you talking about? For the next couple of lines, read this paragraph once more.
The mythical 'perfect' IB is analogous to a dipolar souce insofar as if the latter were large enough to prevent you from ever hearing any of the rearward radiation from the driver.
No, you can never have it all. There is always compromise. Period. We live in the real world.
No, IB subs are prefered (usually) due to their transient response. You've got low Fs units with fairly heavy cones, massive linear excursion & high power handling, usually coupled to dedicated plate-amps & frequently with some Eq applied. Not traits FR drivers are famous for. Some can be used effectively in this way, many can't.
Re suddenly thinking you've found a panacea WRT IBs, what about the driver's mass-corner & LF roll-off? Midrange or HF peaking? Its... you get the idea.
'Proper' placement depends on the room, drivers, preferences etc.
Scottmoose said:
Hi,
Yes you are correct, a poor choice of wording.
Depends on what phase is being inverted and where. 1/4 wave TL's
invert phase at the mouth, the whole thing adds up to a wavelength.
What I meant was though is that the addition of phases is not
by some delay mechanism equivalent to 180 degree phase shift.
Both vented boxes and TLs employ tuned systems that oscillate
in phase with frontal output driven by the out phase rear signal.
/sreten.
Hi,
Oh dear ....
Once Steve Deckart starts being referenced I should politely leave.
Only goes to prove a little knowledge is extremely dangerous, an
affliction SD exploits to the full, well he can, it is his website .......
I'd stop reading such pseudotechobabbledrivel and get a grounding
in the real physics of the matter rather than idealistic "perfection".
/sreten.
Oh dear ....
Once Steve Deckart starts being referenced I should politely leave.
Only goes to prove a little knowledge is extremely dangerous, an
affliction SD exploits to the full, well he can, it is his website .......
I'd stop reading such pseudotechobabbledrivel and get a grounding
in the real physics of the matter rather than idealistic "perfection".
/sreten.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.