Reply
 
Thread Tools Display Modes
  #81   Report Post  
Richard Crowley
 
Posts: n/a
Default More cable questions!


"dangling entity" wrote in message
m...
Kevin McMurtrie wrote in message

...

It's all irrelevant for audio frequencies and normal lengths of wire.
Having the two conductors side by side is perfectly good. Just don't
split the wires and route them to the speaker along opposite walls.


Just curious, but what *would* happen if you did that?


The listener(s) would be inside a 1-turn magnetic loop.

Actually, some hearing-assist systems work by exactly that
method. They use "receivers" with pickup coils and amps
that drive the headphones. And some hearing aids will pick
it up directly (from the coils they use to pick up telephone
receiver audio.)


  #82   Report Post  
Richard Crowley
 
Posts: n/a
Default More cable questions!


"dangling entity" wrote in message
m...
Kevin McMurtrie wrote in message

...

It's all irrelevant for audio frequencies and normal lengths of wire.
Having the two conductors side by side is perfectly good. Just don't
split the wires and route them to the speaker along opposite walls.


Just curious, but what *would* happen if you did that?


The listener(s) would be inside a 1-turn magnetic loop.

Actually, some hearing-assist systems work by exactly that
method. They use "receivers" with pickup coils and amps
that drive the headphones. And some hearing aids will pick
it up directly (from the coils they use to pick up telephone
receiver audio.)


  #83   Report Post  
Bruce
 
Posts: n/a
Default More cable questions!

"Richard Crowley" wrote in news:vvb788bo5ro346
@corp.supernews.com:

"Bruce" wrote ...
Transmission line theory is useless below about a 1/10
wavelength. At 20 kHz, this is several thousand feet.

Why would anyone want to worry about 100kHz, when
most people can no longer hear 20kHz?


If you have to ask then you'll never get it! :-))



Riiiiggghhhhtt.....
  #84   Report Post  
Bruce
 
Posts: n/a
Default More cable questions!

"Richard Crowley" wrote in news:vvb788bo5ro346
@corp.supernews.com:

"Bruce" wrote ...
Transmission line theory is useless below about a 1/10
wavelength. At 20 kHz, this is several thousand feet.

Why would anyone want to worry about 100kHz, when
most people can no longer hear 20kHz?


If you have to ask then you'll never get it! :-))



Riiiiggghhhhtt.....
  #85   Report Post  
Bruce
 
Posts: n/a
Default More cable questions!

"Richard Crowley" wrote in news:vvb788bo5ro346
@corp.supernews.com:

"Bruce" wrote ...
Transmission line theory is useless below about a 1/10
wavelength. At 20 kHz, this is several thousand feet.

Why would anyone want to worry about 100kHz, when
most people can no longer hear 20kHz?


If you have to ask then you'll never get it! :-))



Riiiiggghhhhtt.....


  #92   Report Post  
Bob-Stanton
 
Posts: n/a
Default More cable questions!

"Rusty Boudreaux" wrote in message


So how do you enter into the computer the load impedance being it
is a complex fuction of frequency?


There a two ways one could do it.

1) One could use a general lumped constant model of a speaker as the
termination of the line.

2) The better way is: measure the input impedance of a given speaker
at many frequencies (at 100 frequencies would be good). Mathemetically
convert the input impedance into S11 (input reflection cofficient).
One can then put in dummy values for S22 (output reflection
cofficient), S12 (reverse transmission cofficient) and S21 (forward
transmission cofficient). Create a two-port (data file) device from
the data. Terminate the cable (in the computer) with the two-port
(data) device.

The two-port device will present exactly the impedance you measured,
at the frequencies you measured. The computer will interpolate values
in between the frequencies you measured, to give the correct complex
impedance, at all frequencies.

Simple, no? :-)

Bob Stanton
  #93   Report Post  
Bob-Stanton
 
Posts: n/a
Default More cable questions!

"Rusty Boudreaux" wrote in message


So how do you enter into the computer the load impedance being it
is a complex fuction of frequency?


There a two ways one could do it.

1) One could use a general lumped constant model of a speaker as the
termination of the line.

2) The better way is: measure the input impedance of a given speaker
at many frequencies (at 100 frequencies would be good). Mathemetically
convert the input impedance into S11 (input reflection cofficient).
One can then put in dummy values for S22 (output reflection
cofficient), S12 (reverse transmission cofficient) and S21 (forward
transmission cofficient). Create a two-port (data file) device from
the data. Terminate the cable (in the computer) with the two-port
(data) device.

The two-port device will present exactly the impedance you measured,
at the frequencies you measured. The computer will interpolate values
in between the frequencies you measured, to give the correct complex
impedance, at all frequencies.

Simple, no? :-)

Bob Stanton
  #94   Report Post  
Bob-Stanton
 
Posts: n/a
Default More cable questions!

"Rusty Boudreaux" wrote in message


So how do you enter into the computer the load impedance being it
is a complex fuction of frequency?


There a two ways one could do it.

1) One could use a general lumped constant model of a speaker as the
termination of the line.

2) The better way is: measure the input impedance of a given speaker
at many frequencies (at 100 frequencies would be good). Mathemetically
convert the input impedance into S11 (input reflection cofficient).
One can then put in dummy values for S22 (output reflection
cofficient), S12 (reverse transmission cofficient) and S21 (forward
transmission cofficient). Create a two-port (data file) device from
the data. Terminate the cable (in the computer) with the two-port
(data) device.

The two-port device will present exactly the impedance you measured,
at the frequencies you measured. The computer will interpolate values
in between the frequencies you measured, to give the correct complex
impedance, at all frequencies.

Simple, no? :-)

Bob Stanton
  #107   Report Post  
Ian
 
Posts: n/a
Default More cable questions!


"Stewart Pinkerton" wrote in message
...
On 3 Jan 2004 04:39:58 -0800, (Bob-Stanton)
wrote:

"Rusty Boudreaux" wrote in message

So how do you enter into the computer the load impedance being it
is a complex fuction of frequency?


There a two ways one could do it.

1) One could use a general lumped constant model of a speaker as the
termination of the line.

2) The better way is: measure the input impedance of a given speaker
at many frequencies (at 100 frequencies would be good). Mathemetically
convert the input impedance into S11 (input reflection cofficient).
One can then put in dummy values for S22 (output reflection
cofficient), S12 (reverse transmission cofficient) and S21 (forward
transmission cofficient). Create a two-port (data file) device from
the data. Terminate the cable (in the computer) with the two-port
(data) device.

The two-port device will present exactly the impedance you measured,
at the frequencies you measured. The computer will interpolate values
in between the frequencies you measured, to give the correct complex
impedance, at all frequencies.

Simple, no? :-)


Indeed yes. Now tell us how you optimise a transmission line between
the sub-ohm source impedance of the amplifier, and the wildly varying
multi-ohm load impedance of the speaker. Sheesh, whatta maroon!
--

Stewart Pinkerton | Music is Art - Audio is Engineering


Go to the Linear Technology web site, download the (free) Spice
modelling program, and try out various cable configurations.

Or measure it.

Really, if you measure it someone might learn what is significant, and
what is not.

Regards
Ian

(Stewart, not getting at you, I agree with your .sig)


  #108   Report Post  
Ian
 
Posts: n/a
Default More cable questions!


"Stewart Pinkerton" wrote in message
...
On 3 Jan 2004 04:39:58 -0800, (Bob-Stanton)
wrote:

"Rusty Boudreaux" wrote in message

So how do you enter into the computer the load impedance being it
is a complex fuction of frequency?


There a two ways one could do it.

1) One could use a general lumped constant model of a speaker as the
termination of the line.

2) The better way is: measure the input impedance of a given speaker
at many frequencies (at 100 frequencies would be good). Mathemetically
convert the input impedance into S11 (input reflection cofficient).
One can then put in dummy values for S22 (output reflection
cofficient), S12 (reverse transmission cofficient) and S21 (forward
transmission cofficient). Create a two-port (data file) device from
the data. Terminate the cable (in the computer) with the two-port
(data) device.

The two-port device will present exactly the impedance you measured,
at the frequencies you measured. The computer will interpolate values
in between the frequencies you measured, to give the correct complex
impedance, at all frequencies.

Simple, no? :-)


Indeed yes. Now tell us how you optimise a transmission line between
the sub-ohm source impedance of the amplifier, and the wildly varying
multi-ohm load impedance of the speaker. Sheesh, whatta maroon!
--

Stewart Pinkerton | Music is Art - Audio is Engineering


Go to the Linear Technology web site, download the (free) Spice
modelling program, and try out various cable configurations.

Or measure it.

Really, if you measure it someone might learn what is significant, and
what is not.

Regards
Ian

(Stewart, not getting at you, I agree with your .sig)


  #109   Report Post  
Ian
 
Posts: n/a
Default More cable questions!


"Stewart Pinkerton" wrote in message
...
On 3 Jan 2004 04:39:58 -0800, (Bob-Stanton)
wrote:

"Rusty Boudreaux" wrote in message

So how do you enter into the computer the load impedance being it
is a complex fuction of frequency?


There a two ways one could do it.

1) One could use a general lumped constant model of a speaker as the
termination of the line.

2) The better way is: measure the input impedance of a given speaker
at many frequencies (at 100 frequencies would be good). Mathemetically
convert the input impedance into S11 (input reflection cofficient).
One can then put in dummy values for S22 (output reflection
cofficient), S12 (reverse transmission cofficient) and S21 (forward
transmission cofficient). Create a two-port (data file) device from
the data. Terminate the cable (in the computer) with the two-port
(data) device.

The two-port device will present exactly the impedance you measured,
at the frequencies you measured. The computer will interpolate values
in between the frequencies you measured, to give the correct complex
impedance, at all frequencies.

Simple, no? :-)


Indeed yes. Now tell us how you optimise a transmission line between
the sub-ohm source impedance of the amplifier, and the wildly varying
multi-ohm load impedance of the speaker. Sheesh, whatta maroon!
--

Stewart Pinkerton | Music is Art - Audio is Engineering


Go to the Linear Technology web site, download the (free) Spice
modelling program, and try out various cable configurations.

Or measure it.

Really, if you measure it someone might learn what is significant, and
what is not.

Regards
Ian

(Stewart, not getting at you, I agree with your .sig)


  #110   Report Post  
Dick Pierce
 
Posts: n/a
Default More cable questions!

(Bob-Stanton) wrote in message . com...
(Stewart Pinkerton) wrote in message
While true, this is irrelevant to the fact that lumped theory is
perfectly adequate for audio frequencies in domestic situations.


Yes, I agree that lumped constant models are perfectly adaquate for
audio frequencies, if one doesn't have the software necessary for
modeling a true transmission line.

BTW, have *you* ever tried modeling a *real* speaker cable using
lumped constants?


BTW, have *you* EVER bothered to see if your "theory" results
in predictions that you have then compared with ACTUAL
measurements?

I'm curious to see what you used as a lumped
constant model. Please show us a model of 100 ft of standard (Home
Depot), 12 gage cable (terminated by an ideal 8 Ohm load).


Please show us who is using 100 ft of standard (Home Depot)
12 gauge cable in a typical home listening situation.

Please show us ANYONE whose speaker cables are terminated by
an ideal 8 ohm load.

Mr. Stanton, your model is **** NOT because its a transmission
line or any other model, it's **** becuase of your grossly incorrect
assumptions and the fact that these assumptions simply don't exist
in actual situations.

BTW, have you ever tried modelling a *real* loudspeaker as the load?


Dick Pierce in the past, has presented a lumped constant loudspeaker
model. I'm sure it is adaquate for use as a load.


And that is yet more evidence of how far from physical reality
your "model" is. I have presented a NUMBER of lumped parameter
(not "lumped constant") models, as each and every speaker system
presents a significantluy different load.

So, let's review your assumptions behind your "model" once again:

1. You assume that people are using 100 feet of cable.
But people VERY RARELY use 100 feet of cable, it's more
typically 1/10th that distance, making the necessity of
a transmission line model even more irrelevant and
unnecessary.

2. You assume that the cable is terminated by an ideal
8 ohm load.
But NO speaker is anything approaching an ideal 8 ohm
load.

3. You have looked at ONE example of a non-ideal load.
But, apparently, you have never incorporated such a non-
ideal load in ANY of your models.
Further, you have apparently ignored the fact that one
lumped parameter model simlpy is not representative
of the enormous variations in actual speaker loads.

And, finally:

4. You have never once presented a single shred of physical
evidence in support of your "theory" that demonstrates
its superiority or even its very efficacy. You insist
your "theory" is right, but are unable or, more likely,
simply unwilling to do ANY of the work YOU need to do
to support it.

YOUR theory, based on your gross missapplication of transmission
line principles, your preposterous assumptions of operating
conditions, and your long-demonstrated inability to relate it to
any real-world performance issues indeed makes YOUR theory useless.


  #111   Report Post  
Dick Pierce
 
Posts: n/a
Default More cable questions!

(Bob-Stanton) wrote in message . com...
(Stewart Pinkerton) wrote in message
While true, this is irrelevant to the fact that lumped theory is
perfectly adequate for audio frequencies in domestic situations.


Yes, I agree that lumped constant models are perfectly adaquate for
audio frequencies, if one doesn't have the software necessary for
modeling a true transmission line.

BTW, have *you* ever tried modeling a *real* speaker cable using
lumped constants?


BTW, have *you* EVER bothered to see if your "theory" results
in predictions that you have then compared with ACTUAL
measurements?

I'm curious to see what you used as a lumped
constant model. Please show us a model of 100 ft of standard (Home
Depot), 12 gage cable (terminated by an ideal 8 Ohm load).


Please show us who is using 100 ft of standard (Home Depot)
12 gauge cable in a typical home listening situation.

Please show us ANYONE whose speaker cables are terminated by
an ideal 8 ohm load.

Mr. Stanton, your model is **** NOT because its a transmission
line or any other model, it's **** becuase of your grossly incorrect
assumptions and the fact that these assumptions simply don't exist
in actual situations.

BTW, have you ever tried modelling a *real* loudspeaker as the load?


Dick Pierce in the past, has presented a lumped constant loudspeaker
model. I'm sure it is adaquate for use as a load.


And that is yet more evidence of how far from physical reality
your "model" is. I have presented a NUMBER of lumped parameter
(not "lumped constant") models, as each and every speaker system
presents a significantluy different load.

So, let's review your assumptions behind your "model" once again:

1. You assume that people are using 100 feet of cable.
But people VERY RARELY use 100 feet of cable, it's more
typically 1/10th that distance, making the necessity of
a transmission line model even more irrelevant and
unnecessary.

2. You assume that the cable is terminated by an ideal
8 ohm load.
But NO speaker is anything approaching an ideal 8 ohm
load.

3. You have looked at ONE example of a non-ideal load.
But, apparently, you have never incorporated such a non-
ideal load in ANY of your models.
Further, you have apparently ignored the fact that one
lumped parameter model simlpy is not representative
of the enormous variations in actual speaker loads.

And, finally:

4. You have never once presented a single shred of physical
evidence in support of your "theory" that demonstrates
its superiority or even its very efficacy. You insist
your "theory" is right, but are unable or, more likely,
simply unwilling to do ANY of the work YOU need to do
to support it.

YOUR theory, based on your gross missapplication of transmission
line principles, your preposterous assumptions of operating
conditions, and your long-demonstrated inability to relate it to
any real-world performance issues indeed makes YOUR theory useless.
  #112   Report Post  
Dick Pierce
 
Posts: n/a
Default More cable questions!

(Bob-Stanton) wrote in message . com...
(Stewart Pinkerton) wrote in message
While true, this is irrelevant to the fact that lumped theory is
perfectly adequate for audio frequencies in domestic situations.


Yes, I agree that lumped constant models are perfectly adaquate for
audio frequencies, if one doesn't have the software necessary for
modeling a true transmission line.

BTW, have *you* ever tried modeling a *real* speaker cable using
lumped constants?


BTW, have *you* EVER bothered to see if your "theory" results
in predictions that you have then compared with ACTUAL
measurements?

I'm curious to see what you used as a lumped
constant model. Please show us a model of 100 ft of standard (Home
Depot), 12 gage cable (terminated by an ideal 8 Ohm load).


Please show us who is using 100 ft of standard (Home Depot)
12 gauge cable in a typical home listening situation.

Please show us ANYONE whose speaker cables are terminated by
an ideal 8 ohm load.

Mr. Stanton, your model is **** NOT because its a transmission
line or any other model, it's **** becuase of your grossly incorrect
assumptions and the fact that these assumptions simply don't exist
in actual situations.

BTW, have you ever tried modelling a *real* loudspeaker as the load?


Dick Pierce in the past, has presented a lumped constant loudspeaker
model. I'm sure it is adaquate for use as a load.


And that is yet more evidence of how far from physical reality
your "model" is. I have presented a NUMBER of lumped parameter
(not "lumped constant") models, as each and every speaker system
presents a significantluy different load.

So, let's review your assumptions behind your "model" once again:

1. You assume that people are using 100 feet of cable.
But people VERY RARELY use 100 feet of cable, it's more
typically 1/10th that distance, making the necessity of
a transmission line model even more irrelevant and
unnecessary.

2. You assume that the cable is terminated by an ideal
8 ohm load.
But NO speaker is anything approaching an ideal 8 ohm
load.

3. You have looked at ONE example of a non-ideal load.
But, apparently, you have never incorporated such a non-
ideal load in ANY of your models.
Further, you have apparently ignored the fact that one
lumped parameter model simlpy is not representative
of the enormous variations in actual speaker loads.

And, finally:

4. You have never once presented a single shred of physical
evidence in support of your "theory" that demonstrates
its superiority or even its very efficacy. You insist
your "theory" is right, but are unable or, more likely,
simply unwilling to do ANY of the work YOU need to do
to support it.

YOUR theory, based on your gross missapplication of transmission
line principles, your preposterous assumptions of operating
conditions, and your long-demonstrated inability to relate it to
any real-world performance issues indeed makes YOUR theory useless.
  #116   Report Post  
Stewart Pinkerton
 
Posts: n/a
Default More cable questions!

On 4 Jan 2004 04:05:10 -0800, (Bob-Stanton)
wrote:

(Stewart Pinkerton) wrote in message

Indeed yes. Now tell us how you optimise a transmission line between
the sub-ohm source impedance of the amplifier, and the wildly varying
multi-ohm load impedance of the speaker. Sheesh, whatta maroon!


I realise your knowledge of available circuit analysis programs is
somewhat limited. It is not my primary intent to put you down, rather
I writing this to inform.


Indeed, I've only been a professional electronics engineer and
precision analogue specialist for about thirty years, so my knowledge
is certainly limited, Not so limited as some others, of course......

Some circuit analysis programs can handle the conditions you started
above.


Not without a good model of the load impedance, they can't!

With them one can model a circuit using either a voltage source
or a 50 Ohm source, at audio and RF frequenies, with the output
results either in S-parameters or in volts (dB). With these programs
one can use mixed audio and RF components, such as a speaker cable
terminated with a one-port or two-port data model. The only limitation
being that if someone wants to have the results in terms of
S-parameters, there must be both a resistive source (typically 50 or
75 Ohms) and a resistive termination.


Hence they are of course not practical for use between audio
components, which have neither.

Your experience is probably limited to programs (such as SPICE) that
can not handle RF circuits, and RF programs that can not handle audio
circuits.


You are of course simply ducking and diving here, in order to avoid
answering the question. Regardless of which tool you use, you *must*
have a good model of the load impedance - which you don't have. Now,
tell us how you optimise a transmission line between the sub-ohm
source impedance of the amplifier, and the wildly varying multi-ohm
load impedance of the speaker.

Sheesh, whatta maroon!
--

Stewart Pinkerton | Music is Art - Audio is Engineering
  #117   Report Post  
Stewart Pinkerton
 
Posts: n/a
Default More cable questions!

On 4 Jan 2004 04:05:10 -0800, (Bob-Stanton)
wrote:

(Stewart Pinkerton) wrote in message

Indeed yes. Now tell us how you optimise a transmission line between
the sub-ohm source impedance of the amplifier, and the wildly varying
multi-ohm load impedance of the speaker. Sheesh, whatta maroon!


I realise your knowledge of available circuit analysis programs is
somewhat limited. It is not my primary intent to put you down, rather
I writing this to inform.


Indeed, I've only been a professional electronics engineer and
precision analogue specialist for about thirty years, so my knowledge
is certainly limited, Not so limited as some others, of course......

Some circuit analysis programs can handle the conditions you started
above.


Not without a good model of the load impedance, they can't!

With them one can model a circuit using either a voltage source
or a 50 Ohm source, at audio and RF frequenies, with the output
results either in S-parameters or in volts (dB). With these programs
one can use mixed audio and RF components, such as a speaker cable
terminated with a one-port or two-port data model. The only limitation
being that if someone wants to have the results in terms of
S-parameters, there must be both a resistive source (typically 50 or
75 Ohms) and a resistive termination.


Hence they are of course not practical for use between audio
components, which have neither.

Your experience is probably limited to programs (such as SPICE) that
can not handle RF circuits, and RF programs that can not handle audio
circuits.


You are of course simply ducking and diving here, in order to avoid
answering the question. Regardless of which tool you use, you *must*
have a good model of the load impedance - which you don't have. Now,
tell us how you optimise a transmission line between the sub-ohm
source impedance of the amplifier, and the wildly varying multi-ohm
load impedance of the speaker.

Sheesh, whatta maroon!
--

Stewart Pinkerton | Music is Art - Audio is Engineering
  #118   Report Post  
Stewart Pinkerton
 
Posts: n/a
Default More cable questions!

On 4 Jan 2004 04:05:10 -0800, (Bob-Stanton)
wrote:

(Stewart Pinkerton) wrote in message

Indeed yes. Now tell us how you optimise a transmission line between
the sub-ohm source impedance of the amplifier, and the wildly varying
multi-ohm load impedance of the speaker. Sheesh, whatta maroon!


I realise your knowledge of available circuit analysis programs is
somewhat limited. It is not my primary intent to put you down, rather
I writing this to inform.


Indeed, I've only been a professional electronics engineer and
precision analogue specialist for about thirty years, so my knowledge
is certainly limited, Not so limited as some others, of course......

Some circuit analysis programs can handle the conditions you started
above.


Not without a good model of the load impedance, they can't!

With them one can model a circuit using either a voltage source
or a 50 Ohm source, at audio and RF frequenies, with the output
results either in S-parameters or in volts (dB). With these programs
one can use mixed audio and RF components, such as a speaker cable
terminated with a one-port or two-port data model. The only limitation
being that if someone wants to have the results in terms of
S-parameters, there must be both a resistive source (typically 50 or
75 Ohms) and a resistive termination.


Hence they are of course not practical for use between audio
components, which have neither.

Your experience is probably limited to programs (such as SPICE) that
can not handle RF circuits, and RF programs that can not handle audio
circuits.


You are of course simply ducking and diving here, in order to avoid
answering the question. Regardless of which tool you use, you *must*
have a good model of the load impedance - which you don't have. Now,
tell us how you optimise a transmission line between the sub-ohm
source impedance of the amplifier, and the wildly varying multi-ohm
load impedance of the speaker.

Sheesh, whatta maroon!
--

Stewart Pinkerton | Music is Art - Audio is Engineering
  #119   Report Post  
Bob-Stanton
 
Posts: n/a
Default More cable questions!

(Dick Pierce) wrote in message


BTW, have *you* EVER bothered to see if your "theory" results
in predictions that you have then compared with ACTUAL
measurements?



Thank you for again giving me credit for creating established
transmsission line theory (by your calling it "your theory").

I don't feel it is necessary for me to reprove basic transmission line
theory. The basic theory could be wrong of course, but I'll leave that
up to you to prove by measurements.



I'm curious to see what you used as a lumped
constant model. Please show us a model of 100 ft of standard (Home
Depot), 12 gage cable (terminated by an ideal 8 Ohm load).


Please show us who is using 100 ft of standard (Home Depot)
12 gauge cable in a typical home listening situation.

Please show us ANYONE whose speaker cables are terminated by
an ideal 8 ohm load.



I was just trying to keep the problem as simple as possible. The
termination value is irrelevent since it doesn't change the
transmission line itself.

So Dick, since no one has put forth a lumped element, transmission
line model, perhaps you would like to show us one.



Mr. Stanton, your model is **** NOT because its a transmission
line or any other model, it's **** becuase of your grossly incorrect
assumptions and the fact that these assumptions simply don't exist
in actual situations.



Any real-world cable and any real-world speaker-impedance can be
*easily modeled*.


And that is yet more evidence of how far from physical reality
your "model" is. I have presented a NUMBER of lumped parameter
(not "lumped constant") models, as each and every speaker system
presents a significantluy different load.



That is correct, even if it is rather obvious.



So, let's review your assumptions behind your "model" once again:

1. You assume that people are using 100 feet of cable.
But people VERY RARELY use 100 feet of cable, it's more
typically 1/10th that distance, making the necessity of
a transmission line model even more irrelevant and
unnecessary.



10 ft is fine. One can model a transmission line for X number of ft.


2. You assume that the cable is terminated by an ideal
8 ohm load.
But NO speaker is anything approaching an ideal 8 ohm
load.


One should use an ideal load when measuring any cable's
characteristics. By making the computer model with an 8 Ohm load, the
computer's results can be compared to actual measurements.


3. You have looked at ONE example of a non-ideal load.
But, apparently, you have never incorporated such a non-
ideal load in ANY of your models.
Further, you have apparently ignored the fact that one
lumped parameter model simlpy is not representative
of the enormous variations in actual speaker loads.



Any real-world speaker-impedance can be modeled with a one-port (data)
device.


And, finally:

4. You have never once presented a single shred of physical
evidence in support of your "theory" that demonstrates
its superiority or even its very efficacy. You insist
your "theory" is right, but are unable or, more likely,
simply unwilling to do ANY of the work YOU need to do
to support it.

YOUR theory, based on your gross missapplication of transmission
line principles, your preposterous assumptions of operating
conditions, and your long-demonstrated inability to relate it to
any real-world performance issues indeed makes YOUR theory useless.



Any cable, driving any speaker, can be modeled using "one-port" and
"two-port devices". One simply measures the S-parameters (or the
Y-parameters) of the speaker-cable and then creates a two-port (cable)
device. One then measures the impedance of a speaker and creates a
one-port device. Put the two devices together, and you'll get an
accurate model of real-world performance. If you doubt this, maybe you
should go back to school. Perhaps your knowledge of circuit analysis
is a little out of date?

Bob Stanton
  #120   Report Post  
Bob-Stanton
 
Posts: n/a
Default More cable questions!

(Dick Pierce) wrote in message


BTW, have *you* EVER bothered to see if your "theory" results
in predictions that you have then compared with ACTUAL
measurements?



Thank you for again giving me credit for creating established
transmsission line theory (by your calling it "your theory").

I don't feel it is necessary for me to reprove basic transmission line
theory. The basic theory could be wrong of course, but I'll leave that
up to you to prove by measurements.



I'm curious to see what you used as a lumped
constant model. Please show us a model of 100 ft of standard (Home
Depot), 12 gage cable (terminated by an ideal 8 Ohm load).


Please show us who is using 100 ft of standard (Home Depot)
12 gauge cable in a typical home listening situation.

Please show us ANYONE whose speaker cables are terminated by
an ideal 8 ohm load.



I was just trying to keep the problem as simple as possible. The
termination value is irrelevent since it doesn't change the
transmission line itself.

So Dick, since no one has put forth a lumped element, transmission
line model, perhaps you would like to show us one.



Mr. Stanton, your model is **** NOT because its a transmission
line or any other model, it's **** becuase of your grossly incorrect
assumptions and the fact that these assumptions simply don't exist
in actual situations.



Any real-world cable and any real-world speaker-impedance can be
*easily modeled*.


And that is yet more evidence of how far from physical reality
your "model" is. I have presented a NUMBER of lumped parameter
(not "lumped constant") models, as each and every speaker system
presents a significantluy different load.



That is correct, even if it is rather obvious.



So, let's review your assumptions behind your "model" once again:

1. You assume that people are using 100 feet of cable.
But people VERY RARELY use 100 feet of cable, it's more
typically 1/10th that distance, making the necessity of
a transmission line model even more irrelevant and
unnecessary.



10 ft is fine. One can model a transmission line for X number of ft.


2. You assume that the cable is terminated by an ideal
8 ohm load.
But NO speaker is anything approaching an ideal 8 ohm
load.


One should use an ideal load when measuring any cable's
characteristics. By making the computer model with an 8 Ohm load, the
computer's results can be compared to actual measurements.


3. You have looked at ONE example of a non-ideal load.
But, apparently, you have never incorporated such a non-
ideal load in ANY of your models.
Further, you have apparently ignored the fact that one
lumped parameter model simlpy is not representative
of the enormous variations in actual speaker loads.



Any real-world speaker-impedance can be modeled with a one-port (data)
device.


And, finally:

4. You have never once presented a single shred of physical
evidence in support of your "theory" that demonstrates
its superiority or even its very efficacy. You insist
your "theory" is right, but are unable or, more likely,
simply unwilling to do ANY of the work YOU need to do
to support it.

YOUR theory, based on your gross missapplication of transmission
line principles, your preposterous assumptions of operating
conditions, and your long-demonstrated inability to relate it to
any real-world performance issues indeed makes YOUR theory useless.



Any cable, driving any speaker, can be modeled using "one-port" and
"two-port devices". One simply measures the S-parameters (or the
Y-parameters) of the speaker-cable and then creates a two-port (cable)
device. One then measures the impedance of a speaker and creates a
one-port device. Put the two devices together, and you'll get an
accurate model of real-world performance. If you doubt this, maybe you
should go back to school. Perhaps your knowledge of circuit analysis
is a little out of date?

Bob Stanton


Reply
Thread Tools
Display Modes

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
FA: Neve, Manley, TT patch cables, Eventide, Neumann, Coles, bulk cable, connectors, etc. Lowndes Pro Audio 0 March 6th 04 06:01 PM
Some serious cable measurements with interesting results. Bruno Putzeys High End Audio 78 December 19th 03 04:27 AM
cabling explained Midlant Car Audio 8 November 14th 03 04:07 AM
Digital Audio Cable Question(s) Hugh Cowan High End Audio 11 October 8th 03 07:15 PM
Quad snake cable Justin Ulysses Morse Pro Audio 8 July 3rd 03 05:10 PM


All times are GMT +1. The time now is 10:34 AM.

Powered by: vBulletin
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 AudioBanter.com.
The comments are property of their posters.
 

About Us

"It's about Audio and hi-fi"