In article , Don Pearce
writes
On Sun, 18 Jun 2006 14:45:02 +0100, Glenn Richards
wrote:

Don Pearce wrote:

OK Glenn, maybe you can't be bothered (yeah, right) but I can. I've
done the work for you and simulated the two effects - biwire, or
parallel wire, joined at the far end - in PSpice. The results
vindicate my position completely.

No they don't.

You've only placed a "lossy" cable on one leg of the speaker cable. So
you've got effectively a 5 metre run of cable from the positive speaker
output, then a short (as close to zero ohms as possible) run back to the
ground point on the amplifier.

Also, most crossovers I've seen often only have a series inductor (for
LF) or series capacitor (for HF), and don't bother strapping L or C
across the load. Admittedly some do, but I just took the plate off the
back of one of my Mission 760iSE speakers, there's a single inductor in
series with the LF driver and a single cap in series with the HF driver.
Clearly a case of KISS (Keep It Simple, Stupid).

Of course, Spice could be totally invalid along with a couple of
hundred years of theory, I suppose...

So how do you explain the fact that when the speakers are bi-wired they
sound better (clearer and sharper treble, more detail etc)? And when
single-wired they sound muddy by comparison?

Hint: the standard uk.rec.audio cop-out of "it's all in your mind" is
not a valid response. Of course it's all "in your mind", your ear only
gathers sound and turns it into electrical signals. It's your mind (or
more technically your brain) that interprets those electrical signals as
noises, whether pleasant or unpleasant. Just in case you feel like
getting pedantic.

Remember that computers do what you tell them to do, not what you want
them to do. So if your model isn't tying up with the observed effects
then your model is obviously wrong.

I'll give you a B+ for effort, E for accuracy. Your computer model does
not match the observed effects. Now, take the effort you're about to
expend on trying to prove me wrong and channel it into finding out why
the observable and repeatable effects occur.

Head teacher's comment: Plenty of enthusiasm, needs to channel self
better. D-.

Good grief - there really is no end to it.

OK, I've done all I can - anybody else feel like trying?

d



'ere guv, how can I bi- or tri wire me ESL63's cos if the squirrelsounds is
right.. I want some of what he's on)
--
Tony Sayer

FWIW I tend to prefer the Maplin cables as they give low series resistance
for just a few quid per cable. The above URL uses them for example, so
people can decide for themselves what might suit for the lengths they need.

Slainte,




Maplin guv?, pon me life their pricy..

Someone left some cables from the 400 kV re-wire their doing round this way
and the insulators..you've never seen anything like 'em.

Missus has said either the Pylons go or I do.. so she's packin 'er bags))
Jim


--
Tony Sayer

On Sun, 18 Jun 2006 15:57:05 +0100, tony sayer
wrote:

'ere guv, how can I bi- or tri wire me ESL63's cos if the squirrelsounds is
right.. I want some of what he's on)
--
Tony Sayer

Troll juice, maybe?

d

--
Pearce Consulting
http://www.pearce.uk.com

On Sun, 18 Jun 2006 16:00:00 +0100, tony sayer
wrote:

FWIW I tend to prefer the Maplin cables as they give low series resistance
for just a few quid per cable. The above URL uses them for example, so
people can decide for themselves what might suit for the lengths they need.

Slainte,




Maplin guv?, pon me life their pricy..

Someone left some cables from the 400 kV re-wire their doing round this way
and the insulators..you've never seen anything like 'em.

Missus has said either the Pylons go or I do.. so she's packin 'er bags))
Jim


Do what I do - say to the bloke "I'll give you two quid for what's
left on that roll". Anything less than half a roll, and they will
generally go for it rather than be bothered measuring off what's left.

d

--
Pearce Consulting
http://www.pearce.uk.com

On Sun, 18 Jun 2006 13:36:03 +0100, Jim Lesurf
wrote:

In article , Glenn
Richards wrote:
Serge Auckland wrote:

No it isn't! Your experiment is being done with dc. Repeat the
experiment with 50Hz to one bulb and 10kHz to the second bulb, both
bulbs being fed through a single amplifier and fed through a simple
capacitor or inductor "crossover". You will now see that the bulbs
don't change their brightness.

Makes no difference, AC or DC.

Do the experiment Serge described and you will find that the results show
he is correct in what he says. Then as Don has suggested, investigate
diplexing.

Bear in mind that what he describes is the kind of technique routinely
employed and studied by electronic engineers. Indeed, I'd suspect that
more than one undergrad lab may have what Serge describes as an
experiment to show this to students. It is the basis of frequency
division multiplexing in transmission line systems. :-)


Series resistance is still the same, regardless of line frequency.


Actually, not necessarily so. It is quite possible for a cable to
have a resistance that varies with frequency. :-) However even if we
ignore this, what Serge says is correct.


That experiment was to illustrate a concept, not specifics. The
principle is exactly the same.

Only for the specific case you gave. The snag is that loudspeakers aren't
simply resistors.

Slainte,

Jim

I think that there is a far more fundamental problem at work here,
Jim. Glenn does not appear to understand that what we have is a simple
voltage divider, comprising the cable and the speaker impedance. A 1
ohm cable, combined with an 8 ohm speaker will result in a loss of
about a dB at all frequencies, and it doesn't matter what signals are
present.

If the effect he is describing were to be real, then picture a 1kHz
signal in combination with a 10Hz signal. The current due to the 10Hz
would be changing from a maximum to zero 20 times per second, so the
loss at 1kHz would be changing 20 times per second. The 1kHz would
actually be amplitude modulated by the 10Hz, rather than simply
superposed, which we know to be the case. His misunderstanding of the
physics really is happening at a rather fundamental level.

d

--
Pearce Consulting
http://www.pearce.uk.com

In article , Glenn
Richards wrote:
Don Pearce wrote:


Also, most crossovers I've seen often only have a series inductor (for
LF) or series capacitor (for HF), and don't bother strapping L or C
across the load.

Well, most of the speaker crossovers I've seen were far more complex than
that...

Admittedly some do,

I wonder if anyone has any reliable figures for just what fraction of the
models of speaker on sale are as simple as you imply? It would seem that my
experience differs from yours.

Of course, Spice could be totally invalid along with a couple of
hundred years of theory, I suppose...

So how do you explain the fact that when the speakers are bi-wired they
sound better (clearer and sharper treble, more detail etc)? And when
single-wired they sound muddy by comparison?

Please give us details of some reliable reports of tests that were carried
out in a way that allows their results to be assessed as evidence and
which support what you claim. :-)

Until such time, your use of the term "fact" above may have to be regarded
with some caution... It may mean "error", or "for some other reason that
didn't occur to the listener at the time".

Hint: the standard uk.rec.audio cop-out of "it's all in your mind" is
not a valid response.

Alternative hint: Simply reporting along the lines of, "I listened to A,
then listened to B, and decided I could hear a difference" isn't reliable
evidence for what you assert. :-)


Of course it's all "in your mind", your ear only gathers sound and turns
it into electrical signals. It's your mind (or more technically your
brain) that interprets those electrical signals as noises, whether
pleasant or unpleasant. Just in case you feel like getting pedantic.

Also for the 'pedants': You may also have to bear in mind a variety of
uncorrelated variables, some of which have already been mentioned in this
thread. These mean that unless a listening test is carried out in an
appropriate way that takes these into account, we may simply be unable to
tell if a claim that an 'audible difference' is to due the 'reason'
asserted has any worth. Alas, a result which could mean anything, may mean
nothing at all - regardless of being called a 'fact'.


Remember that computers do what you tell them to do, not what you want
them to do. So if your model isn't tying up with the observed effects
then your model is obviously wrong.

This assumes that the claims about "observed effects" are based on a test
method, etc, that was relevant and reliable for the purpose. It is, alas,
quite easy to design poor tests which then return misleading 'results', or
ones the people involved interpret incorrectly. The classic example being
the kind of "magazine reviewer's test" I described above. Unfortunately,
people make assertions, but may not give the relevant details of how they
arrived at their conclusions. In such cases the 'facts' may have no
assessable meaning or value and become indistinguishable from an 'opinion'
presented as a 'fact'.

I've seen many models/analyses/theories of things which were utter twaddle.
However I've also seen many experimental/test/measurement arrangements and
protocols that were simply not fit for purpose and so returned nonsensical
or useless results. e.g. Lost count years ago of how many flawed
experiments I've seen reported that 'proved' faster-than-light propagation
in free space. Would be nice if it were true, but examination of the
experimental proceedure dissapoints... Sometimes with subtle flaws,
sometimes laughable ones. :-)

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html

Don Pearce wrote:
On Sun, 18 Jun 2006 13:36:03 +0100, Jim Lesurf
wrote:

In article , Glenn
Richards wrote:
Serge Auckland wrote:
No it isn't! Your experiment is being done with dc. Repeat the
experiment with 50Hz to one bulb and 10kHz to the second bulb, both
bulbs being fed through a single amplifier and fed through a simple
capacitor or inductor "crossover". You will now see that the bulbs
don't change their brightness.
Makes no difference, AC or DC.
Do the experiment Serge described and you will find that the results show
he is correct in what he says. Then as Don has suggested, investigate
diplexing.

Bear in mind that what he describes is the kind of technique routinely
employed and studied by electronic engineers. Indeed, I'd suspect that
more than one undergrad lab may have what Serge describes as an
experiment to show this to students. It is the basis of frequency
division multiplexing in transmission line systems. :-)


Series resistance is still the same, regardless of line frequency.

Actually, not necessarily so. It is quite possible for a cable to
have a resistance that varies with frequency. :-) However even if we
ignore this, what Serge says is correct.


That experiment was to illustrate a concept, not specifics. The
principle is exactly the same.
Only for the specific case you gave. The snag is that loudspeakers aren't
simply resistors.

Slainte,

Jim

I think that there is a far more fundamental problem at work here,
Jim. Glenn does not appear to understand that what we have is a simple
voltage divider, comprising the cable and the speaker impedance. A 1
ohm cable, combined with an 8 ohm speaker will result in a loss of
about a dB at all frequencies, and it doesn't matter what signals are
present.

If the effect he is describing were to be real, then picture a 1kHz
signal in combination with a 10Hz signal. The current due to the 10Hz
would be changing from a maximum to zero 20 times per second, so the
loss at 1kHz would be changing 20 times per second. The 1kHz would
actually be amplitude modulated by the 10Hz, rather than simply
superposed, which we know to be the case. His misunderstanding of the
physics really is happening at a rather fundamental level.

d

Absolutely! Glenn's theory would only be valid if cables were
non-linear, but we know that cables are extremely linear, pretty much to
the limits of measurement, so no modulation can take place.

S.

tony sayer wrote:

'ere guv, how can I bi- or tri wire me ESL63's cos if the squirrelsounds is
right.. I want some of what he's on)

You would be better off multi-amping them.
Get rid of that poxy HT delay line and replace it with a digital one
driving a separate amp and transformer for the centre and each annulus
and bob's your uncle.

--
Eiron

No good deed ever goes unpunished.

Don Pearce wrote:

Good grief - there really is no end to it.
OK, I've done all I can - anybody else feel like trying?

Come up with a computer model that matches the observed effects and I
might start taking you seriously.

--
Glenn Richards Tel: (01453) 845735
Squirrel Solutions http://www.squirrelsolutions.co.uk/

IT consultancy, hardware and software support, broadband installation

Glenn Richards wrote:

Don Pearce wrote:

OK Glenn, maybe you can't be bothered (yeah, right) but I can. I've
done the work for you and simulated the two effects - biwire, or
parallel wire, joined at the far end - in PSpice. The results
vindicate my position completely.

No they don't.

You've only placed a "lossy" cable on one leg of the speaker cable. So
you've got effectively a 5 metre run of cable from the positive speaker
output, then a short (as close to zero ohms as possible) run back to the
ground point on the amplifier.

God almighty ! Don't you know that's the same as modelling with the resistance
equally distributed ?

Graham