Russ Andrews and Ben Duncan :-)
 

Keith G wrote:

"David Looser" wrote in message
...
"Keith G" wrote in message
...


The single strand is nothing to do with fuses (what's the whole wire
then - a *higher rated fuse*?) - it was merely to illustrate the
point that I believe 'conventional wisdom' actually promotes and
encourages 'snake oil' (referenced in the OP) by doing something
rather unconventional....


I wondered what on earth you thought your "single strand of wire"
demonstrated; and your "explanation" above does little to clarify the
point. Quite *how* you think that " 'conventional wisdom' actually
promotes and encourages 'snake oil" is a mystery that is unlikely ever
to be solved.


OK, it goes like this, but first I had to substitute a speaker cable for
a mains lead as I couldn't easily experiment with mains leads (same
principles):

Current 'conventional wisdom' says a speaker (or any other) cable should
be able to carry the required current (and be long enough) and it
generally accepted that summat like a 79 strand copper wire of a certain
gauge is quite good enough. At any rate, most speaker cables that are
available today online and in the shops are around this mark; Snake Oil
Merchants build on this by adding in exotic and expensive materials and
inevitably increasing cable cross-sectional sizes.

I reduced an ordinary 'conventional' speaker cable to a single strand to
see what happened.

Nothing happened - as I suspected, it makes no difference to the sound
that I can detect, other than mess the image about - which I suspect
could be easily sorted by reducing the other speaker wire to a single
strand also. It is still working fine and sound levels have been pushed
up to 'uncomfortable' a number of times.

The point is/was/would have been that, although a single strand of wire
isn't practicable (unless strong enough to survive), it makes a
'conventional' wire so *overkill* that an even more OTT 'snake oil
cable' wouldn't be considered for a moment by enough people to make the
exercise worthwhile or to grow the size RA seems. Or, to put it another
way, snake oil merchants get to sell more today because 'normal' speaker
cables are so much bigger than they used to be - even when using the
'powerhouse' amps of the 70s!

I think you've missed something in your analysis. You need to consider the cable length.
Conventional wisdom is that your speaker wires should be less than 5% of the impedance of the speakers,
and I don't see anything wrong with that. So I would be perfectly happy using a pair of 0.2mm diameter
wires,
but only up to a length of about 2 ft for normal speakers. (I haven't calculated the exact length.)
And for the ESLs, the cable length would be rather less than half the width of the speaker.
So in many cases, 79 strand cable is overkill, but not by as much as you think.

--
Eiron.

Russ Andrews and Ben Duncan :-)
 

In article ,
Keith G wrote:
Current 'conventional wisdom' says a speaker (or any other) cable should
be able to carry the required current

Maybe your 'conventional wisdom' But it's wrong.

--
*I'm not your type. I'm not inflatable.

Dave Plowman London SW
To e-mail, change noise into sound.

Russ Andrews and Ben Duncan :-)
 

In article ,
Keith G wrote:
Kitty is forever trying to re-invent the wheel. Dunno why he thought
adding a very small series resistance to a speaker circuit would make
things sound different. But judging by the jump leads melting story, he
needs to re-invent Ohms law too...


Oh dear, **** reaches into his 'smartarse things to say on handy
wipe-clean cards box' and is waving 'Ohms Law' about!

If he had any *real outdoor knowledge* of the situation I described (and
have witnessed on a number of occasions) he would know that jump leads
for commercial vehilcles are long, heavy (very) and expensive (very) -
nobody I ever knew had more than the one set to choose from.

Twerpy Know Nothing For Real probably thinks the current kills the leads?

Wrong! It's when the guy on the starter button can't see/hear the guy
watching the leads - I'll leave it to you people to fill in the blank
parameter....

Pray tell.

--
*7up is good for you, signed snow white*

Dave Plowman London SW
To e-mail, change noise into sound.

Russ Andrews and Ben Duncan :-)
 

In article 4a419cd7.881292296@localhost, Don Pearce
wrote:
On Tue, 23 Jun 2009 09:45:32 +0100, Jim Lesurf
wrote:


If I were trying to analyse the response of a random, unknown piece of
cable, the first thing I would do is find out its impedance by a simple
jX test on an eighth wave piece hanging off a network analyser. The
fixture would then be designed to transform the 50 ohms to that
impedance (at both ends).

The method I tend to use is to measure the input impedance vs frequency of
a measured length. But I do this three times, changing the output load -
short, then open, then a standard value like 50 Ohms if the analyser is
feeding in using 50 Ohms. If the measurements are vector the results should
normally allow me to work out the complex impedance as a function of
frequency. ...and of course I'd tend to measure the output end as well as
the input if both are accessible.

Of course, I'd also check the analyser with standard/open/short loading and
use these to calibrate. Generally checking the loads I used above. All very
fiddly, but necessary IMHO if you want the results to be checkable. May be
my NPL experiences that influence me with questions like, "So how do we
know this is 50 Ohms, then?" ;-


Only then could I make a measurement free from the stupid VSWR leaps.

Here the VSWR mismatch seems to be 'the measurement'. :-)


As for calibration, I would want to de-embed the fixtures by making up
open, short and load calibration pieces that plugged in where the actual
cable went. Tricky, I know, but for results up to a few hundred megs I
would be reasonably happy with the results.

Ditto, as above.

For the "through" calibration I would make sure the mains lead had male
and female versions of the same connector both ends, and simply plug the
fixtures together - that would make for an insertable calibration, which
is always the safest.

Yes, plus the termination load checks.

FWIW I'd probably also check with cable runs of the same type but different
length to see if the results where consistent.

In fact, I did use much the above methods for the measurements I did on
speaker cables that HFN published last month. (Alas, they did trunkate the
article, but I can put up a fuller version on the web in due course!)

Thinking about this, I can find no earthly reason for that BNC curve to
be presented. Do you have any idea what purpose it serves?

Not sure at this point, so have to make the familiar comment that "I need
to re-read the papers". I do need to do that a few times and think to
ensure I'm not missing things as yet.

But my current impression is that it is meant to be some sort of
'demonstration that the system works OK'. All it indicates to me, though,
is that the source and load had impedances similar to the BNC cable, and
that the BNC cable probably had modest losses. Not enough to determine if
any impedance mismatch effects with other cables were being accurately
measured.

However one of the things that intrigues me is the difference in overall
loss proportional to frequency between the BNC and all the others. Does
look like a measurement system effect to me at present.

However I need to re-read the detailed papers a few times and think
about them. My feeling, though is that all the results in the
initial pdf show is that the cables have different Zc values. The
relevant measurements seem to have been done with no mains supply or
loading PSU. Just with what seem to be claimed to be 50Ohm
terminations.

My thoughts exactly. The source impedance should be whatever you get
from a few miles of twin coupled to a transformer, a few thousand
light bulbs, a bunch of motors, many TVs and loads of fluorescents.
That shouldn't be too hard to model ;-)

...but possible to measure. :-)


Perhaps they could go to every potential customers house and measure it,
then produce a curve of probable improvement...

That might be a service worth selling to people. "Give us dosh to connect
to your mains and tell you that you should buy more expensive cables." :-)


I can understand the wish to make measurements as easy and simple as
possible. Particularly when time is money. However the snag is to avoid
making them so 'simple' that they cease to be relevant to the
real-world situation which you want to use the 'results' to describe.
:-)


Who was it who said that in science everything should be described as
simply as possible - but no simpler.

Probably the other guy with wild white hair. 8-]

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

Russ Andrews and Ben Duncan :-)
 

"Laurence Payne" wrote in message
...
On Tue, 23 Jun 2009 12:58:46 +0100, "David Looser"
wrote:

"Conventional Wisdom" says that a speaker cable needs to be of largish
cross-sectional area in order to bring the resistance down to an
acceptable
value. This is what determines the CSA that "Conventional Wisdom"
recommends, not the current-carrying capacity of the cable. This is based
on engineering.

Can you remind us the difference between "low resistance" and
"current-carrying capability" please?

Certainly. Current-carrying capacity is based mainly on temperature rise. If
the temperature rises too much then a variety of undesired effects come into
play, ranging from degradation of the plastic of the insulation, via fire
risk (when cables are enclosed or close to combustible substances) to, at
the extreme, melting of the metal of the wire. You may notice that the
current carrying capacity of insulated wires is specified differently
according to how the wire is to be used, e.g., in free air, bundled with
other wires, enclosed in conduit etc.

In domestic audio use the temperature rise of a speaker cable will be
negligible unless of a resistance which is sufficiently high to: a/ waste a
lot of the audio power, and b/ seriously degrade the frequency response of
the system.

David.

Russ Andrews and Ben Duncan :-)
 

"Eiron" wrote


I think you've missed something in your analysis. You need to consider the
cable length.
Conventional wisdom is that your speaker wires should be less than 5% of
the impedance of the speakers,
and I don't see anything wrong with that. So I would be perfectly happy
using a pair of 0.2mm diameter wires,
but only up to a length of about 2 ft for normal speakers. (I haven't
calculated the exact length.)
And for the ESLs, the cable length would be rather less than half the
width of the speaker.
So in many cases, 79 strand cable is overkill, but not by as much as you
think.


It was only a quick 'I wonder what if' and was never meant to be anything
more than that - I have no means (or interest) to *measure* anything here.
If it helps, substitute 'current common practice' for 'conventional wisdom'
where it will do the most good.

For speaker cables I have always been happy to use mains flex (extension
lead) and have recently dropped down even from that to '13 strand Fig 8 twin
speaker flex' from B&Q and find that perfectly satisfactory. Can't remember
what it costs for a 10m pack but it's a hell of a lot less than 'speaker
cable' from the local hifi shop!!

From now on I would be equally happy with 'lamp flex'..!!

Russ Andrews and Ben Duncan :-)
 

"David Looser" wrote


Certainly. Current-carrying capacity is based mainly on temperature rise.
If the temperature rises too much then a variety of undesired effects come
into play, ranging from degradation of the plastic of the insulation, via
fire risk (when cables are enclosed or close to combustible substances)
to, at the extreme, melting of the metal of the wire. You may notice that
the current carrying capacity of insulated wires is specified differently
according to how the wire is to be used, e.g., in free air, bundled with
other wires, enclosed in conduit etc.


Nicely put - now tell that to Pucci who can't work out why:

Wrong! It's when the guy on the starter button can't see/hear the guy
watching the leads - I'll leave it to you people to fill in the blank
parameter....

Russ Andrews and Ben Duncan :-)
 

On Tue, 23 Jun 2009 13:39:30 +0100, "Keith G"
wrote:


"Eiron" wrote


I think you've missed something in your analysis. You need to consider the
cable length.
Conventional wisdom is that your speaker wires should be less than 5% of
the impedance of the speakers,
and I don't see anything wrong with that. So I would be perfectly happy
using a pair of 0.2mm diameter wires,
but only up to a length of about 2 ft for normal speakers. (I haven't
calculated the exact length.)
And for the ESLs, the cable length would be rather less than half the
width of the speaker.
So in many cases, 79 strand cable is overkill, but not by as much as you
think.


It was only a quick 'I wonder what if' and was never meant to be anything
more than that - I have no means (or interest) to *measure* anything here.
If it helps, substitute 'current common practice' for 'conventional wisdom'
where it will do the most good.

For speaker cables I have always been happy to use mains flex (extension
lead) and have recently dropped down even from that to '13 strand Fig 8 twin
speaker flex' from B&Q and find that perfectly satisfactory. Can't remember
what it costs for a 10m pack but it's a hell of a lot less than 'speaker
cable' from the local hifi shop!!

From now on I would be equally happy with 'lamp flex'..!!



Look, here is how it goes.

For a speaker cable, there is a single, simple "figure of merit" that
tells you if it is ok, and that is the resistance. Two things
influence that, length and thickness - and of these two the thickness
is much the more important. If you double the length, you double the
resistance (a bad thing), but if you double the thickness, you divide
the resistance by four (because it depends on the area, not the
thickness). So thick is good. But thick enough is good enough.
Provided you end up with a resistance (for the full round trip) under
about half an ohm, that will be fine.

Now, what happens if the resistance is too high? Mostly nothing. You
set the volume control to where you like the sound, and that is that.
The fact that the volume control is a bit more clockwise than it might
have been doesn't matter - unless you are trying to be loud, that is.
Your amp will run into clipping and distortion at a much lower volume
if your speaker leads are too thin (and too long).

So in the house, with leads just a few feet long, lamp flex is fine.
If you want them out in the garden for the barbie, use something
thicker.

Finally your little test - meaningless, I'm afraid. That odd inch of
single strand added essentially no resistance to the whole cable
because it was so short.

d

Dear Jim...
 

"Don Pearce"


** This vile, ASD ****ed pommy **** makes Jim look like a saint.

Look up " ASD "

Look up the real meaning of "****".

****ed = ruined or made worthless.

Don't just react to my carefully chosen words emotionally.

Cos Jesus Christ I mean them.





..... Phil

** Dear Jim,
 

"Jim Lesurf"

** Dear Jim,

you are one of the most totally ****ed in the head, retarded autistic
pukes alive in the UK - and that is saying something - cos the whole
stinking **** hole is just crawling with them.

People like YOU constitute a serious public menace, for the sole reason
that you are so damn ****ing STUPID.

Not a damn thing you can do about THAT - of course.

But beware, there is plenty OTHERS can do about YOU !!!

The terminally stupid in society cause FAR FAR more trouble than all the
nutters and psychos put together.

Please please please, for the benfit of humanity

get very ill - very soon and ****ing die.

So well can all stop worrying about what

STUPID ****ING CRAP

you will decide self publish next.




...... Phil