In article , Don Pearce wrote:
How can two currents oppose each other if they run in the same direction?

That's easy. If the amplifier accepts two currents in different directions,
then two currents in the same direction oppose each other and cancel.

I think I see where our understanding differs here. You're thinking of the two
legs of the balanced circuit as two separate circuits *into the amplifier*,
whereas I'm thinking of it as one circuit going "through* the amplifier input.

My experience of balanced circuits began with the use of transformers for
output and input, where this is literally the case, i.e. a single current
flows round a single continuous electrical circuit through two transformers,
rather than two separate currents output by two separate output amplifiers
being applied to two separate inputs.

That, and the dictionary definitions of "oppose", and "opposite".

Just call me old-fashioned, but I like words to mean what they say.

Let's not even start on "phantoms". :-)

Rod.

In message , Roderick
Stewart writes
How can two currents oppose each other if they run in the same direction?


You receive a signal by looking at the DIFFERENCE in signal level
between the two wires. Let's call one wire 'A' and the other 'B'. The
desired signal current flows up the 'A' wire, through the termination
load resistor (or transformer winding), and back along the 'B' wire.
This produces the wanted signal across the load resistor.

Now an induced common-mode signal is identical on both the 'A' and 'B'
wire of the pair. If you were now to monitor the difference in signal
level between A and B you won't see anything. Thus the common-mode
pickup doesn't produce a signal at the output of the diff-amp. (Or the
secondary of the transformer as appropriate).
--
Chris Morriss

In article ,
Roderick Stewart wrote:
No - the induced signals are not in opposite directions in the two
halves of the circuit - they are in the same direction. It is the
wanted signal that runs in opposite directions in the two circuit
halves. It is a terminology thing, really - you just need to revise
to get into line with the rest of the industry.

How can two currents oppose each other if they run in the same direction?

I suppose it depends on what frame of reference you adopt when referring
to "direction", i.e. physical direction in space, or the direction of
currents in a circuit. To me it seems contrary to reason to talk of any
two things "opposing" or cancelling out if they run in the same
direction, so the only valid frame of reference for defining direction
for currents *within a wire* is direction *within the wire*, whatever
its external topology.

This really is too silly to continue.

Much easier to consider a sine wave. With a balanced line the phase on one
leg is reversed at the output, so the 'peaks' will both be positive, as it
were, and the troughs negative. Any interference will be a form of AC
signal and induced on both legs in the conventional way - ie positive peak
and negative trough. At the end of the line, the phase of one leg is
reversed so you get a conventional sine wave back for the signal. But the
interference AC signal is cancelled out as each peak is superimposed over
each trough.

HTH.

--
*Frankly, scallop, I don't give a clam

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

It is a differential signal in a balanced (floating) cct when measured at
Sending end between legs a and b (or vice versa). Call this value 1.
Assume a low-Z source between a and b.

At the Receiving end, the nodes become a' and b' due to insertion (long
twisted pair cable or eventually HF slope due to capacitance) loss.
Assume bridging Z mode across a' and b'.

So the waveform of all parameters in one leg are the inverse of the
waveform of all parameters in the other leg, which amount differentially to
the signal of
1 (and 1' at its far end). But they sum additively by cancelling to 0 and 0'
(the same)
which is meaningless and ignored at the far end. Any possible, environmental
electro-magnetic
interference picked up by the two cores in the cable (being common mode) is
therefore
rejected. And an overall braided screen at Ground potential virtually
eliminates the entry of such longitudinal interference.

"Roderick Stewart" wrote in message
om...
In article , Don Pearce wrote:
Balanced wiring reduces interference from extraneous electromagnetic
fields
because these induce equal and opposite currents in the two signal
wires
(which
are close and parallel), so that the currents tend to cancel out.

Rod.

If they are equal and opposite in a balanced system, they will add.
The
trick is that they are equally induced in both legs in the same
phase -
hence no differential signal across the two lines. This is why it's
called "
common mode ".

I think we're having trouble with words here. In my dictionary,
"opposite"
means that they oppose each other. As the signal wires are close and
parallel,
interference currents will be induced in the same physical direction,
which
will be in opposite electrical directions in the two halves of the
circuit, so
they will oppose. I did say "equal and opposite currents".

Rod.

No - the induced signals are not in opposite directions in the two halves
of the circuit - they are in the same direction. It is the wanted signal
that runs in opposite directions in the two circuit halves. It is a
terminology thing, really - you just need to revise to get into line with
the rest of the industry.

How can two currents oppose each other if they run in the same direction?

I suppose it depends on what frame of reference you adopt when referring
to
"direction", i.e. physical direction in space, or the direction of
currents in a
circuit. To me it seems contrary to reason to talk of any two things
"opposing" or
cancelling out if they run in the same direction, so the only valid frame
of
reference for defining direction for currents *within a wire* is direction
*within
the wire*, whatever its external topology.

This really is too silly to continue.

The interference still cancels, though.

At least we agree on that.

Rod.

In article , Chris Morriss wrote:
How can two currents oppose each other if they run in the same direction?


You receive a signal by looking at the DIFFERENCE in signal level
between the two wires. Let's call one wire 'A' and the other 'B'. The
desired signal current flows up the 'A' wire, through the termination
load resistor (or transformer winding), and back along the 'B' wire.
This produces the wanted signal across the load resistor.

You can describe it as the difference between the two currents flowing *into
the amplifier* via the two signal wires. However, if the input uses a
transformer (which was the original way of implementing balanced signal
circuits), then you are looking at the *same current* flowing in two
different parts of the *same circuit*. The current will flow into the
amplifier through one of the input terminals into one end of the input
transformer's primary winding, and out of the other end of the transformer
winding and the other terminal. It is the same current. The circuit in which
it is flowing is a closed loop comprising the secondary winding of the output
transformer of the source, the two wires, and the input transformer of the
amplifier. There is one current and one loop of wire, and the current flows
round this circuit in one direction.

Apparently you are considering the two signal wires of the same balanced
circuit as two separate circuits into the amplifier, (using ground as a
return I suppose, though how the working is explained if there is no ground
is unclear) and I am told this is now the standard way of regarding these
things. I guess it just depends on how you look at things.

Rod.

In article , Roderick
Stewart wrote:
In article , Don Pearce


No - the induced signals are not in opposite directions in the two
halves of the circuit - they are in the same direction.

How can two currents oppose each other if they run in the same direction?

Because the receiver operates by sensing the *difference* between the
current levels at its two terminals. (Thinks of 'difference' as
'subtract'.) :-)

I suppose it depends on what frame of reference you adopt when referring
to "direction", i.e. physical direction in space, or the direction of
currents in a circuit. To me it seems contrary to reason to talk of any
two things "opposing" or cancelling out if they run in the same
direction, so the only valid frame of reference for defining direction
for currents *within a wire* is direction *within the wire*, whatever
its external topology.

Well, IIRC the standard is as Don describes. Indeed, to be able to define
the unbalanced resultant you have to define which input current terminals
is the '+' one and which the '-' so as to get the differential 'the right
way around' of you are concerned to obtain the intended polarity. This
implies the standard Don is describing.

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

In article , Roderick
Stewart wrote:
In article , Don Pearce
wrote:
How can two currents oppose each other if they run in the same
direction?

That's easy. If the amplifier accepts two currents in different
directions, then two currents in the same direction oppose each other
and cancel.

I think I see where our understanding differs here. You're thinking of
the two legs of the balanced circuit as two separate circuits *into the
amplifier*, whereas I'm thinking of it as one circuit going "through*
the amplifier input.

My experience of balanced circuits began with the use of transformers
for output and input, where this is literally the case, i.e. a single
current flows round a single continuous electrical circuit through two
transformers, rather than two separate currents output by two separate
output amplifiers being applied to two separate inputs.

in such cases the H-field may simply fail to induce any.

However as indicated elsewhere, Don and others are describing the standard
approach - which allows us to also define the sign of the result recovered.

That, and the dictionary definitions of "oppose", and "opposite".

Just call me old-fashioned, but I like words to mean what they say.

So do I. That involves us all agreeing on a standard definition, and then
using it. :-)

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

All those negatives !
Did anyone suggest that you might actually think about listening to them ?
Russ Andrews does sale or return. Buy some on e-bay then re-sell if it
makes no difference

Regards Richard
New Ash Green Hi-Fi Club
"deja" wrote in message
oups.com...
I've been reading up on the subject of main cables, conditioners
etc...I'm rather sceptical of the improvement such expensive things
will add to the sound of my hifi but have a couple of questions if
anyone would care to enlighten me
- are mains conditioner blocks supposed to be used with the basic mains
leads supplied with everyday hifi, or with the custom powerkord type
leads? eg take Russ Andrews the PowerLink, would you add in Powerkords
or not?
- if considering a mains block extension lead that costs £££, would
I not be better just installing a couple more double sockets in the
wall of my lounge and plugging in the hifi directly? The money I'd save
could then be put towards better quality mains leads for my CD player &
amps (assuming one is convinced of the need to do that!)

And just how much improvement do folks get with these products? Mains
leads are around £50-£200 and blocks are anything approaching £1k!
(looking at the Russ Andrews products)

For example (to power CD player + 2 mono blocks)

The Silencer Block (with generic leads) = £150
or
PowerLink + 3 PowerKords - £230
or
DIY extra sockets (free), 2 x Powerkords and a reference PowerKord for
the CD player = £225
or the option I rather favour:
Maplin Multiway Mains Plug VD02C £6.99!!!

Cheers,
David

In article , Dave Plowman (News) wrote:
No - the induced signals are not in opposite directions in the two
halves of the circuit - they are in the same direction.[...]
How can two currents oppose each other if they run in the same direction?

I suppose it depends on what frame of reference you adopt when referring
to "direction", i.e. physical direction in space, or the direction of
currents in a circuit.[...]
Much easier to consider a sine wave.

Thanks for all the explanations of how signals can either add or cancel out,
but this isn't the source of the misunderstanding. It's about currents in
circuits and the question of how we define the direction of flow.

I've known about balanced signal circuits, and had practical experience of
their use, for the best part of forty years. The first ones I encountered used
audio transformers, and maybe some of them still do, though it is possible to
emulate the behaviour of a balancing transformer using active semiconductor
circuitry.

A transformer-balanced audio circuit will work with only the two signal wires,
and though it may benefit from screening in areas of strong interference, it
does not need to use the screen as an earth return. (In fact, other problems
can sometimes be caused if it does, and it is not uncommon to connect the
screen only at one end).

If source and load both use transformers and the only connections between them
are the two signal wires, then that is the circuit, and the current flows
around it. ONE closed circuit loop with ONE current, flowing in the SAME
direction everywhere around it, because we are of course using the wire
itself, including the two transformer windings, as the frame of reference with
which to define direction of current flow. It would be difficult to explain
the behaviour of currents in circuits in a manner that would be understood by
Mr Kirchoff if we used any other convention.

Maybe it's a consequence of being taught about balanced circuits only in terms
of their modern practical implementaion that makes some people want to analyse
them as if each signal wire were a separate unbalanced circuit, each with its
own definition of current direction, as if they were completley unrelated
until the two separate signals arrived at the subtracting circuits in the
input of the load equipment. However, this leaves a question. If each signal
leg is a separate circuit, how does it work with only one wire and no return?
Easy to explain if each signal wire is the return wire for the other, and no
problem seeing where the circuit loop is, but if they are completely separate,
how do two separate voltages appear across two separate pairs of input
terminals if there is no ground or return connection from the ground of the
source equipment? Where are the two circuit loops around which the two
currents flow?

Rod.

On Fri, 26 Aug 2005 12:01:42 +0100, Roderick Stewart wrote:

Thanks for all the explanations of how signals can either add or cancel out,
but this isn't the source of the misunderstanding. It's about currents in
circuits and the question of how we define the direction of flow.

Just think of direction as a physical arrow in space. Is it flowing towards
the window - or the settee? That is what we mean by "in the same direction"
when talking about induced currents in a balanced pair.

d