[sdiy] Using SSM2164 in stereo

[David G. Dixon]

> > The electrons lose potential, but the same number of them flow.
> 
> But an electric potential of a conductor is defined by the amount of
> electrons in that conductor. How do electrons 'lose potential'?

NO.  The electric potential is sort of like the pressure in a water pipe.  A
water pipe can be at high pressure or at low pressure, but it contains the
same amount of water in either case.

> And if potential is lost (= electrons are lost) how are we conserving
> the current across the resistor?

Because your equation is wrong.  Potential lost =/= electrons lost (does not
equal).  Electrons are NEVER lost (except in nuclear reactions).  That's the
whole basis of Kirchhoff's Laws!

> I understand the current on the output of the current source stays the
> same. But that does not imply that the current at 'the end' stays the
> same, does it? I thought current sources were infinite charge pumps
> that terminated in ideal sinks. That just might be me, though, so
> correct me if I'm wrong.

I don't really understand the question.  However, an "ideal" current source
will supply X amount of current into any voltage.  Think of it like a dam or
weir or spill-over where the reservoir behind it can be any height up to
infinity.  A "real" current source can source X amount of current into any
potential up to a certain limit.  For example, a current source powered from
a +15V rail cannot source current into +16V.  That breaks the Second Law of
Thermodynamics.  Hence, our reservoir is only so high; as long as the water
can FALL over the dam into the sink, we can supply a constant current, but
the water will never defy gravity.  Does that make sense?

Trying to pump a current into an inappropriate potential leads to the whole
issue of "output compliance".  This is something which still causes me a bit
of grief now and then, not having had the benefit of an electrical
engineering education (and, in fact, only having graduated from the school
of "Radio Shack Electronics Learning Lab" about two years ago -- and I
actually dropped out before graduation!).

> 
> D.
> 
> On Thu, Jan 21, 2010 at 07:28, David G. Dixon <dixon at interchange.ubc.ca>
> wrote:
> >> But a true virtual current source does not need to be limited, since
> >> it is going to supply a constant (or controlled) current, right?
> >
> > Quite right.  And it would supply that current regardless of the
> resistor
> > appended to it, unless of course the resistor pushed the source voltage
> > right to the rail, in which case one would exceed the "compliance" of
> the
> > current source.  Probably another reason not to put one on.
> >
> >> Don't resistors only limit current from voltage sources?
> >
> > Yes.
> >
> >> With current
> >> sources, we have U = IR, so that we're 'limiting' voltage. I
> >> understand that the losses in current would be due to the resistor
> >> heating up which is not a very efficient way of limiting current.
> >
> > Hot resistors don't lose current.  They lose energy in the form of heat.
> > The electrons lose potential, but the same number of them flow.
> >
> >