"Eeyore" wrote in message
...


Arfa Daily wrote:

"Eeyore" wrote in message
Arfa Daily wrote:
"Eeyore" wrote in message
Arfa Daily wrote:
"Eeyore" wrote
Arfa Daily wrote:

I wouldn't dispute for one minute that changes to the interstage
coupling
electrolytics in a semiconductor amp, would result in any kind
of
improvement that anyone could hear

Are you saying that electrolytics are or aren't audible ? I can't
figure that one.

Think what you like pal, I've grown utterly tired now of this
thread,
and your tedious replies ...

Show me a capacitor ( other than those I've mentioned ) that
distorts
will you !

You sound very gullible to me.

Please don't descend into trying to trade insults. As I said, believe
what you like.

It's not a question of believing. That's what the audiophools do.

I've actually gone to the trouble of measuring ! With Audio Precision
test
gear that has a generator residual THD of 0.0007% ( -103dB ).

Not a single film capacitor of any type I've checked has any distortion
whatever. Nor would I expect any from first principles.

What's your experience in the matter ?


Did I actually say that any capacitor per se had distortion ?

You suggested so.


Only in your head ...


The problem is
that you have such a bee in your bonnet about this relatively minor
point,
that you are either not listening to what I am saying or deliberately
misconstruing it in order to continue with this senseless lambasting.

Clearly there must be differences between different constructions of caps
from different manufacturers, otherwise, there wouldn't be different
types
with the same value and voltage rating, and different price tags.

What's that got to do with it ?

If you *really can't see, then you must be dumber than you appear.


The only real thing of concern is the dielectric and whether the plates
are
metallised or foil ( for high current applications ) with a resultant
effect on
ESR.

Some winding techniques reduce parasitic inductance but this isn't an
issue at
audio frequencies.


You
wouldn't for instance use a cheapo cap intended for decoupling purposes,
as
the timing element in an oscillator.

Define 'cheapo cap'. You might get a shock about the prices I pay for
parts for
products in volume manufacture btw !


Oh please ! Are you going out of your way to be obtuse ? Do you *really*
think that the word 'cheapo' refers just to price ? Where have you been all
of your life ? Head down over a precision signal generator with point 36
noughts seven THD perhaps ?


If you wanted it to last, you wouldn't
use a bog standard cheapo electrolytic in a switch mode power supply. But
there's my point.

Define 'cheapo cap again ! SMPSs require low ESR caps in the output
filters -
it's nothing to do with being cheap or not. That's what's *needed*.


But it is !! A good grade high temperature low ESR electrolytic costs more
than a cheapo ( see above ) bog standard electrolytic. Nobody, least of all
me, would dispute that a low ESR cap is NEEDED, but trust me, I repair
stacks of these things every week - it is not what you actually FIND in
them. If manufacturers can buy good quality low ESR caps for the same price
as the poor quality crap that they do fit, why wouldn't they ? How do the
manufacturers of low quality electrolytics stay in business when according
to you, everyone can buy a superior product for the same money ?

Manufacturers do - not because the designer wanted them
to, but because cost dictated it. This results in inferior performance of
the power supply over what the designer initially intended, and ultimate
failure.

Inferior ? If parts with the correct specs are used the only concern
between
makes in this application is usually lifetime.


Of the correct NOMINAL specs. Anyway, if all parts are as good as each
other, irrespective of price, as you seem to be implying, why is lifetime
suddenly an issue ?


You may also be getting confused over what I am treating as 'distortion'.
I
am including anything that not only makes a piece of equipment deviate
from
the ideal, but also from what the designer intended. If *any* substituted
component on cost grounds causes any deviation from what the designer
originally measured on his development model with the components he used
and
specified, then this under my definition, is distortion. It need not
necessarily result in mishaping of the output waveform in any way that
would
normally be considered as distortion, which may or may not be audible,
but
could include unwanted phase shifts or gain non-linearities.

Phase shifts are related to the *value* of a capacitor. All have a
tolerance on
them regardless of price.

They do, but some have a tighter tolerance spec than others, and you pay
more for that. A 1% cap is going to bring a design closer in performance to
what the designer intended, than a 20% one might, if at the edge of its
spread. The cumulative effect of wide tolerance components - not just caps -
in an item, may cause its performance to deviate quite substantially from
that which the designer intended, but never-the-less keep its performance
within what the manufacturer's marketing people consider acceptable for the
product. No matter what you say or believe, the market is so competitive
that equipment is built down to a price, and that sometimes includes using
inferior quality components.

WTF is a 'gain non-linearity' in your book btw ? Why would a capacitor
cause it
?

Language, language. I haven't sworn at you now, have I, no matter how much
I've wanted to ...
The designer of an amp will want it to have a flat gain across its whole
intended bandwidth. Various factors such as interstage coupling design, will
try to influence this, and the designer will try to mitigate it with his
choice of component values and types. When you see the gain / bandwith curve
for any amplifier plotted, it is never a totally smooth line. It's a fact of
life. Where tone control networks are fitted, these are highly dependant on
the values of capacitor fitted, to produce a flat characteristic at the
control midpoint. Any deviation from the designer's calculated values, will
result in unintended lifts or drops in the curve at particular frequencies.


Graham

Anyway, I don't want to discuss it any more. I actually have a life. If it
makes you happy,
you're right. I'm wrong. The whole of my working life has been for nothing.
Clearly I've understood nothing of electronics over the whole 40 years I've
been involved with it. My english must also be really poor, since you have
such trouble understanding it. I am a **** for having beliefs, no matter how
well founded they might be, and any advice I have ever given to anyone,
should be disregarded and treated as wrong. Oh how I wish that I had your
vast intellect, and eloquent manner. I could have been such a success ...

OK now ?

Arfa