In article , mick
wrote:
On Sat, 08 Aug 2009 09:46:15 +0100, Jim Lesurf wrote:

In article , mick
wrote:
On Fri, 07 Aug 2009 17:31:43 +0100, Eeyore wrote:

Alas, the amplifier designer has no control over what loads the user
connects. And this will change with the choice and length of the
loudspeaker cables.



Fair enough, but surely the only amps that are likely to have much
output (i.e. enough to drive the cable - never mind the speaker) above
1MHz or so are likely to have severe problems anyway (such as
overheating) aren't they? Ok, maybe not if the oscillations are
triggered on audio peaks I suppose, but how would that be audible?

The 'output' would come from the amplifier oscillating. If the amplifier is
not unconditionally stable and you connect it to a load that means the amp
and load combination is unstable, then it can oscillate without any input
signal prompting it to do so.


I'm still trying to get the hang of this - please excuse my stupidity
and put it down to old age! Ok, cable always influences RF performance,
I'm happy with that. I'm incredibly skeptical about whether it affects
audible results though. The L, C and R and/or Z values are just too low
at audible frequencies.

Probably. But 'almost always' isn't 'always'. And the longer the cable
runs, the bigger the lumped values will be for RCLG.


Now I have to get around getting enough RF out of an audio amp to make
those values important. At the frequencies you are talking about a
simple zobel, or even a 1nF capacitor across the output should remove
enough RF to swamp the effects.

No, sorry. Putting an RC snubber across the amp output won't protect the
amp from the cable-transformed load having dips. Not against other things
like it being a very large shunt capacitance. Indeed, if you just put
a shunt capacitor across the output that might provoke oscillations if
the amp isn't already stable for other reasons.


If you want to rely on a passive output network you generally need the
amp to have an added output series impedance *outside* its feedback loop.
Usually an inductor and resistor in parallel. Have a look at the

http://www.audiomisc.co.uk/HFN/Cables2/OhmAndAway.html

which is the second article in the series on loudspeaker cables and
amplifier-cable-load interactions. This outlines how instability can
depend on the loading, and discussess the general type of output
passive network designers have employed.

FWIW The Isolda cable will normally include its own series inductor
for reasons like this since the cable itself has an exceptionally
high shunt capacitance per metre.


I'm not convinced that knowing the RF cable properties tells you
anything at all about the audio performance.

It allows you to determine the cable RCLG values which you can then apply
at audio and near audio. It also shows you if the transformed loads might
be severe so you'd need to take care with an amp that was not
unconditionally stable.

There shouldn't be any RF present so the information is redundant.

Alas, there will be is the load means the amp and load are an unstable
arrangement and the system then oscillates. if there is a load impedance
dip this oscillation may then draw large RF currents from the amp.

Afraid none of this is simply theory, I've seen it on many occasions. the
good news is that none of it should matter if the amplifier is a *good* one
and is unconditionally stable, etc. So it should be a problem the designer
and maker fixed for you before the unit left the factory. However Eeyore
has said that some amps are "pure ****". I can't state I'm sure he is
right. But I am aware that no-one seems to be checking this in reviews,
etc. So for me the worry is absence of data on this point.


I'd be interested to see measurements for a bit of mains twin and earth,
if the mood to test takes you again... :-D

Funnily enough, I'm currently casting a skeptical eye over some of the
recent claims by Russ Andrews and Ben Duncan about their fancy mains cables
being 'RFI filters'. I can hint that I don't really agree with everything
that claim. :-)

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