In article , Glenn Richards
wrote:
Don Pearce wrote:

Glen,, please give it a rest. You aren't funny any more and this is
all becoming just a little boring.

And herein lies the problem with this group.

I ask a serious question, expecting some sensible responses, and instead
all I get are sarcastic comments.

The theory is sound anyway, if sticking an LCR filter across the mains
input does indeed reduce the noise floor.

I note the "if" qualifier in the above statement. :-)

The problem is that you also need to know the source and load impedances in
between which you are fitting any filter. e.g. if the load is essentially a
short, then a shunt capacitance across it won't do much.

The problem is that the source impedance can be expected to vary with
location and frequency, and that the load impedance can be expected to vary
during the mains cycle. Also, if you change the source impedance seen by
the PSUs in the units you are trying to 'protect' you might cause some
problems to become worse, not better.

As reducing the noise floor (by whatever means) improves dynamics.

This makes the unstated assumption that the PSU isn't already ensuring the
'noise' from the mains is - at the output of the unit - already so low as
to be inaudible.

[snip]

So, I repeat. If you've got a lot of electrically noisy equipment on the
same ring main as the hi-fi (eg switching power supplies in PCs etc)
that's kicking out a lot of RF interference on the ring main, would it
not follow that eliminating (or at least greatly reducing) this noise,
and therefore providing a clean power feed to the amplifier and other
equipment, would lower the noise floor?

It may do so *if* the noise was previously at an audible level. It is easy
enough to measure the output of a unit to look for signs of this - although
simply listening and finding you can't actually hear anything may be a
useful guide.

The key point is that the designer of the equipment may well be sensible
enough to know he/she lives in the real world, and will already have:

1) Designed the unit with a PSU that rejects mains interference to a
suitable degree.

2) Designed the circuits in the unit so as to reject fluctuations on the
power rails, and correctly references itself to the defined system ground
level.

Once again, I'm not for a moment advocating spending £250 on something
from Russ Andrews. What I'm actually talking about is getting a standard
IEC lead for a couple of quid, then fitting a filtered 13A plug to it,
also for a couple of quid. At trade prices in single quantities this
will cost less than a fiver (which is actually less than you pay for a
standard IEC cable at places like PC World).

So, would someone like to provide a serious answer now?

There is unlikely to be any harm in buying and using some simple mains
filters. Although note the caution I give above.

FWIW I use 'RS' filtered blocks on the systems I use. The primary reason
being that the Quad 34's I use are prone to picking up 'clicks' from the
mains, and these filters dispose of the problem. As does the TV I use as a
display in the living room.

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