On 11/06/2019 09:26, Brian Reay wrote:
The Natural Philosopher wrote:
On 10/06/2019 19:30, Brian Reay wrote:
On 10/06/2019 13:27, Roger Hayter wrote:
Brian Reay wrote:

The Natural Philosopher wrote:
On 10/06/2019 08:15, Brian Reay wrote:
The Natural Philosopher wrote:

A lot of unearathed metalwork floats at 110vAC due to there being RF
filter caps to live and neutral from the metal case.


It shouldn't be designed like that, it is pointless.Â* It doubles the Z
between L&N to dump noise via the C's (C's in series rule) and
introducing
the stray 110Vac point.Â* As I posted in another group, this
situation can
arise due to a fault condition if, for example, the earth at the
plug (top)
is lost or somewhere else before the filter.

It isn't pointless from an RF POV.





Read what I posted, carefully.Â* The two C's would be in series,
halving the
overall value (assuming they are the same), that doubles the
impedance Z to
decouple the noise. One C would between L&N would be more effective and
avoid the stray 110V AC point.

It would be more effective for possible differential noise, but totally
ineffective for common mode noise, which is more likely.Â*Â* As someone
said it is common to have both, as well as coupled series inductors and
further capacitors if it is a posh mains filter.


The 'mid point' isn't connected to ground. That is the point. Stop
trying to score points and read the thread!

The mid point IS connected to the case though. That is, RF wise, a local
'ground' , of a sort

So you €˜pump any noise onto your local (and signal) ground rather than to
earth where good design dictates you should send it.

Novel approach.







Wel te planet itself is only a 'local (and signal) ground'

--
"And if the blind lead the blind, both shall fall into the ditch".

Gospel of St. Mathew 15:14