Message-ID: <C6ABA8DEC115D311A29D00105ABD3264A96BFD@nmex01nt.hsd.utc.com>
From: "Soegiono, Gamal" <soegiono@nm.hsd.utc.com>
To: rsgb_lf_group@blacksheep.org
Subject: LF: Mains Borne Noise
Date: Thu, 25 Jan 2001 09:07:40 +0100
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>
Content-type: text/plain; charset=windows-1252; format=flowed
MIME-Version: 1.0
Content-transfer-encoding: 8bit

Hello Costas, hello all,

thank you for your interest on my isolation
transformer solution and the suggestions you made.


CK> The only problem with this approach is that in case of 
CK> electrical fault or lightning strike one of the two 
CK> earths conductors is at a high potential (voltage) 
CK> in respect to the other. As the earth conductors are 
CK> connected also to various metal objects (water pipes,
CK> equipment chassis etc) this situation can easily cause 
CK> electrical shock or fire.

[use monospaced font for best view]

P -------W S S  W----------EMF-----p"P"
         W S S  W
N -------W S S  W
           S S-CT
G ---------+ |  W
             |  W
             |  W----------EMF-----p"N"
             |
             +---------------------pG


P = phase conductor of standard mains
N = return conductor of standard mains
G = safety conductor of standard mains

S = shields of transformer
CT = center tap
W = windings of transformer
p"P" = private phase conductor
p"N" = private return conductor
pG = private safety conductor
EMF = electromagnetic fuse (double sense, double throw)

Note: p"P", private phase conductor and p"N", private 
return conductor in fact are more or less symmetrical 
in potential referred to pG and (capacitively) to G.



Regarding case of electrical faults:

fault-1 (p"P" touches pG)
fault-2 (p"N" touches pG)
fault-3 (p"P" touches p"N")

The EMF will detect these 3 situations and will
shut off both, p"P" and p"N" conductors. 


I think electrical shock is impossible as long as 
all connetions made with conductor pG are low ohmic 
and as long as EMF does not fail (when did you check 
resistance of the safet conductor in your house 
wiring last time? ;-).


Regarding case of lightning strike:

Lightning cannot strike G, nor can it strike pG
directly (those conductors are underneath earth 
surface).

Lightning can however strike outdoor equipment 
such as the antennas. Antennas are protected with 
safety spark gap (approx. 15 kV flashover voltage)
connected to mentioned private grounding rod
(direct electrical continuity with pG, indirect
electrical continuity with G via soil).

Antennas are connected via LC circuits with 
coaxial cables. Shields are directly connected 
with private grounding rod. Center conductors
have arrestors outdoors and at the shack end
(protection levels 1kV/250 V respectively).

Coaxial cables run in steel conduit, 70 cm below
soil surface for 7m length, before entering the 
house.

I do not see any chance of built up of lethal
voltages in between conductors G, pG and local
soil potential. In case of direct hit of a
lightning strike into antenna, the conductor
pG and all metallic objects connected to it
will jump to a potential Istrike times
ground rod to earth resistance (6 Ohms in
my case). This will happen as well if I directly
connect conductors pG and G. So where is the
benefit of using conductor G exclusively?


CK> A few years ago somebody tried a similar isolation 
CK> system but conencted together the two earth 
CK> conductors with an inductor made of thick
CK> electrical wire on a ferrite toroid. The coil 
CK> had very small reactance at power frequency 
CK> (50/60 Hz) but blocked conducted noise at radio
CK> frequencies.

This is a good idea as such and was taken into
consideration as well. Finally I did not implement
this for the following reason:

Receiption on my QTH is/was impaired by common mode
(noise) currents coming along the safety wire,
flowing on the outside of equipment chassis,
over coax shield surface and being coupled into
the coax at the antenna. Those currents seem to
origin from relative high impedance voltage sources.
Preventing them from flowing along a conductor,
an even more higher series impedance is required.
To be effective in LF range, large values of
inductivity are required. Large inductors have a
great amount of self capacity which in return
again provides a conducting path for frequencies
above LF.

The cure in my case was to provide a private
earth connection at the antenna (shorting out the
noise voltages from shield to ground) and break up
the conducing path from G to pG, leaving the
leakage capacity in between both as the residual 
path for noise currents.


best 73 de Gamal


> -----Original Message-----
> From:	Costas Krallis [SMTP:sv1xv@eexi.gr]
> Sent:	Tuesday, January 23, 2001 2:44 PM
> To:	rsgb_lf_group@blacksheep.org
> Subject:	Re: LF: Mains Borne Noise
> 
> At 11:44 23/01/2001 +0100, you wrote:
> >>> Has anybody on the list got a "whole house" mains filter 
> >>> installed at their QTH?
> >
> > Not a "whole house" mains filter, but a radio room
> > mains "filter".
> >
> > All main supply to my radio equipment is run via a
> > single phase isolation transformer,...
> 
> 
> The only problem with this approach is that in case of electrical
> fault or lightning strike one of the two earths conductors is at
> a high potential (voltage) in respect to the other. As the earth
> conductors are connected also to various metal objects (water pipes,
> equipment chassis etc) this situation can easily cause electrical
> shock or fire.
> 
> A few years ago somebody tried a similar isolation system but conencted
> together the two earth conductors with an inductor made of thick
> electrical wire on a ferrite toroid. The coil had very small reactance
> at power frequency (50/60 Hz) but blocked conducted noise at radio
> frequencies.
> 
> Costas
> 
> 
> 
>  +------------------------------------------------------------+
>  | Costas Krallis SV1XV      *   LOC KM18UA                   |
>  | P.O.Box 3066              *   FAX: +30-1-3811362           |
>  | GR-10210 Athens           *   E-Mail: sv1xv@eexi.gr        |
>  | GREECE                    *   PGP key: 0x3BDBBC34          |
>  +------------------------------------------------------------+
>