From: "Dave Sergeant" <dave@davesergeant.com>
To: rsgb_lf_group@blacksheep.org
Date: Wed, 04 Aug 2010 11:29:31 +0100
MIME-Version: 1.0
Message-ID: <4C59410B.7665.112465B@dave.davesergeant.com>
In-reply-to: <5A5E700F08A44F948387B7FD7FA8DF44@JimPC>
References: <BLU146-W201F9B3FE3E982D483D89BCFA30@phx.gbl>, <002a01cb32f8$bcf915f0$4001a8c0@lark>, <5A5E700F08A44F948387B7FD7FA8DF44@JimPC>
Content-description: Mail message body
Subject: LF: Re: Noise R75 - rs232 from smpsu -
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org


I have an interesting example of this with my PC and router.

Initially when I set assembled my new PC a couple of years ago I found 
terrible hash from its psu on 80m and 40m. This was traced to the 
infamous omitted filter components in the ATX psu. Filtering on the 
mains lead made no difference, but it went away when I fitted the 
required missing components - ie a series dual choke and ceramic caps 
down to the earth line. Components were salvaged from a scrap TV board.

When I changed my broadband to O2 last year I found the hash on 80m had 
reappeared, but only when I unplugged the supply to the PC last thing 
at night. Provided the PC was plugged in (but powered down) I didn't 
get the noise. Presumably the router psu - the router is a Thomson 
TG585V7 - has no filtering and the hash finds it way to earth via the 
PC earth connections.

73 Dave G3YMC

On 4 Aug 2010 at 10:21, James Moritz wrote:

> My experience would agree with Alan - several times with laptops, and
> also a "compact" desktop with external PSU, making a direct connection
> between the mains ground and the chassis of the computer has made a
> substantial noise reduction. It seems to be fairly common that there is
> no ground  continuity between input and output of PSU. With no ground
> connection to the output side of the PSU, any common-mode noise due to
> the PSU has to find its way to earth via the RF connections usually, and
> can get into the signal path in the process. Winding as many turns as
> possible of the DC output lead round a large split ferrite core can also
> be helpful in reducing the noise current. I find >10 turns are needed to
> make a useful difference.
> 
> Cheers, Jim Moritz
> 73 de M0BMU 
> 


http://www.davesergeant.com