[Elecraft] RS232 Noise

[Mike Harris]

Hi Jim,

I have your paper and will have another read.  I've been meaning to make 
up a new cable for a while.

The noise has only become noticeable since I increased the sensitivity on 
6m and if I switch out the PR6 I can't hear it.  6m is reasonably quiet 
here, particularly when I beam out to sea (north).  As mentioned in a half 
finished posting that launched, I don't have any benchmark so it is not 
known if the "problem" existed before I shorted the chokes on the data 
return lines as offered as an option in the app note.

The spooky sigh as the WSJT clock display changes shows up quite well and 
the decode has its own longer noise.  I turned the monitor off but it is 
still there.

Regards,

Mike VP8NO

----- Original Message ----- 
From: "Jim Brown" <jim at audiosystemsgroup.com>
To: "Reflector Elecraft" <Elecraft at mailman.qth.net>
Sent: Thursday, April 23, 2009 10:10 AM
Subject: Re: [Elecraft] RS232 Noise


| On Wed, 22 Apr 2009 19:11:51 -0400, Mike Harris wrote:
|
| >This extra noise vanished when I removed the "RS232" cable from the
| K3.
| >The cable is actually the one I made up for the K2.
|
| There are several problems in the setup. First, the K3 connector shells
| are not properly bonded to the chassis, so the cable shield is useless.
| Second, the cable Elecraft sold with the K2-series products used
| parallel conductors, not twisted pairs. That's a recipe for noise and
| RFI.
|
| I live in a fairly quiet area in the Santa Cruz Mountains, and lots of
| small electronics produce noise that I can hear, both HF and VHF. Much
| of that noise is radiated as a common mode signal on the various cables
| connected to the noise source. I have several suggestions for you.
|
| 1) Build the serial cable shown in my RFI application note.
| http://audiosystemsgroup.com/RFI-Ham.pdf  It uses CAT5 cable. When VHF
| noise is a concern, shielded CAT5 may offer further improvements.
|
| 2) Use ferrite chokes on every cable to kill common mode current. A
| single turn through a #43 core peaks the choking action around 150 MHz.
| Two-three turns moves the peak down to 6M and provides about three
| times the choking impedance. If you're concerned about both bands, use
| multiple cores in series. The equipment at both ends of any cable can
| be the noise source, so if the cable is longer than about 1/4 wave, use
| chokes at both ends. I did all of this in my station (and around my
| XYL's computers) and killed a lot of noise.
|
| 3) Fix the pin 1 problem in the K3 by bonding the DB-connectors to the
| chassis. Pin 1 problems couple RFI in both directions -- that is, they
| let RF into equipment and they put RF onto cables that radiate it.
|
| 73,
|
| Jim Brown K9YC