[Elecraft] PX3 noise generated on KX3-2M

[Alan]

I finally got around to checking this.

On 02/23/2016 11:37 AM, Jim Brown wrote:

> I've been bugging Wayne about these issues for
> more than ten years, and he still hasn't gotten the message. This is a
> great example of ignoring what EMC guru Henry Ott calls "the hidden
> schematic lurking behind the ground symbol," and which is a primary
> cause of EMC failures (RFI, hum, buzz). In the pro audio world, we call
> it "the Pin One Problem," whereby cable shields fail to contact the
> shielding enclosure at the point of entry, but instead go to the circuit
> board, finally reaching the chassis via a path established by the whim
> of the PC board layout artist.

It is not a "pin 1" problem. The metal shell of the chassis-mount power 
supply connector is mounted directly to the metal side panel of the PX3 
and the paint is masked around the hole to make sure it makes a good 
connection.

On 02/23/2016 10:26 AM, David Anderson wrote:

 > Since I purchased my PX3 I have suffered from radiated noise at 144
 > MHz.
...
 > I had already checked that the DC Jack was grounded properly to the
 > side panel. However what I hadn't appreciated is that the common of
 > the PCB is connected to a switch in the Jack and is not directly
 > connected to the chassis of the Jack. What was happening was the
 > barrel of the plug was connecting to the pole of this switch and the
 > PCB common, but only loosely to the chassis of the Jack. Movement of
 > the plug sideways with enough pressure did connect the plug better to
 > chassis and resulted in the noise going away.

I'm trying to understand that.  The power jack has three pins, the 
connector shell (tied to the printed circuit board ground plane in the 
PX3), the center pin (connected to the +12V input) and a switch.  The 
switch shorts to the connector shell when nothing is plugged in and 
opens when a power plug is inserted.

But the thing is, the switch is not used and that pin is not connected 
to anything inside the PX3.  (The only reason I used a jack with a 
switch was that it was the only suitable chassis-mount type I could 
find.)  Besides, that pin is internally disconnected whenever the plug 
is inserted anyway.

I just confirmed all that by double-checking the PCB artwork and by 
measuring an actual PX3 with an ohmmeter.  If you'd like to check it 
yourself, the three connector contacts are arranged like this:  The 
switch pin is the one on the side of the connector.  The +12V pin is the 
one farthest from the panel.  The shield/ground pin is the one that is 
almost hidden underneath the connector.

 > I opened up the back of
 > the PX3 and connected the chassis part of the Jack to the PCB common
 > side of the jack's switch by soldering a small piece of copper EMI
 > tape across the gap. This ensures that the PCB common is stoutly
 > connected directly to chassis at the DC Jack.
 >
 > This has had the desired effect, no more intermittent  increases in
 > spurs heard in the receiver as the leads are touched or moved,

I'm not sure from that description exactly what was connected to what so 
it's hard to figure out how that could have affected the noise.  It 
sounds like you probably connected the switch contact to the connector 
shell, but I don't see how that could have affected anything since that 
contact is disconnected whenever a plug is inserted and it's a 
no-connect on the PC board.

You might check to make sure the mounting nut on the power connector is 
good and tight and the connector body is making a solid contact to the 
inside of the side panel.  I'm guessing that's the most likely cause of 
the noise radiation.

Alan N1AL