[Elecraft] K3 RFI in transmitted audio

[Jim Brown]

On Sat, 11 Sep 2010 20:06:50 -0400, Joe Subich, W4TV wrote:

>This is good advice ... and always bond the chassis of each piece
>of equipment to the common ground point and not "chassis to chassis."

This is the only advice that I disagree with. Low impedance bonding 
chassis to chassis of equipment that mave an unbalanced signal 
connection between them is a VERY GOOD THING, because it can 
effectively short out any voltage that may be present between them 
(like the IR drop being discussed here). 

Don said:

>Strange modulation of the V- line to the accessory can result, 
>especially with SSB transmission, 

Nothing strange about it, it's simple Ohm's Law. :)  You have DC 
current on the V- lead that reaches a peak value of at least 20A, so 
there is voltage drop on that lead as predicted by Ohm's Law. That 
voltage drop becomes a problem if it appears between equipment that 
has unbalanced connections, like a computer sound card and a rig, or 
between an outboard audio interface and a rig. That's why an interface 
device SHOULD be powered from the end of the power cable nearest to 
the rig, which is the basis of Joe's advice. 

W8JI was the first guy I know of who brought this problem to our 
attention. That IR drop from current that varies with SSB modulation 
from about 1A to about 20A will sound very much SSB audio in an AM 
detector, and just like RF in the audio. So you THINK you have RFI, 
but what you really have is simple IR drop. That's why I've asked 
about power and bonding in this discussion.  

Another part of this equation is whether V- is bonded to the chassis 
in the power supply. In general, it should NOT be bonded. Most power 
supplies are built so that they are either NOT bonded, or have a 
single bond that can easily and safely be removed. The linear Astrons 
are an example -- the entire V- bus floats all the way through the 
supply, but is bonded near the V- terminal. This can very easily and 
safely be cut. 

Now, getting back to Felipe's RFI problem. If, as I suspect, it's a 
pin 1 problem in the computer sound card, the ferrite choke could fix 
that problem. But if it's the DC power problem, removing the V- bond 
in the Alinco power supply (if there is one) would be a likely 
solution. I think the DC power problem is unlikely in this situation, 
because, as I understand it, it's the FT2000 RF that is getting into 
the K3 audio, and it's running on it's own internal supply. 

73, Jim K9YC