[Elecraft] K3: 6M SSB audio hash

[Jim Brown]

On Sat, 7 Jun 2008 17:50:13 -0400, Joe Subich, W4TV wrote:


>Jim, 

>Which is it?  In this reply you say: 

>> >... but not as good as if the shield were connected to the 
>> >chassis (usually PTT ground) and the mic treated as a 
>> >floating/balanced input.  
>> 
>> WRONG! What you have just described is perfect example of a pin 1 
>> problem, and a VERY common cause of hum, buzz, and RFI. 

>but in the reply to Jamie you say: 

>> The proper connection for that is to connect the two mic 
>> wires (the hot and the shield) to the balanced input AND 
>> to connect the shield DIRECTLY TO THE CHASSIS by a very 
>> short path.   

>Your description says the very same thing as I did ... mic wires 
>floating with the shield connected directly to the chassis.  

I'm not ambiguous, you are! :)  "PTT ground" is ambiguous, and implies 
that it goes somewhere and eventually finds the chassis. I'm talking the 
CHASSIS. Period. 

>The point is that Icom, Yaesu, Kenwood and Heil all float the 
>mic return above the chassis and then connect the shield of the 
>mic cable to the floated mic return!  

>The shield should be 
>connected to the PTT return - which is connected to the chassis. 

The shield should be connected STRAIGHT to the CHASSIS, not to something 
that is connected to the chassis. That "connected to" covers a multitude 
of sins -- like a circuit trace that several inches to a phone jack 
that's screwed to the chassis. That trace has inductance, and the voltage 
drop across that inductance causes mischief!   

>Elecraft connect the mic return to chassis inside the K3 *BUT* 
>Pins 7 and 8 (PTT and mic return) have series chokes which result 
>in about 5 Ohms of resistance between both pins and the case. 
>The same is true for the shield/sleeve of the rear panel mic 
>jack ... the sleeve is floated by an RF choke with about 5 Ohms 
>of DC resistance.  

Yes, they do, and I have a BIG problem with that. I had that discussion 
with Wayne about three years ago. I'm not at all happy with the audio 
interface on the K3. For one thing, those transformers are unshielded, 
and pick up hum from power transformers nearby (like the power supply for 
my power amp, which right under the operating desk). There's low level 
hum when my line input is turned on (so I can playback contest messages 
from the computer, or send AFSK). The hum is there with or without a 
cable plugged in. This hum pickup is so bad on my neighbor's K3 that it 
regenerates to full power!  I suspect he has an unusually strong magnetic 
field. 

The one saving grace of what they are doing with the shield is that it is 
also bypassed to the KIO audio daughterboard. It isn't clear to me how 
the bypass capacitors on that daugherboard gets to the chassis. If it's a 
short (small fraction of an inch), great. If it isn't, it's a potential 
RFI problem, especially at higher frequencies. A clue (not an encouraging 
one) -- the retaining screws for  the DB connectors are 8 ohms off of the 
chassis of the radio!  Not a good thing! I haven't measured the audio 
daughterboard chassis, but I suspect the same problem there. So far, I 
have not had RF feedback problems, but my QTH is not challenging in that 
regard -- only my 160M vertical is closer to the radio than 100 ft. 

>The best solution for all amateurs is to connect the mic return 
>pin on the mic connector to the shell to tie both mic return and 
>shield to the chassis by the lowest possible resistance path. 

The cable shield must go to the chassis. If that is the mic return, so be 
it.  BUT  it doesn't help if the shell isn't connected to the chassis!  I 
use an EV RE16 with my ham rigs (a balanced dynamic mic). The shield goes 
to the connector shell, the mic lines go to the mic input and the mic 
return. My Yaesu FT1000MP has a pin 1 problem at its mic input that 
causes it to have RF feedback on 15M and 75M. I can measure that pin 1 
problem by the test method shown in my AES paper on the topic, and, 
indeed, it peaks on 15M and a bit above 75M. That paper is on my website. 

Putting it simply, MANY ham rigs are built with pin 1 problems. That is, 
they have a design flaw that causes hum, buzz, and RFI when we connect 
cables to them. 

73,

Jim Brown K9YC