On 12/21/2011 2:21 PM, Jim Barber wrote:
> Curious if anyone in the group had any strong feelings about how to go
> about RF-proofing RS-4XX interfaces? (RS-232, 422, 485)
Two important measures. First, avoid Pin One Problems at both ends.
Second, use twisted pair (CAT5/6/7 is a great choice) for each signal
circuit. If the circuit is unbalanced, connect each signal return to the
CHASSIS at both ends, not the PC board. If he circuit is balanced, and
if there is a cable shield, connect it to the CHASSIS at both ends, and
not the PC board. With many (most?) computers you can find the chassis
at the shell of a DB connector.
In general, twisting is FAR more important than shielding of
interconnect wiring if you are in the near field of the interfering
signal (the TX antenna, or leakage flux from a power transformer). When
I lived in Chicago, my primary TX antenna on 160M and 80M was a long
wire that ended in the shack, and ran a few feet from the computer that
keyed my rig (a K2/100). The serial cable that Elecraft designed for it
was simple parallel wires, and the serial line locked up with only 15W
on 160M. I diagnosed the problem by replacing that cable first with
shielded CAT5, then unshielded CAT5, and finally by shielded twisted
pair mic cable (for pro applications, then firing up the Titan amp to
see how effective each cable was. In all cases, I wired the cables as
described, and I loaded that wire on all bands from 160M to 10M (only
for testing -- it was a lousy TX antenna above 80M).
The result was that the unshielded CAT5 allowed me to run full power
from 160M through 20M, but I needed shielded cable for full immunity at
15M and above. And this was for a TX antenna whose feedpoint (high
current point) was within 5 ft of both the serial cable and the computer
(an IBM T22 laptop).
For more details about Pin One Problems and RFI in general, see
http://audiosystemsgroup.com/RFI-Ham.pdf
73, Jim Brown K9YC
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|