In an earlier message, Jerry, K3BZ wrote:
"Although it might not pinpoint exactly why, it would at least indicate the
source of the problem is the cable. Since we have several customers
successfully using even longer cables, I think that might be sufficient to
justify cable replacement."
While I was out planting trees today, this problem was continuously running
through my mind... why does running 3 EHU's work fine, but adding the 4th
create a problem? After I was finished outside, I reviewed the message thread
for clues. Then I saw the above message from Jerry stating that even "longer"
cables have worked fine at other installations. He didn't say how long, but his
statement basically rules out this 500 foot run as being the first installation
with this length of cable and creating some unforeseen engineering problem. I
hereby retract my original statement regarding engineering this problem at the
factory!
So... while digging in the dirt, a lightbulb of logic snapped on. I really
believe there is indeed a "mis-wired" twisted pair in that 500 foot cable.
Here's my logic:
The problem only occurs when the final EHU is installed. For simplification,
lets say wires 5&6 (a pair) run EHU 3 and wires 7&8 (another pair) run EHU 4.
If the cable is cross-wired it would use wires 5&7 for EHU 3, this is most
likely working "good enough" even though those two wires are NOT grouped into a
"pair". But when current is applied to wires 6&8 (not a pair) to run EHU 4, now
you have contention within the two legitimate twisted pairs. The pulses driving
EHU 3 will be coupling with the pulses driving EHU 4 and vis-a-vis.
Like Jerry originally stated, I also highly suspect the cable run. Not
necessarily the cable, the conduit, or the length, but the high probability of
a cross-wired pair of conductors.
73 de Bob - KØRC in MN
P.S. This is all theory on my end... I have never bought, installed, used, or
even seen a SteppIR in person. But my next antenna will be a SteppIR!
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|