Sir,

I have seen several MSR-2000 Power Amplifiers go spurious when the electrolytic capacitors dry out.

Unless you looked at the output of the PA with a spectrum analyzer, I wouldn't rule it out.

However I did see here that you said you get no desense with a dummy load connected instead of the PA so I will concede my initial suggestion may be wrong.

One of the best things you can do in this case is lock your transmitter on and turn the power level down to the lowest you can and still have a good amount of desense.?

Then use a handheld tuned to your input frequency and a step attenuatior if you have one, and walk around the site to see where you get maximum noise.

In the past I've had problems with CFL lights, boat and motorcycle battery chargers and FM radio stations that were spaced 600 kHz apart on the FM dial.

Modern cellular radios are bolted behind the antennas and run fiber to the building, but the 48 volt power is usually supplied to the radio with unshielded 6 or 8 gauge wire. Could be the rectifier charging the batteries is creating your noise problem.

Another thing to check is the Telewave TPRD-1556 duplexer can be prone to dissimilar metal corrosion between the coupling loop plates and the top of the can.

?

>? The rest of the antennas are not connected to anything and the feedlines
have (unfortunately) been hacked off just below the catwalk.?

Unlike fine wine, abandoned antennas don’t get better with age.? And, with
their cables cut off, that doesn’t help matters either.? Any of them could
be creating noise when excited by your transmitter RF; it’s the same as the
proverbial “rusty guy wire” problem.

> I cranked down the power on the TX and the desense was still there all the
way down to nothing.?

It would be helpful to know the amount of desense at a few different power
output levels, like 100 watts yields X dB of desense, 10 watts yields Y dB
of desense, and 1 watt yields Z dB.

> I have been told that the feedline has been swept with a NanoVNA

Swept as in sweeping the line via time-domain, or swept for return loss as
if sweeping the antenna?

> so I’m taking the word that ‘it looked fine’ with a huge grain of salt.?

"Trust but verify".? Or depending on the source, don't even trust.

> whether it was a heliax splice or two N-Connectors

I hate Heliax splices.? I'm glad they quit making them.

> Against my advice, the antenna was hooked directly to the feedline without
a jumper.

7/8" is a hell of a moment arm?on a fixed N connector...

> I don’t know much about the G7-144, but from what I’ve seen here in the
archives it appears
> to be ok for repeater service with the exception of not being particularly
good in the weather.?

That pretty much sums it up.? For whatever reason, G7's seem to be lightning
magnets.? I don't know why.? But, assuming lightning wasn't involved, the
other common failure is water getting into the base coil assembly and
corroding the works.? Related to that is another common problem which is
installer-induced - blocking the weep hole at the N connector, thus trapping
water inside.? Since you mention someone attached a 7/8" connector directly
to the antenna, it seems quite possible, if not likely, that when they
sealed the connection they blocked the weep hole.?

> I did see one place at the end of the ice-bridge where someone had stepped
on a piece of 5/8” heliax and crimped it good

Unless it's crimped to the point of the center and shield coming in contact
(which is damn near impossible), a dent isn't going to cause desense.

> The only thing that worries me are the 5G panels at the top of the water
tower.?

I wouldn’t put the cell carriers as being high on my list of suspects.? I'd
put your antenna at the top of the list, your feedline/connectors/splices
second, and other potential noisemakers in your antenna's near field third
(and that includes any/all abandoned antennas).


? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ---- Jeff WN3A