[NLRS] [BC'ers] SWR infinite on 432.....rats!

[Dr. Gerald N. Johnson]

On 9/6/2016 8:34 PM, David Palm wrote:
> So I've been having a bit of intermittent and puzzling issues on 432 MHz
> for the last few weeks.  I got through the UHF contest okay, but as of two
> days ago the SWR has decided to go infinte and stay infinite.  I swapped
> out the N connector on the LMR-400 down-cable right at the rig, which after
> a lot of horsing around to figure out the problem, had started to seem
> flaky anyway.  No change.

My FT857Ds have neen drifting to show higher SWR as time passes
>
> I've run the rig into a dummy load and I know for sure I'm getting power
> out and that into a 50 ohm load everything is as it should.
>
> So....it seems that the only thing is to break this down by subsystem.
> First, climb up there, unwrap the waterproof connection between the LMR400
> down-cable and the flexible rotor loop, terminate the main down-cable with
> a 50 ohm load, and see if that's the culprit.  If not then climb all the
> way up, detach from the antenna and test the rotor loop by itself.
> Finally, test the antenna with a known-good cable.

The SWR won't be infinite (reflected power equal to forward power) if 
the problem is any distance away from the shack. Even LMR-400 has loss 
especiaslly at 432.

Have you checked the length of the type N center pin? Push your index 
finger against the connector end. You will have an imprint of the ground 
connection ring and if the center pin is long enough a dent in the 
center of that ring imprint. Classic type N connectors (UG-21 family) do 
not anchor the center pin and conductor and flexing the cable tends to 
cause the center pin to retract. There is a UG-1185 connector that 
anchors the center pin to keep it from retracting. If you have a 
directional wattmeter, you can compare the reflected power to the 
forward power and presuming the fault is an open or shorted coax you can 
compute the distance to the fault by the two way power loss. It doesn't 
take a lot of capacitance at a broken center conductor connection to 
lower the reflected power so you can't depend perfectly on the power 
loss between forward and reflected to determine the distance to the 
break. A pulse generator and a scope can make a time domain 
reflectometer to detect the time delay to the fault. The faster the 
scope and the pulse rise time the better the resolution.

Sometiems it is possible to detect the length to the fault by checking 
the SWR over the whole band, there can be peaks where the distance to 
the open is a multiple of half waves, and a lower SWR to the open when 
the distance is an odd multiple of quarter waves. Just the opposite if 
the fault is a shorted cable. Unfortunately it won't give the number of 
halfwaves to the fault.

Corrosion of the barrels as Jeff has noticed is not out of the question. 
You save one connector interface by using male on one cable, and female 
on the mate over using a barrel. And some imported connectors are not 
really made for outdoor use at RF and don't stand up.

Scotch #33 is pretty good at waterproofing connectors, I think better 
than other products. it works even better to put on a layer of 33 
strectched a bit, then lay on a layer of Scotch Kote sealant, then 
another layer of tape. In decades past, electricians applied up to 5 
layers of each and directly burried the splices and got good lifetimes. 
Our club repeater antenna on top of a water tower with partial air 
insulated 9913 was installed by one of those electricians (KI0Q) with 
the multilayer tape and ScotchKote at the connections and was used 
without any problems for more than 15 or 20 years. The coax was taken 
down by the water tower painting crew and Dave found a length of 7/8" 
heliax that is the run now increasing the repeater range without 
changing its 25 watts transmitter power.
>
> Any hints or sage advice out there on making this easier?  Or more probably
> it just is what it is.
>
> 73,
>
> David  W9HQ
>
73, Jerry, K0CQ