[TowerTalk] Shunt Feed Mystery

[RT Clay]

 I'm not sure that shows that the problem is the 40m loop. Certainly you have a problem with a ferrite heating up. That ferrite is probably somewhere between the main tower and an insulated element. As it heats up, the coupling to the insulated element will change, changing the top loading on the tower (and the SWR). That changes the current distribution on the whole tower, and will also change how much RF gets coupled into the 40m loop.

It is easy to verify by taking the loop down and seeing if the problem persists. In ON4UN's book there is a description of a large feedline trap that can be used to decouple a dipole. Basically you have the dipole feedline wound in a large air-wound coil (4 inch sewer pipe) resonated by a capacitor. I am using this to decouple the 80m dipole on my tower.
I had similar problems when first shunt-feeding my tower (100 ft tall) on 160. Transmitting at 1.5 KW the SWR would gradually change. The problem turned out to be the ferrite balun on the two element 40m Force 12 yagi at the top of the tower. The heating in the balun (Force12, beads on coax in a pvc pipe) was enough to partially melt the pvc pipe, which ended up banana-shaped! I fixed it by using an air-coil choke balun on the 40m yagi and grounding the center of parasitic elements.  There are two other 20m antennas on the same tower (one at 50ft, one interlaced with the 40m at the top), and there isn't a heating problem in the same baluns on those antennas- it was only with the large 40m driven element.

My case was several years ago and I hadn't seen the center-tapped choke solution. In your case I would be worried about using that on a 40m antenna. If there is enough reactance in the choke to have no effect on 40m, it seems that there might be a significant reactance on 160/80. Did you check that the problem happens with only the new tribander and stock baluns? (i.e. before installing the center-tapped chokes)?

Tor N4OGW

   On Friday, December 13, 2019, 12:23:04 PM CST, K9MA <k9ma at sdellington.us> wrote:  
 
 Another very interesting data point:

The 40m full wave is supported near the top of the tower on a short side 
arm. That's the feedpoint, 1/4 wave from one end. It's like a lopsided 
inverted V or sloper. There's a balun, but the feedline is isolated from 
the tower, and pulled as far from it as possible. Still, there's no 
doubt that some power is coupled into it when shunt feeding the tower on 
160.

I connected a dummy load and a power meter to the 40m coax, just to see 
how much power came back. Initially, 5 W was coming from that antenna. 
However, when "something" heated up and the shunt feed SWR rose, that 
power dropped to 1.2 W. This seems to make it almost certain that it is 
the 40m balun which is heating. (And that would take way more than 5 W.) 
So, I think I've found the culprit.

I'm not sure what to do about it. I only use that antenna for receiving, 
but it's useful. It is a bead balun, which would have pretty low 
impedance on 160. One possibility would be to make a receive only balun 
with very high impedance on 80 and 160.  It might not survive the 
initial tests, but it wouldn't cost much. Any other ideas?

73,
Scott K9MA

On 12/10/2019 23:31, K9MA wrote:
> Here's the mystery: After a long transmission on 160 at 1.5 kW, the 
> SWR suddenly, but smoothly, rises to 1.4:1, but rises no further. A 
> pause of a few seconds allows the SWR to drop back to 1:1, but it will 
> rise again when transmission is resumed. Retuning the gamma capacitor 
> will reduce the SWR, but not quite back to 1:1. Clearly something is 
> heating up, but I can't think of an explanation. There are some N750 
> doorknob capacitors in the matching box, but if one of them were 
> heating and drifting, why would it stabilize? Could a ferrite balun be 
> heating enough to reach its Curie temperature? But since the element 
> ends of those are already shorted to the mast or boom through shunt 
> chokes, why would that affect the shunt feed.
>
> The 70 foot tower has an A3 at 50 feet, a JK Mid-Tri at 29 feet, and a 
> D40 at 75 feet. It is shunt fed it on 80 and 160 with separate gamma 
> matches, switched to a single coax with a relay. This has worked 
> reliably for over 25 years, with a TH7 where the Mid-Tri is now. The 
> TH7 had grounded elements, while the Mid-Tri does not. There are 
> bead-type baluns on the A3 and D40, and a toroidal balun on the 
> Mid-Tri. All three antennas now have shunt chokes at the feedpoints 
> ala N9NB/W5JAW. (None did before.) The coax shields are bonded to the 
> tower at its base, but not to the booms or mast at the baluns. The 
> feedlines all are fastened to tower legs or booms, except at the 
> rotator loop. There is a 40 meter full wave supported by a sidearm at 
> its feedpoint near the top of the tower, with a balun, but it's 
> feedline is NOT on the tower, but pulled away at about a 45 degree 
> angle. That's been like that since long before the problem appeared. 
> There's a photo with the TH7 at qrz.com.
>
> Any ideas?
>
> 73,
>
> Scott K9MA
>
>

-- 
Scott  K9MA

k9ma at sdellington.us

_______________________________________________



_______________________________________________
TowerTalk mailing list
TowerTalk at contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk