[TowerTalk] ...and another coax adapter question

[Jack Brindle]

The problem isn’t so much resistive buildup, but rather arcing and flash-over. The reason the right angle connectors are such a problem is that the center conductor can either  move or is bot well-placed to start with. At 100 watts it is OK, but at 500 it may arc. This tends to be destructive to the amplifier. Good amplifiers will protect themselves, leaving you to wonder why it just shut down.

Similarly, coax in semi-tropical environments exposed to the sun will see migration of the center conductor over time, changing the coax characteristics (which a TDR should see), but ending in contact between the center conductor and shield, which again will be problematic to the driving amplifier. I have seen this take less than a year in South Florida.

With hamfest-variety connectors and adapters there is not much care taken in assuring the center conductors are both well-placed and secure in their positions. Add to that the insulators used (phenolic or something other that Teflon) will weather and age with time, breaking down to
allow center-conductor movement. 

All this is why we use really good connectors. Most of the suggestions are for UHF family connectors, but there is junk even in type-N and BNC connectors as companies push to lower costs to increase sales. These are easy to find, unfortunately.

Also note that for mismatched lines, which we all use, things get far worse. The scenario here is an ATU driving coax to a wire antenna. The ATU will make the transmitter and amplifier happy, providing a good impedance to those devices. But beyond the ATU you will have very high losses in everything from the ATU components, connectors and coax. A lot of power will be dissipated in these things, with surprisingly little being radiated from the antenna. N6BV did some excellent work in this area, which can be found in his published articles and the ARRL Antenna Book, which Dean edited during his time at ARRL. His presentations show how much the components in the antenna system are stressed.

73,
Jack, W6FB


> On Nov 25, 2023, at 12:51 PM, Scott Townley <scott at nx7u.net> wrote:
> 
> If that's the case, then wouldn't an insertion loss test reveal such an issue?
> 
> For example, I was deconstructing the power handling of a Charter Engineering B5-series coaxial relay (N-female connectors).  If I map "max insertion loss at x GHz" to the average power rating on their published chart, at every point the dissipated power comes out to 11 watts.  That tells me that "good" N-connectors can dissipate 5 watts, and for that mechanical form factor I should be able to derive a power limit based on measured insertion loss.
> 
> Or are the physics of power handling in "not-so-good" connectors different from what would be revealed by an insertion loss measurement?  Assuming of course that we are not in the realm of high voltage breakdown of the dielectric (or maybe that's exactly the difference?).
> 
> 
> 
> On 24/11/2023 17:29, Jack Brindle wrote:
>> One of the big problems with off-brand connectors and adapters is their power handling capabilities. This is especially noted in the right-angle adapter. We see a lot of arcing when higher power is applied.
>> Higher power in this case is anything above just a few hundred watts. Whenever I see an amplifier log report with a lot of high reflected power or PA Dissipation faults, I immediately start looking at the
>> connectors and feed line for problems. These are usually found in one of several places - right angle adapters, and baluns that were meant for low power, but have been over-stressed with high
>> power.
>> 
>> My advice? If the right-angle adapters aren’t Amphenol, throw them out.
>> 
>> 73,
>> Jack, W6FB
>> 
>> 
>>> On Nov 24, 2023, at 12:59 PM, Scott Townley <scott at nx7u.net> wrote:
>>> 
>>> Say I have a pile of coax adapters on my bench (we all do, right?).
>>> 
>>> How do I separate the wheat from the chaff?  I have a fully equipped bench...I would think the go-to would be a TDR measurement.  Any better/additional suggestions?
>>> 
>>> TIA,
>>> 
>>> -- 
>>> Scott NX7U
>>> 
>>> _______________________________________________
>>> 
>>> 
>>> 
>>> _______________________________________________
>>> TowerTalk mailing list
>>> TowerTalk at contesting.com
>>> http://lists.contesting.com/mailman/listinfo/towertalk
> 
> 
> -- 
> Scott NX7U
> 
> _______________________________________________
> 
> 
> 
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com <mailto:TowerTalk at contesting.com>
> http://lists.contesting.com/mailman/listinfo/towertalk