Foundations of Amateur Radio

///////////////////////////////////////////
Coax Loss vs Connector Loss - now with more coax

Posted: 24 Mar 2018 10:00 AM PDT


Foundations of Amateur Radio

Recently I spent some time discussing the losses associated with connectors
between your radio and your antenna. The traditional wisdom, and I use the
word "wisdom" ironically, says that each connector introduces loss into the
feed line. There is an understanding that the more connectors you have the
worse it is and the more loss you have.

Jim W6LG did the test, connected up 30 odd connectors and measured. His
measurements were done on 14 MHz and on 50 MHz, using 50 microvolts and 1
kilowatt. No discernible difference.

Of course after I mentioned this out loud the questions started. Why didn't
he test this at a usable frequency, something like 145 MHz, or in the GHz
band? Then there were those who said that this wasn't a real test and that
it should be tested with coax in between the connectors.

I discussed this all at some length and one idea we had was that perhaps
the intersection between the coax and the connector was the problem, that
each transition between coax and connector and back was introducing the
loss.

I wondered if there was a way to test this.

Turns out that somebody already did. Back in July 2015 Jim K9YC decided
that this needed to be tested. That's right, another Jim.

He set up a test with a dozen 100 foot lengths of low loss coax, that's
just over 365m of coax. This included two dozen PL259 connectors and 11
barrel connectors. He tested using a calibrated HP generator/voltmeter rig.
The total loss was and I quote: "one dB or so less than the loss specified
for the cable by the manufacturer". So, the run with connectors was
actually better than a single run of coax.

In case you're wondering, he tested this up to 500 MHz.

Jim K9YC points out that there is a grain of truth in the loss when using
junk connectors which can introduce excessive loss and can overheat because
the centre conductor is too small.

I should mention that this might now debunk the connectors and loss issue,
at least up to 500 MHz, but there is something to be said about reducing
the number of failure points along the way. Having 35 connections instead
of two is an added risk of water ingress, loose connections, short circuits
in the connector and potential for other unexpected things like an
intermittent connection.

In the broader scheme of things, on a field day, or a temporary antenna set
up, there's clearly nothing wrong with using some connectors to join
together some coax. It also means that my investment into coax terminated
SO239 connectors was based on poor information, though it does mean that I
don't need to carry nearly as many barrel connectors around.

Perhaps it's time to, as Jim puts it, put this "old wives' tale" to bed.

In the same document Jim discusses many other questions in relation to coax
and stub filters in your HF station. I came across the document while I was
looking for information about coax stub filters, since I just participated
in another contest where two stations in the same location were interfering
with each other and I want to be prepared for next time.

There's a lot to discuss in relation to coax stub filters, but in essence
you create a quarter wave and half wave lengths of coax that are resonant
at a range of different frequencies and the combination of these will
either pass or block the band you care about. Given that I have a roll of
Quad Shield RG6 lying around, I thought I'd try my hand at making a set of
these for my next outing. No doubt I'll share my adventures with you as I
explore and dig through the pile of information.

Coax and connectors, stubs and filters, it's all in a days experimentation
in this amazing hobby we call amateur radio.

I'm Onno VK6FLAB
This posting includes a media file:
http://podcasts.itmaze.com.au/founda...teur-radio.mp3

In article , rec-radio-info-
says...

Of course after I mentioned this out loud the questions started. Why didn't
he test this at a usable frequency, something like 145 MHz, or in the GHz
band? Then there were those who said that this wasn't a real test and that
it should be tested with coax in between the connectors.

I discussed this all at some length and one idea we had was that perhaps
the intersection between the coax and the connector was the problem, that
each transition between coax and connector and back was introducing the
loss.

I wondered if there was a way to test this.

Turns out that somebody already did. Back in July 2015 Jim K9YC decided
that this needed to be tested. That's right, another Jim.

He set up a test with a dozen 100 foot lengths of low loss coax, that's
just over 365m of coax. This included two dozen PL259 connectors and 11
barrel connectors. He tested using a calibrated HP generator/voltmeter rig.
The total loss was and I quote: "one dB or so less than the loss specified
for the cable by the manufacturer". So, the run with connectors was
actually better than a single run of coax.



Connector test have been ran by many over the years. One way to weed
out the connectors is to put about 1 kw of power through them to a dummy
load and feel for heat after around 30 to 60 seconds.

While the coax connectors made it ok, I melted out an inexpensive 2
position antenna switch and their cheep connectors and the loose way
they were connected to the body of the switch. The antenna had a high
swr and I was using a tuner between the amp and the switch.

There was another quality switch before the inexpensive one and it took
the power just fine. It did not even get warm while the other was
melting out.