[Amps] every db lost re Tubes vs. Solid State

[Glen Zook]

I had a distributor who decided to start tuning all of the two-way radios that he sold instead of having my company tune them for the desired frequency.  He set up a test bench and tried some of the radios which we had tuned.  Out of a 35-watt output 450 MHz radio he was getting like 6 or 7 watts!  He was using a 50-ohm dummy load and a Bird 43 wattmeter with the proper element.  Therefore, I got a telephone call!

I took along both a Bird 6154 "dummy load" wattmeter and a Bird 4304 wattmeter.    With the 6154 wattmeter hooked directly into the units the power output from every one was more than 35-watts!  The distributor had his Bird 43 connected just in front of the dummy load.  So, I replaced his wattmeter with my Bird 4304.  I was getting readings in the 6 to 7-watt range!  Something was drastically wrong!

Then, I looked at the cabling between the radio and the wattmeter.  The cabling consisted of 3-sections of 50-ohm coaxial cable with PL-258 and PL-259 connectors joining the sections.  As each section of cable was removed, the power indicated on the wattmeter increased.  Then, I substituted a longer single section of cable between the radio and the wattmeter and the power was close to what was indicated on the Bird 6154.  Finally, the short length of cable between the wattmeter and the dummy load was replaced and then the power output read basically the same as on the Bird 6154.

As each section of the original cable was disconnected, the connectors were definitely warm.  There was a Radio Shack near the distributor's office and he had purchased the coaxial cable and the connectors from them.  After checking each PL-259 and PL-258 connector, it was discovered that each of these particular connectors had right at 1 dB of loss per connector at 450 MHz!  The insulating material was so lossy that a significant amount of power was being dissipated in each connector!  I had never run into a situation before where the losses in a single connector was anything like those particular ones.

Unfortunately for me, the distributor still had his own newly hired technician do the work.  But, he did go out  an purchase some "decent" coaxial cable and connectors instead of trying to "save a buck" by using those from Radio Shack.
 
Glen, K9STH


Website:  http://k9sth.com


________________________________
 From: Mike Waters <mikewate at gmail.com>
To: amps at contesting.com 
Sent: Monday, April 30, 2012 6:10 PM
Subject: Re: [Amps] every db lost re Tubes vs. Solid State
 
Actually, the loss from a "UHF" connector would be in the coax, not all
concentrated in the connector as I previously stated. I just got the
following private e-mail that nicely explained that:

"While I would concur that coax connectors, even the dated PL-259/SO-239s,
[UHF] are pretty good at HF frequencies, I think what "losses" they do
impose do not occur in the connector itself.  Rather, they create impedance
bumps, which in turn cause reflections, which result in a standing wave on
the transmission line.  Almost all the heating from this effect, therefore,
occurs in the coax, not the connector itself."

The heating due to the condition stated above would be distributed over a
longer length of coax instead of being concentrated right at the connector
itself.

But that's not what that one fellow (and all the others who have beat this
same drum over the years) implied. He didn't state a particular type of
connector, but rather lumped all connectors and switches together as all
being lossy. And that would include constant impedance connectors and
switches. Any heat from losses in a constant Z connector would be created
right at the connector itself.

In any case, it's been proven that the length of the impedance bump inside
the junction of a PL-259 and SO-239 is such a tiny fraction of a wavelength
below 200 MHz that losses from a mismatch is of no practical concern,
either for RX or TX.

73, Mike
www.w0btu.com

It wouldn't take hardly any loss at all for a connector, switch, or whatever to get warm fairly quickly when power was transferred through it. If it did have significant loss, you would burn your fingers on the connector at well below the 1500 watt level.