At 12:36 PM 2000-01-25 PST, Jeff Wolf wrote:
>
>The answer to this problem is unrealistically simple...
>
>Their is reactance on the line. When changing the length of the coax
>changes the SWR on the line you have proven this point.
>
>Either the input to your amp has a reactive input, or the output of your
>transceivers is not matching correctly to 50 ohms...
Only the former. If the cable is matched to the load (amplifier input
impedance), the impedance at the radio-end of the cable won't change due to
cable length, and it doesn't matter what impedance the radio's output is.
The only effect then will be the efficiency of power transfer from the
radio to the load impedance presented by the matched cable-amplifier combo.
In this case, the radio's internal tuner would be appropriate and useful.
>I would suspect the input of the amplifier being the real culprit,
>however, you can check this by taking the output of the exciters and putting
>them into a real (Non-reactive) 50 ohm dummy load and measuring forward and
>reflected power.... If they both measure 0 watts reflected, the
>transmitters are fine.
Always include the cables to be used, too, as it's easy to forget sometimes
and put in that beautiful hunka stuff you picked up at last year's swap
meet for a dime but didn't realize is actually 75 ohms RG-11 as there may
not be any marking on it. Or worse yet, 93-ohm stuff...
Also, "the transmitters are fine" statement is true presuming the harmonic
emission of the radios is sufficiently low as to not be contributing to the
measured reflected power from the amplifier. Remember that a return loss of
20 dB presented by the amplifier input, at forward power of 100 watts,
means 1 watt reflected back to the radio, which can be seen on almost any
power meter. 20 dB return loss is a pretty good match.
But reflected power due to exciter-generated harmonics isn't a practical
consideration unless one is using, for example, a solid-state brick with no
internal harmonic filtering which may generate 2nd harmonic levels as high
as 20 dB down at saturation. In that case, 1 watt at 56 MHz would reflect
back from the 28 MHz amplifier input circuit and be seen on the power
meter. But the exciters being used here should have harmonic levels better
than -60 dBc which will be impossible to read with a 100-watt element in a
Bird 43.
So simply because you read no reflected power when driving a good load
doesn't entirely clear the exciter since the load also absorbs, and thus
prevents reflection of, harmonic energy.
>Also, you should be measuring power input to the amp at the input SO239
>connector... Using a double PL / PL connector. If you are not, then their
>is the chance that you are not seeing the "true" input power to the amp,
>especially if the input to the amp is reactive.
Not a "chance", but a fact. However, if the load impedance is not close to
50 ohms, then the actual power reading will not be accurate because the
detected or sensed voltage on the line will be different due to the
abnormal load impedance. How far off it will read depends upon the
frequency of operation, the load VSWR, and the frequency response of the
element itself, not to mention even the short coaxial adapter between the
meter and amplifier input connector.
>Finally, your exciters are showing different power output levels for input
>power levels because of the inability to deal with the reactive loads
>presented.
Not necessarily the "inability", but more likely different power transfer
efficiencies between each exciter and the abnormal cable-amplifier impedance.
>Remember that a piece of coax WILL act as a transformer, and the inability
>of your solid state exciters to deal with a reactive load is what is causing
>this phenomenon.
>
>Just a note from a lowly CBer...
Well, based on your response, you must be one of the louder and more
efficient ones, then :o))))
73, Steve K0XP
--
FAQ on WWW: http://www.contesting.com/ampsfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|