Hi Steve,
> I made several controversial statements yesterday. One was that the
> reflected energy is dissipated in the transmitter if the source
> impedance is 50 ohms.
The source must be a dissipative resistance, not a 50 ohm impedance. The
only case where that could be guaranteed is if the source drives the line
through a large value attenuator pad.
That is traditionally why, when I or others make lab measurements,
generators are operated through high attenuation pads.
That is the ONLY circumstance where mismatch loss would apply to a ham
transmitter, since they are virtually all non-linear devices whose
impedance changes with drive level and load impedance.
> antenna reflected wave is infinite. Again, some energy will be
> reflected if the VSWR seen by the antenna reflected wave is not 1:1
> and some will be absorbed and dissipated as heat OR some will be
> canceled (maybe a new concept for some).
Cancellation only occurs from re-reflection. The resistive component of
tuners is very small, and primarily confined to the inductors. Heating is
solely a function of the current in the inductor. Run a typical T network
tuner into a 50 ohm line terminated in a 50 ohm load, and it will get MUCH
hotter than into a 50 ohm half wave line terminated in 500 ohms!
Heat for a given tuner is a load impedance problem. Nothing more, nothing
less. It has nothing to do with line SWR past the tuner, except how that
SWR affects the impedance the tuner sees.
The same is true for the rig.
> My discussions and examples assumed a source with an impedance of 50
> ohms. I clearly stated this and many real transmitters and
amplifiers
> operate in this manner.
Nonsense.
Most amplifiers are less than class A, with source impedance changing with
load impedance changes. Not only that, most PA's appear as a
non-dissipative resistance. That "resistance" is a function of the
time-varying non-linear across and through vectors at the output device as
time averaged by the flywheel effect of the tank system and modified by
losses and feedback in the PA system.
Any amplifier that looks like a pure 50 ohm dissipative time and load
independent resistance would be an ideal class A PA stage with fixed
impedance matching and no RF feedback. Almost zero ham or commercial rigs
fit that category. I do have one PA that does. It is a class A 3CXP5000
used for IMD testing of PA's. I operate it at 200 watts maximum output, and
it perfectly terminates a hybrid coupler at any drive level.
> Ask yourself why transceiver manufacturers, even ones with built in
> tuners, specify a maximum VSWR load (like 3.0:1) and why the output
> power is lowered when operating into a high VSWR? If all of the
> antenna reflected power was reflected back to the antenna, why would
> it make a difference? And yes, some transceivers and tuners do burn
> up because of reflections.
As a designer of PA's, I can tell you why. Because of load impedance
variations.
When the PA is misterminated, voltages or currents can soar. Dissipation,
with incorrect termination (due to reflected power) can go up, down, or
remain the same. Power can go up, down, or remain the same. Voltages can go
up, down, or remain the same. Current can go up, down, or remain the same.
> thinks there are no reflections). Three things can happen to the
> energy at the tuner - some will be reflected back to the antenna
> because the reflected wave no longer see a matched condition, some
> will be dissipated as heat in the tuner and some can be canceled as
> described in the previous paragraph.
That's a far cry from your first statement, when you proclaimed all
reflected power was dissipated. The amount of power dissipated is dependent
ONLY on voltages and currents in the tuner, and dissipative resistances in
the tuner. A tuner can run cooler and safer into a 20 :1 SWR than 1:1 SWR
on a 50 ohm line. As a matter of fact, that is typically the case.
73 Tom
--
FAQ on WWW: http://www.contesting.com/towertalkfaq.html
Submissions: towertalk@contesting.com
Administrative requests: towertalk-REQUEST@contesting.com
Problems: owner-towertalk@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|