> If all the reflected power was absorbed by the transmitter PA then there
should
> be no increased forward power reading on the bird meter whether there is
> reflected power or not?
I ran some tests this morning using my Ten-Tec Omni Six and Kenwood
TS-850/SAT with auto-tuner. I used my Bird 43 wattmeter as the measuring
device at the source end into 100-feet of LMR-400 and terminated with a
20-meter half-wave diplole which was removed to create an open termination
during testing.
First test:
I used my Omni Six to establish that the antenna and line were working
properly. I removed the antenna from the line and replaced it with a
combination 50-ohm Bird Termaline load and wattmeter (Model 6154).
Transmitting with the Omni Six at 100-watts as read on the Bird 43, the
measured power at the line end read just under 100-watts. Reflected power
on the Bird 43 measured zero watts. The lower reading is due to a
combination of slight loss in the LMR line and the fact that the Bird 6154
is only accurate to within +/- 10-percent or so.
Next, I removed the load and transmitted into an open termination. The Bird
43 measured 100-watts forward power and just under 100-watts reflected (the
difference being attributed to power absorbed in the line due to line loss).
I am assuming that since the Omni Six, having a reasonably close 50-ohm
resistive source, and feeding a 50-ohm characteristic impedance line, that
all power is being absorbed by the Omni's PA section, less the line loss
power absorption. Where else is the power going? As the forward wave
reaches the unterminated end of the line, a reflected wave is produced, but
exactly 180-degrees out of phase with the forward wave. The phase reversal
is always 180-degrees regardless of the degree of mismatch. The amplitude
of the reflected wave IS a function of the degree of mismatch.
As the reflected wave travels back to the Omni Six, it too is travelling on
a 50-ohm characteristic-impedance line and is terminated into the Omni's
50-ohm source. The reflected wave sees no mismatch and is completely
absorbed by the Omni's PA.
Second Test:
I replaced the Omni with my Kenwood TS-850/SAT. I disengaged the the
auto-tuner. Using the same procedure as with the Omni Six, my Bird 43
measured 25-watts forward power and just under 25-watts reflected power
(Again, the difference being attributed to line losses). Unlike the Omni,
the TS-850 incorporates a SWR foldback/protection circuit to deliberately
limit the output to a fraction of 100-watts. The degree of power foldback
is a function of the the mismatch as seen by the transmitter's SWR circuit.
Third Test:
I engaged the auto-tuner on the TS-850 which was pre-adjusted for
20-meters. Since, the tuner will not tune into an open termination, I used
the saved tuner value. I have no idea what complex network setting the
tuner was saved as. The Bird 43 measured 10-watts forward and just under
ten watts reflected. Again, without a load at the terminating end, the
forward and reflected reading measure approximately the same. I tried this
on several bands and the results are the same.
Fourth Test:
I re-connected the 20-meter dipole and tuned the TS-850 off the dipole's
resonance frequency to produce a 3:1 SWR reading. With the auto-tuner
disengaged, I read 35-watts forward power and approximately 10-watts
reflected power.
Next, I engaged the TS-850's auto-tuner and did not touch the power control.
Now, the Bird 43 reads just under 45-watts of forward power and reflected
power satyed the same at 10-watts. However, the output power on the TS-850
still reads 35-watts. I suspect this is because the TS-850's SWR circuit is
not a true directional coupler.
My conclusions:
1) As expected, the Bird's directional coupler adds the amount of reflected
power to the forward power reading. Caveat: the addition is only true up to
some limit of mismatch. Into an open termination, no power is being
absorbed by a load. At this extreme, reflected power does not add to the
forward power reading.
2) The TS-850's power indicator is not based on a true directional coupler
and hence the forward power indicator never adds the amount of refelcted
power to its readings.
3) A 50-ohm resistive source, fed into a 50-ohm characteristic impedance
line with an open termination, yields total absorption of the reflected wave
into the source and hence all reflected power, less power absorbed in the
line due to line loss. The Bird reads 100-watts forward power and close to
100-watts reflected power. Again. the difference being power lost in the
line due to line loss.
4) Here's where things get complicated: A mismatched impedance from the
source to the line (e.g., auto-tuner, ATU, pi-network), and transmitting
into an open termination creates some intersting results. Power cannot be
absorbed by an open load; there is none. Power can be absorbed by the line
to the degree that the line has loss. If the reflected wave sees a complex
impedance back at the source, the wave sees this as another mismatch,
reverses itself precisely 180-degrees and travels back to the open
termination. The amplitude of this re-reflection is also governed by the
degree of mismatch at the source. Again, in this example, I have to
conclude that 100 percent of the reflected power is absorbed at the
transmitter (less line loss), despite the attempt by the transmitter's
complex network (again, the Pi, ATU, or auto-tuner) to redirect the wave
back to the unterminated load.
-Paul, W9AC
|