Many so called SWR meters are rough approximations and their error
depends to a great degree on the local voltage and current on the line.
A good meter would use those to derive power, but the monimatch style
(copied for most varieties of 11m compatible SWR meters), does show
different results at different points of the line.
As you said, SWR meters tend to cause some mismatch. Bird recommends
adding a length of 50 ohm coax to their meter to make the total length
exactly a half wave. Then the impedance seen by the Bird is the same as
seen by the radio. The greater the sensitivity of the meter for working
at low power, the tighter the coupling required and the more the SWR
meter will change the impedance as seen through it. Even Birds will
introduce a SWR with the low power closely coupled slugs. And Bird uses
a 30 microamp meter to allow loose coupling. Bird also keeps the length
of the coupled line short to minimize wavelength effects.
Bird meters and monimatch type meters use the capacitive coupling of the
coupled line to couple voltage and magnetic coupling to the coupled line
to couple current. Then the diode sees the vector sum of the two
couplings.
In Bruene's original Collins wattmeter, he used a pair of variable
coupling capacitors for the voltage and an electrostatically shielded
toroid current transformer for the current. That way he kept down the
stray interaction. The current transformer secondary is terminated in a
low resistance. There is a variation on the circuit with only one
voltage divider. I don't believe it works nearly as well. Still there's
current and voltage coupling from the main line to the measuring
circuits and that can introduce some change of line impedance as seen
through the meter. On of his goals was to have a 30 dB directivity. E.g.
to have the reverse power indication be more than 30 dB down when
connected to a matched load. I have a prototype that I've used for
nearly 40 years with all kinds of transmitters that have been quite
satisfied with the load, but it was tossed out because the directivity
wasn't as good as desired. Anytime a SWR meter won't get really good
directivity on a fine load, it will be in error and if it does show a
minimum in reverse power, it can be for a somewhat mismatched load. Good
directivity requires considerable attention to shielding and balanced
construction where the forward and reverse sections are identical.
Internal radio SWR meters tend to be simplified, some actually are only
RF volt meters, connected to some point in the output low pass filter or
matching network, often considerable distance in lumped element line
sections from the antenna connection. To the point that some show a
reduced voltage for a high impedance load. I've seen some that shared
output for SWR/ALC protection and for metering and where the meter load
changed the ALC effects. There's been a tendency to leave out the
necessary shielding between a toroid and the line and so to make the
reading dependent on placement of the line conductor which varied from
radio to radio. Nearly as bad a product as the old CB monimatch style
SWR meter.
73, Jerry, K0CQ
--
Entire content copyright Dr. Gerald N. Johnson, electrical engineer.
Reproduction by permission only.
|