Gerald,
I have done this exact problem this many times for a high power ISM
application at 13.56MHz.
1.) For example put an "L" network between the amp and the exciter.
2.) Adjust the "L" network for a 1:1 match at the exciter at full amp power.
3.) Assume the amp impedance at full power is 40-j10 (but you don't know
this value)
4.) For a low pass "L" network with the shunt capacitor at the
transmitter side a match occurs with L=340nH and C=112.5pF (you do not
need to measure the component values)
5.) Now take the "L" network, reverse it, and measure the impedance with
any VNA with the "L" network terminated in 50+j0.
6.) You will measure 40+j10 with the analyzer.
7.) The amp input impedance is the complex conjugate of the measured
value or 40-j10.
The Smith Chart is not useless to solve this problem.
73,
Larry, W0QE
On 6/11/2017 3:12 PM, Gerald Williamson via Amps wrote:
Don, all you say is correct and good. BUT (seems there always is a but), as
a practical matter, the input R+j of an amplifier varies with drive level
and usual test equipment will not provide useful data. If you drive it with
a transmitter there is no way to extract R+j data at that power level.
Without good R+j data the smith chart is unusable for this problem.
73,
Gerald K5GW
In a message dated 6/11/2017 1:13:19 P.M. Central Daylight Time,
donroden@hiwaay.net writes:
"True SWR" doesn't change, but *WHERE* it is measured will vary widely.
exception : perfect Z load for perfect Z line where Zload=Zcoax
You are measuring VOLTAGE inside the coaxial line and the voltage
varies along the length of the line like a sine wave.
Some distances from the wrong Z load = peak voltage
Some distances from the wrong Z load = minimum voltage
Some distances make for a close-to-perfect input match.
1/4 wave away from the above perfect match is a point that
your transmitter probably won't like.
In other words, changing the coaxial length *COULD* improve
your situation, but it could also make it worse.
The Smith Chart is your friend, or just make five jumpers of
1,2,4,8,16 feet with a handful of couplers and start substituting
jumpers between the meter and the load. One combination will *MATCH*
or make a length that the transmitter likes. Others will make it
worse.
This is why it's important to measure the Z *AT* the load and not
through an arbitrary length of coax.
Don W4DNR
## On a related note, why is it, when I increase the length of 50
ohm coax from my drake amps to the xcvr, that the
input swr rises ?? Its flat with 4 ft or less.... but rises
when say 5-8 ft is used. Thats on all bands.
<REPLY My understanding is that's a false reading caused by current
flowing
on the outside of the coax braid. The true SWR does not change.
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|