What happens when two paralleled transmission lines are of different
lengths? How does one analyze such a problem? For instance, what
impedance is seen at the source when a 30 foot run of 450 ohm ladder-
line is paralleled with a 20 foot run of the same line when the common
load is 100+j200 ohms? Will EZNEC handle such a problem? What if the
two different length paralleled transmission lines are not the same VF
and/or not the same Z0? Does superposition work for such a problem?
--
tnx & 73, Cecil, w5dxp.com

Well, my first impulse is to actually hang the two lines and measure
the resultant impedance at the source end...
I am assuming the lines are spread apart so that they are not closely
coupled...
One approach to get you in the ballpark would be to model the 30
footer alone to get a source impedance, then model the 20 footer
likewise alone, then model the resultant of those two complex
impedance in parallel...
An interesting question...

denny / k8do

On Apr 20, 12:37*pm, Cecil Moore wrote:
What happens when two paralleled transmission lines are of different
lengths? How does one analyze such a problem? For instance, what
impedance is seen at the source when a 30 foot run of 450 ohm ladder-
line is paralleled with a 20 foot run of the same line when the common
load is 100+j200 ohms? Will EZNEC handle such a problem? What if the
two different length paralleled transmission lines are not the same VF
and/or not the same Z0? Does superposition work for such a problem?
--
tnx & 73, Cecil, w5dxp.com

I'm puzzled, Cecil. I wouldn't think a simple linear system like this
would be difficult for you to analyze. (Or perhaps you do know how,
and are only testing to see if others do also.) With respect to
R.R.A.A., under what circumstances would such a configuration ever
arise in practice? And...why would you think to use an antenna
modeling program to analyze a linear circuits problem? Why not pick a
tool appropriate to the task?

Cheers,
Tom

On Apr 21, 12:43*pm, K7ITM wrote:
I'm puzzled, Cecil. *I wouldn't think a simple linear system like this
would be difficult for you to analyze. *(Or perhaps you do know how,
and are only testing to see if others do also.)

Nope, not a trick question. The thought occurred to me after a liter
of Merlot. It is akin to the analysis of a fan dipole except shorted
at both ends. If 30 feet is not the optimum length for the matching
section of a G5RV used on 30m, what if we paralleled the non-optimum
length section with the optimum length section? What happens then?

It appears that what happens is that we lower the resistance while
maintaining the reactance at a high level - not a good thing. File it
under " unfortunate wet dreams".
--
73, Cecil, w5dxp.com

Cecil Moore wrote in news:d6ea0bf4-7bb4-4089-9c88-
:

What happens when two paralleled transmission lines are of different
lengths? How does one analyze such a problem? For instance, what
impedance is seen at the source when a 30 foot run of 450 ohm ladder-
line is paralleled with a 20 foot run of the same line when the common
load is 100+j200 ohms? Will EZNEC handle such a problem? What if the
two different length paralleled transmission lines are not the same VF
and/or not the same Z0? Does superposition work for such a problem?

Here is an opportunity for you to do some leg work Cecil.

I have modelled in RFSIM99 a load of 100 ohms in series with 9.1µH at
3.5MHz with 30' of lossless 400 ohm line, vf=0.9, and 20' of lossless 450
ohm line, vf=0.8 as you describe. The input impedance is 767+j296.

Do you get the same result with a valid model in EZNEC?

Owen

On Apr 22, 1:35*pm, Owen Duffy wrote:
Cecil Moore wrote in news:d6ea0bf4-7bb4-4089-9c88-
:

What happens when two paralleled transmission lines are of different
lengths? How does one analyze such a problem? For instance, what
impedance is seen at the source when a 30 foot run of 450 ohm ladder-
line is paralleled with a 20 foot run of the same line when the common
load is 100+j200 ohms? Will EZNEC handle such a problem? What if the
two different length paralleled transmission lines are not the same VF
and/or not the same Z0? Does superposition work for such a problem?

Here is an opportunity for you to do some leg work Cecil.

I have modelled in RFSIM99 a load of 100 ohms in series with 9.1µH at
3.5MHz with 30' of lossless 400 ohm line, vf=0.9, and 20' of lossless 450
ohm line, vf=0.8 as you describe. The input impedance is 767+j296.

Do you get the same result with a valid model in EZNEC?

Owen

Remarkably, Owen, I get exactly the same result using LTSpice.

Spice and RFSim99 are the two that came to mind when I wrote about
more appropriate modeling tools yesterday... It might be fun to find
the general solution to the set of linear equations relating the
forward and reverse voltage and current at each end of each line and
in the load, for a given excitation, to end up with a relatively
simple (?) formula for translating the load impedance to the impedance
seen at the input end of the lines, and maybe that already exists in a
text or reference book somewhere, but there are a few too many
equations in the set to get me interested in doing it by hand.

It's also interesting that in both this example and in another
different one I tried yesterday, I found that, at some frequency
(16.993MHz for Owen's example), the impedance at the input goes to
infinity. That is to say, there is no current in the load: the load
is at a point along the loop of two lines that is at a zero-voltage
point of the standing wave. It shouldn't be a surprise that this
happens, but you might not think about it when just blindly connecting
two lines of different lengths "in parallel," and end up getting
bitten by it.

Cheers,
Tom

K7ITM wrote in news:6dcca41e-e4d9-4cd7-bde3-
:

....
Remarkably, Owen, I get exactly the same result using LTSpice.
....

Well Tom, perhaps it is that "great minds think alike".

Cecil's question on eHam and then here specifically asked if EZNEC can
solved it, but mused that it needed information about the reflection at
(presumably a source side) ATU. Of course, the source side circuitry is
irrelevant to a steady state solution of the input impedance that we both
calculated.

I am not sure that it has any specific application, but the rat race
hybrid is an expanded problem of the same type.

As far as I can see, NEC must have the ability to solve networks of
elements that can be represented as NT cards, just that the GW elements
that you use to hook it all together are subject to mutual coupling
(which is appropriate if it were a real feedline / antenna scenario).

It took me less time to build and check the RFSIM99 model than it took
Cecil to write his post. It would take only a little longer in NEC.

Owen

On Apr 22, 8:43*pm, Owen Duffy wrote:
Cecil's question on eHam and then here specifically asked if EZNEC can
solved it, but mused that it needed information about the reflection at
(presumably a source side) ATU. Of course, the source side circuitry is
irrelevant to a steady state solution of the input impedance that we both
calculated.

It appears that you didn't understand the purpose of my original
posting. Days before, I had already run the EZNEC impedance results
but was questioning the validity of those impedance values. Tuner
settings are certainly not irrelevant when calculating system
efficiency which was my main goal. The musing was whether the results
of two different tuner settings could be superposed.

It took me less time to build and check the RFSIM99 model than it took
Cecil to write his post. It would take only a little longer in NEC.

Days before my original posting using EZNEC, I had already paralleled
a 20 foot section of ladder-line with the standard 30 foot section on
a G5RV but I had no quick and easy way of independently verifying
EZNEC's impedance results. It was only after a lot of effort that I
still had questions. Thanks for the information - if not for jumping
to false conclusions based on ignorance.
--
73, Cecil, w5dxp.com