[TowerTalk] Antenna analyzers and parallel conductor transmissionlines

[Steve]

Hi Jim,

Thanks for the insight.

Any idea of what degree of inaccuracy might be introduced by the
parasitic asymetric capacitance in the HF range if one were to connect
ladder line to an analyzer's SO-239?

With respect to the "extremes of resistance and reactance" mentioned
in my note, I was particularly interested in how these devices perform 
when analyzing dipoles which are electrically short. Using EZNEC and 
TLW, I find instances when the resistive component is in the 5-10 ohm 
range and the reactance is in the 1000 ohm range. I was wondering if the 
the units in the market place are all about the same or if one stands 
out from the herd with respect to measuring these "extremes"?

Thanks,

Steve K8JQ

Jim Lux wrote:
> ----- Original Message -----
> From: "Steve" <steveac at charter.net>
> To: <towertalk at contesting.com>
> Sent: Wednesday, February 02, 2005 5:41 PM
> Subject: [TowerTalk] Antenna analyzers and parallel conductor
> transmissionlines
> 
> 
> 
>>The antenna analyzers one finds marketed to hams all seem to have an
>>SO-239 connector.
> 
> The top of the line MFJ (the one that does UHF) has an N.
> 
> 
>>Can one use these devices with antennas fed with parallel conductor
>>transmission lines such as open wire or ladder line?
> 
> 
> Oh Ho.... yes, but...  Theoretically, you need a well characterized
> transformer/balun, but, of course, the ham analyzers don't have any means to
> input "adapter coefficients". (I suppose that's one of the things that
> differentiates the $50K Agilent PNA from the $300 MFJ).
> 
> 
> 
>>If so, how do you connect the parallel conductor line to the antenna
>>analyzer? One wire to the center pin and the other wire to the threaded
>>body of the SO-239?
> 
> 
> You could do this, and as long as you don't touch the analyzer and keep it
> away from surroundings, (the analyzer is totally floating, so the
> balanced/unbalanced thing isn't really an issue) the readings will be
> correct, except:
> 1) The analyzer is optimized for 50 ohms.. 50 ohm open wire line is somewhat
> unusual.
> 2) It's really an unbalanced measurement device, so there's parasitic C from
> the case of the analyzer which will be asymmetric to the line you're
> measuring.
> 
> So... what you need is a 50:450 ohm or a 50:300 ohm transformer with decent
> properties (low loss, not much in the way of parasitics, etc.).  This would
> be quite the challenge if you were interested in really wideband
> measurements to a gnat's eyelash.
> Transformers are your friend... and since they don't have to handle high
> power, they're not that tough to build
> 
> 50:450 is 1:9 (a 1:3 turns ratio)
> 50:300 is 1:6 (not an even turns ratio.... but, 1:4 (a 1:2 turns ratio)
> would transform 300 to 75, which is probably reasonable for an analyzer...
> assuming you can tell the analyzer that it should use 75 ohms as the base)
> 
> What you would want is some good dummy loads (at the line impedance) to
> check things with.  You slap a 450 ohm load on the transformer and verify
> that your analyzer actually reads 50 ohms. You put a suitable L or C on the
> transformer (pick something with an X of several hundred ohms at the
> frequency of interest) and see that it reads what it's supposed to.
> 
> 
>>Do you lose any antenna analyzer functionality when using it with
>>parallel conductor transmission line as compared to coaxial transmission
>>line?
>>
>>Are any particular antenna analyzers better suited to parallel conductor
>>transmission lines?
> 
> 
>>Are any particular antenna analyzers favored for measuring the extremes
>>of resistance and reactance (and particularly correctly indicating the
>>sign of the reactance) that one encounters at the shack end of a 100ft
>>dipole fed with parallel conductor transmission line operated and in the
>>160 through 10 meter amateur bands?
> 
> 
> Sign of reactance is a function that either the analyzer has, or it doesn't.
> If it does it for coax, it does it for everything.
> 
> When you say "extremes" what do you mean?  10kohms? 1 ohm? or 200-600 ohms
> on a 450 ohm line.
> 
> What sort of accuracy are you looking for?  1%, 10%, "transmitter doesn't
> catch fire"
> 
> As you can imagine, impedance measurement can be either simple or complex.
> 
> 
>