I'm new to the device.

The goal is to measure the impedance an antenna driven by balanced line (here window line) without a balun.

At the moment connecting one side of the line to center of Port 1 and the other side to device ground sounds reasonable if the VNA itself is unconnected to rf ground.

There's a way to set the Smith chart reference impedance to the hypothetical 420 ohm line, I think.


Thanks,

Chris,
NQ8Z

Mini-Circuits makes quite a variety of small impedance transformers.I use the 50-75's all the time.

Just pick one that has the right impedance ratio.
In your case you want to bring 420 Ohms to 50 Ohms
This is pretty close to 9 to 1.   so pick at transformer with a 9 to 1 ratio and the frequency range of your antenna.  .For 300 Ohm line you use a 6 to 1 to get back to 50 Ohms.    Now before you get out your scientific calculator and say it is really 8.40 to one, have a close look in the part's data sheet.   9 to 1 is sort of an average.And if 'Close' is not good enough, a few resistors can tweak the impedance.Good luck with your experiments.  Kent WA5VJB

Owen Duffy uses a NOELEC balun board for measuring balanced line.
https://owenduffy.net/blog/?p=32273

73 de va3rr

he goal is to measure the impedance an antenna driven by balanced line (here window line) without a balun.

What type of antenna are you using that requires grounding?



--

F1Amm

François

On Mon, Jul 15, 2024 at 06:07 AM, <mail@...> wrote:

The goal is to measure the impedance an antenna

If your goal is to measure the impedance of the antenna itself, rather than the impedance at the near end of the balanced line, you will need to measure at a point in the line that is nx1/2 wavelength from your antenna feedpoint at the desired frequency. The complex impedance will vary along the line (go in circles on a smith chart), and only match the antenna impedance at multiples of 1/2 wavelength. (The magnitude of the impedance, swr and return loss will not change much along the length of the feedline - just a bit due to feedline losses).

Yes, using the display->portZ menu you can set the nanovna to use 420 ohms as the characteristic impedance (Z0) in its displays/calculations - i.e. the center of the smith chart will now be 420 ohms, and the impedances calculated correctly for that Z0. This doesn't change the hardware impedance of the S11 port itself - for that, you need a transformer as others have suggested, or you can just use it without a transformer and the nanovna will still give you useful information. You can search the messages of this group for prior threads on this topic from a year or so ago.

QUOTE: The goal is to measure the impedance an antenna driven by balanced
line (here window line) without a balun.

At the moment connecting one side of the line to
center of Port 1 and the other side to device
ground sounds reasonable if the VNA itself is
unconnected to rf ground.

For HF frequencies, the NanoVNA is small enough that eliminating the balun
should not present a bias in measurement of the impedance at the end of the
windowline. However, the NanoVNA must not be connected to ANYTHING and
supported above earth from something like a reasonably sized cardboard box
or styrofoam. Do not hold the VNA while making the measurement.

One side of the windowline should be connected to the center pin of the
source port (you're doing a reflection measurement). The other side of the
windowline should be connected to the backshell of the same connector.

There's a way to set the Smith chart reference
impedance to the hypothetical 420 ohm line, I think.

The measured impedance will not be that of the transmission line alone!
The transmission line modifies the impedance seen at the feedpoint. This
modification of impedance depends on the length of the, the Vp of the line,
and other characteristics of the line. Unless you calibrate out the
influence of the line, what you measure is the impedance at the feedoint
modified by the effects of the transmission line. Therefore, there is no
need or reason to set the Smith Chart center to the line impedance,
420-ohms.

Dave - WØLEV



<http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Virus-free.www.avg.com
<http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

On Mon, Jul 15, 2024 at 1:08 PM mail via groups.io <mail=
cvkimball.com@groups.io> wrote:

I'm new to the device.

The goal is to measure the impedance an antenna driven by balanced line
(here window line) without a balun.

At the moment connecting one side of the line to center of Port 1 and the
other side to device ground sounds reasonable if the VNA itself is
unconnected to rf ground.

There's a way to set the Smith chart reference impedance to the
hypothetical 420 ohm line, I think.


Thanks,

Chris,
NQ8Z

--

*Dave - WØLEV*


--
Dave - WØLEV

Thanks for the useful replies!

I neglected to say that this is for a specialized fan doublet antenna operating below 15 MHz. In actual operation a traditional 1:1 current balun is in place prior to the connection with the radio. The impedance output of the balun is unbalanced and presumably easy to measure.

Both balanced line wires have a very high DC resistance to ground; they're shorted to ground when not operating for lightning protection. At the antenna end of the transmission line a 50K ohm resistor is across the lines for DC continuity testing.

Chris

Hello

I have developed with my NanoVNA multi-doublet antennas and even a double-doublet with traps (17-15-12-10 m). Whether you subsequently use a line of parallel wires or not, the important thing is to measure the impedance according to the frequency of your antenna alone, in position.

The final antenna descent of my multi-doublets is a coaxial cable (between 10 and 30 m). The cable is in place and it is grounded flush with the station. I take the measurement from the lower end of the cable having calibrated at the upper level of the cable with my NanoVNA connected to the PC on which nanovan-saver is running.

The measurements are not a problem and they are correct because, then, I correct the impedance of the antenna at the center of the multidoublet by discrete elements (capacitor coils) whose values ​​I calculate from the impedance measured by the NanoVNA. Once the discrete components are in place (and measured), what I see with the NanoVNA is consistent with my calculations.

** So the initial measurement is good. **

So:
• No need for a balun
• No need to isolate the NanaoVNA
• You can use nanovna-saver at the bottom, it will save you from tearing your eyes.

But, in your case, you need a coaxial that goes up, at least for the measurement, to the center of the antenna. You could integrate the end of the parallel line during the calibration, with 50 Ω , but, I don't have the experience. I nevertheless checked that it worked with 93 Ω coaxial.

I see no point in carrying out the descent of a multidoublet, almost bi-resonant, with a two-wire line while it works very well with coaxial, without balun but that is another story than the measurement of the antenna.

On which bands do you want to operate your multidoublet?
--
73
François

I mis-spoke: The impedance at the end of the transmission line, not the actual antenna impedance at the junction between the antenna and antenna line, is what I'm looking for. Given that the window line is short, ~80 ft, and lossless, once I have the impedance at the end it can be rotated ("towards load") on the Smith chart to get the impedance at the antenna at a particular frequency.

Might try TDR mode to get the actual length of the window line.

I'm hoping to run on 160, 80, 60, 40, 30, and 20 m bands. All at low power, 40W.

Chris

If your design is to use this as an 80 foot two-wire line, then measuring the impedance at the center of the multidoublet is, in fact, not of much interest.

On the other hand, if, to develop the lengths of your multidoublet, in order to match the various resonances, already the agreement of each strand reacts on the others, the fact of measuring through the bifilar line, will go a long way complicate interpretation. Let me explain :

It may be that a resonance (antenna + line) appearing too low in frequency will be even lower if you shorten a strand.

Either way, measuring through the caoxail, calibrating on the low end of your two-wire line, without a balun, will work just fine. Interpreting the results to correct the antenna will be more complicated.
--
F1AMM
François