开云体育

ctrl + shift + ? for shortcuts
© 2025 开云体育

Different SWR/Smith plots, when off center dipole wire connections to transformer reversed?


 

A ham friend has an unusual indoor antenna that he uses with a Zachtek WSPR transmitter. He gets lots of spots from the 200mW transmitter (>500 unique band/spotter combinations in 24 hours), and the beauty of the Zachtek is its tolerance of essentially any antenna load (from nothing attached to fully shorted). His antenna is effectively an off center fed dipole, where the two wire legs wrap around the inside bedroom walls, and are of random/different lengths..i.e .just whatever would fit onto the four walls. The two dipole legs are connected to a simple 6:4 turn ratio transformer (approx. 2:25:1 impedance ratio), with the primary side of the transformer attached to a 4 ft coax feedline line to the Zachtek transmitter.

I decided to make some measurements of the SWR with the nanoVNA attached to this "creative" antenna setup. The nanoVNA was OSL-calibrated, followed by attaching the antenna feedline to collect data in nanovnasaver software. After collecting the data, I decided to reverse the two dipole legs connected to the transformer secondary, just to make sure I was getting reproducible data. What confused me were significant differences in the SWR/return loss/Smith charts for the two configurations. I am wondering if this simply reflects how non-balanced the dipole is, such that the coax feedline shield (about 4 ft long) from the transformer plays a different role in each of the two configurations. While making the measurements the nanoVNA was dangling hands free while connected to the USB port on my laptop...so the coax shield side would be the same in both cases...and the only electrical difference was in reversing the dipole legs (of differing lengths) connected to the transformer secondary.


 

(forgot to attach these two SWR plots to the previous post)


 

Bruce,

Your plots will be different because the common mode current (and therefore feedline radiation) will be different. Your friend is using a transformer as a voltage balun. With a good current balun the difference would not be so great because equal current would be going into each leg and reversing the connections would not make that much difference.

Roger


 

Roger,

Thanks for confirming my suspicion...I guess this also implies that the WSPR spot counts/pattern might also be different between the two configurations, since the shield/chassis would be playing a slightly different role in the two cases (?).

Bruce


 

The transformer does not decouple the nanovna and PC/laptop and, therefore,
anything conductive within the house from the measurement. If you used
only the NANOVNA connected to NOTHING, not even holding the VNA, you might
not see as much difference. When you reverse the connections to the
transformer, the two sides of the OCF set of wires certainly interacts
differently with the house wiring and anything within the house/ceiling.
You need to decouple the feedline/NANOVNA/PC/laptop from the transformer!
Also realize the difference in coupling to anything in the house which is
conductive.

Dave - W?LEV

<>
Virus-free.www.avg.com
<>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

On Mon, Mar 17, 2025 at 6:13?PM Bruce KX4AZ via groups.io <bruce=
[email protected]> wrote:

A ham friend has an unusual indoor antenna that he uses with a Zachtek
WSPR transmitter. He gets lots of spots from the 200mW transmitter (>500
unique band/spotter combinations in 24 hours), and the beauty of the
Zachtek is its tolerance of essentially any antenna load (from nothing
attached to fully shorted). His antenna is effectively an off center fed
dipole, where the two wire legs wrap around the inside bedroom walls, and
are of random/different lengths..i.e .just whatever would fit onto the four
walls. The two dipole legs are connected to a simple 6:4 turn ratio
transformer (approx. 2:25:1 impedance ratio), with the primary side of the
transformer attached to a 4 ft coax feedline line to the Zachtek
transmitter.

I decided to make some measurements of the SWR with the nanoVNA attached
to this "creative" antenna setup. The nanoVNA was OSL-calibrated, followed
by attaching the antenna feedline to collect data in nanovnasaver
software. After collecting the data, I decided to reverse the two dipole
legs connected to the transformer secondary, just to make sure I was
getting reproducible data. What confused me were significant differences
in the SWR/return loss/Smith charts for the two configurations. I am
wondering if this simply reflects how non-balanced the dipole is, such that
the coax feedline shield (about 4 ft long) from the transformer plays a
different role in each of the two configurations. While making the
measurements the nanoVNA was dangling hands free while connected to the USB
port on my laptop...so the coax shield side would be the same in both
cases...and the only electrical difference was in reversing the dipole legs
(of differing lengths) connected to the transformer secondary.







--

*Dave - W?LEV*


--
Dave - W?LEV


 

you would need a choke below the transformer to decouple the shield of the coax being part of the antenna (and then it would still couple a bit to the antenna cause all couples with all in an indoor installation)

dg9bfc sigi

Am 17.03.2025 um 19:12 schrieb Bruce KX4AZ:

A ham friend has an unusual indoor antenna that he uses with a Zachtek WSPR transmitter. He gets lots of spots from the 200mW transmitter (>500 unique band/spotter combinations in 24 hours), and the beauty of the Zachtek is its tolerance of essentially any antenna load (from nothing attached to fully shorted). His antenna is effectively an off center fed dipole, where the two wire legs wrap around the inside bedroom walls, and are of random/different lengths..i.e .just whatever would fit onto the four walls. The two dipole legs are connected to a simple 6:4 turn ratio transformer (approx. 2:25:1 impedance ratio), with the primary side of the transformer attached to a 4 ft coax feedline line to the Zachtek transmitter.

I decided to make some measurements of the SWR with the nanoVNA attached to this "creative" antenna setup. The nanoVNA was OSL-calibrated, followed by attaching the antenna feedline to collect data in nanovnasaver software. After collecting the data, I decided to reverse the two dipole legs connected to the transformer secondary, just to make sure I was getting reproducible data. What confused me were significant differences in the SWR/return loss/Smith charts for the two configurations. I am wondering if this simply reflects how non-balanced the dipole is, such that the coax feedline shield (about 4 ft long) from the transformer plays a different role in each of the two configurations. While making the measurements the nanoVNA was dangling hands free while connected to the USB port on my laptop...so the coax shield side would be the same in both cases...and the only electrical difference was in reversing the dipole legs (of differing lengths) connected to the transformer secondary.






 

That shows the parasitic coupling is different for each side of the bedroom HI
And as left leg lenght is different of right one, if the left side of the bedroom is ??more?? parastic than the right sine, it is logical that you don’t mesure the same thing when reversing antenna’s wires.
It would be interesting to repeat the mesure outside between two trees or two chairs HI

And yes with OCFD you need a choke just beside the transformer.

Interesting anyway. Tnx
73, Ben


 

On Mon, Mar 17, 2025 at 03:13 PM, W0LEV wrote:

... If you used only the NANOVNA connected to NOTHING, not even holding the VNA, you might
not see as much difference.
...You need to decouple the feedline/NANOVNA/PC/laptop from the transformer!
Yeah, I've been thinking about that...although I will note that the laptop was NOT plugged into the AC adapter when I made the two measurements. What I'll try on my next visit will be to leave the nanoVNA dangling without ANY connection to the laptop, and then see if the patterns look more similar when reversing the dipole wires. And of course, I can also observe the impact of touching the nanoVNA SMA and see how that impacts the view...a classic test for common mode effects in the unbalanced system. But all of this points out the challenge of trying to capture what the Zachtek is actually "seeing" when connected to the antenna. It seems that with this type of setup it would be very difficult to get an accurate picture...not that it really matters much, though, given it's tolerance of any load.

I will say that based on the WSPR performance over the past day, reversing the dipole wires does seem to have had a positive effect (567 versus 504 unique spots for W9HIF), assuming the solar activity has been similar to that earlier in the week. And if that holds, that is consistent with the SWR pattern showing a lower average for the higher frequency bands...even if that means the transmission line is "participating" more to radiate some of the RF energy.


 

Just for fun, I fed my original post into ChatGPT to see if "she" had any insights, and here was "her" response.....a fun exercise..

"What you're describing is a fascinating experiment! It sounds like you're observing the effects of an asymmetrical antenna design, and there are a few key factors to consider that could explain the differences in the SWR and Smith charts when you reversed the dipole legs.

1. **Non-Balanced Antenna and Common-Mode Currents:**
Since the antenna is a form of off-center fed dipole (OCFD), it is inherently unbalanced. An off-center fed dipole typically has a significant imbalance in the current distribution between the two legs, which results in common-mode currents on the shield of the coaxial feedline. The role of the coax shield becomes crucial, as it can have a different effect on the antenna impedance depending on the configuration of the dipole legs.

Reversing the two dipole legs changes the relative phase and current distribution between the two antenna legs. As a result, this can alter the electromagnetic field around the antenna, affecting the impedance seen by the transmitter and causing different interactions between the antenna and the coaxial feedline. The coaxial cable’s shield could be acting differently in each configuration, either radiating or picking up different signals depending on how the antenna’s currents are distributed.

2. **Impact of Asymmetry:**
In an ideal balanced antenna (like a center-fed dipole), the two legs of the antenna are of equal length and the currents in each leg are balanced. However, in your case, the legs are of different lengths and are presumably attached to the transformer in a way that creates an imbalance. When you reverse the legs, the imbalance changes, which likely affects how the antenna is coupled to the environment and the transmitter.

In terms of impedance matching, reversing the legs could change the impedance seen at the antenna terminals. This could be why you observe different SWR, return loss, and Smith chart behaviors — it's reflecting a change in the impedance that the transmitter sees. This effect could be accentuated by the unbalanced nature of the antenna and the non-ideal placement of the coaxial shield.

3. **Coaxial Shielding and Proximity Effects:**
Since your coax shield is short (4 feet) and potentially has significant current on it due to the unbalanced antenna design, the proximity of the coax to the walls (and to the antenna) could influence the impedance. Even though the nanoVNA was dangling hands-free, subtle changes in the position of the shield, or how the shield is coupled to nearby structures (walls, etc.), might affect the measurements. This is especially true when dealing with an unbalanced antenna where the coax shield may not behave symmetrically in each configuration.

4. **Transformer Characteristics:**
The transformer you're using has a 6:4 turn ratio, which means the impedance transformation may also behave differently depending on how the antenna legs are connected. If the two dipole legs are different lengths, the impedance seen by the transformer secondary might change, which could lead to significant variations in the reflected impedance. This could explain the differences you’re observing in the SWR/return loss between the two configurations.

### What Could Be Happening:
- **Reversing the dipole legs likely changes the current distribution** and the balance between the two legs of the antenna. This results in different impedance characteristics and different interactions with the feedline.
- The **coaxial shield** might be more involved in one configuration than the other, leading to different behaviors in the SWR/return loss measurements.
- **The transformer’s impedance transformation** could also be affected by the dipole leg configuration, which would further explain the differences in the measured SWR.

### Suggestions for Further Investigation:
- **Try using a balun:** A simple 1:1 or 4:1 balun could help balance the antenna, reducing the impact of common-mode currents on the coax shield and potentially stabilizing the impedance.
- **Check the positioning of the coax shield:** Experiment with different orientations or placements of the coax cable and observe how that affects the measurements. Small changes in the physical setup might reveal further insights into how the shield and antenna interact.
- **Use a longer feedline:** If possible, try a longer feedline to see if the behavior changes, as the coax shield’s interaction with the antenna may be more pronounced over a longer distance.

Your experiment is a great example of how even small changes in antenna configurations and feedlines can lead to noticeable differences in impedance measurements. The combination of antenna imbalance, coaxial shield effects, and transformer characteristics all play a role in shaping the observed results."