|
Looking at the PHASE data on a single band 80m center fed dipole.
See the two attachments: one without antenna tuner, one with an antenna tunder
I've had so much fun and have learned so much about antenna parameters with this little Nano VNA device.
One of the parameters that I was not familiar with is to look at the PHASE data, say on an 80m dipole: a 130 ft long center-fed dipole with about 150 feet of low-loss RG213 50-ohm coax.
The antenna on 80m had a natural resonance around 3580 kHz (where I like to operate digi modes on NBEMS nets).
Remember at resonance, the reactance is zero (regardless of the SWR). The PHASE is zero degrees.
But that antenna had an awful SWR at the top of the band, where the phone nets play around 3990 kHz.
At 3990 kHz, the SWR was more like 4:1 and my rig was not happy.
So, how do I QSY to the top of the band with decent SWR without cutting 10 feet off my dipole?
Well, with an antenna "tuner" in the shack, I can null out the reactance of the antenna at 3990 kHz, and get an almost 1:1 SWR on the coax going to the transmitter.
My rig is "happy" at 3990 kHz with that low SWR. But the SWR is still 4:1 on my feed line.
The SWR is only reduced on the coax between the tuner and the rig.
How would a PHASE plot look for my antenna AFTER the RF passes thru my antenna tuner (T network)?
Well, a PHASE plot shows me the original resonance of the 80m dipole around 3600 kHz (without the tuner) and the SWR and RETURN LOSS graphs verify this as well. Now, post antenna tuner, matched to the TOP of the 75m band, there is a second NEW resonance at around 3900 kHz post antenna tuner. Thus, the tuner RE-REFLECTS the reflected wave at 3900 kHz back in the direction of the antenna. Something W2DU Walt Maxwell claimed was how it worked.
Fascinating stuff. The tuner (in my shack) provides a "conjugate" impedance to offset the reactance at the feed line/antenna boundary. Thus an impedance of R + jX (in my shack) combines with an impedance of R - jX in the tuner ==> resulting in an impedance of R + j0 for the radio, with R about 50 ohms. Beautiful.
And, my low-loss RG213 feed line at 4 MHz, even with a 4:1 SWR, has less than a 1 dB "loss" due to the 4:1 SWR.
A 1 dB loss is barely noticeable.
So why worry about broad-band 80m antennas (gazillion articles in QST)?
Just put a decent tuner in your shack and QSY where you want with very little attenuation in low-loss cable (on 80m).
There is also very little "loss" in the tuner itself with decent size inductors and capacitors.
So cool to see all of this concept verified on the NANO VNA PHASE plot using VNA SAVER software (Win 10).
de Barry k3eui
Phila
|
Hallo Barry,
looking at the phase plot you see the phase of the complex reflection faktor and not the phase of impedance Z. Having a perfect match at a certain frequency, the reflection factor is going to 0 and the phase of it can not be determind. As you can see from smith chart, the curve, coming from lower frequencies, hit the center of the chart nearly in a right angel to the real axis. In the phase chart you can see this behavior in a jump of phase from -180¡ã to +180¡ã.
73, Guenter, dk5dn
Am 13.12.2020 um 15:45 schrieb Barry Feierman:
toggle quoted message
Show quoted text
Looking at the PHASE data on a single band 80m center fed dipole.
See the two attachments: one without antenna tuner, one with an antenna tunder
I've had so much fun and have learned so much about antenna parameters with this little Nano VNA device.
One of the parameters that I was not familiar with is to look at the PHASE data, say on an 80m dipole: a 130 ft long center-fed dipole with about 150 feet of low-loss RG213 50-ohm coax.
The antenna on 80m had a natural resonance around 3580 kHz (where I like to operate digi modes on NBEMS nets).
Remember at resonance, the reactance is zero (regardless of the SWR). The PHASE is zero degrees.
But that antenna had an awful SWR at the top of the band, where the phone nets play around 3990 kHz.
At 3990 kHz, the SWR was more like 4:1 and my rig was not happy.
So, how do I QSY to the top of the band with decent SWR without cutting 10 feet off my dipole?
Well, with an antenna "tuner" in the shack, I can null out the reactance of the antenna at 3990 kHz, and get an almost 1:1 SWR on the coax going to the transmitter.
My rig is "happy" at 3990 kHz with that low SWR. But the SWR is still 4:1 on my feed line.
The SWR is only reduced on the coax between the tuner and the rig.
How would a PHASE plot look for my antenna AFTER the RF passes thru my antenna tuner (T network)?
Well, a PHASE plot shows me the original resonance of the 80m dipole around 3600 kHz (without the tuner) and the SWR and RETURN LOSS graphs verify this as well. Now, post antenna tuner, matched to the TOP of the 75m band, there is a second NEW resonance at around 3900 kHz post antenna tuner. Thus, the tuner RE-REFLECTS the reflected wave at 3900 kHz back in the direction of the antenna. Something W2DU Walt Maxwell claimed was how it worked.
Fascinating stuff. The tuner (in my shack) provides a "conjugate" impedance to offset the reactance at the feed line/antenna boundary. Thus an impedance of R + jX (in my shack) combines with an impedance of R - jX in the tuner ==> resulting in an impedance of R + j0 for the radio, with R about 50 ohms. Beautiful.
And, my low-loss RG213 feed line at 4 MHz, even with a 4:1 SWR, has less than a 1 dB "loss" due to the 4:1 SWR.
A 1 dB loss is barely noticeable.
So why worry about broad-band 80m antennas (gazillion articles in QST)?
Just put a decent tuner in your shack and QSY where you want with very little attenuation in low-loss cable (on 80m).
There is also very little "loss" in the tuner itself with decent size inductors and capacitors.
So cool to see all of this concept verified on the NANO VNA PHASE plot using VNA SAVER software (Win 10).
de Barry k3eui
Phila
|
Barry, Just for emphasis, the Smith Chart is a plot of the reflection coefficient which is the RATIO of the voltage across the transmission line due to the reflected signal to the voltage across the transmission line due to the signal (output of the transmitter) heading up to the antenna ( rho = Vref/Vfwd). The phase angle is the phase between these two voltages at a given point on the transmission line. For impedance, R +X, these are simply overlaid on the chart corresponding to the reflection coefficient at each point. The phase angle for the impedance is the arctan(X/R) and it represents the phase angle between the voltage and the current at each point along the transmission line.
|
Hi Barry!
If you want to operate without a tuner, but with resonance at both 3.589MHz and 3.990MHz, you could add an additional dipole (with resonance at 3.990MHz) 284mm apart from your "3.580MHz dipole". Just use som spreathers and without any connection between the two elements, only the one for the lowest frequency is to be feeded...
I made a small program to calculate the distance vs. resonance frequencies, but I think I found the formulae for this in the ARRL Antenna Handbook (don't remember the issue, but it was an old one!)
Best regards
Karl Jan - LA3FY
|
Karl Jan, Can you share your program with us? This looks like an ideal way to deal with both the 80 and 160M bands! Thanks, Earl WA4KBT On Mon, Dec 14, 2020 at 10:10 AM Karl Jan Skontorp < karl.jan.skontorp@...> wrote: Hi Barry!
If you want to operate without a tuner, but with resonance at both
3.589MHz and 3.990MHz, you could add an additional dipole (with resonance
at 3.990MHz) 284mm apart from your "3.580MHz dipole". Just use som
spreathers and without any connection between the two elements, only the
one for the lowest frequency is to be feeded...
I made a small program to calculate the distance vs. resonance
frequencies, but I think I found the formulae for this in the ARRL Antenna
Handbook (don't remember the issue, but it was an old one!)
Best regards
Karl Jan - LA3FY
-- Earl Griffith WA4KBT Soddy Daisy,TN EM75Kg
|
"Thus, the tuner RE-REFLECTS the reflected wave at 3900 kHz back in the
direction of the antenna. Something W2DU Walt Maxwell claimed was how it
worked."
YES, YES, YES !!!!!
73, WB4HXE, Carey
toggle quoted message
Show quoted text
On Sun, Dec 13, 2020 at 9:45 AM Barry Feierman <k3euibarry@...> wrote: Looking at the PHASE data on a single band 80m center fed dipole.
See the two attachments: one without antenna tuner, one with an antenna
tunder
I've had so much fun and have learned so much about antenna parameters
with this little Nano VNA device.
One of the parameters that I was not familiar with is to look at the
PHASE data, say on an 80m dipole: a 130 ft long center-fed dipole with
about 150 feet of low-loss RG213 50-ohm coax.
The antenna on 80m had a natural resonance around 3580 kHz (where I like
to operate digi modes on NBEMS nets).
Remember at resonance, the reactance is zero (regardless of the SWR).
The PHASE is zero degrees.
But that antenna had an awful SWR at the top of the band, where the phone
nets play around 3990 kHz.
At 3990 kHz, the SWR was more like 4:1 and my rig was not happy.
So, how do I QSY to the top of the band with decent SWR without cutting
10 feet off my dipole?
Well, with an antenna "tuner" in the shack, I can null out the reactance
of the antenna at 3990 kHz, and get an almost 1:1 SWR on the coax going to
the transmitter.
My rig is "happy" at 3990 kHz with that low SWR. But the SWR is still
4:1 on my feed line.
The SWR is only reduced on the coax between the tuner and the rig.
How would a PHASE plot look for my antenna AFTER the RF passes thru my
antenna tuner (T network)?
Well, a PHASE plot shows me the original resonance of the 80m dipole
around 3600 kHz (without the tuner) and the SWR and RETURN LOSS graphs
verify this as well. Now, post antenna tuner, matched to the TOP of the
75m band, there is a second NEW resonance at around 3900 kHz post antenna
tuner. Thus, the tuner RE-REFLECTS the reflected wave at 3900 kHz back
in the direction of the antenna. Something W2DU Walt Maxwell claimed was
how it worked.
Fascinating stuff. The tuner (in my shack) provides a "conjugate"
impedance to offset the reactance at the feed line/antenna boundary. Thus
an impedance of R + jX (in my shack) combines with an impedance of R -
jX in the tuner ==> resulting in an impedance of R + j0 for the
radio, with R about 50 ohms. Beautiful.
And, my low-loss RG213 feed line at 4 MHz, even with a 4:1 SWR, has less
than a 1 dB "loss" due to the 4:1 SWR.
A 1 dB loss is barely noticeable.
So why worry about broad-band 80m antennas (gazillion articles in QST)?
Just put a decent tuner in your shack and QSY where you want with very
little attenuation in low-loss cable (on 80m).
There is also very little "loss" in the tuner itself with decent size
inductors and capacitors.
So cool to see all of this concept verified on the NANO VNA PHASE plot
using VNA SAVER software (Win 10).
de Barry k3eui
Phila
--
Carey Fisher
careyfisher@...
|
Yes, no problem Barry!
A bit off topic, but I hope it's ok with the others in this group!
One problem is the language, it's in norwegian, but it's a rather simple program. I can have a look and find the source code and change to english...
Please send a personal message if you have questions...
Here is the link: (case sensitive!)
Karl Jan
|
Barry a little while back there was an article in qst trying to solve this same issue. I used it to do exactly what you want to do.
I used the nano to figure out how much capacitance I would need to add, in the shack, to move the resonance up to the ssb part of 80. From the cw portion. I then cut a piece of coax to the length needed (a bit of trial and error) and added that to one side of a switch so I could, when I wanted to move up to SSB, switch in the coil of coax.
Works great and almost free!
Ron
N4XD
|
Thank you all for your comments and suggestions.
I'm learning more about how to interpret the Nano VNA data from the end of my 150 ft feed line.
I was reminded that the PHASE data, as well as the resistance, reactance, and impedance and the Smith Chart view CHANGES with the feed line length. Thus the PHASE graph I get from my shack is not the same PHASE graph I would get if the Nano VNA were right at the antenna. I guess for most of the readers, that is obvious, but it was not obvious that the RESONANT frequency (where X is zero) CHANGES as the feed line length changes. How did I miss that? Do most of the readers on this list know WHY that is so?
My next "project" is to take that 80m inverted V 130 ft dipole with 150 ft of RG213 50 ohm coax and ADD 10 more feet of feedline, and take a new VNA plot. Then add another 10 ft, then 20 ft, unitl I "see" that when I get to 1/2 wavelength of additional coax (considering the velocity factor) I get the Smith Chart rotated back to its original position ==> then I will be a believer in how the feed line is part of the antenna system.
Resonant frequency (measured at the antenna) is NOT the same resonant frequency measured at the end of X feet of feed line. The "resonant" frequency depends on what the feed line does to PHASE.
What does NOT change (much) with additional feedline is SWR, reflection coefficient, and RETURN LOSS.
I think I undetstand why those quantities are (mostly) independent of feed line length, with some additional loss due to the additional feed line via normal attenuation.
On the question of "broad-banding" a typical 80 meter dipole: much has been written over decades about how to modify the antenna (more wires, thicker wires, cage dipole, fan dipole, adjacent wires, etc.) but still the easiest way to operate on the entire 80/75m band with a simple dipole 130 fee long and center-fed is to just put a wide-range antenna tuner in your shack and not mess with the antenna iitself. The "lost" power at 3.5 MHz due to an SWR of 5:1 on RG213 coax is not worth worrying about. If you "lose" an additional 1 dB from the higher SWR in 100 ft of RG213 coax, no one at the receving end is going to noitice that loss.
I've given two Nano VNA talks to clubs over the past two months, but with a very limited understanding of how to interpret these VNA SAVER graphs in order to improve your setup.
Thank you to all of you who wrote to me off-list. You know who you are.
Now I wonder why we hams have not used RETURN LOSS in place of SWR.
When was the last time you saw an analog meter that read Return LOSS, or a meter with a scale that included BOTH return loss and SWR?
73, and stay safe
This IO group has a lot of good information.
Barry Feierman k3eui
West Chester, PA
|
Hi Barry!
If you use a number of half wavelength coax cables, you will get the same result as if you made the measurement direct in the feedpoint. You can use the smith chart, one "full circle" is the same as one half wavelength. If you start with a measurement without any cable and open input, you will have a "point" on the edge to the left in the chart. If you connect a cable, 1/4 length, you will end ut at the left side and get a "short", extending the cable to the double, you will be back to where you started. If adding a new half wavelength, you get a "new" circle, a litle closer to the center. This is because there will be some loss in the cable. Set the sweep start to 10kHz and adjust the sweep stop to get a bit more than a full circle. Now you can cut the cable until you get one or two or x whole number of circles - each circle represents one half wavelength.
So, keep in mind using a number of half wavelengths of cable to do measurements "remote" OR you could of course make the calibrations at the end of a random length cable. The calibration will "cancel" the cable length...
Karl Jan - LA3FY
|
Ron,
I too am working with the issue of broad banding a 80CW dipole into the phone band.
Can you provide a closer estimate or be specific to when you saw the QST article?
The coax you added, was it in series with the existing feed line, or a stub; open or shorted?
I'm assuming this is in the shack?
|
I think the article you are referring to is on page 43 of Feb 2020 titled ¡°SWR Shifting Tee.
There is an excellent free program Sim Smith that can solve all these problems. Measure the length of your coax using the VNA¡¯s TDR function and sim smith will shift your measured values to the actual feed point impedance.
Dallas N5FEE
|
Barry K3EUI
schweppe