What causes SWR to change along a feed line?

[Jackson Richter]

Engineering has taught us that unequal impedances between feed line and antenna feed point will result in return reflections along the feed line back to the source. The mixing of the forward and reflected waves cause nodes which consist of variations min and max voltages along the line resulting in impedance variations along the line. At the same time, SWR measurements taken along the feed line are supposed to be the same any where along the line.
My question to all, What are the contributing factors that would cause SWR to change?
Thanks JR Richter

 

[tech99]

My question to all, What are the contributing factors that would cause SWR to change?
Thanks JR Richter

This points to a VSWR indiactor or reflectometer with poor directivity. It is seeing reflected and forward energy at the same time.
VSWR also improves if the line is lossy as we move away from the load.

 

[GrahamN-UK]

What are the contributing factors that would cause SWR to change?

Losses in the feed line will reduce the reflected power and make the indicated SWR become smaller (i.e. look better). The further away you measure from the antenna the better the SWR will look. However the SWR at the antenna will still be as bad.

Additional discontinuities in the feed line can increase the SWR seen on the feed side. These could be caused by a bad joint or using a section of cable or plug/socket of a different characteristic impedance to the rest of the system.

 

[Jackson Richter]

Losses in the feed line will reduce the reflected power and make the indicated SWR become smaller (i.e. look better). The further away you measure from the antenna the better the SWR will look. However the SWR at the antenna will still be as bad.

Additional discontinuities in the feed line can increase the SWR seen on the feed side. These could be caused by a bad joint or using a section of cable or plug/socket of a different characteristic impedance to the rest of the system.

Through my studies, I agree with you that one of the major contributors to swr measurement error is feed line loss. If taking swr measurements close to the source causes lower than realistic measurement data then by the same token, measurements taken directly at the feed point of the antenna would be higher than the realistic swr measurements, depending on loss characteristics. I am also wondering if any mismatch between feed line Z and antenna Z would add to the swr measurement error? Validity of the swr measurement appears to be dependent on the level of loss in the feed-line but what about the mismatch too? Your thoughts

 

[tech99]

I am also wondering if any mismatch between feed line Z and antenna Z would add to the swr measurement error?

Mismatch between the feedline and the antenna is the swr.

 

[GrahamN-UK]

If taking swr measurements close to the source causes lower than realistic measurement data then by the same token, measurements taken directly at the feed point of the antenna would be higher than the realistic swr measurements, depending on loss characteristics.

On a lossy line with a single discontinuity at the load end, the SWR will vary continuously along the line, being largest at the load end and smallest at the source end. No single figure can be said to be the correct or realistic measure. Assuming your SWR meter is correctly calibrated then whatever it reads at any given point on the line is the correct reading for that point on the line. Changes in the SWR reading along a lossy line is not an error - it is what is actually happening. Only if you have a lossless line will the SWR be constant along the line.

If you need to decide where to measure the SWR you need to consider what you want the measurement for. If you want to know if the antenna is properly matched to the line you'll want a SWR reading at the line/antenna junction. Moderate to high power RF transmitters tend to specify a maximum SWR they will work into, so you might want a SWR reading at the source end of the line to ensure you are within the specified limit. (Well, transistorised transmitters do, valve/tube ones can be more relaxed.)

If you're concerned at losses in your feedline then markedly different SWR readings at either end of the line are an indication that you need to invest in a better quality and lower loss (=more expensive) line.

I am also wondering if any mismatch between feed line Z and antenna Z would add to the swr measurement error?

tech99 has nailed this; the SWR is a measurement of the mismatch between the line and the antenna.

 

[Baluncore]

The mixing of the forward and reflected waves cause nodes which consist of variations min and max voltages along the line resulting in impedance variations along the line.

Forward wave energy, reflected by a mismatched impedance at the end of the line, travels back along the line as a reflected wave. The forward wave and the reflected wave in a linear transmission line are quite independent.

The nodes formed as a standing wave on the line are only seen when you insert a meter in the line that cannot distinguish between the independent forward and reflected waves. The traveling waves see a constant characteristic impedance. The characteristic impedance of the transmission line does not change due to the standing waves as they are not traveling on the line, they are standing waves.

In a lossy line, continuous attenuation of the forward wave reduces it's amplitude towards the point of reflection. The reflected wave is also being continuously attenuated as it returns to the source. As expected, the SWR measured on that line will represent the SWR only at that point on the line where the measurement is made. There will be no sudden change in SWR when crossing any part of the standing wave pattern on the line because the SWR meter differentiates between the direction of wave travel.

 

[GrahamN-UK]

I agree with you that one of the major contributors to swr measurement error is feed line loss.

No, that's not what I meant. Variation of SWR readings along a lossy line is not a measurement error. The line loss will cause the SWR to actually vary along the line and a correctly functioning SWR meter will show this.

This means you can't state a single figure for SWR for all of a lossy line. If you want to state a single figure you must also state whereabouts on the line you measured it.

 

[Svein]

https://en.wikipedia.org/wiki/Transmission_line

 

[Jackson Richter]

Forward wave energy, reflected by a mismatched impedance at the end of the line, travels back along the line as a reflected wave. The forward wave and the reflected wave in a linear transmission line are quite independent.

The nodes formed as a standing wave on the line are only seen when you insert a meter in the line that cannot distinguish between the independent forward and reflected waves. The traveling waves see a constant characteristic impedance. The characteristic impedance of the transmission line does not change due to the standing waves as they are not traveling on the line, they are standing waves.

In a lossy line, continuous attenuation of the forward wave reduces it's amplitude towards the point of reflection. The reflected wave is also being continuously attenuated as it returns to the source. As expected, the SWR measured on that line will represent the SWR only at that point on the line where the measurement is made. There will be no sudden change in SWR when crossing any part of the standing wave pattern on the line because the SWR meter differentiates between the direction of wave travel.

 

[Jackson Richter]

I truly understand that swr is created because of impedance mismatch between feed line and antenna feed point. Any deviation in swr along a feed line is due to the initial loss characteristics of that particular feed line and of course the applied frequency. I certainly did not make myself clear in previous statements, (sorry). I want to be very clear about the statements made here. "Standing waves are not traveling along the line". If they're not on the line then where are they? "Impedance is constant for the traveling waves". I believe impedance does change with length otherwise 1/4 wavelength matching stub would not work. Please clarify, thanks

In reference to earlier statement regarding losses, I have charts that show the added feed line losses due to swr. I was just wondering if the added losses affect swr measurement? It sounds like lossless feedline is the only way to get a real swr measurement.

 

[Baluncore]

"Standing waves are not traveling along the line". If they're not on the line then where are they? "Impedance is constant for the traveling waves". I believe impedance does change with length otherwise 1/4 wavelength matching stub would not work. Please clarify, thanks

You have failed to understand that the forward and backward waves on the transmission line are quite independent traveling waves. You are assuming only the special case where the forward and a backward wave are sinewaves, with the backward wave being generated by the reflection of the forward wave.

Consider instead a 50 ohm impedance line, with a correctly matched and independent broadband noise source at each end. Energy will propagate in both directions, there will be no reflections. The characteristic impedance of the line will be 50 ohms everywhere, in both directions. Line length and wavelength will have no meaning since the broadband signals have no one frequency and there are no reflections.
That situation occurs in digital networks where metal lines or optic fibre are used for duplex data transmission, for example twisted pair telephone lines or coaxial cable data networks.

Standing waves are produced in stationary positions where waves with the same wavelength, traveling in opposite directions are allowed to interact. That is why they are called standing waves, because they do not travel along the line. Standing waves can only be perceived where you break the transmission line and so destroy the independence of the directional waves. The directional waves form a vector sum in the termination impedance at that point.

 

[Jackson Richter]

I am trying to understand what you are saying about "The sine wave being a special case" in reference to forward and reflected waves. Your theory should hold true for all applications. "Sinewaves are the only special case where backward wave being generated by reflection of forward wave"?

What is generating the real impedance changes along a feedline? Is it simply, only because of the characteristics of the feedline and its VF? Since the impedance does change along a feed line, it has to be related to something, especially when feeding a multi-band antenna.

If I understand you correctly, any mismatch between feedline and the antenna feed point impedance has no affect on impedance changes along its length, since "nodes are real, but apparently only measurable" when doing swr measurements.

 

[Baluncore]

What is generating the real impedance changes along a feedline?

You will be stuck until you accept that there are no real impedance changes along the line. The current you measure at a fixed point on the line is the difference between two current waves propagating in opposite directions.
You have convinced yourself that total line voltage divided by the difference between the forward and backward currents is physical line impedance. But that is not the case as you must treat the directional waves propagating on the line independently, until you terminate the line.

 

[Jackson Richter]

Ok, When looking anywhere along the feed line that has a mismatch, swr will be present. As a result, "nodes formed as a standing wave on the line are only seen when you insert a meter in the line that cannot distinguish between the independent forward and reflected waves." This concept is new to me but understandable. Thank you, that's why I am here to learn.

When considering a lossless line at a fixed frequency, every 1/2 wavelength along the line, the source impedance and load impedance will look the same but if you were to look at every 1/4 wavelength, if the source impedance were shorted, the load impedance would show an open. These are valid impedance changes along the feed line, strictly due to the characteristics of that line and the applied frequency. Again, these are real impedances changes along the feedline. Matching stubs are used to transform one impedance to another.

 

[Baluncore]

These are valid impedance changes along the feed line, strictly due to the characteristics of that line and the applied frequency. Again, these are real impedances changes along the feedline.

No, they are not.

SWR is computed from the relative forward and backward energy propagating on the line.
Consider using a hypothetically perfect directional coupler to measure only the the forward wave energy at all points along a poorly-terminated slotted-line. Then turn the coupler round to measure only the backward wave energy at all points along the line. That directional data will show only a steady attenuation of the independent directional signals, from which you can compute the SWR as it changes along the line, due to attenuation. The SWR will only change gradually along that line.

Now open or short the far end of that sinewave driven line to get 100% energy reflection. Insert a small coupling loop into the slot to sense the line current (= the vector sum of the two directional currents). As you slide the loop along the line you will see variations in current from 0% to 200% at different positions, determined by the position in wavelengths along the line from the reflector. If the line has resistance, some parts of the line will stay cold while others will be twice as warm as when perfectly terminated.

You must cut your line to length before you hypothesis about where the forward and reflected waves will be summed. Only then can you say "the impedance of the line varies depending on where you have cut and terminated the line".

 

[Jackson Richter]

© 2009-2018 Kevan Hashemi, Brandeis University
"Because an imperfectly-terminated transmission line causes power to be reflected back to the source, the impedance seen looking into such a line is not equal to the characteristic impedance of the line, but some function of the reflection coefficient at the far end, and the length of the line. If we are working at one particular frequency, or a narrow range of frequencies, we can use a finite transmission line, or stub to create any impedance we like. We can use such stubs as matching networks." This comment was posted in the conclusion of a paper written by Kevan Hashemi an Engineer at Brandeis University in Waltham, Massachusetts.

My earlier question was " What are the contributing factors that would cause SWR to change?" I think that was covered nicely by a few of you. Its directly related to loss along the feed line. The following question still haunts me (How and why does impedance change along a feed line?) The above paper written by Mr. Hashemi, says its caused by a mis-match and length of line and others say its not.

Clarification is needed. Thanks

 

[Baluncore]

The following question still haunts me (How and why does impedance change along a feed line?)

Your Kevan Hashemi quote does NOT say "the impedance varies along the line".
It says terminal port impedance depends on time delay due to the length of the line.

A transmission line is simply a two port network with a characteristic impedance and a delay time.

If one port is mismatched, the impedance seen at the other port will vary due to the length of the line. The impedance certainly DOES NOT vary along the line between the ports. In that space the directional propagating signals are quite independent. You have no access to that space.