OwenDuffy.net 


 

VSWR Myths

My antenna has a VSWR of 1.000x:1

It is unlikely that you can measure VSWR to that precision, let alone accuracy. If the forward power is 100w and the reflected power is 10mW, VSWR is 1.02.

In practice, it is difficult to measure return loss figures greater than about 40dB with good quality equipment, most ham VSWR meters could not achieve that for several reasons. Indeed, in my observations such instruments are likely to introduce more VSWR than that, I have observed insertion VSWR (the VSWR caused by inserting the device in a line to a dummy load) as high as 1.5 at the high end of the claimed frequency range of the instrument.

Any measurable VSWR (reflected power) is bad

Some transmission lines are intended to operate at VSWR that is not unity, and may be designed to be quite large (approaching infinity for stub elements).

Here is an example of three alternatives for feeding100W to a 50 ohm antenna on 14MHz

  • Open wire (TV line)
    • VSWR = 4, matched line loss = 0.38B, line loss due to VSWR alone = 0.49dB, power delivered to the antenna = 81W
  • Open wire (600)
    • VSWR = 8, matched line loss = 0.08B, line loss due to VSWR alone = 0.26dB, power delivered to the antenna = 93W
  • Coax (RG58C/U)
    • VSWR = 1, matched line loss = 1.5dB, line loss due to VSWR alone = 0dB, power delivered to the antenna = 70W

Note that high VSWR does not of itself damn a solution, here the configuration with highest VSWR delivers most power to the antenna.

For simplicity the above example assumes a 50 ohm antenna and ignores the requirements (ie the losses) for baluns. In a practical installation baluns may be required in one or other configuration and they do add losses, but for the purpose of considering the effect of VSWR alone, the example shows that high VSWR on a low loss transmission line increases the losses only marginally and may still be much lower loss than the matched line loss of some other types on line.

This explains why long open wire lines are often used in commercial HF stations, and they are not necessarily operated under very low VSWRs.

Here is a practical design case that runs at high VSWR on an inexpensive low loss line.

I have a 7.1MHz dipole, radiation resistance of 70 ohms. I feed it with open wire feeder with a single stub tuner to get 50 ohms at the transmitter. I connect a s/c stub (2.4m) of 300 ohm open wire TV feeder in parallel with the antenna. I connect 300 ohm open wire TV feeder (20.3m) from the antenna to a 1:1 balun at the transmitter. The VSWR on the 20m line is 6, the line loss will be .06dB matched line loss plus 0.13dB due to VSWR. See the Smith Chart solution. The transmitter delivers rated output into the 50 ohm load, and the total line loss of 0.2dB (ignoring the stub losses which are very low) means 95% of the power is delivered to the antenna.

If I used RG213 all the way to a 1:1 balun at the antenna, I would be lucky to get 70 Watts forward at the transmitter due to the 1.35 VSWR, and would get only 63 Watts into the antenna.

I have ignored the balun losses in each case, they will be similar and low.

A transmission line with VSWR > 1 radiates and will cause interference

Radiation is a result of the net effect of the radiation of all conductors in the transmission line.

If it is an open wire line, the radiation is dependent on the net current (the imbalance). If it is a coaxial line, it is dependent on the current flowing on the outside of the outer conductor. The VSWR does not inherently affect the balance, and open wire transmission line is often used with high VSWR and good balance. However, in some installations a change (increase or decrease) in VSWR from normal (not necessarily unity), may indicate a fault that may also cause unbalance and attendant transmission line radiation.

VSWR doesn't matter at all

VSWR greater than unity does increase transmission line loss, but the extent depends on the VSWR and the matched line loss. Mismatched loss=matched loss * (1+S**2)/2/S where S = VSWR.

If the VSWR is significantly different to the design VSWR, then it may indicate a faulty implementation or a  fault.

If low VSWR is required to meet the transmitter's requirement for a load impedance, high VSWR may limit the power than can be obtained from the transmitter and could cause the transmitter power amplifier to operate outside the designed safe limits or to activate protection circuits and reduce output power in a controlled way.

My VSWR bridge is infallible

VSWR bridges necessarily disturb the transmission line into which they are inserted.

  • They are a transmission component themselves and may be quite imperfect in some implementations
  • They couple energy from the main line, and if they couple too much energy they become inherently inaccurate event though they indicate nil reflected power
  • The coupler element is frequency sensitive and is a compromise between sensitivity and upper and lower frequency cutoffs / insertion VSWR.
  • The detectors are non-linear at lower power levels (especially when at high sensitivity settings on units with variable sensitivity), and the result in a change in scale shape that is not usually reflected in the meter calibration.
  • Poor instruments may not be accurately calibrated for nil reflected power and indicate 1:1 at other than true 1:1 VSWR.
  • Poor coupler implementations are invariably 'calibrated' to show nil reflected power on a dummy load, but commonly don't read accurately for higher VSWR, especially towards infinity thereby significantly limiting their usefulness.
  • Dual coupler models (ie with a separate forward and reverse coupler) are often poorly matched, meaning that the result will depend on which direction the forward power flows through the instrument.
  • Poor couplers may insert much more VSWR than is the target value for system operation.
  • Under certain operating conditions the characteristic impedance of the line may depart significantly from the specified nominal value.

You can only measure VSWR with a tuned line length.

If you:

  1. break a transmission line and insert a VSWR meter and its line(s); and
  2. the VSWR meter and its lines are matched to the line impedance (ie they do not cause significant VSWR of themselves); and
  3. the VSWR is unity

then the impedance presented to the transmitter is unchanged, it is the same as the characteristic impedance of the line.

If conditions 1 and 2 above apply, but not 3, then the impedance presented the the transmitter is not the same as the characteristic impedance of the line, and it will vary with line length and depend of VSWR. If the effective length of line that is inserted with the VSWR meter and its lines is an exact multiple of a half wave, and the inserted line loss is low, then the impedance presented to the transmitter is essentially unchanged.

Using the tuned line allows you to tune the transmitter (which optimises the transmitter output for the actual load impedance), knowing that when you remove the SWR meter and its lines, that the load impedance should be the same and the settings remain applicable. You also get a power measurement that is applicable to the actual load impedance.

If the length of the inserted line is very short compared to a wavelength, then there is very little effect due to the insertion. So for instance, if you use a standard say 1m patchcord from the transmitter to a patch panel, then replacing it with a VSWR meter and lines of the same electrical length should achieve the same result. (Don't forget to allow for the velocity factor of the lines and the SWR meter, which may all be different - the VSWR coupler is typically air spaced and has a velocity factor of 1.)

Remember, if all three conditions apply, line length is not critical. If the first two conditions apply and VSWR is low, then it won't make much difference.

If condition 2 does not apply, get a good VSWR meter. VSWR meter specifications should state the insertion VSWR (ie the VSWR that they cause when connected to a perfect dummy load). I can't recall ever seeing this parameter specified for VSWR meters intended for the ham user, even instruments costing hundreds of dollars. It is relatively easy to make a VSWR meter that indicates zero reflected power when connected to a very good dummy load, but that is not to say that it is not causing reflected energy itself.

Note that at under some conditions the actual complex characteristic impedance of the line may depart significantly from the specified nominal (purely) resistive value. This occurs most noticeably for lines operating at low frequency with high loss and high velocity factor.

VSWR is the same at the transmitter end as it is at the antenna end

This is only true if line losses are zero. This is not possible with real transmission lines, but if the line losses are low they may be ignored without impacting the accuracy of the result unduly.

Line loss doesn't matter

Loss under mismatched conditions depends on the matched line loss and VSWR.

Mismatched loss=matched loss * (1+S**2)/2/S where S = VSWR. This is not a very practical formula, because it really expresses mismatched line loss per unit length, and S is the VSWR at that 'point'. You can get reasonable results by averaging the VSWR at each end of the lossy line and using that value for S in the formula.

Low loss lines can be operated at high VSWR without a large increase in loss, whereas higher loss lines need to be operated at low VSWR to preserve reasonable transmission efficiency.

In the example above, the high VSWR=5 doubles the line loss, but that only adds 0.12dB. A VSWR of just 1.5 on the coax line would add about 0.12dB due to the higher matched line loss. 

Low loss line is worth its cost / inconvenience.

The only way to evaluate the choice of lines is to find the power delivered to the antenna for the different scenarios.

If you want to have the lowest VSWR in town, lossy line helps mask antenna mismatch. In fact, a length of line that gives say 17dB of loss, terminated in a short circuit makes a pretty good dummy load (VSWR wise - 1.05 ).

If you use expensive low loss line, the antenna  mismatched is more transparently presented to the transmitter as a high VSWR at the transmitter end of the line. I have seen instances where low loss line on a poorly matched antenna caused the transmitter VSWR protection circuit to reduce transmitter output such that there was less net power delivered to the transmitter than when a lossier line was used. Than answer is to fix the antenna.

VSWR protection circuits typically have a threshold before activating, and good transmitter output is achieved up to that threshold, but rapid reduction in output occurs above the threshold. Mostly, you can't calculate (predict) this effect, but have to measure it. Transmitter output in any event depends on the magnitude and particularly the angle of the load impedance, so again it is best to measure it.

Of course, if the antenna is truly a good match to the line (ie indicates a low VSWR when you measure it at the antenna, or measure it anywhere and take the losses into account), then you can calculate the improvement and make the value choice.

 


© Copyright: Owen Duffy 1995, 2017. All rights reserved. Disclaimer.