OwenDuffy.net 


Feeding a G5RV

Overview

The G5RV is described by its inventor, Louis Varney (G5RV), as a limited space antenna that will work on the (then pre WARC-79) HF bands from 80m to 10m.

The essence of a G5RV is a dipole that is 3λ/2 long at 14.15MHz, fed by a λ/2 balanced line "matching" section (approx 520 Ω Zo) and an arbitrary length of coax or low Zo balanced line to a tuner. Varney's articles suggest that an inverted-v configuration of the dipole legs is acceptable, though he recommends the included angle should be greater than 120°. (Varney did also describe a configuration using only open wire line of approx 520 Ω Zo, but that configuration is not nearly as popular as the high Zo / low Zo combination.)

This article analyses the feed arrangements for a G5RV in a typical, low, shallow inverted-v configuration, using ladder line for the "matching" section and a range of options from that point to a transmitter providing a total length of transmission line from the dipole centre to the transmitter of 25m.

The analysis is based on the feedpoint impedance from an NEC model of the dipole, computed every 0.1MHz from 1MHz to 30MHz. The real performance of transmission line elements in terms of impedance transformation and loss, and a practical L-match tuner network components and loss are calculated to form a view of the overall losses in the feed arrangement from transmitter to dipole centre.

This analysis does not consider conductor loss in the dipole, ground loss, directivity, polar pattern of the radiator. (Copper radiator conductor sufficiently strong to reliably withstand the wind has conductor losses that are low compared to typical feed system losses.)

The radiator behaves as a low dipole. At frequencies where it is relatively short, the pattern is nearly omni directional, and the pattern transitions to multiple lobes and nulls at frequencies where the dipole it is multiple wavelengths long.

Table 1 sets out the design parameters that are common to all of the configurations modelled in this article.

Table 1: Common design parameters
Item Value
Design frequency 14.15MHz
Height of centre 10.0m
Height of ends 4.5m
Length of half dipole 15.5m
Included angle 137°
Antenna conductor 2mm diameter copper
Ground "real" ground (σ=0.001S, ε=4)
Matching section line type Wireman 554 "ladder line", dry, assumed copper conductivity
Length of "matching" section 9.85m

Note that the length of the dipole legs and matching section are not of the same physical length as described by Varney, but they are of the same electrical length having each been adjusted to be resonant at 14.15MHz in the model scenario.

The model uses Wireman 554 performance data measured by N7WS, but his measurements would not have exposed skin depth inadequacy at low HF, so the extrapolation in this analysis assumes copper conductivity.

Table 2 sets out assumptions underlying the models.

Table 2
Assumptions
Antenna is erected over "real" ground (σ=0.001S, ε=4).
Ideal balun is used at lower end of "matching" section.
L-match tuner used to transform load to 50+j0 Ω to suit a transmitter.
Total transmission line from antenna centre to transmitter is 25m.

Figure 1 is a plot of the R and X values at the base of the matching line against frequency, the so called impedance spiral. The plot is cropped to show only the low R and X ranges, values off the plot will have unacceptable losses with practical lengths of coax line.

Fig 1: Impedance spiral at base of the "matching" section

Figure 2 is another plot of the R and X values at the base of the matching line against frequency. The plot is cropped to show only the low R and X ranges, values off the plot will have unacceptable losses with practical lengths of coax line.

Fig 2: Impedance at base of "matching" section

The models below explore different configurations for connecting the base end of the "matching" section to a transmitter requiring a 50Ω load.

The L match used for tuner loss is in general, the most efficient way to transform the load impedance. Using the loss of a practical L-match tends to  underestimate the loss in any other tuner configuration. For example, for the scenario in Fig 4, the losses in a typical T-match at 3.6MHz are 0.8dB against 0.1dB for a typical L-match.

Models

This article explores the feedline and tuner losses for five configurations of feed arrangements of a G5RV. The first two configurations are arguably the most common (though a significant departure from Varney's description).

G5RV with RG58C/U to tuner/transmitter

The configuration comprises balanced line matching section of 9.85m of Wireman 554, extended to the "local" tuner by 15.15m of RG58C/U coaxial cable.

It is an inexpensive solution, but has poor losses that are acceptable only on the 3.5, 7, and 14 MHz bands.

Fig 3: 15.15m RG58C/U - tuner

G5RV with RG213 to tuner/transmitter

This configuration addresses some of the loss by operation of a lossy coax line at high VSWR by using a less lossy coax line. It comprises an balanced line matching section of 9.85m of Wireman 554, extended by 15.15m of RG213 coaxial cable to a "local" tuner.

It is a high cost solution "local" tuner solution, and has moderate losses that are acceptable only on the 3.5, 7, 14 and 21 MHz bands.

Fig 4: 15.15m RG213 - tuner

G5RV with RG6/U to tuner/transmitter

The configuration comprises an balanced line matching section of 9.85m of Wireman 554, extended by 15.15m of RG6/U coaxial cable to a "local" tuner. It is an inexpensive "local" tuner solution with losses acceptable on the 3.5, 7, 14, and 21 MHz bands.

Fig 5: 15.15m RG6/U - tuner

G5RV with ZIP cord to tuner/transmitter

This configuration comprises an balanced line matching section of 9.85m of Wireman 554, extended by 15.15m of US style "ZIP" cord to a "local" tuner.

Varney suggested 75Ω balanced line in this role, and there is a popular misconception amongst hams that Figure-8 flex or ZIP cord is near to 75Ω and, and being a balanced line, that it has low losses. They are wrong on both counts.

The feed system losses with this configuration are acceptable on only the 14MHz band, and accordingly it is rated unsuitable.

Fig 6: 15.15m ZIP cord - tuner

Figure 6a shows a similar configuration, it comprises an balanced line matching section of 9.85m of Wireman 554, extended by 15.15m of Australian Figure-8 flex (0.5mm2 24/0.20) to a "local" tuner.

Fig 6a: 15.15m Figure-8 flex - tuner

The feed system losses with this configuration are acceptable on only the 14MHz band, and accordingly it is rated unsuitable.

Novel using a remote auto-tuner at the base of the matching section and RG58C/U to the transmitter

This configuration is designed to minimise coax operation at high VSWR by relocating the tuner to be as close to the end of the "matching" section as possible. It comprises a balanced line matching section of 9.85m of Wireman 554, extended by 1m of RG6U coaxial cable to a "remote" (auto) tuner, extended to the transceiver by 14.15m of RG58C/U coaxial cable.

It is the overall least loss solution, with losses acceptable on all HF bands 3.5MHz to 30MHz (including WARC bands). The remote auto-tuner more expensive than manual tuner, and the auto-tuner has limited power handling (100w to 200W max).

Fig 7: 1m RG6/U - tuner - 14.15m RG58C/U

Ladder Line

Though in this analysis, a low grade balanced line transmission has been used, it is quite sufficient for the task, so long as it is dry. Ladder line performance degrades significantly when wet, and a better type of balanced line should be considered in environments that are commonly subject to rain, fog, frost, snow, ice etc. Using XLZIZL to explore the impact of wet ladder line, at 3.6MHz the loss on the 9.85m of Wireman 554 would increase from 1.7dB dry to 3.9dB wet.

Figure 8 shows the impact of wet ladder line, it is the same configuration as shown in Figure 3, but with wet ladder line. The losses are not acceptable on most bands. If you had in mind operating on 80m at night in an environment where dew settles on the feedline, you need to use a better feedline.

Fig 8: Wet Wireman 554 ladder line - 15.15m RG58C/U - tuner

A Better G5RV

High VSWR operation of the transmission line between the end of the "matching section" and the transmitter is the greatest problem with the G5RV. The "Better G5RV" configuration discussed here addresses some of the issues that arise with a G5RV, proposing a solution that is based on modern technology.

The configuration proposed is an inverted-v configuration (as modelled above), an air spaced matching section to reduce loss and particularly to reduce the degradation when wet, a balun and remote auto-tuner at the base of the "matching section", and 50Ω RG58C/U coax to the transmitter. The model includes 0.5m of RG303/U on the output side of the tuner for a W2DU style balun using 50 to 100 Amidon FB-73-2401 beads. The length of the 50Ω coax to the transmitter is not critical, the models here use 15m of RG58C/U for loss calculation, longer runs should use lower loss cable such as RG8X or RG213.

Note that the remote tuner is earthed to reduce RF currents flowing on the outside of the coax, and the tuner control cables.

Figure 9: A better G5RV

Figure 10 shows the feed system loss for the Better G5RV. Losses are acceptable across all of HF from 3.5MHz to 30MHz.

Fig 10: Better G5RV

An important issue in the deployment of a remote auto-tuner such as the Icom AH-4, Alinco EDX-2 etc, is that they have a quite limited capacity to withstand voltage. Figure 11 shows the voltage impressed on the remote auto-tuner in the configuration at Figure 9. Note that operation at 5MHz and 10MHz causes the greatest stress (clearly the configuration is unsuitable at 1.8MHz). It may be wise to fine tune the length of the "matching section" to obtain an impedance minimum at the base of the "matching section" at 3.6MHz so that the first voltage minimum is centred in the 80m band.

Fig 11: Voltage at remote Auto-tuner at 100W power level

Classic tuned feeder

A question that occurs is would a quality open wire transmission line all the way from the feedpoint to a balanced tuner adjacent to the transmitter have better performance. (This configuration was also described by Varney.)

Figure 12 models the G5RV radiator with 23m of 600Ω open wire air spaced line (2mm dia, 150mm spacing) to 1m of RG6 (balun) and 1m RG58C/U (fly lead to tx). This configuration is lower loss and is less affected by water, but there are issues in bringing balanced line into the shack. Note that the voltages present at the tuner on low frequencies would be excessive for common auto-tuners.

Fig 12: Losses for classic tuned feeder configuration

Conclusions

Table 3 is a summary of the characteristics of the different configurations. The ranking is a overall ranking of each solution, 1 is best. Total feed losses above 3dB are considered unacceptable.

Table 3: Summary of characteristics
Configuration Fig Comments Ranking
Balanced line matching section of 9.85m of Wireman 554, extended by 15.15m of RG58C/U coaxial cable to a "local" tuner. 3
  • least cost "local" tuner solution;
  • poor loss "local" tuner solution, losses acceptable on 3.5, 7, 14 MHz bands.
4
Balanced line matching section of 9.85m of Wireman 554, extended by 15.15m of RG213 coaxial cable to a "local" tuner. 4
  • high cost "local" tuner solution;
  • moderate loss "local" tuner solution, losses acceptable on 3.5, 7, 14, 21, 25 MHz bands.
3
Balanced line matching section of 9.85m of Wireman 554, extended by 15.15m of RG6/U coaxial cable to a "local" tuner. 5
  • least cost "local" tuner solution;
  • lowest loss "local" tuner solution, losses acceptable on 3.5, 7, 14, 21, 25 MHz bands.
2
balanced line matching section of 9.85m of Wireman 554, extended by 15.15m of "ZIP" cable to a "local" tuner. 6
  • low cost "local" tuner solution;
  • worst loss "local" tuner solution, losses acceptable on 14 MHz band.
5
Unsuitable
balanced line matching section of 9.85m of Wireman 554, extended by 15.15m of Figure-8 flex to a "local" tuner. 6a
  • low cost "local" tuner solution;
  • worst loss "local" tuner solution, losses acceptable on 14 MHz band.
5
Unsuitable
Balanced line matching section of 9.85m of Wireman 554, extended by 1m of RG6U coaxial cable to a "remote" (auto) tuner, extended to the transceiver by 14.15m of RG58C/U coaxial cable. 7
  • overall least loss solution, losses acceptable on all HF bands 3.5MHz to 30MHz (including WARC bands);
  • auto tuner more expensive than manual tuner, auto tuner has limited power handling (100w to 200W max).
1
Balanced line matching section of 9.85m of wet Wireman 554, extended by 15.15m of RG58C/U coaxial cable to a "local" tuner. 8
  • unacceptable losses on most bands.
6
Unsuitable
 

It can be seen that in all configurations, at 80m and above, losses in the open line  "matching" section and in the tuner are relatively small and quite acceptable. The coax transmission line makes the greatest and dominant contribution to total loss.

The performance of a G5RV in its common configuration of an balanced line "matching section" (λ/2 at 14.15MHz) with an arbitrary length of coax or balanced line of "low" characteristic impedance ("Classic G5RV") depends on the height and configuration of the radiator and choice of transmission line types and length.

The Classic G5RV does not deliver a low SWR 50Ω load on any band without an impedance matching network (tuner / ATU). Though a tuner is required for operation, tuner losses are acceptable.

Contrary to oft stated views, the Classic G5RV is not an all band antenna, it:

Zip cord, Figure-8 two core flex, or whatever it is known as in your locality is not a suitable transmission line for extending the G5RV "matching" section to the transmitter, it is far too lossy to operate at high VSWR (as it does in the G5RV).

Though many stations use configurations identified above as poor or unacceptable and claim that they work, it goes  more to their meaning of the word "works" than attestation of the performance of the antenna.

Varney designed the G5RV as an antenna for a limited space and it does, with care, give adequate multi band performance, but it is not an all-band antenna. Modern day alterations, eg using ladder line in wet conditions and using 50Ω coax, compromise the antenna's performance to a greater or lesser extent.

The configuration in Figure 3 is probably the most common implementation followed by that in Figure 4, they are poor performers, beaten by the lower cost configuration in Figure 5. The best performance is obtained by avoiding high VSWR operation of lossy coax by locating the tuner as close to the end of the  "matching" section as possible.

FAQ

Q: How about the configuration in Figure 3, but using the fabulous RG8X?

A: RG8X is better than RG58C/U (Fig 3) but not as good as RG213 (Fig 4). In this model, using  RG8X does not achieve less than 3dB of feed system loss over all of the 21MHz band so I rate its loss as unacceptable on that band. This configuration is undoubtedly better than RG58C/U, but to be consistent, I must rate it as acceptable only for the 3.5, 7, and 14 MHz bands. For much less money, you could use RG6/U and get a small flexible cable with better performance in this application.

Q: Is a balun really necessary?

A: A balun is any device that permits a transition from balanced to unbalanced or vice versa. Baluns are not perfect, and the extent to which they permit that transition is a measure of their perfection. This article states an assumption of an ideal balun for the models, an ideal balun does not compromise the current balance in the feedline, so there will be no radiation from the feedline in a symmetric configuration. If you do not use a balun where it is needed in the G5RV, it will still "work", but it "works" differently as the feedline is likely to carry a net current, contributing to the radiation, and affecting the pattern and impedance. Though Varney initially recommended a balun, then advised against a balun in a later article because of issues with "voltage" baluns, those objections are probably overcome with "current" baluns and a balun of that type should well in this application.

Q: Doesn't a G5RV work on 20m without a tuner?

A: The behavior of your transmitter on high VSWR loads is important to the answer. If it incorporates protection circuitry that reduces output on high VSWR, you may need to keep the VSWR low to achieve rated power output. In the models above, the impedance at the base of the matching section at resonance on 20m is 90 ohms. If your dipole and matching section is resonant, and you use 75 coax of an odd number of quarter waves in electrical length, you will transform that to about 63 ohms, for a VSWR (50) of about 1.25. (If you use 50 ohm coax, the length is not critical and the VSWR will about 1.8 for any practical coax section.)

Links

How was it done?

The analysis is based on an NEC model of the radiator, using EZNEC. The EZNEC model of the G5RV Inverted Vee is available here. EZNEC was used in scan mode to create a report of feedpoint impedances from 1MHz to 30MHz in 0.1MHz steps.

A Perl script was written to parse the EZNEC report, and do the transmission line calculations (impedance transformation, loss, voltage etc) and L-Tuner simulation. The results were written to a tab delimited text file which was loaded into Excel for the final analysis and graphics. One graph (the impedance spiral) was created from Excel data in Dplot.

Transmission line parameters come from the Transmission Line Loss Calculator or XLZIZL.

(Note that none of the results depend on approximations based on VSWR, in fact VSWR is not calculated during the analysis.)


Changes

Version Date Description
1.06 10/10/2005 Fixed error in "acceptability" of certain feed arrangements on 12m (W5DXP).
1.07 11/11/2005 Fixed error in calculation of losses in wet ladder line (VK3QD).
1.08 15/01/2006 Added Australian Figure-8 flex analysis (Figure 6a etc).
     

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