What I learned from building my own G5RV / ZS6BKW Antenna – Part 2

Part 2.

Back to the Beginning for a Second.

There seems to be many misunderstandings about the G5RV antenna. I am not sure why, but if my experience with it, is anything to go by, it is not hard to see how these myths are perpetuated.

As a bit of background, I’ve only recently upgraded from Foundation. Beginning hams like me, often choose the G5RV because it is often promoted as a simple multiband antenna, of course this requires a tuner (match). Some even make the claim you don’t need a tuner! Unfortunately many beginners don’t have the knowledge or skills to sort myth from fact, and build sub-standard antennas. Lacking the knowledge to deal with problems encountered we not only believe some of the myths, but further perpetuate them by sharing our experiences.

I Made a Bad Antenna Good- But How?
My own experience with my homebrew G5RV was interesting. At first my G5RV didn’t work very well. I made some changes to eliminate receive noise, and then after reading about the ZS6BKW. I change the antenna slightly to the configuration of the ZS6BKW and the antenna worked much better. But Why?.

My own failure, then subsequent success with the G5RV/ZS6BKW left me wondering what was myth and what was fact. As I began to understand more about antenna systems, I also wanted to know what changes I made that had resulted in eventual success with this antenna. I originally chose the G5RV for the reasons mentioned in my previous post. I, like many other F calls do, spent a bit of time trolling the net and came across the G5RV. I didn’t want to make a great big fan dipole for each band or a trapped dipole or anything like that, and with my lack of knowledge, I assumed there must be something special about the G5RV that made it work across all the amateur bands where as I believed a simple dipole would not. The final decision came down to the fact that the G5RV seemingly could work on all bands with a tuner and it fitted my available real estate.- my Icom IC703 has a built in tuner, so I reasoned that the G5RV would be suitable whereas a simple dipole which I imagined to be limited to the frequency of design, would not be. How wrong I was, and I wonder how many other beginners make the same mistake, assuming there is something magical about the 102ft doublet fed with 34 ft of open wire.

Sharing What I’ve Learned Along the Way
I now want to share with you what I have learned, and maybe along the way I can help dispel a few myths. I’ve spent quite a bit of time learning about antenna systems since I first put up the G5RV. I am no expert, in fact I am a real beginner, but I’ve learned enough that I wish to share some of what I’ve learned in order to help other beginners.

The Random Length Dipole
The G5RV is simply a version of what is known as a random length dipole. In order to understand the G5RV and the variant known as the ZS6BKW we need to first look at these random length dipoles in general.

A random length dipole is simply a dipole with a total length at least ¼ wavelength at the lowest frequency of use. The G5RV is simply a version of this. So that is the first realisation- there is essentially very little that is special about the G5RV versus any other random length dipole, including the ZS6BKW.

A centre fed ¼ wavelength dipole is 95% effective when compared to a ½ WL dipole. A quarter wave dipole will have a terminal impedance of  14-j1300 Ohms, which yields an SWR of  244:1 with 600 Ohm feedline and 424:1 with 300 ohm feedline. Such an SWR will result in very sharp tuning response. So what happens when the dipole is longer, lets say 3/8 WL.- then the terminal impedance is 34-j400 Ohms which yields an SWR of 25:1 with 600 Ohm feed and 24:1 with 300 Ohm feed.

So there is No Magic??
So what am I saying??? You can have any length dipole greater than ¼ wavelength at the lowest operating frequency, feed it with an appropriate open wire feed line and a tuner, and it will work as an efficient antenna?? Yes, that is exactly what I am saying. I am as shocked as you hihi.

Many amateurs believe the 102ft of wire employed by the G5RV to be somehow magical. So what is the significance of the 102 ft length of the G5RV.  Luis Varney, the inventor selected 102 ft for two simple reasons.

1.    He wanted multi-lobe radiation pattern at 14 MHz
2.    He wanted a low feed-point impedance at 14 MHz.

Warning Science Content
On the 20m band, the feed-point of the G5RV is at the centre of the central ½ wavelength portion. Hence the impedance is only moderately higher than if the outer ½ wavelength sections were eliminated ( leaving a ½ WL dipole). At mid 20m band, the free space impedance is 100+j0 Ohms and at reasonable height above ground 90 + j0 Ohms.  Remember Varney designed this antenna in 1958 when antennas were more likely to be fed with 70 to 100 ohm line. Today this results in a mismatch of 1.8:1 relative to the 50 ohms typically required by a transceiver. Pretty good.

So now to the 34 foot feed-line.  Varney chose this length because it makes a 1:1 impedance matching transformer. Varney probably used open wire feeder of somewhere between 270 and 400 ohms. On the 20m band therefore, the 90 ohms impedance is transferred to the feed line input, which in my case is the coax termination. On all other bands the impedance is much higher resulting in a much higher SWR.

So What Did That Mean to Me?
Obviously as I was an F call, I was not operating on 20m. Mostly I operated on 80m and 40m and occasionally when Es was working, 15 metres and 10m bands. I would have therefore experienced very high impedance at the feed line terminals with resulting high SWR. This would have been adequately handled by the tuner, so what else was at work?

Do I Need A Balun?
Now to the point about whether a balun is required or not. You can still read lots of web sites that tell you a balun is not required for a G5RV. I will quote once again W2DU.

“….thus with a suitable choke balun, to make a transition from balanced to unbalanced line, the low 1.8 :1 mismatch allows connection to a 50 Ohm line feasible.”

As you will recall from my previous post (Part 1 below) I had at one stage made a choke also known as an ugly balun, by winding 20 or so turns of RG58 around a six inch PVC former but it did nothing to alleviate the problems. So why did the choke not work?  Well as we will discover, such chokes are not very effective at frequencies below 14MHz. So on 80m band, it was probably of no use at all. What I ended up installing is a current balun, is a choke, but works more effectively for reasons we will discuss later. I just happened upon it in my efforts to reduce receive noise.

It is the installation of this balun that I believe resulted in the much-improved performance. But before I examine this, lets look at a bit more theory.

What Happens If I don’t fit a Balun?
Lets look at the no-balun situation and what can happen. To understand how a balun works, we need to look at the current paths at the feed point of a dipole. Because of their symmetrical relationship, the dipole arms couple energy of equal and opposite phase onto the feed line, thus cancelling out induced current on the outside of the feed line
.
What many of us don’t know is that there are three paths for current flow in a coaxial cable. I certainly didn’t know that until I began investigating this antenna. At radio frequencies, skin effect means there can be a current on the inside of the coax braid, as well as the outside of the coax braid. At RF these currents do not interact, they are separate. So you have a current on the coax inner, and two separate currents on the braid. So what happens when we transmit? Whenever current is flowing from arm 1 of the dipole which is connected to the centre conductor, back along the arm of the dipole to the centre conductor of the coax, the current in the opposite arm which is flowing in the opposite direction (opposite phase) separates where the braid meets the arm of the dipole. It flows up the inside of the braid, then separates. Some current flows along the arm of the dipole, the rest flows along the outside of the braid.

If the effective path length is an odd multiple of ¼ wavelength the impedance will be high, and the current flow low. If the path length is a multiple of ½ wavelength, the impedance will be low and the current flowing on the outside of the braid will be high. This results in unequal radiation from the arms of the dipole (inefficiency) and radiation from the feed line. Note the currents flowing inside the coax cannot produce radiation sine they are equal and opposite phase and housed inside the coax. However the current flowing on the outside of the coax does radiate and forms a third arm of the dipole. It is this third arm of the dipole that effectively changes the antenna length by adding the length of the coax to the equation, thus when no balun is present, changing frequency or changing the length of the coax will change the SWR.  The function of the balun is to block the current from getting to the outside of the coax.

An RF choke in the outer conductor of the feed line presents high impedance to the current that would otherwise flow on the outside of the coax. A simple choke balun made by turning several turns of coax around a  six inch PVC former provides enough series reactance to minimize this current if the frequency is between 14 and 30 MHz. Unfortunately it is not very effective below 14 MHz according to Walter Maxwell W2DU. So that is why my coil of coax (ugly balun) did not work.

Many commercial baluns are transformer baluns and voltage baluns. The G5RV antenna does not require a matching transformer– it already has one (the open wire feed line). It doesn’t require a voltage balun. If the open wire feedline goes straight to the tuner, it requires no balun at all. But here is the catch- if you connect the open wire or ladder line to coax, you will need to fit a 1:1 current balun. Since an air-wound choke is a kind of 1:1 current balun but since such chokes are not effective above 14 MHz, we need another way to create a 1:1 current balun. The W2DU balun is apparently a very effective way to do this. The balun I used came about through my attempts to reduce receive noise at my very noisy location. It just so happens, it was precisely the type of balun required to reduce the current flow on the outside of the coax.

So to the myths/realities of the G5RV

1.    SWR is low on all bands – FALSE.
2.    SWR is high on all bands except 20m. – TRUE. If built to specification the SWR at mid-band of 20m band should be 1.8:1 or thereabouts.
3.    The feed line acts as part of the antenna that is the feed line radiates.. FALSE. Equal currents of opposite phase run in the feed line. At the close spacing of the open wire feed line these cancel each other out. More about this in Part 3.
4.    The length of this feed line is critical. TRUE but not because it radiates, it doesn’t. It is critical because it is a matching transformer. What is not critical is the impedance of the open wire feeder, as long as it is somewhere around 270 to 450 ohms and as long as it is aprox 34 ft in length. More about this in part 3.
5.    If using coax between rig and open wire feeder, the coax must be cut to the specific length that gives lowest SWR. FALSE. If you are experiencing this, you have a problem, best solved by fitting a 1:1 current balun or W2DU choke balun
6.    A balun is not required for the G5RV. TRUE and FALSE. If you run open wire feed line from the antenna to the (balanced) tuner, then no balun is required. If you terminate the open wire feeder to coax (balanced to unbalanced) then you need a 1:1 current balun- not a voltage balun. If you don’t fit one, and your coax is a multiple of half wavelength at the frequency of use, you will experience degraded performance due to the outside of the coax radiating. This explains why some people say a balun is not needed, and others say it is. Antenna science says it is needed.
7.    The G5RV was made for the pre-WARC bands and is no good on modern bands. The ZS6BKW modification is better on modern bands. That my friends is the subject of Part 3. Hihi

73 de VK4MDX