Monday, January 19, 2009

At last a good Zepp - or a J-Pole that actually works!

There seem to be a multitude of antennas based on the Zeppelin. Above HF they are usually know as J-Poles but whatever they are called the principle is the same. The basic antenna consists of a ¼ wave stub to match a ½ wave radiator. An antenna of this type is reasonably simple to construct for HF but due to the stub it tends to be a single band affair, also at ¾ of a wavelength its pretty long which isn’t very useful for HF.  

That said you could probably build it as an inverted L feed and it wouldn’t require a ground plane so it may have its uses.

Maybe a tuneable stub combined with a fan arrangement for radiating elements would work but at ¾ of a wavelength its going to be very long at anything below 20M!

On the VHF and UHF bands the concept becomes a lot easier to realise and there are almost as many J-pole designs on the web as there are web sites. If you need to understand the principles of how this antenna works have a look in the ARRL handbook.

For me three things attracted me to the j-pole  

1 – no Ground plane required - Quite an issue at 6M or below!

2 – it’s a ½ wave so exhibits more gain than a ¼ wave

3 – Its end fed so realising it as a vertical antenna is mechanically simple

Like many others these features prompted me to build several J-Poles; some from the web, and one from the ARRL handbook. To date I have been unsatisfied with the match I could obtain with any of the published designs  - finally I found one exception, a J-Pole design that actually works!

I have a reasonable J-pole on 6m which I use for spotting. I made it from aluminium pole with a steel whip at the top to reduce the wind resistance. However my 6m J-pole is a bit of a Heath Robinson affair. It performs okayish but although its tuned to 51 Mhz its still about 1.2:1 at its lowest point. This isn’t a big deal as the auto tuner in my transceiver can take care of the mismatch.

On UHF I have been getting by with a simple ¼ wave ground plane. I made this out of an N-socket and five pieces of insulated copper wire. This exhibits quite a low SWR and was light and easy to construct.  But of course I wanted a bit more gain and something a little sturdier so I started looking for a suitable J-pole design.

Eventually I thought would have a go at the design from WB8ERJ -

What attracted me to this design was his pragmatic approach, i.e. he built the antenna and then started adjusting it for best VSWR, the dimensions he gives work even if the dimensions seem odd. Most of the designs I have seen for this type of antenna make lots of false claims backed up by theory and K-factors and they have a lousy SWR, I really doubt whether half of them work as well as claimed. 

The most important thing to note in the design is the distance between the radiating element and the stub. This is ¾” between centres which works out as ¼”!!

To achieve this spacing I had to attack the t-connector and 90 degree bend with an angle grinder. After some careful soldering I had the copper cactus assembled. Pay careful attention to the dimensions! The measurements are to the centre line of the horizontal (Not that there’s much of it!) 

I tuned the antenna by attaching it to an old chair and adjusting the feeder until it gave the best match at 435Mhz (Centre of the UK 70cm band). I just held the feeder against the antenna for this exercise and marked the spot with an indelible pen.

The design suggests using jubilee clips but this isn’t very elegant and could corrode. To ensure this was reliable I used PTFE coax which can be connected to the pipe using some light work with the blow torch and solder (The same one I used for assembling the antenna). Obviously you still have to be careful, use a very small flame, get the solder on the pipe then the coax. N.B don’t inhale any fumes if you burn the PTFE insulaton!

I finished the antenna with a few turns of coax as a choke an used amalgamating tape to protect the end of the coax. The finished antenna was mounted on a short stub pole so its more than ¼ wavelength from my 6M j-pole and gives a near 1:1 match at 435 Mhz rising to about 1.2:1 at the band edges.

And yes it works very well, I have about 2s points improvement on the GB3BN repeater in Bracknell (About 30 miles away) and some repeaters I can hear for the first time. I notice the dimensions of this antenna are for 440Mhz so maybe if I increase the overall length slightly the bandwidth may improve?  I now appreciate the importance of the stub distance, maybe this is the real secret to successful J-pole designs?

I notice WB8ERJ constructed a two meter version, I’m sure it works just as well. Maybe I will pull down that 6M J-pole and see if  I can get it to match properly…