Sunday, April 9, 2017

Project - Cobwebb (or cobweb) antenna


Living in a densely populated area antennas are an issue here. Officially I can place one antenna provided it is situated at the back of my house and not higher than 5 meters above the point where the antenna or mast meets the roof line.

The challenge: how to get good results on multiple HF bands from the home QTH with an antenna that is not too much in your face. Following the official rules I could still install an enormous beam (provided the width fits my 9m wide lot). However this would create problems with the XYL, neighbours and probably the official body that has a say on the aesthetics of any change you make at or around your house.

In 2010 I worked mainly using a Ventenna dipole in my attic (inside the house) - not a real DX killer. In 2011 I started using an end fed half wave wire that I set up sloping from my attic window into the garden. This worked a bit better but I need to change the wires for different bands and I keep on getting better results when I work portable with the same end fed wire vertically on a fibreglass mast.

Reading a lot of enthusiastic stories on the Cobwebb antenna (originally by Stephen Webb, hence the weBB) and considering its dimensions I thought it would make a good compromise for my situation.

There are two main configurations for the Cobwebb: the original T-match version as developed by G3TPW and the 1:4 balun version as developed by G3TXQ.
Comparing the two versions it appeared to me that the G3TXQ version was easier to build and tune with no apparent disadvantages.

A few good sources I used:

After soaking in all the information I set out to get components that would make the antenna less visible, sturdy enough to keep the antenna in one piece in windy conditions and that would allow me to use the antenna as a portable antenna if I would choose to do so.

The following are my main building blocks (click on the pics for a larger version):

1. Support poles 2. Antenna wire 3. Center plate 4. Metal tube (cut)

FT-240 toroids,
RG-316 coax,

5. U-bolts 6. Steel angle 7. Balun box 8. Balun parts

  1. Support poles - looking at the various fishing poles used by people building this antenna I was not too enthusiastic. The fibreglass tubes used in fishing poles are rather thin. This means you have to either choose a large diameter or accept drooping poles. It turned out however there is a fibreglass producer in the Netherlands that sells tubes through a webshop. I visited them to find the best diameter versus stiffness. It turned out to be a 16mm diameter tube with 2mm thick walls. I bought 4 x 2m + 1x 1m of the stuff.
  2. Wire - I chose to get me very thin yet very strong wire from DX-Wire in Germany. Aiming at low visibilty yet flexible wire (for portable use) I selected the UL version of their wire that has an outer diameter of 1,5mm. I was already familiar with DX-Wire through my various end fed wire antennas.
  3. Center plate - after looking for some sort of plate - preferably aluminium - I came across the plate shown in the picture. It is a very thin metal plate called a "nailing sheet" over here. The way I was planning to connect the support poles the thickness of the plate is not very relevant. The advantage of this plate is that it holds a lot of holes that can be used to affix parts and that it is very light (good property for a portable antenna).
  4. Fixture for the support poles - I found a 50cm (20") metal axis with an inner diameter of 16mm. This suited me nicely as the fibreglass poles were just under 16mm outer diameter. The idea was to fix the metal tubes on the center plate and insert the fibreglass poles into the metal tubes.
  5. U-bolts for the metal tubes - not having any u-bolts of a convenient size handy I decided to take a length of M5 (metric) wire and cut and bend that to form U-shapes
  6. Steel angle - to be able to attach the center plate to a mast I included this piece of metal. My idea is to use the vertical strip for holding a u-bolt. For a more sturdy (permanent) setup I can extend the strip adding a plate the same width (that allows for more u-bolts) or adding a strip that is wider (that allows larger u-bolts).
  7. Junction box - the antenna needs a 1:4 (12.5 : 50) balun. The picture shows the box already prepared for the balun. You will find more info on how to construct it below - after the center plate construction info. I chose a junction box of  110mm x 110m x 60mm (4.3" x 4.3" x 2.4"). If you are looking at different sizes of boxes: the minimum I would recommend is 110mm x 100mm x 50mm. Note that I am constructing a QRO balun and therefore use large toroids.
  8. Balun parts - for the QRO balun you additionally need: an SO-239 chassis connector, two toroids (FT-240) and 2.5m (8') of RG-316 coax.
..and a hand full of washers, nuts, bolts and tie wraps.

Construction of the center plate

2. adding u-bolts 3. adding the metal angle (mast support) 4. adding tubes and center bolt

  1. I cut the metal axis in three pieces: 2 x 20cm (to hold 2 support poles each) and 1 x 10cm (to hold the support pole for the balun). 
  2. I started adding my home made u-bolts through the right holes in the nailing sheet. I had to enlarge the holes slightly (from 4 to 5 mm).
  3. I added the metal angle that would go under one of the metal tubes, using two small bolts to keep it in the right position.
  4. I added the two larger tubes, drilled a hole through the whole stack and put a bolt through it - this is not really necessary to keep the tubes in place (the u-bolts will do that) but the center bolt marks the end of the insert for each of the four main fibreglass poles.
  5. I added the fifth metal tube. To attach that one I used one u-bolt and one bolt through the end of the tube. I chose this configuration as there was not enough room for two u-bolts
Construction of the balun
The G3TXQ version of the Cobwebb requires a 1:4 balun. I mistakenly grabbed a commercial 4:1 balun I had from an abandoned antenna project. Although called a "1:4" balun it was meant to be used to transform a high impedance antenna feed point down to 50 Ohm. Browsing the web I found that 1:4 and 4:1 are frequently used to describe the same balun - not the one you need for this antenna (you need one that is going to transform the low - 12 Ohm - impedance of the antenna up to 50 Ohm). Time to build the Guanella current balun described in detail in G3TXQ's website...

Note that G3TXQ describes a 100w balun. If you are building a QRO version like me you will need different toroids. Luckily NH7RO built one as well and shared the info.

3. RG-316 on toroid 6. stacked chokes 8. tube clamps 10. balun in box

  1. Cut four pieces of RG-316 of 540mm (21.25") each and strip the ends
  2. Lay two pieces along side each other (use heat shrink tubing or tape to keep the two pieces together) and connect the ends braid to braid and center to center
  3. Wind the double RG-316 with 7 turns onto one of the toroids, creating a choke
  4. Repeat step 2 and 3 for the remaining two pieces of RG-316 and the remaining toroid
  5. Stack the two newly created chokes in such a way that the coax ends of both chokes are aligned and bind the chokes together (using tie wraps)
  6. Connect the coax ends in two different ways:
    1. On one end connect the two combined coaxes (or 4 coax ends) braid to braid and center to center. This end will be used to connect to the antenna wires ("the antenna end").
    2. On the other end connect the braids of one choke to the center of the other choke. The two remaining "free" ends (center and braid) will be used to connect to the SO-239 connector ("the feedline end").
  7. Add two short pieces of wire (of equal length) to the antenna end of the balun - so you will be able to bring these connections to the side of the balun box.
  8. Prepare the balun box: add the SO-239 connector, two bolts to opposite sides that will hold the antenna wires and any clamp to the bottom of the box that allows you to attach the balun to the fibreglass tube later on (refer to picture 8 for a suggestion).
  9. Solder the feed line end of the balun to the SO-239 connector 
  10. Attach the two wires of the antenna end of the balun to the bolts in the sides of the balun box (I used cable shoes) 
This concludes the balun construction. I found connecting the tinned center braids the most challenging part as they tend to break off easily.

Putting it all together
Adding the fibreglass poles is very far from rocket science once you have the center plate. So I will refrain from going into detail. I chose to let the antenna wires keep the poles in place - so I did not affix the poles to the metal tubes.This way I should be able to take it apart relatively easily if I want to use this antenna /P. Time will tell if this is sturdy enough to withstand a storm.

The frame for the antenna wires is now ready: a cross of fibreglass of 4m each side and a fifth shorter leg (1m in my case) holding the balun. Now it is time to add the wires.

G3TXQ has a very handy table of wire lengths he needed to get his antenna tuned (including the center to wire fix point length along the fibreglass pole). I repeat it here adding the metric numbers in as well:

Band Wire length Wire length (metric) Centre-to-corner length Centre-to-corner (metric)
20m 201" 510,5cm 72.25" 183,5cm
17m 157.5" 400cm 56.75" 144cm
15 135" 343cm 48.25" 122,5cm
12m 114" 289,5cm 40.25" 102cm
10m 100.5" 255,5cm 34.75" 88,5cm

Note that you use 2 wires for each band - the numbers represent one leg of the folded dipole. I added a few centimetres when cutting the wires making sure both dipole ends were exactly the same length. It is always easier to tune upwards cutting the wire than it is to tune downwards (never found an easy way to lengthen a wire ;-)

With 10 wires of the right length you can start affixing them to the fibreglass cross, one tube after one tube, one wire after the other (starting at the 10m wire and working outwards kinda makes sense). I used the centre-to-corner lengths specified above and marked them on the fibreglass tubes. That way I knew where to (approx.) wrap the tie-wrap around the wire and tube.

I then added cable shoes to each of the wires at the balun to make for a neat connection of all the wires to the balun:
Balun with all the wires connected

The other end of the wires are the ends of the folded dipoles. They need to stay some space apart. The length of the wires and (therefore) the amount of space in between will change during the tuning process. To keep the ends at an adjustable distance I tried several things but finally went for electrical strip connectors (single pieces) and tie wraps.

Dipole ends connection before and during tuning

For every end of wire I folded over a piece of the wire, slid the connector over it and screwed it tight. That way I created small wire loops on all wire ends that could be connected using long tie wraps. By folding over too much wire at the beginning of the tuning process I knew the wire ends would only get closer to each other during the tuning. That is what you want if a tie wrap is keeping the ends together (as a tie wrap is easily tightened but not so easily loosened).

After the tuning I removed the electrical strip connectors and replaced them by heat shrink tube and a small tie wrap, like so:
Dipole ends connection after tuning

Whatever you select as the material to keep the wire ends apart be careful that the material does not soak up water and influence the SWR of the antenna.

Tuning the antenna
I tuned the antenna today (April 22, 2012) using only the built in SWR meter of my FT-857d (very doable). I started with the 10m band dipole. I read about different results people got when changing the effective wire lengths either by folding over more or less wire or by cutting of wire. On the 10m dipole I cut most of the extra length. On the other bands I left most of the excess wire. I think the latter approach is the best as you can always tune down the antenna if you ever need to.

Note that I found the antenna can be tuned by only folding over more or less wire (no cutting).

The Cobwebb during tuning
From 10m I went outwards to all the other dipoles. I did not find a noticeable change in the resonance of the dipoles I already tuned when tuning the other dipoles.

First live tests
During the tuning process I tried to get some signal reports and did some comparisons to my tried and tested half wave end fed wire I set up vertically. My mast was not sturdy enough to carry the Cobwebb on a height above 6m (18ft). Therefore I focused on 10-12-15m to at least have some relative height.

Conditions were rather poor on the higher bands with deep QSB. Three Italian stations came back to my calling on 15m but either the signals or their fluency in English was just not good enough - anyway there was no opportunity to get different reports on the two antennas. On 17m I heard and was able to work EY7 (5000km) using the Cobwebb. The first DX on this new antenna!

10m came slowly alive but I had to pack up soon so I just compared the signal levels on the Cobwebb and the vertical end fed wire (using a handy antenna switch). The QSB made it difficult but switching quickly back and forth I found that both antennas were most of the time very close (not even 1 S unit apart), sometimes 2-3 S units apart with no clear winner - sometimes the vertical was stronger, sometimes the Cobwebb. There was not enough time to find a pattern (like distance or antenna used by the other side). I will need to do more testing to establish that.

What this first test shows me is that the Cobwebb is a very practical multi-band antenna (my end fed wires need to be changed to cover the 5 bands) with a very acceptable performance. It also shows me that for a quick portable setup the end fed wires set up vertically on a fishing pole are hard to beat.

Cobwebb in action during and after tuning

No comments:

Post a Comment