First, set your traps


I want my antenna to cover the 6m (50 MHz) band as well as the usual HF bands.  My FT-847 can route the 6m band to the HF antenna socket (instead of the dedicated 50 MHz socket), and my YC-20 matching unit can also deal with 50 MHz.

The trap design program tells me that I want 2.8 turns of my coax on my 2.8 cm o/d former, for a total of 31.45 cm of cable.  Let’s say 2.75 turns, with slightly more coax poking through into the centre of the former.  I’m going to cut my coax at 31.5 cm, and wind 2.75 turns, and see where that leaves me.

Well … it resonated on about 48 MHz.  By pulling the turns apart, I can get it up to 50.5 MHz, but that is not really satisfactory.  One of the problems must be that where the coax dips down into the former, it does not do so perpendicularly, but tends to continue in the direction it was going.  So effectively it adds an eighth of a turn or so.

According to the design program, the turns sensitivity is about 2.1 MHz per cm.  But the end sensitivity is 0.75 MHz per cm.  So I reckon if I drop down into the former about 0.75 cm earlier at each end, it should raise the resonant frequency by about 2 MHz.  So that will be my next attempt.

BTW, finding the resonance with my new MFJ-259B is not easy.  The GDO coils are not very effective, and it is difficult to judge the resonant point.  However, the combination of my old Cirkit dip meter, and the frequency counter in the MFJ-259B, is just what you need: the Cirkit dip meter to find the resonance, and the MFJ-259B to read the frequency.

First attempt at a coaxial trap


As a first attempt, I tried making the 28.85 MHz traps from the original NU3E design. Four turns on my 35.5 mm former for a total of 51.4 cm. The slightly larger former meant that I did not have much coax poking into the former. But, the worst bit was that it resonated on about 25 MHz, not the 28.85 MHz it was designed for.

Not being one to waste the effort, I took the coax off the former, cut a 54.0 cm section, and wound it back on, making a trap for 24.940 MHz in the 13 m band.  I’ll come back to that later.

The next task was to work out why.  My suspicion fell immediately on the coax, so I cut a 2 ft (61 cm) section and measured it on my capacitance meter.  66pF.  Standard RG58 is supposed to be 30.8 pF per foot, so we are about 10% high.  Just to be sure, I cut a longer section, 9.95 m, and measured it.  1052 pF, or about 32.3 pF per foot, or 106 pF per meter.  I also check quite a number of fixed capacitors just to be sure my meter was reading reaqsonably accurately, and it was.  BTW, I am asuming the frequency my meter uses for this test is sufficiently low to avoid any transmission-line effect.

Looking at various cable catalogues on the web, I am surprised what a range there is to the capacitance per meter of nominally RG58 cable.  I have seen figures from 79 F per meter to about 108 pF per meter.

So: capacity of cable = 106 pF per meter.

Does it matter?  If I use the trap design program, I should be able to use any cable. However, if my capacitance is greater, then the inductance will have to be less to achieve resonance, i.e. less cable and fewer turns.  This means the shortening effect of each trap will be rather less than in the original design.  We shall see.

The NU3E Attic Coaxial-Cable Trap Dipole


For some time now, I have been wanting to get back on the air with a reasonable selection of HF bands.  Not having the courage to provoke the neighbors by mounting a 5-element multiband beam on a huge mast, I have a selection of wire antennas shoehorned into the attic.  It isn’t very satisfactory because there isn’t much space.  The ridge of the house is about 7.5m long and runs about NW-SE (actually aboput 150° – 330°).  There are about 11 m from eves to eves, running about NE-SW (60° – 240°).

I am starting off with two dipole antennas, connected to the same balun:

  1. A 20m dipole runs like an inverted-V under the slope of the roof.  The dipole runs NE-SW, and so it fires its maximum signal at 330° and 150°.  This should be OK for North America, especially Canada, for Africa, and for Australia via the long path.
  2. A 40m dipole runs horizontally under the ridge of the house (NW-SE) for about 4 m on each leg, then drops vertically down about 3.5m to the floor of the attic, then wonders aimlessly to one side for a further 2.5m.  In theory I can use this on both 7 MHz and 21 MHz.  In fact, it seems to give a better SWR on 21 MHz than on 7 MHz.  In theory, this antenna should shoot its signal at 60° and 240°, making it OK for South America, Asia, and Australia via the short path.

What I really want to do though, is to get on 80m.  My first QSO ever was on 80m, on 29th April 1972, with G3FXC.  Alf Watts was ADC for scouts, and was something of a mentor to me when it came to Amateur Radio.  (He also tried to teach me to roll a canoe, a skill I have never mastered to this day.  I always ended up pulling the tabs on the cover and swimming my way out.)  Sadly he passed away a couple of years later, but I do owe him for sparking my interest in this hobby.

Getting on 80m will be a hard task in the limited space, so I have decided to try and build a trap dipole.  Trap dipoles have two huge advantages: they are multi-band, and they are shorter than full dipoles.  I found an interesting and easy-looking design by John DeGood, NU3E.  Click here for a description of the antenna.

 Trap materials

The first difference is that here in Europe I cannot get hold of exactly the same formers that NU3E used.  The nearest materials I can get at my local DIY centre are:

  • 28 mm o/d couplings designed to join 25 mm electrical conduit.  At 1.1″, this is considerably smaller that the formers used by NU3E, but they may be useful if I want to add the 6m band to the original design.
  • 35.5 mm  o/d couplings designed to join 32 mm electrical conduit.  At 1.4″, these are near enough to NU3E’s 1.375″ couplings.  They are 10 cm long, which is rather longer than I need, but they have a groove round the middle, which makes it easy to saw them into two formers each 5 cm long.
  • 44 mm o/d couplings designed to join 40 mm drain pipe.  This works out to 1.693″.
  • 53 mm o/d couplings designed to join 50 mm drain pipe.  This works out to 2.090″.
  • 78 mm o/d couplings designed to join 75 mm drain pipe.  This works out to 3.071″

I also have a roll of coax that I think is RG58, but I am going to have to do some research on it.

It is evident I am going to have to tweak the traps a bit, so I wrote to NU3E for some advice.  He pointed me at a useful little program for designing traps.

OK, that’s it for now.  I actually have got quite a bit further in my experiment, but I need to take a break from the keyboard.  To be continued ….


Doing some research about my new MFJ-259B antenna analyser, I came across a really interesting video magazine, It has lots of interesting articles relevent to Radio Amateurs. In particular, episode 21 has a great review of the MFJ-259B at about 0:37:40 onwards.

Hello world!


Welcome to my new blog.  Some of my readers will know my CCIE Study blog, and will have come here out of curiosity, or because they share my interest in the hobby.

I first got interested in Amateur Radio through the scouts and “Jamboree On The Air”.  I was first licenced in 1972 at the tender age of 15, with callsign G4AZO.  In 1993, I moved to Luxembourg, and converted my licence to a local callsign, LX2KD.  However, I still keep the UK licence too.

I enjoy experimenting with radio, mainly HF, and that is what this blog will be about.  Currently I have lots of antenna projects that need writing up, and I shall post about them in the next few days (or after the Cisco Networkers conference).

So,  here goes ….