This is a recent project that I have carried out. A simple balun
A balun using ferrite ring
A balun using ferrite rod
Baluns - you either swear at them or swear by them!
So what do they do? They simply allow a balanced item like a centre fed dipole to be fed with an unbalanced system like coax (or balanced feeder (open wire balanced) to an unbalanced ATU) to try to reduce RF currents causing EMC or tuning problems. Now the next step is which one to use, a voltage or current balun?. Now I have only built and used a simple current balun so these thoughts are based on my experience.
Back in time before I held a license and was still a SWL, I built a trap dipole and needed a 1:1 balun at the feed point, saw a kit for a balun and sent off for it. Having got the kit and instructions my next problem was trying to understand what was meant by trifilar windings. I thus learnt it means three parallel windings (and of course bifilar is two windings!). Having understood this, I made it and it seemed to work. Later on I started to read up more on baluns and trifilar and bifilar windings. A look in Les Moxon’s book on aerials explained more and thus made me realize how simple baluns are to make. A 4:1 balun is even simpler to make than a 1:1 balun due to having one less winding to make!
Making a simple current balun
I feel that anything you make you learn a lot from both the practical and theory so here’s some tips to help you make 1:1 and 4:1 current type Baluns that will save you some cash! You need a ferrite rod or core, a smallish box, suitable connectors some wire (either hard drawn copper or heavy duty PVC covered different colour wires) fixings and your soldering iron, cutters etc.
Now to get the ferrite rod. You may have a defective MW portable radio, which a great source for the ferrite rod for a balun if not then a ferrite ring can be used.
The wire I have used is 18 SWG enameled wire.
Straighten the wire (tie one end around a door handle, pull gently until straight if you don’t have a vice) to remove any kinks etc and cut three lengths around 18’ long.
Lay the three straight wires in a triangular bundle with no twists and wrap in PTFE tape (as used by plumbers and stocked by most DIY stores) firmly but not too tight. (Helps to keep the wires straight and in one line! Wind 12 – 18 turns of the bundle onto the ferrite former (rod/ring) leaving NO GAPS between turns (wind slowly and carefully to ensure this). The method I used was to find the middle of the bundle and wind onto the former from this point towards the two ends. I then use a tie-wrap to hold the two ends in position on the former. If using a rod, ensure that the bundle usually lays flat on the rod but if not gently squeeze in the vice jaws taking care not to break the rod or scratch the enamel). Now define which end of the windings are your start and which is your finish. Scrape off the enamel on the end of the separate wires, buzz through to identify the individual windings (A, B & C) and connect up as per circuit diagram. (I show the connections for both 1:1 and 4:1 and using a switch you could make a 1:1 and 4:1 balun in one unit) If you are making a 1:1 balun connect a 50 Ohm dummy load to the balanced side (or output side) and aerial analyzer to the unbalanced (or input side) via a short coax lead. (If you're using a low power transmitter ensure the 50 Ohm load can handle the power, 4 x 220W 1w resistors in parallel should be ok) to check the VSWR. (Likewise if you built a 4:1 balun, put a 200 Ohm load on the output or balanced side). If the measured VSWR is high i.e. greater than 2:1 check your wiring!
I used a simple ABS box to mount the transformer with a SO259 connector on the input and bolts on the output for the aerial feeder or elements.. The transformer can be glued in the bottom of the box for mechanical strength.
I have used both ferrite rings and ferrite rods. The pictures, I hope explain more. I also attach a Excel plot of one balun I made, fairly flat upto around 33 MHz!
And to prove it here is a picture of it being measured on 29Mhz.
Info on types of cores
Plotted response curve