Based on comments and experiments from Monk ZS4SF, I was interested to see if I could build a larger coupler for my home deployed EFHW using a large toroid T-200-2 which I happen to have on hand on my work bench. This would allow me to use higher power and in theory would be more efficient than my field coupler which uses a small T-50-2 toroid. I learnt a lot during the process.
I realized by looking at the XL/XC reactances at my target resonant frequency of 7020Khz that the challenge was going to be the turns ratio and small capacitance. I needed a 10:1 ratio to convert from the 5000ohms expected at the end of the EFHW to 50ohms on the link coupling. I started with 30 turns of hookup wire on the secondary and 3 turns on the primary. This resulted an inductance in the tank circuit of 11uh and a small capacitance needed to resonate the circuit at 7020Khz of about 47pf (Xc/Xl=485ohms). I realized the capacitance by using a piece of RG58/U coax connected in parallel with the inductor. This having a capacitance of 28.8pf/ft. Thus the coax piece was about a 1.6ft long. The circuit was very unstable. Small amounts of capacitive coupling resulted in a major shift in the resonant frequency.
I experimented with a number of turns ratios and discovered that any combination resulting on a coax length of less than about 2 feet was unstable due to capacitive coupling effects from surrounding objects including my hand.
I reduced the turns down to 18Turns on the Secondary (4uh). I was able to get a good match using 2turns on the link coil and by using some stiff stove type wire, insulated which I bought from Builders Hardware. This required a coax length of 4 feet (~115pf). I was able to get a 1:1 SWR using a 4.7K resistor as the terminating impedance on the bench. The circuit was stable and showed no capacitive or inductive coupling effects to nearby objects. This shows approximate values of 4uH for the inductor and 115pf for the coax capacitor to yield a resonant frequency of 7020KHz.
When I connected the EFHW wire to the coupler I was able to obtain the best match at 6.8Mhz by driving the system with my MFJ259B antenna analyzer. By tuning the analyzer up to 7020Khz I noticed a change in SWR to about 1.6:1. Since I had already resonated the tank circuit at 7020KHz against a pure 4.7Kohm resistor I was hesitant to tweek the capacitor by trimming it shorter. Instead I trimmed about 6 inches off the EFHW wire and was pleased to see a 1:1 SWR match on 7020Khz. This confirmed my feeling that the wire was showing a 5000 ohm pure resistance at resonance (in spite of being pulled through a tree at one end!). In looking at the write up on the web by AA5TB http://www.aa5tb.com/efha.html I noted that an EFHW will exhibit a 5000 ohm resistance with a 0.45 wavelength counterpoise. I must conclude that the mass of the antenna analyzer and the short coax connecting to the analyzer of a few feet is sufficient to couple the system to ground and simulating a quarter wave counterpoise (I really don't know however). I also experimented by linking the 'cold' end of the secondary to the 'cold' end of the link coupling. I noted with interest that this had no effect at all on the SWR of 1:1 at 7020Khz. The system is quite stable.
Much more experimentation is warranted on this interesting antenna. Especially for that case where the rig may not be near the ground or coupled to ground in some way. In this case a short counterpoise will likely be needed. Also the effects of different deployments and wire length variations on SWR is room for much experimentation.
Experiment to try in future. Orient the rig and coupler well above the ground in the air so that there is no (little) coupling to ground. Then measure the SWR variation. Then attach a 0.05 wavelength counterpoise to the 'cold' end of the secondary. In theory the antenna end impedance should drop to about 1800 ohms. This would require a 6:1 turns ratio. In other words another turn on the link coil to bring it to 3 turns and 50ohms match. According to the AA5TB graph, at this counterpoise length the reactance should be zero at the end of the wire and the system should be stable.
Observation. The practical point of this type of experiment would be to see if it is possible to use a coupler in the field with fixed components. This would greatly simplify the deployment to a blackbox that requires no tweeking. Furthermore....could this coupler be integrated into a single box along with the transmitter and receiver. In other words how critical is the length of coax between the coupler and the rig. Is this acting as a counterpoise in reality?
Bandwidth. The bandwidth of my system at the 2:1SWR points is approximately 145Khz. What is the effect on Q of different component uses and combinations (thicker wire and air variables)? For me 140KHz bandwidth is more than enough. Referring to a previous blog where I define the scope of my field operation I would trade bandwidth for a better Q since this (I think) would mean less resistive loss in the coupler (I think). Of course a very high Q would mean a less stable coupler from the point of view of field deployment and the effect of nearby objects on the system.
|Bench Test using a 4.7Kohm terminating load and coax capacitor|