About a week ago, I saw this “UPC Unique wire tuner” on tweedehands.nl, a Dutch equivalent of ebay.com. As I’m very much into end-fed antennes because they seem to work well and are very easy to erect or span away, I decided to buy the good old 70’s made in USA L/C tuner. And what a pretty tuner it is, look for yourself:
and one huge roller inductor that seems to be of very good quality, probably around 25 to 35 uH. The roller inductor has a nifty two digit counter and on the inductor dial knob there’s also a 1-10 indication which allows for very precise registration of settings.
All the parts seem to be of very good quality and also craftmanship is very neat. The circuit is a typical L/C circuit. The tuner comes with handy plugs to short out one of the PL connectors to ground. At first I was a bit confused, but of course this lets you choose if you want the inductor in series and capacitor to ground (that’s what I’m familiar with and tend to call “LC configuration”) or vice versa (“CL configuration). According to PA0FRI this methode as similar but requires a little less capacitance so if a wire is tuned with full capacitance but that’s still not sufficient, the CL configuration might come in handy.
I did some testing in my backyard. I used a 7m stack on fishing pole with some length of wire that I once used for a 40m end-fed experiment. Our 1 week “autumn holiday” has just started and I had to take advantage of the nice weather:
As I can recall, the total wire length I used is around 14m of which 7m is coiled up somewhere in the middle on a 50mm PVC pipe (not visible in the picture). As a counterpoise, I used a random (I think around 8m) of speaker wire that I used for some kind of 40m holiday groundplane once. I connected the tuner input to my RigExpert AA-54 (7.1 MHz), set the capacitor to about 1/3 capacity and started turning the inductor dial looking for an SWR dip. When found, I increased capacity to see if I could get the SWR further down. If that was not the case, I decreased capacity and turned inductor to minimum SWR and so on. It turned out I could tune the L/C circuit to resonance on 80m, 40m, 20m, 15m and 10m easily. Once tuned, the Rx was (very close) to 0 which kind of confirms my understanding of this “tuned circuit”. Even though the wire is not resonant, the tuner makes up for the extra or missing wire and as a whole, it is a resonant circuit. Pretty cool!
Here’s some proof; an LCR graph of the wire tuned to 40m:
As you can see in the graph, the Rx = 0 somewhere around 7.120 MHz. And the impedance (Z) equals the resistance (R) meaning there is no imaginary resistance meaning the “system” is in resonance. I was surprised by the bandwidth by the way. Look at the SWR graph:
As you can see over the whole band, the SWR is better than 1:1:3 !! Probably because the wire I used was near a half a wave on 40m, so the wire itself was close to resonance as well (??? not sure about this ???). On 80m it seems to tune OK as well, look a the graphs below:
Again the R(esistance) equals the impedance (Z) on 3.7 MHz meaning the Rx = 0 and system is in resonance. Note that the bandwidth is much much narrower, only around 70 kHz for an SWR better than 1:2. Still, for /P use it’s nice to be able to tune a length of wire to resonance (it is resonant, not just 50 Ohms!). Same story on 20m:
Bandwidth is slightly better, a little over 100 kHz for an SWR better than 1:2 which is acceptable for /P use I guess. I got similar results on 15m and 10m. Bandwidth on 10m was better, very low SWR over full 1 MHz spectrum. Not sure why it’s so narrow on 20m, maybe somebody out there knows the answer to this question? Oh, of course I realise these measurements do not tell anything about the performance in practice, but I will be doing some tests coming week (if the weather OK) so will follow up this article.
On the previous pictures you probably noticed the fixed “door knob” capacitors on the back. As I’m no tuner elmer, it took me a little reasoning before I figured out what it did and how it worked. I actually raised the question on a local 80m QSO and while I was describing the circuit, I realised the extra capacitors could convert the L/C circuit (in C/L configuration) into a T-circuit with fixed output capacitors. Check the backside:
As you can see, on the input side (lower right), I’ve connected the counterpoise wire. The PL connector is the input of the system (that’s where I had the analyser connector). This input is connected to the variable capacitor, so currently the capacitor is in series with the input and I configured for a C/L circuit. The PL connector on the left side is shorted to ground with a neat jumper. The wire (ceramic insulator) is connected in between the capacitor and inductor making a typical C/L circuit.
Now note the extra jumpers in the middle top. Currently, these are shorting out the door knob capacitors but when one, two or both are set horizontally, you add capacitance to the output (before the wire). If the system is configured with C in series and L to ground, you add capacitance to the output making it a traditional T-circuit. The capacitance is fixed of course. I have figured out how to set the jumpers. Considering the fact they are 4 logical bits (on/off, true/false, 0/1); 2 bits with jumpers horizontally and 2 bits with jumpers vertically, I would have expected 8 combinations. Later I realised that “all off” is the same combination vertically as well as horizontally so 1 combination less leaving 7 combinations. With some reasoning I found only 4 settings:
both jumpers on (horizontally OR vertically): no output capacitance
lower jumper horizontally: no output capacitance
top jumper horizontally: 70 pF (2 door knobs in parallel in series with 1 door knob)
right jumper vertically: 100 pF (bypasses the two door knobs in parallel)
left jumper vertically: 200 pF (bypasses the single door knob)
Not sure when to use the T-circuit (probably when L-circuit cannot match the wire?), will found in practice I guess.