(updated 2 April 2020: added 15m trap)
After some thinking, I decided to build a dedicated HWEF, just because I can. I did some measuring and calculating on different cores (also read my article “What toroid(s) to use”), I decided on a single FT240-52 toroid with 3:21 ratio. Again, just because I was curious I made a true auto transformer where the first three primary turns actually are the the first three secundary turns. In other words, it’s just 21 turns and primary is tapped at the 3rd turn. Because I don’t have any experience with this material, I made it 3:22 with taps at the 20th, 21st and 22nd turn.
So there it is. As you might know I’m not really a good DIY’er but I really did my best and this is how it turned out. So no seperate (twisted) primary and because the input is on the other side of the output I decided not to pass through the core like the conventional 1:49 transformer designs… No scientific reason, just curious. While it looked pretty good, I put it in an enclosure and soldered 2k5 Ohms carbon composite resistor between the output and ground.
Yes my soldering is terrible and again (my trademark) I touched the enclosure with my soldering iron but you can’t see that when the lid is on HI. Looking good, I was really looking forward to testing it with 10.1 meters of wire but I’m glad I first did some measuring with the 2k5 resistor because…
…it’s terrible!! (HI). The SWR graph doesn’t say much but in this case enough. The R(eal) part of the complex resistance was around 50 Ohms, but the X is way too high on 14 – 30 MHz so back to the drawing board. I’m planning to update this page in very near future hoping I will be able to build a better solution. I’m not even going to test it with 10.1 meters of wire, going to make a new transformer the conventional way and hoping it will work better.
It’s the next day, I made a new transformer on a single FT240-52 with a dedicated 3 turn primary and 21 turn secundary coil:
At first the spacing of the primary and secundary coil where as narrow as possible (green line in graph below), not very good. The red line is with secundary coil widely spaced, the orange line with primary widely space and pink line with primary even more space.
Results are worse than yesterday. I do notice however that R is and stays around 50 Ohms but the imaginary X part is way too high. Optimum seems around 6.2 MHz (about the same as yesterday) and that’s not where I want it.
So next stap was building a conventional transformer with 2:14 ratio on a single FT240-52 with primary turns twisted in first two secundary turns. Here it is…
I did some fiddling with the coil spacing, here are the results:And here are the results:
Now that is a pretty picture… The blue graph is with maximum spacing on first two twisted turns + four extra secundary turns and maximum spacing on seven secundary turns (bear in mind passing the core is one extra turn!) on the other side. The red line is with narrow spacing on the first two (twisted) primary/secundary turns, the pink line is narrow spacing for input side. Making spacing narrow / wide on the output side didn’t seem to make much difference.
The day after I tested the system with 10.4 meters of wire (about 20cm folded back). The results were *very* promising to say the least. Nice low SWR right in the middle of 20m band and low SWR around 28.5 MHz. I noticed that on both bands, the R was on the low side, around 40 – 44 Ohms. This means the transformer has too many secundary turns, it needs to transform a little less. Therefore I opened up the transformer housing and took one secundary turn off making is 2:13 so impedance transformation is now 1:42 instead of 1:49. I’m very happy with the results:
As you can see the X=0 and R around 50 Ohms on 14.175 and 28.500, exactly where I want ‘m. I’m going to put up the feedpoint a little higher; the wire is sloping in 45 degree angle in South-West direction. I did some comparing with my Hy-Gain 12AVQ, on most stations the HWEF is better and (!) quieter. Compared to the 80/40 HWEF (higher up and less sloping so probably lobes in N and S direction) some signals are better (I think from the West) and some a little less (signals from the South, this can be explained by how both antennas are spanned (is that a good English word?) away.
I’m very happy with the results, I will be testing 20m tomorrow and this weekend, hoping for some openings on 10m as well. Already thinking of taking the 12AVQ down and replace it with a dualband 15/10 end vertical, the project that failed but started all this tinkering with FT240-52 cores.
Stay safe es 73 de PA3HHO
As promised I would update this article with the experiment of adding a 15m trap making the HWEF half a wavelength on 15m (wire before trap) and half a wavelength on 20m with some loading (the 15m trap is inductive on 20m) so a little less efficiency and bandwidth compared to the full 10 meters of wire.
Because traps are kind of universal and not specific for the HW EF, I published that article onder “projects”. If you’re interested in how I made the 15m trap, click here: 20m L/C trap on Fritzel isolator.
So I’ve added the trap… Coincidently, I cut the wire to the exact correct length for 15m, I think around 7,4 meters. The tail piece of wire took some more cutting, first it was too long, then I cut it too short, added a little pigtail but too long again, clipped a few cm’s and now I’m happy.
The 15m SWR graph (return loss around 19 dB on 10m with the full wire it was 29 dB):
And here’s the 20m SWR graph (return loss around 31 dB)
Impedance on 15m is a little under 50 Ohms, impedance on 20m is a little over 50 Ohms. I reckon if I’d add a 10m trap, the impedance will be around 40 Ohms. I’ve seen this phenomenom before with my holiday vertical EFHW antennas.