OCF Dipole with Balanced Line and Balun: Symmetry?

[ae4cw]

I am seeing non-symmetrical results when reversing the leads of the ladder line feed and need the wisdom of this board to help me see why this is happening or point out my measurement technique error. Here's the setup:

The antenna system is a 136.5' OCF (33.3%) dipole fed with 300 ohm ladder line (100 to 120') and coupled by a Elecraft BL-2 balun (switched 1:1 or 4:1) to 9" of RG-58x directly to the 4170. The LL runs perpendicular to the Dipole for 50' and slightly off-perpendicular (as much as 15 feet) for the remaining 50-70'. BTW, the BL-2 balun appears by measurement to be quite linear across the HF and low VHF spectrum

I made two calibrations: 1) the basic 4170 calibration using the supplied calibration BNC connectors, and 2) at the output of the Balun set to 1:1. The first is to measure the antenna "system" impedance as seen by the 50 ohm transceiver and the second is to measure the impedance at the input to the 300 ohm ladder line so I can calculate the loss in the line using TLW or TLD.

Here's the problem: I am seeing significant differences in impedances when the ladder line connection is reversed, i.e. simply swapping connections on the output of the balun. The differences are most evident below 10MHz and more pronounced as the frequency decreases. There are differences up through 30MHz but not as pronounced. I am seeing these differences using both calibration setting described above. With setting 2) the balun "should" be calibrated out, so I don't think this is the problem. Perhaps the feed-line being not perfectly perpendicular to the dipole could be the problem.

So, I'm stumped. Anyone have an explanation on why the asymmetry or ideas on errors in my measurement technique?
Thanks to all!

[k7buc]

Maybe the balun doesn't quite have enough inductance for the freq your scanning at? OCF's have a large imbalance, so it takes a pretty good balun to get rid of the common mode current flow to where its insignificant.
-Del

[wa3off]

I agree, the inductance may be too low. I'm not sure what core Elecraft uses for the BL2 balun. I suggest using the AIM4170 to measure inductance from the BL2's RF-in center pin to the A terminal with the balun in 1:1 mode. In this mode, the windings on each of the two cores are actually connected in parallel, halving the choking inductance. Let us know the inductance value you see and how it compares at various frequencies (tells something about the characteristics of the core material).

Next, terminate the balun (in 1:1 mode) with 50 ohms from A to B and scan the balun across the frequency range you plan to use it. Then, with the termination still in place, scan it again with the B terminal grounded. The results should be the same. Now for the big test, instead of grounding the B terminal, ground the A terminal instead. The results of this will show you how well the balun performs in a worst case situation.

I would also test the balun in 4:1 mode. You'll need to terminate it with 200 ohms for the tests I suggested.

EDIT: I did a bit more looking at the BL2's manual. From its picture and the intended use, my guess is that the core is a Amidon BN-43-7051. With 2 of those cores in parallel (1:1 mode) and 2.5 turns on each, I think the balun would be okay down to 80M, but below that it would be marginal. It looks like Elecraft's specs on the BL2 are measured in 4:1 mode where I would expect the balun to operate reasonably well down to 160M in the worst-case test configuration.

Dave

[ae4cw]

Dave and Del, thanks for your replies and analysis.

Dave, regarding measuring the inductance: are you suggesting connecting the 4170 BNC center pin to the BL-2 BNC center pin (without the shield) and then connecting the BL-2 A terminal back to the 4170 BNC shield? I actually tried this using a BNC banana adapter and the results were not what I expected, so perhaps I misunderstood your direction.

I did successfully run the tests with 50 and 220 (didn't have 200) ohm loads. The results with terminal B shorted to ground on 1:1 were very close to non-shorted results. On 4:1 the results were considerably different but not dramatically so. With terminal A shorted to ground there were significant differences at the lower frequencies. I've stored those in bitmap format and invite you to take a look here. The circle annotated lines are the unshorted results and the straight lines are the shorted results:
www.mediamax.com/ae4cw/Hosted/AIM%204170#

Please let me know what you think, and thanks again!

EDIT: I've now added bitmap SWR plots across the HF bands of the original and reversed ladder line connections to the balun in the url link above. As you will see the basic SWR/resonance patterns don't change dramatically, but they do change, especially at the lower frequencies!

P.S. Elecraft's specification rates the BL-2 from 500 KHz to 55 MHz with less than 1.25 VSWR connected to a 200 ohm resistive load. There is no mention of performance with reactive loads.

Chuck

[wa3off]

Sorry, I failed to notice your latest post until now.

Yes, the first test I suggested to measure the inductance doesn't seem to work vary well. You can get some idea of the inductance if you look at the lower end of the frequency range. Above that, perhaps the losses in the 43-mix core are causing the resistive component to dominate.

Your plot of 1:1 mode terminated with 50 ohms and the A terminal grounded are as I expect for 2.5 turns on a BN-43-7051 core. It's actually pretty good. I've compared it to a balun I have that is 6 turns of RG-316 wound on an FT240-43 core. It has a bit less inductance than what you would see with one of your 2.5 turn binocular cores, but more than you see with your two cores in parallel. The slight increase in Theta with increasing frequency might say that the BL2's wire pair when wound on the core has an impedance slightly higher than 100 ohms, but still pretty close to 100 (the two cores wired in parallel on the BNC side present a combined impedance of roughly 50 ohms). My balun with the 6 turns of 50 ohm RG-316 seems to present a transmission-line impedance of something closer to 60 ohms. When terminated with 60.5 ohms (no grounding of the terminated side of the balun), I see an SWR of less than 1.01:1 and Theta virtually 0 from 160M through 10M.

Given that your BL2 balun has what seems to be reasonably sufficient inductance for use at 160M, I guess I'm a bit puzzled as to why the difference you see when you swap leads of the ladder line is as large as it is. Makes me want to see what happens when I swap the leads of the 450 ohm ladder line I have between my antenna and a 9:1 balun I am using. My wire antenna is quite unbalanced, fed very close to one end. Either I got lucky the way it is or I can get some improvement.

Dave

[ae4cw]

Dave, I'll be interested to see if you see any difference with your unbalanced antenna as you swap the leads of your balanced feed line. Please keep us posted!

Chuck

[wa3off]

Well, I found out that my 9:1 balun is not a transmission-line balun. I guess it's a transformer balun. Since the balanced 450 ohm side and unbalanced 50 ohm side are isolated from each other, reversing the leads made no difference. So I connected the analyzer directly to the 450 ohm twin lead and did see a big difference when I swapped the leads. I'm still a bit puzzled by that since the laptop and analyzer were running on batteries.

Dave, WA3OFF