OCF dipole tuning: balun down near the ground, twinlead to the feedpoint
Hello, I am homebrewing an off-center-fed dipole for 40m and shorter. It hangs between two poles about 100 feet apart, about 32' above ground. The two limbs are roughly 2/3 and 1/3 length of half-wave at 7.100 MHz. I took the design from W8JI and a little from W0HC. W8JI explained how the placement of the feedpoint allows you to change feedpoint impedance, to find best compromise in impedances for multiple bands. A third ham pointed out in an article that performance on 15 and 10m could be improved with a second set of limbs cut to 30m, fed from same feedpoint and hanging below the main limbs with some separation angle.
I'm trying that with my OCF and running into problems. My strategy has been to adjust overall length of the dipole to get resonance at the longest band, by taking up slack at the feedpoint. Then use that slack to move the feedpoint back and forth to get workable SWR on as many bands as possible. Using Autek RF1.
When I adjust overall length of the dipole to minimize SWR on 7.100, the SWR instead heads for the roof, so high a tuner can barely bring it in. Tried this with and without the accessory dipole for the higher bands connected to the feedpoint.
Note that I don't have a balun at the feedpoint. Instead, I have a section of twin-lead running from the feedpoint down to near the shack, and a balun right where coax comes out of the house from the radio equipment.
I am guessing that the length of the twinlead is adding to the length of the dipole as I get the dipole close to resonance. Would prefer not to have the balun up there at the feedpoint, with the weight of it and the coax to feed it. Any ideas on what I'm doing wrong? Would a specific length of twinlead feedline, some function of lambda at 7.100 MHz, prevent this?
Am also considering a means to get this antenna about 25' higher, it is probably an NVIS in 40m band.
73 de KF7CJJ
Last edited by KF7CJJ; 03-23-2012 at 02:46 AM.
Reason: added: feedline length, also ideal height vs freq
OCF's vary in feedpoint impedance all over the place based on where the feedpoint is. With a 33% off center fed as you describe, a 6:1 balun is the norm to try to match impedance to 50 Ohm coax. Some of the "great" OCF's 33% off center fed used 9:1 baluns.
Why would you use twin lead to feed a very unbalanced antenna?
I'd use a 6:1 balun or so at the feedpoint, and then coax. But in all cases, this is an antenna that will require a tuner.
A government which robs Peter to pay Paul can always depend on the support of Paul.
-- George Bernard Shaw
yessir, am using 6:1 balun and LDG tuner.
The antenna is made of 17ga aluminum electric fence wire. Strong, but I don't want to push my luck using it to hold that balun and the weight of RG58 (let alone RG8) up there. The lower loss of the twin lead is also desirable. My QTH has very strong winds.
I think if you want to use twinlead you pretty much have to center feed, I did the center fed 80m dipole with ladder and it worked very well. Dale
The balun is part of the antenna and you need to feed that with coax or you will have another unspecified Z transformation with twinlead between ant and tuner !
I would rethink that design.
I never liked "all band" dipoles because of the radical difference in X and Z from band to band.
I finally decided to experiment with a 90 ft dipole fed with TV twinlead and a MFJ bal line tuner. It does ok on 60-20M for me but I also have a fan dipole for 30/20M that requires no tuner !
My butternut vert has traps and stubs to make it "all band" and also needs no tuner.
FISTS #3853,cc 455
SKCC # 1395,tribune #12
Official US Taxpayer
An OCF will never work with twinlead as the feedline. The feedline will transform the impedance to something completely different from what you are expecting. It will also have very high common mode currents on the feedline and that feedline will be a radiating part of the antenna. If you want to read about what the feedline does, look at Cecil's website.
You can either center feed it with twinlead, in which case the feedpoint impedance will be very high and you can use the tuner to match it, or use it as an OCF by placing the balun at the feedpoint and try to get the feedpoint impedance to be a reasonable match for 50 ohms.
KC7CJJ to answer your original question yes there is a specific length of twin lead might improve things for the 40, 20, 15, and 10 meter bands. That electrical length is 180 degrees at 7.1 MHz.
What is the physical length? That depends on the VF (Velocity Factor) of your twin lead. Using a VF of 0.82 the length is 57 ft. The impedance at the OCF feedpoint will appear unchanged at the end of the 180, 360, 540, or 720 degree transmission line. IF it is resistive at the OCF feedpoint it will be resistive at the end of the transmission line.
Now what about the coax? That too will transform the impedance and it can be made 180 degrees at 7.1 MHz. For a VF of 0.66 (solid polyethylene coax) the length is 46 ft. For foam dielectric coax it is about 58 ft.
If you have an antenna analyzer, such as an MFJ-259, you can measure the VF of your coax and twin lead and cut them to the correct length. If you have only a VSWR meter you can still find the correct length by using a 50 ohm load.
Last edited by WX7G; 03-23-2012 at 03:19 PM.
If the antenna was at 70 ft height or higher and the half wavelength of feedline was routed straight down and there was a choke at the junction of the feedline and the coax (no 6 to 1 balun), you might have a chance of making this work on more than one band, although the SWR won't as good as with most OCFs. However with the antenna at 30 ft, that means more than half the feedline will not be vertical, and may be close to the ground. Since using twinlead with an OCF will result in the feedline being a radiating part of the antenna, that feedline will change the feedpoint impedance (in addition to the normal transformation due to impedance mismatches), and being close to the ground, and bent, it will change a lot, and that ground proximity will also reduce the gain of the antenna. If you play with the length of the feedline long enough you can probably get the SWR low on a single band, although you may need a small adjustment in antenna length.
Originally Posted by WX7G
You would be much better off constructing a conventional OCF.
One of the guys already told you - using twin lead is just asking for trouble!
Originally Posted by KF7CJJ
If you worry about the weight of the balun then why not use what's called a 'messenger' support wire, like the electrical service to the house? Get some 1/2" Dacron or similar rope and use that for the main span. Hang the balun from its top hook from the rope. Angle the antenna wire down slightly with maybe an 18" piece of 3/8 cheap grey PVC tubing with holes drilled where necessary to act as a spacer. This will handle the main wire and you can handle the 30 m wire the same as you've been doing.
Make sure it's a current balun! The ONLY OCF design that uses a voltage balun is the Carolina Windom, because it WANTS RF on the feedline! It uses a portion close to the feedpoint as a vertical radiator, isolated at the bottom with another balun to choke or block the shield RF.
Messenger wires work great! Takes all the worries away! Some guys just loosely twist the antenna wire around the rope, not bothering with the spacers. I wonder about rainy weather losses from the antenna to the wet messenger but if there's enough rope trailing from each end maybe I worry too much.
The best fit I can get in NEC for a 40, 20, 15 and 10 meter OCF is 68' long fed 27' from the end.
To keep the VSWR below 3:1 over these four bands it needs to be fed with a 4:1 current balun having a common-mode impedance of 2000 ohms or greater. The balun common-mode impedance was modeled as 1400 +j1400 ohms on each band. Higher is better. The balun is placed at the antenna and 50 ohm coax runs to the shack.
Last edited by WX7G; 03-23-2012 at 06:40 PM.