DJ4VM Quad

[Steve - K4HX]

As has been discussed numerous times here, using even a 40 meter dipole fed with open-wire line (OWL) on the higher bands will result in pattern break up (more narrow main lobe, numerous smaller lobes and the associated nulls). Although pattern break up is not the end of the world, it can produce less than optimal azimuthal coverage on the higher bands. So, what to do?

Check out the DJ4VM Quad. It's really a quad split in two as shown in the attached drawing. Each side is one-quarter wavelength long. OWL or ladder line is used to connect the top to the bottom. The feedline, OWL or ladder line, is connected the the points marked X. This system maintains a very consistent fat figure-8 pattern on all bands from 20 through 10 meters. The gain increases with frequency amounting in 2-3 dB more on 10 meters compared to 20 meters. Check out the attached azimuth patterns for 20, 15 and 10 meters.

[Tom WA3KLR]

I’ll throw this out as maybe this will be of help.  I think most guys know that a dipole used at higher bands than the half wave resonance particularly 4X frequency and above, the azimuth pattern breaks into lobes with nulls greater than 25 dB.  The over-used G5RV dipole is based on this principle and he knew how to orient the antenna to put the lobes for specific higher band skeds (20, 15, and 10) to the target bearings.  

But for general work the nulls can be a killer.  A few years ago I discovered using simulations that a dipole used at higher bands you can reduce the depth of the azimuth pattern nulls by having the ends of the dipole hang down vertically.  The configuration is a lazy C with only the center 1/3 of the antenna horizontal.  Interestingly the top-half of the DJ4VM quad is the same thing.  This configuration also had the advantage of not needing as much horizontal real estate.

The Radio Works guy, Jim W4THU, notes in his literature that a dipole used at the higher bands it is very important to keep the whole antenna flat and horizontal, not an inverted V, otherwise the elevation angles of the lobes are higher, killing gain at the needed lower angles.

So for multi-band operation it looks like flat dipole or the quad loop or the cut-in half variation are the ways to go.  If one was interested in 160 and good 20 meters (and higher band) performance, the thing to do is to keep the lazy-C dipole’s basic length as short as possible.  With center feed and open-wire line and large wire diameter, you can get an 80 meter length dipole to work efficiently on 160 meters.  The feedline can’t be of length that puts the voltage node at the tuner though!. (Murphy has done this to a number of stations).    Then if the antenna is oriented so that the 20 meter lobes are in the desired direction, things may be great.  This is just my theory as I have not implemented this. 

I am fortunate to be able to put up separate dipole antennas here for 160, 80, 40 and a 20 – 10 vertical. I’m a believer in single band antennas, but am getting interested in implementing a 40 – 6 meter discone to add to the farm.

[ke7trp]

I ran 130 ft of wire at 55 ft with about 15 to 20 ft hanging down stright on the ends. Fed with 450 and then later 600 line.  It worked fantastic on the higher bands and thanks to you, I now know why.  WHen I got the tower up. I put the antenna at an inverted V. The higher bands really really suffered.  Guys on talked to every day on 15 and 20 AM where way down.   Thanks Tom.


C

[Steve - K4HX]

Good info Tom. I'll have to model that arrangement to see how it works out.

As with most antennas systems, there are trade offs. Letting the ends hang down will reduce the effective height of the antenna and reduce the gain some on the lower bands. If you only have limited space, then you would probably do this anyway. If the space  is available for a full length dipole, I'd go with the full length and then hang a 20 meter dipole off the same feedpoint. Or go with the DJ4VM quad or the G6XN SDL.

Anyway, it's good to have options. 

[Tom WA3KLR]

Yep, a separate dipole for 15 or 20 is also a good way to go also as they are small antennas.  Since that antenna would be resonant I would feed it with coax and using a balun at the feedpoint is a must for the best performance.  Antenna height is critical.  It should be 1/2 wave or a hair lower as lower dipoles have a 90 degree elevation pattern.  When the dipole is over 1/2 wave high, the elevation pattern breaks up into two lobes, one low elevation and one high; might be a good thing for a sked with a close and one far station though.

For number of years I had a sked with a ham in South Dakota to my QTH in the Philadelphia suburbs on 20 using only a dipole.  For the distance it is at the limit of a single hop yielding a very low elevation angle.  This was a real killer.  The dipole was coax fed and when I added the feedpoint balun the difference was dramatic.  Decoupling the transmission line allows the pattern to be close to the theoretical expected.  You can't implement the feedpoint baluns practically for open-wire line.

[Steve - K4HX]

Indeed. The feedline radiation may create other lobes or fill-in nulls. Most often in my experience, the non-isolated feedline seems to pick up local noise, which hurts on receive. A nice high dipole with a decoupled feedline can be surprisingly quiet on receive.

[W2VW]

Indeed. The feedline radiation may create other lobes or fill-in nulls. Most often in my experience, the non-isolated feedline seems to pick up local noise, which hurts on receive. A nice high dipole with a decoupled feedline can be surprisingly quiet on receive.

One reason off center fed antenna are a bad idea for many.

[Steve - K4HX]

I guess you can use some hefty chokes to isolate the feedline, but it seems like a PIA to me. One company purposely lets about 30 feet of the feedline radiate claiming improved performance. Dunno.

[Tom WA3KLR]

Dave W2VW and W2OBR Old Buzzard Rich are 2 guys who did a lot of incremental improvements to 75 meter dipoles center-fed with open-wire line used on 160 meters to the point where the antennas performed very efficiently.

Heathkit marketed a large coil balun for open-wire line in a box way back, large.  They are very rare.

[W2VW]

Dave W2VW and W2OBR Old Buzzard Rich are 2 guys who did a lot of incremental improvements to 75 meter dipoles center-fed with open-wire line used on 160 meters to the point where the antennas performed very efficiently.

Heathkit marketed a large coil balun for open-wire line in a box way back, large.  They are very rare.

Rich asked Dave: "Why again are we using this large wire".

[Tom WA3KLR]

"For the absolute best signal possible Rich".

[W2VW]

"For the absolute best signal possible Rich".

Or how to make 20 amperes of RF current out of the coupler with only 5 going in the 50 ohm port.

One engineer told me that violates the law of conservation of energy.

[Tom WA3KLR]

Fire that engineer if he is a r.f. engineer.

[W2VW]

Fire that engineer if he is a r.f. engineer.

Too many forms to fill out. Maybe just put the guy on an IT project...

[KB2WIG]

"Fire that engineer if he is a r.f. engineer."

Why not just sequester him/her?

klc

[Steve - K4HX]

I've done some modeling runs on the bent dipole.

Below are plots on a 75 meter dipole with the horizontal section (H) with a length of 65 feet and vertical sections (V) on each end of 32.5 feet. The height of the horizontal section (Z) is 50 feet. The azimuth pattern looks good up through 20 meters. Some changes begin on 15 meters. The pattern remains bidirectional but is turned 90 degrees from the broadside patterns on the lower bands. The peak gain is down a few dB compared to the DJ4VM quad and the take-off angle is pretty high at 48 degrees. There is a lower angle lobe, so all is not lost, but its gain is down a few more dB.

The cloverleaf remains intact on 10 meters but becomes a bit sharper. But there are many higher angle lobes. The 3D plot shows the extent of the pattern break up. In the broadside direction, the DJ4VM is up to 20 dB better than the bent dipole.

A 40 meter version of the bent dipole would probably be the best choice for a clean pattern on 15 and especially 10 meters.

[Tom WA3KLR]

I ran simulations of a lazy C 40 meter 1/2 wave dipole at 29 MHz.  The antenna is 32 feet high, center fed, the top horizontal section is 23 feet across, and the ends are 23 feet long, hanging vertically, coming down to 9 feet from the ground.  

The simulation uses real ground, real 14 AWG wire.

Attached are 2 pdf plots, one for azimuth at 20 degrees elevation angle, and the other plot is an elevation plot at 45 degrees azimuth from the antenna plane.  The 10 meter lobes are broadside to the antenna.  The gain on 10 meters at 20 degrees elevation at the peak azimuth is +5.3 dBi, the sides (in-line with the ends of the dipole) are 9 dB down from the peak.  The first main notch in the elevation pattern is at 32 degrees elevation.  Some other antenna heights and designs have the first elevation notch at 20 degrees elevation which is not good.

[Tom WA3KLR]

For comparison, an azimuth plot at 20 degrees elevation pdf is attached for a 40 meter 1/2 wave dipole 66 feet long at 32 feet high also, all horizontal, operating at 29 MHz.