New yagi for 144 Mhz
After low impedance antenna bad experience, we decided to try some new design based on 50 ohm direct feed dipole concept. Low impedence antenna such as dk7zb, k1fo, hi-gain etc don't seem to work properly as well as shown in simulation as low impendence makes tuning not easy at all. All the low inpedance aerials we built up were born following the desiner suggested measures but, after tuning procedure, every antenna had different measures. We realized it is due to very critical design to achive lower dipole impedence close to 25ohm. Comparing simulaton of antenna with same boom lenght, at first sight, shows an higher gain from 25ohm antenna respect to 50ohm one, but the software can't consider some other circumstances that makes "extra-gain" disappear.
Something to consider before assembling low impedence stuff:
- It's difficult achive high return loss value
- the coaxial transformer is cumbersome and has losses
- the dipole is absolutely not simmetrized, and lot of RF runs across the dipole when you drive it with QRO. That's the reason why coaxial cable explode!
- There are to many items around the dipole such as the coaxial transformer and the box where cable is connected
- Hairpin, t-match, Beta-match, gamma-match, folded dipole with 4:1 transformer, parallel cable transformer are unusefull stuff for 50 ohm aerials
For all reasons above we decided to convert low impedance yagi to 50 ohm, and built some new 50 ohm antenna based on g4cqm design with some changes after remodeling his design with modeling software.
These are test results:
- the antenna is capable to be driven with several power without cable arching across boom. You can achive better result just bending cable and make a little curl near the feeding point or, if you can, let the cable runs out of the antenna from the back side thru reflector.
- Incredible return loss < -30 dB !!! for example, thats mean: power in 1000 W ==> power back 1W !!! it's 0,1 % S.W.R. (Analog Devices 8307)
- Driven element is a split dipole, very easy to build.
-The balun is easy to bilt too and works fine. It is a coaxial balun made with a wire straight thru the dipole internal radius, fixed end to end. No other items is required to simmetrize the antenna such as Tonna and some other Brand were used to do throwing the cable into a quarter wave pipe grounded at its end ( sometimes it is not grounded too! )
- 50 ohm design is much more silent
Some dipole pictures:
a view into the box
the box is closed
the dipole is ready to be mounted on antenna
We've been building aerials since 5 yaers and new designs are cooming up. After some test we can say that the evolution is going fast thanks to new project based on 50 ohm impedence at the driver . (nomore burned cables, and we never work with sat-cable).
More test will follow to get top results on the single antenna and than make a stack.
At first tuning we get a stable -25 Db return loss and we look forward the get lower values working on the driver diameter.
After some trouble due to ground influence in the tuning procedure, we decided to make antenna measurements beaming the cold sky with 45 degrees elevation and the reflector element at 2,5 m above the ground.
Now we're surprised to get a -34dB of return loss at the power meter.
Next step is to build an OWA yagi just with one more element in the first driver cell among Ref and D3. We realized that main lobe is too narrow and side lobes are not optmized as well stacking distance between 2 aerials is still above 3.50 meters. We hope the further parastatic element to equalize the RF magnitude in the first cell elements and makes sidelobe decreasing for lower H and V stacking distance.
You can see those new aerials we built up this winter season and still under test, but at first on air test seems outperform old group. By the way, we will produce futher information after spring contest as one of the most important thing for antenna builder is don't stop after simulation but built your own prototipe and make several test under different propagation and weather condition. Nevertheless, it's very noticeble that after tuning procedures all the 3 tested antenna come out with same dimensions and spacing (according with the design sheet) into the 4 elements of the driver cell group and the rest of parastatics of course. We just realized during December and Jenuary contest that new antenna have larger and clean main lobe and not a narrow lobe with deep nulling post as the old one. The added of the current control elements make the sidelobe profile more confortable and makes an easy beaming of week signals in the contest activity.
Old aerials come from low impedence design but with the driver cell redesigned to achive 50 ohm at the driver and may have some 0,2 db more gain at simulation, but field operations allow us to say that get a 25 ohm driver means only makes feeding much difficult witn no kind of real advantage in performance. In our humble opinion, these are only two different design concept but nothing to be defined as a new "theory" about low impedance as you will get quite same results and patterns with a 50 ohm driver cell and the same spacing and measures of the following parastatic elements. The only thing that you can be sure is that you're getting hard life to manage trasformation and current neutralisation due to the diffent inpedence values of radiator and feeding line.
In the pictures below you can see the 3 lambda design under test, but we have 4 lambda element just ready to be assembled on the boom.This time we decided to make full isolated element instead of fixed in the middle point ( tnx Giuliano I5HQG for isolators ).
Distance between elements on boom is enough to unconsider any elements correction factor. We suggest you to visit a very good yu7ef homepage to read something about it.
We get good results with old radiator element but we are going to replace them with some coaxial dipole made with 1/2 " cellflex which garantees full simmetrize and , we hope, easy tuning on the desired frequency by simply trimming the inner wire tail without adjust element spacing in the driver cell.
some pictures:
We're going to win something to drink and we already know who's so kind to pay for
This is the new vertical stack for season 2008! Now we have changed cables with a longer set of Ecoflex-10. Power back is now about less than 2%.
more to follow....