Open-Folded W8JK Antenna (2024)

Experimentation

Open-Folded W8JK 10 to 30m (and 6m) : the turning radius is only 2.7 up to 3 meters only. I build this antenna during the 80s and the performance are really great . During the 80s my W8JK was built with regular size compare thisthe new one which is folded due to to environment constraints in my area (i have no choice , since i am staying in a very dense urban area , rising a full sizeYagi is impossible).

The finalizedantenna details :

-2.70m boom (Driven elements are 2.63m)

-The longest driven element (monopole) is 2.5m long so about 5m long in total

-Turning radius is only 2.7m

During the past few days i manage to collect all the hardware i need and perform some tests with the antenna at only 3 meters high.

At this stage the results are above my expectations and i can even access to the 30m band (No gain) and 50Mhz with a significant gain.

According NEC and MMANA simulation the results are really close and are matching which is the most important.

According the test i can perform prior to put the antenna on the top of a tower , all signals (Receive) are better compare to my Vertical and End-Fed antennae. It look promising for when the antenna will be at 12 meters high.

Antenna details :

All details are provided according the measurement i perform with a Network Analyzer (From my office) KeysightHp8714, A navo VNA , A SDR receiver , SK100 and i was using a RPI with RPITX as a generator located in the Garden and driven with VNC by remote

-Antenna Gain : from 2.5up to 3.5dBd. On 10Mhz the gain is almost at the Isotropic dipole performance. On 50Mhz the gain is about 4 to 5dbd

-Front/back attenuation : None

-Side attenuation : From 15 up to 20db except for the 50mHz is up to almost 20db

-Boom lenght and size : 50x50x3mm – 270cm Long (Aluminum)

-Partslenght : 2.70m for the Boom and 5m (2.5m + 2.50m)for the driven element.

-Driven wire dipole length : 10.40m up to 10.65m

-Phase line driver : 450 Ohms twinlead with 5 cm spacers

-Turning radius : 2.70m

-Max power : 500W (to be confirmed because it is related to the D10 wire i am using)

It is pretty much the size of the MA5B beam antenna from Cushcraft

Folded W8JK Pro and Cons :

Pro :

-Small size

-Simplicity : this antenna is easy to build and the assembly take about 30 to 45 minutes (I will add some pictures)

-Performanceregarding the size of the antenna compare to half wave dipole.

-Transmission line loss using a 450 Ohm twinlead compare to coaxial cable is really better

-Beaming : With this antenna you just need 180 degree

-Resistivity and radiation impedance : It was a great surprise . i was expecting the impedance will be a little bit higher compare the genuine W8JK but according with the VNA measurement the impedance is higher than expected. The good thing is less loss and easy tuning even with a commercial tuner likely the MFJ969

-Compare to a Yagi , you have almost nothing to setup as far you wire is about half wavelenght of the lowest frequency and you antenna tuner is efficient (Mac Coy).

Please refer to this webpage for an efficient tuner for this antenna:

-You can cover from 30m up to 6m without any issue and the antenna is providing gain on all bands except 30m.

-You will have the same performance compare to a 2 element Yagi with Coil or trap.

-Weight : Using Fishing rods and aluminum for the boom will give you about 7 to 9Kg antenna only .Mine is 8kg only.

Cons :

-The transmission line using twinlead is a bit tedious for the routing down to the Shack

-You need a dedicated Tuner andusing a MAC-COY is the recommended one. The Impedance is low and without a good tuner the loss will be significantduring transmit.

-Less performance compare to a full size Yagi

-Bi directional . The F/B will not assist you to cover spurious from the rear of the antenna.

This antenna is a good compromise between a dipole and a full size Yagi.

Hereafter you will find the simulation results from NEC4 and MMANA

The W8JK antenna is2 (Two) centre-fed double-dipole fed by a pair of anti-phase signals. With a Center-fed we can simply use a balun in order to feed the dipolebut with the W8JK it is not possible or not easy to achieve it. In order to do it , a dedicated out of phase and current divider should be built to ensure a proper current and power distribution in both dipoles.

The two dipole elements are fed by a pair of anti-phase signals designated below with – and + (V+ and V-)

Each dipole element can be divided into two parts or monopoles . The W8JKcan be associated as four monopoles. Because of the anti-phase characteristics of the excitation signals and , the currents on the arms of the 2 dipoles (4 monopoles) are anti-phase at the feed points (Noted Feeding in my Schematic below).

2 Monopoles can be associated as a dipole and regarding the current , they will be identical (They should be identical) in the 2 equivalent dipoles. The 2 dipoles shall resonate at the same frequency with identical power and current distribution (This is the specificity of the W8JK antenna , if not the antenna will not perform well and performance can drop significantly). This , should be taken into consideration during the build and you should use the same material for both dipole. Before raising the antenna in the final position on the top of the mast , you should perform a simple test with a grip dip or more simply with a hertz loop and a small buld or LED.

How to test it :

-Idle you transmitter at 50 up to 100W

-With your Hertz Loop or grip dip , check all the current all along the antenna and the phasing line. When the bulb , the LED or the Grip dip is a the maximum , it means the current is at the maximum. Both side of the antenna the result SHALL BE IDENTICAL . If not , it means the power and current division from the Feeding network (Anti-Phase line) is not perfect .

The phasing line or anti-phase feeding network consists of a pair of parallel striplines strickly identical in order to ensure perfect distribution of the current and power are opposite at the feed points. Bear in mind that the feeding network is also a power divider and obviously the power distribution shall be equivalent in all monopoles.The phasing line should also have exactly the same length and build with the same material. needless to say that the soldering and all connection shall be performed with a special attention. If not , the phasing of the antenna will be disrupted with significant effect on the impedance, and mostly on the radiation pattern (Directivity will be affected).

The monopoles or dipoles shall be strictly identical with the same length . A discrepancy for the length will provide some ”current lagging” between a dipole to the other. This current lagging will ”disturb” the antenna radiation pattern and Gain. Bear in mind the dipoles have also their own capacitance and obviously this capacitance should be identical.

Building this antenna will teach you a lot regarding antenna arrays. I really recommend new amateur radio this antenna build . Definitely it will assist you for a good understanding for how it works .

FOLDED W8JK 10 to 30 m*14.15***Wires***130.1195,0.02,0.0,0.1195,2.525,0.0,7.500e-04,0-2.61738,0.01551,0.0,-2.61738,2.52551,0.0,7.500e-04,-10.1195,-0.02,0.0,0.1195,-2.525,0.0,7.500e-04,-10.1195,-0.02,0.0,-1.24894,0.0,-0.01864,7.500e-04,-1-2.61738,-2.52701,0.0,-2.61738,-0.01701,0.0,7.500e-04,-1-2.61738,0.01551,0.0,-1.24894,0.0,-0.01864,7.500e-04,-1-1.24894,0.0,0.01551,0.1195,0.02,0.0,7.500e-04,0-2.61738,-0.01701,0.0,-1.24894,0.0,0.01551,7.500e-04,0-1.24894,0.0,0.01551,-1.24894,0.0,-0.01864,7.500e-04,-10.1195,-2.525,0.0,0.1195,-0.1,1.1,7.500e-04,-1-2.61738,-2.52701,0.0,-2.61738,-0.08136,1.10788,7.500e-04,-10.1195,2.525,0.0,0.1195,0.1,1.1,7.500e-04,-1-2.61738,2.52551,0.0,-2.61738,0.08136,1.10001,7.500e-04,-1***Source***1,0w9c,0.0,1.0***Load***0,1***Segmentation***400,200,1.6,2***G/H/M/R/AzEl/X***1,12.0,1,450.0,120,60,0.0###Comment###F5NPV

W8JK antenna performance

The antenna is now finalized and i manage to perform some tests FT8, CW and SSB low power about 70w Max during the different tests i performed. I was stunned by the performance considering the size (Radius) of the antenna

Compare to my the other HF antennas , the Signal level is always better by at least 1 point from the S-Meter, the directivity and mostly the antenna side attenuation is great.

The big surprise was regarding the 50Mhz , On this frequency the antenna is working very well.

The disappointment is regarding the 10m band , but to be honest it is something i was expecting since the boom is not tailored for this band.

Bear in mind this antenna according the currentdesign is tailored for the 20m band but like the center-fed dipole with the suitable antenna tuner, the W8JK can also resonate on other bands in particular with the multipliers : 2F, 4F and so on. Obviously, the radiation patterns popup some additional one which are not suitable but regarding the antenna design it cannot be avoided . The antenna will operate at higher frequencies, but the pattern breaks down into multiple lobes as the electrical length of the elements increases at higher frequencies.. To solve this issue it should be possible to reduce or enlargethe boom lenght which is not easy to do. A multiband beam antenna is always a compromise.

The final result is providing about 3 up to 3.5 dbd against the Dipole reference or isotropic antennawhich is not so bad considering the size and the funky design of the antenna. A 50Mhz the antenna gain is almost 5dBd

Transmission Line , Feedpoint antenna tuner

The antenna is fed with 30 meters of 450ohms twinlead and i have 2 antenna tuneri can use.

-MFJ969 : This tuner surprisingly works very well with this antenna and i can tune it from 30Mhz downto 10Mhz

-Mac Coy autotransfo : Since i am using the MFJ for other antenna or experimentation i prefer to dedicate the Mac Coy antenna tuner to this antenna . This transformer is really simple to build and ditto no issue was noticed and according the latest test perform today with an Italian ham radio , the tuner is performing well and did not noticed any losses .

The W8JK antenna is a flat top beam built with 2 half wave dipoles (4 monopoles) fed by a out of phase line and the 2 dipoles are located about a 1/8wave from each other. The radiation resistance is about 3 to 5 ohms, the gain about 4 to 6 dbi mean about 2.5 up to 3.9 dbd.

The directivity and front/back ration is null , since this antenna by design is bi-directional and the impedance at the feed point of the dipole isstrickly identical (sould be identical).

The main constraints of the W8JK antenna is regarding the radiation resistance which is really low. this specificity is involving high current circulation all along the dipoles and the antenna tuner. All connection and soldering shall be performed carefully and with a lot of attention. Therefore the particularity of the low impedance with High current will assist this antenna to produce a gain with each dipoles acting as a reflector. This specificity will provide bi-directional antenna : The W8JK

The simulation had provided some results but according my measurements , with the VNA with a 1/1 balun , this is the final figures for an antenna at 2 meters high (R +/- jX)

The impedance at resonanceof a half-wave folded dipole is up to four times that of a regular half-wave dipole., practically about 240 -280 Ω. This is not so far to the characteristic impedance of 450 Ω twin-lead cable. According the design i am testing means an open-folded dipole , this currently the figure i measure (Obviously it is not a perfect open space and soil and the high was only 2 meters. According the following results , feeding this antenna with a 450 Twinlead and a 4:1 balun should be tested

Frequency –>R +/- jX

10.1Mhz –>3 + 12

14.1Mhz –> 15+ 80

18.1Mhz –> 15 + 280

21.1Mhz –> 66 – 450

24.1Mhz –> 100 -256

28.2Mhz –> 1– 400

50.1Mhz –>57 + 15

For a better understanding i use “Folded” but i guess the correct wording should be “bent”or “Open Folded”

The Very low R at 28Mhz is not helping at all . For all other frequencies the results are confirming what i notice with my 2 antenna tuner : The antenna is very easy to tune even with a commercial tuner like the MFJ969. With this kind of configuration , i mean a folded W8JK , compare to a standard non-folded one the R is really higher and this is assisting for antenna impedance which are higher. A higher Impedance is providing less losses and a better efficiency.

The main goal of this project was to ensure a decentimpedance of short monopoles with optimum current distributions. The efficiency was really my concern regarding this project.

The efficiency rapidly decreases with decreasing radiation resistance. For constant radiated power, the current must be greater for smaller values of radiation resistance. Greater current means greater loss and consequent reduction in efficiency; therefore, the power loss in a center-loaded monopole is more than that of a top-loaded equivalent antenna and the loss in the bottom-loaded case is more than that of the center-loaded case. A combination of both top and bottom loading, however, gives a radiation impedance which in some cases reduces the loss to an exceptionally low value compared to the bentW8JK.

For low radiation resistances and practically nothing can be done to improve their efficiencies to a reasonable value, except possibly by using a multi-wire system to raise the impedance butthe folded solution was surprisingly a good choice. The efficiencies of longer quarter-wave monopoles may be increased considerably by a proper folding.

During this experimentation , i perform a lot of test with different wire length , folded angle and driven element spacing. The best result waswith a folded anglenot less than 65 degree. With less angle the radiation resistance dropped significantly and the current was really too high.

According this configuration , the wire length was really the key and with with less than a quarter-wave for each monopole , the Radiation resistance and impedance dropped also significantly. The extend of this experimentation and antenna configuration , the monopole shall be at least at aquarter wave length (The lowest frequency).

Since i cannot reduce the lenght of the dipole without a significant drop regarding the performance , the main idea was to fold them in order to reduce the antenna footprint . This experimentation is providing decent results and performances.

Wire vs tube

Generally speaking , for Beam antennae we are using tube but considering the specificity of the W8JK antenna and the very low resistance at the feedpoint i need to increase a bit the resistance in order to provide the best efficiency and minimize losses . For my experimentation choosing to fold the wires has increase a bit the resistance. Using low section wire for a beam antenna is minimizing the bandwidth but also increasing the resistance which is not suitable for a dipole or a beam (Yagi) antenna. This constraint is in fact assisting to rise a bit the resistance for the W8JK antenna and becomes finally a benefit.

The wire i am using is 1mm only section and working perfectly upon 400 up to 500w

Cons : The small wire section will limit the amount of power in the antenna . I did not make some test above 500w but with 300 up to 500w the antenna is able to sustain such power.

W8JK variant few experimentation

One thing very important , during my several experimentation i compare a standard full size W8JK with the Open-Folded one:

-The full size non folded W8JK has a better gain (about roughly +1dbd) but less side attenuation. In addition the radiation resistance drop and the tuning was a bit tedious to achieve with the commercial antenna tuner (No pb with the Mac Coy tuner). According with WEBSdr and measurement with the spectral analyzer , the Full size antenna is providing a better radiated energy. Currents in the dipoles are more uniform and consistent.

-The Open-Folded W8JK has less gain but a better side attenuation . The radiation resistance was a bit higher and the tuning was easier. The side attenuation is impressive and initially i was thinking the antenna was not working at all until i beam it towards the signal and the emitter location. Currents in the dipoles are less consistent especially on 10m band.

-I have tried also some other length for the dipoles , but the best efficiency was achieved with at least 10.40m long. Eg: with 8.50 meters long the performance was really poor. The best performance was achieve from 10.65m down to 10.40m

-I have tried also some different angles and center pole length , but definitively the minimum angle is about 65 degree. With less the current all along the antenna was all upside down with significant inconsistency involving poor performances and gain.

Remaining test i need to perform:

Since the feeding point impedance at the resonance is higher than the non-folded W8JK i need to test the feeding point with a current or voltage 4:1 Transformer.

Ultimate Design

Testing and experimentation is really the key and 2 weeks ago i manage to perform additional tests with different antenna size. The following results are the best performance i can achieve with simply modifying the lenght of the horizontal monopole size to 3 Meters (Instead of 2.5 meters). This modification is increasing the turning radius a bit to 3 meters (Previously it was 2.70 meters )which is still acceptable for limited space and constraints.

The boom lenght remains identical with 2.63m

FOLDED W8JK 10 to 30 m*14.15***Wires***130.232,0.02,0.0,0.232,3.005,0.0,7.500e-04,0-2.72988,0.01551,0.0,-2.72988,3.00551,0.0,7.500e-04,-10.232,-0.02,0.0,0.232,-3.005,0.0,7.500e-04,-10.232,-0.02,0.0,-1.24894,0.0,-0.02297,7.500e-04,-1-2.72988,-3.00701,0.0,-2.72988,-0.01701,0.0,7.500e-04,-1-2.72988,0.01551,0.0,-1.24894,0.0,-0.02297,7.500e-04,-1-1.24894,0.0,0.01551,0.232,0.02,0.0,7.500e-04,0-2.72988,-0.01701,0.0,-1.24894,0.0,0.01551,7.500e-04,0-1.24894,0.0,0.01551,-1.24894,0.0,-0.02297,7.500e-04,-10.232,-3.005,0.0,0.232,-0.96121,0.77549,7.500e-04,-1-2.72988,-3.00701,0.0,-2.72988,-0.95876,0.95759,7.500e-04,-10.232,3.005,0.0,0.232,0.96121,0.77549,7.500e-04,-1-2.72988,3.00551,0.0,-2.72988,0.9555,0.77225,7.500e-04,-1***Source***1,0w9c,0.0,1.0***Load***0,1***Segmentation***400,200,1.6,2***G/H/M/R/AzEl/X***1,12.0,1,450.0,120,60,0.0###Comment###F5NPV

Results :

• The antenna F/B gain is a bit better with roughly 0.5 db increase compare to the 2.5m horizontal monopoles size.
• The radiation impedance at 28 Mhz is about 8 ohms which is far better compare to 2.5m horizontal monopoles size.
• The side attenuation remains very good and still better compare to the full-size W8JK (Non-open folded).
• The size and footprint of the antenna remains acceptable for limited space constraints.
• The noise level remains really low compare to my 2 other HF antennae.
• The radiation impedance are identical compare to the previous design (2.5m horizontal monopoles size)
• This alternative design is providing in all respects better performances versus a dipole or a 1/4 wave vertical antenna. Roughly it is at least 3.5db and even more in some cases.
• The tuning still very easy to achieve and please refer to 1KW universal antenna tuner page for more details regarding the transformer i use (4:1 Guanella current transformer)

Mac Coy antenna tuner

Antenna BOM

By the way it was a great experimentation project during the 2 months lockdown due to the COVID 19 pandemia.

Since this antenna is really simple the total cost was about 125euros only :

-Boom aluminum : 50x50x3mm and 2.65 Long : 45Euros (weight 3.5kg)

-4 Fiber glass fishing rod , 5 meters long : 32Euros (Since i am not using the full size weight is 1kg only for the total weight)

-Wire and twinlead for the phasing line: 10 euros

-Bolt , screw about 20 euros

-Misc : 20 euros

Final weight is only 8kg (the MA5B from Cushcraft is about 12kg)

The good thing regarding my design is you can convert the folded antenna to a full size W8JK antenna with ease. The conversion to a Moxon, ZL SPECIAL or a Maria-Maluca will be easy also. This design will provide you some flexibility.

Manage to contact Thailand , only 45watts with -7db signal

FOLDED W8JK build step by step (Final build)

The build is straight forward , therefore the following shall be considered:

-During the experimentation , i notice that any piece of metal , collar or any conductive parts may affect the antenna performance .

The main result is i need to ban all metal parts during the build and this why for example i use wood rod to carry the fishing rod ; Initially i use also aluminium rod instead of wood and definitely it was affecting the antenna performance and the radiation pattern.

My initial intention is to use wood for the boom , but unfortunately i cannot find the suitable one or the one i found was really too heavy.

-The phase line shall be islolated and apart from the boom with at least 5 cm .i am using nylon spacer,nylon bolts and nylon screew.

-The fishing rod shall be perfectly isolated from the boom ; I saw some document with rods part of the boom . The simulation and the different test i perform just show something : it is not working at all and the dipoles SHALL be isolated from the boom

-I am using is D10 telephone line wire; i am using this wire for my vertical antenna and should be able to handle up to 500 to 600W (this wire with my vertical antenna can handle about 700 to 800w butSince the current can be very high for the W8JK antenna 500W i guess is something i can expect. During my test with 500W no heat or sided effect was detected).

-The center pole are parts of the fishing rod since i am not using the entire length of the rod.

2023 update

Since i am now located in Brunei Darussalam for few years, this is an updated release of the antenna .

FOLDED W8JK 20M TO 6M*14.15***Wires***130.232,0.02,0.0,0.232,3.52,0.0,7.500e-04,0-2.72988,0.01551,0.0,-2.72988,3.51551,0.0,7.500e-04,-10.232,-0.02,0.0,0.232,-3.52,0.0,7.500e-04,-10.232,-0.02,0.0,-1.24894,0.0,-0.02297,7.500e-04,-1-2.72988,-3.51701,0.0,-2.72988,-0.01701,0.0,7.500e-04,-1-2.72988,0.01551,0.0,-1.24894,0.0,-0.02297,7.500e-04,-1-1.24894,0.0,0.01551,0.232,0.02,0.0,7.500e-04,0-2.72988,-0.01701,0.0,-1.24894,0.0,0.01551,7.500e-04,0-1.24894,0.0,0.01551,-1.24894,0.0,-0.02297,7.500e-04,-10.232,-3.52,0.0,0.232,-1.77823,0.52786,7.500e-04,-1-2.72988,-3.51701,0.0,-2.72988,-1.77524,0.52786,7.500e-04,-10.232,3.52,0.0,0.232,1.77823,0.52786,7.500e-04,-1-2.72988,3.51551,0.0,-2.72988,1.7731,0.52573,7.500e-04,-1***Source***1,0w9c,0.0,1.0***Load***0,1***Segmentation***400,200,1.6,2***G/H/M/R/AzEl/X***2,12.0,1,450.0,120,60,0.0###Comment###F5NPV - V85NPV

Conclusion & Summary

This antenna is definitively not the best performer , especially considering the design . Therefore , according the size , this antenna is really interesting if you have some environment constraints and limited space available.

If you have some space available , you should consider the W8JK/MOXON described in this website. The W8JK/MOXON will provide a better gain and no constraints regarding the 10m band

The Antenna is providing a coverage with fairly decent performance from 10Mhz up to 50Mhz including all WARC bands in between (0dbd on 10Mhz and roughly from 2.5 dbd up to 3.5dbd , on 50Mhz the gain is very close from 5db which is a surprise considering the boom length is not fitting the 50Mhz Wavelength. For 50Mhz the radiation impedance is almost 50 ohm.

Against a dipole or a 1/4 wave vertical antenna , the W8JK is performing better in most of the case and really assist me during pill-up with some rare DX stations.

The side attenuationwith this folded configuration is excellent (It was a great surprise).

Building this antenna was a challenge and a great experience. This kind of project will bring you back to radio-electricity basics and the essence of the Ham Radio hobby.

As mentioned in this study , this W8JK design is providing the opportunity for alternative design and elements lenght in order to cope with your requirements and constraints. Modification are really easy thanks the specificity of the mutual coupling arrays with W8JK design, this is not implying tedious setup to perform. At least the minimum lenght of the dipoles should be in accordance the frequency you are using . For this study , the lowest frequency is 14Mhz and this imply a minimum lenght of 10.40 m (1/2 wavelenght) for the dipoles in order to keep the best radiation pattern and performances. You do not need to cut or add some wire.

The W8JK is an Easy Peasy antenna for sure which providing decent performance !