The "
BQH9-6c" Antenna - A  9 Element Yagi for 50.150 MHz with 12.2 dBd gain, and a 50 ohm direct feed
(thanks to W1IPL for finished antenna photos)

This antenna was designed to be an easily-constructed, very strong and durable antenna, with over 12 dB gain (compared to a dipole), and capable of being used effectively for EME contacts when aimed at the horizon.  With the use of weak signal digital modes, this antenna should be able to produce contacts with larger stations, even while elevated off the horizon.  A pair of these antennas, or an array of four such antennas, would make an excellent steerable 6m EME array.

:  The details of this design must be followed strictly to obtain the anticipated performance.  In other words, you MUST use the same spacing and mounting of the insulated elements, the same element diameters, the correct lengths shown, etc.

Many thanks to the efforts of VE7BQH for his fine contribution, diagrams and notes, which are shown below.

The "
BQH9-6C" Construction Notes

1)   BOOM material is assumed to be 2" diameter aluminum tubing, although slight variations in diameter are not considered significant.You will need to plot the element locations on paper versus the joints in the boom, to make sure no elements end up on a transition joint. It is normally a pretty easy thing to adjust the transition joint to not coincide with an element by either moving where the antenna starts on the boom or changing where the joint happens.

2)   TRUSSING of the boom at right angles to the elements can be done as required, with turnbuckles and steel cable to keep boom sag minimal.  Any lines added to reduce side-to-side bending in the plane of the elements should be done with non-conductive material such as Phyllistran, and should be held out to the sides with non-conductive material such as fiberglass tubing.  Furthermore, any such lines should be positioned so they do not bear down across the elements when encumbered by ice or snow.
3)   ELEMENT TO BOOM PLATESare made from 1/4" thick Lexan or HDPE insulated material.  There is no boom correction for the type of insulated element mounting used here. That is, the boom correction is under a 1/64" and is beyond any reasonable cutting tolerances. This assumes that you are using small metal U bolts to fasten the elements to the lexan plates; alternatively, you could connect the elements to the plates with a pair of small diameter stainless steel screws and nylon locking nuts (such as 10-32) on each side of the centerline of the plate.  In eithe case, this should introduce only a minimal disturbance (provided they are only as long as required to hold the materials together).  

Each 1/4" thick lexan (or HDPE) mounting plate is mounted to the boom with a stainless steel "muffler" style clamp so the bottom of each element is 5/8" above the boom.  This dimension IS critical.  If your particular mounting U bolt has a saddle that introduces a different spacing, use a thicker piece of insulated plate so that the elements are still separated from the boom by 5/8".

4)   DRIVEN ELEMENT is constructed using 1" diameter aluminum tubing, with some short sections of 7/8" diameter tubing sticking out on each end, to allow for some final VSWR adjustment.  In addition, D1 may have to moved slightly to adjust VSWR. It is difficult to capture every capacitance in the program consequently the reason for some real world touch up.

For mechanical stability, it is suggested that a 7/8" diameter piece of  teflon be inserted in the ends of the two driven element halves, to provide good alignment and rigidity where they are attached to the mounting plate. 

5)   CONNECTION OF COAXIAL CABLE to the split dipole must be kept as short as possible. The inside ends of the split dipole should be no more than 1/2" apart to allow the coax connection to be kept short.  VE7BQH recommends you use crimp on spade lugs and also solder them with silver bearing non lead solder (available in any large building supply etc).  If the antenna is going to be left installed for an extended period of time, you can then seal all the connections with non-taminating (non acetic acid curing) silicone.

6)   FERRITE CORES with the proper mix, such as the 1/2" RFI CORE (available from THE RADIO WORKS at, are used to choke the RF currents in the outside of the coax.  The number of ferrites you will need is dependent on the power you will use. 1 ferrite will handle up to 600 watts; 2 ferrites will handle 1200 watts, etc.  The test is simple, the ferrite should only be warm to the touch under sustained power. The ferrites must be positioned as close to the feed point as possible. As the ferrites are snap on, they need to be well wrapped to ensure the two faces of the core are kept dry and firmly together. Large size shrink tubing works well for this.

7)   PARASITIC ELEMENTS are constructed using center sections of 5/8" diameter tubing, with the outer portions of each element being 1/2" diameter tubing.

8)   ELEMENT LENGTHS shown in the table below are the HALF length. Therefore, to obtain the overall lengths for the directors and reflector, take the .625 length and the .5" length,  add them together and then multiply by TWO.

Example: Reflector = 12" + 46.4445" X 2 = 116.889". The driven element is done the same way but uses the 1" and .875 diameters but will be kept in two halves.

There is no allowance shown for the extra material that will have to be used when the smaller tubing is inserted into the larger tubing.  It is anticipated that each joint in the elements would probably require several additional inches, and that the  joints would be secured with one or more short stainless steel screws with nylon locking nuts.

BQH 9 Element 6 Meter
50.000 50.150 50.300 MHz

REFLECTOR   0.6250    1.0000    0.8750    0.5000 DIAMETER OF ALUMINUM TUBING IN INCHES

  0.0000   12.0000    0.0000    0.0000   46.4445 REFLECTOR
 40.0000    0.0000   50.0000    5.6407    0.0000 DRIVEN ELEMENT - DIRECT 50 OHM FEED
 62.0000   12.0000    0.0000    0.0000   42.9632 DIR 1
115.9397   12.0000    0.0000    0.0000   41.6255 DIR 2
201.3638   12.0000    0.0000    0.0000   40.5285 DIR 3
285.6098   12.0000    0.0000    0.0000   39.8537 DIR 4
385.4554   12.0000    0.0000    0.0000   39.8850 DIR 5
467.9373   12.0000    0.0000    0.0000   40.3807 DIR 6
536.0000   12.0000    0.0000    0.0000   39.0541 DIR 7
9)   STACKING BEAMS is recommended as follows:  E plane= 33.4', H plane = 30.63'  You could go down to 90% of these values with minimal gain loss.

Last updated May 22, 2010