Building A Home-brew 'spiderbeam' Antenna

Have you heard or worked someone who was using a spiderbeam antenna?  Have you ever thought about building one?  I home-brewed the 5-band version covering 20m to 10m inclusive.

This page introduces the spiderbeam antenna and the rationale behind home-brewing such an antenna.   The reasons for adjusting the antenna design are discussed, followed by details of the resulting changes to the structural design and electrical design.   Assembly of the antenna, rotator cage fabrication, and the means of erecting the telescoping mast are described.  The page concludes with a note concerning the antenna performance, and possible future optimisation.

Back to Miscellaneous Technical Page

The spiderbeam antenna

The original spiderbeam antenna designed by DF4SA is a full size, lightweight, tri-band yagi for 20-15-10m.  It is optimised for portable use, being constructed from fibreglass and wire.  The design has transmogrified into a family of antennas including a 5-band version (20-17-15-12-10m), a WARC version (30-17-12m), and others.

Spiderbeam GmbH, the company which sells the spiderbeam* antenna and related products, has kindly provided a very detailed construction guide in the public domain for free - how many antenna companies do that!  This guide serves as both the assembly instructions for the antenna kit bought from Spiderbeam, and the basis for "rolling your own" spiderbeam antenna.  Visit the Spiderbeam website to obtain this documentation.

* In this document I will use 'spiderbeam' to refer to the antenna design, and 'Spiderbeam' to refer to Spiderbeam GmbH.
Note: I have no affiliation with any of the companies or businesses mentioned on this web page ... I am merely a satisfied customer. :-)
Back to top

Why home-brew a spiderbeam antenna?

Since November 2011, I had been using a Hy-Gain TH3JR antenna with quite a degree of success.  This is a popular, 3-element, trapped yagi operating on 20-15-10m, and is well suited for amateur radio operators with limited space (eg a suburban backyard) who want a beam antenna but cannot install full size mono-band or multi-band yagis.  I had always wanted one after first seeing them in the early 1980s and subsequently using them during field days, when visiting other amateur radio operators' "shacks", etc.  (Possibly the higher solar activity and consequent better propagation conditions for that solar cycle biased my judgement?)  This particular TH3JR has served me well (after I bought it for the equivalent price to its scrap value, and refurbished it); but after not working D4 (Cape Verde) in yet another contest, in October 2013 I decided that another antenna was in order.

Ironic fact: shortly after embarking on building the spiderbeam antenna, I worked D4 on 20m using the TH3JR fed with 100W during the CQ WW CW Contest in November 2013. However; since then I worked D4 using 400W PEP on the 10m short path and 15m long path using the spiderbeam during the CQ WPX SSB Contest in March 2014!  :-)

The following considerations influenced my choice towards the spiderbeam:

... (last, but not least) ...

As I wanted to make the antenna a permanent feature of the antenna farm (that is, as permanent as any other antenna); considerations such as portability, ease of assembly and disassembly for portable operation, etc, were not applicable in my case.  Our weather in the Southern Tablelands can be somewhat antenna-challenging, so I opted for building the heavy duty version of the spiderbeam antenna.


Nighttime shot of spiderbeam
              antenna
A night-time view of my home-brewed spiderbeam antenna.

Back to top

Adjusting the antenna design

If the antenna design is already proven, then why fiddle with it?

It's a good question; and the answer depends on whether or not the builder is using exactly the same components (especially the wire) as those used in the original design.  If you're using exactly the same wire, then leave the design alone - build it per the construction manual.  But, if you're using different diameter wire, or bare wire, or different insulation (including thickness and material), or different wire material (you might use aluminium wire instead of copper), or any combination thereof, then you will probably need to adjust the wire dimensions.  This is mentioned specifically in the spiderbeam construction guide.

Some amateur radio operators say that these parameters don't matter.  Strictly speaking, this turns out not to be the case.  A wire carrying radio frequency currents - even a single conductor up in the air, such as one "leg" of a horizontally-polarised half-wave dipole - has a velocity factor**.  It's common knowledge that a such an antenna (for a given frequency) constructed from thick wire (or tube) needs to be physically shorter than one constructed from thin wire: the two antennas have the same electrical length, but different physical lengths ... that's a dead give-away that velocity factor is involved somehow.  The same principle applies for differences in insulation dimensions and/or material, and conductor material.  Even the cross-sectional shape of the conductor and/or insulation will have some effect - it's a matter of quantifying these effects and determining if they need to be accounted for in the design.  Sometimes they do, sometimes they don't - so why bother worrying about it in the case of the spiderbeam?  The construction guide stresses the need for accuracy (in terms of millimetres) when measuring and cutting the wire lengths for the elements - this suggests the designers found that small variations in wire lengths from those specified will adversely impact the performance of the antenna.  From this, one can conclude that changes arising from using a different wire with a different velocity factor will result in changes in the wire physical lengths needed to achieve the necessary electrical lengths; which in turn, if not accounted for, will have similarly adverse effects as not cutting the correct wire lengths in the original design.

** Velocity factor: the ratio of the velocity of an electromagnetic wave propagating along a conductor (or in a medium) to that of the wave propagating in free space.  This parameter is commonly used with transmission lines, especially coaxial cable.  For example, if a particular coaxial cable exhibits a velocity factor of 67% at a given frequency, then the physical length of a full wavelength inside the coaxial cable is 67% of that in free space - if the frequency of interest was, say, 21.2 MHz; then a full wavelength in free space is approximately 14.2m; whereas in the coaxial cable it is only 67% of this, or 9.5m.  This is because the wave at 21.2 MHz in free space propagates at a velocity of approximately 300 000 km/s, whereas in this coax cable it travels at a comparatively glacier-like 201 000 km/s.  The same principle applies for a wave travelling along an antenna conductor, only it tends to travel faster than it would in a coaxial cable featuring a dielectric other than air.

When building a horizontally-polarised half-wave dipole, many amateur radio operators calculate the half-wavelength in free space, then apply the magical 95% correction factor - the origins of which are lost in the mists of amateur radio history - thus allowing for "end effect", erect the dipole, and trim the length of the dipole "legs" (or "arms") to achieve minimum SWR at the frequency of interest.  In this process, they never see the difference due to velocity factor (which they've unwittingly already allowed for, but called it "end effect" instead), unless they're using wire diameters and/or insulation material wildly different from that used for a similar antenna they've seen elsewhere.  In practical terms for such antennas, velocity factor is not a big impactor in terms of making it work - you build it, put it up, and trim it for minimum SWR.  But; when building a yagi antenna from wire (or metal tube, or any other physical form of conductor), the "cut-and-try" method of adjusting for velocity factor arising from different conductor and insulation properties becomes more complicated due to the increased number of elements and their interactions.  In the case of the 5-band spiderbeam there are 14 elements, so the cut-and-try method rapidly becomes unworkable.

One way to determine the difference in velocity factor, and consequently the change in wire length from that used in the original design, is to erect a horizontally-polarised half-wave dipole made from the originally-specified wire and adjust it for resonance; then build another horizontally-polarised half-wave dipole for the same frequency from the wire you want to use in exactly the same location, and adjust it for resonance (after removing the first antenna, of course).  Taking the ratio of the two antenna physical lengths will provide the necessary scaling factor for converting the original wire dimensions to the new wire dimensions.  This method requires a suitable length of the originally-specified wire, and a suitable location to erect the dipole.  It also requires a means of measuring resonance, as opposed to SWR referenced to a 50 Ohm system - these parameters are not necessarily the same!

Some amateur radio operators confuse resonance with minimum SWR, or with 50 Ohms (resistive).  The definition of resonance is the impedance where the reactive component is zero.  To illustrate this point: a certain antenna may exhibit a feedpoint impedance of 70 Ohms at resonance, ie the load it presents to the transmitter is equivalent to a 70 Ohm resistor with no inductive or capacitive reactance in series.  This will have a SWR of 1.4:1 for a 50 Ohm system.  However; if the antenna is trimmed further it may exhibit a feedpoint impedance of 51 - j 9 Ohms, ie the load it presents to the transmitter is equivalent to a 51 Ohm resistor in series with a capacitive reactance of 9 Ohms.  This will have a SWR of 1.2:1 - this is clearly lower than that for the 70 Ohm impedance case, and will be a better match for a transceiver designed to operate into a 50 Ohm resistive load, but the antenna is no longer resonant.  Whether the antenna being non-resonant matters or not in this particular case is another issue entirely - yagi elements each exhibit their own resonant frequency (assuming operation at a fundamental frequency, and not a harmonic).

For determining the change in velocity factor, I used the method of analysing a horizontally-polarised half-wave dipole with a NEC-based antenna modelling program; viz. EZNEC, because that's what I have to hand on my PC.  (Other programs, including some not based on NEC, could also be suitable.)  This had the advantage of allowing me to "try" different wire configurations representing what I had in the "junk box", and I didn't have to build a thing - I could reserve my building energies for the spiderbeam itself, and not get distracted with other "side" projects.  It also meant that the complete lack of the originally-specified wire in my "junk box" didn't prevent me from adjusting the design and building this antenna.  The ratio of the resonant antenna physical lengths for my preferred "junk box" wire to the originally-specified wire provided the scaling factor required to adjust the wire lengths specified in the spiderbeam construction guide.

Back to top


Changes to the structural design

The spiderbeam construction guide contains information to the "build to print" level, thus allowing someone to duplicate very closely the published spiderbeam design when constructing their own antenna.  This includes components such as the centre hub, etc.  However; I wanted to use components already available in my "junk box" or available in the local hardware store.  This meant departing from the structural design described in the spiderbeam construction guide.

A critical constraint in this departure from the published design was to maintain (or even improve upon) the lightweight attribute of the original spiderbeam antenna design.  The antenna is an integrated structure, and the mechanical strength of the assembled antenna is greater than that of the sum of the parts.  When designing such structures, it is very easy to "over engineer" the individual components, thus increasing the weight of the antenna while realising greater mechanical strength than what is really required.

This is analogous to designing the fuel pump for the first stage of a rocket booster used for lifting satellites into orbit around the Earth.  The actual service life of such a booster is measured in minutes - its job is to lift the stack including the payload to the altitude where the next stage can take over.  That fuel pump needs to operate reliably for only those few minutes - a fuel pump which will operate for much longer may be heavier, thus constituting a reduced mass margin for the booster (meaning a consequent reduction in deliverable payload mass), not to mention wasted engineering effort, etc.  Designers of Formula One racing cars take a similar approach: an engine which lasts one minute longer than the race may be lighter than one which lasts 10 minutes longer than the race; and for a given chassis mass (and driver skill, track conditions, tyres, etc) the car with the lighter engine will be fastest.  A similar approach can be used with antennas - try not to make them stronger and heavier than they have to be!

Boom & spreader poles

Heavy duty, telescoping, coarse fishing poles (also known as "squid poles") were obtained from Haverford Pty Ltd for the boom and spreader poles.  This pole features an 8mm diameter tip at the small end, complete with a rubber cap.  It is constructed from fibreglass, and is advertised as having "nil carbon for radio antennae usage/interference".  The base of each pole features a plastic end cap cemented onto the outside of the bottom section; and incorporates a screw cap allowing access to the individual pole sections.

The figure below shows how I obtained the necessary 5m length from a 7m pole.  (The resulting length of all poles was slightly more than 5m.)  The plastic end cap could not be easily removed from the bottom section, and accommodating this section in the design would have resulted in a more complicated centre hub assembly.  Accordingly, I removed the bottom section from all of the poles.

For each spreader pole (one each side of the centre hub), the load imposed by the wire elements is concentrated near the tip.  A higher rigidity in a such a cantilever-loaded pole is desired.  To achieve this, I removed the upper section of the 7m pole, thus using the middle five sections of the original seven section pole.  I sealed the new tip section (being the small end of the original sixth section) with a black plastic furniture leg plug obtained from the local hardware store.

The load imposed on each boom pole (again, one each side of the centre hub) by the wire elements is distributed along the pole, albeit not uniformly.  Because the pole is somewhat conical, and much of this loading occurs at the larger diameter sections which have more resistance to bending moments, the rigidity requirements are somewhat relaxed compared to that of the spreader pole.  By removing the second bottom section, thus using the upper five section of the original seven section pole, I could realise a sufficiently-rigid pole but lighter than that of the spreader pole.

The ends of each pole closest to the centre hub were sealed with furniture leg plugs slid inside the pole - some of these were built up in diameter with a suitable layer of tape to create a "press fit".

Poles Drawing (not to scale) showing how the 5m spreader and boom poles were realised from the original 7m telescoping pole.  Unscrewing the base of the end cap (shown at the left-hand end of the original pole) allows all pole sections to be removed from inside the lower pole section, thus allowing easy modification of the pole assembly.

As mentioned in the spiderbeam construction guide, the extended lengths of telescoping coarse fishing poles are somewhat variable.  The length is a nominal value; and the four poles I used all had slightly different lengths - no two of them were the same!  Therefore, it is somewhat risky when trying to duplicate an antenna built from such poles by specifying parameters such as "so-and-so millimetres beyond the junction of the third and fourth section", etc.  It is best to specify measurements in terms of being from the base of the pole, or some other reference point such as the centre of the sub-mast, etc.

One potential disadvantage with using telescoping poles is that a section may slip back into the next bigger section while in use.  Methods for preventing this include gluing or taping the joints.  However; I used a hose clamp and a home-made rubber gasket at each joint.  The gasket was cut from some rubber strip from the "junk box".

Section clamp A hose clamp and rubber strip installed at the pole section overlap, thus preventing the pole from telescoping in on itself. The rubber strip serves two purposes: to prevent any chafing of the fibreglass by the hose clamp, and to distribute the compressive force of the clamp and mitigate the risk of longitudinal cracking in the pole.

Centre Hub

I didn't use the centre hub design described in the spiderbeam construction guide, where the centre hub is mounted directly on the mast which will support the antenna.  Instead, I used a sub-mast consisting of 20x20mm aluminium square hollow section with 2mm thick walls.  This sub-mast is approximately 1m long, and serves to anchor the centre hub assembly and the attachment points for the upper and lower stays.  The lower half of the sub-mast carries two mast clamps which are used to secure the antenna to the mast on which it will be mounted.

Another advantage of this approach is that the entire antenna can be built on the sub-mast, which in turn can be mounted on a short post sticking out of the ground.  This allows the antenna to be assembled at a more manageable height than, say, on top of a telescoping mast assembly which may be some 2m above the ground.

Aluminium angle section, 30x30mm, 3mm thick, was used to anchor the squid poles.  Some short lengths of 50x50mm aluminium angle section, also 3mm thick, was used to connect the angle section anchoring the boom poles to the angle section anchoring the spreader poles.  Gusset plates manufactured from 100mm aluminium plate, 3mm thick, provided some further mechanical stiffening.

Hose clamps with rubber gaskets were used to secure the bases of the squid poles to the aluminium angle sections.  These clamps were bent to fit around the angle profile prior to installation of the squid poles.

Centre hub The centre hub assembly for spiderbeam antenna.  Note the square cross-section sub-mast, aluminium angle sections supporting the fibreglass boom and spreader poles, and profiled hose clamps attaching the fibreglass poles to the aluminium angle sections.

Stays

I have used the term "stays" to describe the guy ropes which run from the top and bottom of the sub-mast to the boom and spreader poles, and between the spreader poles themselves.  (I use the term "guys" later to describe the guy ropes running from the insulators at the ends of the elements to the tips of the spreader poles.)

There are three sets of stays: the upper stays, which are anchored at the top of the sub-mast and run to the middle and tips of the boom and spreader poles; and lower stays, which are anchored at the bottom of the sub-mast and run to the tips of the boom and spreader poles; and the inter-pole stays, which connect the middle of each adjacent pole.

I used 600 lb (272kg) breaking strength mono-filament line for the upper stays and inter-pole stays, and 500 lb (227 kg) breaking strength line was used for the lower stays.  It is almost impossible to tie knots in this line; and the line manufacturer also provides aluminium crimp sleeves for the purpose of securing items to the ends of a piece of line.  The line is sold as "leader line" for game fishing - this and the sleeves were purchased from a local fishing tackle supplier.  Crimping is easily achieved with a stock-standard automotive electrical wiring crimping tool.

At the sub-mast, each line was looped through a 2mm thimble and crimped.  At the poles, a short length of garden watering "dripper" hose was used to cover the line looping around the fibreglass pole.  Each loop bears up against a hose clamp.  The dripper hose protects the fibreglass pole and mono-filament line from mutual abrasion, and also protects the line from being abraded or cut by the metal hose clamp.

Tip of boom pole The tip of a boom pole, showing the rubber end cap supplied with the original 7m telescoping fibreglass pole.  In the centre of the picture is an upper stay bearing against a hose clamp - note the household garden "dripper" hose protecting the mono-filament line from the hose clamp, and protecting the fibreglass pole from the mono-filament line. At the far right-hand side of the picture is a hose clamp anchoring the centre of one of the 20m parasitic elements.

Per the spiderbeam construction guide, additional upper stays were installed for our high-wind environment.  There are four of these, and each one runs from the top of the sub-mast to a point approximately 3m along the pole from the centre hub.

Stay clamps A view of some of the hose clamps anchoring the stays.  The clamp on the right-hand side doubles as a section clamp (ie, it prevents the pole from telescoping in on itself) as well as anchoring two inter-pole stays.  Note the household garden "dripper" hose protecting the mono-filament line from the hose clamp, and protecting the fibreglass pole from the mono-filament line.  The clamp on the left-hand side is dedicated to the inner upper stay (added for high-wind resilience).  Originally, this stay bore against the section clamp on the right-hand side; however, as the elements were installed and the boom and spreader poles became loaded, re-tensioning of the upper stays was required - the easiest way to do this was to install another clamp outwards of the section clamp, and adjust its position in order to change the tension of the stay.  The additional weight and complexity arising from fitting four extra clamps and rubber strips was not prohibitive.  In the foreground is a close-up view of some 2mm diameter "army cord" guy rope used to secure an end-of-element insulator to the end of the spreader pole.

Lower stays The lower stay attachment points at the bottom of the aluminium square cross-section sub-mast.  Note the thimbles for anchoring the mono-filament line to the sub-mast, and the aluminium crimp sleeves.  Note also the mast clamp, and the star picket on which the antenna was mounted during construction.

Guys

These are the lines connecting the insulators at the ends of the wire elements to the spreader ends.  The spiderbeam construction guide specifies mono-filament line for the guys - instead, I used 2mm diameter "army cord" which was already in the "junk box".  These lines do not carry appreciable tension, and solar ultra-violet (UV)) radiation resistance was a primary concern - I have used this cord for outdoor antennas for a few years, and have not experienced any failures attributable to UV-induced degradation.  The "ring" on which the guy lines terminate was fashioned from 25mm (1") diameter PE agricultural irrigation pipe, selected for UV resistance.

Guy rope termination The guy rope termination at the end of a spreader pole.  (It does not look pretty, but it works.)  The outer end of the spreader pole is out of picture on the right-hand side.  Each guy loops back to a tie-off loop (out of picture to the left-hand side) where it is tied off after removing excess slack.  On the left-hand side of the termination is the end of the upper stay, while on the right-hand side is the end of the lower stay.

Back to top

Changes to the electrical design

I made a number of changes to the electrical (or radio frequency) design of the 5-band antenna described in the spiderbeam construction guide.  This was purely due to what was available in the "junk box", and nothing to do with any attempts to "improve" the original design.  As mentioned above, the element lengths were adjusted to accommodate the different wire used.  Also modified or substituted were the insulators, feedlines, and the balun.

Elements

This antenna uses approximately 104m of wire!  The spiderbeam construction guide specifies a Wireman wire which has good properties for an antenna needing to survive multiple set-ups and tear-downs in portable environments featuring adverse weather conditions.  I didn't have any of this wire on-hand, but I did have lots of 1mm diameter bare copper wire which I used instead.  The slightly different diameter of this wire, and the fact that it was bare as opposed to insulated, meant it has a different velocity factor than that of the Wireman wire specified in the spiderbeam construction guide.  As discussed earlier, this means the element lengths using this wire are different from those stated in the spiderbeam construction guide.

Insulators

I didn't have any of the very nice looking insulators described in the spiderbeam construction guide, either.  But, living on a rural property means I do have some handy offcuts of UV-resistant, PE, agricultural irrigation pipe which could be used for fabricating insulators.

End-of-element insulators

The end-of-element insulators were fabricated from 20mm lengths of 25mm (1") diameter PE agricultural irrigation pipe.  (The insulators were cut from a length of pipe which the dogs had chewed, thus rendering it useless for carrying water, but quite recoverable for the purpose of converting into insulators ...)

End-of-element insulator A view of an insulator located at the end of a wire element.  This is fabricated from 25mm (1") diameter PE rural "poly" pipe.  A 20mm long section of pipe is cut off with a hacksaw, and two 3mm diameter holes are drilled through the wall of the pipe, diametrically opposite to each other.  The picture shows the wire is fed through the hole on the right-hand side, and secured on the inside of the pipe section - this can be by tying a knot (per the Spiderbeam Construction Guide), or by soldering (which is what I did).  The 2mm diameter green "army cord" is fed through the hole on the left-hand side, and a knot tied to secure it to the insulator.  Excess cord is trimmed with a pair of sidecutters.

Driven element centre insulators

The driven element centre insulators were fabricated from 50mm (2") ID PE "rural" poly pipe (colloquially known as "green stripe poly").

Driven element feedpoint insulator A view of a driven element centre insulator located on the boom.  The twin-lead feedline enters the insulator via a hole on the right hand side of the picture.  The element wires exit the insulator via holes on the near and far side of the insulator as seen in the picture.  The red- and blue-coloured heatshrink tubing slid over the twin-lead feeder wires matches similar pieces of heatshrink at the balun end of the feeder, thus mitigating the risk of connecting the a driven element out of phase.  The insulator is secured to the boom pole with three cable ties.

Feedlines

For constructing the four feedlines running between each of the 20, 17, 12 and 10m driven elements and the balun; the spiderbeam construction guide specifies the same Wireman wire as it specifies for the elements.  As I didn't have any of this wire; instead I used some PTFE-insulated, stranded wire from Storm Products which I had on-hand.  Short lengths of heatshrink are used to tie together the two wires forming each of the feedlines.  Using this wire afforded a low characteristic impedance line with lower loss than that of a similar line fabricated from PVC-insulated "figure 8" wire.  I considered using 300 Ohm perforated TV twinlead, but was concerned that the higher characteristic impedance of this line would transform the feed point impedances of the higher frequency yagis to an undesirable impedance at the balun, thus degrading the overall performance of the antenna.

Balun

I used the GM3SEK high-frequency balun design with two additional turns of RG-142 coaxial cable.  (I didn't have any large toroidal ferrite cores on-hand, but I did have the cores specified by GM3SEK for his balun designs.)  By coincidence, I had a polycarbonate box of the same type specified in the spiderbeam construction guide; so, as far as outward appearances go, the balun looks just like the specified balun.

Balun
The GM3SEK-style balun used for the spiderbeam antenna, featuring a SO-239 (UHF) coaxial connector, and brass fasteners on the top and sides of the enclosure for the balanced connections.  Note that the adjacent side and top connections are paralleled.

The 15m driven element, and the feedlines for the 17m and 20m driven elements, connect to the terminals on the sides of the enclosure.  The feedlines for the 10m and 12m driven elements connect to the terminals on the top of the enclosure.

Back to top


Antenna assembly

The antenna was manufactured and assembled in the following sequence:

1.    Fabricated the centre hub assembly and sub-mast.

2.    Installed the centre hub on a post approximately 1.5m high in the back yard.

3.    Installed the boom and spreader poles on the centre hub, and installed the hose clamps which ensure the poles did not telescope back into themselves.

4.    Installed the upper stays then the inter-pole stays.  (The lower stays were not installed at this time, as they were not needed at this point in the construction, and would get in the way whenever I had to duck under the fibreglass poles to access other parts of the antenna.)

5.    Constructed and installed the parasitic elements, but did not tension the guys.

Wire clamp for parasitic element A hose clamp securing a parasitic element wire to the boom pole.  Note the rubber strip under the hose clamp, which helps distribute the compressive force of the clamp, thus mitigating the risk of longitudinal cracking in the pole.  The gap in the rubber strip accommodates the wire, which is covered with insulation tubing to prevent chafing of the fibreglass pole, and reduce damage to the wire by the hose clamp.  Beyond the wire clamp is another hose clamp and rubber strip - this is installed at the pole section overlap, thus preventing the pole from telescoping in on itself.

6.    Constructed and installed the driven elements.  Again, the guys were left loose.  (This step and the preceding step combined probably took the most time of all.)

7.    Tensioned the guys, starting with the inner elements and working out towards the ends of the boom by alternating from one boom pole to the other (ie one before mast, one after mast, next one before mast, next one after mast, etc).  This method meant all of poles stayed straight when viewed in the horizontal plane - the poles are quite flexible and will bend out-of-true in order to equalise the tensions in the system.

8.    Re-adjusted the upper stays so that the poles curved upwards slightly.

9.    Constructed the balun, installed it near the top of the sub-mast, connected the feedlines running to the 20/17/12/10m driven elements, and directly connected the 15m driven element.

Sub-mast The view of the sub-mast assembly and inner part of boom, looking towards the rear of the antenna.  Note the four feeders connecting the driven elements to the balun; and the anchor points at the top of the sub-mast assembly for the upper stays.  The upper mast clamp can be seen just under the aluminium angle sections supporting the fibreglass poles.

10.    Installed the lower stays.

Another view of the antenna on ground Another view of the sub-mast assembly looking along the boom towards the rear of the antenna.  The 10m and 12m driven elements are in the foreground, and connect to the pair of terminals on the top of the balun enclosure.  The 20m and 17m driven elements are in the background, and connect to the side terminals on the balun enclosure.  The 15m driven element is connected directly to the side terminals on the balun enclosure.

11.    Connected the coax feedline to the balun, applied water-resistant tape to the connector, and cable-tied the coax in place.  (This was easier to do with the antenna mounted on the ground, because the coax connector on the balun is some 3m above the ground when the antenna is mounted on a collapsed, telescoping mast which in turn in mounted on a rotator cage.)

Completed antenna on post The completed spiderbeam antenna on post, prior to installation on telescoping mast.

12.    Installed the entire antenna assembly on the mast.  The overall weight of the antenna was such that I could lift it with one hand; however, the 10m largest dimension makes it somewhat unwieldy when moving it.

Back to top


Rotator cage

One advantage of using a not-very-heavy antenna such as the spiderbeam, and a guyed mast which can rotate within the guy sets, is that the rotator can be left on the ground.  This reduces the weight and topload carried by the mast, reduces the work required to raise and lower the mast and antenna, reduces the length of cable required to connect to the rotator (and helps reduce potential RF decoupling issues associated with that cable on the mast), and enables easy maintenance of the rotator itself.  The rotator is housed in a home-brew cage, as shown in the pictures below.  This particular rotator cage employs a bolt-together construction; however, a welded cage would do just as well.

Wide screen shot of the rotator A view of the home-brewed rotator cage showing the rotator sitting on its support plate; the mast and upper bearing assembly; and three tensioned stabilising guys constructed from fencing wire, each connecting the top of the cage to star pickets via a ratchet tensioner.  The star picket ends are covered with plastic caps for safety reasons.
Close-up view of the rotator cage
A close-up view of the rotator cage showing the upper bearing assembly.  This is constructed from pieces of 50mm (2") ID PE rural "poly" pipe, which are secured to the mast using large-diameter hose clamps.  Four pieces of "U" section aluminium channel bear lightly on the PE pipe, thus keeping the mast centred at the top of the cage.  Also shown are two of the ratchet tensioners - these are sold in rural supply stores for fencing work.  The rotator cable (carrying DC power and position-indicating voltages) is shown disappearing to the right-hand side of the picture.

Back to top


Erecting the telescoping mast

Because the spiderbeam antenna is so lightweight, it can be mounted on a relatively light mast or tower.  I used a Spiderbeam 14.5m heavy duty telescoping mast which can be rotated within its sets of guys.  As described above; this mast is mounted on the home-brew rotator cage, thus allowing the whole mast and antenna to be remotely steered to the desired azimuth angle.

To erect the mast with the antenna on top (and a length of coaxial cable hanging from it), I used the procedure for erecting the system single-handed, as described on the Spiderbeam website, albeit with some variations.  I didn't need the tripod described in the procedure, as its purpose was served by the rotator cage.  After lifting the first few sections of the telescoping mast, the all-up weight of the antenna plus the deployed mast sections becomes substantial, and can be somewhat difficult to lift safely without resorting to some form of mechanical assistance.  For this, I used a "gin pole" (a term used in the radiocommunications tower construction industry) featuring a block and tackle system as the lifting mechanism.

Pole and block and tackle for
              lifting mast A view of the gin pole with "double tackle" for lifting the telescoping mast sections.  The pole assembly is shown on the right-hand side, with a double block (ie containing two pulleys) fixed at the top.  (The bottom of the pole is secured to the rotator cage.)  On the left-hand side is the 14.5m telescoping mast, showing the mast section being lifted at approximately half way through the lifting action.  The locking clamp which secures this section of mast to the section below it can be seen towards the bottom of the picture.  Just over halfway up the picture is the moveable clamp attached to the mast section being lifted.  This clamp carries another double block - the moving block.  The rope is fixed to the top of the gin pole and reeved through the four pulleys - ie the tackle is "rove to disadvantage" - and the downhaul can be seen running down the left-hand side and behind the mast towards the bottom of the picture.  This arrangement provides 4:1 mechanical advantage, thus allowing me to erect the mast single-handed.  The other ropes in the picture are guy ropes secured to a mast guy plate out of scene at the top.

Back to top


Performance

To date, only SWR measurements have been conducted on the antenna, as tabulated below.  These measurements were obtained with the antenna at approximately 8m height.  The measuring device was a MFJ-269B Antenna Analyser, and measurements were taken at the end of a 26m length of RG-8A/U coaxial cable which runs down the mast to the cable interconnect point.

Band
Lower 2:1 SWR Frequency [MHz]
Minimum SWR Frequency [MHz] Upper 2:1 SWR Frequency [MHz] 2:1 SWR Bandwidth [kHz]
Minimum SWR
20m
13.84
14.04
14.50
660
1.1
17m
17.84
18.02
18.42
580
1.2
15m
20.79
21.23
21.50
710
1.6
12m
24.74
24.87
25.04
300
1.5
10m
27.59
28.79
29.37
1780
1.2

The SWR values are commensurate with those listed in the spiderbeam construction guide.  Clearly, some further adjustment could result in the amateur radio bands sitting centrally within the respective 2:1 SWR bandwidths.  This is especially the case with the 10m band, where the antenna doesn't cover the upper portion of the band.  As at the time of writing, this adjustment has not been performed.

SWR is one thing, but it doesn't tell the radio operator anything about important antenna performance parameters such as main beam or sidelobe gain, front-to-back ratio, etc.  Some of these parameters are not the easiest quantities to measure for a HF antenna.  For example, to date I have not attempted to measure gain.  Instead, I have to rely on other indications; eg the fact that, as I steer the antenna in azimuth, signals from certain parts of the world rise in strength while signals from other parts of the world weaken - this indicates there is gain in the desired direction, but doesn't indicate how much.  Variables such as changing ionospheric conditions resulting in changing signal strength and varying angle of arrival for received signals (even in the short term) means the measurement of these quantities using signals heard "on the air" can be plagued with inaccuracy, and give rise to misleading conclusions.

Back to top


Conclusion, and where to next?

Building a home-brew spiderbeam antenna is not like building a home-brew dipole antenna - it's not a task which can be completed in an afternoon.  And, the task is even more complicated when altering the original design, and having to solve design and construction problems "on the fly".  After putting in the time and effort to build this antenna, I am tempted to leave it as is, and just use it.  However; I believe it would be most instructive to model this as-built version of the antenna using NEC-2, and determine if there are any realisable performance improvements which could be made.

Spiderbeam antenna at dusk The spiderbeam antenna at dusk.  The guys and end-of-element insulators can be clearly seen.  The boom runs from bottom right to top left of the picture, and the spreader pole ends are in the bottom left and top right corners.  The loop in the coaxial cable allows the cable to run around the upper guy ring when the antenna is rotated through the rotator's full travel range of 450 degrees.  The antenna is pointing to the top left, ie Northwest, which is the short path to Europe from my QTH.

Another shot of spiderbeam antenna at night Another shot of my home-brew spiderbeam antenna at night, taken adjacent to the telescoping mast.  The boom runs from bottom right to top left, and the spreader pole tips are at top right and bottom left. Note the square of light formed by reflection from the inter-pole stays, and the pattern of the end-of-element insulators seen as dark dots against the sky.  (The lower guy set and mast sections are over-exposed due to a floodlight positioned at the base of the mast for this shot.)

Back to top

This page was created by Mike Dower VK2IG: 15 Oct 2014; last updated 21 Feb 2015.  Material may be copied for personal or non-profit use only.