Installing a New Tower and a Mosley TW-33-XL Beam

September 2006

    During the spring of 2006, with the solar cycle near the bottom, I found myself operating most of the time on 30, 20 and 17 metres (10.1, 14.2 and 18.1 MHz.)  20 metres was great as I was using my Hy-Gain TH-5 beam.  I found 30 and 17 metres to be much more of a challenge because of a simple dipole on 30 and a Windom on 17.  The lack of gain and directivity combined with the lack of sunspots meant I was missing a lot of DX on those bands.  There wasn't a lot I could do about the progress of the solar cycle, but I knew I could make improvements in my antenna system.  One approach I had been considering for over a year was to get a dual band antenna for 17/12 metres.  We are a few years away from any significant opening on 12 metres, but if I was going to do anything about 30 and 17, I thought I might as well do it for 12 metres too. 

    Many of my fellow DXers are using the 12/17 metre Cushcraft A3WS antenna with good success.  I came close to choosing this as a solution, but I wanted to do something for 30 metres as well.  I did quite a bit of research and found that Mosley Electronics sell a TW-33-XL antenna.  This is a WARC triband beam (30/17/12 metres) with 3 active elements on each band.  It weighs about 55 pounds and has an 18-foot boom.  It is very close in wind surface area and wind load to my TH-5.

    The gain and front to back figures looked good (measured in dBd, the forward gain/FB: 12 metres = 7.2/20, 17 metres = 6.8/20, 30 metres = 6.0/18.)  I did a lot of comparison and thinking about this, including several phone calls to Gary Wurdack Sr of Mosley Electronics.  In the end, I decided to go with this antenna, but the obvious question was "where do I put it?"  My existing Delhi DMX-48 was near its capacity holding the TH-5 in the air, so stacking the Mosley there was not an option.  I certainly did not want to take the TH-5 down and replace it with the new WARC band antenna.  The only solution was to put up a second tower.  In July I ordered another Delhi DMX-48 tower and the Mosley Beam.  The tower arrived on 29 August (I asked the vendor not to deliver it earlier as I did not plan on putting it up until September.)


The Tower

    The first thing I had to do was to dig a hole for the base.  Delhi recommend a 4 X 4 X 5 foot concrete base.  My existing tower base is a slightly larger one, and hasn't moved since it was put there in 1990.  I decided to play it safe and use the same size.  This is 5 X 5 X 5 feet, which is a 5-foot (1.52-metre) cube.

Marking off the 5-foot hole where the tower base will go

    I have never dug a hole for a tower base by hand.  Those who have say they never will again!  The soil here in most of Nova Scotia is tightly packed, hard and has more than its share of rocks.  The only way to dig a hole is to hire a backhoe.  I certainly preferred that method over trying to displace 125 cubic feet of earth with a pick and shovel.


Starting the hole

    My lot isn't that big, so I had to do a lot of head scratching regarding where to put the new tower.  I ended up selecting a site beside our doghouse (we no longer have a dog, but the name of the previous resident is still on the house) and in front of our clothesline pole.  The clothesline pole had a 20-foot (6.1-metre) mast clamped to it and doubled as a support for a 144 MHz quad antenna.  This quad is no longer in service so, for the first time in 20 years, the clothesline pole is not be supporting an antenna of any kind.

    Given the restrictions I have due to lot size, I was only able to get the towers approximately 70 feet (21.3 metres) apart.  This allowed me to re-position my 7 MHz dipole between them at a height of about 45 feet.  I am not sure what to do about 3.5 MHz yet, but there are a number of tall trees in the yard that could work in conjunction with one or both towers to give me a resonant antenna for that band as well.


Hole completed!

      The hole was dug on Friday, 31 August.  The entire job took about an hour, as opposed to a week if I had tried to do it by hand.  I asked the backhoe operator to dig as close to a 5-foot cube as possible. That's (5 X 5 X 5 = 125) / 27 = 4.6 cubic yards (3.52 cubic metres) of concrete . . . I ordered 4.5 cubic yards (3.44 cubic metres.)  I kept a few big rocks around in case I ran short of concrete -- when the concrete was about half poured, we tossed them in . . . rocks are cheaper than concrete.  A 5-foot cube of concrete will not move.  You can't get enough weight or wind load on a 48-foot heavy duty DMX tower to do so.  The tower will fold over or collapse before the concrete block moves.  Concrete weighs approximately 150 pounds (68 kilograms) per cubic foot, so 125 cubic feet weigh 18,750 pounds (8,494 kilograms.)  That's over 9 tons.  It isn't going to move! 
     There are many formulas used for concrete.  I am not a civil engineer, and the only measure I hear commonly used is the "compressive strength."  As I understand it, this varies with the formula used, and is measured by making test cylinders 6 inches in diameter and 12 inches in height.  These cylinders are cured for 28 days and tested by compression until they are crush.  3000 PSI crushes when 3000 PSI is applied, 4000 PSI when 4000 PSI is applied and so on.

    I used 5000 PSI contractor grade concrete.  I don't know if this is far more than needed or not.  When I ordered concrete for the first tower 16 years ago, the company asked if I wanted "4000 pound or 5000 pound."  I went with 5000 PSI then, and I have had no problems since.  If it was good enough for the first tower, then it will be good enough for this tower as well.
    In my view, it is better to use a bit of overkill, and do it right the first time.  If you are serious about the hobby, you want the tower to stay put for 30-40 years, not 8-10 months. 
Tower base with freshly poured concrete
    The next morning, 1 September, the concrete arrived.  I didn't use forms for the base.  We made a 5-foot square box out of 2 X 4 and put it around the top of the hole, with about 1/2 of the 2 X 4 above ground.  We let the concrete go where it wanted in the hole.  The 2-3 inches of exposed 2 X 4 makes the base look like a perfect square . . . although underground it is only as square as the skill of the backhoe operator.  He was extremely good at this, as you can see from the picture of the hole above.  An hour or so after the concrete had been poured, but before it has started to set up hard, I tapered off the edges just a bit so that the centre of the block was about 1/2-inch higher than the edges.  I did this so that water, etc. would run off and not freeze in the winter. 
Bottom section with "form" removed
    This base looks great, but it is a slightly distorted underground.  This makes absolutely no difference.  After a few days when the concrete had set firmly, I dug out the frame and raked a bit of ground up against the block.  It may look like it was done by a professional, but I am just an amateur.  Several days later I had the excess soil and rocks removed and things were starting to look more or less normal in the back yard.  Although those who do not appreciate Ham radio and their associated towers, etc. may think otherwise.
Grounding system installed
    There are two types of grounding for towers and antennas in general, in my opinion.  They are lightening protection grounding and RF grounding.  For lightening protection, I've used two approaches.  On my first tower, I bolted a piece of heavy (8 gauge) copper wire to the bottom of one of the stubs (legs) that go into the concrete block, and attached this to a 5-foot rod that I drove into the ground in the bottom of the hole.  (You obviously prepare all this before the concrete is poured.)  This gives a path to ground from the tower through the steel stub, into the heavy wire and then into the rod, all through/beneath the concrete block.  It does nothing as an RF ground.  I have since heard that in rare cases a heavy lightening strike can cause the concrete block to crack if the path to ground runs through it, as described above.
    In order to avoid this possibility, I took a slightly different approach with this tower.  I attached a copper lug to two of the steel legs where they bolt to the stub (I just used one of the three tower bolts.)  I then ran this to a couple of 8-foot copper clad steel grounding rods (the type used in electrical wiring for houses.)  This provides a good path to ground that avoids the concrete base.  I have no way of knowing if it is better than the way I did my first tower.  In my opinion either method should work just as well.
Close up of lightening ground system
    This is a close-up of one of the two lightening grounds.  In the unlikely event that the tower is hit, this should provide the shortest and easiest path to ground.  However, the physics of lightening strikes is not well understood.  More properly put, at such high voltages and currents, electricity does not always behave in a predictable manner.  All a person can do is prepare for the worst, hope for the best and take what you get.
    As regards to RF grounding, on my first tower I also attached a copper lug to each of the three steel legs where they bolt to the stubs.  I ran this to buried radials that spread out from the tower. The radials are similar to what you would use for a vertical antenna (although I am not using the tower as a shunt fed vertical or anything like that.)  I probably have 75 to 100 of them of various lengths.  I also have buried "ground mats" made of chicken wire here and there around the tower base and in other places in the yard.  They are all tied together with buried copper wire of various types and sizes.  I plan on doing something similar with this second tower and attaching both systems together.

    My theory is that the best thing you could ever have is a copper plated back yard!  Since this isn't practical, I just bury as much surplus copper wire as I can find and attach it to these "radials."  It will increase the apparent conductivity of the soil, and it makes, in my opinion, a better ground plane for signal take off.  All of this may be a waste of time and wire, but over a 20 year period I've worked a lot of DX, and I feel the copper buried all over the yard has helped.  You won't find this in any radio books or discussions on antenna theory.  It makes me feel better, and I believe it helps.
"Lining up" the tower on the ground
    In theory, each 8-foot tower section is a perfect triangle, and it tapers from bottom to top, getting smaller as the tower gets higher.  In reality, due to slight differences in production at the factory and/or distortion during shipping, there is always a "best" orientation where the sections slide together easiest.  They will usually go together in any of the three positions, but there always seems to be one or two that work better.  Over the years, I've learned that it is a good idea to determine the best orientation while the tower is on the ground, and then mark the corresponding legs with a felt marker.  Then, when a section is hoisted into the air, the person(s) on the tower can just drop them into place.  It's a lot easier to line up these sections on the ground than to do it in the air.
Rotator, mast and thrust bearing installation
    When I installed my TH-5, I lowered the rotator plate from the factory default of 39 inches in the top section to the middle of the next section down.  I did this because I was using a 20-foot mast with a 144/440 MHz Comet vertical on top.  I was concerned about the sideways leverage on the rotator.  With this tower, I am only using an 8-foot mast and there is nothing above the TW-33-XL Yagi.  The vast majority of the weight and leverage is about a foot above the top plate.  I therefore decided to stay with the factory default configuration.
     My existing rotator is a Yaesu G-800SA, and the thrust bearing is a Yaesu GS-065.  I have had excellent success with these, so I decided to use the same models on the new WARC tower.  I had to drill holes in the top plate to accommodate the GS-065 bearing, and also in the rotator plate for the G-800SA.  The locations of these holes is fairly critical.  The reason for this is that the rotator and thrust bearing have to line up so that the mast is vertical.  If either is slightly off, the mast may not be able to turn easily, and could bind.  Both the rotator and the bearing have adjustments that will compensate for this a bit, but it is best to get the them aligned as close to centre as possible.  The mast is thick walled aluminium and I re-enforced it by inserting a tight fitting steel pipe.  I drilled a hole through the then near the top and inserted a bolt.  This probably wasn't necessary, but it is easier to do a little extra than to have something come loose at 50-60 feet in the air.
The Antenna
    My Mosley TW-33-XL arrived on 26 September 2006.  It came in two boxes around noon, and I made sure that all the parts were there.  Everything looked good so I decided to start assembling it.  The instructions were clear and well written.  The antenna was pre-drilled and colour coded (as are all Mosley antennas.)  No measuring and adjusting are required.
Diagram of the TW-33-XL antenna
    Although everything is supplied pre-measured and ready to go together, I read the instructions several times.  It's easy to make a simple mistake like reversing the direction of a trap that will not be obvious to the eye.  I've assembled a number of Yagis and the old carpenter's saying of "measure twice and cut once" applies here.  It probably should be "read three times and assemble once", but however you word it, it's far better to take your time putting it together than to rush.  There seems to never be enough time to do it right the first time, but there always is plenty of time to take it down and re-do it because the SWR is 8:1! 

    The first step was putting the 18-foot boom together.  It consists of three 6-foot sections of thick walled 2-inch aluminium. They are bolted together with eight stainless steel bolts.  Then the mast plate is attached to the balance point, which is between the radiator (driven element) and reflector locations.


TW-33-XL Assembled
    As previously mentioned, all Mosley antennas are pre-drilled and colour coded so no measuring and adjusting are required.  It still took 6 hours to complete the assembly.  This antenna looks larger than I had expected, although I knew the longest element (the reflector -- closest to the house) was almost 35 feet.  They always look larger on the ground than when they are 50 feet (15 metres) in the air.  This thing may be interesting to get up on the tower given the size of my lot, but I haven't seen an antenna yet that can't be installed.  Sometimes it just takes creative thinking!
Keep it Simple
    This is typical of Mosley antennas.  They try to keep things as simple as possible.  Most Mosley beams do not require Baluns or other balancing devices to provide a balanced radiation pattern.  The braid ("ground") side of the coax goes to one half of the radiator and the centre conductor ("hot") goes to the other.  Note the grounding strap that ties the braid and half of the radiator to the boom, and that the other half of the radiator is electrically "floating."  The simpler things are, the less there is to go wrong.  With this design, Baluns and similar devices only introduce loss and increases the complexity of things.  Some Hams disagree with this, and they use Baluns with Mosley beams anyway.  They don't hurt (other than to add a miniscule bit of loss), but they don't do any good either.  All you need is ordinary RG-213 coax made waterproof at the Y with 3M tape and ScotchKote (or similar water resistant materials.)  The reason for this is because linear radiators have a driving point impedance close to 52 ohms when open at the centre.  The best and simplest way to feed such a radiator to connect a 52 ohm line directly.
    Other designs (such as my Hy-Gain TH-5) use Gamma or T-Match systems that usually need to be adjusted to present a 52-ohm impedance.  Many of these antennas do benefit from the use of a Balun.
Coaxial Choke
    Sometimes RF will "leak" back down the coax shield and cause interference problems in household electronic devices such as TVs, computers, telephones, etc.  One of the easiest ways to prevent this is to use a coaxial choke.  Some Hams confuse a choke with a Balun.  They are not the same.  A Balun is a transformer that converts unbalanced feed lines to balanced feeds, and as described above, it is not needed with this antenna.  A coaxial choke is just a coil of about 10 turns of feed line with about a 10-inch internal diameter.  It is made by rolling up your coax like a rope in the size mentioned and taping it in place.  I used a plastic 10-inch form to make it perfectly round, but this is not really necessary . . . just looks neater.  This is attached as close to the radiator as practical and then tied or taped securely to the boom.
    If you are not experiencing RF down the feed line, a choke is unnecessary.  Remember, simpler is better!  I don't have one on my 20/15/10 metre TH-5.  The reason I put one on my TW-33-XL is because every 30-metre antenna I have used in the past 20 years has had RF leakage.  This is not a flaw in the TW-33-XL design or anything that is universal to 10.1 MHz.  Some Hams experience the problem at other frequencies and some never see it at all.
    On 30-metres, it has to be something to do with my location, such as closeness to an artificial ground (roof flashing, metal gutters, other antennas -- anything that can give a capacitive affect to the antenna.)  Whatever it is, I never have been able to find it.  I simply use a coaxial choke on anything that transmits on 10.1 MHz.
Installation Day
    We began to install the tower and antenna on the morning of 30 September 2006.  We had initially planned on doing so the day before, but it rained that day.  There was also a considerable wind blowing, so it had to be put off until the next day.  The weather on 30 September was perfect for this sort of work, with little to no wind, clear skies and the temperature between 18-20 C.
VE1JDS Preparing to Add Another Section
        I decided to ask Jim Sutherland, VE1JDS, to give me a hand installing the tower because he is a professional rigger.  He has all of the equipment and the expertise to do it right, and to do it safely.  He came to my house around 8:00 AM and the two of us put the tower up.  Actually, Jim did most of the work and I just looked after things on the ground.  I am not comfortable climbing, and it is better to have a professional do that part than take the chance of falling off.  By 8:30 AM we had the tower half completed.  Jim used a gin pole to haul the tower sections up and then I then helped lower them in place with the rope (from the safety of the ground!)  As soon as the tower section was in place, I put the bolts for the section in a container tied to the rope and he pulled them up and bolted the section in place.
VE1JDS Completing Tower
    By about 9:30 AM the tower was completed.  Not bad for about 90 minutes work!  I already had the rotator, thrust bearing and mast installed and aligned in the top section.  When it was time to install it, we just hauled it up the same as the rest and Jim bolted it in place.  Note that if you look at the tower from the bottom, it appears to be tilted slightly to the right.  This is an optical illusion caused by the clothesline pole.  That pole is actually tilted a bit toward the house and it makes the tower appear to be leaning slightly.  When we put the first section into the concrete base, we were very careful to ensure it was straight.  As the sections were added and the tower got higher, it became clear that we had done the job correctly.  (And if it is leaning slightly to the right, we did it on purpose because that is the direction of the prevailing wind!)
VE1JDS Attaching Beam
    I thought getting the antenna up to the top of the tower was going to be a problem with all of the trees, etc.  However, it went up very nicely with the two of us pulling it up and guiding it with ropes.  Once we got it on the top plate, Jim bolted it to the mast, attached the coaxial choke, RG-213 feed line and rotator cable.  Then he aligned the direction of the boom to match the rotator controller.  The last job was to waterproof the joints with 3M tape and ScotchKote.  It took about an hour to get the antenna from the ground to the top, and do all of the connections, etc described above.
VE1TK Making Final Adjustments
    After Jim left, Mike VE1TK and I relocated my 40-metre dipole from its old home in the trees to between the towers.  You can see one end (the white insulator tied to the rope) above.  The ends are fed through small pulleys so the dipole can be lowered if a storm is on the way, or if adjustments need to be made to it.  Mike also installed some pieces of old garden hose over the spots on the rotation loop that rub up against the tower leg when the antenna is rotated.  This stops the coax from chafing against the leg and wearing the outer shield off over time.  It is one of the many "secrets" Mike has come up with over the years to prevent problems from happening.  It is a lot easier to do a little preventive maintenance on a nice sunny day than to be without the use of your equipment if something wears or breaks in the winter . . . when the cold, ice and wind make climbing the tower next to impossible.
Feed Point
    This is the feed point at the radiator (driven element) with the coaxial choke installed.  The choke has a PL-259 connector on either side so that it can be bypassed or removed if necessary.  Generally it is best to have a continuous piece of RG-213 feed line from the radio to the beam.  In this case, I decided to make a PL-259/barrel "disconnect" point before and after the choke.  That way, if problems develop, either the 2-foot piece of coax feeding the driven element, the coaxial choke or the main piece of feed line can be replaced quickly and easily.  Some Hams make the case that PL-259/barrel splices are potential sources of problems and can cause loss.  Regarding loss, at frequencies below 30-MHz, this will be negligible.  It is only at 50-MHz and above that one really has to be concerned about impedance bumps, etc that can be introduced by these splices.  As for waterproofing, I use stretchy 3M tape that seals like rubber, and then I coat it with ScotchKote to make a water tight seal.  I have never had one get wet in it in over 20 years.
Looking up!
    This picture was taken from the bottom of the tower looking up at the antennas as soon as we'd finished everything.  If you look closely, you can see the 40-metre dipole on the left side near the top, heading off toward the other tower.
If One is Good, Two Must be Better
    This picture was taken across the road from my house.  Both towers are exactly the same height (48 feet -- 15 metres), but the one on the left with the Hy-Gain TH-5 and the Comet dual-band vertical appears to be higher.  At present I have no plans to stack any other antennas on top of the right tower. 
    We finished around 3:30 PM and put all of the tools away.  I checked the SWR curves for all three WARC bands and they were very low (and close to what Mosley had obtained and supplied with the documentation.)  It was a bit late in the day to try 17-metres, although I did make several contacts with European stations.  I decided to concentrate on 30-metres.  In 4 hours I made dozens of contacts and worked 20 individual DXCC countries: SP, OX, SV, ZA, TF, GW, LX, VE (VY0ICE in Iqaluit), EA, UA3, UT, G, 4X, LZ, VK, FG, HA, F, I and HP.  The most surprising contact was with VK6RZ, who was 579 at 22:50 UTC.  I would never have even heard him on my wire dipole that I'd previously been using on 30-metres.  VK6 (the west coast of Australia) is the antipodal point from Nova Scotia.  The distance is approximately 11,000 miles (17,703 kilometres!) 
    I found the forward gain and front to back to be extremely good.  There is no easy way to compare it to Mosley's published figures, but I am certain it is as good or better than they say.  It will take a few weeks to get a good feel for this, as well as the side rejection, but my initial impressions are that it works very well.

[Main Page[Amateur Radio]

Last updated on Sunday, 23 October 2011