Thursday, October 12, 2017

Bad Vibrations

When we think about fasteners for antennas perhaps the most common questions are about stainless steel and yield strength. There is more to it than that. This was brought home to me recently when I went up my Trylon 70' tower to remove a wire antenna, to do a general inspection and to plan coax routing improvements for the Hy-Gain TH7 and the Cushcraft A50-6.

That's when I saw the following on the TH7. Can you see what's wrong?


The splice bolt holding the boom sections together has lost its nut and lock washer and had worked itself halfway to freedom. The one on the other side of the boom was not quite as bad since the nut had loosened but not yet fallen off. Left uncorrected the boom would soon be held together only by the tension of the boom truss. I put other projects on hold since this called for an emergency repair.

The bolts were out of reach and the antenna could not be easily rotated due to its weight and the 6 meter yagi stacked above. A closer visual inspection uncovered other problems. The antenna had to be lowered to the ground for a full maintenance overhaul despite having been up for only 4 months.

Obviously I had made mistakes. I was in a rush at the time and took a few shortcuts. Nothing serious, or so I thought. Or at least nothing that couldn't be remedied before winter, which was my plan and why I did this inspection climb.

I had several days leeway since I had to wait for helpers to resume work on the big tower. Once the weather was cooperative I rigged the tower for lowering the antenna with my lawn tractor. It came down surprisingly well considering I did it myself. I was able to handle the tag line while the tractor rolled forward.

Once on the ground I gave the antenna a thorough inspection. The number of problems was depressing. Just because I write a blog about towers and antennas does not mean I do every job perfectly. Far from it. I make mistakes and I take shortcuts like any ham.

First about those boom splice bolts. There was an interior start pattern lock washer on the one bolt that had not yet entirely unscrewed itself. Presumably the other had the same. Lock washers on round surfaces, especially small diameter tubes (2" in this case) don't bite well. I redid the bolts using the pattern for the Cushcraft XM240: lock washer under the bolt head and nyloc nut on the other end.


Before undertaking repairs I went to my favourite fastener store and stocked up on ¼" stainless lock washers and nyloc nuts. I had an interesting conversation with the clerk helping me out. He pointed out the labels they stuck on the stainless steel fastener bins: "beware of galling!" They'd gotten weary of dealing with irate customers who would try to return stainless fasteners that were over-tightened (excessive torque) and not lubricated. Careless handling can ruin a stainless steel fastener in an instant. I lubed the boom splice bolts before firmly applying torque.

Lubrication of fasteners

Lubricating the threads of a bolt (or nut) before use is often good practice. Aside from prevention of galling of stainless steel fasteners there are other benefits. One is that non-stainless, non-galvanized fasteners succumb to rust far slower when coated with grease, and grit has more difficulty being driven into the threads by weather. Later removal of the fastener is easier. Common grade 5 bolts and nuts fall into this category.

The second benefit is setting the correct torque. When nuts are tightened on a bolt some of the applied torque is due to friction in the threads rather than axial force of the thread surfaces. This results in increased risk of insufficient torque leading to future loosening due to vibration or thermal cycling. A thin film of grease reduces this risk by eliminating friction, without increasing risk of the parts "sliding" apart.

There are unique lubes available for every imaginable fastener alloy. Those are the best though not always necessary. A thin coating of a non-optimum lube is better than none at all. Most often I use white lithium grease. Over many years I've had good success with it. It's cheap and widely available.

Lock washers vs. double nutting

Writing about lock washers on round tubing reminded me of a conversation I had with a tower pro. He said that they are increasingly skipping lock washers entirely since they have had too many instances of lock washers failing in service. That is, they break in two. Whether this is due to the rigours of weather and stress cycling or washer quality wasn't made clear to me.

The alternative they and many others often use is to start with a flat washer rather than a lock washer between the work surface and the nut. After applying the required torque they then put on a second nut. The first is held with a wrench as the second nut is tightened. The claim is that the locking action is excellent and due to the nut being thicker than a lock washer there is less chance of fatigue failure or loosening.

I cannot vouch for whether this is indeed superior. It is a technique I've used in other applications with good effect but never on towers and antennas.

Singing

Most hams have experienced or are aware that yagi tubing elements are prone to high frequency oscillation when the wind blows. The sound may be audible and thence comes the name of the phenomenon.

Damping can be used to reduce singing, and eventual element failure from metal fatigue. Old Hy-Gain mono-band yagis are especially notorious in this regard. Rope inserted in the elements is a common method of mechanical damping. Newer Hy-Gain element tips are roped. This may be inadequate.

15 meter reflector (second element from the end of the boom) singing in the wind

I mention this since during my inspection of the TH7 I discovered that the 15 meter (mono-band) reflector was singing, even though the tips are roped. After lowering the antenna I reconfirmed that the ropes were in place. None of the other elements was singing at the time or during a later climb. I find that the trapped elements are more immune from singing, likely due to the tube diameter transitions from small to large to medium around each trap.

I decided to leave the damping alone for now. The singing was intermittent rather than continuous so it may be that the tip ropes alone keep metal fatigue under control. I will check again next year.

More problems

There were additional mechanical problems with the TH7 that I had to address. Several of the boom-to-element clamps were inadequately tightened, and some bolts lacked a lock washer. They had not loosened: I didn't tighten them enough. A few of the elements had rotated on the boom a few degrees. This is not enough to be a problem, but it would surely become one in future.

I carefully tightened all the clamps and realigned the elements. Or perhaps not so careful since I managed to miss one of the driven elements. The driven elements are close to the tower so this oversight was easily remedied.

When the TH7 first went up in June I noticed that the SWR minimums were at higher frequencies than expected (per the assembly instructions) on 15 and 10 meters. I rechecked all measurements. I discovered a few small (< 1") errors and that the first 10 meter director was 3" off its correct position on the boom. By placing the antenna 6" higher on the mast I was better able to route the rotation loops of the this and the 6 meter antenna above it so that they ran orthogonal to the boom and driven element phasing harness rather than parallel.

The SWR problem did not go away. I tested it with an analyzer at the antenna. The ancient RG213 feed line show a small amount of age-related irregularity and could affect the measurement. I now suspect the balun since the pattern of the TH7 seems to be fine; that is, it is a feed point problem. I will test that hypothesis later by attaching the analyzer directly to the feed point.

While the TH7 was off the tower I measured the SWR of the A50-6. It improved a small amount. This is evidence of unwanted mutual impedance that could be affecting 6 meter performance. This is not unexpected for yagis with 20'+ boom separated only 7' on the mast. I plan to push the A50-6 higher up the mast to reduce coupling. But not right now.

The TH7 boom truss mast support that I cobbled together from spare parts was unstable and ready for replacement. I had done this because not all the original parts for the truss came with the antenna (I bought it secondhand). I constructed a better support bracket and installed it when the antenna went back on the tower.

Hopefully no more distractions

Trouble on the Trylon was not accounted for in my fall antenna and tower planning. Luckily the cost in time was not excessive. What it did was pretty well push me off the air for close to two weeks, and missed one contest I had intended to operate: the California QSO Party.

Now it's back to work on the big tower. Progress has gone in bursts depending on availability of friends for ground crew. I am gradually getting closer to having antennas nice and high.

No comments:

Post a Comment

All comments are moderated, and should appear within one day of submission.