Ron Stone, Vine Antennas wrote:
Comments interspersed.
> Hi chaps.
>
> I recently built two five eles for 28 MHz from the ARRL antenna book
> design. This is 24 foot boom. Element center pieces are 1/2" for the
> first three feet on each side. Then we taper to 5/8" to the tip. This is
> the "light duty design", but follows my own usage of 2" square boom and
> flat plates on top of the boom for securing the eles. This keeps the
> eles in line much more readily than round boom, and makes boom to mast
> plate mounting very accurate and direct.
Unless I've made a mistake reading the above, the element has a reverse
taper with the small diameter tube connected to the boom and the large one
out near the tip. If this is correct, it is contrary to common practice for
designing cantilevered beams for stress.
The commonly accepted approach is to use the smaller sections farthest from
the boom and the larger ones near the boom.
For 10 meter elements this may be insignificant as it is really hard to
build a 10 meter element that cannot take 100 mph.
In the case where one has built an element that will experience high
amplitude oscillations in the wind, it may be exacerbating the original
design problem. The larger section, with larger mass is out at the
unsupported end of the element causing any sinusoidal motion to be
amplified. This amplified motion will result in higher stress levels during
the peaks of each cycle and result in faster work hardening of the material.
When it gets work hardened it looses all of its ductility and becomes very
brittle. Then the slightest movements cause very high stress levels and the
section fails.
I would recommend that the element taper is reversed to have the largest
tube connected to the boom.
> All was well for about four weeks, and then about one foot of the outer
> 5/8" tubing of D2 of the upper antenna of the stack vibrated off. It
> looked almost like it had been cut with a tube cutter. I noted when I
> climbed the tower that other eles were vibrating even in light wind. It
> looked like the element tip was vibrating at many c/s back to a point
> about a foot from the tip, where it appeared to be stationary. Then the
> vibrations started up again and reached another maximum about 18" in
> from that point. the flexion was about one inch max travel. Bad news.
> The element was singing with the effort and no wonder D2 didn't last
> long.
>
> I have put rope in the tubing at the tips. As much nylon rope as I can
> stuff in. The rope fills about 5/8 of the inner diameter of the tubing,
> and is pushed back to the 1/2" dia section where pop rivets get in the
> way of any further pushing. Then I have put plastic end-caps on the tip.
>
> Chaps. Have I done it right, and what else should I be doing to overcome
> this little problem please ? So far it looks OK, but it has been fairly
> windy, and this problem looked at it's worst in light winds and
> particularly on REF, D2 and D3.
>
> Comments based on real experience will cut more ice than theorising
> about it thanks !
Well, we'll see what we can do here! There are lots of theorems, not
theories, that describe the behavior of the universe we live in. Lots of
people get to learn a new term today.
There is an aerodynamic phenomenon that is called "Vortex Shedding." It has
to do with how wind flows over a shape and what happens on the downwind side
of non-aerodynamic or relatively blunt shapes. On the downwind side the
laminar air flow separates from the surface of the section creating a low
pressure area. It does not form the separations uniformly and this results
in unbalanced loads on the section that start the oscillations. Once the
oscillations start, the non-uniformity of the separations increases adding
to the loads that feed the oscillation.
For every shape, size, air density, and wind speed, there is a certain range
of windspeed that will start and support the process. When the windspeed
goes above or below the thresholds for the section and conditions the
problem goes away.
That's the laymans explanation, from memory, I'm not an expert on this
subject. I just went and found reference material, read it and tried to
figure out how to defeat it. Maybe there are some aerodynamicists on
frequency to offer a better description.
Ok, this phenomenon is real, I would guess that we have hundreds on this
reflector that have climbed a tower and seen it! Hundreds more that have
found aluminum in the xyl's garden.
I've been up and seen a very light breeze not cause the oscillations, then a
stiffer one make it occur, then higher wind speeds came and the oscillations
stopped. Just like the guys in the textbooks said! BTW, this phenomenon has
been studied for decades, it's not a new revelation.
So, what to do about it?
That's next. First some more discussion.
As with everything electrical, all mechanical things also have a resonant
frequency. If we design an antenna with the right dimensions to have it be
mechanically resonant for the "Vortex Shedding" phenomena, we will get to
see it shake itself apart.
The tough problem is figuring out analytically how not to do this.
This is in the too hard box for most of us (including me).
So what do we do?
Here is a suggestion, based on the information I've been able to gather.
Option #1:
Put some smaller sized rope inside the elements, as you have tried. The
comments of others about not packing it in there tightly are appropriate.
The rope dampener only works when it has some room to flap about. It needs
to be of sufficient mass to be able to offset the energy of the oscillating
tube. When the aluminum tube oscillates one direction, the rope inside goes
the other way and smacks the opposite wall of the tube to slow down its
movement.
Rope stuffers don't stop the phenomenon, it always occurs on its own merit,
they just limit the amplitude of the oscillations, hence increasing the work
hardening life of the structure.
In other words, rope dampeners just prolong the life of improperly designed
elements.
Option #2:
This is pretty simple. If we make the entire element with one diameter and
wall, we get a structure that will ALWAYS go absolutely nuts into
oscillation at some windspeed.
If we make it with two different sections, we have two shorter mechanical
resonators that will want to vibrate, due to "Vortex Shedding," at different
windspeeds.
When we use 3,4,&5 different sections we get a whole collection of
mechanical resonators that all want to oscillate at significantly different
windspeeds. Then, there are no longer enough of them to act together, at
overlapping windspeed thresholds to make the element go nuts.
The mechanical solution to the problem is to use more sections with more
taper.
Don't let the number of connections in the element scare you off. If you can
make one good reliable connection, you ought to be able to do it several
times.
My experience tells me that any two section element that is made from slip
fit tube, not swaged tube, will shake itself apart, eventually.
You've got to use either swaged tubes with large changes in diameter, or
make more diameter changes via more sections with the slip fit sizes.
73, Kurt
--
YagiStress - The Ultimate Software for Yagi Mechanical Design
Visit http://www.freeyellow.com/members3/yagistress/
Or contact NI6W@contesting.com
--
FAQ on WWW: http://www.contesting.com/towertalkfaq.html
Submissions: towertalk@contesting.com
Administrative requests: towertalk-REQUEST@contesting.com
Problems: owner-towertalk@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|