[TowerTalk] Guyed + self supporting /2 ??

[Jim Lux]

On 10/16/14, 5:30 AM, john at kk9a.com wrote:
> I think that the concern (on towertalk) is that guying adds compression to
> the tower legs.  All of the self supporting towers that I have seen have
> stronger legs than equivalent guyed towers, at least towards the bottom.
> While I would also recommend following the manufacture's design and I am
> not an engineer, I do not see how adding guys especially toward the bottom
> of a tower can make it weaker.


A lot depends on how the tower was designed.  Let's say you have a self 
supporter and the load increases smoothly from the top to the bottom in 
arbitrary units 0-100.  The designer might choose to make the tower 
tapered, so that the failure load is always a fixed amount more than the 
expected load, say going from 10-110 units.

So, adding some guys might decrease the bending forces, but will 
increase the down forces.  More importantly, it changes the distribution 
of the loads.  The original designer anticipated a particular 
distribution of loads, and designed with that in mind.  So instead of a 
smooth distribution of strength that nicely matches the smooth 
distribution of load, you might have a load concentration that exceeds 
the design strength at some point.

A good example might be a thin, flexible mast that is designed to bend 
in the wind.  If you add some guys at the middle, the bottom part is now 
loaded more highly in compression, and the bending is concentrated at 
the guy attachment point.

Most designers put margin into their designs (which is why 50 year old 
unmaintained towers with too big antennas usually don't fall down), but 
the margin isn't the same everywhere.

I've been struck, when looking at the detailed analyses on some towers, 
at how clever the engineer was in distributing the loads and not over 
designing.  For instance, on a lattice tower like the familiar Rohn 25 
or 45, there's a load on the vertical tubes, on the horizontal tubes, 
and on the diagonal braces.  In some of the members, column buckle 
loading is a bigger factor, in some it straight compression or tension. 
In general, they all have about the same percentage design margin (e.g. 
the load on the diagonal members are low in most guyed towers, and 
they're also long and skinny, so buckling is an issue, but you can do 
that, even with a fairly small diameter member).

  And then, you have to consider the combination of forces (Is the tower 
bending, compression overall? is there a torsional load?).  Kurt 
Andress's (sp?) page has some of this analysis.

A standard problem in early engineering classes is analyzing a welded T 
(one member with another sticking out the side), and then analyzing it 
with triangular gussets added to "strengthen" it, which actually makes 
it fail sooner, because it changes the distribution of the stress.

The overall story is that without a fair amount of analysis (or 
practical experience with lots of different installations, not anecdote) 
it's pretty hard to predict what the effect of doing something unusual 
will be.