Now consider the case when the wind is from the single leg side of the
tower and towards the opposite face. The single windward leg will be in
tension and the two opposite legs will be in compression. Unlike the first
example with the wind applied on the face, the two leeward legs of the tower
resist the downward force from the wind with twice the counteracting force
of a single leg. To get this tower's two legs to buckle will require twice
the wind force on the opposite leg vs the single leg example above.
Actually, this isn't necessarily the case, because straight tension
failure is very different than straight compression failure is very
different from buckling failure.
The tower leg that in tension will fail because the material strength is
not enough: it's all about cross sectional area * failure load per unit
area.
The tower leg that is in compression can fail either by straight
material failure or by buckling. Buckling is more about the
length/diameter ratio than the material properties, and a lot about
symmetry and lack of dents.
We can see a practical demonstration of this:
I would think that it is very difficult for a person to pull an aluminum
can apart (without twisting and tearing it first), but it is easy to
crush an aluminum can (column failure by buckling), long before it fails
by simple material failure.
It is true that an Engineer may design a lattice tower so that the
individual members receive loads that will buckle at about the same
overall structural loading that they would fail by straight
compression/tension. Efficient designs tend to have all failure modes
occur at the same time/load, and excessive strength in one mode often
means excessive stiffness in some way, which can lead to premature
failure. (the classic example is a 1000 lb breaking strength rope and
1000 lb breaking strength piano wire in parallel, not holding 2000 lbs)
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|