[TowerTalk] Guywires
jimlux
jimlux at earthlink.net
Sat Jan 16 13:33:58 PST 2010
Kevin Normoyle wrote:
> Is this analysis wrong?
> http://wiki.contesting.com/index.php/Guyed_tower_study
>
> If so, where?
>
> It's a 45g analysis. I would think that means 25g sees more deflection
> at the top.
>
> I didn't understand people talking about 1" deflections.
>
> What are the typical worse case deflections 100' 25g towers see in bad
> conditions? (at the top say, given whatever guy conditions you like?)
>
> I guess the analysis above doesn't address dynamic oscillations or
> torsion issues.
>
> Assuming you don't have enough damping, it would seem flexing is good
> (so stress isn't transmitted to the base?)
>
Some flexing is good. What the typical analysis does is look at the
forces on the vertical tubes and diagonal braces (the braces are longer
and skinnier, so even though they're not loaded as heavily in absolute
terms, they may be closer to their buckling point). For the most part,
these are thin columns loaded in compression, so buckling is one of the
concerns.
When you start looking at multiple guy points, as well as nonuniform
wind loads, the calculation gets complex.
There's a fair amount of margin in the manufacturer's recommendations to
cover variations in installation, variations in material properties and
manufacturing (weld size, for instance), as well as aging and usage
effects (that ding in the side of the tube has a big effect on buckle
resistance, as anyone who has stood on an aluminum beer can knows).
That margin is why anecdote isn't a good guide; that is, "I put up 130
feet unguyed, and sure it waves a bit in the wind, but I just hold on to
a leather belt wrapped around a cross brace with my teeth to hang on
while I beat the rotor into submission" stories are totally believeable,
but not a good basis for design.
And, hey, what's acceptable for a ham with 40 acres (if it didn't fall
down last winter, it wasn't big enough) isn't necessarily the case for a
public safety agency putting up a VHF repeater antenna at HQ.
>
More information about the TowerTalk
mailing list