At 09:45 AM 8/18/2005, Al Williams wrote:
>Steve's (k7lxc) prime directive not withstanding (do what the mfr says), I
>think that Garry at U.S. Tower might be correct in stating "...the
>winch...should not be a problem"; However the real concern is the
>tilt-raising fixture assembly strong enough to safely raise the tower.
>Remember that the safe load is usually several time less than the working
>strength load. I hope that some experts will
>correct and/or further explain the following:
>1. It is typical that metal structures bend/bow until they suddenly collapse
>with a little increased load. Does this bending also progressively weaken
There's two basic "styles" of failure here. The simpler one is simply
bending past the elastic limit. Grab a hunk of rebar, and carry it across
the yard. It will flex as you carry it, but it won't "take a set". Then,
stand on it, and bend the end up. You've now got a "failed" piece of
rebar, because it has permanently deformed. In this sort of case, the
deflection gradually increases as the load increases, and it's only when
the load is removed that you know it has failed. (Stretching a piece of
licorice or taffy until it fails might be similar, too..)
Then, there's "buckling". Empty an aluminum can of your favorite beverage.
Put it on the ground, balance on one foot standing on the can. It will
likely support your weight, but, with a small deflection, it will collapse.
(Interestingly, you can also do this with a single sheet of paper rolled in
a tube, but it takes a very steady hand and foot.) There's a sort of
"positive feedback" at work. As long as the forces are all balanced, the
structure is intact. If you unbalance a lot, it locally deforms
(elastically, as above), which then changes the distribution of the loads,
typically concentrating more of the load at one point, until the structure
fails. The load limit for most latticework towers is determined by the
buckling of the members (the tubes in the small sense, and the "cage"
formed by the lattice work of tubes in the larger sense).
There's also a "fatigue failure", where something that is repeatedly loaded
to less than the elastic limit fails after many cycles. Aluminum is
notorious for this, of course. (Look up the ill-fated DeHavilland Comet)
There's something else to consider and that is "brittle-ness". Some
materials are very stiff (high modulus) and don't deflect very much before
reaching their elastic limit (tungsten is a fine example, as are glass or
boron fibers). You can have something that is very, very stiff, and you
won't see any deflection before it fails, just because the deflection is so
small (it's still there, because nothing has infinite modulus)
>2. What does the winch rating of 1500lbs really mean? One mfr states that
>winches are often rated by their pull capabiliity. The pull capability is
>determined by how much pull can a typical man exert on the crank handle
>multiplied by the leverage of the winch handle and the gear ratio of the
>winch. This has nothing to do with the safe rating. Instead the safe
>rating of the winch concerns how strong is the raising ratcheting mechanism
>and the clutch mechanism of the brake part of the winch when lowering, and
>how do they deteriate with usage?
That would be a "carefully read the spec sheet" issue. There's also the
aspect of whether the rating is based on a full drum or only with one layer
wrapped on the drum. As more cable accumulates on the drum, the size of
the effective "pulley" increases.
One some of the small motorized winches I've seen (like those from
Grainger), the brake mechanism is a couple friction pads on a disk. The
combination of that friction, and the friction of backdriving the worm gear
is what would determine the max load holding capability.
Winches with a worm drive can be designed so that they cannot be backdriven
by the load (it's a matter of the pitch on the helical worm gear and the
friction coefficient of the steel gears).
>3. I read the response of the winches 1500lb rating with only a 900lb +
>200lb load as leaving a safety margin. However there are two leverages not
>being considered. The first leverage is a linear one caused by not lifting
>the load at the center of gravity. For simplicity assume that the tower,
>antenna, and mast weight are approximately evenly distributed. With the
>winch cable attached to the tower at approximately 7' from one end of the
>tower and approximately 24' from the other end (21 + 10 -7), the leverage is
>24/7. Of course the actual leverage is less because the weight is not evenly
>distributed but the leverage is still considerable.
You have to look at the geometry. In any case, lifting straight up, with
the bottom of the tower pinned to the ground (the fulcrum), the ratio would
be 31/7 (4.5:1), not 24/7, if the load were all at the top end. If the
load is distributed evenly, it would be (31/2)/7, or about 2:1.
>4. The second leverage is an angular one caused by lifting the load at an
>angle away from vertical. In this case the height of the raising fixture is
>about the same as the distance from the bottom of the tower that the cable
>is attached resulting in and angle of about 45 degrees. This means that the
>only .707 of the pull of the cable is applied to lifting the tower. Also
>.707 of the pull of the cable is applied to trying to pull the tower towards
>the concrete base. I need some help here. Does this mean that the leverage
>is 1/.707 or 1/(.707 + .707)?
The load on the winch is 1.414 (=1/.707) times the gravity load pulling
>Now multiplying the 1100lb tower and antenna load by the two leverages, say
>3:1 for the linear leverage and 1.414 for the angular leverage shows that
>the winch and raising fixture is being stressed to around 4000 pounds!
Yep.. that's a good ballpark figure... I'd figure more like 3000 pounds
(because the load is more like 2200 lbs than 3300)
> Again, will the real experts please correct and explain better my analysis?
And now you see why you need to ask the manufacturer why they think a 1500
pound rated winch is good enough. For instance, is there an extra pulley
pass in there, so the winch cable is doubled, in which case the winch load
is half the lifting load. The angle might not be 45 degrees (particularly
if the far end of the tower is supported off the ground. And, the load on
the cable will decrease as the tower comes up .
See: http://www.mscomputer.com for "Self Supporting Towers", "Wireless Weather
Stations", and lot's more. Call Toll Free, 1-800-333-9041 with any questions
and ask for Sherman, W2FLA.
TowerTalk mailing list