[TowerTalk] FW: Guy article in CQ

Thu Jan 14 21:28:20 PST 2010

I'm sure there many on here who know more about this than I do and could 
likely phrase it better.
Guy wires serve one purpose. That is to keep the tower vertical, 
although you could within the same  vein say they also prevent 
oscillations and need to be spaced such as to prevent said oscillations.

To operate properly the wire angles should be such that "all in any 
particular tier" have the same vertical and horizontal vector tension.  
With EHS wire which is *heavy* length is also important. Even with 
Phillystran the length does not want to be long enough that the natural 
resonance of the guy is any where near a resonant point in the tower. As 
the tower is guyed at several heights with variable spacing/distance 
between the tiers the tower will have several resonant points. These 
points will not only depend on the natural resonances in the tower using 
the guy points and base as pivot points for the bending moments. However 
the ability of any of these points to move due to both the elasticity 
and catenary of the guys will add (or subtract) from or to these natural 
resonant frequencies of the tower sections.

Based on the mechanics involved I see little problem of using longer 
guys when using Phillystran, at least to a point. IOW I see no problem 
of using 45 degrees as the guy angle for each level. Of course this 
means a guy anchor on the ground for each guy which could be a bit 
inconvenient.

Usually the top guy in any installation is on the order of 60 degrees. 
Changing that to 45 degrees should reduce the stress on the tower as the 
translation from horizontal wind load to pul on the guy is considerable 
as well as to the load on the tower in the vertical axis. That would 
make the center tier much less of an angle and as long as there isn't 
too much mass in the catenary to create a resonant problem it should 
work, but in this case the weight of the guy line translates to a 
lateral (sideways) pull on the tower. Hence a good idea to use the guy 
bracket system that puts the tension on the mount and not the tower as 
the tower more or less sets inside the bracket. With a 3 tier 
arrangement this makes for a very low angle on the bottom tier.  Again 
if using a guy bracket this puts little stress on the tower, but guy 
resonances should be a concern if of EHS.

There is a lot of lattitude available when it comes to the distance from 
the tower to the guy anchor base, BUT depending on the mass of the 
guy/material I'd not go far without doing a stress analysis on the 
tower.  Why I favor the Phillystran is the light mass puts the resonance 
for even relatively long runs well above any resonances in the tower 
itself of the kind that are likely to do damage. Longer runs of EHS OTOH 
are massive and likely to develop resonances that can be destructive by 
themselves. I've seen power and telephone lines oscillate to the point 
of snapping off the power poles.

Just East of Alma MI on the North side of Lincon road  is a long run of 
telephone and cable.  Quite a few years back (just shortly after being 
installed) we had some strong winds that caused the lines to oscillate. 
I saw them moving over 10 feet between the poles. Later that day they 
actually busted off a number of the poles.  When you go through there 
now there are some triangular fittings or wings attached and the 
resonance has changed enough the cables move very little although it's 
still noticeable.I would expect the same phenomena with a tower guying 
system. 

So, although I don't see any problem going longer with guy lines within 
reason,  as soon as you get outside the manufacturers specs it's time to 
do, or have an analysis done on the system which is not at all simple. 
On aircraft with smaller parts we can do what is called a vibrational 
analysis and come up with a dynamic model that will give an upper speed 
limit for a particular design as configured. It also lets us rebalance 
control surfaces to eliminate resonances which show up as flutter. (a 
control surface vibration, or oscillation with positive feedback - not a 
good thing)

Dick Green WC1M wrote:
> TowerTalkians --
>
> In followup to my post about computing guy lengths over sloping terrain,
> below is an exchange with W2RU that clarifies my approach and adds important
> information. Bud's question is at the bottom, my response at the top.
>
> 73, Dick WC1M
>
> Hi Bud,
>
> I agree that the tension must be the same in all three directions, but not
> only in magnitude. It must also be the same for the horizontal and vertical
> vector components of the force along each guy wire. I don't believe that,
> for a given angle, simply duplicating the tension along the guy can
> reproduce the relative proportions of vertical and horizontal force of a guy
> that's at a different angle. 
>
> I think you can visualize this by picturing a tower with two guys at, say,
> 45 degrees to the tower, and a third guy at 15 degrees to the tower. That
> third guy can easily be set to the same tension as the other two, but it's
> going to have a much greater proportion of vertical (pull down) force than
> horizontal (pull out) force. I suspect there's no value of tension that
> could be used to exactly duplicated the vertical and horizontal tension
> vectors of the other two guys. That proportion is a function of the angle of
> the guy to the tower.
>
> Regardless of the physics involved, one of my objectives was to follow the
> Rohn-specified layout for the tower as closely as possible, on the theory
> that the manufacturer knows best. Rohn specifies guy attachment heights and
> the distance from the tower base for the anchors, which results in a certain
> set of guy angles and horizontal/vertical vectors. I wanted to get as close
> as I could to that and still allow clearance for the middle SteppIR. I had
> to push the anchors out about 5% to accommodate the middle SteppIR, but
> that's within the Rohn-specified tolerance for the distance. Then the task
> was to reproduce the guy angles and forces over my sloping terrain.
>
> Another way to picture it is to draw a horizontal line from the tower base
> to each guy as computed for my terrain. I've done that in the attached
> tower/guy diagrams. The line intersects the downhill guy well before it
> reaches the ground. The line goes past the uphill guy before it intersects
> the line of the guy if it were extended at the same angle into the ground.
> The line hits the guy on level ground exactly at the anchor. The length of
> all three lines will be 92', which is what I computed to be necessary for
> middle SteppIR clearance. The horizontal line drawn from the tower base to
> each guy is therefore the flat ground that's missing. The picture
> essentially shows that the tower has the same geometry as it would over flat
> ground.
>
> Now, if you look closely at the guy diagrams, you'll see that what I just
> said isn't true for the Northwest guy. With the extreme slope of my terrain
> in the downhill direction, it's not possible to exactly reproduce the angles
> for all three guy wires when attached to the same anchor position. You can
> see that when the guy anchor is placed to properly reproduce the desired
> angle for the top guy, the other two guys intersect the flat plane at
> distances other than 92 feet. At 85 feet, the middle guy isn't too far off,
> but the bottom guy is off by about 22 feet, which is a lot. When the guy
> attachment points are moved to be the same as the top guy, the angles of the
> other two guys must increase. If you look at the guy angles for all three
> directions, you can see that the middle and bottom guys are not at the same
> angle in all three directions. The middle guy is close to nominal all three
> directions, but the bottom guy angles are quite different.
>
> The correct way to deal with this is to use multiple guy anchors, either
> three separate anchors or at least two: one for the top and middle guys and
> another for the bottom guy. But I wasn't willing to go through the
> considerable trouble and expense to do that. My feeling was that the most
> important angle to get right was the top guy angle, and that if the other
> two guy angles had to be different, it was at least better for the
> horizontal component to be greater than spec than for the vertical component
> to be greater than spec (in the horizontal direction, the other guys anchors
> resist the force; in the vertical direction, the tower and base resist the
> force.)
>
> That's my story, and I'm stickin' to it!
>
> 73, Dick WC1M
>  
>
>   
>> -----Original Message-----
>> From: W2RU - Bud Hippisley [mailto:W2RU at frontiernet.net]
>> Sent: Thursday, January 14, 2010 9:22 AM
>> To: wc1m73 at gmail.com
>> Subject: Re: Guy article in CQ
>>
>> Dick --
>>
>> I would love to have a copy of all your drawings, calculations, and 
>> spreadsheets.  We're in the process of locating a new house and one of 
>> the two finalists has small rolling hills that will necessitate guy 
>> wire anchors being at varying heights relative to the tower base and to
>>     
> each other.
>   
>> You said one thing I'm not sure I understand the need for:
>>
>> On Jan 13, 2010, at 7:07 PM, Dick Green WC1M wrote:
>>
>>     
>>> The idea was to ensure that the angle of the guys to the tower, when 
>>> placed on my hill, would be identical to the angle of the guys over flat
>>>       
> ground.
>   
>> Visualizing a tower with one set of guys (for simplicity), for the 
>> tower to be in equilibrium and vertical the horizontal tension in the 
>> three guys must be equal in all azimuthal directions.  If one guy wire 
>> has a different angle relative to the tower or vertical axis, it seems 
>> to me it just means the tension along that guy wire must be different than
>>     
> the tension in the others.
>   
>> However, if your objective was to ensure clearance of all guy wires by 
>> rotating beams, then I guess I understand why you said that.
>>
>> Or did I miss yet another constraint?
>>
>> Thanks!!!!
>>
>> Bud, W2RU
>>     
>
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>
>   