Tower leg reinforcement

[towerengr]

Attached are sketches from a proposed tower leg reinforcement scheme proposed by a consultant. The tower leg is overstressed between 108-114%. The new reinforcement piece is the same size as the original tower leg.

With the new angle, the composite section properties are computed and new l/r is computed. But the load is carried only by the existing member. The allowable compressive stress is increased by the addition of the new member which does not carry any load.

If you look at the sketches carefully, the existing and new leg angles are not connected to each other at fixed spacing. the new angle is not connected to the tower panel points to horizontal struts are diagonals.

Any comments on the acceptability of this type of reinforcement will be much appreciated.

Thanks in advance.

 

[transmissiontowers]

I would probably specify ASTM A-394-T1 high strength bolts in lieu of the A325 on the drawing. They have the same shear strength and smaller heads and nuts. The A394 are specifically for transmission towers where the load is transferred by bolt shear and not by clamping 2 friction surfaces together.

AFA, the built up section, it looks like it should work but check the L/r between the clips for each angle to make sure it is lower than the overall composite L/r. You don't want the individual angles to buckle.

_____________________________________
I have been called "A storehouse of worthless information" many times.

 

[towerengr]

Transmission Tower,

Thanks for your response.

You have to look at the details carefully to evaluate it.As the new angle is not connected at the panel points, how can we expect it to provide lateral bracing? Without a proper support system how can it function?. That is why I have reservations about this. I agree that the new angle concept has been used many times over but with continous angle connected at the panel points.

 

[transmissiontowers]

I looked it over a little more. In Detail 1-16 and 1-15, you would be better off if the added angles extended to more of the panels but I can see what they were trying to do. When the panel tries to buckle, the largest deflection will occur in the middle and rotate about the panel points. Since the original legs are continuous and not pinned individual members, you will get rotational restraint at the panel points (they hoped?)

Let me ask another question that someone should have asked already. What about the foundation? Traditionally the foundation is designed for the maximum leg load, so what happens when you increase the leg capacity? Are the foundations strong enough?

Another possible solution is not to strengthen the existing leg, but to shed some of the load. If the load case that is too large for the legs is caused by the wind loads and especially an oblique wind, you can guy the tower at several elevations along the line. We do this all the time when we add PCS antennas to the top of our towers.

If your max leg load is due to ice or just larger or more conductors, there is not much you can do to shed load.

_____________________________________
I have been called "A storehouse of worthless information" many times.

 

[VoyageofDiscovery]

To act compositely and utilize the radius of gyration from the new inertia, the section must be able to transfer interface shear. If the clip between the two sections is adequate for approx 2% of the compressive force in the original section to create bending in the composite section by lateral point load at clip locations, then it would be okay. So it needs to be checked for combined compression and bending.

 

[VoyageofDiscovery]

"So it needs to be checked for combined compression and bending."

should say

"The composite section then needs to be checked for combined compression and bending from eccentric axial compression."

The new section (not composite section) should be stiff enough for the 2% lateral force to provide as support to the original section.

This seem hokey as it would be better connecting at gussets.

 

[transmissiontowers]

Lattice transmission towers have been analyzed as truss systems for the last 100 years or so and the ASCE 10 manual compression equations were developed for truss action and the L/r is adjusted for end fixity and framing eccentricities based on many tests. The angles stitched onto the leg are there only to increase the radius of gyration so the L/r is decreased. If the individual L/r of the original leg between clip attachments is smaller than the combined section L/r, then the combined section L/r will control.

Now, if you decide not to go by ASCE 10 compression equations and consider bending and frame action, then you could use a general FE frame analysis program (we use GTStrudl for non-transmission tower frames) and check the tower per AISC 13, but you need to model all the redundants, extra leg bracing members and connection eccentricities to get a meaningful model.

_____________________________________
I have been called "A storehouse of worthless information" many times.

 

[towerengr]

Thanks to both of you for your input.

I think one thing is clear. It is very difficult to accept this design concept based on a quick review of the details. This approach must be justified on paper with computations and a list of assumptions.

I find it hard to believe that the two angles will act as a unit but will not share the load. If both angles act as single unit, for compatibility of deformations how can the new angle not take any load? I think it is very hard to swallow this explanation.

Voyage's explanation is inline with my thinking. I think composite action with increased moment of inertia can be assumed only when we have provided enough connection between the two angles using the flexural shear values.

A more simplistic approach seems to be to assume the new angle acts as a brace. It takes 2% of the compressive force in the main leg in flexure to prevent the main leg from buckling. The only condition we must satisfy with this approach is to connect the new angle at existing secondary brace and main panel points so that the 2% force can be resisted in truss action by the tower truss system. This assumption will require that the new angle be made longer than what is shown on the drawing.

Obviously using secondary members to increase the l/r will be much cheaper and is something we are used to.

 

[transmissiontowers]

I guess I think of the added angle as a parallel redundant brace. Up higher in the tower a new redundant was added to cut the unbraced length in half.

Think of the parallel member if it were spaced about 2 feet away from the main leg and connected at the panel points. The clips would go from the parallel member back to the old leg to cut down the unbraced length. This parallel member takes no axial leg load from the main leg but is just a brace to lower the L/r.

I'm not saying I agree with the approach. I am just trying to explain what I think they are trying to accomplish. I would insist on some vigorous documentation that this method has worked before. I might also suggest a full scale test be done to prove the concept.

Given that we design the towers for a 50 year return period storm, it is not likely that a load strong enough to test the concept will occur in the next few years.

_____________________________________
I have been called "A storehouse of worthless information" many times.

 

[towerengr]

Transmission tower,

I am not disputing your suggestions. It is just hard for me to accept this as adequate reinforcement for the tower leg, based on the material provided so far.

Thanks

 

[transmissiontowers]

If I were in your shoes, I would be skeptical too. I think I understand the theory they are trying to apply and am trying to explain it, and like I said before, I would probably stitch the leg from bend to base just to be safe, but Engineers by nature are conservative. Ultimately you will need to become comfortable with the solution (unless management is forcing the solution down your throat) and I would have them do a full scale test of the reinforced section.

_____________________________________
I have been called "A storehouse of worthless information" many times.

 

[towerengr]

transmissiontower,

I was able to locate a AISC paper on double angles in compression with test reports for various double angle configurations. Using this paper I was able to convince the Consultant on revising their design with continuous reinforcing angle with more intermediate connections. This design is very similar to reinforcement schemes used 50 years ago.

 

[transmissiontowers]

Good news. I'm glad you stuck to your guns. Ultimately you have to believe the solution will work. I think the previous fix "probably" would have worked but I would like to have seen a test to prove the concept.

_____________________________________
I have been called "A storehouse of worthless information" many times.