Power Line Crossarm Analysis

[Albert20]

Hi there,

As you can see from my handle I'm an electrical engineer by training. However I'm trying to understand a power line terminal crossarm and I'm wondering how to approach it.

I've attached a drawing for reference. The crossarm consists of two back to back angles connected at the conductor attachment points and cross-braced. The crossarm is attached to two wood poles via four steel channels. The channels are bolted to the pole at 90 degrees to the pulling action of the conductors. The crossarm is five metres long, the poles are 2.5m apart and the channels are 0.2m apart.

As its a terminal structure the conductors pull in the direction of the line. This pull is resisted by stay wires which are bolted to the top of each channel. The conductor tension is therefore taken as a strut load in the poles and a tension in the stay wires.

My question is this: when analysing the loads on the channels and crossarms can I assume the each pair of channels acts as one connection point and the crossarms can be analysed as a simple beam with two supports? Or, do I have to analyse it as a continuous beam with four supports?

Many thanks for any help, much appreciated.

Albert20.

 

[dhengr]

Albert20:

Is this design configuration a standard for your industry, company, or region? Or, is a design which is open to suggestions and changes?

It’s a simple fairly symmetrical beam structure as relates to its loading and reactions, but it is kind of a monster because of the way you have configured it, with all its various pieces, bolted connections, and unspecified conductor connections.

1. Conductor loads are self weight, wind loading, snow/ice loading, etc.
2. Where they are connected to the beam member, through an insulator of some sort, they will cause bending horizontally and vertically, and might cause a significant torsional loading to you trussed beam member.
3. The trussed beam member (2 angles with some bracing and channels) has a 2500mm center span, and two 1250mm cantilevers, one on each end. The reactions go to the two poles, approx. 1.5(single conductor loads) plus the vertical and lateral components from your four strut wires (we call these guy wires). All of these forces must be resolved into the reactions at the poles, and at lease consider that the three conductor loads may not all be identical.
4. Consider one of the cantilever ends: at its tip it has a vert., horiz., and likely a torsional loading; with the moments and stresses being max. at the reaction (pole). Angles may be easier to haul up the pole, but they are not a real good cross section for the loads we are applying to them in this case. Due to vert. load component, both angles have tension in the top fiber. Due to the horiz. component, one angle will be in tension and the other in compression, like the top and bottom chords of a truss. Then these disconnected angles must act in torsion, and they are just not very good for that.
5. The two bolts between the two channels and a post are probably overloaded. The two channels can reasonably be assumed to be one reaction, at one pole, if done right.

For a first timer, you give quite a good description and sketch, as a starting point. Think over the above, and answer my first to questions, and post again.

 

[paddingtongreen]

I spent a sad couple of years doing T & D work. Sometimes, I was told not to bother trying to analyze standard structures such as this dead end structure. They didn't even want to check the guy tensions.

What you is a quasi truss. The interior braces are usually not in alignment, causing eccentric connections, the bends in the angles don't coincide with them either. A real analysis would have to take these things into account. If you can find that this is a client standard with a successful history, then use that as the qualification.

I was looking for an overhead messenger for lightning protection, it must be done another way.

dhengr, the bolts will probably be okay, the poles have little moment capacity at the ground so the line pulls and the guy tensions tend to be equalized by deflection of the top.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[BAretired]

I wouldn't worry too much about continuity because the supports are not hinged. They are spring supports which depend on the length of the stays and the initial tightness to determine the spring constants.

There will be a tendency for the outer channels and stays to carry more load, so I would be inclined to design them to carry the full shear of one conductor. The inner channels would tend to carry only 0.5 times one conductor load, but you should just make it the same as the outer one.

To help alleviate the torsional issue mentioned by dhengr, you should use a stiff separator between the vertical legs of the two angles at Conductor #2 to prevent the angles from racking. The two angles will participate equally in carrying the horizontal force, but not the vertical force at Conductor #2.

The other issue which should be addressed is the possibility of one or more of the conductors breaking or carrying different loads. This changes the statics quite dramatically.


BA

 

[Albert20]

Thanks very much for your replies. The design is a standard one in my country. It was done approx 25 years ago for a certain maximum conductor size but the source calculations seem to have disappeared during industry privatisation etc! People are starting to talk about using larger conductors with a corresponding increase in weight and wind / ice loads. I'd like to have at least an understanding of how the structure is reacting to the conductor loads before I start dealing with this.

Electrically its an unearthed / ungrounded design so there are only the three phase conductors. There's no lightning protection / earth conductor.

After reading your replies I realised I'd missed one (fairly major!) component from my sketch. There is a 12mm thick x 100mm wide steel plate bolted across the horizontal legs of the angles just to the right of the conductor two position. This is for a post insulator to sit on. I imagine this would provide some assistance in preventing the angles from racking at this position.

Also, in order to calculate the loads in the cross braces I take it I'd need to analyse the structure as a truss?

Thanks again for your help.

 

[BAretired]

For horizontal loads, you would analyze the cross brace forces like a truss. If the cross brace makes an angle [θ] with the crossarm, the force in each cross brace is H/2sin[θ] where H is the horizontal component of Conductor #2.

For vertical loads in Conductors #1 and #3, the bending moment in each angle is VC/2 where V is the vertical component of the load and C is the cantilever length of each angle.

For vertical load in Conductor #2, the eccentricity of load will affect the result. If the eccentricity from the c.g. of the two angles is 'e', the vertical load on each angle will be V/2 +/- Ve/d where d is the distance between the shear center of angles. One angle will be loaded downward, the other upward.

BA

 

[MikeDB]

I would consider each pair of channels as one support, thus making it a simple beam with two supports. I would consider the angles as a simple beam instead of a truss for both vertical and horizontal loads, which should be a conservative assumption. If it does not meet expectations, then I might consider some truss action.

 

[dhengr]

With a 25 year old design, which probably assumed 2 guys (2 poles) with climbing belts, a parts bag and climbing spikes; that design probably had a lot more to do with smaller pieces hauled up the poles and fairly easily assembled in the air; than it did with any rigorous structural and stress analysis. The angles were an easy to get std. shape, which could be bent and punched (holes) with simple fab. equipment. Paddington suggests one approach, it’s worked for 25 years, that’s proof enough, so forget about it, leave it alone. BA gives you some good advice on load distributions and approx. forces and moments in some of the various members, which should get you started if you do decide to put some numbers on it. I would suggest that you get some books on strength of materials, and elementary structural analysis, and steel design. All of the pieces in your current design configuration are unsymmetrical in load or shape cross section, poor actors in torsion, and represent potential buckling problems. As I said above, the basic loads and reactions are fairly straight forward, but the individual members are each kinda messy at each analysis point in the whole structure.

I don’t mean to beat a dead horse too long, but your cross arm configuration is a fairly difficult little structure to do a simple structural analysis on. However, with the consideration of heavier conductors, and the possible longer spans, this might be the ideal time to at least take a look at a new design configuration, in good part because of the heavier loads. A light hollow structural section (HSS), rectangular shape would really make your structural design easier. And, it would eliminate the bracing pieces and their bolting. It would need light end caps, and drain holes; and have two steel plate tabs at each pole, .2m apart, to bolt to the backs of the two vert. channels above the pole top. These tabs would also have holes for your stay wire end hardware. With newer terminal hardware designs and insulators, etc. this might be much cleaner, and no heavier, a design.

If you change anything, you will not be able to fall back on the old design having withstood the test of time. And, in today’s market you probably will need a fairly complete design for the new standard.

 

[paddingtongreen]

You'd think 25 years ago was back in the dark ages, we had cranes and cherry pickers back then LOL.

I would agree with the HSS only if they galvanize inside as well as outside.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[Albert20]

Thanks very much for all your help. You've all shone a little light on the situation!