Wind load on trussed tower 2

[jmurph36]

There was a discussion in my office today about applying wind load to a trussed tower. The engineer at my new employer thinks that if wind is blowing against the windward face of a trussed tower that the wind load should also be applied in the same direction on the leeward side. I am of the opinion that the windward side would shield the leeward side and therefore it is only necessary to apply the load on the windward side. Does anyone have any helpful information on this topic? I just want to know if my thought process is correct.

 

[Shu Jiang PEng SE]

Use section 29.5 of ASCE 7 2010 calculate wind load. Apply the load to any member not being shielded in the leeward wind directions.

—————————————————————
J&J Structural Consulting Inc.
Structural design, analysis, inspection, drawing review and stamping, and connection design

 

[jmurph36]

The tower is made of identical cross section. So, if the wind is blowing perfectly normal to the face, the members on the back would be directly in line and should be shielded. So, again, my question is, would that constitute placing the wind load only on the windward face?

 

[ENGINEER92]

It is not a straight forward answer. Here is a diagram from the TIA-G tower code. The tower code will also get more detailed about shielding.

 

[jmurph36]

I am attaching a shot of a question that is answered in the ASCE wind load companion manual to ASCE-7. According to this, it seems that wind loads should only be applied to one side of a tower because the code accounts for both sides of the tower within the equations.

 

[rb1957]

so should 1/2 the load be applied to the windward side and 1/2 to the leeward ? (or 60:40, or 60:60)

applying all to the windward side is conservative (if it is the total load).

and all faces are windward (for some wing direction). You won't say that a West wind blows harder than an East wind … would you ?

another day in paradise, or is paradise one day closer ?

 

[Agent666]

If its an open tower, then you should be working out the drag on each individual structural section based on the section shape and your wind code and applying these loads.

Usually there is a shielding factor/multiplier for subsequent frames (i.e. the leeward side). But the same drag coefficient and wind pressure apply to members in both planes, just with shielding on members behind others due to the wind flow being disturbed by the windward side. The sides also have drag (i.e. any diagonals), there should be in your code a method for altering the resultant line load based on the angle that the wind flow hits the sloping member.

You should be applying the wind load where it is applied in reality (i.e. on members as a line load, at the intensity calculated based on wind pressure, shielding and drag coefficients). Simply working off frontal areas as it sounds you might be may or may not be conservative compared to working it out correctly.

 

[JoshPlumSE]

ASCE-7 doesn't do a very good job of addressing trussed towers. I'd be more inclined to start with TIA-G as a basis a trussed tower. TIA is really intended for towers supported cell or radio antenna or such. But, it's a well thought out code and it is based off of ASCE. Just with a lot more practical methods for those doing trussed towers.

 

[JAE]

Wind NEVER blows at a perfect 90 degrees so there is really very limited, or difficult to quantify how much shielding occurs.

I'd be inclined to do what Agent666 says or follow JoshPlum's advice with TIA-G.



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[jmurph36]

I agree. And my understanding is that the force coefficients given in figure 29.5-3 are higher to account for the effect of the leeward side so it is not necessary to apply the wind load to both faces.

 

[WinelandV]

I design trussed towers very regularly (bucket elevator and catwalk support, not transmission towers). Typically square in plan view, not uncommonly rectangular (aspect ratio never really more than 2 when it happens). The towers have struts every ten feet, so for wind analysis it gets broken into 10 sections, with a 5 foot section at the bottom and top. Using 29.5 (ASCE 7-10), I calculate the force at each strut level. qz varies with the elevation of each strut. Cf is defined in 29.5-3. There's even a modification for diagonal wind on a square tower. At any rate, after determining the force at each strut level, if I'm analyzing in an analysis program, I split the force into 1/4's and apply that to each node.

As far as your wind load companion image, it's not saying to NOT apply it to the rear face, just that you're calculating the total force by summing both faces (hence why I divide it by 4, as noted above). Unless you're dealing with a VERY short tower, the actual wind pressure on each member will not be a major contributor to the member sizing, as resisting the OT moment will control the design (typically).

Now then, you haven't actually stated what the purpose of your trussed tower is. Transmission lines? Industrial facility? Grain co-op?

 

[jmurph36]

Its a truss the supports a conveyor for moving aggregate at a rock quarry.

 

[jmurph36]

So when you say you split it into 4's, you mean that you take the total calculate design fore, divide it by 4, then apply 1/4 total calculated load to each member both front and back? So a line load of 1/4 of the total force would be what I need to apply?

 

[WARose]

IIRC, the windward side is what gets loaded up (as per the truss tower method).....and that accounts for the wind load for about a bay or so. If the structure keeps going....you'll apply that load at about every other column line.

 

[skeletron]

For detailed analysis:
Take into account the drag factor on each member.
Account for the wind acting at different angles to the faces too. As the wind rotates, the "face area" will be the resultant of the perpendicular and transverse wind components. Typically 15-30degrees will be the maximum case. There is also a shielding factor applied based on the solidity ratios.

If you had too you could probably make a really conservative estimate based on the solidity ratios and shielding, but I don't see this as a back-of-the-envelope type problem. The drag factor, non-orthogonal wind components, and the slenderness of the structure make it a little complex. I don't think the "pressure+suction" applies (like on a building) because these are generally open structures that are affected by breaking up the wind flow unlike walled building structures.

 

[WinelandV]

jmurph36,

I apply a force at each node at the corner of the tower at that elevation, no line loads. Again, unless your tower is super short and not supporting anything, the actual pressure on the member won't control the member design.

 

[Agent666]

With differing members and hence different drag forces on individual members meeting at nodes it is possibly easier from a perspective of accounting for all the loads to simply apply them to members. From a global perspective it will make little difference either way to the global response.

You should also either way account for the member moment due to the loads being on individual members, as often things like angles have pretty poor bending strength and you can use up a reasonable percentage of the capacity by the bending which needs to be accounted for under assessment of the combined actions checks required by any code.