Torsional irregularity - with no diapghram

[dkrpink]

I am working on a canopy project that utilizes cantilevered columns for the sesimic resisting system. The canopy is simple structure - Steel column, steel beam, steel purlins, all supported on drilled peir foundation. There is no diaphram, all lateral loads are transferred through bending. The structure is about 20’ wide and 116’ feet long with 6 columns. Each column supports a “bay” area of 20’ by 20’. The columns are spaced at 20’ o/c along the 116’ length (purlins cantilever about 8’ on each end). In the width of the structure the column is located 4’ from one end and 16’ from the other. There is no frame so in section view there is only one column. The beam cantilevers on one side of the column 4’ and on the other 16’.

I have a plan checker that feels that the structure is irregular because it has a torsional irregularity per ASCE 7-05. He believes this because if you were to look at the story drift in the direction parrallel to the 116’ length the weak direction deflection at the end of the 4’ cantilever beam is significant less than that of the weak direction deflection at the end of the 16’ length of the beam. Basically because the weak direction deflection at the end of the 16’ length of the beam is more than 1.2 times the average of the deflection at the ends of the 4’ and 16’ beam length.

So back to my questions:

Can there be a torsional irregularity when a structure does not have a a diapghram?

Can there be a torsioanl irregularity when there is no rigid diapghram?

Can there be a torsioanl irrengularity when there is no rotation of the structure just lateral deflection?

Does a torsional irregularity exist per ASCE 7-05 in the above described structure?

 

[hokie66]

I agree with the plan checker that the structure is torsionally irregular. Whether you use a diaphragm or discrete bracing does not affect whether there is global torsion, and I don't know why you would say there is no rotation. If the lateral deflection varies, then that is rotation. ASCE 7-05...I don't know.

 

[ishvaaag]

In Fig. 10 Note 5

"Exception: One story buildings with h less than or equal to 30 ft (9.1m), buildings two stories or less framed with light frame construction, and buildings two stories or less designed with
flexible diaphragms need not be designed for the torsional load cases."

The note, however need not be understood, I think, a waiver of the general requirement of the prevention of torsional effects. These will appear at every column plus beam frame as soon you subject each separate of such frames to the dissymmetical load appearing on the beams frome their different projection for the longitudinal wind, that at the distances quite likely may not be understood shielded by the first frame.

 

[SAIL3]

As I understand it, torsion irregularity due to a rigid diapghram is to account for the possibilty of unintended/irregular shear distribution that can be delivered by the rigid diapghram. If there is no rigid diapghram present this delivery/magification of shear in certain portions of the structure can not occur.However, there could be torsion present in this case but it's just regular torsion and treated as such with no assumed magnification and is distributed according to the stiffeness of the various elements of the structure.
If there is no rotation then there should be no torsion present on the overall structure.

 

[DST148]

@ dkrpink : Ref: ASCE 7-05
In reference to the four questions you posted-----
#1) Yes. See Table 12.3-1. Torsional irregularity requirements apply to structures in which the diaphragms are rigid or semirigid. The word diaphragm is used in a generic sense. In your case the discrete members serve as diaphragm.
#2) Yes. It also applies to semirigid diaphragm. Table 12.3-1
#3) Theoretically No. (Torsional irregularity is calculated under torsional loads including accidental torsion therefore the displacements at the two ends will vary and hence result in rotation / torsion)
#4) Yes.

 

[MiketheEngineer]

Me thinks I would design it as cantilevered posts. If it can take that - then I am happy

 

[dhengr]

Dkrpink:
Without the benefit of having the latest Eds. of ASCE 7, I am inclined to agree with SAIL3 and technically I think I agree with DST148's answers to your question, but I’m not absolutely sure I fully understand your framing system without a sketch. I think you have a cross section which looks like a TEE above the piers: the stem of the TEE is your stl. cols.; the roof beams cantilever 4' on one side of the column and 16' on the other side, each of these TEEs is a unsymmetrical frame cantilevered up off a pier. You have 116' long purlins on top of the roof beams with some sort of a roof diaphragm and covering on the purlins. Any differences in beam tip lateral deflections must be taken care of in this roof framing/diaphragm/x-bracing system. Otherwise, your canopy should act as unit in the long direction. This difference in tip deflection is different than story drift, which happens at the tops of the columns, and relates to the whole structure. But, in fact the story drift could be different, parallel to the roof beams, at the two ends (116' apart) of the canopy, both in magnitude and direction; for example from a lateral wind gust and an uplift at one end of the canopy, in either direction perpendicular to the 116' length. And, to this extent you might have some irregular torsional action which you want to consider in your design, but the design affects should not be worse than your worst lateral load case, in either direction, as you look at typical lateral loadings.

As I understand it, the irregular torsion issue in ASCE 7 has more to do with irregular bldg. shape or framing systems, as you look down in plan view. And, that you expect the bldg. to act about one rotational center over its full height or from story to story (thus story drift is in play). But instead, because of irregular bldg. shape or shape changes, or because of irregular framing systems or changes (fl. to fl.) the bldg. reacts about a different center of rotation and induces unexpected torsional affects on the bldg. For example.... A nice rectangular bldg. with several interior lines of columns and a very consistent exterior wall system, with a centrally located elevator and mech. core, would show few irregular torsional issues. I do assume a consistent floor diaphragm at each level. In its simplest form we can sum the stiffnesses of the lateral load resisting elements, and find the center of rotation in plan, x and y, almost as we find the C.G. of a built-up steel beam. That same bldg. with a very stiff core off to one end, and the exterior walls of conc. blk. or tilt-up panels at that end, but then the other 2/3s of the bldg. with all glass walls, could have some serious torsional issues. That’s irregular torsion, as I understand ASCE 7 to define it in its broadest sense. Structural framing changes from fl. to fl. could cause some of these problems, and again it is assumed that the hroiz. diaphragms (floors, roof) can adequately distribute loads to all columns, frames, walls, etc. Structural mass changes can cause some of these same torsional problems when addressing the EQ issue.