|
|
|
|
|
We have changed the way the program incorporates forces from shearlines on upper levels in the rigid diaphragm analysis on the level below.
In versions before Shearwalls 11.2:
For seismic design, and the non-torsional wind cases (ASCE Envelope Procedure form Figure 28.3-1 and ASCE Directional Procedure, Case 1 from Figure 27.3-8) Shearwalls included the unfactored shearline forces derived from flexible diaphragm analysis on the level immediately above and added them to the applied loads on the level for which rigid analysis was being performed. The shearline force was then factored by the 0.6 load combination factor.
For ASCE Directional Case 2 loads, Shearwalls included the applied loads from all levels above and added them to the applied loads on the level for which rigid analysis was being performed. The shearline force was then factored by 0.6 for wind design
For both wind and seismic design, the shearline forces arising from rigid diaphragm analysis from the level immediately above are applied to the rigid diaphragm analysis on the floor below. For each shearline, the worst case of forces from positive and negative accidental eccentricity on the level above are those which are transferred.
When the rigid diaphragm procedure was implemented along with multi-story analysis in version 2002 of the program, there was concern that the torsional effect of the accidental eccentricities applied on the level above should not be carried through to the level below for two reasons:
It was thought that accidental eccentricity is entirely to account for inaccuracies in the distribution of mass in a structure, and that it was unlikely that these inaccuracies or variabilities would occur on each level and in the same direction.
There was a concern that the eccentricity would unnecessarily compound, that is, the effect of eccentricity on the level above creates amplified forces that would be further amplified by the eccentricity on the level below.
We received guidance at that time that it would be acceptable to apply loads derived from flexible forces on the floor above to rigid analysis on the level below.
The Case 2 Directional torsional wind requirement was determined to apply to flexible diaphragms and added to the program for version 10, in 2013. In order to avoid compounding this eccentricity, applied loads on the levels above were used in lieu of shearline forces. This was an internal change only, there is no effect on the resulting torsional moments.
ASCE 7 C12.8.4.2 says that it is "typically conservative to assume that the center of mass offsets of all floors and roof occur simultaneously and in the same direction." That is, as we cannot know for certain which levels and directions the accidental eccentricities occur in, it is best practice to assume they occur on all levels.
Typically, accidental eccentricity adds approximately 10% force to the shearline experiencing the greatest effect. Compounding this effect on the floor below by adding 10% of 10% = 1% is not a significant increase and is much less than the inaccuracy that arises from using the flexible diaphragm forces from the shearlines above. Flexible diaphragm analysis can result in a markedly different distribution of force than the rigid diaphragm force, and by using flexible forces, the program was not transferring the inherent torsion from the floor above to the one below.
The new procedure uses the direct forces due to shear wall stiffness on the level above, rather than those arising from flexible analysis. As we are assuming rigid diaphragms on all levels, this is a significant improvement in the accuracy of the procedure.
The new procedure transmits inherent torsional forces due to physical offset between the center of mass and center of rigidity, and these forces contribute to torsion on the level below. This is the correct procedure, as an object when subject to torque twists throughout its whole length. The old procedure did not transmit torsion between levels.
On each line, the program transmits the maximum force from positive and negative accidental eccentricity. As such, the new procedure will transmit the increase in force due to accidental torsion but will not transmit the torsional effect of that force, as the worst case of the positive and negative accidental torsions on opposite extremities of the structure tend to cancel.
We believe this is an acceptable compromise between the conservative procedure described in ASCE 7 C12.8.4.2, and the non-conservative procedure of not including accidental eccentricities from the floor above on the floor below at all.
For All heights wind load generation procedure, when a wind shearline force was added directly to a shearline using the Add as a factored force directly… input in the Add Load input form, for rigid diaphragm analysis the force had an effect on only the level it was added on and was not being transferred below for inclusion in the torsional analysis on the level below.
This has been corrected in the course of implementing the changes in Bug 2733, above, Vertical Distribution of Rigid Diaphragm Shearline Forces.
In the case of buildings with flat roofs and on the upper level of the structure, the generation and distribution of low-rise wind loads using ASCE Fig 28.3-1 were incorrect in several ways. In such buildings, the longest direction of the building is considered to be the direction parallel to the ridge line for buildings with roofs. The following problems were identified and corrected
For generation of wind loads, when a block with roof overhangs was adjoined to another block, the program would create loads on the roof the block was adjoined to in the portion shadowed by the block, even if the Exclude roof portion covered by other block checkbox in the Load Generation input was checked. This happened when using the load generation procedures for both low buildings and all other buildings.
The following problems pertaining to wall out-of-plane seismic force Fp from ASCE 7 12.11.1, which is shown in Elevation view, have been corrected.
The program was using only the weight W of ½ the wall height in the calculation Fp= max( 0.4 SDS Ie* W, 0.1W) when it should have been using the full wall height.
Fp is now shown for both rigid and flexible diaphragm analysis. Earlier, it was shown only for flexible analysis.