RAM | STAAD | ADINA - RAM Frame - Dynamic Analysis FAQ

	Product(s):	RAM Structural System (Ram Elements or STAAD.pro)
	Version(s):	Any
	Area:	Analysis; Dynamic Analysis

How does a dynamic analysis using response spectra for seismic loads differ from a static analysis using equivalent lateral forces?

Dynamic load cases are not story forces in the same sense as the static load cases. When a dynamic load case with a code selected or user-generated response spectrum is analyzed, the program finds the Eigen solution and determines the modal response for each of the modes included in the analysis. The modal response is based on the response spectra assigned to the dynamic load case, which defines the acceleration vs. period spectrum curve. The modal responses are then combined in some fashion to determine the total response of the structure.

How can I change the number of modes used in the dynamic analysis?

Create an Eigen solution dynamic load case and enter the number of modes to consider. The number of modes can be changed at any point by changing this load case in RAM Frame (Loads > Load cases).

How many modes should I consider?

Generally, building codes require that the dynamic analysis include a sufficient number of modes to obtain a combined modal mass participation of 90% of the total building mass in each of the two orthogonal directions (see ASCE 7-05 12.9.1). The mass participation can be found in the Period & Modes Report under the section titled “Modal Effective Mass Factors” The “%Mass” values represent the mass participation for one particular mode; the “%SumM” values represent the cumulative mass participation. The X and Y modes with the highest mass participation represent the fundamental modes; the program will use the period and frequency associated with these modes in the calculation of the static seismic and wind loads when these values are chosen to be calculated by the program.

There is not a fixed number of modes that need to be specified in order to obtain the 90% mass participation. Rigid diaphragms have 3 degrees of freedom for each diaphragm (x-translation, y-translation, and z-rotation). If you need to include more modes than 3*number of diaphragms to obtain 90% mass participation, then you may need to increase the stiffness or look for instabilities in the model.

Note that when semirigid or pseudo- flexible diaphragms are used, there are 2 mass degrees of freedom for each node (x-translation and y-translation). As a result, default number of modes used by the program may be very large when no Eigen solution dynamic load case is created. In such cases, the number of modes can be changed as noted above.

When I look at the Period & Modes Report, I see that the cumulative mass participation for rotation is always 0% no matter how many modes are included in the analysis?

If a diaphragm includes a two-way deck or it is defined as a semirigid diaphragm, it is meshed and represented with finite (shell) elements. Each node of the finite elements has a nodal (point) mass associated with it. In other words, the diaphragm mass is represented with a network of spatially distributed nodal masses. Note each node includes point mass defined in the global X and Y-directions but not include rotational mass moment of inertia. Because the array of masses does not include rotation, the mass participation for this degree of freedom will always be 0. For the same reason, the mass participation for the rotational degree of freedom will also be 0 when pseudo-flexible diaphragms are used.

It should be known that this type of modeling still suffices for capturing all essential dynamic properties. In other words, the proposed solution accurately captures any dynamic actions related to rotational inertias or any twisting modes due to having center of rigidity and mass center at different locations.

What scale factor should I apply to the response spectrum?

The response spectrum analysis should be factored by the quantity Ie/R for both forces and drifts. Some codes require the dynamic forces (but not drifts) to be scaled so that the dynamic base shear is at least a specified percentage of the base shear calculated from the equivalent lateral force procedure (see ASCE 7-05 12.9.4, for example). The total base shear is reported in the Building Story Shear Report (Report > Building Story Shear). When the ground level is set to a level other than the base of the model, it is best to use the value reported for the level immediately above the specified ground level.

If you are not considering sign in the analysis, then the perpendicular dynamic base shear value is not really relevant to the X direction scale factor (it could be a result of torsion on the structure only), so we usually recommend scaling the Dynamic load based on the base shear parallel to the load angle.

Note that in version 17.01 we expanded the ASCE 7 Dynamic Response spectrum analysis to allow for different scale factors for each eccentric load case. See Release notes.

Should I use SRSS or CQC modal combination?

The CQC method is recommended as the SRSS method can give inaccurate results for 3D structures. When 2D structures are analyzed, the methods will produce similar results.

Does the dynamic analysis consider accidental torsion?

Most building codes require the inclusion of accidental torsion for both the equivalent static load procedure and dynamic analysis. The code requirements for evaluating accidental torsion for dynamic load cases are implemented in RAM Frame when rigid diaphragms are used. To include eccentricity, set the eccentricity to “+ and –" for the x and y directions in the Response Spectra dialog when the dynamic load case is created. The % eccentricity is defined in the Mass dialog in RAM Frame (Loads > Masses). Accidental torsion effects can be included in analysis for both rigid and semirigid diaphragms. For pseudo-flexible diaphragm, it is assumed that a flexible diaphragm is not able transmit diaphragm torsional moments, and hence accidental torsion effect is ignored for flexible diaphragms.

Why do the analysis results for all members in the model have positive values?

The modal response (nodal deflections, member forces, reactions) determined from a response spectra analysis have signs associated with them. However, once the analysis results are determined by combining the modal results using either the SRSS or CQC method, each response has only a positive value. Because the sign of the result is important for member design, there is an option in the RAM Frame (General > Criteria) to consider the sign in the results. When this option is selected, the sign of the analysis result will match the sign from the predominant modal result. You should not design continuous footings, gusset plate connections, or other elements based on the forces from multiple members using dynamic results unless the sign of the analysis result has been considered.

Why do I see a rift in the dynamic load deflected shape?

When the option to consider sign in dynamic analysis is considered, Ram Frame independently determines the signs for member forces, nodal displacements, diaphragm shears, etc. It uses the predominant mode to establish the sign for these various results. When the nodal displacements for two proximate nodes are determined to have opposing signs, the deflected shape will appear to have a rift in it. This tends to happen in structures with semi-rigid diaphragms or two-way decks where there are lots of close nodes to show displacements for.

Since nodal displacements are not directly used in any aspect of the design, and since dynamic results are inherently reversible, this has no consequence to the results obtained from the program. The absolute magnitude of the deflections are fine.

Turn off the option to consider sign in the Criteria - General to see an all-positive deflected shape.

Why don’t the values reported for the building story shear correspond to the values reported for the change in story shear?

The combination of modal results can affect interpretation of the results, especially building story shear and change in story shear. The program calculates a total story shear and a change in shear for each mode in the analysis. The results are each combined using either the SRSS or CQC combination. As a result, the reported change in shear is not linearly related to the reported building shear. For the same reason, the reported building story shears should not be expected to match the sum of the reported frame story shears. If one is trying to determine equivalent static forces to apply to the diaphragms to match the total base shear, the reported shear from one level should be manually subtracted from the shear reported in the adjacent level rather than using the change in shear values reported in the Building Story Shear Report.

How do I generate load combinations with dynamic loads cases?

Dynamic loads cases are included in the default load combination installed with the program. This was a feature enhancement in v14.05.03. See following web page for release notes: RAM SS v14.05.03 Release Notes

In previous versions of programs, dynamic load cases were not included in the load combination templates. In these versions, you can create load combinations with dynamic load cases using the custom load combinations dialog. You may find it helpful to first generate load combinations with the templates using equivalent lateral force seismic load cases (E) and then modify the load combinations to use the dynamic load case (Dyn) by changing E to Dyn. You can also copy and paste load combinations from another load combination dialog or text document into the custom load combination dialog or create a custom load combination template to include the dynamic load cases.

How can I account for 100/30 orthogonal load effects when using dynamic load cases?

Experts have recommended using a SRSS combination of two orthogonal response spectra analyses to determine the critical forces in a dynamic analysis (see reference below). This method is not implemented in the current version of RAM Structural System. You will need to use custom load combinations that include the dynamic loads in each direction with appropriate load factors to account for this code requirement.

“A Clarification of the Orthogonal Effects in a Three-Dimensional Seismic Analysis,” E.L. Wilson, I. Suharwardy, and A. Habibullah, EERI Earthquake Spectra, Vol. 11, No. 4, Nov. 1995.

I ran my model with a Response Spectrum analysis using a custom .rsp file, but the results are all zero, what's wrong?

The program interpolates acceleration values from the user provided table, but it does not extrapolate acceleration values past the last data point. It's critical that the last period value in the table is longer than the first mode shape to get accurate results.

For further details on custom response specta tables see the Ram Manager manual section titled, "RAM Frame Response Spectra Tables". For general help on using custom tables in Ram Structural System see RAM Table Editing.

RAM Frame - Dynamic Analysis FAQ