Examination of various methods to apply acceleration excitation to a dynamic analysis.

Christos Sakkoulas_21687 · July 2023

Introduction:

In this article, we review three different modeling practices for performing an FRF analysis by enforcing acceleration to the structure. The methodologies are simulating the phenomenon of enforced acceleration with the use of the following three cards:

SPCD
ACCEL1
GRAV

Frequency Response Analysis is used to investigate how the structure responds to being shaken at different excitation frequencies. Vibrating the part at the mode frequency is likely to excite the mode, particularly if shaking occurs in the direction in which the mode moves. By performing this type of analysis, it is possible to envisage issues early on.

Excitation frequencies can come from various sources:

Engine, transmission, propeller, wind, waves etc.

Frequency Response Analysis is used to compute the response of the structure.

OptiStruct contains a range of frequency response solvers. There are two types of Frequency Response Analysis (FRA):

Modal Frequency Response Analysis
Direct Frequency Response Analysis

Use case of each card:

SPCD: This entry can be used to define enforced displacement, velocity, or acceleration value for analysis.
ACCEL1: This card defines acceleration loads at specific GRID points in the model.
GRAV: This card defines the gravity vectors for use in determining gravity loading for the structural model.

These three cards were used to define the EXCITEID field (Amplitude "A") of dynamic loads in RLOAD2 Bulk Data Entry for the dynamic solution sequence.

Methodology independent set up:

The modeling process is the usual of the frequency response analysis and useful information can be found by clicking on the links 4-6 at the end of the article.

The component that was chosen for this article is a 127x50.8x5 mm plate and it is the same for all the methodologies. In the following picture the plate is presented.

Figure 1 The component of the model

The material and the property values are presented in the following pictures.

Figure 2 Material and the property values

The needed FREQi and EIGRL cards the input values are:

Figure 3 FREQi and EIGRL cards

Figure 4 TB inputted curve

Methodology dependent set up:

The next part of this article presents the specific inputs and the differences between the three modelling procedures.

Model with SPCD card

In this methodology, the boundary conditions were defined with the SPC card and in the two following pictures the inputs are presented:

Figure 5 SPC cards

It is necessary for the correct creation of the SPCD card to ensure the existence of SPC at the application nodes and at the same DOF. So, the SPCD was applied to the same nodes as the SPCs and in the figure 6 the inputted values are presented. One crucial deviation of the three methodologies is the type of RLOAD2. In the RLOAD2 card the loading curve and the entity that is being excited are defined. In the case of the model that uses the SPCD card the type of RLOAD2 has to be ACC due to the fact that the SPCD enforces acceleration to the structure.

Figure 6 SPCD card Figure 7 RLOAD2 card

2. Model with ACCEL1 card

The simulation process for this methodology is pretty much similar to the aforementioned one. The SPC cards are defined in the same way. During the creation of the ACCEL1 card, it is necessary for the correct creation, all the nodes of the structure to be selected. In the following picture the ACCEL1 card is presented:

Figure 8 ACCEL1 card

Regarding the RLOAD2 card the type has to be set to LOAD since the ACCEL1 card is a load applied to the structure:

Figure 9 RLOAD2 card

3.Model with GRAV card

On the methodology with the applied gravity to the structure, during the creation of the GRAV card, it isn’t necessary to select the nodes since the gravity is assigned by default to all the nodes which have mass (density assigned). It is important to mention that the boundary conditions of the models are identical to each other. So, the GRAV and RLOAD2 cards inputs are:

Figure 10 GRAV and RLOAD2 cards

In this methodology the type is set to LOAD since this is also a load applied to the structure.

Results

In the following picture the results of the models are presented:

A picture containing screenshot, colorfulness, text, diagramDescription automatically generated

Figure 11 Displacement

A picture containing screenshot, colorfulness, text, diagramDescription automatically generated

Figure 12 Velocity

A picture containing screenshot, text, colorfulness, diagramDescription automatically generated

Figure 13 Acceleration

For the case of the SPCD in order to see the same results with the model of the ACCEL1 and GRAV, it is necessary for a PARAM card to be created with the ENFMOTN option enabled. The displacements/velocities/accelerations output during Modal Frequency Response Analysis and Modal Transient Response Analysis are relative to the results due to enforced motion specified in the model. Lastly, the direction of the enforced acceleration should opposite from the other two cases.

Conclusion

As is clear from the pictures, the results are identical on all three cases. The choice between each of the cards should be made in relevance to the excitation origin. If the vibrations come from the boundary conditions, then the SPCD methodology should be chosen. Otherwise, the other two methodologies are more suited.