Week - 5 - Modelling Spotwelds

OBJECTIVE:

Spot welding is a type of resistance welding used to join metal sheets. The objective for this assignment is to model the spot weld for the given assembly of parts. This assembly is tested for a crash test similar to the previous crash box impact test (challenge 4). The spot welds should be modelled using beam and solid elements separately.

GEOMETRY:

The given geometry is shown in fig 1.1 below. The upper and lower metal sheets are meshed with shell elements. These two metals sheets have to be connected with spot welds using beam and solid elements separately.

Fig 1.1: Given part assembly

PROCEDURE:

The crash test on this model is done similarly to challenge 4 by creating the rigid wall away from the component by 10mm distance. The same velocity boundary condition is given to both the metal sheets (13.889 mm/ms). The spot weld modelling needs some extra card definitions which will be discussed below.

Basically the spot weld can be modelled in the following ways,

1. Using beam element and tied contact
2. Using constraint spotweld
3. Using constrained generalized
4. Using solid element and spotweld contact.

We will model spot weld for this assignment by the first and fourth method.

Spot weld by beam element formulation:

The spot weld of beam element is created by using the element generation option through element and mesh utility. It can be easily done by picking the desired nodes. For beam formulation, the section property is assigned by using *SECTION_BEAM card. It is important to assign element formulation (ELFORM) as 9 because *MAT_SPOTWELD material model only applies for ELFORM 9 formulation. The cross-section type is selected as tubular. Default cross-section (rectangular) gives zero cross-section area during the simulation. The beam thickness at both ends is given as 3mm. Shell section is also defined for metal sheets.

Fig 1.2: Spot welds modelled with beam elements

Fig 1.3: Section properties defined for beam

Then the material property is defined by *MAT_SPOTWELD card, the material properties are defined similarly to steel properties. The time-dependent failure criteria is assigned. The spot weld fails at 1.2ms. Other failure criteria such as effective plastic strain, resultant stress, axial stress are assigned very high values to consider only time-dependent failure criteria. Elastic material property has been defined to the upper and lower metal sheets.

Fig 1.4: *MAT_SPOTWELD card assignment

For spot weld defined by beam elements, desired contact type is tied shell edge to surface. Here, it is advisable to have all spot welds as a slave segment and spot weld material/part as a master segment. All the spot weld nodes are defined by nodes set ID and the spot weld parts are defined by part set ID. The metal sheets are defined with automatic single surface contact.

Fig 1.5: Contact definition for spot weld defined by beam element

Spot weld by solid element formulation:

The material card is not changed and used as it is to assign material properties to spot welds defined by solid element formulation. The spot weld of the solid element is created by using element generation option through element and mesh utility as discussed earlier. Section properties are defined by using *SECTION_SOLID card. Here, for solid section, it is important to assign element formulation (ELFORM) as 1 because *MAT_SPOTWELD material model only applies for ELFORM 1 formulation.

Fig 1.6: Spot welds modelled with solid elements

Fig 1.7: Section properties defined for solid

The desired contact type for spot welds defined by solid element is *CONTACT_SPOTWELD. The master and slave segments are modelled in a similar way as discussed above.

Fig 1.8: Contact definition for spot weld defined by the solid element

All the other settings are kept similar to the challenge 4 setup. In addition, SWFORC database card is added which gives the nodal constraint reaction forces for spot welds. Through energy plots for the system, it can be seen that there is a significant increase in hourglass energy. So, there are few methods to control the hourglass energy like changing element formulation, number of integration points, adding hourglass control etc. Here in this study, the *CONTROL_HOURGLASS card is used to control the hourglass energy. Figure 1.9 shows higher hourglass energy and is compared with the internal energy of the system. Whereas figure 1.10 shows the hourglass energy after using hourglass control card, this controls the hourglass energy and brings down into desirable limits (5-10 % of internal energy). Hourglass energy is basically work done to resist the hourglass modes and this extra energy is taken away from physical energy. So figure 1.9 and 1.10 shows the difference between internal energies with higher and lower hourglass energies respectively.

Fig 1.9: Hourglass energy before using *CONTROL_HOURGLASS card

Fig 1.10: Hourglass energy after using *CONTROL_HOURGLASS card

RESULTS AND DISCUSSION:

Various other output parameters are evaluated and are discussed in the following section.

Acceleration plot of spot weld nodes:

The nodes selected to extract the acceleration data are from one of the spot weld connection. From figure 1.11 and 1.12 it can be seen that after the impact the acceleration is not constant, at first the acceleration values decreases and then fluctuates. The lower metal sheet bounces back due to the impact whereas the upper metal sheet connected with spot weld continues to travel in the same direction due to momentum.

Fig 1.11: Nodal accelerations in the X direction for spot weld defined by beam element

Fig 1.12: Nodal accelerations in the X direction for spot weld defined by solid element

Force plots for spot weld:

SWFORC database card gives the nodal constraint reaction forces for spot welds. The axial and shear forces are compared below. The model has four spot welds and being symmetric, force values for only two spot welds at different location are compared in figure 1.13 and 1.14. Higher axial forces can be seen compared to shear forces for spot welds modelled by 1D beam element. The force values drop to zero after the spot weld fails. Whereas, higher shear forces are notices comparatively to axial forces for spot welds modelled by solid element. The force acting on the spot weld is in the transverse direction, so due to force resisting ability of solid element in the transverse direction, higher shear force values are notices for a solid element than beam element.

Fig 1.13 Axial and shear force on spot welds defined by beam element

Fig 1.14 Axial and shear force on spot welds defined by solid element

Energy plots of the system:

Fig 1.15: Energy plots for spot weld crash simulation defined by beam element

Fig 1.16: Energy plots for spot weld crash simulation defined by solid element

From figure 1.15 and 1.16 we can see the kinetic energy decreases just after the impact with the rigid wall. Simultaneously the internal energy increases due to the collision. The total energy overall remains constant as there is no energy loss or gain except small hourglass energy. It can be seen that the kinetic energy has a higher drop for spot welds modelled by solid elements. This signifies that the velocity of spot weld assembly modelled with solid elements reduces more than spot weld assembly modelled with beam elements after the impact.

CONCLUSION:

The spot welds have been modelled by using beam and solid elements separately.

The present spot weld model can be calibrated to actual spot weld by assigning known failure criteria such as plastic strain, axial force, resultant force etc from the experimental data.

The solid element gives a better representation of the actual spot weld connection.

The main difference between both modelling approaches is the contact definition. Contact tied shell edge to surface is used for beam element formulation and contact spotweld is used for solid element formulation.