Crush Tube Analysis with 6 Different Cases and Comparison

Objective:

• To mesh the Bumper model provided and to simulate the Crush Tube with different types of Interface cases.
• Following are the 6 cases according to the which the interfaces had to be set up:

1. Run the crash tube model as it is.
2. Change the Inacti=6 and run.
3. Create the type 11 contact and run.
4. Remove both notches and remove boundary conditions on a rigid body node then run.
5. Create a new notch in the middle, select the whole section, and run.
6. Create a new notch with nodes only from opposing 2 faces and run.

Introduction:

• Interfaces/Contacts are defined to model how parts interact when they come into contact with each other.
• Penalty method used in interfaces: This method treats the behavior between slave nodes and master segments as springs that generate resistive forces as a function of penetration.
• Some of the most used interface types are:

1. Type 7: General node to surface
2. Type 11: Edge to edge
3. Type 19: Type 7+Type 11
4. Type 27: General node and surface to surface
5. Type 2: Tied node to surface

• Along with this, there are different parameters like I_gap, Gap_min, Inacti, I_stf, I_form, St_min, I_del, etc. that are to be set in each of the interfaces depending on the type of the interface being used.

Procedure:

• The Bumper model was meshed first using the ‘auto mesh’ option. The ‘mesh type’ was set as ‘quads’ and the ‘mesh size’ was set as 5mm. After toggling a few nodes, the mesh was finalized as below.
  
  

Case 1: Run the crash tube model as it is.

• The file ‘crush_tube_0000.rad’ was imported into the HyperMesh using the ‘import solver deck’ option and was simulated as it was.
• Here the contact interfaces used are Type 2 and Type 7. The Type 2 contact is used in the welds, while Type 7 is a general interface used to define contact between node and surface. In this type, the penalty method is used. Depending on the resistive forces generated, the position of the slave node is determined concerning the master node
• The simulation was done using the ‘Radioss’ option from the ‘Analysis’ tab. The destination for saving the output and animation files was specified and with ‘-nt 4’ in the ‘options’, the operation was performed.
• Later, 0001.out file was opened to check the cycles, energy error, and mass errors to determine whether they are in the limit and if doing the further analysis is feasible or not.
• The following results were found in 0001.out file
  Cycles: 83600
  Energy error: -3.8%
  Mass error: 0.000
  
  

• Since the values were well within the limit of 5%, it was good to go for further analysis. For viewing the animation, ‘HyperView’ was opened and the ‘.h3d’ animation file was imported. It was found that the notches provided in the tube caused deformations a little quicker at a lower load.
• From the ‘contours’ option, the plot for Von-Misses stress was selected. The maximum amount of Von-Misses stress was found to be 6.968e^-1 KN/mm². The following animation shows the Von-Misses stress plot.
  
  
• Similarly, the plot for Plastic Strain was selected. The Maximum Plastic strain in the model reached up to 1.204.
  
  
  
  
• Further, to plot the different types of graphs of rigid wall forces, and energies, the HyperGraph2D was opened and the ‘T01’ template file was loaded. The screen was split into 2 parts, and first, the rigid wall forces were plotted. The maximum values of rigid wall Normal forces and Tangential forces, obtained were 1350KN and 4KN respectively. The images show both the ‘Rwall’ forces.
  
  
  
  
• Similarly, the internal energy, kinetic energy, and contact energy was plotted. The highest values of internal energy and contact energy were approximately 42500J and 2000J respectively.
  
  

Case 2: Change the Inacti=6 and run.

• In this case, keeping all the other parameters constant, the ‘Inacti’ parameter in the Type 7 contact was changed to 6 in HyperCrash. ‘Inacti’ is used when there is some initial penetration in the system. In Inacti=6, the gap is variable with time. If there is some initial penetration, corrective actions are taken and the nodes are de-penetrated.
  
  
• After making the changes, the RADIOSS files were exported and opened in Hypermesh for simulation. Like the previous case, radioss was performed. The 0001.out file was opened in the notepad and the final values of energy error and mass error were checked along with the cycles. Those values were also the same as the Case1. Therefore later in the HyperView, the .h3d animation file was opened.
  
  
• The simulation, in this case, was exactly similar to the previous case i.e. case1, since no geometrical changes were made to the model, and the properties were also kept as they were. Though the Inacti parameter was changed, it didn’t have any change in the model as there was no initial penetration to cope up for. The following animation file shows Von-Misses stress. The maximum amount was equal to the previous case i.e. 6.968e^-1KN/mm².
  
  
  
  
• Similarly, the plastic strain was loaded, and the maximum value of the plastic strain was found to be 1.204.
  
  
• After this, the template file was loaded in the HyperGraph2D and the graph for R-wall forces and different types of energies were plotted as seen below. These graphs also were similar to the initial case.
  
  
  
  

Case3: Create the Type 11 contact and run.

• For this case, along with the other contact interfaces, new contact of Type 11 (/INTER/TYPE11) had to be created. Type 11 is an edge to edge type contact. In this type, the interaction between the edges of the master and slave is computed to determine the deformation. This type of contact is used to overcome the disadvantages of Type 7 of reduced time steps.
• This contact was created in the HyperCrash. The ‘contact interfaces’ tab was opened and in that the Type 11 contact was created with the recommended properties as seen below.
  
  
• It was then exported and opened in HyperMesh. Radioss operation was performed on it. The errors and cycles were checked in the engine output file to determine whether the values are acceptable or not. The following values of cycle, energy error, and mass error were observed in 0001.out file:
  Cycles: 99600
  Energy error: -3.8%
  Mass error: 0.000
  
  

• The energy error was acceptable and it was feasible to proceed for further analysis. The .h3d animation file was opened in the HyperView. The deformation was slightly different than both the previous cases. The von-misses plot was selected and the maximum value of the stress reached up to 6.659e^-1KN/mm², which is slightly lower than Case1 and Case2.
  
  
• The maximum amount of Plastic Strain was 1.195 which was also slightly lower than the previous scenario.
  
  
• Later, the rigid wall forces were plotted in the HyperGraph2D.While the Normal resultant force of this case was similar to Case1, the Tangential force had a higher value of 9.5KN as opposed to that of 5.4KN in Case 1 and 2. Also, the spike in the TF was a bit late than case 1.
  
  

• While plotting the internal energy, kinetic energy, contact energy, and total energy, it was found that all the energy values were almost the same as the previous case except contact energy which had a value slightly lower than 2000J.
  
  

Case 4: Remove both notches and remove boundary conditions on a rigid body node then run.

• All the notches in the model had to be removed in this case. This was done in the HyperMesh with the help of the ‘align nodes’ option. This option was selected from the ‘node edit’ panel (F7). All the notches were removed from the body by aligning the nodes in the notch with the nearby nodes as seen below.
  
  
• Along with that, in the HyperCrash, the ‘boundary condition’ collector was also deleted to remove the boundary conditions on the rigid node. After this, similar operations were carried out in the HyperMesh. The radioss was performed and the cycles and the errors were checked in the 0001.out. The following were the values observed at the end of the simulation.
  Cycles: 93400
  Energy error: -2.4%
  Mass error: 0.000
  
  
• As the values were well within the limit, the animation was thus viewed in HyperView. Since there were no notches present in the crush tube, the buckling started at the contact area of the rigid wall and the tube. Whereas, in the previous cases, the initial buckling or kinking used to happen at the notch area.
• The Von-Misses stress generated in the body was 6.360e^-01KN/mm². Along with this, the Plastic Strain was a bit higher than the previous cases due to the removal of notches. It was found to be 1.295. The following gifs show the von-misses stress and plastic strain.
  
  
  
  
  
  
• Later, the rigid wall forces were plotted in the HyperGraph2D. The Normal resultant forces had a lower value of 1150KN, while the Tangential forces were high at 12.5KN.
  
  
• Similarly, the internal energy, kinetic energy, contact energy, and the total energy were plotted. The following image shows the values and the graph of the different energies.
  
  

Case 5: Create a new notch in the middle, select the whole section, and run.

• As per the condition, a notch was created in the middle of the crush tube model from the previous case, while all the other parameters were kept constant as they were as seen in the image below.
  
  
• The radioss operation was performed on the model, and the engine output file was viewed for the values of cycles, energy error, and mass error. The values observed in the file were:
  Cycles: 94800
  Energy error: -2.8%
  Mass error: 0.000
  
  
  
  
• Since all the values were in the limit of 5%, HyperView was opened and the animation file .h3d was imported. Now in this case, since there was a notch created in the middle, the buckling started at the notch unlike the previous case, where it happened at the contact area.
• The Von-misses stress was plotted and the maximum value of the stress was recorded as 6.381e^-1KN/mm², while the plastic strain was at a lower value of 1.077 due to the creation of the notch.
  
  
  
  
• Later the rigid wall forces and the internal, kinetic, contact, and total energies were plotted in the HyperGraph2D as seen below. The spike in the Tangential Resultant force graph was a little earlier at approximately 19^th ms due to the notch at the middle, where the tube buckled.
  
  
  
  

Case 6: Create a new notch with nodes only from opposing 2 faces and run.

• In this case, two notches were created, each at the opposite face of the crush tube. This was done in HyperMesh with the help of the ‘translate’ command, while all the other parameters were kept unchanged.
  
  
  
  
• Radioss was performed and the cycles and the errors were checked. The following values were obtained from the 0001.out file:
  Cycles: 93100
  Energy error: -0.8%
  Mass error: 0.000
  
  
• The animation was plotted in the HyperView by opening the .h3d file. The buckling happened at the notches created on the opposite faces. The Von-misses stress was plotted and the maximum value was noted as 6.599e^-1KN/mm², and on the other hand, the Plastic Strain reached up to 1.209.
  
  
  
  
  
  
• Later the graphs for R-wall forces and the energies were plotted in HyperGraph2D. The Tangential Force reached an all-time high of 95KN. The following images show the R-wall forces and energies.
  
  
  
  
  
  

Results and Conclusion:

• Interfaces define the type of contact between the master and slave node during a simulation of a component under a load. These parameters play an important role in determining the behavior and the way the model is going to deform under the applied load, with a very close representation of real-life scenarios.
• In this assignment, different types of contacts were used to analyze the simulations of a crash tube, which is a vital component in the front portion of a vehicle in situations of a crash. Under a constant load, the behavior was observed by varying the geometry and changing the contact types.
• The following results were obtained in all the 6 variations of the contacts and geometry.

  	Parameters	Cycles	Energy Error	Von-Misses stress (GPa)	Plastic Strain	R-wall NF Forces (KN)	R-wall TF Forces (KN)	Internal Energy (J)	Contact Energy (J)
  Case 1	No change	83600	-3.8%	0.6968	1.204	1350	5.4	42500	2000
  Case 2	Inacti=6	83600	-3.8%	0.6968	1.204	1350	5.4	42500	2000
  Case 3	Type 11 contact	99600	-3.8%	0.6659	1.195	1350	9.5	42500	1950
  Case 4	No notch and boundary condition	93400	-2.4%	0.6360	1.295	1150	12.5	44000	1250
  Case 5	Middle notch	94800	-2.8%	0.6381	1.077	1350	8	44000	1600
  Case 6	Opposite notches	93100	-0.8%	0.6599	1.209	1250	95	42500	1750

  
  *All the above values are approximated
  
  
• Though the ‘Inacti’ parameter was changed to 6 in case 2, the results were exactly similar to those of case 1 since there was no alteration in the geometry or other significant parameters of the model. The ‘Inacti’ parameter is only considered when there is an initial penetration of slave nodes into the master. But in this case of crush tube, there was no penetration. Therefore, ‘Inacti’ was not considered.
• In the 3^rd case, Type 11 contact was used. Using the Type 11 contact along with the Type 7 contact, the problem of locking of nodes is solved. Locking condition occurs when the slave nodes fully penetrate the master segment which further leads to computational errors. Since in this type of contact, the impact occurs between a master and slave line, a slave line can impact on one or more master lines. There were some changes in the graph of the R-wall force of this case. Using the Type 11 contact, the Tangential Resultant Force had risen to 9.5KN. But the disadvantage of this type is, it requires more computational time as it uses complex mathematical algorithms.
• Later, for Case 4, all the notches were removed. Therefore, the deformation started at the contact area between the rigid wall and the tube. Also, due to the absence of notches, the plastic strain rose to 1.295, and the R-wall TF was also higher than the previous case and reached up to 12.5KN.
• In case 5, a notch that was added in the middle of the tube helped to initiate the deformation at a relatively earlier stage. This also helped to reduce the plastic strain, and increase the stress absorption. The TF also was relatively lesser than before.
• Two notches were created on the opposite faces of the tube in the 6^th case, which reduced the energy error, and helped, in better absorption of stress. This way, the energy which would have traveled until the passenger compartment would get reduced. The tangential resultant forces generated in this case reached a whooping 95KN!

Additional GDrive link for files of all cases: Assignment 5 files