Mass Study

Folder Path

GrabCAD\MSXIV sims\Spring 2021 Reinforcement Sims\MassStudy.wbpj

Setup

The goal is to determine how the mass distribution impacts the results. The CAD model being used for this study is 2021-01-17 Chassis and the collision scenario being used is the top 30 collision, since this is one of the worst collision scenarios for our chassis. The mesh will use 20 mm elements and tetrahedral elements for the gussets, in accordance with previous mesh studies. The max deformation and max combined stress will be recorded, as well as pictures of the simulations.

The cases that will be simulated are:

1. No Mass - The chassis will be simulated without additional masses, just the mass of the chassis.
   
   

2. Distributed Mass - The additional mass of the passengers, battery box, and dynamics assemblies (totalling 620 kg) will be distributed across all chassis components.
   
   

3. Point Masses - The masses will be separated into 9 points that represent objects in the car with significant masses (four 80 kg passengers, the 100 kg battery box, and four 50 kg dynamics assemblies). These points will be positioned in the vehicle and assigned to specific tubes that the objects will be mounted to. Passengers masses will be assigned to tubes with seatbelt mounts, the battery box will be assigned to tubes with mounting holes for the catamaran cover, and the dynamics assemblies will be assigned to tubes with mounting holes in the A and C planes.

   1. Another point mass case will be simulated where the dynamics assembly masses are distributed across the A and C bulkheads and the passenger masses are only attached to the 2 seat mounting tubes, rather than the 4 seatbelt mounting tubes. This is the way that the masses are currently being assigned.
      
      

4. L-Shaped Seats - The passenger point masses will be replaced by L-shaped volumes to replicate the seat geometry. These seats are touching all beams with seatbelt mounting points, but also some other beams, gussets, and the B-bulkheads. The material being used for the simulation is aluminum, and the volume has been set to be within 1% of 80 kg (each) based on aluminum’s density. Point masses will be used for the other masses, as described in case 3. Point Masses.

Open

Figure 1: Point mass locations. The rear dynamics assembly masses are hidden behind the C bulkhead.

Another test to confirm if the simulations are giving reasonable results is to simulate the gradual addition of mass and see how the stress responds. It is expected that the trend will be close to linear, if the trend is not linear that may be an indication that the simulation is not setup properly.

Results

The results from the 3 main cases can be seen in the following table:

Case	Max Combined Stress (MPa)	Max Deformation (mm)	Pictures

Case	Max Combined Stress (MPa)	Max Deformation (mm)	Pictures
No Mass	249.6	9.4	Open Figure 2: Stress plotted on deformed chassis with no additional mass (deformation scaled 29x).
Distributed Mass	2448.4	92.9
Point Masses (Seatbelts)	2781.5	91.5
Point Masses (Seat Tubes)	2995.8	98.8
L-Shaped Seats	3448.7	87.4

Gradually adding masses onto the chassis did result in a linear stress response (R2 = 0.96). This suggests that the simulation setup is working as expected.

Discussion

The no-mass case already reaches stresses of 250 MPa. Ideally, this is the stress we’d want to see from the fully loaded vehicle with a safety factor of 1.5-1.75, so it makes sense that we will need to add reinforcements to pass this collision scenario. Unfortunately, simulating the distributed mass case shows stresses over 2,400 MPa, large deformations, and multiple points on the chassis that would yield. This is not close to passing. Separating the masses into point masses increased the max stress but, considering that these results are well outside the linear, elastic region of Chromoly 4130, it isn’t too meaningful to compare them. The current method of assigning point masses (without using the seatbelt mounting points) has approximately 6% greater max stress, so in future simulations the seatbelt mounting points should be used for more realistic stresses.

Switching the passenger masses to L-shaped seats further increased stress, even though the deformation decreased. In the pictures, it can be seen that the seat locally decreases the stress and deformation in front of the B-bulkhead and increases the stress and deformation behind the B-bulkhead. There are also some high stress points at the front of the vehicle, even on the side of the vehicle that is not directly hit by the collision object. This difference can likely be attributed to the bottom of the seat increasing the rigidity of the chassis structure, which will depend on the specific seat properties being used.

Despite the linear stress response to gradual changes in mass, 8 out of 9 data points lie outside of the linear elastic region for Chromoly 4130. This means the simulations are behaving as expected, but this trend is not realistic. Implementing a multilinear plastic hardening model (MPHM) or a bilinear plastic hardening model (BPHM) would fundamentally change the relationship between stress and mass but would probably give more realistic results.