Torsional analysis of the chassis and its validation through Finite. Element Analysis

Transcription

1 Torsional analysis of the chassis and its validation through Finite Ayush Anand Student(Production) BIT Mesra,Ranchi, Jharkhand ,India Element Analysis Keywords: Roll cage, Torsional analysis, HyperMesh Abstract The most important aspect of the vehicle design is the frame part or roll cage. The frame has the operator, engine, brake system, fuel system, and steering system, it must be of adequate strength to protect the driver in the event of a rollover or impact. BAJA SAE roll cage is generally constructed of steel tubing, with minimum dimensional and strength requirements dictated by Society of Automotive Engineers (SAE). The torsional stiffness of a vehicle s chassis has a significant effect on its handling characteristics and is, therefore, an important parameter to measure. In the present paper, the torsional stiffness of BAJA roll cage is experimentally computed which was then, validated by the Finite Element Analysis (FEA) using HyperWorks Introduction The biggest effect of torsional deflection has been found on lateral load transfer distribution between the front and rear axle. During a steady state turn, an infinitely rigid chassis will cause the both front and rear roll angle to be same, which is assumed when suspension design calculations are performed [4]. Allowing for a twist in the chassis will redistribute some amount of weight transfer between the front and rear tires, causing these values to deviate from their designed values [5]. Most racecars design frame to stiffer than suspension roll stiffness. To find out the torsional stiffness of the frame experimentally, a fixture has to be made to constrain the roll cage from all the directions and then, a load has to be applied from one direction and the deflection would be measured using a dial indicator. A similar frame had been modeled in CATIA which was then trimmed and mid-surface is generated in CATIA itself. The model had been processed in hyper mesh and similar load had been applied. The measured deflection had been compared with the experimental value for its validation. Process Methodology A. Torsional analysis using FEA (Finite Element Analysis) Material and property Fig 1. Assigning components and properties 1

2 As it could be observed from fig 1, all the parts like chassis, A-arms, solid springs were assigned different component and properties as per their thickness and material using different cards available in the hyper mesh. This had to be done to get more accuracy in the result and get more realistic view. Constraint and load Fig 2. Load and constraint applied Two equal and opposite forces of 75 lbs were applied at the ends of the front suspension as shown in fig.2. The rear part was constrained in all the degree of freedom while the front part was constrained in all translational 3 degrees of freedom but was set free in rotational motion. The shocks were modeled as solid links. B. Torsional analysis using experimental setup Design and fabrication All the fixture were made and welded in the college premises with the suitable material and good thickness. All of the components fabricated for the fixture were made of mild steel. This choice was made because mild steel is cheaper than aluminum, does not require heat treatment after welding and is easy to machine. The stands were made 14 inches tall so that the chassis could clear the torsion fixture. The adapter plates were made large enough to accommodate the bolt patterns of the chassis. Big I-beam section was used to hold the G-clamps and the entire roll cage. Adapter mounting plate The adapter mounting plate is a very crucial part in the fixture and it should be validated using FEA that there is not any significance error in the plate. Fig.3. Meshing of plate Fig.4. Load applied Fig.5. Result The plates were thick enough to enable the wheel hub studs to pass through the other side. An FEA with a worst case scenario load of 2000N revealed that deflection would be less enough, less than the resolution of our dial gauges. All the procedure had been shown in fig. 3. 2

3 Procedure The rear hubs were constrained in all the degree of freedom by mounting it tightly to the adapter plate via the block to the I-beam section. The adapter plate had been welded firmly to the block.it had been shown in fig. 4. Fig.4. Rear mounted firmly Front mounting points were constrained in all degree of freedom except one rotational degree of freedom along X-axis. For that, a roller had been made in the front system. The front torsion stands attach the front hubs of the chassis to an I-beam. The axis of rotation of the beam is parallel to the longitudinal axis of the chassis. Weights are now placed on the end of the pivoting beam alternatively, causing it to pivot. This raises one stand and lowers the other by an equal amount while keeping the rear stands fixed. The moment or torque applied is the amount of weight multiplied by its distance from the center of rotation. It had been shown in fig.5. Fig.5. Front mounting has been given one degree of freedom The twist of the frame is found by measuring vertical displacement of the pivoting I-beam. In this experiment, a dial indicator was used since proof of concept and not accuracy was the goal. Four dial indicator had been used in this experiment near all the suspension mounting points. 3

4 Fig.6. Complete setup Result In this section, the procedures used to calculate torsional stiffness using the twisting fixture for the chassis are discussed. Measurements from a fully assembled car are used to illustrate the uncertainty in the design. Finite element analysis (FEA) result Fig.7. Post-processing result A vertical displacement of 0.51 mm was recorded as could be seen in fig. 7, which has been evaluated from post-processing in hyperview. The torsional stiffness value for this particular displacement would be calculated as follow - Θf = 2δ Lf (radian) (1) =2* =0.002 rad =0.114 Where Lf is the lateral distance between the front dial indicators. Force reactions at the left front and right front, denoted Rf and Rr, respectively, are given by putting weights. The torque is calculated from T= {(Rf + Rr)/2}*Ls (2) = {(75+75)/2}*0.77 =56.77 lb-ft. 4

5 Where Ls is the distance between the pivoted tube and the point where force had been imposed. The twist angle is adjusted by subtracting the deflection at the rear. The twist angle at the rear, θr, is calculated from vertical deflections measured near the rear spring perches. θr= (δf+δr)/lr (Radian) (3) =0(approx.) Where δf and δr are the right and left vertical deflections measured (by dial indicators) near the rear spring perches and Lr is the lateral distance between the rear dial indicators. The torsional stiffness at each increment is calculated by K=T/θ (4) =56.77/0.114 =498 lb-ft/deg θ= θf- θr. Experimental result Dial indicators are used to measure the equal and opposite applied vertical deflections δ, at the left and right front spring perches. The increment is chosen such that at least 5 data points are obtained. Table 1: Left side (weight added) Table 3: Right side (weight added) Table 2: Left side (weight removed) Table 4: Right side (weight removed) The deflection in the rear side was approx. zero and so we neglected the deflection in the rear part. Based on the above data values, a graph has been plotted between moment and angle to find out any discrepancy in between the left and right side or the weight added and weight removed plot. Also, an average plot has been made in the graph itself to find out the average torsional stiffness value. The torsional stiffness value while loading weight on left side comes out to be 423 lb-ft/deg as calculated from the graph in fig.8. 5

6 Fig. 8. Left side The torsional stiffness value while loading weight on right side comes out to be 433 lb-ft/deg. as calculated from the graph in fig 9. Fig.9. Right side So, the average torsional stiffness value is 428 Lb-ft/deg. Discussion and conclusion To prove the repeatability of the torsion fixture, weights were applied on either end (right, left) of the pivoting beam in two separate experiments. Now, error percentage = [498(from FEA) 428(experimental value)] 428 = % A small amount of fluctuation was observed. This can be attributed mainly to the clearance in the holes on the suspension mounts. The overall curve was linear which shows the torsion fixture is giving the value of the roll cage torsional stiffness. Other Inference: Roll stiffness of the suspension was 165 Lb-ft/deg in front and 270 Lb-ft/deg in the rear. The torsional stiffness value is quite good as compared to the roll stiffness of the chassis. Good torsional stiffness ensures that no extra bracing is required between LFS and SIM member. 6

7 Assumptions: All fasteners, tabs and welded joint on the roll cage were assumed to be rigid. The stands, solid shocks, fasteners, I-beams, and all other fixture components are assumed to be rigid and stationary. The 75-pound weights are assumed to be exact. Acknowledgement I sincerely thank the production department of BIT Mesra, Ranchi, India for funding this project work.the author would like to express their gratitude to Altair Engineering India Pvt. Ltd. for helping in carrying out all the work. References [1] Thompson L., Lampert K., Law H., Design of a Twist Fixture to Measure the Torsional Stiffness of a Winston Cup Chassis, SAE Paper [2] Claisse, A.; Featherston, C.A.; Holford, K.M, C.A.; Holt.; Manning, D.: Measuring the Torsional Stiffness of a Space Frame Chassis using 3D Motion Capture Techniques, Applied Mechanics, and Materials, 2006, pp , [3] Law, E.H.; Raju, S.; Sone, P.H.; Thompson, L.L.: The Effects of Chassis Flexibility on Roll Stiffness of a Winston Cup Race Car, Motorsports Engineering Conference Proceedings, [4] Hartog, J.P.D.: Advanced Strength of Materials, McGraw-Hill Book Co, New York, NY, [5] Brown, J.C., Robertson, A.J.; Serpento, S.T.: Motor Vehicle Structures, Butterworth-Heinemann, Oxford, [6] Crocombe, A.; Sampe, E.; Somiotti, A.: Chassis Torsional Stiffness: Analysis of the Influence on Vehicle Dynamics, SAE 2010 World Congress & Exhibition, [7] Nitin S. Gokhale.: Practical Finite Element Analysis, Altair,