Diesel Engine Crankshaft High Cycle Fatigue Life Estimation and Improvement Through FEA

Download Full-Text PDF Cite this Publication

Text Only Version

Diesel Engine Crankshaft High Cycle Fatigue Life Estimation and Improvement Through FEA

Mr. Suraj K. Kolhe

    1. ech (CAD/CAM), Shri Guru Gobind Sighji Institute of Engineering and Technology,

      Nanded.

      Mr. Prafulla Ghare

      Sr. Manager R&D, Greaves Cotton Ltd.

      Aurangabad.

      Mr. Amit Chaudhari

      Manager R&D, Greaves Cotton Ltd.

      Aurangabad.

      Abstract- In an internal combustion engine, crankshaft is one of the most critical components. The main objective of this paper is to eliminate fatigue failure and improve fatigue life of crankshaft. FEA is carried out in ANSYS with the actual boundary conditions to simulate the physics. In this study a detailed stress analysis of crankshaft of single cylinder diesel engine is done. This paper describes the exact nature and magnitude of stresses at crankpin fillet and journal fillet during crankshaft operation. This paper also describes theoretical stress at critical locations and nature of stresses on crankshaft during engine operating condition and how this can be simulated in FEA. Despite huge amount of papers written on crankshaft stress analysis, detailed nature of stress and procedure to simulate the stress condition in FEA is lacking in literature. This paper also describes the possible modification in crankshaft to increase the fatigue life without modifying other interacting components of crankshaft.

      Keywords crankshaft failure; fatigue life; FEA; PRO-E; ANSYS.

      1. INTRODUCTION

        Crankshaft is one of the most important components of internal combustion engine. It is a long component with complex geometry consisting of cylinder bearings, and plates with crank web. Geometry section in crankshaft causes stress concentration at fillet areas where bearings are connected to crank web. The crankshaft converts the reciprocating displacement of piston to a rotary motion with a four link mechanism. The most of crankshaft load is applied to perpendicular to its rotational axis, and reaction forces are thus transmitted through the rod and main bearing, thus the force acts as a bending load. However in addition to this load, torsional load also acts on crankshaft along its axis of rotation. Compared to bending load, twisting load is very small, and does not have much impact on stress analysis. Angled drilling in crankshaft carry pressurized lubricant from main bearing to each of the rod bearing. To avoid stress concentration these holes are chamfered at the end and sharp edges are broken that might damage the bearings [1].

        A. Crankshaft Fillet Development

        Design of a durable crankshaft starts with a single throw and the alternating stress across the web between the rod

        and main fillets. Fig. 1 shows loading when crankshaft is at TDC during firing of a particular cylinder, the top of the rod throw is loaded by force transmitted from the connecting rod, and reaction force acting at the bottom of main bearing is transferred into the main bearing cap. The resulting crankshaft deflection places the rod fillet in tension and the main fillet in compression as shown in Fig. 1. After one revolution of crankshaft, loading at TDC is shown in Fig 2. The inertia forces at piston assembly now predominate and connecting rod cap loads the underside of the rod bearing surface. The reaction force is now transmitted into the block along the upper surface of the crankshaft main bearing. This place the rod bearing fillet in compression and the main bearing fillet in tension as shown in Fig. 2. In summary each operation cycle of the engine creates an alternating stress component of simple bending across the crankshaft web. High cycle fatigue due to this bending mode is thus primary failure mechanism. For crankshaft durability, material properties and dimensions are most important. As the fillet radii are increased, the peak stress magnitude decreases, but bearing width decreases as well, in some cases the radii are undercut in order to increase the radii and simultaneously maintain the bearings area. As the crankshaft is being developed, it is important to determine the stress profile across the fillet radius and to identify the location of maximum stress along the radius [1].

        Fig. 1 Nature of stresses on fillet under compressive loading

        Fig. 2 Nature of stresses on fillet under tensile loading

      2. LITERATURE SURVEY

        Failure of the crankshaft makes an engine unworkable, resulting into costly procurement and replacement. An extensive research in the past shows crankshaft is subjected to multi-axial loading (Bending and Torsion), stress gradient and stress concentration and effect of variable amplitude loading. Effect of twisting load on the specified crankshaft is far less than the bending load, hence only bending load is considered for analysis for most of the crankshafts.

        R. Metkar [2] et al described finite element method as the most favorite method to solve stress and fatigue analysis and it is commonly used for analyzing engineering problems. He also studied stress life, strain life and LEFM methods to solve fatigue analysis. There are lot of softwares available for use in finite element analysis applications, such as, ANSYS, Abaqus, Nastran, and MSC.

        F. Monte [3] et al studied two crankshafts of same material. The analysis was done to predict root cause of failure of shaft, both crankshafts were failed due to fatigue but crack on first crankshaft was at fillet and on second crankshaft was at oil hole.

        Amardip Jadhav [4] et al studied the marine engine crankshaft and analyzed fatigue failure. They also performed experiment for evaluation of fatigue failure. Crack initiation due to the elliptical arrest lines, which comes in the plane or area between the crankpin and main

        crankshaft and found out improvement in results as compared to original geometry of crankshaft.

        Vijaykumar Khansis [7] et al studied stress, balancing and fatigue life of crankshaft. They modified geometry by adding very small amount of material at bevel section of crankshaft. Maximum stresses generated at bevel section in original crankshaft were minimized in modified geometry.

        Also none of the authors showed nature of stresses generated in crankpin and journal fillet explicitly, which is described in this paper.

      3. GEOMETRY OF CRANKSHAFT

        As per design dimensions, model has been created. Modeling of crankshaft is done using the PRO-E modeling software. Fig 3 shows crankshaft geometry with gear at one side and PTO at other side.

        Fig. 3 Crankshaft geometry

      4. LOADING CONDITIONS

        1. Force due to maximum gas pressure

          After expansion of gas, force generated due to maximum pressure in cylinder is applied to crankshaft. The slider-crank mechanism converts the maximum gas pressure into vertical force which is applied on the piston head and then transmitted to the joint between crankshaft and connecting rod [5].

          Fg = (Maximum Gas Pressure) × (C/s Area of Piston)

          2

          journal, LEFM technique is used. Failure is not due to wear

          Fg = P

          max

          ×

          4

          (4.1)

          but due to significant magnitude of bending and torsional load.

          Farzin H. Montazersadgh [5] et al suggested the modifications for improvement in fatigue life such as changing the main bearing radius, crank pin radius, fillet radius or main journal pin fillet and changing the type of material in crankshaft, which are very common modifications usually done in crankshaft geometry.

          Amitpal Singh Punewale and Amit Chaudhari [6] et al studied torsional vibration with modified crankshaft

          Fg = 39499.8 N

        Because of rotating as well as reciprocating components (e.g. connecting rod) the crankshaft is subjected to inertia force and this force increases with the increase of engine speed. Inertia force is a function of rotating speed and acceleration of rotating components [5].

        Fi = (mass of Reciprocating components i.e. piston assembly.) × (Acceleration)

        F = M 2 R (cos + 2) (4.2)

        i r

        geometry by adding some material at face inclination on

        Fi = 6112.34N

      5. FINITE ELEMENT ANALYSIS OF

CRANKSHAFT

The FE model of the crankshaft geometry is meshed with tetrahedral elements. Mesh refinement are done on the crank pin fillet and journal fillet, so that fine mesh is obtained on fillet areas, which are generally critical locations on crankshaft [8]. The meshed crankshaft is shown in Figure 4.

Fig. 4 Mesh model of crankshaft geometry.

  1. Boundary condition applied on crankshaft geometry

    As alternating loads are acting on crankshaft which differ in magnitude as well as direction, so loading conditions are very difficult task for analysis. For fatigue analysis, combination of two solutions (compressive and tensile loading) is most suitable and simplest way to obtain the stresses and life. In FEA, boundary conditions are applied on crankshaft are based on engine configuration and component mounting conditions. This crankshaft is mounted on same type of bearings but length of bearing is different [8]. Table I shows the different boundary conditions used to simulate FEA and corresponding results, but out of these conditions force and remote displacement support gives good results. Fig 5 shows application of maximum compressive force and constraints on the bearing area. Fig 6 shows application of maximum tensile force and constraints on the bearing area.

    Table-I: Different boundary conditions applied on geometry and corresponding results

    Sr.

    No.

    Boundary conditions

    Results

    1

    Force at crank end and Fixed support at bearing.

    Nature of stresses are incorrect.

    2

    Force at crank end and frictionless support (constraint in vertical direction) at bearing.

    Nature of stresses are incorrect.

    3

    Force at crank end and displacement support (constraints Y and Z direction, X direction is free to deform) at bearing.

    Nature of stresses are incorrect.

    4

    Force at crank end and remote displacement support (constraints Y and Z direction, X direction is free to deform, Rotation is free along X axis) at bearing.

    Nature of stresses are correct at both fillets.

    Fig. 5 Compressive loading

    Fig. 6 Tensile loading

  2. Results of Stress analysis

Crankpin fillet, journal fillet, oil hole region is considered for documentation. According to loading conditions applied on crankshaft geometry, results shows maximum and minimum stress locations [9]. Fig 7 shows normal stress of 484 MPa generated at crankpin fillet for compressive loading and Fig 10 shows 64 MPa compressive stress generated at crank pin fillet for tensile loading. Fig 11 shows equivalent stress plot for combustion loading. Fig 8 and 9 shows maximum and minimum stress locations at fillet.

Fig. 7 Normal stress for compressive loading

Fig. 8 Maximum stress location at crankpin fillet

Fig. 9 Minimum stress location at journal fillet

Fig. 10 Normal stress for tensile loading

Fig. 11 Equivalent stress for compressive loading

on crankpin fillet diameter, maximum stress is approximately at 1800 from the top and it is tensile in nature while at the same point it is compressive in nature when load is tensile. Fig 13 shows variation of stress along the y axis. Fig 14 shows variation of stress along x axis, it shows that there is sudden increment of stress at oil hole, which is located at a radial distance of 23mm thus indicating there is stress concentration near the oil hole. Also same variations are shown for journal fillet. From these figures it can be concluded that maximum stress is at outer layer of crankpin fillet, hence fatigue crack will originate from outer surface of fillet only.

Normal stress (MPa) Vs Crank pin diameter angle

Normal stress (MPa) Vs Crank pin diameter angle

combustion load

inertia load

combustion load

inertia load

600

500

400

300

200

100

0

600

500

400

300

200

100

0

-100

60

160

260

360

-100

60

160

260

360

-200

Crank pin diameter angle (degree)

-200

Crank pin diameter angle (degree)

-40

-40

combustion load inertia load

combustion load inertia load

Normal stress (MPa)

Normal stress (MPa)

Normal stress (MPa)

Normal stress (MPa)

Fig. 12 Graph between Normal stress Vs Angle on Crank pin fillet

Normal stress (MPa) Vs Crank pin diameter along Y-axis

Normal stress (MPa) Vs Crank pin diameter along Y-axis

600

500

400

300

200

100

0

-1-400

-200

6

16

26

36

46

600

500

400

300

200

100

0

-1-400

-200

6

16

26

36

46

Crank pin diameter (mm)

Crank pin diameter (mm)

Fig. 13 Graph between Normal stress Vs Crank pin diameter fillet along Y-axis

Fig. 8 Maximum stress location at crankpin fillet

Fig. 9 Minimum stress location at journal fillet

Fig. 10 Normal stress for tensile loading

Fig. 11 Equivalent stress for compressive loading

on crankpin fillet diameter, maximum stress is approximately at 1800 from the top and it is tensile in nature while at the same point it is compressive in nature when load is tensile. Fig 13 shows variation of stress along the y axis. Fig 14 shows variation of stress along x axis, it shows that there is sudden increment of stress at oil hole, which is located at a radial distance of 23mm thus indicating there is stress concentration near the oil hole. Also same variations are shown for journal fillet. From these figures it can be concluded that maximum stress is at outer layer of crankpin fillet, hence fatigue crack will originate from outer surface of fillet only.

Normal stress (MPa) Vs Crank pin diameter angle

Normal stress (MPa) Vs Crank pin diameter angle

combustion load

inertia load

combustion load

inertia load

600

500

400

300

200

100

0

600

500

400

300

200

100

0

-100

60

160

260

360

-100

60

160

260

360

-200

Crank pin diameter angle (degree)

-200

Crank pin diameter angle (degree)

-40

-40

combustion load inertia load

combustion load inertia load

Normal stress (MPa)

Normal stress (MPa)

Normal stress (MPa)

Normal stress (MPa)

Fig. 12 Graph between Normal stress Vs Angle on Crank pin fillet

Normal stress (MPa) Vs Crank pin diameter along Y-axis

Normal stress (MPa) Vs Crank pin diameter along Y-axis

600

500

400

300

200

100

0

-1-400

-200

6

16

26

36

46

600

500

400

300

200

100

0

-1-400

-200

6

16

26

36

46

Crank pin diameter (mm)

Crank pin diameter (mm)

Fig. 13 Graph between Normal stress Vs Crank pin diameter fillet along Y-axis

VIII.MODIFICATION IN CRANKSHAFT

GEOMETRY

Original geometry has been modified as some material is added on the face inclination area and balancing is done according to weight added so that crankshaft geometry is completely optimized. Changing the main bearing radius, crank pin radius, fillet radius or main journal pin fillet as suggested by authors, requires change in dimension of different components of crankshaft assembly, sometimes requiring change in manufacturing process. But modification suggested in this paper does not require any change in mating components of the crankshaft and does not require different manufacturing process. Fig 19 shows the modified geometry of crankshaft. After modification in geometry same boundary condition as that of original were applied on the crankshaft geometry and verification was done in ANSYS. The results of original geometry and modified geometry were then compared.

Addition of material on face inclination

Addition of material on face inclination

Fig. 19 Modification done on crankshaft geometry

  1. Results of modified geometry

    According to the applied loading conditions on modified crankshaft geometry, the results show maximum and minimum locations of stresses. Fig 20 shows normal stress of 393 MPa generated at crankpin fillet for compressive loading. Fig 21 shows 48 MPa compressive stresses generated at crank pin fillet for tensile loading. Fig 22 shows fatigue life of the component as 2.0998×108cycles.

    Fig. 20 Normal stress on modified crankshaft for compressive loading

    Fig. 21 Normal stress on modified crankshaft for tensile loading

    Fig. 22 Fatigue life for modified crankshaft

  2. Comparison of results

    Normal stress (MPa)

    Normal stress (MPa)

    Comparison between two results clearly shows that the stress is reduced by significant amount on modified crankshaft. Fig 24 and 25 shows comparison between normal stress at crankpin fillet and normal stress at journal fillet respectively on original as well as modified crankshaft. Maximum reduction in stress is almost 20%. Bar chart describes comparison of life at crankpin fillets on original as well as modified geometry and also indicates that maximum 15 times improvement in fatigue life of crankshaft.

    Normal stress (MPa) Vs Crank pin diameter angle

    Normal stress (MPa) Vs Crank pin diameter angle

    combustion load

    inetria load

    combustion load

    inetria load

    500

    400

    300

    200

    100

    0

    500

    400

    300

    200

    100

    0

    -100

    60

    160

    260

    360

    -100

    60

    160

    260

    360

    -200

    -200

    Crank pin diameter angle (degree)

    Crank pin diameter angle (degree)

    -40

    -40

    Fig. 23 Graph between Normal stress Vs Angle on Crank pin fillet on modified geometry

    Normal stress (MPa) Vs Crank pin diameter angle

    original modified

    600

    Normal stress (MPa) Vs Crank pin diameter angle

    original modified

    600

    CONCLUSION

    400

    400

    Normal stress (MPa)

    Normal stress (MPa)

    Detailed structural and fatigue analysis procedure for crankshaft analysis is described. Nature and magnitude of stresses at crankpin and journal fillet location are clearly shown. Also crankshaft geometry has been successfully modified and results are compared and verified using ANSYS.

    200

    200

    0

    -40

    -200

    60

    160

    260

    360

    0

    -40

    -200

    60

    160

    260

    360

    Crank pin diameter angle (degree)

    Crank pin diameter angle (degree)

    Normal stress (MPa)

    Normal stress (MPa)

    Fig. 24 Comparison of stresses between original and modified geometry on crankpin fillet

    Normal stress (MPa) Vs angle on journal diameter

    original modified

    Normal stress (MPa) Vs angle on journal diameter

    original modified

    100

    50

    0

    -4-500

    -100

    -150

    -200

    -250

    -300

    60

    160

    260

    360

    100

    50

    0

    -4-500

    -100

    -150

    -200

    -250

    -300

    60

    160

    260

    360

    Journal diameter angle (degree)

    Journal diameter angle (degree)

    Life at right crankpin fillet on original geometry life at right crankpin fillet on modified geometry

    2.10E+08

    1.35E+07

    Fatigue life (cycles)

    Life at right crankpin fillet on original geometry life at right crankpin fillet on modified geometry

    2.10E+08

    1.35E+07

    Fatigue life (cycles)

    Fig. 25 Comparison of stresses between original and modified geometry on journal fillet

    Fig. 26 Comparison of life at fillet locations

    Maximum stress location is observed at the same location as that of actual failure location for baseline crankshaft, with maximum tensile stress at crank pin fillet and minimum compressive stress at journal fillet.

    Slight addition of material on crankshaft face inclination, gives 20% reduction in stresses and 15 times increment in fatigue life for modified configuration.

    REFERENCES

    1. Vehicular engine design by Kevin L. Hoag, SAE publication.

    2. R.M. Metkar, V.K. Sunnapwar, S.D. Hiwase A fatigue analysis and life estimation of crankshaft- a review, International Journal of Mechanical and Materials Engineering (IJMME), Vol.6 (2011), No.3, 425-430.

    3. F. Monte, P. Duarte, L. Reis, M. Freitas, Failure mode analysis of two crankshafts of single cylinder diesel engine, Engineering Failure Analysis,2015.

    4. Amerdip Jadhav, Vijaykumar Chalwa, Fatigue Failure Analysis Of Marine Engine Crankshaft, International Journal of Engineering Research & Technology ,Vol. 2 Issue 6, June – 2013

    5. Farzin H. Montazersadgh and Ali Fatemi, Dynamic Load and Stress Analysis of a Crankshaft, SAE Technical Paper No. 010258, Society of Automotive Engineers, 2007.

    6. Amitpal Singh Punewale, Mr. Amit Chaudhari, Prof. N. M. Khandre, Analysis & Optimization of Torsional Vibrations in a Four-Stroke Single Cylinder Diesel Engine Crankshaft, International Journal of Engineering and Technical Research (IJETR), Volume-3, Issue-4, April 2015.

    7. Vijaykumar Khansis, Manoj Ukhande, Girish Tilekar, Rajesh mane, Girish Shegavi, Crankshaft design and optimization to improve dynamic Balancing and fatigue strength, International journal of automotive engineering,6(2015),59-66.

    8. Farzin H. Montazersadgh, Ali Fatemi, Stress Analysis and Optimization of Crankshafts Subject to Dynamic Loading, Project report, University of Toledo, 2007.

    9. Lakshmanan N., Ramachandran G.M. and Saravanan K., Dynamic Stress Analysis of a Multi cylinder Two-stage Compressor Crankshaft, Research Journal of Engineering Sciences, Vol. 1(4), 34-40, October (2012).

    10. Jalal Fathi Sola, Farhad Alinejad, Fatigue Life Analysis of an Upgraded Diesel Engine Crankshaft, 11th World Congress on Computational Mechanics (WCCM XI) , July 20 – 25, 2014.

    11. Prakash, V., Aprameyan, K., and Shrinivasa, U., An FEM Based Approach to Crankshaft Dynamics and Life Estimation, SAE Technical Paper No. 980565, Society of Automotive Engineers, 1998.

    12. Jonathan Williams, Farzin Montazersadgh, and Ali Fatemi, Fatigue Performance Comparison And Life Prediction Of Forged Steel And Ductile Cast Iron Crankshafts, Published in

      Proceeding of the 27th Forging Industry Technical Conference in Ft. Worth Texas , 2007.

    13. Zoroufi, M. and Fatemi, "A Literature Review on Durability Evaluation of Crankshafts Including Comparisons of Competing Manufacturing Processes and Cost Analysis", 26th Forging Industry Technical Conference, Chicago, 2005.

    14. Wei Li, Qing Yan, Jianhua Xue, Review-Analysis of a crankshaft fatigue failure, Engineering Failure Analysis, 55(2015).

      /li>

    15. Sanjay B Chikalthankar, V M Nandedkar, Surender Kumar Kaundal, Finite Element Analysis Approach for Stress Analysis of Crankshaft under Dynamic Loading, international journal of scientific & engineering research, volume 4, february-2013.

    16. A. Ktari, N. Haddar, H.F. Ayedi. Fatigue fracture expertise of train engine crankshafts. Engineering Failure Analysis 18 (2011) 10851093.

    17. Rajesh .Metkar, Vivek Sunnapwar, Subhash Hiwase, Evaluation of FEM based fracture mechanics technique to estimate life of an automotive forged steel crankshaft of a single cylinder diesel engine, Procedia Engineering (2013).

    18. Ilya piraner, Christine Pflueger, Oliver Bouthier, Cummins crankshaft and Bearing analysis process, (Cummins Inc.)

    19. K. Thriveni, Dr.B.JayaChandraiah, Modeling and Analysis of the Crankshaft Using Ansys Software, International Journal of Computational Engineering Research, Vol, 03Issue, 5.

    20. Mr.B.Varun, Stress Analysis and Optimization of Crankshafts Subject to Static Loading, International Journal Of Engineering And Computer Science, Volume 3 Issue 5 May, 2014.

One thought on “Diesel Engine Crankshaft High Cycle Fatigue Life Estimation and Improvement Through FEA

Leave a Reply

Your email address will not be published. Required fields are marked *