A Thesis entitled

Fatigue Behavior and Life Predictions of Forged Steel and Powder Metal Connecting Rods
by

Adila Afzal

Submitted as a partial fulfillment of the requirements for the Master of Science Degree in Mechanical Engineering

Advisor: Dr. Ali Fatemi

Graduate School

The University of Toledo May 2004

The University of Toledo

College of Engineering

I HERBY RECOMMEND THAT THE THESIS PREPARED UNDER MY SUPERVISION BY Adila Afzal
ENTITLED Fatigue Behavior and Life Predictions of Forged Steel and Powder Metal Connecting Rods

BE ACCEPTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF Master of Science in Mechanical Engineering

Thesis Advisor: Dr. A. Fatemi

Recommendation concurred by Committee Dr. Mehdi Pourazady On

Dr. Hongyan Zhang

Final Examination

Dean, College of Engineering

AN ABSTRACT OF

Fatigue Behavior and Life Predictions of Forged Steel and Powder Metal Connecting Rods

Adila Afzal

Submitted as partial fulfillment of the requirements for The Master of Science in Mechanical Engineering

The University of Toledo May 2004

This study investigates and compares fatigue behavior of forged steel and powder metal connecting rods. A literature review on several aspects of connecting rods in the areas of load and stress analysis, durability, manufacturing, economic and cost analysis, and optimization is also provided. The experiments included strain-controlled specimen testing, with specimens obtained from the connecting rods, as well as load-controlled connecting rod bench testing. Monotonic and cyclic deformation behaviors, as well as strain-controlled fatigue properties of the two materials are evaluated and compared.

ii

Specimens from C-70 connecting rods were also tested and results compared with forged steel and powder metal connecting rods. C-70 connecting rod is considered to be an economical alternative to powder metal and conventional steel connecting rods. Experimental S-N curves of the two connecting rods from the bench tests obtained under R = -1.25 constant amplitude axial loading conditions are also evaluated and compared. Fatigue properties obtained from specimen testing are then used in life predictions of the connecting rods, using the S-N approach. The predicted lives are compared with bench test results and include the effects of stress concentrations, surface finish, and mean stress. The stress concentrations factors were obtained from FEA, and the modified Goodman equation was used to account for mean stress effect. Fractography of the connecting rod fracture surfaces were also conducted to investigate the failure mechanisms. A discussion of manufacturing cost comparison and recent developments in crackable forged steel connecting rods are also included.

iii

ACKNOWLEDGEMENTS
First of all I would like to acknowledge my parents, Mr. Afzal and Surrya, and my husband, Babar Majeed, who supported me through this task. They not only supported me financially with their limited resources but also help me emotionally. I will always be grateful for their love and support. My brother Usman is the true driving force behind me who truly understand my strengths and inspired me to choose this path. It was privileged to work under Dr. Ali Fatemi who influenced this work in many ways. He not only guided me in the right direction but also broaden my thoughts. I would also like to acknowledge the American Iron and Steel Institute for funding this project. Atousa Plaseied, Hui Zhang, Fengjie Yin and Bing Li are acknowledged for providing experimental assistance and raw data collection.

iv

TABLE OF CONTENTS
ABSTRACT..........ii ACKNOWLEGEMENTS....iv LIST OF TABLES..viii LIST OF FIGURES.......x NOMENCLATURE..xviii CHAPTER 1: INTRODUCTION.......1 1.1 Background.......1 1.1.1 Function of connecting rod in an engine.....2 1.1.2 Service loads and failures experienced by connecting rods....3 1.2 Motivation and Objectives of the Study..5 CHAPTER 2: MATERIALS AND MANUFACTURING PROCESSES.......12 2.1 Drop-Forged.......12 2.2 Powder Forging......16 2.3 Die Casting ....18 2.4 Comparison of Forged Steel and Powder Metal Connecting Rods....19 CHAPTER 3: LITERATURE REVIEW..........29 3.1 Load and Stress Analyses...29 3.2 Mechanical Properties and Fatigue Behavior.34 3.3 Density Distributions and Dimensional Accuracy.....39 3.4 Shot Peening and Residual Stress Effects......40 3.5 Optimization Studies..41

3.6 Material Selections, Manufacturing Process and Economic Cost.....45 3.7 Summary....48 CHAPTER 4: STRESS ANALYSIS AND FEA.......118 4.1 Finite Element Modeling......118 4.1.1 Generation of the geometry and FEA mesh of the component.118 4.1.2 Loading and boundary conditions.....119 4.1.3 Mesh size sensitivity analysis....122 4.2 Finite Element Analysis Results and Discussion.123 CHAPTER 5: FATIGUE BEHAVIOR, COMPARISONS, AND LIFE PREDICTIONS.......156 5.1 Material Fatigue Behavior and Comparisons...156 5.1.1 Material, specimen, and test equipment....156 5.1.2 Test methods and procedures....159 5.1.2.1 Monotonic tension tests......159 5.1.2.2 Constant amplitude fatigue tests.160 5.1.3 Experimental results and comparisons...161 5.1.3.1 Monotonic deformation behavior....161 5.1.3.2 Cyclic deformation behavior.......163 5.1.3.3 Constant amplitude fatigue behavior..........165 5.2 Component Fatigue Behavior and Comparisons......168 5.3 Life Predictions and Comparisons with Experimental Results171 5.3.1 Stress-life (S-N) curves for forged steel connecting rod ..173 5.3.2 Stress-life (S-N) curves for powder metal connecting rod........175 5.3.3 Comparisons with experimental results.....175

vi

5.4 Discussion of Results and Potential for Improvements176 CHAPTER 6: SUMMARY AND CONCLUSIONS.....231 REFERENCES...234

vii

LIST OF TABLES
Table 2.1: Summary of cost breakdown of manufacturing processes for conventional forged steel and powder metal connecting rods (Clark et al., 1989).....22 Table 3.1: Summary of loading, stress analysis and fatigue studies survey........51 Table 3.2: Optimization studies survey...........66 Table 3.3: Manufacturing process and economic studies survey........73 Table 3.4: Other studies...........80 Table 3.5: The local stress amplitude obtained for the connecting rod critical areas (Sonsino, 1996)...83 Table 3.6: The typical engine parameters of 4-cylinder engine (Folgar et al., 1987)..84 Table 3.7: Maximum stress values and locations in a connecting rod (Webster et al., 1983)...84 Table 3.8: Comparison of three major manufacturing processes of connecting rod...85 Table 4.1: Summary of forged steel connecting rod parameters used for FEA.129 Table 4.2: Summary of connecting rod FE analysis load parameters....129 Table 4.3: Summary of von Mises stresses at different element sizes...130 Table 4.4: Summary of stress concentration factors......131 Table 4.5: Stresses at the nodes in region I shown in Figure 4.12 (a) (transition region between the shank and the crank end) under tension and compression load of 26688 N (6000 lbs)......132 Table 4.6: Stresses at the nodes in region II shown in Figure 4.12 (b) (region between two transition regions) under tension and compression load of 26688 N (6000 lbs)...133 Table 4.7: Stresses at the nodes in region III shown in Figure 4.12 (c) (transition region between the shank and the pin end) under tension and compression load of 26688 N (6000 lbs).134

viii

Table 4.8: Alternating and mean stress (MPa) components for the nodes in region I..135 Table 4.9: Alternating and mean stress (MPa) components for the nodes in region II..136 Table 4.10: Alternating and mean stress (MPa) components for the nodes in region III.....137 Table 4.11: Equivalent von Mises alternating, mean, and maximum stresses for region I...138 Table 4.12: Equivalent von Mises alternating, mean, and maximum stresses for region II.........139 Table 4.13: Equivalent von Mises alternating, mean, and maximum stresses for region III............140 Table 5.1: Chemical composition by percent weight for three materials......178 Table 5.2: Summary of mechanical properties for three materials....179 Table 5.3: Summary of monotonic tensile test results.......180 Table 5.4: Summary of constant amplitude completely reversed fatigue test results for forged steel and powder metal.........181 Table 5.5: Summary of constant amplitude completely reversed fatigue test results for C-70 steel.............182 Table 5.6: Summary of dimensions and weights for forged steel and powder metal connecting rods.........183 Table 5.7: Summary of connecting rod fatigue test parameters and results..184 Table 5.8: Fatigue life prediction parameters for S-N line equation, Sa = A(Nf)B185

ix

LIST OF FIGURES
Figure 1.1: Schematic of a typical connecting rod (Schreier, 1999)..7 Figure 1.2: The four strokes in an engine, (a) intake cycle, (b) compression cycle, (c) power cycle, and (d) exhaust cycle (Schreier, 1999).......8 Figure 1.3: Even firing four strokes, four-cylinder engine (Nortan, 1994)........9 Figure 1.4: The origin of stresses on a connecting rod (Sonsino, 1996)..10 Figure 1.5: Two modes of bending that occur in connecting rods (Wright, 1993)...11 Figure 2.1: Schematic of drop-forged process of connecting rod (Todd et al., 1994)..23 Figure 2.2: Schematic of fracture splitting jig for C-70 steel (Fukuda and Eto, 2002)....23 Figure 2.3: Schematic of connecting rod cracking machine ultimate precisions rod fracturing machine can fracture powdered metal or C-70 type forged steel rods at ambient temperature (www.upinc.net/excello.htm).......24 Figure 2.4: Schematic of tooling configuration for powder-forging of connecting rods (Weber, 1993)........25 Figure 2.5: Schematic of fracture splitting process for powder forged connecting rod (Weber, 1993). 25 Figure 2.6: Schematic of plate for the cast connecting rod (Niebel et al., 1989).....26 Figure 2.7: Schematic of manufacturing process of cast blank connecting rod (Weber, 1993)...........26 Figure 2.8: Summary of forged steel connecting rod manufacturing processes......27 Figure 2.9: Summary of powder metal connecting rod manufacturing process..28 Figure 3.1: A free body diagram of piston forces in cylinder engine (Shiao and Moskwa, 1993).......86 Figure 3.2: A free body diagram of connecting rod (Shiao and Moskwa, 1993)....86 Figure 3.3: A free body diagram of forces on crank system (Shiao and Moskwa, 1993)...87

Figure 3.4: Schematic of load distribution (Folgar et al., 1987)........88 Figure 3.5: Schematic of cylinder pressure versus crack angle at 5000 r.p.m. (Majidpour et al., 2002)...88 Figure 3.6: Boundary conditions used for static analysis (Sugita et al., 1990)......89 Figure 3.7: Comparison between FEA calculated and strain gage measured value of stress at critical locations in static load analysis (Sugita et al., 1990)...90 Figure 3.8: Boundary conditions for quasi-dynamic FE analysis (Sugita et al., 1990)91 Figure 3.9: Schematic of link mechanism to measure stress in connecting rod under actual engine operating condition (Sugita et al., 1990).....92 Figure 3.10: The calculated values of lightweight connecting rod (Sugita et al., 1990)......93 Figure 3.11: Distribution of tension loading of the connecting rod (Webster et al., 1983)......94 Figure 3.12: Distribution of compressive loading of the connecting rod. (Webster et al., 1983).......94 Figure 3.13: Schematic of the deformation of the connecting rod. (Webster et al., 1983)...95 Figure 3.14: Stretch and out of roundness as a result of inertial load. (Balasubramanian et al., 1991).........95 Figure 3.15: Crankpin pressure distribution under inertia load. (Balasubramanian et al., 1991)...96 Figure 3.16: Deformation and out of roundness from a firing load. (Balasubramanian et al.,1991)......99 Figure 3.17: Crankpin pressure distribution as a result of firing load. (Balasubramanian et al., 1991......97 Figure 3.18: Comparison of fatigue strength of powder forged (Fe-3Cu-0.55C-0.35) and hot forged (SAE 40L559) connecting rod (Imahashi et al., 1984)......98 Figure 3.19: The effect of temperature on fatigue limit of smooth and notched specimens made of microalloyed steels (Kuratomi et al., 1990)................98 Figure 3.20: The fatigue strength of as-sintered powder metal connecting rod as

xi

compared with forged steel connecting rod (Suzuki et al., 1988)...........99 Figure 3.21: Typical fatigue test rig of connecting rods (Suzuki et al., 1988).......100 Figure 3.22: Weibull probability plot of powder forged and forged steel rods (Marra et al., 1992)........101 Figure 3.23: Differentiation of fatigue cycles on the connecting rod (Beretta et al., 1996)......102 Figure 3.24: S-N plot of hot forged carbon steel under cyclic alternating loads (R = -1) (Berreta et al., 1996).........103 Figure 3.25: S-N plot of type-B cast iron at different stress ratio (a) R = 0, (b) R = -1, (c) R = -1.84 (Berreta et al., 1996).........104 Figure 3.26: Staircase fatigue test results of 1151 steel connecting rods (R=-1) (Olaniran and Stickels, 1993)... .105 Figure 3.27: Staircase fatigue test of new alloy connecting rods (Olaniran and Stickels, 1993)........105 Figure 3.28: Density distribution for powder metal connecting rod (Araki et al., 1993)...106 Figure 3.29: The effect of density on mechanical properties of 4640 powder forged steel (Imahashi et al., 1984)...........106 Figure 3.30: The effect of density on fatigue strength of 4640 powder forged steel (Imahashi et al., 1984)...107 Figure 3.31: Residual stress profile in fatigue at 20 A intensity (Chernenkoff et al., 1995)....108 Figure 3.32: The staircase fatigue test results for shot peened and unpeened connecting rods (Chernenkoff et al., 1995)...........109 Figure 3.33: The statistical comparison of shot peened and unpeened connecting rods (Chernenkoff et al., 1995)..........110 Figure 3.34: The influence of shot peening on fatigue strength (Chernenkoff et al., 1995)..111 Figure 3.35: The comparison of FE calculated and strain gage measured stress (Sugita et al., 1990)....111

xii

Figure 3.36: The boundary conditions used in shape optimization study by Yoo et al., 1984........112 Figure 3.37: Geometry and loading of an engine connecting rod (Yoo et al., 1984)..112 Figure 3.38: Optimum design resulting in 25% weight reduction (Yoo et al., 1984).....113 Figure 3.39: Views of the several designs including final bearing optimal design by Goenka and Oh, 1986........114 Figure 3.40: The manufacturing process of new developed connecting rod (Kuratomi et al., 1995).......115 Figure 3.41: Use of microalloyed forged steel employed for heavy duty diesel engine connecting rod (Banerji, 1996)..........115 Figure 3.42: A trial powder forged connecting rod production process (Araki et al., 1993)...116 Figure 3.43: Cost ratio of powder forged vs. hot forging (Imahashi et al., 1984)..116 Figure 3.44: Manufacturing energy comparison of powder forged vs. hot forging (Imahashi et al., 1984)...........116 Figure 3.45: Comparison of raw materials (Imahashi et al., 1984)....117 Figure 4.1: Detailed generated geometry of the connecting rod........141 Figure 4.2: Schematic of parabolic tetrahedron element shape used for FEA meshing142 Figure 4.3: Loading and constraint used for FEA of the connecting rod (a) Tension loading at crank end, (b) compression loading at crank end, (c) tension loading at pin end, and (d) compression loading at pin end.......143 Figure 4.4: von Mises stress locations used for mesh size sensitivity analysis (loading at crank end and constraint at pin end)..144 Figure 4.5: Schematic of connecting rods at different element mesh sizes...145 Figure 4.6: Plot of von Mises stress vs. element size at 26688 N (6000 lbs) tension load at the crank end for locations identified in figure 4.4...146 Figure 4.7: von Mises stress (MPa) contours of connecting rod 26688 N (6000 lbs) load magnitude. (a) Tension at the crank end, (b) compression at the crank end, (c) tension at the pin end, and (d) compression at the

xiii

pin end.............147 Figure 4.8: Displacement contours at magnification factor of 5 in X-direction. (a) Tension at the crank end, (b) compression at the crank end, (c) tension at the pin end, and (d) compression at the pin end........148 Figure 4.9: Stress distribution profile at the crank end transition region in the loading direction (x-axis)........149 Figure 4.10: Comparison between calculated and FEA obtained stress values at 26688 N (6000 lbs) tension load.....150 Figure 4.11: Locations of strain gauges on front of the connecting rod (two gauges are also used on the backside, behind gauges 1 and 2)...151 Figure 4.12: Locations of nodes where stresses were obtained.....152 Figure 4.12 (a): Region I (transition region between crank end and shank) nodes...153 Figure 4.12 (b): Region II (region between crank end transition and pin end transition) nodes.....154 Figure 4.12 (c): Region III (transition region between pin end and shank) nodes155 Figure 5.1: As-received connecting rods (a) forged steel, (b) powder metal, and C-70 steel.......186 Figure 5.2: Photomicrograph in the longitudinal direction for forged steel. (a) Microstructure at 500X after etching. (b) Inclusions at 100X....187 Figure 5.3: Photomicrograph in the longitudinal direction for powder metal. (a) Microstructure at 500X after etching. (b) Inclusions at 100X....188 Figure 5.4: Photomicrograph in the longitudinal direction for C-70 steel. (a) Microstructure at 500X after etching. (b) Inclusions at 100X.189 Figure 5.5: Locations of two specimens obtained from each connecting rod....190 Figure 5.6: Specimen configuration and dimensions (all dimensions are in mm).....191 Figure 5.7: True stress versus true plastic strain for (a) forged steel, (b) powder metal powder metal, and (c) C-70 steel..192 Figure 5.8: Monotonic stress-strain curves for (a) forged steel, (b) powder metal, and (c) C-70 steel. .......193

xiv

Figure 5.9: Superimposed monotonic stress-strain curves for three materials......194 Figure 5.10: True stress amplitude versus normalized number of cycles for (a) forged steel, (b) powder metal, and (c) C-70 steel....195 Figure 5.11: True stress amplitude versus number of cycles for (a) forged steel, (b) powder metal, and (c) C-70 steel......196 Figure 5.12: True stress amplitude versus calculated true plastic strain amplitude for (a) forged steel, (b) powder metal, and (c ) C-70 steel........197 . Figure 5.13: True stress amplitude versus true strain amplitude for (a) forged steel, (b) powder metal, and (c) C-70 steel..198 Figure 5.14: Superimposed plots of true stress amplitude versus true strain amplitude for three materials..199 Figure 5.15: Composite plots of cyclic and monotonic stress-strain curves for (a) forged steel, (b) powder forged, and (c) C-70 steel......200 Figure 5.16: Superimposed plots of cyclic and monotonic stress-strain curves201 Figure 5.17: Composite plots of midlife hysteresis loops for (a) forged steel, (b) Powder metal, and (c) C-70 steel..202 Figure 5.18: True stress amplitude versus reversals to failure for (a) forged steel, (b) powder metal, and (c) C-70 steel.203 Figure 5.19: Superimposed plots of true stress amplitude versus reversals to failure...204 Figure 5.20: Calculated true plastic strain amplitude versus reversals to failure for (a) forged steel, (b) powder metal, and (c) C-70 steel.......205 Figure 5.21: Superimposed plots of calculated plastic strain amplitude versus reversals to failure..206 Figure 5.22: True strain amplitude versus reversals to failure for (a) forged steel, (b) powder metal, and (c) C-70 steel.....207 . Figure 5.23: Superimposed curves of true strain amplitude versus reversals to failure.....208 Figure 5.24: Schematic of fatigue test fixture............209 Figure 5.25: Photo of fatigue test fixture used for connecting rod testing.210

xv

Figure 5.26: Load amplitude vs. cycles to failure......211 Figure 5.27: Experimental stress amplitude vs. cycles to failure for forged steel and powder metal connecting rods.......212 Figure 5.28: Digital photographs of FS6 connecting rod (a) as fractured and (b) close-up of the overall fracture surface.....213 Figure 5.29: Digital photographs of FS9 connecting rod (a) as fractured and (b) close-up of the overall fracture surface.....214 Figure 5.30: Digital photographs of FS1 connecting rod (a) as fractured and (b) close-up of the overall fracture surface.....215 Figure 5.31: Phototmicrograph of the crack initiation site for FS6 connecting rod at 50X.....216 Figure 5.32: Phototmicrograph of the crack initiation site for FS9 connecting rod at 50X.217 Figure 5.33: SEM image of FS6 connecting rod fracture surface showing (a) crack origin site with undesired morphology, and (b) microvoid coalescence at the crack site..........218 Figure 5.34: Photomicrograph of FS9 connecting rod showing close-up of crack initiation site followed by crack propagation....219 Figure 5.35: Digital photographs of PM10 connecting rod (a) as fractured and (b) close-up of the overall fracture surface.........220 Figure 5.36: Digital photographs of PM7 connecting rod (a) as fractured, and (b) close-up of the overall fracture surface.............221 Figure 5.37: Digital photographs of PM14 connecting rod (a) as fractured, and (b) close-up of the overall fracture surface.........222 Figure 5.38: Photomicrograph of the crack initiation site for PM7 connecting rod at 50X....223 Figure 5.39: Photomicrographs of PM7 connecting rod (a) close-up of crack site, (b) close-up of porosity, and (c) close-up of porosity at outer surface at different site..........224 Figure 5.40: S-N lines for a component without mean or residual stress..........225 Figure 5.41: S-N lines for forged steel connecting rod......226

xvi

Figure 5.42: S-N lines for powder metal connecting rod.......227 Figure 5.43: Comparison of predicted and experimental data for forged steel connecting rod...228 Figure 5.44: Comparison of predicted and experimental data for powder metal connecting rod.......229 Figure 5.45: Comparison of experimental lives with predicted lives.....230

xvii

NOMENCLATURE
Ao, Af b, c n, n E, E e HB, HRB, HRC K, K Kf Kt Ks Lo, Lf N50%, (Nf)10%, (Nf)50% 2Nf Pa Pm Pmax Pmin R r S Sa initial, final area fatigue strength, fatigue ductility coefficient monotonic, cyclic strain hardening exponent monotonic, midlife cycle modulus of elasticity engineering strain Brinell, Rockwell B-Scale, Rockwell C-Scale hardness number monotonic, cyclic strength coefficient fatigue notch factor stress concentration factor surface finish factor initial, final gage length number of cycles to midlife, 10% load drop, 50% load drop reversals to failure alternating load mean load maximum load minimum load stress ratio radius of the notch root engineering stress stress amplitude

xviii

SNF Sm Su to, tf wo, wf YPE YS, UYS, LYS, YS %EL %RA e, p, f, f a, m, e, p , f, f a, m, eq(a) eq(max) eq(m) min, max

fatigue strength mean stress ultimate tensile strength initial, final thickness initial, final width yield point elongation monotonic yield, upper yield, lower yield, cyclic yield strength percent elongation percent reduction in area true elastic, plastic, total strain true fracture ductility, fatigue ductility coefficient strain amplitude, mean strain, strain range elastic, plastic strain range characteristic length depending on the material true stress, true fracture strength, fatigue strength coefficient stress amplitude, mean stress, stress range equivalent alternating stress equivalent maximum stress equivalent mean stress minimum, maximum stress poissons ratio

xix

Chapter 1 INTRODUCTION
1.1 Background
Connecting rods are widely used in variety of engines such as, in-line engines, Vengines, opposed cylinder engines, radial engines and oppose-piston engines. A connecting rod consists of a pin-end, a shank section, and a crank-end as shown in Figure 1.1. Pin-end and crank-end pinholes at the upper and lower ends are machined to permit accurate fitting of bearings. These holes must be parallel. The upper end of the connecting rod is connected to the piston by the piston pin. If the piston pin is locked in the piston pin bosses or if it floats in the piston and the connecting rod, the upper hole of the connecting rod will have a solid bearing (bushing) of bronze or a similar material. As the lower end of the connecting rod revolves with the crankshaft, the upper end is forced to turn back and forth on the piston pin. Although this movement is slight, the bushing is necessary because of the high pressure and temperatures. The lower hole in the connecting rod is split to permit it to be clamped around the crankshaft. The bottom part, or cap, is made of the same material as the rod and is attached by two bolts. The surface that bears on the crankshaft is generally a bearing material in the form of a separate split shell. The two parts of the bearing are positioned in the rod and cap by dowel pins, projections, or short brass screws. Split bearings may be of the precision or semi precision type.

The precision type bearing is accurately finished to fit the crankpin and does not require further fitting during installation. It is positioned by projections on the shell that match reliefs in the rod and cap. The projections prevent the bearings from moving sideways and prevent rotary motion in the rod and cap (http://em-ntserver.unl.edu).

1.1.1 Function of connecting rod in an engine The function of connecting rod is to transmit the thrust of the piston to the crankshaft. Figure 1.2 shows the role of connecting rod from reciprocating motion into rotary motion. A four-stroke engine is the most common type. The four strokes are intake, compression, power, and exhaust. Each stroke requires approximately 180 degrees of crankshaft rotation, so the complete cycle would take 720 degrees. Each stroke plays a very important role in the combustion process as shown in Figure 1.2. In the intake cycle, while the piston moves downward, one of the valves open. This creates a vacuum, and an air-fuel mixture is sucked into the chamber, see Figure 1.2 (a). During the second stroke compression occurs. In compression both valves are closed, and the piston moves upward and thus creates a pressure on the piston, see Figure 1.2 (b). The next stroke is power. During this process the compressed air-fuel mixture is ignited with a spark, causing a tremendous pressure as the fuel burns. The forces exerted by piston transmitted through the connecting rod moves the crankshaft, see Figure 1.2 (c). Finally, the exhaust stroke occurs. In this stroke, the exhaust valve opens, as the piston moves back upwards, it forces all the air out of the chamber and thus which completes the cycle of crankshaft rotation, see Figure 1.2 (d).

1.1.2 Service loads and failures experienced by connecting rods The function of connecting rod is to translate the transverse motion to rotational motion. It is a part of the engine, which is subjected to millions of repetitive cyclic loadings. It should be strong enough to remain rigid under loading, and also be light enough to reduce the inertia forces which are produced when the rod and piston stop, change directions and start again at the end of each stroke. The connecting rod should be designed with high reliability. It must be capable of transmitting axial tension, axial compression, and bending stresses caused by the thrust and pull on the piston, and by centrifugal force without bending or twisting. An explanation of the axial forces acting on connecting rod is provided by Tilbury (1982). The top-dead-center (TDC) position of the induction stroke is denoted by 0 0 in Figure 1.3. From 0 0 to approximately 75 0 of the crank angle the connecting rod is in compression as the piston is accelerated down the cylinder bore. The connecting rod begins to decelerate from 75 0 to approximately 285 0 of the crank angle. From 285 0 to 360 0 the connecting rod is in tension as it reduces the rate of acceleration of the piston up the cylinder bore. Between 0 0 and 180 0 the pressure is slightly below atmospheric, as air is drawn into the engine through the spark plug hole, causing compression in the connecting rod. Between 180 0 and 360 0 the pressure is above atmospheric as the air is expelled from the engine. Connecting rod is submitted to mass and gas forces. The superposition of these two forces results in the axial force, which acts on the connecting rod. The gas force is determined by the speed of rotation, the masses of the piston, gudgeon pin and oscillating part of the connecting rod consisting of the small end and the shank. Figure 1.4 shows

axial loading (Fay) due to gas pressure and rotational mass forces. Bending moments (Mb,xy, Mb,zy) originate due to eccentricities, crankshaft, case wall deformation, and rotational mass force, which can be determined only by strain analyses in engine (Sonsino, 1996). The connecting rod experiences inertia forces plus direct forces that produce bending in a plane perpendicular to the crankshaft longitudinal axis as shown in Figure 1.5, where part (a) shows bending perpendicular and part (b) shows bending parallel to the crankshaft axis (Wright, 1993). Connecting rod is typically designed for infinite-life and the design criterion is endurance limit. It experiences axial tension/compression with constant amplitude loading and multi-directional bending with variable amplitude, as inertia force, torque and moment are all functions of engine speed (rpm). The ratio of a (axial stresses) to b (bending stresses) is 1: 1 to 1: 0.1 (Sonsino, 1996). Failures of connecting rods are often caused by bending loads, as shown in Figure 1.5 acting perpendicular to the axes of the two bearings. Failure in the shank section as a result of these bending loads occurs in any part of the shank between piston-pin end and the crank-pin end. At the crank end fracture can occur at the threaded holes or notches for the location of headed bolts. Pin-end failures can occur from fretting in the bore against a fitted bushing (Wright, 1993).

1.2 Motivation and Objectives of the Study

The intense competition in the marketplace for superior products has motivated the evolution of high performance components, which must be designed and fabricated economically. In spite of higher cost of powder metals raw material, some automotive manufacturers have switched from forged steel connecting rods to powder metal connecting rods. A major reason for this has been the cost associated with the machining operations to separate the rod body from the cap and to provide a sound fit between the matching joint faces. Powder metal connecting rods do not require this additional machining. Recently, the development of C-70 and microalloyed splittable steels is attracting attention. These steels reduce production cost by eliminating the machining cost associated with matching surfaces of the cap and body. Simultaneously, they provide the benefit of high fatigue strength as compared to powder metal connecting rods. The objective of this research project was to obtain and compare the fatigue behavior of forged steel and powder metal connecting rods. Specimen tests were carried out to compare mechanical and fatigue properties for forged steel, C-70 steel, and powder metal materials. Component tests were conducted to investigate and compare the differences in fatigue behavior of forged steel and powder metal connecting rods. Elastic finite element analyses were carried out to predict forges steel connecting rod stresses, stress concentrations, stress states, and stress ratios. The S-N approach was used to perform the fatigue life predictions and to compare the predictions with the experimental results. Chapter 2 provides description of commonly used connecting rod manufacturing processes. Cost comparisons of forged steel and powder metal processes are also

discussed. Chapter 3 gives a literature review on loading, fatigue, optimization, and manufacturing processes and costs of connecting rods. Chapter 4 discusses the finite element analysis, modeling of the connecting rod, boundary conditions, and the obtained results. Chapter 5 gives a detailed description of specimen and component experiment program and results. Fracture locations and fracture surfaces are also discussed. Stressbased approach is used to predict fatigue lives and compared with experimental data for forged steel and powder metal. Chapter six summarizes the conclusions drawn based on the experimental and analytical results obtained in this work.

Piston pin end

Shank section

Crank pin end Figure 1.1: Schematic of a typical connecting rod (Schreier, 1999).

(a)

(b)

(c)

(d)

Figure 1.2: The four strokes in an engine, (a) intake cycle, (b) compression cycle, (c) power cycle, and (d) exhaust cycle (Schreier, 1999).

Figure 1.3: Even firing four strokes, four-cylinder engine (Nortan, 1994).

Figure 1.4: The origin of stresses on a connecting rod (Sonsino, 1996).

10

Figure 1.5: Two modes of bending that occur in connecting rods (Wright, 1993).

11

Chapter 2 MATERIALS AND MANUFACTURING PROCESSES

Automotive companies are continuously searching for new and better cost effective ways to manufacture connecting rods. The common process technologies for the bulk production of connecting rods are drop-forged, casting, and powder-forged.

2.1 Drop-Forged
Forging is a plastic deformation process in which the work piece is compressed between two dies, using either impact or gradual pressure to form the part. This process is used to shape metal objects that must withstand great stress. The forging operation can be carried out at room temperature (cold working) or at elevated temperatures. Forging offers a high strengthto-weight ratio, good toughness and resistance to impact and fatigue. Metals commonly used for forging include carbon steel, alloy steel, stainless steel, aluminum, copper, bronze, brass and magnesium. A forged steel connecting rod is a production of drop-forged closed die process. The round steel stock as being forged to a connecting rod is shown in Figure 2.1. Hot working proportions the metal for forming the connecting rods. Fullering, which is the portion of the die, is used in hammer forging primarily to reduce the cross section and lengthen a portion of the forging stock. The fullering impression is often used in

12

conjunction with an edger or edging impression. Bustering converts square section bar into a preform to reduce the cross-section and lengthen it. Blocking operation forms the connecting rod into its first definite shape. This involves hot working of the metal in several successive blows of the hammer, compelling the work piece to flow into and fill the blocking impression in the dies. Flash is produced, which is the unformed metal around the edge of the connecting rod that was forced away from blocking die impressions by the successive blows of the forging hammer. The exact shape of each connecting rod is obtained by the impact of several additional blows in the hammer, which force the stock to completely fill every part of the finishing impression. The flash or unformed metal around the edge of the finished forging still has to be trimmed. Flash is removed by different ways with trim dies in mechanical press or in special circumstances by sawing and grinding. The trimmed connecting rod is ready for heat-treating and machining. To acquire closer tolerances the coining (which is the operation of applying heavy pressure to a relatively small portion of a forgings surface in order to obtain closer tolerance) and sizing process is required. Coining may be performed hot or cold. In order to obtain an excellent surface finish, cold coining is preferred. It should be mentioned that in net shape flashless forging, the preform is totally enclosed in the die cavity so that no flash formation is allowed. There is no material waste as in impression forging (Vazquez and Altan, 2000). Heat treating: After final forging and before machining, proper heat treatment methods are used to acquire optimum grain size, microstructure and mechanical properties.

13

Shot peening/blasting: During forging scale is formed as a result of chemical action between oxygen in the furnace and the heated metal. For further operations scale is removed by shot-peening, tumbling, or pickling. In shot peening, the parts are bombarded by high-velocity streams of steel or iron shots, hardened wire clips, aluminum oxide grit or sand which is propelled by centrifugal force or compressed air. The particles abrade the surface of the parts mechanically chipping, breaking and dislodging the scale. In addition, they induce compressive surface residual stresses, which are beneficial to fatigue life (Jensen, 1966). Machining: This is one of the major steps after the forging process. It includes machining of outer contours, drilling of oil bores, machining of cap bolt bores, machining of bolt head and nut seats, broaching of pin and crank end bearings, etc. (Weber, 1991) Separation of cap from rod: If the cap and rod are forged separately, separation of the piece is not an issue. The conventional method of forged rod manufacturing as a single piece requires sawing to separate the cap from the rod. This requires extensive machining of sawed surfaces to ensured a perfect match of the cap with the rod. In this case the bolts also cannot ensure perfect doweling of the cap and the rod upon bolting together because of diametrical clearance. Under operating conditions the microshifting of the cap and the rod affect bearing life. An alternative to sawing is splitting. Splitting of single piece rod was initially done by struck on one side with a sharp blow. The other method was to split the cap and rod with a wedge-expendable mandrel placed in the crank bore opening. Another approach was an electron beam used along a desired splitting plane. For all the conventional fracture splitting techniques the major disadvantage is the required machining after fracture splitting (Hoag and Yeager, 1990). To meet this challenge, the

14

manufacturing of connecting rods has undergone evolutionary changes. Recently, the development of C-70 and other so called splittable or crackable innovative microalloyed steels have eliminated the need for this additional machining, resulting in reduced production cost. Fracture splitting method for the C-70 steel connecting rod is shown in Figure 2.2. This method is a mechanical method in which the big end is fractured by splitting, hydraulically, from two notches at the crank end. The jig is mainly composed of (a) Ushaped jig, (b) fixture of big end and (c) side fixture of big end. The fracture splitting method is as follows: After the small end and big end of connecting rod are fixed to the jig, a U-shaped push jig is fitted to the bearing surface of the big end. Then, a push key (7o) is inserted into the big end, resulting in sliding of the cap of the U-shaped jig. Thus, the big end is split from the rod (Fukuda and Eto, 2002). A connecting rod cracking machine is shown in Figure 2.3. With this technology the fracture area is perfectly matched. Inspection: Connecting rods receive inspection for dimensional accuracy, weight and surface condition after finishing operations.

2.2 Powder Forging

Powder forging is a process in which powders such as iron and copper are compacted, heated and forged so that their density increases up to that of wrought steel. The technology involves the following steps: Stage I: A controlled amount of mixed powder used for connecting rods is automatically gravity-fed into a precision die and is compacted usually at room temperature and at high

15

pressure (this normally depends on the density requirement of the part) up to 200 to 400 MPa, as shown in Figure 2.4. The resulting mass of powder is a green compact and has very little cohesive strength requiring further operations. At the end of the cold closed compaction stage the powder mass has achieved a density of 60 to 70 percent of the fully dense metal stock. Stage II: The green component is ejected out of the dietool system and placed on wide endless mesh belt, which moves slowly through a controlled atmosphere-heating furnace. This is called sintering and it develops the metallurgical bonds between the green compacted grains. The green compacts are heated below the melting temperature of the base metal and held at the sintering temperature for an appropriate time and then cooled. At the end of the sintering process, the powder material has achieved a density of 70 to 85 percent of the fully dense metal. Stage III: The sintered powder compact normally requires further densification in a hot forging press to achieve complete densification comparable to the fully dense metal. In the forging step, shear flow of the material is achieved to gain a complete filling of the die cavity and this further results in full densification of the part achieving relative density equal to 0.91. The temperature and pressure required for the forging step depend on the type of powder and its application. Minimum contact time and high speed is very crucial; otherwise, porosity, decarb, and oxide penetration occurs. At the end of forging, the connecting rod is ready for further machining operations before use (Jinka, 1997). The causes of porosity, decarb, and oxide penetration, which can occur in powder forged connecting rods are briefly discussed below.

16

Porosity exists when there are pores or holes in the material. Forged parts are not to have any air spaces between the grains of powder, once forged to the minimum 7.8 g/cm3. The two main reasons for porosity are excessive water or moisture on the tooling and not enough material in the preform. When there is excessive moisture, it forms steam and gets trapped in the air spaces in the preform. As the part is forged, the steam gets compressed but still keeps the edges of grains of powder from touching, causing pores. In the second case, there is not enough material to fill the die cavity and the material doesnt move much, leaving the pores of the preform undisturbed. Decarb occurs when carbon is removed from the microstructure. Decarb results when the hot part is not surrounded by a protective blanket of nitrogen for an extended period of time. This could happen if air leaks into the furnace, or if there is a delay in the cycle time from exit of furnace to forging. Oxide penetration is common for powder-forged rods, where a layer of oxide forms on the rod. It is the scale of the part, which is formed from the warm part in atmosphere air. Oxide penetration is when some of this scale penetrates into the interior of the part. This condition is usually found with porosity. Excessive moisture on the tooling, a crack in the preform, not enough material in the preform, or a piece of extra material falling into the die cavity getting forged into the part are the prime causes of oxide penetration. Stage IV: Shot peening: is implemented after forging. Before shot peening, some connecting rods receive primary milling/rework. Bad connecting rods are thrown to scrap baskets.

17

Stage V: Fracture splitting process: The procedure for the separation of the cap from the connecting rod (Figure 2.5) was explained by Weber (1991). A notch is optimized during compacting process and molded into the preform bore at the location where separation of the cap from the rod is desired. The notch is closed during densification by forging to a fracture-initiating crack, which is defined exactly with respect to position and depth. The cap is split from the connecting rod by applying hydraulically generated force inside the bore. The fracture starts from the two crack initiating notches. This process does not require further machining, because fracture faces perfectly fit each other. Stage VI: Inspection: During this process presence of any cracks and flaws are inspected. Inspection of any flaw on the surface can be observed by white light. Magnetic particle inspection is done randomly to check for hidden cracks, fillet or inner contour machining, jamming or forging marks. Finally, the rods are inspected for weight and length, and any overweight connecting rods are rejected. Magnetic particle inspection is used to inspect for cracks (Jinka et al., 1997).

2.3 Die-Casting
Die-casting is accomplished by forcing molten metal under high pressure into reusable metal dies. Cast connecting rods were introduced in 1962. Riser is designed to achieve directional solidification by tapering the central web so that solidification begins at a point remote from the risers, in this case the center of the thin arm. Directional solidification is achieved from the center of the arm to each end despite gating from the small end. The I-beam section of the arm provides additional chilling, so that solidification begins at the arm center where it is critically stressed and travels to the

18

risers. Cooling fins are added to the boss and column rail to provide better local solidification. A match plate shows the die casting of connecting rods (Figure 2.6). After trimming and machining, all connecting rods pass through conveyorized sonic, ultrasonic, and magnaglow inspection stations, where defective castings are removed from the production lines (Niebel et al., 1989). After casting few machining steps are shown in Figure 2.7. Die casting machines, large or small, vary fundamentally only in the method used to inject molten metal into the die. The rapidity of operation depends upon the speed with which the metal can be forced into the die, cooled and ejected, the casting removal, and the die prepared for the next shot. These are classified and described as either hot or cold chamber die casting machines (www.diecasting.org).

2.4 Comparison of Forged Steel and Powder Metal Connecting Rods

The two most competitive high volume manufacturing processes of connecting rods are forged steel and powder metal processes. There has been a significant increase in the production of powder metal connecting rods in North America in the last decade. The main driving force for this trend has been cost effectiveness of PM connecting rods resulting from near net shape manufacturing as well as fracture splitting of the cap from the rod, introduced in 1990. Near-net shape achieved in powder metal forged connecting rods results in substantial reduction in the material used to make them. In spite of the substantially lower weight of the material used, however, the cost of the powder forged roughstock could be higher than that for the conventional hot drop-forged roughstock,

19

because of additional operations of powder formation, preform formation, presintering, and sintering (Gupta, 1993). Figure 2.8 summarizes the manufacturing processes for conventional as well as splittable forged steel connecting rods. As shown in this figure, the conventional connecting rods were originally forged as one piece from high carbon wrought steel. They were then sawed to separate the cap from the rod. An alternative manufacturing process shown in Figure 2.8 is for the rod and the cap forged separately. For both of these methods extensive machining is required to mate the cap to the rod. The most recent process used for forged steel connecting rod is the use of splittable steel. In this process, most of the machining processes are eliminated (see Figure 2.8). Figure 2.9 summarizes the manufacturing steps for the powder metal connecting rod. By comparing the steps in manufacturing processes in Figures 2.8 and 2.9, it can be seen that the fracture splitting step is the main difference between conventional forged steel and powder metal connecting rod manufacturing processes. Table 2.1 summarizes the cost breakdown of manufacturing processes for the conventional forged steel and powder metal connecting rods. This table indicates 57% higher cost in machining of the forged steel, as compared with the PM connecting rod (i.e. $3.32 for the forged steel versus $2.11 for the PM connecting rods). Clark et al. (1989) indicate that the machining steps for the case of the connecting rod and cap are identical, irrespective of whether they are forged separately or as one piece. However, with recent introduction of new materials such as C-70 splittable steel, this key advantage of powder metal connecting rods no longer exists, as machining of matching surfaces of splittable steel is no longer required. Similar to the powder metal

20

connecting rod, fracture-splitting method for the C-70 steel connecting rod is a mechanical method in which the big end is fractured by splitting, hydraulically, from two notches at the crank end. As mentioned in the literature review, fracture-split technology as applied to forged steel connecting rod cuts the total cost by 25%, compared to the conventional forged steel connecting rod (Repgen, 1998). The C-70 steel has higher strength than the powder metal material. Other alternative steels are also being developed with higher fatigue strength. Micro-alloyed 36 MnVS4 steel shows better fatigue strength than C-70 steel. For this new micro-alloyed steel the component tests on materials showed 15% to 20% increase in fatigue strength. Also, there was increase of tool life for this new steel (Repgen, 1998). This allows providing the benefits of both higher strength and lower manufacturing cost, simultaneously, as compared to powder metal connecting rods. As a result, some automotive manufacturers are starting to switch back from PM to forged steel connecting rods.

21

Table 2.1: Summary of cost breakdown of manufacturing processes for conventional forged steel and powder metal connecting rods (Clark et al., 1989).

Forged steel
Process Building Raw material Heating Sizing Forging Trimming Coining Heat Treating Shot Blast Shearing Inspection Machining Total $/part $0.02 $0.84 $0.18 $0.04 $0.45 $0.13 $0.11 $0.10 $0.04 $0.04 $0.09 $3.32 $5.36 Percent 0.40% 15.60% 3.40% 0.80% 8.30% 2.40% 2.00% 1.80% 0.80% 0.80% 1.60% 62.00% 100.00%

Powder metal
Material Blending Compaction Sintering Forging Machining Building Sawing Inspection Total $0.9074 $0.0789 $0.5073 $0.3695 $0.7861 $2.1162 $0.1176 $0.0703 $0.0843 $5.0376 18.01% 1.57% 10.07% 7.33% 15.60% 42.01% 2.33% 1.40% 1.67% 100.00%

22

Figure 2.1: Schematic of drop-forged process of connecting rod (Todd et al., 1994).

Figure 2.2: Schematic of fracture splitting jig for C-70 steel (Fukuda and Eto, 2002).

23

Figure 2.3: Schematic of connecting rod cracking machine. Ultimate precision's rod fracturing machine can fracture powdered metal or C-70 type forged steel rods at ambient temperature (www.upinc.net/excello.htm).

24

Figure 2.4: Schematic of tooling configuration for powder-forging of connecting rods (Weber, 1993).

Figure 2.5: Schematic of fracture splitting process for powder forged connecting rod (Weber, 1993).

25

Figure 2.6: Schematic of plate for the cast connecting rod (Niebel et al., 1989).

Figure 2.7: Schematic of manufacturing process of cast blank connecting rod (Weber, 1993).

26

Raw material Cutting to length of material

Temperature Heating of billet

Hot forging Trimming, piercing, coining etc. Heat treatment Shot peening/blasting
Forged (one-piece)
Machining of outer contours Deburring of outer contours Preliminary grinding of side faces Broaching of pin and crank end bearings Machining of cap bolt bores Drilling of oil bores Machining of bolt head and nut seats

Crackable steel (i.e. C-70, microalloyed steels)

Machining
Rod/cap forged seperately
Machining of outer contours Deburring Broaching of crank and pin end bearings Machining of cap bolt bores Drilling of oil bores Machining of bolt head and nut seats

Straightening

of

rod

and

surface finishing of both cap and rod

Separation of cap from rod

Sawing to separate the cap from the rod Grinding of rod/cap contact surfaces Alignment of the cap with the rod

Fracture splitting

Milling cap and rod to required width

Straightening of rod and surface

Assemble of rod
Insertion of bolt sleeves Finish-grinding of side faces Final drilling of small and big end bearings Milling of bearing shell positioning grooves Honing of crank end bearing Insertion of pin end bearing bush Insertion of crank end bearing shells Machining of balance pads (not required for splittable steel)

Inspection for cracks and flaws

Visual Inspection (surface for cracks, corrosion, pitting, galling or other damage) Weight/length Inked

Testing
Chemical analysis ? Microstructure observations ? Hardness measurements ? Mechanical strength measurements ? Fatigue tests ? Magna tests ? Coining ?

Figure 2.8: Summary of forged steel connecting rod manufacturing processes.

27

Powder stage or mix Compacting Delube Sintering

Pressure Density Die-system Wax out Oil out

Temperature Atmosphere (no oxidizing) Die-Temperature Forging time Forging dies Gas (cooling after forging) Deflash Primary milling/Secondary milling Rework

Forging

Shot peening
Machining of cap bolt bores Drilling of oil bore

Fracture splitting Assembly

Finish-grinding of side faces Final drilling of pin and crank end bearings Milling of bearing shell positioning grooves Honing of big end bearing Insertion of small end bearing bush Insertion of big end bearing shells

Inspection for cracks and flaws

White light Inked Magnetic particle inspection Weight/length

Testing

Chemical analysis ? Density measurements ? Microstructure observations ? Hardness measurements ? Mechanical strength measurements ? Fatigue tests ?

Figure 2.9: Summary of powder metal connecting rod manufacturing processes.

28

Chapter 3 LITERATURE REVIEW

Many recent papers in the literature indicate the resurgence of interest in the use of lightweight connecting rod materials and designs, inertial force reduction, increased speed, and cost effective methods of manufacturing. The literature survey shows a variety of selective information and comparison of connecting rods in the areas of load analysis, stress analysis, durability, manufacturing, economic/cost analysis, and optimization. Approximately 50 studies have been reviewed. These studies are sorted by loading, stress analysis and fatigue studies (Table 3.1), optimization studies (Table 3.2), manufacturing process and economic studies (Table 3.3), and other studies (Table 3.4).

3.1 Load and Stress Analyses

Sonsino (1996) analyzed the loading experienced by connecting rods. Connecting rods are submitted to mass and gas forces. The superposition of these two forces produce the axial force, which acts on the connecting rod. Connecting rods also experience bending moments due to eccentricities, crankshaft and rotational mass forces. Axial loading can be calculated by the knowledge of engine pressure and rotational speed, whereas bending moment can be determine by strain analysis in an engine. Stress analysis of critical areas and maximum local stresses of powder-forged connecting rods were also discussed by Sonsino (1996). The critical areas are the transition regions from the small end to the shank and from the shank to the big end.

29

Table 3.5 shows the local stress amplitudes obtained for the critical areas. The R ratios were calculated for the most severe conditions with a gas pressure of p = 660 kPa, rotational speed of 6800 rpm, and an operational load amplitude of Fa = 19.2 kN. Because of mass distribution, different mean loads, and therefore R-ratios, for particular areas of the connecting rod were obtained, as listed in Table 3.5. The magnitude of piston side load forces and dynamics errors for the two-lump mass connecting rod model and actual rod model were discussed by Shiao and Moskwa (1993). They concluded that an error in the moment of inertia of connecting rod using this model could predict incorrect engine forces and dynamics. With the actual connecting rod model, the inertia forces of the distributed mass model become more complicated to analyze. The small end of the rod has a reciprocating motion along the bore, whereas, the big end of the rod moves in rotational motion. However, the center of rod mass path describes an ellipse, which makes it difficult to analyze. Therefore, a twolump mass connecting rod model was introduced for simplification. The connecting rod is divided into reciprocating and rotating masses so that the distributed connecting rod mass is replaced by lumped masses at each end located at the bore center. The following equations have to be satisfied to make the rod model equivalent to the actual rod model: mc = mc,a + mc,b mc,a L1 = mc,b L2 Jc = mc,a L12 + mc,b L22 + Jf (3.1) (3.2) (3.3)

where mc is the mass of connecting rod, mc,a and mc,b are the connecting rod reciprocation and rotating masses, L1 and L2 are the distances from connecting rod mass center to

30

piston pin center and to the crank pin center, Jc is the actual moment of inertia, and Jf is the inertia error. For a single cylinder engine, there are three moving parts for transmitting engine force: piston system, connecting rod system, and crankshaft system. Figure 3.1 shows a free body diagram of forces on a piston, as well as dynamic equations. Figures 3.2 and 3.3 show free body diagrams and dynamic equations of a connecting rod and a crankshaft, respectively. Typical key engine parameters such as firing pressure, rpm and piston weight for a 4-cyclinder engine are shown in Table 3.6 (Folgar et al., 1987). The distribution of inner pressure in tension can be estimated by cosine function. Figure 3.4 shows the loading distribution experienced by connecting rod in tension and compression. FEA analysis conducted by NASTRAN on full connecting rod indicated small axial displacement in the shank (Folgar et al., 1987). Tilbury (1982) measured axial forces and bending moments within the connecting rod by transducers. The study compared the theoretical calculations with the experimental results of three tests. It was concluded that the experimental accelerations and axial forces were in close agreement with the predicted results. However, the bending moment distribution and the experimental results did not agree due to the bending moment influenced by the inertia effects of the linkage, viscous drag on the linkage as it was rotating through the engine oil in the sump, and friction effects at the gudgeon pin. A study to develop dynamic stress analysis using FE techniques and stress-time history generations is discussed by Majidpour et al. (2002). The force due to the gas load on the piston is a function of crank angle and engine speed. A plot of cylinder pressure

31

versus crank angle diagram is shown in Figure 3.5. The maximum force is attained at 370o crank angle. Inertia forces are composed of two parts. The first part is inertia of reciprocating masses and acts on the pin end and its direction changes with respect to the piston acceleration. The second part includes centrifugal forces, which act on the connecting rod in a distributed manner. The centrifugal force is normal to the longitudinal rod axis and produce bending stresses. The maximum moment occurs when the crank is perpendicular to the rod. The maximum tension and compression forces occur at the top dead center point during each cycle. It was observed that inertia load is proportional to the engine speed. Stress-time graphs at different r.p.m. speeds were obtained. It was concluded that the experimental methods restrict to measure the dynamic stress on the connecting rod because of the limited surface available for strain gauges to stick on the connecting rod body, while FE techniques were considered to be the best method to find connecting rod stress distribution during operation. Static analysis, quasi-dynamic analysis and design of a lightweight connecting rod were discussed by Sugita et al. (1990). Figure 3.6 shows the boundary conditions used for static finite element analysis under tensile load. The force acting on the piston end is due to piston and piston rings under maximum inertia force. In this case the area on the cap in contact with the crank pin is constrained. The press-fit stress at the piston end is obtained by applying the piston pin displacement. Figure 3.7 shows compressions of the maximum principal stress values obtained at the critical locations based on finite element analysis and strain gage measurements under static loading. In order to simulate actual engine operating conditions, the distributed inertia force was considered. Figure 3.8 shows the boundary conditions for this quasi-dynamic analysis. In this case the piston pin

32

end under goes both tension and compression load at each engine speed. The connecting rod was constrained from one end. Under these boundary conditions, the maximum stress amplitude and mean stress values were calculated. Strain gauges were also used to measure stresses in an engine and Figure 3.9 shows the mechanism used. Figure 3.10 compares the calculated and measured values and also shows the design limit line, which was obtained considering the fatigue limit of the material and a margin of safety. Webster et al. (1983) explained the loading of connecting rod in an engine. In this study the tension and compression were obtained from experimental results. For tension loading the crank end and piston ends were found to have a sinusoidal distribution on the contact surface with pins and connecting rod. Figures 3.11 and 3.12 show the load distribution in tension and compression, respectively. Figure 3.13 shows the deformation of connecting rod for compression and tension loadings with and without bolt pretension. It was concluded that the highest stress levels occurred in four locations: the upper area of the cap end on the axis of symmetry, the transition region of the bolt section and the lower rib, the transition region of the lower rib and the shaft, and the connecting rods bolt head. Table 3.7 shows the magnitude and location of maximum stresses. Maximum stress values reached 63.2 po for tension and 51.9 po for compression, where po is the normal pressure constant (i.e. peak pressure in the pressure distribution profile). In a study by Balasubramanian et al. (1991) finite element analysis was used to optimize the connecting rod model. The connecting rod load was broken down into various individual loads for the simulation and the actual stress was obtained by superposition. The individual loads were the inertia load, firing load, press fit of the bearing shell, and bolt forces. These individual load cases were analyzed on a case-to-

33

case basis: alternating load, lateral acceleration, buckling, and free-free vibrations. They state that using 2-D shell model ignores the fact that crank pin and piston pin deflect when subjected to firing or inertia loads, and that the load perpendicular to the plane of the connecting rod is not constant. With 3-D model it is possible to take account of bolt geometry. In both 2-D and 3-D models, the influence of lubrication oil film for firing and inertia loads were ignored. The inertia load is dependent on the engine speed and it is maximum at top dead center. In this case the connecting rod stretches as seen in Figure 3.14. The bolts and the bearing cap experience the maximum stress. A critical area in connection with inertia load is the mating surface of bearing cap and connecting rod. The reaction force distribution between bearing shell and crankpin in this case is shown in Figure 3.15. The firing load shortens the rod, as shown in Figure 3.16 and is decisive for the design of the connecting rod cross-section. The reaction forces associated with this load are shown in Figure 3.17. In this case the connecting rod shape in the area of the big end bearing is designed in such a way that the contact area between bearing shell and crankpin is as large as possible and this leads to a low contact stress in this area.

3.2 Mechanical Properties and Fatigue Behavior

Typical hardness for PF connecting rods range between 76 and 88.5 HRG (Chernenkoff et al., 1994). Olaniran and Stickels (1993) investigated a new crackable alloy forged steel for connecting rod application. Specimens of the new alloy and 1151 steel for charpy impact testing were machined and tested. The hardness of this new alloy was in the range 77-82 HRG. It was found that the fracture separation of caps is feasible

34

if the carbon content of the rods is increased to reduce the ferrite content for the developed steel alloy. In order to maintain the hardness, the manganese content must be varied inversely with the carbon content. When the hardness is maintained constant then the machinability is not greatly affected by the reduction in ferrite content. A primary design criterion for the connecting rod is endurance limit. Imahashi et al. (1984) reported that the factors, which effect to fatigue strength in PF connecting rod are hardness of the material, depth of decarbuized layer, metallurgical structure, density, and surface roughness. They conducted constant amplitude, load-controlled component axial fatigue tests on PF connecting rods. The fatigue properties were compared to SAE 10L55 steel. They concluded that hardness has a large impact on fatigue strength. Figure 3.18 compares the fatigue strength of the two connecting rods. Constant amplitude, loadcontrolled axial fatigue tests were conducted on trail PF connecting rods by Illia and Chernenkoff (2001). The load ratios of -1 and -1.87 were used. Research on development of lightweight connecting rods based on fatigue resistance analysis of a microalloyed steel was conducted by Kuratomi et al. (1990). Rotating bending, load-controlled fatigue tests on notched and smooth round specimens were conducted. The study concluded that the microalloyed steel exhibits lower fatigue strength than the quenched and tempered steel for smooth specimens, but equivalent or higher fatigue strengths for notched specimens. The fatigue limit of smooth specimens decreases by increasing temperature. The fatigue limit increases for notched specimen as temperature increases to the same level as that of smooth specimens (see Figure 3.19). The grain size increases by increasing temperature, and when the grain size becomes larger, notch sensitivity of fatigue properties is not observed. This is attributed to the

35

relation between the effective grain size and notch depth. When the effective grain size is the same as or larger than one half the notch depth, notch sensitivity decreases. Figure 3.19 shows the effect of temperature on fatigue limits for three steels used in this study for comparison. SV40CLI microalloyed steel and S40C quenched and tempered conventional steel used for connecting rods were also compared. The study concluded that microalloyed SV 40CLI steel connecting rods exhibit higher fatigue strength than quenched-tempered S10C steel connecting rods and are 10% lighter in weight (Kuratomi et al., 1990). Suzuki et al. (1988) described the development of a powder metal sintered connecting rod. Fatigue tests on the trial connecting rod were carried out under tensilecompressive loading with R ratio of 5. The fatigue test results are shown in Figure 3.20. It was found that newly developed as-sintered connecting rod made of SUMIRON 4100S powder had sufficient fatigue strength, but not equivalent to conventional powder forged connecting rods. The fatigue test rigs used are shown in Figure 3.21. Marra et al. (1992) conducted fatigue testing of powder-forged and forged steel connecting rods under axial tension/compression at R = - 1.38 and ten million cycles. It was observed that the fracture locations of the powder-forged rods were near the wristpin and crankpin end bore due to fretting corrosion. Typical forged steel rod fracture was at the I-beam section. The test results were statistically analyzed based on a Weibull distribution. Based on a medium ranking the analysis indicates that the B-10 and mean life of the powder forged rod test population are statistically improved over the wrought rod test population (Figure 3.22).

36

High fatigue strength free machining microalloyed steel was developed and used for connecting rods (Nakamura et al. 1993). The influence of alloy elements such as C, Mn, Cr, V, S, Pb and Ca has impact on fatigue strength and machinability. A 0.33%C1.05%Mn-0.5%Cr-0.12%V-0.55%S-0.20%Pb-Ca composition was found to be the best composition to improve fatigue strength. The tensile/compression fatigue test on prototype connecting rod showed 26% higher fatigue strength than conventional free machining microalloyed steel. A 15% weight reduction resulted without any reduction of mechanical and fatigue strengths. Beretta et al. (1997) investigated fatigue performance of the connecting rods made of either cast iron or hot forging carbon steel. They state that if a connecting rod working in a car engine is subjected to bench test loading conditions, the different areas of the connecting rod are subjected to peculiar load spectra with different stress ratios. The most critical areas, from the fatigue damage view are the shank and the two ends. The I-beam is mainly subjected to alternating load cycles with R -1. The big-end and small-end have machined surfaces and are subjected to pulsating loads with R > 0. Figure 3.23 shows the fatigue cycle on the connecting rod. The prototype connecting rods made of 0.43% carbon steel and of type-B cast iron were tested in order to obtain axial fatigue data. The fatigue test results are shown in Figures 3.24 and 3.25 for carbon steel and cast iron, respectively. It is shown that cast iron has greater dispersion of the data, compared to the narrow scatter band for the carbon steel. Beretta et al. (1997) described the impact of inclusions or inhomogeneities on the fatigue strength of both cast iron and carbon steel connecting rods. The fracture surface of carbon steel conrods revealed that the defects were fold and cavity, due to forging

37

along midplane of the conrods. The microscope examination of polished surfaces revealed internal defects in the forms of faceted cavity or small oxide flaws. The cavities occurred near the surface (Sub-surface microcracks and microfolds with a depth of 30 m). Fracture surface of cast iron conrods revealed that the defects were either internal microshrinkage cavities or inclusions. Typical dimensions of defects were in the range of 300-2000 m. The impact of decarburization on fatigue life was investigated by Illia and Chernenkoff (2001). Constant amplitude, load-controlled axial fatigue tests were conducted on shotpeened connecting rods with a frequency of 170 Hz and at a stress ratio of -1.50. The connecting rods failed in the minimum cross-section of the I-beam. They concluded that decarburized layer depths up to 0.4 mm do not impact fatigue life of PF connecting rods. The decarburized layer depths higher than 0.4 mm decreases the fatigue life. Olaniran and Stickels (1993) investigated the cap splitting process for a new forged steel alloy. The connecting rods had 45o notches, about 0.5 mm deep, machined on the sides of the bore and on the edges of the rod in the same plane as the bore notches. During production, caps were separated from the rods. Fatigue testing of 1151 steel and newly developed forged steel rods were performed in a resonance machine in a stair case manner with fully reversed axial loading (R = -1). Tests were suspended after 107 cycles. The test data as summarized in Figures 3.26 and 3.27 show a dispersion of data for 1151 steel and new developed alloy, respectively. The load amplitude of 8700 lbs were used for both rods. It was found that rods were not shot peened adequately and the forging

38

scale was not removed on the I-beam surfaces prior to peening. It was concluded that by reducing ductility, fracture separation of new forged rods and caps become feasible. In hot forging, the directional solidification of grains are desired for better fatigue strength. In PF parts, there is no strengthening effect obtained by hot and cold work or any directional solidification (www.forging.org). A study published by Rabb (1996) investigated anisotropy effect of the material. Smooth test bars were cut from forgings in the same direction as the grain flow and perpendicular to it. However, there was no clear indication of anisotropy effects.

3.3 Density Distributions and Dimensional Accuracy

The distribution of density in PF connecting rod varies. Figure 3.28 shows the density distribution in PF connecting rod (Araki et al. 1993). The typical density of PF material is 7.85 g/cm 3 , which is over 99% of the theoretical density as reported by Imahashi et al. (1984). A 1% oversize density in preform can lead to overcharge and rupture the PF tooling (Weber, 1990). Mechanical properties such as tensile strength and fatigue strength of powder forged material increase with density. Figure 3.29 shows the effect of density on mechanical properties of 4640 powder forged steel and Figure 3.30 shows the effect of density on fatigue strength of the same powder forged steel (Imahashi et al. 1984). OBrien (1988) investigated fatigue properties of PM materials. He mentioned that the single most effective method of improving fatigue properties is to increase the part density. The study provides close statistical estimates of the fatigue endurance limits of commonly used PM steels.

39

The dimensional accuracy of powder forgings depends on forging press accuracy, forging die accuracy and preform design and compacting accuracy. For PF connecting rod the tolerance of 0.5 to 1.0% is typically used (Araki et al., 1993). Investigation of metal flow and preform optimization in flashless forging of a connecting rod was performed by Takemasu et al. (1996). The study shows that the material flow and the dimensional accuracy in the initial stage depend on geometry of the forging, initial and intermediate billet geometry, percentage of flash, geometry of the flash opening, and heat transfer between the tooling and the billet.

3.4 Shot Peening and Residual Stress Effects

Shot peening creates a layer of compressive stress on the surface of the connecting rod. The shot peening process usually decreases the tensile stress and increases the endurance limit. Shot peening of connecting rod has the biggest impact on the fatigue life. The effect of shotpeening on PF material used for connecting rod is described by Chernenkoff et al. (1995). They further explained the process parameters for precision peening. These include shot diameter, shot distribution, shot shape, shot hardness, shot type, shot density, shot impact angle, shot flow rate, shot blast pattern, nozzle-workpiece position relationship, and exposure time. Comparisons of shot peened and unpeened specimens were conducted and was found that the fatigue strength increases from 262 MPa for unpeened specimens to approximately 407 MPa for peened specimens at 20A. It was concluded that the maximum fatigue strength occurred at a shot peened intensity in the range of 17-20A. Figure 3.31 shows a typical residual stress distribution for 20A intensity. The depth of peening surface for 20A intensity was

40

approximately 0.37 mm. Any further increment can result in decrease of fatigue life. Comparisons of shot peened and unpeened connecting rods fatigue strength are shown in Figures 3.32 and 3.33. The optimum peening intensity of powder-forged connecting rods resulted in an improvement of 27% of the fatigue strength over the unpeened rods. Overpeening can form cracks at the surface of the connecting rod (Chernenkoff et al. 1995). The effect of shot peening intensity on the fatigue strength is shown in Figure 3.34. For shot peened and upeened connecting rods tested at medium to high strain amplitudes, the typical failure initiation was at the surface. For shot peened connecting rods tested at low strain amplitude failure initiation sites were in the subsurface, where fatigue failures initiated approximately 1.5 mm below the peened surface.

3.5 Optimization Studies

The FEM offers the possibility to design the entire manufacturing process on computer. This leads to reduction of the cost and time in process and tool design, tool manufacturing, and die try out. It makes it possible to iteratively modify the process conditions in the simulation to find the best design condition for a connecting rod (Takemasu et al. 1996). An investigation of metal flow and preform optimization in flashless forging of a connecting rod was conducted by Takemasu et al. (1996). The preform shape was optimized through FEM simulation results. To optimize the preform design, the connecting rod was divided into three parts consisting of small end section, large end section, and I-beam section. Simulations were performed independently for each section.

41

For the large end the initial geometry and the volume distribution of the preform were optimized at the same time by changing the position and the diameter of the hill section under a condition of constant volume. It was concluded that optimization of the geometry of the big end is more difficult due to the fact that the shape of this part is very complex and the material flow is strongly influenced by the initial geometry of the preform. The deformation pattern of the small end is not as sensitive to the initial geometry of the preform as the large end. The initial volume distribution was mainly controlled to fill the die cavity at the end of the punch stroke. The deformation pattern of the connecting rod Ibeam section indicated deformation under plane strain conditions, but the side wall was designed to have enough height after deformation at the end of the stroke. The ribs of the I-beam were filled at the end of the simulations. A study by Sugita et al. (1990) used boundary element method to reduce the weight of the connecting rod. The connecting rod was designed by incorporating a thin Ishaped section column and adopting the two-rib design to the big end. The weight was reduced approximately 20%. The boundary element method results verified that it is possible to estimate stress distribution of the connecting rod under actual engine operating conditions. This system was found to be effective in weight reduction and to allow flexibility in design analysis. Figure 3.35 show the stress values based on the design limit. It was concluded that the stress values are below the design limit (Sugita et al., 1990). Hippoliti (1993) performed 2D and 3D FEA on the connecting rod and reported the time and cost reduction to improve the design. The 2D and 3D models were compared. The 2D model is based on plane stress and requires less modeling effort and

42

analysis, but the fillet effects cannot be obtained. The change in width at various locations of the connecting rods was taken into account by changing the thickness of the four node plane stress elements. The rollers were evaluated by using the Hertizan theory. It was concluded that the 2D analysis is more robust than 3D analysis, that the results from 2D analysis are adequate, and that the CPU time to analyze a 2D model is about 1/5 of a 3D model. The total mass of the connecting rod was optimized based on FEA models with respect to von Mises stresses at critical locations. To avoid obtaining too slender of a design, buckling was checked by utilizing Johnsons buckling formula. Buckling strength was discussed by Kuratomi et al. (1995). Development of lightweight connecting rod made of a low-carbon martensite steel from the standpoint of high fatigue strength and buckling was considered in designing the connecting rod. Buckling is an important issue in optimization and is influenced by many factors including hardness, coining ratio and section area. Shape optimization of an engine connecting rod was investigated by Yoo et al. (1984), where thickness distribution and shape were used as design variables. It was assumed that the rod is in a plane stress state. Triangular elements were used in the FEA. The boundary conditions were denoted by and as shown in Figure 3.36. 1 and 2 are boundaries at which connecting rod touches the crankshaft and piston pins. For the analysis these two variables were kept fixed. 3 and 5 of the shank and the neck regions of the rod were to be determined. 4 and 6 for the shape boundaries were kept fixed. To satisfy the boundary conditions, the movement of the connecting rod was constrained in the X1 direction, as shown in Figure 3.37. Lower and upper bounds were imposed on principal stresses of inertia and firing loads. Steepest descent algorithm was used to

43

obtain the optimum solution. Figure 3.38 shows the optimum design after 15 iterations. This optimum design resulted in 25% weight reduction, which was achieved in the neck region at a reasonable computational cost. A study by Geonka and Oh (1986) investigated the optimum design of connecting rod bearings. The bearing performance was improved based on the oil film thickness and the oil film pressure. The objective was to maximize film thickness and minimize pressure and power loss. Lubrication analysis programs were used. Because of the connecting rod symmetry only one-half model was designed. The study was performed at an engine speed of 2000 rpm. It was found that the peak pressure occurs at about 20o crank angle after top end center in the mid plane of the bearing of the connecting rods big end. Usually, the minimum film thickness is also located in the midplane. Therefore, midplane distribution was examined. FE models were created to support the connecting rod in region B. Figure 3.39 shows different design modification. Figure 3.39 also shows the final optimal design, which was considered to be a compromise between good bearing performance and acceptable stress value after eight different design iterations. The maximum octahedral shear stress of 154 MPa was 10% increase over the conventional design. The minimum film thickness obtained was 2.32 m and the maximum oil film pressure was 77 MPa. The friction for the optimum design was 7% below the conventional design. It was concluded that the optimum design has little effect on power loss, which was improved by increasing the film thickness by keeping the same length to width ratio.

44

3.6 Material Selection, Manufacturing Process and Economic Cost

Connecting rods are manufactured by casting (nodular or malleable cast iron), hot forging (unalloyed carbon steels or micro alloyed steels), and powder forged and sintering. Comparisons of all three major processes is discussed in Table 3.8. The use of lighter metals such as aluminum, magnesium, titanium, and plastics are also under consideration. The application of titanium connecting rods is costly and limited to racing cars. The mechanical properties and fatigue strength are closely equivalent to the tempered and quenched medium carbon steels. The titanium rods have a tendency for surface oxidation during hot forging and high wear in direct contact with steel; which are usually overcomed by a thermal molybdenum coating spray. Aluminum metal matrix rods have one-third the weight of the steel rod. They have one half the thermal expansion, but twice the yield strength and wear resistance. AlLi alloys are under evaluation for better strength, stiffness and fatigue resistance. The carbon fiber composite connecting rods provide 40% lower weight but suffer from inadequate fatigue strength and tensile properties. High costs for the complicated procedures of fiber winding and the continual high matrix cost for fibers are the biggest disadvantages of this materials application (Gupta, 1993). Kuratomi et al. (1995) outlined the mass production process for a new lightweight connecting rod made of low carbon martensite steel (Figure 3.40). For this process the major change was the installation of new press machines (quenching and tempering), magna test for hardness and shot peening equipment. This manufacturing technology was

45

adopted for a newly developed connecting rod used in Nissans new V6 twin-cam VQ engine. Banerji (1996) investigated the application of microalloyed forging for heavyduty diesel connecting rods. Figure 3.41 summarizes the various activities that were necessary to overcome the hurdles and make MA work for such connecting rods. Forged MA steel connecting rods are used for the most demanding, high volume, and higher horsepower diesel engines. A study by Araki et al. (1993) published the forging process of a trial powder forged connecting rod. Figure 3.42 shows the production process. Powder is mechanically mixed and packed into a powder die where it is compacted. The green compact is sintered at 1120 oC. The sintered compact is heated to 1000 oC. The preform is transported into forging die at 300 oC. Then the temperature of forgings is lowered to room temperature. After heating, the forgings are tempered and normalized at 570 oC. The production cost of connecting rods play a major role in the global competitive market. Beside the machining process for powder metal connecting rods, the following operations were also compared with forged steel connecting rods by Gupta (1993): tool costs, on-line labor cost, set up cost, tool change cost, scrap cost, rework cost, inspection cost, inventory cost, amortization of equipment eliminated, maintenance, repair and rebuild of equipment eliminated. The comparative study by Imahashi et al. (1984) shows that the major cost difference between hot forging and PF is in machining (Figure 3.43). Figure 3.44 shows that the energy saving for PF connecting rod is also due to the machining process. As reported by Imahashi et al. (1984), the energy required by PF process is one half of the

46

forged steel. Figure 3.45 shows the weight of steel forging and PF connecting rod raw materials. In spite of the higher raw material weight for the steel forged connecting rod, the raw material cost is significantly lower than the PF connecting rod raw material. The machining savings come from the fracture splitting process introduced in 1990 by PM connecting rod manufacturers (Edser, 2001). However, with introduction of new crackable forged steel connecting rods, this advantage of powder metal connecting rods no longer exists. Olaniran and Stickels (1993) published the separation of caps and rods by fracture splitting forged steel. They developed new forged alloy, which can be cracked as powder forged connecting rod. The typical microstructure of 1151 air-cooled steel rods contains 10-20% ferrite. For connecting rods, the analysis of total cost must include both the rough part and the machining cost. Sintered connecting rods achieve the near net shape but manufacturing process is much more expensive than the forged steel. The material strength is also a major requirement for the connecting rods, which favors forged steel. Sintered connecting rods always show a certain level of porosity that varies from part to part. The sintered connecting rods are less reliable in meeting machined specification (Repgen, 1998). The cost-benefit analysis by engine manufacturers has shown that the introduction of the fracture-split technology cut the total cost by 25%, when compared with conventional forged steel connecting rod. Other alternative steels are being developed with higher fatigue strength. Micro-alloyed 33 MnVS4 steel show better fatigue strength than C-70 steel. The fracture force was the same as for C-70 steel. For this new micro-

47

alloyed steel the component tests on materials showed 15% to 20% increase in fatigue strength. Also, there was an increase of tool life for this new steel (Repgen, 1998). Within the last decade powder metal has dominated the production of conrods in North America and Europe. However, introduction of new crackable C-70 steel and other similar steels has started to reverse this trend and some automotive manufacturers are switching back to forged steel connecting rods for their higher performance.

3.7 Summary
Several aspects of connecting rods in the areas of load and stress analysis, durability, economic and cost analysis, and optimization were reviewed. Each study showed fatigue behavior and manufacturing cost as central to the process of design assessment. A primary design criterion for the connecting rod is endurance limit. Failures of connecting rods are often caused by axial and bending loads acting perpendicular to the axes of the two bearings. Typical fracture locations of the powder metal rods are near the wristpin and crank pin end bores, while typical forged steel rod fracture is in the I-beam section. The typical density of powder metal connecting rod is 7.85 g/cm3. Powder metal connecting rods show porosity, oxide penetration, decarburization or other defects, which is typical of powder metal material. The depth of decarburization should not exceed 0.4 mm. Shot peening with optimum peening intensity improve fatigue strength by 27%. Over peening can form cracks at the surface of the connecting rod. Majority of the fatigue cracks start subsurface, due to compressive residual stress induced by shot peening. In powder metal connecting rods, shot peening removes the oxides on the rod surface, but

48

the oxides entrapped in the channels and pores, below the surface remain, causing the crack to originate below the surface. Tension loading at the crank and piston ends are found to have sinusoidal distribution on the contact surface. The compression loading at the crank and piston ends are found to have a uniform distribution on the contact surface. The inertia load is dependent on the engine speed and it is maximum at top dead center. The most critical areas, from the fatigue damage view are the shank and the two ends. Because of mass distribution different mean loads, and therefore R-ratios exist for particular areas of the connecting rod. The I-beam is mainly subjected to alternating load cycles with R -1. The big-end and small-end have machined surfaces and are subjected to pulsating loads with R > 0. Both 2-D and 3-D finite element analysis approaches have been used to study a variety of issues such as to optimize manufacturing process design, and shape optimization of different light weight prototype connecting rods; FEA is also extensively used to determine the high stress regions and define the deformed shapes for various loading conditions. The 2D analysis is more robust than 3D analysis, the results are often adequate, and the CPU time to analyze a 2D model is about 1/5 of a 3D model. The intense competition in the marketplace has motivated the evolution of high performance components, which must be designed and fabricated economically. The most common process technologies for the bulk production of connecting rods are closed die drop-forging and powder-forging. In spite of higher cost of powder metals raw material and lower fatigue strength, many North American automotive manufacturers have switched from forged steel connecting rods to powder metal connecting rods. A

49

reason for this has been the cost of the machining operations to separate the rod from the cap and machining of the matching joint faces. Powder metal connecting rods do not require this additional machining. Recent developments of crackable steels such as C70 and microallyed steels have eliminated the machining process associated with fracture surfaces and thus reducing production cost. Recent studies show that these newly developed connecting rods are more cost effective than either powder metal on conventional steel connecting rods.

50

Table 3.1: Summary of loading, stress analysis and fatigue studies survey
Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Three forces were discussed. Acceleration force Axial force Bending moment

Main objectives and conclusions

Tilbury [1982]

The prediction and measurement of axial forces, bending moments and accelerations in an engine connecting rod

The comparison of experimental and theoretical results of engine load and cylinder gas pressure

Webster, Coffell, and Alfaro [1983]

A three dimensional finite element analysis of a high speed diesel engine connecting rod

FS

Modeling aspects Boundary conditions Stress analysis

FEA

Transducers were fit to a connecting rod to measure the axial force and the bending moment. Three engine tests were conducted using combinations of engine load and cylinder gas pressure. The experimental and theoretical results of the three tests were compared. The experimental acceleration and axial force were in close agreement with predictions. However, the bending moment variation did not agree with the predicted result. Following reasons were given as contributing factors for the difference between measured and predicted bending moments. -Inertia effects of the linkage -Viscous drag on the linkage as it was rotating through the engine oil in the sump -Friction effects at the gudgeon pin A 3-D model was developed with rod bolt threads and pilots, rod bolt pretensioning, separation of cap end, maximum compression loading, symmetry conditions, and low aspect ratios. Boundary conditions used, were obtained from the experimental results. In tension, a sinusoidal distribution of load of the contact surface area, whereas, in compression, a uniform distribution over the contact area. The stresses found in the shaft and cap exhibited the beam and axial loading distribution. A complex region to perform analysis was rod bolt section as this region requires a complete understanding of the behavior. The isoparametric hexahedron finite element was eliminated because stresses predicted by this element and deformations of any regions formed by this element were questioned.

51
FS - Forged steel, FEA - Finite element analysis

51

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Specimen fatigue tests, loadcontrolled, constant amplitude Comparison of developed powder metal and hot forging connecting rods Component fatigue tests, tensile-compression, loadcontrolled, constant amplitude

Main objectives and conclusions

Imahashi, Tsumuki, and Nagare [1984]

Development of powder-forged connecting rods

PM

Product design

Tensile and fatigue strengths of PF increased by increasing the density after forging process. The density and hardenability for PF is the same as for SAE 1055 steel. Separation of caps in both hot forging and PF were compared. The tests show the factors which affect the fatigue strength are: hardness of material, depth of decarburized layer, metallurgical structure, density 7.85 g/cm3, and surface roughness. There was no difference found in hardness of both materials. Porosity was observed on powder forged surface but surface roughness is better than hot forging. A 0.3% sulfur was suggested to have a decrease in machinability and an increase hardness.

52
Folgar, Widrig, and Hunt [1987] Design fabrication and performance of fiber FP/ metal matrix composite connecting rods Fiber plastic metal matrix composite /casting Stress analysis 2D elastic FEA Fatigue testing Component fatigue tested to 10 million cycles under a cyclic loading of 6000 lbs Non-destructive testing of castings

Prototype con-rod was designed with the aid of a kinematic model and strain analysis was performed by using FEM. Prototype connecting rod survived more than 10 million cycles under cyclic loadings.

Ultrasonic and computerized industrial tomography techniques were considered as means to characterize defects in fiber metal matrix component.

FEA - Finite element analysis, PM - Powder metal

52

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Rotating-bending fatigue test at 3300 rpm with 10 million cycles (R = -1)

Main objectives and conclusions

The objective of the study was to provide close statistical estimates of the fatigue endurance limits of commonly used P/M steels. Statistical estimates of the 99.9% survival stress have shown that fatigue endurance ratio of commonly used P/M steels can vary from 0.16 to 0.47. The use of 0.38 as a rule of thumb for estimating the fatigue endurance limit from static tensile property data can result in large errors. The most effective method of improving fatigue strength is to increase the part density. It was concluded that the porous steels should be ideally fatigue tested from 50 to 100 million cycles rather than 10 million cycles, since some of the runouts were found to have stable crack growth. Sintering at high temperature increases the static tensile properties of the PM and partially prealloyed steels but cannot improve the fatigue properties of the partially alloyed steel because of loss of microstructural heterogeneity. To obtain the optimum shape for as-sintered connecting rod, finite element stress analysis was carried out. The connecting rod with tapered hollow in the beam region with one reinforcement was used, which lead to reduce the maximum stress. The tensile-compressive loading was conducted on the prototype as-sintered connecting rod. It was found that as-sintered connecting rod has sufficient fatigue strength but little less fatigue strength than that of forged connecting rod. As-sintered method expected to be more cost effective than PM forged due to elimination of the forging process.

OBrien [1988]

Fatigue properties of P/M materials

PM

Fatigue properties

53
Suzuki, Toyama and Konda [1988] Development of as-sintered P/M connecting rod for automobiles SUMIRON 4100s low alloy steel powder/oil atomized process (PM) Design concepts FEM elastic stress analysis Fatigue test Component, load-controlled axial fatigue test (R = -5)

PM - Powder metal, FEM - Finite element method

53

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Component, load controlled, axial fatigue test with R = -2, 20 Hz

Main objectives and conclusions

Luong [1997]

Rapid determination of fatigue strength of connecting rods using infrared thermography

The use of infrared thermography

The connecting rods were scanned by the infrared system at different stress levels. The load duration applied was 100 seconds corresponding to 2000 load cycles. The fatigue limit of connecting rods were observed by infrared thermography within a few hours instead of several months when using the standard staircase method. This paper concluded that the infrared thermography offers a non-destructive, noncontactable technique to observe metal degradation or damage. Dogbone shape specimens were machined from the powder metal forging and tensile properties were obtained. Longitudinal tensile properties were obtained using a single loading cycle. The criterion to be met by the connecting rod was to pass axial/compression (+33.3 kN/ 64.5 kN) fatigue test for ten million cycles minimum. The rod met the test requirements and was put in use in a V-8 engine.

Marra and William [1989]

Mechanical properties of a powder forged material for connecting rods

PM

Fatigue strength Tensile properties

Specimen tensile test Component, loadcontrolled, axial fatigue test at a frequency of 20 Hz

54

PF - Powder metal

54

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Rotating bending fatigue test of smooth and notched specimens

Main objectives and conclusions

Kuratomi, Uchino, Kurebayashi, Namiki, and Sugiura [1990]

Development of lightweight connecting rod based on fatigue resistance analysis of microalloyed steel

Microalloyed steel (SV40CLI) / FS

Fatigue properties

Study shows that the consideration of notch sensitivity is important for high fatigue strength material. Properties of microalloyed steel SV40CLI were compared with quenched and tempered S40C steel. The MA (V40CLI) steel exhibits slightly higher tensile strength but 0.2 % yield strength is 20% lower than the quenched and tempered steel. The elongation and reduction in area of the MA steel are much smaller than the quenchedtempered steel. MA steel connecting rods are 10% lighter in weight as compared to S40C steel. Specimen of (SV40CLI) microalloyed steel shows buckling strength 25% lower than the quenched and tempered steel. Therefore, forged parts with small surface notches show results that are different from the smooth specimens. By increasing temperature the grain size becomes larger and notch sensitivity of fatigue properties is not observed (When the effective grain size is the same as or larger than onehalf the notch depth, notch sensitivity decreases).

55
MA - Microalloyed, FS - Forged steel

55

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Fatigue tests under constant amplitude load-controlled with R = -3.22

Main objectives and conclusions

Whittaker [1991]

Sintered materialscan they operate under fatigue loads?

PM/Sintered

Fatigue properties of sintered material

Samples of connecting rod were tested under constant amplitude fatigue test (R = -3.32). Sintered materials showed low cyclic fatigue properties. This material showed lower notch sensitivity than the competitive materials. Crack propagation rate for sintered PM was higher than for the steel because of low ductility. This technology offered higher dimensional accuracy and light weight. Several machining operations can be eliminated during this process and thus saving cost.

56
Marra, Compton, and Skurka [1992] Fatigue testing of a powder forged connecting rod
PM and FS

Fatigue tests

Component axial fatigue test +46.7/-64.5 kN, R = -1.38

Powder forged and wrought steel connecting rods were fatigue tested. The intent of this analysis was to perform increased load fatigue test on both kind of rods. The tensile load was increase above +33.7 kN (the design requirement). The compressive load remains constant. These parameters were chosen based on ability to generate test fractures in less than 10 million cycles. It was concluded that the dominant fracture site of the powder forged rods were near the wristpin and crankpin. The wrought steel rods fracture locations were dominant towards I-beam. The test results were statistically analyzed based on Weibull distribution. Weibull analysis indicates that the B-10 and mean life of powder forged rods were statistically improved as compared to wrought rods. It was concluded that at elevated test loads, the powder forged rods showed better fatigue life to that of the wrought steel rods.

Weibull analysis

FS - Forged steel, PM - Powder metal

56

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Component axial fatigue test, load-controlled with R = -1

Main objectives and conclusions

Olaniran and Stickels [1993]

Separation of forged steel connecting rods and caps by fracture splitting

Hot forged SAE 1151 steel /FS

Development of a splittable steel alloy

Fatigue tests on connecting rods were performed in a resonance machine in a stair case manner with tests suspended after 107 load cycles. The separation of caps in FS is feasible if the carbon content of the rods is increased to reduce the ferrite content. To have the same hardness after air cooling from the forging temperature the Mg content of rods must be varied inversely with carbon content. The machinability of connecting rods is not greatly affected by reduction in ferrite content as long as hardness is maintained. Shot peening has big impact on the fatigue strength. It is believed that differences in shot peening practice account for the differences in strength dispersion. High carbon rods show a greater dispersion in strength.

57
FS - Forged steel

57

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading

Main objectives and conclusions

Nakamura, Mizuno, Matsubara, and Sato [1993]

Development of high fatigue strength steel free machining microalloyed steel for connecting rods

Micro-alloyed steel/FS

Mechanical properties Fatigue properties

Smooth specimen, rotating bending, load-controlled fatigue tests Component, load controlled, axial fatigue tests

High fatigue strength free machining microalloyed steel was developed to decrease the weight of connecting rod without the reduction of machinability. The comparison of the mechanical properties and the fatigue strength between developed steels and conventional steel were carried out. -Tension/compression fatigue test of 0.33%C1.05%Mn-0.5%Cr-0.12%V-0.055%S0.20%Pb-Ca composition showed 26% higher fatigue strength than the conventional steel. - New connecting rod was developed with the new material, resulting in 15% weight reduction.

58
Fatigue crack propagation Cyclic loading of Kmax= 500 N/mm-3/2 with R = - 0.5 Shiao and Moskwa [1993] An investigation of load force and dynamic error magnitude using the lumped mass connecting rod model Models of connecting rod: Actual connecting rod model Two-lump mass connecting rod model Dynamic equations based on Newtons second law of motion

The addition of chromium in connecting rod material improves the impact and the toughness of pearlite. This high toughness prevents crack initiation and propagation under cyclic loading.

It is very common practice to use two-lump mass connecting rod model to predict forces, inertia, acceleration etc. Newtons equations were used to predict forces. Five different engine connecting rods were used to determine the load and dynamics forces. The simple connecting rod model predicts incorrect mass inertia. It is helpful to understand the side force error in advance.

FS - Forged steel

58

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Specimen uniaxial fatigue test with constant amplitude, fully reversed strain-controlled mode

Main objectives and conclusions

Chernenkoff, Hall, Mocarski, and Pease III [1994]

Steel powders for high performance automotive parts

PM

Microstructure

The objective was to investigate the mechanical properties of four powder metals. The non-metallic inclusion content among four powders was the same. The tool life can be affected not only by the amount of inclusions but also by the kind of inclusions and their hardness. (Hard inclusion can cause excessive wear of the cutting tools and significantly increase the cost of manufacturing). The hardness at low end is 76-78 HRG and 86-88.5 HRG at high end is the specification for connecting rods. Cyclic properties are obtained by testing several specimens under uniaxial, strain-controlled conditions at different strain amplitudes.

Machinability

Fatigue test

59
Richter, Hoffmann, Lipp, and Sonsino [1994] Single-Sintered Conrods an illusion?
PM Fatigue test FEM

Component, axial fatigue test with R = -1

A prototype connecting rod was designed with the help of finite element analysis. 0.5% MnS was added into the material for better machinability. -The load was introduced via gudgeon pin in axial direction. -The fatigue strength of sintered connecting rod is equivalent to that of the currently used optimized cast connecting rods. Study concluded that failure occurs due to fretting corrosion in the small end, which can be avoided by optimizing the shrink fit between gudgeon pin and small end of the connecting rod.

PM - Powder metal, FEM - Finite elements method

59

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Component fatigue tests

Main objectives and conclusions

Richter, Hoffmann, Lipp, and Sonsino [1994]

Single-sintered PM connecting rods for passenger cars - an illusion?

PM/sintered

Fatigue tests

The objective was to implement the sintered connecting rods. The 0.5% MnS was added for better machinability. Lab fatigue tests and engine tests were conducted. The connecting rods failed due to fretting corrosion between gudgeon pin and small end, which can be avoided by optimizing the overlap in the shrinkfit. Fracture splitting method can reduce machining cost. Further improvement can be implemented by shot peening and high temperature sintering. Self-hardening PM alloys which can be further compressed in the single-sintering technology, can improve fatigue strength with lightweight. Powder particles did not fully achieve metal bonding, which promotes oxidation. Fatigue strength is affected highly by NSA. Shot peening improves fatigue strength. Development of PM connecting rod showed 30% higher strength and 10% lighter weight.

60
Takahashi, Hoshina, and Ogino [1994] Development of high strength powder-forged connecting rod PM Fatigue strength of connecting rods

Surface defects of PM were summarized as: Decarburized layer. Surface roughness. Non- sintered area (NSA)

PM - Powder metal, NSA - Non-sintered area

60

Table 3.1 Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Constant amplitude, strain-controlled specimen tests at frequencies of 0.510 Hz Constant amplitude, loadcontrolled component tests at a frequency of 180 Hz

Main objectives and conclusions

Chernenkoff, Mocarski, and Yeager [1995]

Increased fatigue strength of powder forged connecting rods by optimized shot peening

PM

Effect of shot peening on fatigue strength

Peened specimens resulted in an increase of 55% fatigue strength as compared with unpeened specimens. Typical failure initiation site for the unpeened specimens and the shot peen specimens tested at medium to high strain amplitudes were at the surface of the specimen. However, for the shot peened specimens the initiation site at low strain amplitude were in the subsurface where the residual stress was tensile. Peened normalized powder metal connecting rods resulted in an increase of 27% fatigue strength as compared with unpeened connecting rods. Shot peen intensity between 15-20 A (Almen) provided the maximum fatigue strength. The initiation site for the component fracture was located approximately 1.5 mm below the surface, which was under a residual tensile stress. Shot peening improved reliability of powder forged connecting rods.

61
PM - Powder metal

61

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Elastic analysis and ABAQUS

Main objectives and conclusions

Rabb [1996]

Fatigue failure of connecting rod

FS

FE analysis

Stresses were investigated in the threaded area of the bolt. Fatigue failure occurred and the crack had initiated in the upper connecting rod threads. Linear analysis was performed to see the differences in a root radius in the connecting rod thread of 0.15 mm with 0.3 mm. For the analysis the screw dimension was M45 X 3 and Kt = 3.7 for 0.15 mm radius and Kt = 2.7 for 0.3 mm radius. The result showed that the maximum principle stress was dominant and the fatigue analysis with sufficient accuracy based on a uniaxial stress assumption and maximum principal stress at the surface.

Non-linear contact analysis

Fatigue

Testing for anisotropy

The engagement of the threads of the connecting rod and the screw was treated as a nonlinear contact problem with friction, plasticity and load cycle. The load applied to this model was the displacement load. The penalty function approach was chosen because this algorithm converges faster. For testing the anisotropy, the tests were carried out on smooth test bars. They were cut in the same direction as the grain flow and perpendicular to it. Mean tensile strength of 904 MPa and yield strength of 757 MPa were observed. It was concluded that there was no clear indication of material anisotropy. Both notched and unnotched fatigue tests were conducted. In both cases mean fatigue limit and standard deviation was obtained. The Haigh diagram was constructed with the help of Gerbers parabola. The fatigue limit at nominal mean stress of 750 MPa was 74 MPa according to Haigh diagram. Load cycling was introduced into the analyses for new designed engine connecting rod. Steady state has been achieved by adding the load steps combustion and unloading to the load sequence.

62
Specimen tension/compression fatigue test (R = -1) Component testing

FS - Forged steel

62

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Load-controlled, reversed axial loading

Main objectives and conclusions

Sonsino [1996]

Fatigue design of high loaded PM parts

PM

Loadings on the connecting rod

FE analysis

Connecting rod is subjected to gas and mass forces. The following parameters were calculated for the severe condition with a gas pressure P = 660 KPa, rotational speed nmax= 6800 rpm and an operational load amplitude Fa = 19.2kN. For the shank Fm = -2.2 kN, with R = -1.26 For the pin-end Fm = -5.5 kN, with R = -1.82 For the crank-end Fm = +7.8 kN, with R = 0.42. The FEA were conducted based on the known fatigue properties. The critical areas were the transition regions from the small end to the shank, from the shank to the big end and the transition area and from the cap to the bolt seat. Connecting rod is required to operate for about 2 x 109 cycles. Test loads, which were much higher than the calculated load, resulted in failures occurring mainly from fretting corrosion between gudgeon pin and the small end. Failures at the transition regions of the shank were rare. PM offers economic manufacturing routes capable of competing with conventional materials and producing parts subject to fatigue loading. Murakami-Endo approach was used to estimate the fatigue strength of components containing defects, inclusions or inhomogenieties. Different type of cast iron connecting rod materials were investigated. Surface residual stress, SEM of fracture origins were discussed. The study suggested a methodology for a confident estimate of the fatigue strength lower bound in a batch of industrial components.

63
Beretta, Blarasin, Endo, Giunti, and Murakami [1997] Defect tolerant design of automotive components Nodular cast iron / carbon forged steel

Fatigue

Strain-controlled, reversed axial loading (R = -1)

The fatigue strength estimation of materials or components containing natural defects, inclusions or inhomogeneities

Fatigue tests: Rotating bending, R = -1 Plane bending, R = -1

PM - Powder metal, FE - Finite element, SEM - Scanning electron microscope

63

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Newtons second law equation Pressure in a four-stroke engine

Main objectives and conclusions

Schreier [1999]

Tension and compression in connecting rods

Loading and free body diagrams

From the second law of Newton various equations were derived. The pressure is small in the intake cycle. In the second stroke a much larger amount of pressure is applied to the piston. The pressure is maximum in the third stroke in which the dangers of engine knock or pre-deformation can occur. Finally, in the exhaust stroke, the pressure is very low. The connecting rods are often made to withstand stresses in excess of 250 Ksi. Fatigue evaluation was conducted on powder metal specimens. The cyclic stress-strain curve was obtained by measuring the stress amplitude from near half-life hysteresis loops. It was concluded that cyclic stress is approximately 20% greater than monotonic stress at 0.01 strain. Strain-based approach was used to analyze local yielding. Strain-life fatigue tests were conducted to predict the effect of periodic compressive loads. It was concluded that variable amplitude loading consisting of a compressive overload cycle followed by smaller cycles, resulted in a fatigue strength reduction of 80% at 107 cycles when compared with constant amplitude loading.

Chernenkoff and Krause [2000]

The influence of variable amplitude loading on the fatigue behavior of a powder-forged steel

PM

Fatigue properties

Specimen strain-controlled, fully-reversed fatigue tests (R = -1)

64

PM - Powder metal

64

Table 3.1: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
Component constant amplitude, load-controlled axial fatigue tests at a frequency of 170 Hz with R = -1.50

Main objectives and conclusions

Illia and Chernenkoff [2001]

Impact of decarburization on the fatigue life of powder metal forged connecting rods

PM

The impact of the decarburized layer on the fatigue life

Weibull test was conducted and the loads were high enough in order to obtain only failures. All the tested connecting rods failed in the minimum cross-section of the I-beam. Study showed that a fatigue crack did not always initiate at the field with the highest decarburized layer depth. Decarburized layer depth up to 0.4 mm does not impact fatigue life. Fatigue life decreases if decarburized layer depth of 0.4 mm or higher exists. Most fatigue cracks initiated at subsurface, due to compressive residual stress induced during shotpeening. Other methods were used to analyze the data such as B life method, Weibayes method, Mean life ratio method and Likelihood contour method.

65
Zhu and Chen [1983] The stability of the motion of a connecting rod Investigating the stability of the motion of a connecting rod Perturbation method Mathieu-type of partial differential equation used

For this study the rod was assumed to have constant cross-sectional area. The ratio of crank radius to the rod length was assumed to be small. The approximation to the transverse motion of the rod satisfies the Mathieu-type equation, from which instability regions of motion were derived. Numerical method was used to calculate the transverse viberational response, both stable and unstable. The approach used in this study was determined to be sufficient in determining the stability of a given design.

PM - Powder metal

65

Table 3.2: Optimization studies survey

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
FEA

Main objectives and conclusions

Yoo, Haug, and Choi [1984]

Shape optimal design of an engine connecting rod

FS

Shape optimization of a connecting rod

Thickness distribution and shape were used as design variables. Stress and thickness were used as constraints. Design sensitivity analysis were performed, where the deformation in the axial direction were analyzed. The simple gradient projection method was used to solve the optimal design problem. The steepest descent method was also used to verify the results. A 20% weight reduction resulted in the transition regions of the connecting rod with reasonable computational cost.

FS - Forged steel, FEA - Finite element analysis

66 66

Table 3.2: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
EHD, DEHD, and FEHD analysis

Main objectives and conclusions

Geonka and Oh [1986]

An optimum connecting rod design study A lubrication view point.

Optimize the bearing performance

The bearing performance was measured by three key factors: the minimum oil film thickness, the maximum oil film pressure and the power loss. For lubrication analysis, DEHD or FEHD was used. It was concluded that DEHD was more accurate but the cost of computing iterations prohibits its use. Therefore, FEHD analysis was performed to do the design iteration. The final performance of the optimum lubrication design was verified using the DEHD program. Both EHD and FEHD analysis results were close with each other. The pressure distribution was obtained by DEHD analysis. It was observed that the high stresses are concentrated in the vicinity of the central support. Eight different generic models were designed to obtain satisfactory stress level. The maximum octahedral shear stress of 154 MPa, the film thickness of 2.32 m and the oil film pressure of 77.1 MPa were obtained. The friction for optimum design was 7% below the conventional design. In summary the clearance has a bigger impact on power loss. The power loss was reduced by changing the clearance, also the film thickness goes down. Optimizing the structure analytically gives the option of reducing the bearing size and getting a significant improvement in both the film thickness and power loss.

67
EHD - Elastohydrodynamic analysis, DEHD - Detailed EHD, FEHD - Fast EHD

67

Table 3.2: Continued

Reference Title Material/Process Aspect(s) Studied Experiments/Analysis conducted & loading
FEM by elastohydrodynamic lubrication theory

Main objectives and conclusions

Murray [1988]

Compliant connecting rod reduces bearing stresses

PM

Optimization of bearing performance by complying with fluid pressure loads

The objectives of the study were to maximize film thickness and minimize power loss. The bearing performance was measured by 3 characteristics -minimum oil film thickness. -maximum oil film pressure. -power loss. Study revealed: -150% increase in the bearing film thickness. -25% decrease in oil film thickness. -Improved pressure distribution results in more durable bearings. Study concluded that downsizing the bearing surface, helps improve fuel economy. To lower the manufacturing cost, it is suggested to loosen tolerances on the mating (bearing and connecting rod) dimension.

68
PM - Powder metal

68

Table 3.2: Continue

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
FE analysis

Main objectives and conclusions

Hippoliti [1993]

FEM method for design & optimization of connecting rods for small twostroke engines

Comparison of 2D and 3D analyses

A 2D approach presents big advantage over a 3D approach, less modeling and meshing difficulties and less CPU time for modeling and analysis. But 2D approach cannot give information about effects of the fillets in 3D. Two different ANSYS sessions were used to generate 2D and 3D FEM models of the connecting rod. 2-D analysis was based on the plane stress approach. The 2D analysis is much robust than the 3D analysis. The results obtained by 2D were satisfactory. The CPU time was 1/5 of that for 3D model. The total mass of the connecting rod was optimized by checking von Mises stresses in six zones. The computer analysis method was used to cut down the time necessary to perform complete check by a factor of 10. It was concluded that 2D analysis was cheaper and faster.

69
Vijayaraghavan, Brewe, and Keith [1993] Effect of out-ofroundness on the performance of a diesel engine connecting rod bearing Dynamic performance of engine connecting rod bearing with circular and out-of-round profiles Analytical formulation: Cavitation algorithm

The connecting rod bearing was analyzed by cavitation and mass inertia effects. Three out-ofround bearing profiles, namely elliptical, semielliptical, and three lobed epicycloids were considered in the analysis. It was concluded that for the circular bearing, the minimum film thickness as a function of crank angle is larger than the values predicted by Jones prediction. The out-ofroundness in bearing geometry reduced the minimum film thickness and a shift in the crank position, the motion of the journal, as observed from the rocking bearing, and an increase in the peak pressures, an increase in the average side leakage rate and the same amount of power loss. Therefore, a circular bearing profile was considered better for its performance.

FE - Finite element

69

Table 3.2: Continued

Reference
Karhausen and Kopp [1995]

Title
Improvement of microstructure in forging of a connecting rod by means of finite element simulations

Material/process

Aspect (s) studied

The use of computer simulation to obtain the microstructure of a process

Experiments/Analysis conducted & loading

FEM 2-D thermally coupled analysis 1.

Main objectives and conclusions

The reproducible inhomogenities, which appear as white spots, are because of unfavorable distribution of recrystalization between two phases. It is advantageous to transfer part of the necessary total strain from the pre to finish forging. After changing the geometry of preforging die, the white spots are successfully avoided in the industrial process. An even distribution of a ferrite concentration of about 16% was observed experimentally.

2.

3.

4.

Jinka and Bellet [1996]

Hot forging of a P/M connecting rod: Three dimensional computer model

PM

Analysis of the densification process during hot powder forging

3-D FE analysis

1.

2. 3. 4.

5.

A 3-D FEA model successfully simulated the frictional phenomenon and plastic yield of porous material. The design of the preform can easily be modified. The low-density regions were predicted at the crank and pin ends. For homogenous density distribution, the initial density of the preform should be inhomogeneous. The numerical model developed in this research applied to the hot forging of a connecting rod.

70

FEM - Finite element model, PM - Powder metal

70

Table 3.2: Continued

Reference
Lu [1996]

Title
The shape optimization of a connecting rod with fatigue life constraint

Material/process

Aspect(s) studied
Optimal design of a connecting rod for a newly developed motor-cycle engine, considering fatigue life constraint

Experiments/Analysis conducted & loading

FEM

Main objectives and conclusions

An optimization approach incorporating fatigue life constraint, based on fracture mechanics was used. The objective was to minimize the total volume of the connecting rod. Stress, fatigue life constraint, load analysis, inertia forces, propagation life and design sensitivity were studied. Optimization algorithm was developed into a computer program. After 19 iterations, the volume was converged, with stresses and fatigue life constraints satisfied.

Takemasu, Vazquez, Painter, and Altan [1996]

Investigation of metal flow and preform optimization in flashless forging of a connecting rod

FS

Design of the preform to forge a connecting rod without flash

3-D FEM simulation using DEFORM3D.

The flash, which accounts for approximately 20 40% of original workpiece of connecting rod, can be eliminated by the approach discussed in this study. The optimum preform design showed that: The large end section geometry is optimized to fill the cavity completely and uniformly. In the small end section, the initial volume distribution and transfer of excessive volume to other features of product is controlled. For the connecting I- beam section, the diameter of the preform was optimized to deform the sidewall nearly in plane strain conditions.

71

FEM - Finite element method, FS - Forged steel

71

Table 3.2: Continued

Reference Title Material/process Aspect(s) studied Experiments/Analysis conducted & loading
3-D FEM Simulation

Main objectives and conclusions

Vazquez and Altan [2000]

Die design for flashless forging of complex parts

FS

New design of preform and tooling concept for the forging of connecting rods with a controlled amount of flash

The proposed tooling concept results in significant material saving, however a more strict control of the volume of the preform is necessary. The concept presented allows a small amount of flash (5%) compared to the conventional forging processes. The tooling geometry was optimized using two-dimensional simulations and the best results were obtained for a tooling geometry with the flash gap in the middle of the connecting rod and a flash thickness of 0.8 mm. Unlike conventional design, the proposed tooling will not be overloaded because the flash is extruded radially without a normal load, i.e., the upper die does not compress the flash.

72
Yang, Dewhirst, Allison, and Lee [1992] Shape optimization of connecting rod pin end using a generic model Optimization of generic model of pin-end 3-D FEM optimization

A generic connecting rod model of upper pin end was used to provide the relationship between the FE model and design variables. A quarter symmetric model was examined. The pin was assumed to be continuous with the rod. A uniform pressure of 100 MPa was applied at the pin end to simulate the firing load. The critical stress of 1000 MPa was chosen for the analysis. For the minimum weight of a connecting rod pin-end, two optimization models were designed, one with 5 design variables and another with 10 design variables. A forward finite difference method was used for updating the finite element model. A modular approach was used for design sensitivity analysis and optimization. Convergence resulted after 6 iterations.

FS - Forged steel, FEM - Finite element method

72

Table 3.3: Manufacturing process and economic studies survey

Reference
Klink [1985]

Title
Multi-stage honing of connecting rods

Material/Process
FS

Aspect(s) studied
Finishing of connecting rod bores by honing process

Experiments/Analysis conducted & loading

Surface finish experiments

Main objectives and conclusions

Different honing procedures were described. Honing process eliminated the rework in the production. Multiple honing provides a high output within a small workspace. High-performance diamond honing sticks can be used to obtain fine boring, which increases the accuracy and makes fitting of bearing shells easier.

Weber [1990]

Powder forged connecting rod state of the art and economics

PM

Manufacturing processes

In this study two methods of PF process described are: Compacting-----weight control------sinter furnace-----inductor (temperature of heated preform is controlled by infrared camera)---forging-----cool down. Compacting ------weight control-----sinter furnace (here the first temperature zone is rapid-burn off of lubricant and in the 2nd zone the parts are sintered for a short period of time)-----forging-----cool down. PF tooling is very similar to powder compacting tooling. Due to closed die forging 1% oversize density in preform can lead to overcharge and rupture of powder forging tool. Accurate weight and low tolerances lead to less machining operation on sintered forged rods. Fracture splitting of cap resulted in remarkable low cost. In PM, savings is achieved by eliminating some machining operations.

73
Configuration of tooling Application example: connecting rod Economic aspects

PM - Powder metal, FS - Forged steel

73

Table 3.3: Continued

Reference
Kimura, Nakamura, Isogawa, Matsubara, Kimura, and Sato [1991]

Title
A free machining titanium alloy for connecting rods

Material/Process
FS

Aspect(s) studied
Investigated process design and surface treatment of titanium alloy connecting rods

Experiments/Analysis conducted & loading

Rotating-bending fatigue tests of specimens and connecting rods

Main objectives and conclusions

Ti-3Al-2V alloy was less machinable but drilling was easily performed. It has poor hot ductility. Induction heating was adopted to eliminate surface oxidation. To assure high fatigue strength, pickling was performed after forging to remove the oxidation layer. To protect against rubbing action between titanium alloy connecting rods and steel crankshafts, PVD of CrN was selected. New developed connecting rods showed high performance. It was concluded that the split CFRP/titanium version results in weight savings of 30%. Economic application of split CFRP connecting rod is not feasible for high volume production. CFRP provides good damping properties, reducing noise emission of the engine. The titanium provides high bending strength and stiffness.

Kuch and Kummerien [1991]

Potential for lightweight design of a hybrid connecting rod for automobile applications

CFRP

Development of CFRP connecting rod used (shank and a tension loop made of CFRP, split crankshaft bearing and small eye made of titanium)

74

CFRP - Carbon fibres, and polyimide resins, FS - Forged steel

74

Table 3.3: Continued

Reference
Weber [1991]

Title
Cost effective finishing of powder forged connecting rods with the fracture splitting method

Material/Process
PM

Aspect(s) studied
Fracture splitting and economic aspects

Experiments/Analysis conducted & loading

Main objectives and conclusions

The study shows savings by eliminating some machining processes such as: separation of rod cap from rod by broaching, grinding of joint faces, drilling of reamed holes, and application of adjusting pins and bushings. Close weight tolerances ( 0.5%), resulting in weight pads no longer required. No secondary heat treatment. Production costs of a powder blank part are higher in comparison to the cast or conventional forged steel connecting rod. Economic analyses of powder forged component should not carried out by simply comparing production price against production price of the other procedures in competition but also consider high precision achieved by powder forged component and consequently the savings on investments and machining operations.

75
Weber [1991] Powder forging of automotive connecting rods PM Fracture splitting procedures and economic aspects

The method for powder forging of connecting rod is explained in this research. Fracture splitting procedure of PM of connecting rod was studied. Saving for PM is achieved from reduction in machining lines and machining operations.

PM - Powder metal

75

Table 3.3: Continued

Reference
Araki, Satoh, and Takahara [1993]

Title
Application of powder forging to automotive connecting rods

Material/Process
PM

Aspect(s) studied
Manufacturing process and fatigue tests

Experiments/Analysis conducted & loading

Load-controlled, axial fatigue tests at R = -1, -1.87

Main objectives(s) and conclusions

The process of trial connecting rod was outlined. Powder is mixed and packed in the compaction. The green compact is sintered at 1120 oC. Then the sintered compact is heated to 1000 oC. The preform is transported into forging die. Temperature of forging is lowered to room temperature. After heating to 870 oC, the forging is normalized and tempered at 570 oC. The ferrite content in connecting rods should be less than 23%. Density of 7.82 g/cm3 was obtained at any point on connecting rod. Specimen testing at stress ratio of -1 and -1.87 showed fatigue limit of 340 N/mm2. Component testing satisfies 107 cycles with 33.3 kN tension and 62.2 kN compression.

76
PM - Powder metal

76

Table 3.3: Continued

Reference
Weber [1993]

Title
Comparison of advanced procedures and economics for production of connecting rods

Material/Process
Die Cast

Aspect(s) studied
Fatigue strength Fracture splitting of cap from rod

Experiments/Analysis conducted & loading

Surface treatment

Main objectives(s) and conclusions

Shows 20% increase in fatigue strength due to surface treatment via shot blasting. Fracture splitting process is one of the major issues manufacturer of conrods are seeking.

Heat-treatable steel/FS

Fracture splitting

Splitting of cap from the rod is not feasible due to nonisotropic structure and residual stresses. In this case spring back effect after splitting results in opening of the cap and hence mismatch of the cap and rod. Fractured joint faces accurately match each other and less machining is required. Less weight and high precision obtained as compared to other processes In this case saving from machining process and weight uniformity Shows unsatisfactory fatigue strength and high cost for the procedure and material Due to economical restriction application of this type cannot be expected in the near future.

PM

Fracture splitting and weight

77
PM - Powder metal, FS - Forged steel

PM

Economics aspects

Carbon Fiber Composite Aluminum, PM

Fatigue and cost

77

Table 3.3: Continued

Reference Title Material/Process Aspect(s) Studied Experiments/Analysis conducted & loading

Main objectives and conclusions

Kuratomi, Takahashi, Houkita, Hori, Murakami, and Tsuyuki [1995]

Development of a lightweight connecting rod made of a lowcarbon martensite steel

Low-carbon martensite steel/ FS

Buckling strength

The elastic limit of the low-carbon martensite steel was found to be lower than that of the conventional steel. Buckling strength is influenced by many factors. By optimizing the section area, hardness, and coining ratio, the buckling strength improved by 30%.

Fatigue strength

Component axial fatigue tests

Residual stresses by shot peening improved the fatigue strength which also reduces the notch sensitivity that forms during forging. Magna test: To measure the hardness of the connecting rods, which is directly correlated with material makeup and impedance. Shot peening: To assure the expected requirements drum type shot peeing equipment were introduced.

Manufacturing technology

78
Banerji [1996] Application of microalloyed forgings for heavyduty diesel engine connecting rods and the other components MA steel/ FS Implementation of microalloyed steel

The application of MA steel for high horse power high-speed diesel engines developed in the mid 1980s. This technology was very well developed in Europe. Reasons for slow growth of MA in the US are: heat treat costs, forging capabilities, steel availability, machinability concerns, and consumer liability. Multi-disciplinary effort was used to make the MA steel connecting rod cost effective. The tooling cost was reduced by 84% and the operating cost for machining line was reduced by 71% from the original level. The overall development of new parameters reduce the cost by 26%.

FS - Forged steel, MA - Microalloyed steel

78

Table 3.3: Continued

Reference Title Material/Process Aspect(s) Studied Experiments/Analysis conducted & loading
Load-controlled, axial fatigue tests at 25 Hz of connecting rods

Main objectives and conclusions

Skoglund and Bengtsson [2001]

Warm compaction cuts the cost of connecting rods

PM

Warm compaction process

Comparison of powder metal and forged steel connecting rod processes were discussed. 1. Survivable fatigue local range was determined by push-pull test. 2. The fatigue test duration was 10 millions cycles at frequency of 25 Hz. 3. Conventional steel-forged rods were found to survive the highest fatigue load of 101 kN. PM standard rods survived a local range of 82 kN. The warm compact prototype rods were not tested but expected to survive 80 kN. 4. The warm compacted rod requires less machining and fewer processing steps and thus significantly reduced the total production cost.

79

PM - Powder metal

79

Table 3.4: Other studies

Reference Title Material/Process

Aspect(s) studied
The development of the strength-design system for bolted-joint case. The stress on the bolt and the separation behavior of the contact plane are discussed.

Experiments/Analysis conducted & loading

Experimental equipment provided by Research lab. Of P.M.E. for local deformation of the connecting rod Stress analysis of the bolt models

Main objectives(s) and conclusions

The analysis is based on the assumption that the contact plane does not separate (separation at the contact plane leads to an over-stressed condition of the bolt). A method of the strength design for the connecting rod bolts is developed in which the connecting rod joint is transformed into the single-bolted-joint model by using the ring model. The local deformation of the connecting rod joint and the stress on the bolt were measured and the analytical results are compared with the experimental results. Both results showed good agreement. By using correct combination of control gas, heating temperature and times, oxide inclusions resulting from powder metal processes can be reduced. Pre-forging heat treatments can alter the forged oxygen content, and effectively increase the inter-inclusion spacing resulting in cleaner, better-branded particle boundaries. As the oxygen level decreased, further parts of oxide inclusions were reduced to metal and fracture behavior of inclusions was changed from brittle fracture to ductile fracture.

Hagiwara [1984]

A method to simplify the strength design of bolted joints-case of connecting-rod bolts

Saritas and Davies [1984]

Fracture behaviour of powder forged steels

PM

The influence of reduction of oxide inclusions on the fracture behavior of powder forge steel during preforging heat treatments were investigated

80

PM - Powder metal

80

Table 3.4: Continued

Reference Title Material/Process Aspect(s) Studied Experiments/Analysis conducted & loading Main objectives and conclusions

Gunyaev, Borovskaya, and Panin [1994]

Using CFRP for the design of a connecting rod for automobile applications

Carbon fibre and bisimide matrix

To develop a composite connecting rod

A composite connecting rod offers some advantages such as weight savings, inertia force reductions, increasing automobile speeds and reducing noise emission from engines. The unsplit connecting rod shows a 71% weight reduction, and split connecting rod shows 63% weight reduction as compared to metal connecting rod. Only tension loadings were used in the study because both split and unsplit connecting rods failed under tension, but passed under compression loading of 10 kN. Study showed unsplit connecting rod has more ultimate tensile strength and less elongation than split connecting rod.

Unsplit vs. split connecting rod

81
Dynamic testing of unsplit connecting rod Compression-tension, loadcontrolled fatigue tests with frequency of 60 Hz Titanium-CFRP connecting rod

For this study the maximum load used was 25 kN compression and 15 kN tension. Compressiontension fatigue loading for 106-107 cycles caused a decrease of tension load. The residual tension load of composite connecting rods was 85-95 and 5367% of static ultimate load under loading for 106 107 cycles. The Titanium-CFRP connecting rod has considerably higher elongation under loading which is the biggest drawback of this application. The study shows a decrease in noise level and also 15% decrease in mechanical losses. Many economical and production problems are still to be solved and much progress required before mass production of CFRP composite connecting rod can be considered.

CFRP - Carbon fiber reinforce plastic

81

Table 3.4: Continued

Reference Title Material/Process Aspect(s) Studied Experiments/Analysis conducted & loading
Casting simulation

Main objectives and conclusions

Chen, Herrera, and Venkatesan [1999]

Solidification modeling and simulation for connecting rods to overcome porosity problems in the squeeze casting process

Casting

Elimination of porosity in connecting rods

The computer application for die-cavity filling and casting solidification simulation, were used to predict the die filling conditions by using different gating methods. It was concluded that solidification analysis showed that the center section has the best area to gate the connecting rod to avoid shrinkage porosity in the casting. Functional solid x-ray view plots and filling sequences comparison showed that the single gating approach was better than the double gating system, the single gating approach showed a much smoother flow pattern to avoid air entrapment for porosity. Hot forging technology required the flowing steps: Process requirement: high billet accuracy, constant tool temperature, constant forging temperature. Tool requirements: exact guidance of tool elements, positioning of the billet inside the die, and reliable closing of the die. Press-requirement: overload protection, tool mounting space, ejector, and constant forging energy. Precision hot forging improves the mechanical properties of connecting rods. o Higher dimensional accuracy of parts. o Decrease the number of deformation steps. o Minimizes the billet mass by forging without flash in closed dies. o Saving of clipping operations by forging without flash. The precision hot forging technology demonstrates its suitability for the products of connecting rod, bevel, gears, straight planet gears, helical gears, clutch gearing or constant velocity joints.

82

Doege and Bohnsack [2000]

Closed die technology for hot forging

FS

Application and importance of closed die technology for hot forging

82

Table 3.5: The local stress amplitude obtained for the connecting rod critical areas (Sonsino, 1996).

83

Table 3.6: Typical engine parameters of 4-cylinder engine (Folgar et al., 1987).

Connecting rod length Crankshaft radius Crank to C. G. Distance Engine R.P.M Maximum firing pressure Maximum compression pressure Piston diameter Piston assembly mass Steel connecting rod mass

157.00 mm 52.00 mm 42.50 mm 4000-8000 rpm 53.70 ATM 10.00 ATM 87.5 mm 585.00 g 662.00 g

Table 3.7: Maximum stress values and locations in a connecting rod (Webster et al., 1983).

Loading condition Tension Compression Tension & bolt pretension Compression & bolt pretension

Maximum octahedral shear stress 19.4 Po 5.6 Po 63.2 Po 51.9 Po

Location Bolt head Shaft Bolt head Bolt head

Po is the maximum pressure in the pressure distribution profile

84

Table 3.8: Comparison of three major manufacturing processes of connecting rod. Forged steel
Advantages
Raw material is cheaper than powder metal. This process provides high strength, ductility, and impact resistance along the grain flow of the forged steel. The density achieve by this process is uniform. Fatigue performance of forged steel rods is higher than die cast and powder metal rods. Connecting rods are produced using permanent mold casting process. The molten material poured in the mold is in the same amount as the mold shape. No flash is produced during this process, except the riser where the molten metal is poured in. Cast connecting rods do not require further steps to obtain final shape as the connecting rod achieved its dimension at the first molded stage. Fracture splitting process is recently introduced for die-cast connecting rods, which can save several expensive manufacturing steps. Powder metal preform connecting rods start with the net shape, that result in no material waste. There is no balancing pad used in this process. Fracture splitting of cap from rod without subsequent machining of matching surfaces is most popular for powder metal connecting rods.

Die-casting

Powder metal

Disadvantages
The roughstock weigh required is more than needed for connecting rod, resulting in some scrap. In conventional forging, flash is produced during forging and requires extra steps of trimming to remove. Recently developed connecting rods are forged by flashless forging. To ensure proper weight balance, conventional connecting rods are provided with excess weight, which is later remove during finishing operations. Conventionally, the cap and rod are sawed apart for one-piece forging, or cap/rod are forged separately. In order to remate the surfaces, grinding of rod/cap contact surfaces are required. Recently, newly developed crackable steel provides perfectly matching mate surfaces without needing additional machining/grinding. For cast connecting rods weight and balancing is provided at the molding process, which is removed during the finishing operations. For this process the die is filled violently and solidification happens quickly, and results in microporosity. The dies for this process are complicated and costly to construct. Raw material of this process is expensive because of operations of powder formation, presintering, and sintering. Sintering is the biggest step after compacting, where connecting rods achieve toughness and bonding of powder metal particles. Lower mechanical properties because of rapid solidification. During compacting trapped oxygen in the powder results in porosity, decarb, and oxide penetration. The density can vary within powder metal connecting rod. Costly part density modification or infiltration is required to prevent powder metal defect.

85

Figure 3.1: A free body diagram of a piston forces in cylinder engine (Shiao and Moskwa, 1993).

Figure 3.2: A free body diagram of a connecting rod (Shiao and Moskwa, 1993).

86

Figure 3.3: A free body diagram of forces on a crank system (Shiao and Moskwa, 1993).

87

Load distribution
Compression load

Tension load

Uniform distribution over 120o

Cosine distribution

over 180o

Figure 3.4: Schematic of load distribution (Folgar et al., 1987).

Figure 3.5: Schematic of cylinder pressure versus crack angle at 5000 r.p.m. (Majidpour et al., 2002).

88

Figure 3.6: Boundary conditions used for static load analysis (Sugita et al., 1990).

89

Figure 3.7: Comparisons between FEA-calculated and strain gage measured values of stress at critical locations in static load analysis (Sugita et al., 1990).

90

Figure 3.8: Boundary conditions for quasi-dynamic FE analysis (Sugita et al., 1990).

91

Figure 3.9: Schematic of link mechanism to measure stress in connecting rod under actual engine operating condition (Sugita et al., 1990).

92

Figure 3.10: The calculated values of lightweight connecting rod (Sugita et al., 1990).

93

Figure 3.11: Distribution of tension loading of the connecting rod (Webster et al., 1983).

Figure 3.12: Distribution of compressive loading of the connecting rod (Webster et al., 1983).

94

Figure 3.13: Schematic of the deformation of the connecting rod (Webster et al., 1983).

Figure 3.14: Stretch and out of roundness as a result of inertial load (Balasubramanian et al., 1991).

95

Figure 3.15: Crankpin pressure distribution under inertia load (Balasubramanian et al., 1991).

Figure 3.16: Deformation and out of roundness from a firing load (Balasubramanian et al., 1991).

96

Figure 3.17: Crankpin pressure distribution as a result of firing load (Balasubramanian et al., 1991).

97

Figure 3.18: Comparison of fatigue strength of powder forged (Fe-3Cu-0.55C0.35) and hot forged (SAE 40L559) connecting rods (Imahashi et al., 1984).

Figure 3.19: The effect of temperature on fatigue limit of smooth and notched specimens made of microalloyed steels (Kuratomi et al., 1990).

98

Figure 3.20: The fatigue strength of as-sintered powder metal connecting rod as compared with forged steel connecting rod (Suzuki et al., 1988).

99

Figure 3.21: Typical fatigue test rig of connecting rods (Suzuki et al., 1988).

100

Figure 3.22: Weibull probability plot of powder forged and forged steel rods (Marra et al., 1992).

101

Figure 3.23: Differentiation of fatigue cycles on the connecting rod (Beretta et al., 1996).

102

Figure 3.24: S-N plot of hot forged carbon steel under cyclic alternating loads (R = -1 (Berreta et al., 1996).

103

Figure 3.25: S-N plots of type-B cast iron at different stress ratio. (a) R = 0, (b) R = -1, (c) R = -1.84 (Berreta et al., 1996).

104

Figure 3.26: Staircase fatigue test results of 1151 steel connecting rods (R = -1) (Olaniran and Stickels, 1993).

Figure 3.27: Staircase fatigue test results of new alloy connecting rods (Olaniran and Stickels, 1993).

105

Figure 3.28: Density distribution for powder metal connecting rod (Araki et al., 1993).

Figure 3.29: The effect of density on mechanical properties of 4640 powder forged steel (Imahashi et al., 1984).

106

Figure 3.30: The effect of density on fatigue strength of 4640 powder forged steel (Imahashi et al., 1984).

107

Figure 3.31: Residual stress profile in fatigue at 20 A intensity (Chernenkoff et al., 1995).

108

Figure 3.32: The staircase fatigue test results for shot peened and unpeened connecting rods (Chernenkoff et al., 1995).

109

Figure 3.33: The statistical comparison of shot peened and unpeened connecting rods (Chernenkoff et al., 1995).

110

Figure 3.34: The influence of shot peening on fatigue strength (Chernenkoff et al., 1995).

Figure 3.35: The comparison of FE calculated and strain gage measured stresses (Sugita et al., 1990).

111

Figure 3.36: The boundary conditions used in shape optimization study by Yoo et al. (1984).

Figure 3.37: Geometry and loading of an engine connecting rod (Yoo et al., 1984).

112

Figure 3.38: Optimum design resulting in 25% weight reduction (Yoo et al., 1984).

113

Final optimal design

Figure 3.39: Views of the several designs including final bearing optimal design by Goenka and Oh (1986).

114

Bar

Forging + Quenching

Shot Blasting

Measurement of Hardness & Weight, Magna test Magnaflux Shot Peening

Coining

Machining

Figure 3.40: The manufacturing process of new developed connecting rod (Kuratomi et al., 1995).

Increased thru heating of mults

Lowered forging temperature

Tighter Chemistry

Moderately elevated sulfur

Tooling analysis

Reduced hardness variation from surface to core

Finer grain size, more ferrite %, reduced hardness

Less hardness variation from heat to heat

More Mn S, finer grain size, more intragranular ferrite

Chemistry & geometry optimization, coolant, etc.

Improved machinability reduced variability reduced cost

Machining cost reduction of 54%

Figure 3.41: Use of microalloyed forged steel employed for heavy duty diesel engine connecting rod (Banerji, 1996).

115

Origina

Compacting

Sintering

Heating

Forging

Heat treatment

Testing

Iron powder V-Mixer 30

Pressure 41/cm2 Density of preform 85% (6.8 g/cc Approx.) Floating die system

Temperature 1120 oC (2048o F) Atmosphere Ammonia cracking gas (75%H, + 25%N2) 20 min

Temperature 1000 oC (1832 o F) Atmosphere No oxidizing (Argon gas) 15 min

Die temperature 300 oC (572o F) Forging time 5 pieces/min Forging die upper die moves lower die fixed Argon gas cooled after forging

Tempered at 570 oC (1058o F)

Chemical analysis

min

Density measurement Micro-structure observation Hardness measurement Mechanical strength measurement Axial loading fatigue test

Figure 3.42: A trial powder forged connecting rod production process (Araki et al., 1993).

Figure 3.43: Cost ratio of powder forged vs. hot forging (Imahashi et al., 1984).

Figure 3.44: Manufacturing energy comparison of powder forged vs. hot forging (Imahashi et al., 1984).

116

Figure 3.45: Comparison of raw materials (Imahashi et al., 1984).

117

CHAPTER 4 STRESS ANALYSIS AND FEA

4.1 Finite Element Modeling

The objective of FEA is to investigate stresses, displacements and hotspots experienced by the connecting rod. The stresses obtained can then be used to predict the fatigue life, and determine the expected failure regions. Forged steel rod was used for the finite element analysis, since this connecting rod was also optimized (Shenoy, 2004). Linear elastic analysis was used since the connecting rod is designed for long life where stresses are mainly elastic. Therefore, only Youngs modulus (E = 207 GPa or 30,000 Ksi) and Poissons ratio ( = 0.3) were needed as material properties. Axial loading was considered for all the analyses, since this is the primary service loading.

4.1.1 Generation of the geometry and FEA mesh of the component Modeling incorporated three-dimensional geometry, tension and compression loading, symmetry conditions and other aspects of designing. A 3-D model geometry was developed in IDEAS-8. Dimensions of the forged steel connecting rod were taken from three different connecting rods, and the average of these dimensions were used to generate the model. The dimensions were obtained by a coordinate measurement machine, as well as a micrometer, digital calipers, dial calipers, compass and depth probe. Due to symmetry of the geometry, the component was first half modeled, and then the entire geometry was created by reflecting (mirror imaging) the half geometry (Figure

118

4.1). Table 4.1 summarizes the few connecting rod parameters used for the analysis in this study. The model was designed without forging flash, bolts and crank/pin bearings, as these details are not expected to have any significant influence on the obtained results at the critical regions, and their removal allowed simplification of the model. The density of 7.806 x 10-6 kg/mm3 (0.282 lbs/in3) was used as material property in the FEA model to calculate the mass of the connecting rod. The mass based on the FEA model was found to be 439 g (0.968 lbs), which included the mass of the two bolts but excluded the two bearings. The average weight of 12 forged steel connecting rods was measured to be 455 g (1.003 lbs), which included the bolts but not the bearings. The mass of the two bolt head was approximated at 11.7 g (0.026 lbs) and subtracted from the FEA-based calculated mass of the connecting rod. The weight difference between the measured and FEA calculated values was found to be less than 1%. This is an indication of the FEA model accuracy. A parabolic tetrahedron element was used for the solid mesh, as this was the default option by I-DEAS for any 3-D analysis. Figure 4.2 shows the element shape used for analysis. Sensitivity analyses were performed to obtain the optimum element size, as discussed in section 4.1.3. The element size of 1.27 mm (0.05 in) was finally considered. A total of 80016 elements and 130210 nodes were generated at 1.27 mm element length.

4.1.2 Loading and boundary conditions Tension and compression loads were applied as pressure on the bearing surfaces of the connecting rod. Webster et al. (1983) found that under actual service condition the pin end experiences tension by the piston pin causing distribution of pressure along the

119

upper half of the inner diameter, which is approximated by the cosine function (Figure 3.12). In compression, the piston pin compresses the bearings against the pin end inner diameter, causing uniform distribution of pressure (Figure 3.13). The same phenomenon of pressure distribution caused by the crankshaft was experimentally measured (Webster et al., 1983) on the crank end of the connecting rod. The connecting rod was constrained in all six degrees of freedom from the bearing surfaces on one end and pressure was applied at the other end. Figure 4.3 shows loading and boundary conditions for tension or compression at either end. The parameters used for FE analyses are shown in Table 4.2. Only longitudinal loading direction was considered (the direction parallel to the longitudinal axis of the connecting rod), as this is the primary loading direction in service. The loading and constraints used for each case are presented as follows: Tension loading: In tension, the connecting rod experiences cosine distribution over 180o of the contact area. The pressure is acting on the contact surface area of the connecting rod. The normal pressure (po) was calculated from the following equations: p = po cos po = (4.1)

2 Pt rt

(4.2)

where,

= Crank angle, 0 degree for top dead center position

r = Radius of crank or pin end t = Thickness of the connecting rod at the loading surface Pt = Force magnitude in tension

120

A 26,688 N (6000 lbs) force (Pt) was used for crank end and pin end tension. Since linear elastic FEA analysis is used, the results obtained are proportional to the applied load. Therefore, stresses for any other load level can readily be obtained. Figures 4.3 (a) and (c) show cosine distribution in tension loading for crank end and pin end, respectively. The obtained pressures on crank end and pin end resulting from this load are different. For the crank end a pressure of po = 35.7 MPa (5.2 Ksi) was obtained, while for pin end a pressure of po = 72.6 MPa (10.5 Ksi) was obtained. The connecting rod has symmetry in two planes; Since the whole model was considered, the following constraints were used in tension: For tension loading in the crank end, the piston pin end is constrained in all six directions (three for translational and three for rotational) through a 180o contact arc. Cosine distribution pressure of po = 35.7 MPa (5.2 Ksi) is applied at the crank end through a 180o contact arc (Figure 4.3a). For tension loading in the pin end, the crank end is constrained in all six directions (x, y, z, xy, xz, yz) through 180o contact arc. Cosine distribution pressure of po = 72.6 MPa (10.5 Ksi) is applied at the piston pin through a 180o contact arc (Figure 4.3c). Compression loading: The compression loading for the crank end and piston end is assumed to have a uniformly distributed loading over the 120o contact arc (Webster et al., 1983). The compression pressure (po) was calculated as: p = po po =
Pc rt 3

(4.3) (4.4)

121

where, r = Radius of crank or pin end t = Thickness of the connecting rod at the loading surface Pc = Force magnitude in compression A force (Pc) of 26,688 N (6000 lbs) was also used in compression. The compression pressure applied on crank end was po = 32.4 MPa (4.7 Ksi), while for the pin end it was po = 65.9 MPa (10 Ksi). Figures 4.3 (b) and (d) show the uniform loading over 120o in compression for the crank end and pin end, respectively. The following boundary conditions were considered for compression loadings: For compression loading in the crank end, the piston pin end is constrained in all six directions (three for translational and three for rotational) through a 120o contact arc. Uniform distributed pressure is applied at crank end through 120o contact arc (Figure 4.3 b). For compression loading in the pin end, the crank end is constrained in all six directions (x, y, z, xy, xz, yz) through a 120o contact arc. Uniform distributed pressure is applied at the piston pin through a 120o arc (Figure 4.3 d).

4.1.3 Mesh size sensitivity analysis

To recognize the effect of element size on the stresses, mesh size sensitivity analyses were carried out. The stress changes as a result of element size changes. These analyses were performed iteratively at different element lengths until the solution obtained appropriate accuracy. For this analysis the rod was constrained at the pin end and cosine distribution pressure was applied at the crank end (see Figure 4.3 a). For the

122

first analysis, a 7.77 mm (0.306 in) default mesh size was used and von Mises stresses were obtained at three different locations. Figure 4.4 shows the locations at which the stresses were obtained. For the second iteration, a mesh size of 2.54 mm (0.1 in) was used, and for the third iteration a mesh size of 1.52 mm (0.06 in) was used. For the final iteration, an element length of 1.27 mm (0.05 in) was used. Convergence of stresses was observed, as the mesh size was successively refined. Table 4.3 summarizes the results of the convergence of stresses at different locations. Figure 4.5 shows the meshing at different element lengths. Figure 4.6 shows the plot of stresses converged at 1.27 mm (0.05 in) element length. The element length of 1.27 mm (0.05 in) was, therefore, considered for the FEA. At this length, 80016 elements and 130210 nodes were observed.

4.2 Finite Element Analysis Results and Discussion

Figure 4.7 shows the von Mises stress contours and critical locations observed under tension and compression loadings. The most highly stressed areas are in the transition regions between the shank and the crank end, as well as the shank and the pin end. Stresses are all symmetric over the entire rod, since geometry and loading were symmetrical. The stresses obtained are below yield strength (elastic); and therefore, the relation between load and stress is linear. If the load applied were increased by a factor of 2, then the stresses would be higher by factor of 2. Figure 4.8 shows the displacement contours. The result shows a maximum longitudinal displacement of 0.302 mm (0.012 in) in tension and 0.114 mm (0.005 in) in compression at the crank end. Similarly, a maximum longitudinal displacement of 0.354

123

mm (0.014 in) in tension and 0.121 mm (0.005 in) in compression were observed at the pin end. The finite element analysis gave an approximated solution for the high or low stresses in the areas where high stress concentration factor (Kt) existed. Kt was obtained from the finite element analysis for the two transition regions in tension. To do this, first maximum stresses were obtained for Sx (stress in the longitudinal direction) and for von Mises stress. These stresses were then divided by the corresponding nominal stresses at the particular point of interest. Table 4.4 summarizes the Kt values. The Kt value at the crank end was subsequently considered because the stresses are higher at the crank end transition region. The Kt value of 1.18 in the longitudinal direction (Sx) was used for the fatigue life calculations, discussed in section 5.3. Kt value was considered more accurate for the case where the load is applied at the crank end and the rod is constrained at the pin end. Figure 4.9 shows the stress contour at the crank end transition region. This figure shows the stresses are highest at the curved outer surface portion of the connecting rod. The boundary conditions used to observe the stress contour in Figure 4.9 are shown in Figure 4.7 (a). Some hand calculations (i.e., P/A) were performed to compare with the FEA results. In order to avoid stress concentration regions where P/A does not apply, the stresses were evaluated in the shank region. For this comparison, tension loading was considered. Figure 4.10 shows comparison of the calculated and FEA stress values, as well as the locations where stress values were obtained and calculated. The crosssectional areas for the shank section at different locations were obtained from the FEA. As can be seen, the stress values from FEA and the hand calculation agree well.

124

In order to verify the finite element calculations, a connecting rod was strain gauged with four strain gauges at the middle of the shank, and then loaded in tension. The strains were measured at the locations shown in Figure 4.11. Under a load of 3000 lbs, 243 micro strain was measured. The strain from the FEA was calculated under 6000 lbs to be 475 micro strain (Figure 4.7 a), which corresponds to 238 micro strain at 3000 lbs load. The small difference of 2% confirms the accuracy of the FEA calculations. From the FEA, it was observed that the shank transition regions are the most critical areas. Stresses at the nodes in the critical locations were obtained (Figure 4.12). These locations are divided into three regions: Region I represents the nodes in the transition region from the crank end to the shank (Figure 4.12 a); region II represents the nodes in the shank section (Figure 4.12 b); and region III represents the nodes at the transition region from the shank to the pin end (Figure 4.12 c). The stresses were obtained in x, y, z directions. Tables 4.5, 4.6 and 4.7 summarize the stress components (x, y, z, xy, xz, yz) for region I, II and III, respectively. The stress in the longitudinal direction (x) was the most prominent stress, as expected. For both transition regions, y, z and xy stresses also show that stress multiaxiality existed in these regions. The xz and yz were nearly zero for all the load cases and in all regions and, therefore, not shown in these tables. Comparison of stresses at the critical locations for region I (nodes from 13013 to 11373 in Table 4.5) between tension loading at the crank end and tension loading at the pin end, and between compression loading at the crank end and compression loading at the pin end show small differences. Similarly, small differences are found for the nodes at critical locations in region III (nodes 24167 to 24161 in Table 4.7). These differences are

125

also small for the nodes in the shank region (region II), as shown in Table 4.6. It can, therefore, be concluded that the two sets of boundary conditions (i.e., loading at one end while constraining the other end and vice versa) do not significantly affect the results at the critical locations (i.e., transition from either end of the connecting rod to the shank region). Even though the differences are small, more accurate results are considered to be for the loading/boundary conditions where the loading is closer to the node being evaluated, and the constraint is further away. The stresses in compression were higher as compared to tension for regions I and III. In the shank region, for both tension and compression, all stresses other than x are zero. Stresses in the shank region in tension and compression were found to be equal because this region is away from loading and boundary conditions, and therefore, not affected by close proximity to these critical locations. Tables 4.8, 4.9 and 4.10 summarize the alternating stresses (xa, ya, za, xya) and the mean stresses (xm, ym, zm, xym) for regions I, II and III, respectively. The alternating and mean stresses for the other two stress components (xz, and yz) are not shown because they were approximately zero. The alternating stress (a), mean stress (m), and R-ratio for each stress component were calculated from:

a ( x, y, z ) =

max min

m( x, y, z ) =
Rx, y, z =

max

2 + min 2

(4.5) (4.6)

min max

(4.7)

126

where, max represents the stress component in tension and min represents the stress component in compression, as listed in Tables 4.5 to 4.7. R-ratio was considered for only the longitudinal direction for all the regions, since x-direction is the prominent stress direction. The R-ratio was considered for the critical regions. For region I, the R-ratio in these regions varies between 1.09 and 1.14, while it varies between 1.0 and 1.20 for region III and between 0.99 and 1.23 for region II. Therefore, under axial load conditions, the mean stress at critical locations is not large, as R = -1 represents completely reversed stressing (i.e. no mean stress). The multiaxial stresses were converted to effective uniaxial stresses using von Mises equivalent stress, and results are summarized in Tables 4.11, 4.12 and 4.13 for regions I, II and III, respectively. These stresses were obtained from:

eq (max)=

1 ( x(max) y(max))2 + ( y(max) z(max))2 + ( z(max) x(max))2 + 6 [ 2 xy(max)+ 2 yz(max)+ 2 zx(max)] 2

(4.8)

eq ( a ) =

1 ( x ( a ) y ( a ) ) 2 + ( y ( a ) z ( a ) ) 2 + ( z ( a ) x ( a ) ) 2 + 6 [ 2 xy ( a ) + 2 yz ( a ) + 2 zx ( a ) ] 2

(4.9)

eq(m) =

1 ( x(m) y(m) )2 + ( y(m) z(m) )2 + ( z (m) x(m) )2 + 6 [ 2 xy(m) + 2 yz(m) + 2 zx(m) ] 2

(4.10)

or

eq ( m ) = mx + my + mz

(4.11)

where, eq(max) is the maximum equivalent stress in tension, eq(a) is the alternating equivalent stress, and eq(m) is the mean equivalent stress. Equation 4.10 always results in a positive equivalent mean stress, whereas, equation 4.11 can be either positive or

127

negative. Therefore, equation 4.11 represents the beneficial effect of compressive mean stress and the detrimental effect of tensile mean stress on fatigue life. The values of eq(max) at the critical transition regions are small as compared to yield strength of the forged steel (700 MPa). At the regions near the pin end and around the oil hole, the stresses are close to yield strength since these regions are at or near the loading and/or constraints. Therefore, stresses at these locations may not be accurate. Note that, the nodes with largest eq(a) do not necessarily have the largest eq(max) (i.e. the difference between static and cyclic stress analysis). For region II, only longitudinal stresses (x), and therefore, a uniaxial stress state exist. For regions I and III, the effect of stress multiaxiality at the critical locations is small, as judged by comparing eq(max) values with x values for nodes at these locations.

128

Table 4.1: Summary of forged steel connecting rod parameters used for FEA.

Forged steel connecting rod parameters Overall length Crank end diameter Pin end diameter Thickness at the crank end Thickness at the pin end Total mass including the bolts but excluding the bearings

Dimensions 187.2 mm (7.373 in) 47.9 mm (1.889 in) 23.6 mm (0.930 in) 19.8 (0.780 in) 19.8 (0.780 in)
455 g (1.003 lbs)

Table 4.2: Summary of connecting rod FE analysis load parameters.

Crank end loading Parameters Load magnitude Load distribution Pressure function Pressure on the surface Tension
26688 N (6000 lbs) Cosine distribution over 180o po =

Pin end loading Tension

26688 N (6000 lbs) Cosine distribution over 180o po =

Compression
26688 N (6000 lbs) Uniform distribution over 120o po =

Compression
26688 N (6000 lbs) Uniform distribution over 120o po =

2 Pt rt

Pc rt 3

2 Pt rt

Pc rt 3

35.7 MPa (5.2 Ksi)

32.4 MPa (4.7 Ksi)

72.6 MPa (10.5 Ksi)

65.9 MPa (9.5 Ksi)

129

Table 4.3: Summary of von Mises stresses at different element sizes.

Size of elements mm (in) 7.77 (0.306) No. of elements 3060 No. of nodes 5996 Location (see Figure 4.4) 1 2 3 1 2 3 1 2 3 1 2 3 von Mises stress MPa (Ksi) 91.4 (13.3) 134.5 (19.5) 85.5 (12.4) 120.5 (17.5) 143.3 (20.8) 135.7 (19.7) 126.1 (18.3) 145.1 (21) 136.3 (19.8) 126.7 (18.4) 147.1 (21.3) 136.7 (19.8) % difference in stress

2.54 (0.1)

21267

37227

1.52 (0.06)

68636

108500

1.27 (0.05)

80016

130210

31.8 6.6 58.7 4.7 1.3 0.4 0.5 1.4 0.4

130

Table 4.4: Summary of stress concentration factors.

Kt at crank end Uniaxial MPa von Mises MPa Kt at pin end Uniaxial MPa von Mises MPa

Loading condition and location of Kt calculation

Kt = Sx/Snom.

Kt = SvM/Snom.

Kt = Sx/Snom.

Kt = SvM/Snom.
SvM = 198, Snom. = 26688/134 = 199, Kt = 1

Sx = 208, SvM = 197, Sx = 222, Snom. = 26688/137 = 195, Snom.= 26688/137 = 195, Snom = 26688/134 = 199, Kt = 1.07 Kt = 1.01 Kt = 1.12

Sx = 229, SvM = 204, Sx = 213, Snom. = 26688/137 = 195, Snom. = 26688/137 = 195, Snom = 26688/134 = 199, Kt = 1.18 Kt = 1.05 Kt = 1.07

SvM = 203, Snom. = 26688/134 = 199, Kt = 1.02

131

131

Table 4.5: Stresses at the nodes in region I shown in Figure 4.12 (a) (transition region between the shank and the crank end) under tension and compression load of 26688 N (6000 lbs).

Loading at crank end Tension (MPa) Node number 12915 12887 12944 12921 12973 12966 13015 13013 13023 13021 13016 13026 13025 13024 13022 13017 13018 1402 1404 1406 1408 1410 1412 1414 11369 11368 11372 11374 11375 11373 x y z xy Compression (MPa) x y z xy 36 68 80 80 71 57 39 27 25 25 27 14 15 16 15 16 19 7 7 7 6 7 7 7 3 3 3 3 4 3 x 92 137 163 175 178 181 187 195 192 192 196 198 194 191 190 194 199 192 187 184 182 184 187 194 175 174 170 171 174 178

Loading at pin end Tension (MPa) y 68 73 59 42 26 13 6 3 2 2 3 1 2 1 1 1 2 0 0 -1 0 0 0 0 0 0 -1 -1 0 -1 z 4 -12 -31 -45 -54 -54 -50 -28 -38 -36 -25 -6 -17 -29 -30 -23 -11 1 -9 -21 -26 -21 -10 1 -8 -16 -21 -22 -13 -4 xy -78 -99 -99 -86 -69 -51 -35 -24 -22 -22 -24 -12 -13 -14 -13 -14 -16 -6 -6 -6 -5 -6 -6 -6 -3 -3 -3 -3 -3 -3 Compression (MPa) x -32 -59 -89 -122 -155 -185 -209 -220 -218 -218 -220 -222 -220 -217 -217 -220 -223 -215 -211 -208 -206 -208 -210 -217 -196 -196 -191 -193 -195 -199 y -23 -31 -31 -28 -22 -7 -4 -3 -2 -4 -2 -2 -1 -1 -2 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 z 15 27 35 41 43 42 40 22 31 29 20 4 14 25 25 19 8 -2 8 18 23 18 8 -2 7 15 20 20 12 3 xy 27 43 53 59 59 52 39 27 24 25 27 14 15 16 16 16 19 7 7 7 7 7 8 7 3 4 3 3 4 3

167 125 17 -142 -44 -31 -6 227 121 -6 -164 -94 -49 7 250 92 -33 -151 -134 -48 22 244 59 -56 -120 -164 -39 35 222 32 -69 -86 -186 -27 44 200 14 -70 -58 -202 -15 47 192 7 -65 -35 -214 -7 46 196 2 -36 -24 -220 -3 26 191 1 -48 -22 -217 -2 36 192 1 -46 -22 -217 -2 34 198 3 -33 -24 -221 -4 24 199 1 -9 -12 -220 -1 6 194 1 -22 -13 -217 -2 17 190 0 -37 -13 -214 -1 27 189 0 -38 -13 -214 -1 30 194 1 -30 -14 -217 -2 23 200 2 -14 -17 -222 -2 10 194 0 0 -6 -212 0 -1 188 0 -12 -6 -208 0 9 183 -1 -26 -6 -204 1 21 181 0 -31 -5 -202 0 26 183 0 -26 -6 -204 0 21 187 0 -13 -6 -207 1 10 195 0 0 -6 -215 0 -1 176 0 -10 -3 -192 0 9 174 0 -20 -3 -192 0 17 170 -1 -25 -3 -187 1 22 171 -1 -26 -3 -189 1 22 174 -1 -16 -3 -192 0 14 179 -1 -5 -3 -196 1 4

132

Table 4.6: Stresses at the nodes in region II shown in Figure 4.12 (b) (region between two transition regions) under tension and compression load of 26688 N (6000 lbs).
Loading at crank end Tension (MPa) Node number 11347 11333 11335 11399 11381 11431 11408 11457 11476 11482 11509 11493 11543 11520 11563 11582 x 176 184 190 194 197 198 198 198 198 198 198 197 195 191 187 184 y 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 z -6 -1 1 1 1 1 0 0 0 0 1 1 0 -1 -5 -11 xy -2 -3 -3 -3 -3 -3 -3 -3 -3 -3 -3 -3 -3 -3 -3 -3 Compression (MPa) x -182 -186 -191 -194 -196 -197 -198 -198 -198 -198 -198 -197 -197 -197 -198 -201 y 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 z 7 2 0 -1 -1 -1 0 0 0 0 0 0 0 1 2 3 xy 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 x 170 175 180 184 186 187 187 187 188 188 187 185 180 172 162 154 Loading at pin end Tension (MPa) y 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 z -6 -2 0 1 1 1 0 0 1 1 1 2 1 -2 -9 -22 xy -2 -3 -3 -3 -3 -3 -3 -3 -3 -3 -3 -3 -3 -3 -3 -3 Compression (MPa) x -184 -188 -191 -194 -196 -197 -198 -198 -198 -198 -197 -196 -195 -192 -190 -190 y 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 z 6 2 0 0 -1 0 0 0 0 0 0 0 0 -2 5 10 xy 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

133

Table 4.7: Stresses at the nodes in region III shown in Figure 4.12 (c) (transition region between the shank and the pin end) under tension and compression load of 26688 N (6000 lbs).

Loading at crank end Tension (MPa) Compression (MPa) Node number 24167 24168 24164 24160 24165 24163 24161 24117 24138 24088 24103 24073 24081 24085 24084 x 204 200 201 201 205 207 207 221 246 283 324 308 198 165 205 y 0 0 0 0 0 0 0 2 6 15 29 40 22 25 23 z xy x -214 -215 -214 -214 -212 -209 -213 -178 -143 -118 -83 -50 -23 -19 -24 y z -1 0 0 -1 -1 -1 0 -2 -4 -6 -7 -7 -3 -2 -3 -1 0 -1 -1 -1 -1 -1 0 0 -2 -4 -5 -1 -1 -1 xy -3 -1 -2 -1 -6 -8 -6 -18 -23 -26 -23 -18 -9 -8 -9 x

Loading at pin end Tension (MPa) Compression (MPa) y z xy x -214 -210 -180 -211 -216 -218 -217 -226 -237 -252 -236 -128 13 35 14 y 0 0 0 0 0 0 0 -2 -6 -14 -22 -13 13 12 10 z 11 16 16 11 17 18 12 16 0 -33 -70 -52 9 17 4 xy -3 -1 -2 -1 -6 -9 -6 -24 -4 -58 -68 -38 14 20 12

-15 2 -21 1 -21 2 -15 1 -23 6 -25 9 -17 6 -28 23 -18 41 12 65 54 94 59 104 3 65 -8 59 9 73

186 0 -47 2 175 0 -49 0 180 -1 -51 2 180 0 0 0 193 -2 -57 6 204 -1 -61 9 196 0 -42 5 286 2 -68 32 400 10 -51 68 530 29 2 123 689 64 101 204 767 114 193 280 607 93 150 224 543 95 132 211 628 93 155 242

134

4.8: Alternating and mean stress (MPa) components for the nodes in region I.
Loading at crank end Node number (Fig. 4.12 a) 12915 12887 12944 12921 12973 12966 13015 13013 13023 13021 13016 13026 13025 13024 13022 13017 13018 1402 1404 1406 1408 1410 1412 1414 11369 11368 11372 11374 11375 11373 xa xm 106 161 192 204 204 201 203 208 204 205 210 210 206 202 202 206 211 203 198 194 192 194 197 205 184 183 179 180 183 188 62 67 58 40 18 -1 -11 -12 -13 -13 -12 -11 -12 -12 -13 -12 -11 -9 -10 -11 -11 -11 -10 -10 -8 -9 -9 -9 -9 -9 Rx -0.26 -0.41 -0.54 -0.67 -0.84 -1.01 -1.11 -1.12 -1.14 -1.13 -1.12 -1.11 -1.12 -1.13 -1.13 -1.12 -1.11 -1.09 -1.11 -1.11 -1.12 -1.11 -1.11 -1.10 -1.09 -1.10 -1.10 -1.11 -1.10 -1.09 ya ym za zm xya xym xa 78 85 70 49 30 15 7 3 2 2 4 1 2 1 1 2 2 0 0 1 0 0 1 0 0 0 1 1 1 1 47 36 22 10 3 -1 0 -1 -1 -1 -1 0 -1 -1 -1 -1 0 0 0 0 0 0 1 0 0 0 0 0 -1 0 12 7 28 46 57 59 56 31 42 40 29 8 20 32 34 27 12 1 11 24 29 24 12 1 10 19 24 24 15 5 6 1 -6 -11 -13 -12 -10 -5 -6 -6 -5 -2 -3 -5 -4 -4 -2 -1 -2 -3 -3 -3 -2 -1 -1 -2 -2 -2 -1 -1 89 116 116 100 79 58 37 26 24 24 26 13 14 15 14 15 18 7 7 7 6 7 7 7 3 3 3 3 4 3 -53 -48 -36 -20 -8 -1 2 2 2 2 2 1 1 2 1 1 1 1 1 1 1 1 1 1 0 0 0 0 1 0 62 98 126 149 167 183 198 208 205 205 208 210 207 204 204 207 211 204 199 196 194 196 199 206 186 185 181 182 185 189 xm 18 7 -15 -40 -65 -86 -102 -109 -108 -108 -109 -111 -109 -108 -108 -110 -111 -108 -106 -105 -103 -104 -105 -109 -98 -98 -96 -97 -98 -100 Loading at pin end Rx -0.35 -0.43 -0.55 -0.70 -0.87 -1.02 -1.12 -1.13 -1.14 -1.14 -1.12 -1.12 -1.13 -1.14 -1.14 -1.13 -1.12 -1.12 -1.13 -1.13 -1.13 -1.13 -1.12 -1.12 -1.12 -1.13 -1.12 -1.13 -1.12 -1.12 ya ym za zm xya xym 46 52 45 35 24 10 5 3 2 3 3 2 2 1 2 1 1 0 0 -1 0 0 0 0 0 0 1 1 0 1 35 53 66 74 78 87 92 96 95 94 97 98 97 95 94 97 100 96 94 92 91 92 94 97 88 87 86 86 87 90 6 10 53 -26 20 8 71 -28 33 2 76 -23 43 -2 73 -14 49 -6 64 -5 48 -6 52 1 45 -5 37 2 25 -3 26 2 35 -4 23 1 33 -4 24 2 23 -3 26 2 5 -1 13 1 16 -2 14 1 27 -2 15 1 28 -3 15 2 21 -2 15 1 10 -2 18 2 2 -1 7 1 9 -1 7 1 20 -2 7 1 25 -2 6 1 20 -2 7 1 9 -1 7 1 2 -1 7 1 8 -1 3 0 16 -1 4 1 21 -1 3 0 21 -1 3 0 13 -1 4 1 4 -1 3 0

135

Table 4.9: Alternating and mean stress (MPa) components for the nodes in region II.
Loading at crank end Node number (Fig. 4.12 b) 11347 11333 11335 11399 11381 11431 11408 11457 11476 11482 11509 11493 11543 11520 11563 11582 xa xm 179 185 191 194 197 198 198 198 198 198 198 197 196 194 193 193 -3 -1 -1 0 1 1 0 0 0 0 0 0 -1 -3 -6 -9 Rx -1.03 -1.01 -1.01 -1.00 -0.99 -0.99 -1.00 -1.00 -1.00 -1.00 -1.00 -1.00 -1.01 -1.03 -1.06 -1.09 ya ym za zm xya xym xa 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7 2 1 1 1 1 0 0 0 0 1 1 0 1 4 7 1 1 1 0 0 0 0 0 0 0 1 1 0 0 -2 -4 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 xm Loading at pin end Rx -1.08 -1.07 -1.06 -1.05 -1.05 -1.05 -1.06 -1.06 -1.05 -1.05 -1.05 -1.06 -1.08 -1.12 -1.17 -1.23 ya ym za zm xya xym 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 2 0 -1 1 1 0 0 1 1 1 1 1 0 7 16 0 0 0 1 0 1 0 0 1 1 1 1 1 -2 -2 -6 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

177 -7 182 -7 186 -6 189 -5 191 -5 192 -5 193 -6 193 -6 193 -5 193 -5 192 -5 191 -6 188 -8 182 -10 176 -14 172 -18

136

Table 4.10: Alternating and mean stress (MPa) components for the nodes in region III.
Loading at crank end Node number (Fig. 4.12 c) 24167 24168 24164 24160 24165 24163 24161 24117 24138 24088 24103 24073 24081 24085 24084 xa xm 209 208 208 208 209 208 210 200 195 201 204 179 111 92 115 -5 -8 -7 -7 -4 -1 -3 22 52 83 121 129 88 73 91 Rx -1.05 -1.08 -1.06 -1.06 -1.03 -1.01 -1.03 -0.81 -0.58 -0.42 -0.26 -0.16 -0.12 -0.12 -0.12 ya ym za zm xya xym xa 1 0 0 1 1 1 0 2 5 11 18 24 13 14 13 -1 0 0 -1 -1 -1 0 0 1 5 11 17 10 12 10 7 11 10 7 11 12 8 14 9 7 29 32 2 4 5 -8 -11 -11 -8 -12 -13 -9 -14 -9 5 25 27 1 -5 4 3 1 2 1 6 9 6 21 32 46 59 61 37 34 41 -1 0 0 0 0 1 0 3 9 20 36 43 28 26 32 200 193 180 196 205 211 207 256 319 391 463 448 297 254 307 xm -14 -18 0 -16 -12 -7 -11 30 82 139 227 320 310 289 321 Loading at pin end Rx -1.15 -1.20 -1.00 -1.17 -1.12 -1.07 -1.11 -0.79 -0.59 -0.48 -0.34 -0.17 0.02 0.06 0.02 ya ym za zm xya xym 0 0 1 0 1 1 0 2 8 22 43 64 40 42 42 0 0 -1 0 -1 -1 0 0 2 8 21 51 53 54 52 29 33 34 33 37 40 27 42 26 18 86 123 71 58 76 -18 -17 -18 -22 -20 -22 -15 -26 -26 -16 16 71 80 75 80 3 1 2 1 6 9 6 28 36 91 136 159 105 96 115 -1 -1 0 -1 0 0 -1 4 32 33 68 121 119 116 127

137

Table 4.11: Equivalent von Mises alternating, mean, and maximum stresses for region I.
Loading at crank end Loading at pin end Node eq(max) eq(a) eq(m) m eq(max) eq(a) eq(m) m number (MPa) (MPa) (MPa) (MPa) (MPa) (MPa) (MPa) (MPa) (Fig. 4.12 a) Eq. 4.8 Eq. 4.9 Eq. 4.10 Eq. 4.11 Eq. 4.8 Eq. 4.9 Eq. 4.10 Eq. 4.11 12915 280 175 105 114 156 104 49 62 12887 349 241 101 103 215 141 67 68 12944 359 249 83 75 240 158 84 53 12921 335 234 56 40 243 167 103 32 12973 296 212 29 8 237 172 124 8 12966 260 196 11 -13 228 181 150 -5 13015 237 188 11 -21 223 188 167 -15 13013 220 198 10 -18 213 199 177 -16 13023 222 190 11 -20 216 193 176 -17 13021 222 191 11 -19 215 193 175 -18 13016 219 200 10 -17 213 201 178 -14 13026 204 207 10 -12 202 208 181 -14 13025 207 197 10 -15 204 200 178 -14 13024 212 189 10 -18 208 193 176 -15 13022 212 188 11 -17 207 192 175 -17 13017 212 195 10 -16 208 199 179 -15 13018 209 207 10 -13 206 208 182 -13 1402 195 203 9 -10 192 203 176 -12 1404 195 194 9 -12 192 195 172 -13 1406 198 183 10 -13 196 188 170 -14 1408 198 179 10 -13 196 183 168 -14 1410 198 183 10 -13 196 187 170 -14 1412 194 192 10 -11 193 195 172 -13 1414 196 205 10 -11 194 205 178 -12 11369 182 180 8 -9 179 182 161 -11 11368 185 175 8 -11 183 178 160 -12 11372 184 167 8 -10 182 171 157 -11 11374 186 169 8 -11 183 172 158 -13 11375 183 176 8 -11 181 179 160 -11 11373 182 185 8 -9 181 186 164 -11

138

Table 4.12: Equivalent von Mises alternating, mean, and maximum stresses for region II.
Loading at crank end Loading at pin end Node eq(max) eq(a) eq(m) m eq(max) eq(a) eq(m) m number (MPa) (MPa) (MPa) (MPa) (MPa) (MPa) (MPa) (MPa) (Fig.4.12 b) Eq. 4.8 Eq. 4.9 Eq. 4.10 Eq. 4.11 Eq. 4.8 Eq. 4.9 Eq. 4.10 Eq. 4.11 11347 179 176 4 -3 173 174 7 -7 11333 185 184 2 -1 176 181 7 -7 11335 190 190 1 0 180 186 6 -6 11399 194 194 0 0 184 189 5 -5 11381 197 196 1 1 186 191 5 -5 11431 198 197 1 1 187 192 5 -5 11408 198 198 0 0 187 193 6 -6 11457 198 198 0 0 187 193 6 -6 11476 198 198 0 0 188 193 5 -5 11482 198 198 0 0 188 193 5 -5 11509 198 198 1 1 187 192 5 -5 11493 197 197 1 1 184 190 6 -5 11543 195 196 1 -1 180 187 8 -7 11520 192 194 3 -3 173 182 9 -12 11563 190 191 5 -7 167 173 13 -16 11582 190 189 7 -13 166 165 16 -24

139

Table 4.13: Equivalent von Mises alternating, mean, and maximum stresses for region III.
Loading at crank end Loading at pin end Node eq(max) eq(a) eq(m) m eq(max) eq(a) eq(m) m number (MPa) (MPa) (MPa) (MPa) (MPa) (MPa) (MPa) (MPa) (Fig. 4.12 c) Eq. 4.8 Eq. 4.9 Eq. 4.10 Eq. 4.11 Eq. 4.8 Eq. 4.9 Eq. 4.10 Eq. 4.11 24167 212 205 7 -14 214 187 18 -32 24168 211 202 10 -18 204 178 17 -34 24164 212 203 10 -18 211 166 17 -18 24160 209 204 7 -15 213 182 20 -38 24165 218 203 10 -16 228 188 17 -33 24163 221 203 12 -15 241 195 19 -29 24161 216 206 9 -12 221 195 14 -26 24117 239 195 31 8 329 241 49 4 24138 263 196 58 44 440 308 111 58 24088 292 207 85 92 557 403 155 131 24103 327 207 120 157 703 464 239 263 24073 316 185 131 173 785 452 334 441 24081 218 122 96 98 624 304 320 443 24085 189 102 84 80 565 264 302 417 24084 228 128 100 105 659 321 338 452

140

All dimensions are in inches unless otherwise specified

Radius fillet: 0.1 TYP

3.436o

0.630 1.299

7.373

1.759 0.780 5.559

141
variable fillet radius: 0.2 0.04 2.189

1.889 with a chamfer of 0.050 2.207

1.167

0.200 with a chamfer of 0.030

R 0.300 0.390

0.075 TYP R 1.366 TYP 0.930 R 2.075 TYP with a chamfer of 0.010

Figure 4.1: Detailed generated geometry of the connecting rod.

141

N1

N2

N3

N4 X

Figure 4.2: Schematic of parabolic tetrahedron element shape used for FEA meshing.

142

(a)

(b)

X Z
143

(c)

(d)

Figure 4.3: Loading and constraint used for FEA of the connecting rod. (a) Tension loading at crank end, (b) compression loading at crank end, (c) tension loading at pin end, and (d) compression loading at pin end.

143

Figure 4.4: von Mises stress locations used for mesh size sensitivity analysis (loading at crank end and constraint at pin end).

144

7.77 mm (0.306 in) default

2.54 mm (0.1 in)

1.52 mm (0.06 in)

1.27 mm (0.05 in)

Figure 4.5: Schematic of connecting rods at different element mesh sizes.

145

175 150 125

von Mises stress (MPa)

100 75 50 25 0
7.77 2.54 1.52 1.27

Location 3 Location 2 Location 1

Element size (mm)

Figure 4.6: Plot of von Mises stress vs. element size at 26688 N (6000 lbs) tension load at the crank end for locations identified in Figure 4.4.

146

(a)

(b)

147

(c)

(d)

Figure 4.7: von Mises stress (MPa) contours of connecting rod at 26688 N (6000 lbs) load magnitude. (a) Tension at the crank end, (b) compression at the crank end, (c) tension at the pin end, and (d) compression at the pin end.

147

(a) 0.302 mm (0.012 in)

(b) 0.114 mm (0.005 in)

148

(c) 0.354 mm (0.014 in)

(d) 0.121 mm (0.005 in)

Figure 4.8: Displacement contours at magnification factor of 5 in X-direction. (a) Tension at the crank end, (b) compression at the crank end, (c) tension at the pin end, and (d) compression at the pin end.

148

Figure 4.9: Stress distribution profile at the crank end transition region in the loading direction (x-axis).

149

Tension stress, x (MPa) 192.6 206.2 198.2 200.9 188.3

Calculated stress S = P/A (MPa) 197.7 197.7 197.7 197.7 197.7

Tension stress, x (MPa) 185.4 203.3 198.4 198.5 192.4

150

Figure 4.10: Comparison between calculated and FEA obtained stress values at 26688 N (6000 lbs) tension load.

150

Figure 4.11: Locations of strain gauges on front of the connecting rod (two gauges are also used on the backside, behind gauges 1 and 2).

151

Region I Region II Region III

Figure 4.12: Locations of nodes where stresses were obtained.

152

12915

1414

12915

12887 12944 12921 12973 12966 13015 13016 13018 13026 13025 13024 13022 13017 13021 13023 13013

1402 1404 1406 1408 1410 1412 1414 11369 11368 11372 11374 11375 11373

Figure 4.12 (a): Region I (transition region between crank end and shank) nodes.

153

11347 11582

11347 11333 11335 11399 11381 11431 11408 11457 11476 11482 11509 11493 11543 11520 11563 11582

Figure 4.12 (b): Region II (region between crank end transition and pin end transition) nodes.

154

24085

24164

24165 24163 24167 24168 24164 24160 24161 24117 24138 24088 24103 24073

24081

Figure 4.12 (c): Region III (transition region between pin end and shank) nodes.

24085 24084

Figure 4.12 (c): Region III (transition region between pin end and shank) nodes.

155

CHAPTER 5 FATIGUE BEHAVIOR, COMPARISONS, AND LIFE PREDICTIONS

5.1 Material Fatigue Behavior and Comparisons
5.1.1 Material, specimen, and test equipment

Powder forged connecting rods manufactured by OEM #1 in the United States used in a mini-van (2.4 L engine) and forged steel connecting rods manufactured by OEM #2 in Japan used in a midsize car (2.3 L engine) were used to obtain test specimens in this study. Specimens obtained from C-70 connecting rods were also included for comparison, because this material is a currently used steel for crackable (also referred to as splittable) connecting rods. The as-received forged steel, powder metal and C-70 steel connecting rods are shown in Figure 5.1. Table 5.1 summarizes the chemical composition of the forged steel, powder metal material and C-70 steel. The chemical composition for C-70 steel was obtained from 2 inch diameter bars, which were not forged. Photomicrographs of the microstructure for forged steel, powder metal and C-70 steel were obtained using an optical microscope. Two specimens were prepared for each material. The specimens were mounted in the longitudinal direction. The grains are shown at 500X magnification in Figures 5.2 (a), 5.3 (a) and 5.4 (a) for forged steel, powder metal, and C-70 steel, respectively. The microstructure was observed after etching. The inclusions/voids for the three materials at 100X magnification are shown in

156

Figures 5.2 (b), 5.3 (b) and 5.4 (c). According to ASTM standard E45, method A, the inclusion rating of type D inclusions was found for forged steel and powder metal materials. Type D inclusions with severity level of >4 was found for forged steel, and >3 for powder metal. The inclusion rating for C-70 steel revealed type A inclusion with severity level of >1, and type D inclusion with severity level of >2. Forged steel specimens from finished connecting rods, and powder metal and C70 steel specimens from unfinished connecting rods were obtained. The tests were conducted at room temperature. Flat plate specimens (rectangular cross-sections) with the longitudinal axis coinciding with the longitudinal axis of the connecting rods were used for the monotonic and fatigue tests. Two samples were obtained from each connecting rod as shown in Figure 5.5. The specimen configuration and dimensions are shown in Figure 5.6. This configuration was chosen such that the gage section length could be minimized to prevent buckling due to the thin specimen thickness. A detailed finite element analysis was performed to evaluate stress concentration at the specimen radius. The FEA results indicate the stress concentration factor to be about 1.05. This is reasonable, considering the limitation on material thickness. All specimens were machined in the Mechanical, Industrial, and Manufacturing Engineering Machine Shop at the University of Toledo. The specimens were initially rough cut to a rectangular strip using a milling machine and then they were ground. Next, they were inserted into a fixture for cutting the required geometry in CNC milling machine. Using the CNC milling machine, final machining was performed to achieve the final dimensions. After machining, the specimens were checked carefully for flatness on a machinists stone-flat surface. Any specimen with camber exceeding 0.1 mm (0.004 in)

157

from end to end was rejected. The specimen gage section edges for all four surfaces were polished using a hand held buffing rotating mandrel with the polishing marks coinciding with the specimens longitudinal direction. The polished surfaces were carefully examined to ensure complete removal of machining marks in the test section. An INSTRON 8801 closed-loop servo-hydraulic axial load frame in conjunction with a Fast-Track 8800 digital servo-controller was used to conduct the tests. The calibration of this system was verified prior to beginning the test program. The load cell used had a capacity of 50 kN. In order to achieve the best alignment of the specimens, two stops were designed and mounted on the hydraulically operated universal wedge grips with flat faces during all tests. These stops helped to align the specimens ends in series with the load train. Any twisting of the specimens was avoided by using precisely machined blocks to prevent any relative rotation of the grips. Total strain was controlled for all tests using an extensometer rated as ASTM Standard E83-96. The calibration of the extensometer was verified using displacement apparatus containing a micrometer barrel in divisions of 0.0001 in. The extensometer had a gage length of 0.2362 in (6 mm) and was capable of measuring strains up to 10 %. In order to protect the specimens' surface from the knife-edges of the extensometer, M-coat D mixture was used to 'cushion' the attachment. The extensometer was carefully positioned at the center section of the specimen uniform gage section. All tests were conducted at room temperature and were monitored using a digital thermometer. In order to minimize temperature effects upon the extensometer and load cell calibrations, fluctuations were maintained within 2 oC ( 3.6 oF) as required by

158

ASTM Standard E606. Also, the relative humidity of the air was monitored using a precision hydrometer. Significant effort was put forth to align the load train (load cell, grips, specimen, and actuator). Misalignment can result from both tilt and offset between the central lines of the load train components. According to ASTM Standard E606, the maximum bending strains should not exceed 5% of the minimum axial strain range imposed during any test program. For this study, the minimum axial strain range was 0.0035 in/in. Therefore, the maximum allowable bending strain was 175 microstrain. ASTM Standard E1012, Type A, Method 1 was followed to verify specimen alignment. A 0.25 in 0.25 in square cross-section bar with eight strain gages was used for this purpose, and the maximum bending strain was found to be much smaller than that allowed by the ASTM standard.

5.1.2 Test methods and procedures

5.1.2.1 Monotonic tension tests

All monotonic tests in this study were performed using test methods specified by ASTM Standard E8. One specimen was used from each material to obtain the monotonic properties. Due to the limitations of the extensometer, strain control was used only up to 10% strain. After this point, displacement control was used until fracture. A stress versus strain plot was obtained automatically for each test. For the elastic and initial yield region (0% to 0.5% strain), a strain rate of 0.003125 in/in/min was chosen. This strain rate was three-quarters of the maximum allowable rate specified by ASTM Standard E8 for the initial yield region. After yielding (0.5% to 10% strain), the strain rate was increased by a factor of three (i.e., 0.0094

159

in/in/min). After the extensometer was removed, a displacement rate of 0.00846 in/min was used. This displacement rate provided approximately the same strain rate as that used prior to switching control modes. After the tension tests were concluded, the broken specimens were carefully reassembled. The final gage lengths of the fractured specimens were measured with a Vernier caliper having divisions of 0.001 in. Using an optical comparator with 10X magnification and divisions of 0.001 in, the final cross-section dimensions were measured. It should be noted that prior to the test, the initial cross section was measured with this same instrument.

5.1.2.2 Constant amplitude fatigue tests

All constant amplitude fatigue tests in this study were performed according to ASTM Standard E606. It is recommended by this standard that at least 10 specimens be used to generate the fatigue properties. For this study, only four forged steel specimens at four different strain amplitudes ranging from 0.2% to 0.5% were utilized, due to the limited number of available forged steel connecting rods. Twelve powder metal specimens were used at six different strain amplitudes ranging from 0.175% to 0.7%. Thirteen C-70 steel specimens at eight different strain amplitudes ranging from 0.174% to 0.7% were used. A strain amplitude larger than 0.7% was not possible due to specimen buckling limitation. Instron LCF software was used in all tests, except for some long life tests in which Instron SAX software was used after changing to load control mode. During each test, the total strain was recorded using the extensometer output. Test data were automatically recorded throughout each test.

160

There were two control modes used for these tests. Strain control was used in all tests, except for long life (greater than 105 cycles) and run-out tests. For these tests, strain control was used initially to determine the stabilized load. Then load control was used for the remainder of the test. The reason for the change in control mode was due to frequency limitations on the extensometer. For the strain control tests, the applied frequencies ranged from 0.1 Hz to 2.2 Hz. For the load control tests, the frequency was increased to 30 Hz in order to shorten the overall test duration. All tests were conducted using a triangular waveform.

5.1.3 Experimental results and comparisons 5.1.3.1 Monotonic deformation behavior

The properties determined from monotonic tests were the following: yield strength (YS), ultimate tensile strength (Su), percent elongation (%EL), percent reduction in area (%RA), true fracture strength (f), true fracture ductility (f), strength coefficient (K), strain hardening exponent (n) and modulus of elasticity (E). True stress (), true strain (), and true plastic strain (p) were calculated from engineering stress (S) and engineering strain (e), according to the following relationships, which are based on constant volume assumption:

= S (1 + e )

(5.1a) (5.1b)

= ln(1 + e)
p = e =
E

(5.1c)

The true stress ()-true strain () plot is often represented by the RambergOsgood equation:

161

 = e + p =

+ K

1 n

(5.2)

The strength coefficient, K, and strain hardening exponent, n, are the intercept and slope of the best line fit to true stress () versus true plastic strain (p) data in log-log scale:
= K

( )
p

(5.3)

In accordance with ASTM Standard E739, when performing the least squares fit, the true plastic strain (p) was the independent variable and the stress () was the dependent variable. True stress versus true plastic strain behavior of each material is shown in Figure 5.7. To generate the K and n values, the range of data used in this figure was chosen according to the definition of discontinuous yielding specified in ASTM Standard E646. The K is strength coefficient and n is the strain-hardening exponent. The data range used was between 0.6% and 9% strain, which is between yield and ultimate strength. The strength coefficient and strain-hardening exponent for forged steel and powder metal are close to each other. For C-70 steel the strength coefficient and strainhardening exponent are higher than those for the forged steel or powder metal. The true fracture strength, f, was calculated from:

Pf Af

=
w

Pf
f

(5.4)
f

where Pf is the load at fracture, wf and tf are the width and thickness at fracture, respectively. The true fracture ductility, f, was calculated from the relationship based on constant volume:

162

A = ln o A f

1 = ln 1 RA

(5.5)

where Af is the cross-sectional area at fracture, Ao is the original cross-sectional area, and RA is the reduction in area. A summary of the monotonic properties for three materials are provided in Table 5.2. Table 5.3 summarizes the monotonic tensile test results for forged steel, powder metal and C-70 steel. The monotonic stress-strain curves for all three materials are shown in Figure 5.8. Superimposed curves of forged steel, powder metal and C-70 steel are shown in Figure 5.9. As can be seen from Table 5.2, the yield strength of forged steel is 19% higher than that for the powder metal, while the ultimate tensile strength of the forged steel is 8% higher than that for the powder metal.

5.1.3.2 Cyclic deformation behavior

Transient cyclic response describes the process of cyclic-induced change in deformation resistance of a material. Data obtained from constant amplitude straincontrolled fatigue tests were used to determine this response. Plots of stress amplitude variation versus applied number of cycles can indicate the degree of transient cyclic softening/hardening. Also, these plots show when cyclic stabilization occurs. The transient response normalized on the rectangular plot for the three materials are shown in Figure 5.10, while semi-log plots are shown in Figure 5.11. Even though multiple tests were conducted at the same strain amplitudes, data from one test at each strain amplitude tested are shown in these plots. Another cyclic behavior of interest was the steady state or stable response. Data obtained from constant amplitude strain-controlled fatigue tests were also used to

163

determine this response. The properties determined from the steady-state hysteresis loops were the following: cyclic strength coefficient (K'), cyclic strain hardening exponent (n'), and cyclic yield strength (YS'). Half-life (midlife) hysteresis loops and data were used to obtain the stable cyclic properties. Similar to monotonic behavior, the cyclic true stress-strain behavior can be characterized by the Ramberg-Osgood type equation:
e = + 2 2 2
p

= 2E

+ 2K '

1 n
'

(5.6)

The cyclic strength coefficient, K', and cyclic strain hardening exponent, n', are the intercept and slope of the best line fit to true stress amplitude (/2) versus true plastic strain amplitude (p/2) data in log-log scale:
p = K ' 2 2
n'

(5.7)

In accordance with ASTM Standard E739, when performing the least squares fit, the true plastic strain amplitude (p/2) was the independent variable and the stress amplitude (/2) was the dependent variable. The true plastic strain amplitude was calculated by the following equation:
2
p

2 2E

(5.8)

These plots for three materials are shown in Figure 5.12. To generate the K and n values, the range of data used for forged steel was chosen from 0.0025 a 0.005. The range of data used was chosen from 0.0020 a 0.007 for powder metal, and 0.0025 a 0.007 for the C-70 steel.

164

The cyclic stress-strain curve reflects the resistance of a material to cyclic deformation and can be vastly different from the monotonic stress-strain curve. The cyclic stress-strain curves are shown in Figure 5.13. Superimposed curves of the three materials are shown in Figure 5.14. From this figure it is observed that the curves for forged steel and powder metal are close to each other, while the curve for C-70 steel falls below those for forged steel and powder metal. In Figure 5.15, superimposed plots of monotonic and cyclic stress-strain curves are shown for the three materials. Figure 5.15(a) shows cyclic softening behavior for forged steel. Figure 5.15(b) shows the powder metal initially cyclic softens, but then it cyclic hardens at strain amplitude higher than 0.5%. Figure 5.15(c) shows the C-70 steel initially cyclic softens, and after about 0.8% strain, the monotonic and cyclic curves coincide. Figure 5.16, shows superimposed plots of the monotonic and cyclic curves for the three materials. Figure 5.17 shows composite plots of the steady-state (midlife) hysteresis loops for forged steel, powder metal and C-70 steel. Even though multiple tests were conducted at the same strain amplitudes, the stable loop from only one test at each strain amplitude is shown in these plots.

5.1.3.3 Constant amplitude fatigue behavior

Constant amplitude strain-controlled fatigue tests were performed to determine the strain-life curves. The following equation relates the true strain amplitude to the fatigue life:
' e p f = + = 2N f 2 2 2 E

) + (2 N )
b ' f f

(5.9)

165

where f' is the fatigue strength coefficient, b is the fatigue strength exponent, f' is the fatigue ductility coefficient, c is the fatigue ductility exponent, E is the monotonic modulus of elasticity, and 2Nf is the number of reversals to failure (which was defined as a 50% load drop, as recommended by ASTM Standard E606). The fatigue strength coefficient, f', and fatigue strength exponent, b, are the intercept and slope of the best line fit to true stress amplitude (/2) versus reversals to failure (2Nf) data in log-log scale:
= 2
' f

(2 N )
f

(5.10)

In accordance with ASTM Standard E739, when performing the least squares fit, the stress amplitude (/2) was the independent variable and the reversals to failure (2Nf) were the dependent variable. The plots for all three materials are shown in Figure 5.18. To generate the f' and b values, the range of data used for forged steel values was 0.002

a 0.005, while this range for powder metal and C-70 steel was 0.002 a 0.007.
Superimposed curves of the three materials are shown in Figure 5.19. From this figure, it can be seen that the slope b for powder metal and C-70 steel is steeper than for the forged steel. The fatigue strength coefficient (f') of forged steel is 1188 MPa, whereas for powder metal it is 1493 MPa and for C-70 steel it is 1303 MPa. Therefore, the powder metal and C-70 steel have somewhat higher strength at short life, compared to forged steel. Forged steel shows higher strength at longer lives. Table 5.4 lists the summary of constant amplitude completely reversed fatigue test results for forged steel and powder metal. Table 5.5 lists the summary of constant amplitude completely reversed fatigue test results for C-70 steel.

166

The fatigue ductility coefficient, f', and fatigue ductility exponent, c, are the intercept and slope of the best line fit to calculated true plastic strain amplitude (p/2) versus reversals to failure (2Nf) data in log-log scale:
2
p

calculated

' f

(2 N )
f

(5.11)

In accordance with ASTM Standard E739, when performing the least squares fit, the calculated true plastic strain amplitude (p/2) was the independent variable and the reversal to failure (2Nf) was the dependent variable. The calculated true plastic strain amplitude was determined from Equation 5.8. These plots for the three materials are shown in Figure 5.20. To generate the f' and c values, the range of data used in this figure was 0.0025 a 0.005 for the forged steel, 0.002 a 0.007 for the powder metal, and 0.0025 a 0.007 for the C-70 steel. The ductility of the material under cyclic loading can be observed from these graphs. Figure 5.21 shows superimposed plots of the three materials, and as can be seen the three curves are close to each other. The true strain amplitude versus reversals to failure plots are shown in Figure 5.22. These plots display the strain-life curve (Eqn. 5.9), the elastic strain portion (Eqn. 5.10), the plastic strain portion (Eqn. 5.11), and the superimposed fatigue data. At large strains, better fatigue resistance depends more on ductility, whereas at smaller strains, it depends more on strength. Figure 5.23 shows superimposed total strain-life curves for the three materials. It can be seen that the total strain-life curve is higher for forged steel, about a factor of 2 in life at short lives and about a factor of 7 in life at long lives. Summary of the cyclic properties for the three materials are provided in Table 5.2. Tables 5.4 and 5.5

167

provide the summary of the fatigue test results for forged steel, powder metal, and C-70 steel.

5.2 Component Fatigue Behavior and Comparisons

Alignment of connecting rod was carefully made before the axial fatigue tests were conducted. Misalignment can result from tilt and offset of fixture pins inserted in the connecting rod. To minimize misalignment, a ball joint pin was introduced at the bottom of the test fixture. However, this makes the system less rigid, not allowing high frequency testing. Schematic drawing of test fixtures with ball joint used for testing of the connecting rods is shown in Figure 5.24 and a photo is shown in Figure 5.25. The same fixture was employed to test both forged steel and powder metal rods. Alignment was verified with four strain gauges (2 in front and 2 in back) attached to the middle of the shank of a connecting rod and loaded in tension. For installation of rods in the test fixture, pressed-in wrist pins were used. It was necessary to have at least 0.001 press fit between the wrist pins and the rods to ensure the pins did not loosen or move during testing. The axial fatigue tests were performed at room temperature and under constant amplitude load control. A sinusoidal wave form was used in all the tests. Loading frequency varied from 2 to 5 Hz with a lower value used for high load levels and a higher value for the low load levels. Each test was performed under a constant load frequency condition. The fatigue tests were conducted at three load levels for both forged steel and powder metal rods, resulting in lives between 4 x 104 cycles and >3 x 106 cycles. Three tests were conducted at each load level, to assess variability and scatter. All the

168

connecting rods were tested at a load ratio of R = -1.25. This ratio was obtained from FEA at the highly stressed locations in the transition regions. Prior to the testing, all the connecting rods were numbered, weighted, and diameters were measured, as listed in Table 5.6. The range of weight for forged steel connecting rods was 454 to 456 grams, whereas, the range for powder metal connecting rods was 566 to 571 grams. The powder metal connecting rods are 25% heavier than forged steel connecting rods. Table 5.7 summarizes the loading conditions, area at fracture, cycles to failure and failure locations for all the tested rods. The fatigue life of the connecting rod was defined as the number of cycles endured until it completely separated into two pieces. The dominant fracture location of the powder metal rods was near the transition region to the pin end. Only one powder metal rod fractured in the crank end transition. The forged steel rods mainly failed in the transition to the crank end region. The load amplitude versus cycles to failure for both forged steel and powder metal connecting rods is shown in Figure 5.26. As can be seen, at the same load amplitude the lives for forged steel connecting rods are higher than the powder metal connecting rods. This is in spite of the fact that the powder metal rods are 25% heavier than the forged steel rods. Figure 5.27 shows the stress amplitude versus number of cycles to failure. In order to obtain the stress, cross sectional areas of 131mm2 for the forged steel and 159 mm2 for the powder metal connecting rods were used. These areas were measured from the crank end transition to the shank straight region. These areas were nearly the same at the two transition regions for each type of connecting rod. Figure 5.27 indicates the scatter in fatigue life at each load level to be less than a factor of 3. Forged steel connecting rod fatigue strength, defined at 106 cycles, is 387 MPa, whereas

169

for the powder metal connecting rod it is 282 MPa. Therefore, the forged steel connecting rod exhibits 37% higher fatigue strength, as compared with the powder metal connecting rod. As can be seen from Figure 5.27, this increased strength results in about two orders of magnitude longer life for the forged steel connecting rod. With longer lives, the two component S-N lines diverge, so that the difference in fatigue performance between the two connecting rods increases. Typical macroscopic appearance of fracture surfaces for forged steel connecting rods are shown in Figures 5.28 through 5.30. Cracks started normal to the I-beam axis and subsurface. Some cup and cone phenomena were observed for the forged steel connecting rods, indicating ductility. Figures 5.31 and 5.32 show typical subsurface crack nucleation sites at 50X. The SEM photomicrographs of some fracture surfaces are shown in Figures 5.33 and 5.34. The internal defects are inclusions, which caused crack initiation, as seen from Figure 5.33 (a). For FS9 connecting rod, the SEM image shows crack propagation along the web of the I beam of the rod (Figure 5.34). The longer crack was sustained, since the load amplitude was low. Figures 5.35 through 5.37 show typical powder metal connecting rod fracture surfaces. The crack origin appeared to be from the surface or subsurface as it showed some discoloration. The typical surface crack originates from the rib of the I-beam section. The fracture surfaces did not indicate any sign of gross plastic deformation, as expected due to the lower ductility of the material, as compared with the forged steel. Figure 5.38 shows the fracture initiation site for PM7 connecting rod at 50X. Figure 5.39 shows the SEM of the fracture surface, and close-up of the crack initiation site. Decarburization/porosity may be the root cause(s) of the crack origin. As discussed by

170

Illia and Chernenkoff (2001), powder metal connecting rods show porosity, oxide penetration, decarburization or other defects which cause the fatigue crack to initiate at the surface or subsurface. Fatigue crack can started from a minute defect. They further mention that most of the fatigue cracks start subsurface, due to compressive residual stress induced by shot peening. Additional discussions of surface or subsurface cracks associated with residual stresses were provided by Chernenkoff et al. (1995). They state that shot peening at 20A intensity provides a residual stress of 500 MPa to 400 MPa at a depth of 0.1 to 0.15 mm. Over peening, however, forms cracks at the surface. For shot peened connecting rods the crack initiation sites were observed to be in the subsurface region, where the residual stress is tensile. They found that at 20A, the depth of peening was approximately 0.37 mm and fracture origin was 1.5 mm below the surface. Post-delubrication peening is discussed by Liu and Hathaway (2001). They mention that in powder metal connecting rods the crack initiates below the surface of an I-beam inner rail tip. They also mention that shot peening removed the oxides on the rod surface, but the oxides entrapped in the channels and pores below the surface remained, causing the crack to originate below the surface.

5.3 Life Predictions and Comparisons with Experimental Results

The stress-based approach is based on analyzing the stress-life curve. It is typically used for life prediction of components subject to high cycle fatigue, where stresses are mainly elastic, as in the case of connecting rods. This approach emphasizes nominal stress rather than local stresses, and it employs elastic stress concentration factor,

171

empirical modification for surface finish effect, an equation such as the modified Goodman equation for residual and/or mean stress effects, and the material stress-life curve. The stress-life curve can be estimated from fatigue strength coefficient, f, and fatigue strength exponent, b, of the material. For smooth component without a stress concentration, fatigue limit is estimated from: Sf = f (2x106)b (5.12)

The first line in Figure 5.40 connects the stress amplitude at 1 cycle to Sf at 106 cycles. To evaluate the effects of the notch or stress concentration on the fatigue behavior, the fatigue notch factor, Kf, is calculated according to:
Kf = 1 +

Kt 1 1+ / r

(5.13)

where the characteristic length, , depends on the material and r is the notch radius. The Kt value is obtained from FEA. S-N line for notch effect connects fatigue strength at 1 cycle to Sf/Kf at 106 cycles, as shown in Figure 5.40. Note that no effect of stress concentration at 1 cycle is predicted due to yielding. The modified Goodman equation is often used to calculate the stress amplitude for mean and/or residual stress effect. The combined effect of mean stress and stress concentration effect is represent by:
S Sa + m =1 Su Sf /Kf

(5.14)

where, Sa is the stress amplitude (alternating stress), Sm is the mean stress, Su is the ultimate tensile strength, and Sf is unnotched fatigue strength with no mean stress.

172

To account for surface finish effect, fatigue strength, Sf, of a smooth component at 106 cycles is multiplied by the surface finish factor, Ks. To account for both the surface finish and stress concentration effects on the fatigue life, Sf at 106 is changed to (Ks Sf/Kt). These lines for completely reversed loading condition, R = -1, are shown in Figure 5.40. Note that if multiaxial stresses are present, equivalent (i.e. von Mises) alternating and mean stresses, as discussed in section 4.2, should be used, rather than uniaxial stresses. However, since the stress state at the critical locations are essentially uniaxial, as discussed in section 4.2, this effect was not considered.

5.3.1 Stress-life (S-N) curves for forged steel connecting rod

To predict the life of the forged steel connecting rod, material S-N line is used: Sa = f (2Nf)b = 1188 (2Nf)-0.0711 = 1131 (Nf)-0.0711 (5.15)

where, f and b are listed in Table 5.8. The material S-N curve for forged steel is constructed by connecting Sa at 1 cycle, Sa = 1131 MPa, to Sa at 106 cycles, Sa = 1131 (106)-0.0711 = 424 MPa. This line is shown in Figure 5.41. Mean stress is calculated from the load ratio, R = -1.25. R=
Pmin = - 1.25 Pmax

(5.16)

Pm =

Pmax + Pmin P + (1.25) Pmax = max 2 2

= - 0.125 Pmax

(5.17) (5.18) (5.19)

Pmax = Pa + Pm = Pa 0.125 Pmax or Pmax = 0.89 Pa Pm = - 0.125 (0.89) Pa = - 0.111 Pa

173

where, Pmin is the minimum load used in compression, Pmax is the maximum load used in tension, Pm is the mean load, and Pa is the alternating load. Based on this equation, Sm = 0.111 Sa . The effect of notch is accounted through the use of a fatigue notch factor Kf, calculated from equation 5.13. A Kt value of 1.18 is used, obtained from FEA as described in section 4.2. The characteristic length, , is 0.04 mm for forged steel at an ultimate tensile strength of 938 MPa (Stephens et al., 2000). Kf is calculated to be 1.18 by equation 5.13, which is the same as Kt. For notches with large radii, the Kf value is close or equal to Kt. The modified Goodman equation (equation 5.14) was used for combined mean stress and notch effects, resulting in Sa = 375 MPa. Figure 5.41 shows the S-N line with the combined notch and mean stress effects. This figure also shows the S-N line with forged surface effect. Surface effects are caused by microstructure, residual stresses, chemical composition, and surface roughness. This factor has more influence at long lives. The surface factor, Ks, depends on ultimate tensile strength and hardness. For asforged surface finish with Su of 938 MPa, Ks = 0.3 (Stephens et al., 2000). The surface finish effect on fatigue strength is then obtained as the product of notched fatigue strength and surface finish factor (Kf Sf), calculated to be 113 MPa as shown in Figure 5.41. The surface compressive residual stresses from shot peening, however, nullify detrimental effects of forged surface finish, as evidenced from subsurface crack nucleation observations.

174

5.3.2 Stress-life (S-N) curves for powder metal connecting rod

For powder metal connecting rod, S-N lines are obtained in the same manner as for forged steel connecting rod S-N lines. Figure 5.42 shows the S-N lines for powder metal connecting rod. The powder metal fatigue properties used are listed in Table 5.2. For powder metal connecting rod, the stress concentration was higher at the transition to the pin end. This was manifested by the fact that most powder metal connecting rods failed in this region. A stress concentration factor of 1.30 at this transition region was estimated from mechanics of materials analytical equations based on the powder metal connecting rod geometry. For powder metal connecting rods, the surface factor, Ks = 0.31 was used, as obtained for a forged surface finish and with ultimate tensile strength of 866 MPa (Stephens et al. 2000). The same discussion with regards to surface finish and surface compressive residual stress effects also applies to powder metal connecting rods.

5.3.3 Comparisons with experimental results

The comparison of experimental data and predicted S-N lines for forged steel and powder metal connecting rods are shown in Figures 5.43 and 5.44, respectively. The experimental data are close to the predicted S-N curves for both forged steel and powder metal connecting rods. As stated earlier, the effect of forged surface finish was not taken into account for predictions due to surface compressive residual stresses. The experimental fatigue lives against predicted lives for both forged steel and powder metal connecting rods are plotted in Figure 5.45. The 45o line shown represents perfect correlation line. Data points above this line indicate over prediction, while data points below this line represent conservative predictions. Factor of 2 and 3 lines are

175

also plotted. The majority of data are within a factor of 2. This indicates that the predicted lives agree very well with the experimental lives.

5.4 Discussion of Results and Potential for Improvements

Comparing Figures 5.19 and 5.27, it can be seen that the difference in fatigue performance between forged steel and PM connecting rods shown in Figure 5.27 (about two orders of magnitude) is more than that observed from specimen testing of the two materials shown in Figure 5.19 (an order of magnitude). This could perhaps be explained by a better response of the forged steel connecting rod to the shot peening process, and higher surface compressive residual stresses, resulting in subsurface crack nucleation. In addition, surface and subsurface defects such as porosity and oxides in the PM connecting rod can affect its fatigue behavior more than the defects in the forged steel connecting rod. Both Figure 5.27 for the component behaviors and Figure 5.19 for the material behaviors indicate increased difference between the fatigue behaviors of the two materials/processes with longer lives. Directional property of the material is another important factor in its durability performance. In hot forging, the directional solidification of grains is desired for better fatigue strength. In PM connecting rods, there is no strengthening effect obtained by forging, whereas the grain flow in one direction (primary loading direction) during the forging process of steel connecting rods can provide additional fatigue strength. The fracture splitting capability of powder metal connecting rods has been a dominant feature for the last decade. An important cost reduction step for forged steel connecting rods is the development of fracture splitting process of C-70 steel. Recent

176

development of new crackable materials such as C-70 and microalloy steels, provide cost efficient methods for forged steel connecting rods from the manufacturing and cost points of view. The new crackable steels eliminate a number of machining steps. Also, increased tool life has been reported. As discussed in the literature review, recent studies show that the fatigue strength level is higher for the splittable forged steel connecting rods, as compared with powder metal connecting rods. In addition, the development of fracture splittable steel connecting rod has resulted in a total cost reduction of up to 25%, as compared with conventional forged steel connecting rod.

177

Table 5.1: Chemical composition by percent weight for three materials. Elements Forged steel 0.33 0.02 0.4 0.07 0.21 0.084 <0.005 0.99 0.04 0.47 0.03 <0.01 <0.005 <0.005 Powder metal 0.4-0.64 0.04 max 0.03 max 0.1 max 1.8-2.2 1070 (C-70)1

C P Si Ni Cu Va Ti Mn S Cr Mo Al Cb Zr B Zn Co

0.3-0.6 0.18 max 0.09 max 0.05 max

0.7 0.014 0.28 0.01 0.015 0.031 33 PPM 0.98 0.024 0.11 0.049 0.031

33 PPM 34 PPM 35 PPM

[1] Khalil and Topper, 1999

178

Table 5.2: Summary of mechanical properties for three materials.

Monotonic Properties Modulus of elasticity (measured), E, GPa (ksi): Yield strength (0.2% offset), YS, MPa (ksi): Ultimate strength, Su, MPa (ksi): Percent elongation, %EL (%): Percent reduction in area, %RA (%): Strength coefficient, K, MPa (ksi): Strain hardening exponent, n: True fracture strength, sf, MPa (ksi): True fracture ductility, ef (%): Hardness, Rockwell C (HRC) Hardness, Brinell Cyclic Properties Fatigue strength coefficient, f', MPa (ksi): Fatigue strength exponent, b: Fatigue ductility coefficient, f' : Fatigue ductility exponent, c: Cyclic yield strength, YS', MPa (ksi) Cyclic strength coefficient, K' , MPa (ksi): Cyclic strain hardening exponent, n': Sf = f' (2Nf)b at Nf = 106, MPa (Ksi) Average E' GPa (Ksi) 1,188 -0.0711 0.3576 -0.5663 620 1,397 0.1308 423 204 (61.4) (29586) (89.9) (202.6) (172.3) 1,493 -0.1032 0.1978 -0.5304 609 2,005 0.1917 334 197 (48.4) (28571) (88.3) (290.8) (216.5) 1,303 -0.0928 0.5646 -0.5861 528 1,739 0.1919 339 193 (49.2) (27993) (76.5) (252.2) (188.9) Forged steel 201 700 938 24% 42% 1,400 0.122 1,266 54% 28 272 (184.0) (203.0) (29157) 102 (136.0) Powder metal 199 588 866 23% 23% 1,379 0.152 994 26% 20 223 (144.1) (200.0) (28861) (85.3) (125.5) C-70 steel 212 574 966 27% 25% 1,763 0.193 1141 28% 23 241 (165.5) (255.7) (30674) (83.2) (140.1)

179

Table 5.3: Summary of monotonic tensile test results.

Material Forged steel Powder metal C-70 steel CR13 PM-3 Spec. ID FS-3 to, mm (in.) 2.67 (0.105) 2.63 (0.104) 2.59 (0.102) wo, mm (in.) 2.63 (0.104) 2.58 (0.102) 2.60 (0.102) tf, mm (in.) 2.09 (0.082) 2.28 (0.090) 2.24 (0.088) wf, mm (in.) 1.97 (0.077) 2.30 (0.090) 2.26 (0.089) Lo, mm (in.) 6.56 (0.26) 6.31 (0.25) 6.65 (0.26) Lf, mm (in.) 8.13 (0.32) 7.78 (0.31) 8.48 (0.33) E, GPa (ksi) 201.0 (29157.0) 199.0 (28861.0) 211.5 (30674.4) YS (offset=0.2%), MPa (ksi) 700.0 (101.5) 588.4 (85.3) 573.7 (83.2) Su, MPa (ksi) 937.7 (136.0) 865.5 (125.5) 965.8 (140.1) K, MPa (ksi) 1,399.9 (203.0) 1,378.9 (200.0) 1763.0 (255.7) 0.193 27% 25% 28% 0.152 23% 23% 26% n 0.122 %EL 24% %RA 42% f 54% f 1266 (183.6) 994 (144.1) 1141 (165.5)

180

180

Table 5.4: Summary of constant amplitude completely reversed fatigue test results for forged steel and powder metal. Forged steel
At midlife(N50%) Specimen ID FS-4 Test control mode strain Test freq., Hz 0.50 /2, % 0.497% p/2 [a] (calculated), % 0.202% p/2 (measured), % 0.189% /2, MPa (ksi) 609.2 (88.4) FS-1 strain 1.20 0.348% 0.075% 0.062% 565.0 (81.9) FS-5 strain load FS-6 strain load 1.50 22 1.80 22 0.200% 0.001% 0.003% 0.251% 0.026% 0.018% 465.3 (67.5) 411.7 (59.7) m, MPa (ksi) -5.9 (-0.9) -9.5 (-1.4) -68.7 (10.0) 126.9 (18.4) 72,450 344,406 344,416 IGL 203 32,768 58,804 58,918 IGL 202 2N50% ,
[b]

2(Nf)10% ,
[c]

reversals 4,096

reversals 8,790

2(Nf)50% , [d] reversals 8,860

Failure location[e] IGL

E' (GPa) 200

48,870

3,574,326

3,574,338

OGIT

209

Powder metal
At midlife(N50%) Specimen ID PM-6 Test control mode strain Test freq., Hz 0.10 /2, % 0.701% p/2 [a] (calculated), % 0.380% p/2 (measured), % 0.313% /2, MPa (ksi) 663.7 (96.3) PM-9 strain 0.10 0.699% 0.373% 0.320% 674.9 (97.9) PM-1 strain 0.50 0.495% 0.189% 0.159% 634.7 (92.0) PM-18 strain 0.50 0.499% 0.200% 0.163% 618.2 (89.7) PM-2 strain 1.20 0.348% 0.092% 0.062% 528.4 (76.6) PM-14 strain 1.20 0.348% 0.091% 0.066% 532.9 (77.3) PM-7 strain load PM-8 strain 1.50 5.0 1.5 0.249% 0.043% 0.023% 0.250% 0.031% 0.019% 452.9 (65.7) 425.2 (61.7) PM-4 strain load PM-5 strain load PM-10 [f] PM-17 [f] strain load load 2.20 22 1.80 12 2.00 30 30 0.175% 0.000% 0.000% 0.174% 0.009% 0.000% 0.199% 0.015% 0.004% 0.199% 0.020% 0.009% 371.5 (53.9) 380.7 (55.2) 341.7 (49.6) (362.9) 52.4 m, MPa (ksi) -10.9 (-1.6) -2.3 (-0.3) -2.9 (-0.4) -0.1 (-0.0) 10.8 (1.6) 7.6 (1.1) -27.2 (-4.0) -3.3 -(0.5) -68.4 -(9.9) 39.4 (5.7) 32.0 (4.6) 0.0 (0.0) 1,349,610 IGL 194 15,600 >10,000,000 No Failure 200 10,000 573,118 573,120 IGL 200 155,000 635,932 635,938 IGL 205 65,536 147,580 154,612 IGL 200 32,100 100,980 100,984 IGL 200 16,384 30,638 31,674 IGL 195 16,384 28,998 30,316 IGL 198 2,048 5,150 5,168 IGL 195 2,048 4,096 6,074 IGL 198 512 1,428 1,444 IGL 195 2N50% ,
[b]

2(Nf)10% ,
[c]

reversals 1,024

reversals 1,896

2(Nf)50% , [d] reversals 1,924

Failure location[e] IGL

E' (GPa) 180

[a] p/2(calculated)=/2-/2E. [b] N50% is defined as the midlife cycle. [c] (Nf)10% is defined as 10% load drop. [d] (Nf)50% is defined as 50% load drop. [e] IGL= inside gage length; OGIT = outside gage length but inside test section;

181

Table 5.5: Summary of constant amplitude completely reversed fatigue test results for C-70 steel.
At midlife(N50%)
Specimen ID CR3 Test control mode strain Test freq., Hz 0.10 /2, % p/2 [a] (calculated), % 0.410% p/2 (measured), % /2, MPa (ksi) 610.6 (88.6) CR5 strain 0.10 0.699% 0.407% 0.374% 617.4 (89.5) CR7 strain 0.20 0.600% 0.327% 0.292% 576.8 (83.7) CR1 strain 0.50 0.499% (0.241%) (0.208%) 546.0 79.2 CR6 strain 0.30 0.500% 0.248% 0.200% 532.2 (77.2) CR11 strain 0.40 0.402% 0.170% 0.132% 489.8 (71.0) CR2 strain 1.20 0.299% 0.087% 0.057% 447.8 (64.9) CR10 strain 1.00 0.299% 0.086% 0.064% 450.5 (65.3) CR12 strain load CR4 strain load CR9 strain load CR8 strain load CR14 load 1.00 2.00 1.80 3.00 1.20 5.00 0.20 30.00 30.00 0.174% 0.000% 0.000% 0.174% 0.007% 0.002% 0.200% 0.013% 0.014% 0.199% 0.018% 0.011% 0.250% 0.054% 0.032% 414.3 (60.1) 384.7 (55.8) 395.8 (57.4) 353.5 (51.3) 353.5 (51.3) m, MPa (ksi) -0.1 (0.0) 2.6 (0.4) 3.3 (0.5) 8.9 1.3 42.8 (6.2) 10.7 (1.5) 30.2 (4.4) 34.8 (5.0) 43.4 (6.3) 36.7 (5.3) -39.7 (-5.8) -13.6 (-2.0) 0.0 (0.0) >10,000,000
NO FAILURE

2N50% ,
[b]

2(Nf)10% ,
[c]

reversals 2,048

reversals 4,958

2(Nf)50% , [d] reversals 5,168

Failure location[e] IGL

E' (GPa) 186.7

0.699%

0.367%

2,048

4,484

OGIT

192.0

2,048

6,442

6,612

IGL

191.3

4,096

11,400

11,726

AKP

186.7

4,096

8,932

9,000

IGL

178.6

8,192

25,108

26,438

IGL

182.6

32,768

66,616

66,930

IGL

186.7

19,152

41,314

43,760

IGL

192.0

32,768

183,488

183,488

IGL

191.0

23,202

1,057,672

1,057,674

IGL

205.6

12,000

565,196

565,196

IGL

215.0

8,400

>10,000,000

NO FAILURE

208.0

[a] p/2(calculated)=/2-/2E. [b] N50% is defined as the midlife cycle. [c] (Nf)10% is defined as 10% load drop. [d] (Nf)50% is defined as 50% load drop. [e] IGL= inside gage length; OGIT = outside gage length but inside test section; AKP = at knife position

182

Table 5.6: Summary of dimensions and weights for forged steel and powder metal connecting rods.

Forged steel
Rod no. FS1 FS2 FS3 FS4 FS5 FS6 FS7 FS8 FS9 FS10 FS11 FS12 Range Average Crank end diameter (mm) 47.85 47.88 47.88 47.87 47.82 47.76 47.75 47.82 47.88 47.79 47.77 47.79 47.75-47.88 47.82 Pin end diameter (mm) 23.80 23.79 23.85 23.80 23.82 23.83 23.81 23.82 23.80 23.79 23.81 23.81 23.79-23.85 23.81 Weight (g) 456 454 456 456 456 456 454 456 454 454 456 456 454-456 455.3

Powder metal
PM1 PM2 PM3 PM4 PM5 PM6 PM7 PM8 PM9 PM10 PM11 PM12 PM13 PM14 PM15 PM16 PM17 PM18 PM19 PM20 PM21 PM22 Range Average 53.07 53.08 53.10 53.10 53.16 53.18 53.07 53.07 53.19 53.07 53.07 53.07 53.07 53.08 53.07 53.07 53.07 53.10 53.07 53.09 53.18 53.07 53.07-53.19 53.10 21.89 21.89 21.89 21.93 21.93 21.89 21.89 21.92 21.91 21.90 21.90 21.90 21.89 21.89 21.89 21.93 21.89 21.94 21.93 21.89 21.93 21.89 21.89-21.94 21.91 570 566 567 571 568 571 569 568 567 567 569 568 570 571 569 568 570 570 569 568 568 569 566-571 568.8

183

Table 5.7: Summary of connecting rod fatigue test parameters and results.
Forged steel
Specimen ID FS4 FS3 FS1 FS8 FS9 FS6 FS5 FS12 FS10 2 Hz 2 Hz 2 Hz 4 Hz 4 Hz 4 Hz 5 Hz 5 Hz 4 Hz Frequency Pmax kN (Kips) 51.15 (11.5) 51.15 (11.5) 51.15 (11.5) 44.48 (10.0) 44.48 (10.0) 44.48 (10.0) 42.26 (9.5) 42.26 (9.5) 42.26 (9.5) Pmin kN (Kips) -63.94 (-14.37) -63.94 (-14.37) -63.94 (-14.37) -55.60 (-12.50) -55.60 (-12.50) -55.60 (-12.50) -52.82 (-11.88) -52.82 (-11.88) -52.82 (-11.88) -1.25 -1.25 -1.25 -1.25 -1.25 -1.25 -1.25 -1.25 -1.25 R-ratio Pa kN (Kips) 57.55 (12.94) 57.55 (12.94) 57.55 (12.94) 50.04 (11.25) 50.04 (11.25) 50.04 (11.25) 47.54 (10.69) 47.54 (10.69) 47.54 (10.69) Pm kN (Kips) -6.34 (-1.44) -6.34 (-1.44) -6.34 (-1.44) -5.56 (-1.25) -5.56 (-1.25) -5.56 (-1.25) -5.28 (-1.19) -5.28 (-1.19) -5.28 (-1.19) 87,380 123,815 168,475 606,172 1,362,132 1,663,613 >5.6 x 10
5

Nf

Failure location 7 3 3 3 5 4 7 3 3

Notes

1 2 3

>2.3 x 106 >2.3 x 106

184
PM15 PM14 PM9 PM10 PM7 PM22 PM18 PM16 2 Hz 2 Hz 2 Hz 4 Hz 4 Hz 4 Hz 4 Hz 4 Hz 51.15 (11.5) 51.15 (11.5) 51.15 (11.5) 44.48 (10.0) 44.48 (10.0) 44.48 (10.0) 38.69 (8.7) 38.69 (8.7) -63.94 (-14.37) -63.94 (-14.37) -63.94 (-14.37) -55.60 (-12.50) -55.60 (-12.50) -55.60 (-12.50) -48.37 (-10.87) -48.37 (-10.87) PM6 5 Hz 38.69 (8.7) -48.37 (-10.87) [1] One of the fixture's bolts broke before rod fractured [2] One of the fixture's washers broke before rod fractured [3] Ball joint broke before rod fractured

Powder metal
-1.25 -1.25 -1.25 -1.25 -1.25 -1.25 -1.25 -1.25 -1.25 57.55 (12.94) 57.55 (12.94) 57.55 (12.94) 50.04 (11.25) 50.04 (11.25) 50.04 (11.25) 43.53 (9.79) 43.53 (9.79) 43.53 (9.79) -6.34 (-1.44) -6.34 (-1.44) -6.34 (-1.44) -5.56 (-1.25) -5.56 (-1.25) -5.56 (-1.25) -4.84 (-1.085) -4.84 (-1.085) -4.84 (-1.085) 44,653 79,160 95,279 214,239 215,646 366,272 >3.1 x 10
6

7 7 2 8 8 8

>3.0 x 106 >3.3 x 106

3 4

6 7

184

Table 5.8: Fatigue life prediction parameters for S-N line equation, Sa = A(Nf)B.

Forged steel
Smooth surface Ks = 1 R= -1.0 R= -1.25 unnotched Kf = 1.0 notched Kf = 1.18
A= 1131 MPa A= 1306 MPa B = - 0.078 B = - 0.071

Forged surface Ks = 0.30 R= -1.0 R= -1.25

A= 1131 MPa A= 1306 MPa B = - 0.158 B = - 0.165

A= 1131 MPa A= 1306 MPa A= 1131 MPa A= 1306 MPa B = - 0.17 B = - 0.083 B = - 0.09 B = - 0.177

Powder metal
Smooth surface Ks = 1 R= -1.0 R= -1.25 unnotched Kf = 1.0 notched Kf = 1.3
A= 1390 MPa A= 1691 MPa B = - 0.114 B = - 0.103

Forged surface Ks = 0.31 R= -1.0 R= -1.25

A= 1390 MPa A= 1691 MPa B = - 0.188 B = - 0.2

A= 1390 MPa A= 1691 MPa A= 1390 MPa A= 1691 MPa B = - 0.2 B = - 0.122 B = - 0.133 B = - 0.218

185

(a)

(b)

(c)

Figure 5.1: As-received connecting rods (a) forged steel, (b) powder metal, and (c) C-70 steel.

186

20 m

(a)

100 m

(b) Figure 5.2: Photomicrograph in the longitudinal direction for forged steel. (a) Microstructure at 500X after etching. (b) Inclusions at 100X.

187

20 m

(a)

100 m

(b) Figure 5.3: Photomicrograph in the longitudinal direction for powder metal. (a) Microstructure at 500X after etching. (b) Inclusions at 100X.

188

20 m

(a)

100 m

(b)
Figure 5.4: Photomicrograph in the longitudinal direction for C-70 steel. (a) Microstructure at 500X after etching. (b) Inclusions at 100X.

189

Figure 5.5: Locations of two specimens obtained from each connecting rod.

190

2.67 for forged steel, 2.63 for powder metal, and 2.56 for C-70 steel

Figure 5.6: Specimen configuration and dimensions (all dimensions in mm).

191

10000

FS-3 =1399.99(p) K =1399.9M Pa n = 0.1221

0.1221

True Stress, (MPa)

R2 = 0.9876 1000

Specimen ID: FS-3

100 1.0% True Plastic S train, p (%) 10.0%

(a)
10000 PM -3
0.1517 =1378.9(p) K =1378.9M Pa n = 0.1517

True Stress, (MPa)

R2 = 0.9934

1000

Specimen ID: PM-3

100 1.0% True Plastic S train, p (%) 10.0%

(b)
10000 C-70 Steel
0.1931 =1763(p) K = 1763M Pa n =0.1931

True Stress, (MPa)

R2 = 0.9791 1000

Specimen ID: CR-13

100 1.0% True Plastic S train, p (%) 10.0%

(c) Figure 5.7: True stress versus true plastic strain for (a) forged steel, (b) powder metal, and (c) C-70 steel. 192

1000 Engineering Stress, S (MPa) 900 800 700 600 500 400 300 200 100 0 0% 1% 2% 3% 4% 5% 6% Engineering S train, e (%) 7% 8% 9% 10%

Specimen ID: FS-3

(a)
1000 900 Engineering Stress, S (MPa) 800 700 600 500 400 300 200 100 0 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% 10% Engineering S train, e (%)

Specimen ID: PM-3

(b)
1000 Engineering Stress, S (MPa) 900 800 700 600 500 400 300 200 100 0 0% 2% 4% 6% Engineering S train, e (%) 8% 10%

Specimen ID: CR-13

(c)
Figure 5.8: Monotonic stress-strain curves for (a) forged steel, (b) powder metal, and (c) C-70 steel.

193

1000 900 800 700 600 500 400 300 200 100 0 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% 10%

Engineering Stress, S (MPa)

194

FS PM C 70

Engineering S train, e (%)

Figure 5.9: Superimposed monotonic stress-strain curves for three materials.

194

True Stress Amplitude, /2 (MPa)

700 650 600 550 500 450 400 350 300 250 200 0.0 0.2 0.4 0.6 0.8 1.0
Strain Amplitudes: (top to bottom) a=0.5% a=0.35% a=0.25% a=0.2%

Cycle Ratio, (N/N f )

(a)
True Stress Amplitude, /2 (MPa) 660 610 560 510 460 410 360 310 260 210 160 110 60 10 0.0

Strain Amplitudes: (top to bottom) a=0.7% a=0.5% a=0.35% a=0.25% a=0.2% a=0.175%

0.2

0.4

0.6

0.8

1.0

Cycle Ratio, (N/N f )

(b)
True Stress Amplitude, /2 (MPa) 650 600 550 500 450 400 350 300 250 200 150 0.0 0.2 0.4 0.6 Cycle Ratio, (N/N f ) 0.8 1.0
Strain Amplitudes: (top to bottom) a=0.7% a=0.6% a=0.5% a=0.4% a=0.3% a=0.25% a=0.2%

(c) Figure 5.10: True stress amplitude versus normalized number of cycles for (a) forged steel, (b) powder metal, and (c) C-70 steel.

195

700 650 600 550 500 450 400 350 300 250 200 150 100 50 0 1E+1 1E+2 1E+3 1E+4 Cycles, N 1E+5

True Stress Amplitude, /2 (MPa)

Strain Amplitudes: (top to bottom) a=0.5% a=0.35% a=0.25% a=0.2%

1E+6

1E+7

(a)
700 650 600 550 500 450 400 350 300 250 200 150 100 50 0 1E+1 1E+2 1E+3 1E+4 Cycles, N 1E+5 True Stress Amplitude, /2 (MPa)
Strain Amplitudes: (top to bottom) a=0.7% a=0.5% a=0.35% a=0.25% a=0.2% a=0.175%

1E+6

1E+7

(b)
650 600 550 500 450 400 350 300 250 200 150 100 50 1E+1 1E+2 1E+3 1E+4 Cycles, N 1E+5 True Stress Amplitude, /2 (MPa)
Strain Amplitudes: (top to bottom) a=0.7% a=0.6% a=0.5% a=0.4% a=0.3% a=0.25% a=0.2%

1E+6

1E+7

(c) Figure 5.11: True stress amplitude versus number of cycles for (a) forged steel, (b) powder metal, and (c) C-70 steel. 196

True Stress Amplitude, /2 (MPa)

1000

0.1308 /2 =1397.3(p/2) K ' =1397.3M Pa n ' = 0.1308

R2 = 0.9494 100 0.01% 0.10% True Plastic S train Amplitude, p/2 (%) 1.00%

(a)
True Stress Amplitude, /2 (MPa) 1000

0.1917 /2 =2004.7(p/2) K ' =2004.7M Pa n ' = 0.1917

R2 = 0.9652 100 0.01% 0.10% True Plastic S train Amplitude, p/2 (%) 1.00%

(b)
1000 True Stress Amplitude, /2 (MPa)

/2 =1738.8(p/2) K ' =1738.8 MPa n ' = 0.1919 R 2 =0.9895 100 0.01% 0.10% True Plastic S train Amplitude, p/2 (%)

0.1919

1.00%

(c)
Figure 5.12: True stress amplitude versus calculated true plastic strain amplitude for (a) forged steel, (b) powder metal, and (c ) C-70 steel.

197

True Stress Amplitude, /2 (MPa)

800 700 600 500 400 300 200 100 0 0.0% 0.2% 0.4% 0.6% 0.8% 1.0% 1.2% True S train Amplitude, /2 (%)
Data Cyclic Stress-Strain Equation

True Stress Amplitude, /2 (MPa)

800 700 600 500 400 300 200 100 0 0.0% 0.2% 0.4%

(a)

Data Cyclic Stress-Strain Equation

0.6%

0.8%

1.0%

1.2%

True S train Amplitude, /2 (%)

(b)
True Stress Amplitude, /2 (MPa) 800 700 600 500 400 300 200 100 0 0.0% 0.2% 0.4% 0.6% 0.8% 1.0% 1.2%
Data Cyclic Stress-Strain Equat ion

True S train Amplitude, /2 (%)

(c)
Figure 5.13: True stress amplitude versus true strain amplitude for (a) forged steel, (b) powder metal, and (c) C-70 steel.

198

800

700

600 True Stress Amplitude, /2 (MPa)

500

400

199

300

200
FS

100

PM C 70

0 0.0% 0.2% 0.4% 0.6% True S train Amplitude, /2 (%) 0.8% 1.0% 1.2%

Figure 5.14: Superimposed plots of true stress amplitude versus true strain amplitude for three materials.

199

800 700 True Stress (MPa) 600 500 400 300 200 100 0 0.0%

M onotonic Curve Cyclic Curve

0.5%

1.0% True S train (%)

1.5%

2.0%

(a)
800 700 True Stress (MPa) 600 500 400 300 200 100 0 0.0% 0.5% 1.0% True S train (%) 1.5% 2.0% M onotonic Curve Cyclic Curve

(b)
800 700 M onotonic Curve True Stress (MPa) 600 500 400 300 200 100 0 0.0% 0.5% 1.0% True S train (%) 1.5% 2.0% Cyclic Curve

(c) Figure 5.15: Composite plots of cyclic and monotonic stress-strain curves for (a) forged steel, (b) powder forged, and (c) C-70 steel. 200

1000
PM Cyclic

900

800
FS Monotonic C 70 Monotonic FS Cyclic PM

700

600 True Stress (MPa)

500

C 70 Cyclic

201

400
FS monotonic curve

300

FS cyclic curve PM monotonic curve

200

PM cyclic curve C 70 monotonic curve

100
C 70 cyclic curve

0 0.0% 0.5% 1.0% True S train (%) 1.5% 2.0%

Figure 5.16: Superimposed plots of cyclic and monotonic stress-strain curves.

201

700 500 True Stress, (MPa) 300 100 -100 -1.1% -300 -500 -700

Strain A mplitudes: (starting o n o utside) a=0.5% a=0.35% a=0.25% a=0.2%

-0.9%

-0.7%

-0.4%

-0.2%

0.0%

0.2%

0.4%

0.7%

0.9%

1.1%

True S train, (%)

(a)
700 500 True Stress, (MPa) 300 100 -100 -1.1% -300 -500 -700 True S train, (%)
Strain A mplitudes: (starting o n o utside) a=0.7% a=0.5% a=0.35% a=0.25% a=0.2% 75% a=0.1

-0.9%

-0.7%

-0.4%

-0.2%

0.0%

0.2%

0.4%

0.7%

0.9%

1.1%

(b)
700 500 True Stress, (MPa) 300 100 -100 -1.1% -300 -500 -700 True S train, (%)
Strain A mplitudes: (starting o n o utside) a=0.7% a=0.6% a=0.5% a=0.4% a=0.3% a=0.25% a=0.2%

-0.9%

-0.7%

-0.4%

-0.2%

0.0%

0.2%

0.4%

0.7%

0.9%

1.1%

(c)
Figure 5.17: Composite plots of midlife hysteresis loops for (a) forged steel, (b) powder metal, and (c) C-70 steel.

202

True Stress Amplitude, /2 (MPa)

1000

-0.0711 /2 =1187.9(2N f ) f ' =1187.9M Pa b = -0.0711

R2 = 0.9696 100 1E+3 1E+4 1E+5 Reversals to Failure, 2N f

Fatigue Data Least Squares Fit

1E+6

1E+7

(a)
True Stress Amplitude, /2 (MPa) 1000

-0.1032 /2 =1493.4(2N f ) f ' =1493.4M Pa b = -0.1032

R2 = 0.9840 100 1E+3 1E+4 1E+5 Reversals to Failure, 2N f 1E+6

Fatigue Data Least Squares Fit

1E+7

(b)
True Stress Amplitude, /2 (MPa) 1000

-0.0928 /2=1302.6(2N f ) f ' =1302.6M Pa b = -0.0928

Fatigue Data Least Squares Fit

R = 0.9434 100 1E+3 1E+4 1E+5 Reversals to Failure, 2N f

1E+6

1E+7

(c)
Figure 5.18: True stress amplitude versus reversals to failure for (a) forged steel, (b) powder metal, and (c) C-70 steel.

203

1000

True Stress Amplitude, /2 (MPa)

204

FS PM C 70

100 1E+3 1E+4 1E+5 Reversals to Failure, 2N f 1E+6 1E+7

Figure 5.19: Superimposed plots of true stress amplitude versus reversals to failure.

204

 p/2

1.00%
p/2 = 0.3576(2N f ) f ' =0.3576 c = -0.5663
-0.5663

True Plastic Strain Amplitude,

R2 = 0.9982 (%) 0.10%

Fatigue Data Least Squares Fit

0.01% 1E+3 1E+4 1E+5 Reversals to Failure, 2N f 1E+6 1E+7

(a)
p/2 1.00% p/2 = 0.1978(2N f ) f ' =0.1978 c = -0.5304 R2 = 0.9793 (%) 0.10%
-0.5304

True Plastic Strain Amplitude,

Fatigue Data Least Squares Fit

0.01% 1E+3 1E+4 1E+5 Reversals to Failure, 2N f 1E+6 1E+7

(b)
p/2 1.00% p/2 =0.5646(2N f ) f ' =0.5646 c = -0.5861 R2 = 0.9794 (%) 0.10%
-0.5861

True Plastic Strain Amplitude,

Fatigue Data Least Squares Fit

0.01% 1E+3 1E+4 1E+5 Reversals to Failure, 2N f 1E+6 1E+7

(c)
Figure 5.20: Calculated true plastic strain amplitude versus reversals to failure for (a) forged steel, (b) powder metal, and (c) C-70 steel.

205

1.00%

True Plastic Strain Amplitude,

p/2 (%) 0.10%

206

FS PM C 70

0.01% 1E+3 1E+4 1E+5 Reversals to Failure, 2N f 1E+6 1E+7

Figure 5.21: Superimposed plots of calculated plastic strain amplitude versus reversals to failure.

206

1.00%

True Strain Amplitude, /2, %

0.10%

Strain-Life Equation Fatigue Data (Plastic) Fatigue Data (Elastic) Fatigue Data (Total)
0.01%

1E+3

1E+4

1E+5

1E+6

1E+7

1E+8

Reversals to Failure, 2N f

(a)
1.00% True Strain Amplitude, /2, %

(2)

0.10% Strain-Life Equation Fatigue Data (Plastic) Fatigue Data (Elastic) Fatigue Data (Total) 0.01% 1E+3 1E+4 1E+5 1E+6 1E+7 1E+8 Reversals to Failure, 2N f

(b)
1.00% True Strain Amplitude, /2, %

(2 )

0.10% Strain-Life Equation Fatigue Data (Plastic) Fatigue Data (Elastic) Fatigue Data (Total) 0.01% 1E+3 1E+4 1E+5 1E+6 1E+7 1E+8 Reversals to Failure, 2N f

(c)
Figure 5.22: True strain amplitude versus reversals to failure for (a) forged steel, (b) powder metal, and (c) C-70 steel.

207

1.00%

True Strain Amplitude, /2, %

208

FS PM C 70 0.10% 1E+3 1E+4 1E+5 Reversals to Failure, 2N f 1E+6 1E+7 1E+8

Figure 5.23: Superimposed curves of true strain amplitude versus reversals to failure.

208

209

Figure 5.24: Schematic of fatigue test fixture.

209

Figure 5.25: Photo of fatigue test fixture used for connecting rod testing.

210

Figure 5.26: Load amplitude versus cycles to failure.

211

600

Stress A m plitude (S a), MPa (MPa)

Sa = 852 (Nf)-0.0572

(2)

Sa = 971 (Nf)-0.0895 (2)

R = - 1.25 200 1E+4 1E+5

O Forged steel X Powder metal 1E+6 1E+7

Cycles, Nf
Figure 5.27: Experimental stress amplitude vs. cycles to failure for forged steel and powder metal connecting rods.

212

(a)

Crack initiation site

(b) Figure 5.28: Digital photographs of FS6 connecting rod (a) as fractured, and (b) close-up of the overall fracture surface.

213

(a)

Crack initiation site

(b)

Figure 5.29: Digital photographs of FS9 connecting rod (a) as fractured, and (b) close-up of the overall fracture surface.

214

(a)

Crack initiation site

(b)

Figure 5.30: Digital photographs of FS1 connecting rod (a) as fractured, and (b) close-up of the overall fracture surface.

215

200 m

Figure 5.31: Photomicrograph of the crack initiation site for FS6 connecting rod at 50X.

216

200 m

Figure 5.32: Photomicrograph of the crack initiation site for FS9 connecting rod at 50X.

217

7m

(a)

6m

(b) Figure 5.33: SEM image of FS6 connecting rod fracture surface showing (a) crack origin site with undesired morphology, and (b) microvoid coalescence at the crack site.

218

555 m

Figure 5.34: Photomicrograph of FS9 connecting rod showing close-up of crack initiation site followed by crack propagation.

219

(a)

Crack initiation site

(b) Figure 5.35: Digital photographs of PM10 connecting rod (a) as fractured, and (b) close-up of the overall fracture surface.

220

(a)

Crack initiation site

(b) Figure 5.36: Digital photographs of PM7 connecting rod (a) as fractured, and (b) close-up of the overall fracture surface.

221

(a)

Crack initiation site

(b) Figure 5.37: Digital photographs of PM14 connecting rod (a) as fractured, and (b) close-up of the overall fracture surface.

222

200 m

Figure 5.38: Photomicrograph of the crack initiation site for PM7 connecting rod at 50X.

223

(a)

(b)

83 m

17 m

(a)

(b)

10 m

(c) Figure 5.39: Photomicrographs of PM7 connecting rod (a) close-up of crack site, (b) close-up of porosity, and (c) close-up of porosity at outer surface at different site.

224

unnotched, smooth, R = -1 f(2)b notched, smooth, R = -1 unnotched, forged, R = -1 notched, forged, R = -1

Sa
b

Sf Sf/Kf
225

KsSf

(Ks/Kf )Sf

106
Cycles to failure, Nf

Figure 5.40: S-N lines for a component without mean or residual stress.

225

1.E+04

notched, smooth, R = -1.25

1306

S a (MPa)

1113

notched, forged, R = -1.25

226

unnotched, smooth, R = -1 notched, smooth, R = -1

424 375 359

1.E+02 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 Cycles to failure, Nf Cycles, Nf 1.E+05 1.E+06

113

1.E+07

Figure 5.41: S-N lines for forged steel connecting rod.

226

1.E+04

notched, smooth, R = - 1.25 notched, forged, R = - 1.25

1691 1390

1.E+03

Sa (MPa)

unnotched, smooth, R = -1
1.E+02

334 268 257

227

notched, smooth, R = -1

83

1.E+01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07

Cycles to failure, Nf

Figure 5.42: S-N lines for powder metal connecting rod.

227

1000

Alternating stress, Sa (MPa)

(2)

228

Forged steel R = - 1.25

100 1.E+04 1.E+05 1.E+06

Predicted Experimental
1.E+07

Cycles to failure, Nf

Figure 5.43: Comparison of predicted and experimental data for forged steel connecting rod.

228

1000

Alternating stress, S

(MPa)

(3) (2)

229

Powder metal Powder metal R=-1.25 R = - 1.25

100 1.E+04 1.E+05 1.E+06

Experimental Predicted

1.E+07

Cycles to failure , Nf

Figure 5.44: Comparison of predicted and experimental data for powder metal connecting rod.

229

1.E+07

1.E+06

Predicted life

(2)

1.E+05

230

O Forged steel X Powder metal

1.E+04 1.E+04

1.E+05

1.E+06

1.E+07

Experimental life

Figure 5.45: Comparison of experimental lives with predicted lives.

230

CHAPTER 6 SUMMARY AND CONCLUSIONS

In this study, first a literature review on several aspects of connecting rods in the areas of load and stress analysis, durability, manufacturing, economic and cost analysis, and optimization was carried out. Forged steel, C-70 steel and powder metal connecting rods were then used to obtain and compare the fatigue properties and behaviors. The experiments conducted used both specimen testing of the three materials, as well as component testing of forged steel and powder metal connecting rods. Tensile tests of specimens were conducted to obtain and compare the monotonic properties. Straincontrolled fatigue tests were performed to obtain and compare the S-N, strain life, and cyclic stress-strain curves and properties. Component fatigue tests for forged steel and powder metal connecting rods were performed under axial load-controlled condition with a load ratio of R = -1.25. The analysis conducted included FEA to obtain stresses, as well as life prediction analysis to compare with the experimental component test results. Based on the literature review conducted, experimental results and observations, and analyses performed, the following conclusions can be drawn: 1. The literature review indicates that a major driving force for the increased use of PM connecting rods has been its cost effectiveness resulting from the fracture splitting of the cap from the rod. With recent introduction of new materials such as C-70 and MA splittable steels, this key advantage of powder metal connecting rods no longer exists. Some automotive manufacturers are starting to switch back from PM to forged steel connecting rods, due to their higher strength and lower manufacturing cost.

231

2. From tensile tests and monotonic curves it is concluded that forged steel is considerably stronger than the powder metal. Yield strength of forged steel is 19% higher than that for the powder metal. Ultimate tensile strength of forged steel is 8% higher than that for the powder metal. The Yield strength of C-70 steel is slightly lower than the powder metal (by 2%) and its ultimate tensile strength is higher than the powder metal (by 12%). 3. Better fatigue resistance of the forged steel material, as compared with the powder metal material was observed. Long life fatigue strength at 106 cycles of the forged steel is 27% higher than that for the powder metal. This results in about an order of magnitude longer life than the powder metal. The C-70 steel exhibits equivalent fatigue strength to powder metal in the high cycle regime. 4. Based on strain-life fatigue behavior, the forged steel provides about a factor of 7 longer life than the powder metal in the high cycle regime. At short life regime the difference is smaller. The C-70 steel exhibits the same strain-life resistance, as compared to powder metal in the high cycle regime. 5. Linear elastic finite element analysis of the forged steel connecting rod under axial loading indicated the two transition regions (crank end to the shank and pin end to the shank) to be the critical regions. A stress concentration factor of 1.18 was obtained at the crank transition region. The stress multiaxiality in the two transition regions was found to be small. 6. Forged steel and PM connecting rod component testing indicate the forged steel connecting rod exhibits 37% higher fatigue strength, as compared with the powder metal connecting rod. This increased strength results in about two orders of

232

magnitude longer life for the forged steel connecting rod. The difference in fatigue performance between the two connecting rods increases with longer lives. 7. The dominant fracture location of the powder metal connecting rods was near the transition region to the pin end. The forged steel connecting rods mainly failed in the transition to the crank end region. For the PM connecting rods the crack origins appeared to be from either the surface or subsurface, while for the forged steel connecting rods cracks started subsurface. 8. The S-N approach predictions are very reasonable, if the predictions are based on smooth surface finish, rather than forged surface finish. The beneficial compressive residual stresses on the surface from the shot peening process nullify the detrimental effect of forged surface finish. Predicted fatigue lives for both forged steel as well as powder metal connecting rods are mainly within a factor of 2 of the experimental lives.

233

REFERENCES
Araki, S., Satoh, T., and Takahara, H., 1993, Application of powder forging to automotive connecting rods, Kobelco Technology Review, Vol. 16, pp. 20-24. ASTM Standard E83-96, 1997, Standard practice for verification and classification of extensometers, Annual Book of ASTM Standards, Vol. 03.01, pp. 198-206. ASTM Standard E606-92, 1997, Standard practice for strain-controlled fatigue testing, Annual Book of ASTM Standards, Vol. 03.01, pp. 523-537. ASTM Standard E1012-93a, 1997, Standard practice for verification of specimen alignment under tensile loading, Annual Book of ASTM Standards, Vol. 03.01, pp. 699706. ASTM Standard E8-96a, 1997, Standard test methods for tension testing of metallic materials, Annual Book of ASTM Standards, Vol. 03.01, pp. 56-76. ASTM Standard E739-91, 1995, Standard practice for statistical analysis of linear or linearized stress-life (S-N) and strain life (-N) fatigue data, Annual Book of ASTM standards, Vol. 03.01, pp. 615-621. ASTM Standard E646-93, 1997, Standard test method for tensile strain-hardening exponents (n-values) of metallic sheet materials, Annual Book of ASTM Standards, Vol. 03.01, pp. 550-556. Athavale, S. and Sajanpawar, P. R., 1991, Studies on some modeling aspects in the finite element analysis of small gasoline engine components, SAE of Japan, 911271, pp. 379-389. Balasubramanian, B., Svoboda, M., and Baur, W., 1991, Structural optimization of I.C. engines subjected to mechanical and thermal loads, Computer Methods in Applied Mechanics and Engineering, Vol. 89, No. 1-3, pp. 337-360. Banerji, S. K., 1996, Application of microalloyed forgings for heavy-duty diesel-engine connecting rods and other components, Mineral, Metals and Materials Society, pp. 375389. Beretta, S., Blarasin, A., Endo, M., Giunti. T., and Murakami, Y., 1997, Defect tolerant design of automotive components, International Journal of Fatigue, Vol. 19, No. 4, pp. 319-333.

234

Bhambri, A. and Pease III, L., 1988, Mechanical properties of P/F connecting rod materials, Modern Development in Powder Metallurgy, Vol. 18-21, pp. 155-181. Chen, P. H., Herrera, N., and Venkatesan, K., 1999, Solidification modeling and simulation for connecting rods to overcome porosity problems in the squeeze casting process, Die Casting Engineer, pp. 58-59. Chernenkoff, R. A., The effects of surface modification on the fatigue performance of powder-forged steel, Ford Motor, Scientific Research Laboratories. Chernenkoff, R. A., Hall, D. W., and Mocarski, S., 1991, Material characterization of Powder forged copper steels, SAE Technical Paper 910155, pp. 1-18. Chernenkoff, R. A., Hall, D. W., Mocarski, S., and Pease III, L. F., 1994, Con-rod powders reveal quality consistency, Metallurgy Powder Report, pp. 34-38. Chernenkoff, R. A., Hall, D. W., Mocarski, S., and Pease III, L. F., 1994, Steel powders for high performance automotive parts, SAE Technical Paper 940423, pp. 1-13. Chernenkoff, R. A., Mocarski, S., and Yeager, D. A., 1995, Increased fatigue strength of powder forged connecting rods by optimised shot peening, Powder Metallurgy, Vol. 38, No. 3, pp. 196-200. Chernenkoff, R. A. and Krause, A. R., 2000, The influence of variable amplitude loading on the fatigue behavior of a powder-forged steel, Metal Powder Industries Federation. Clark, J. P., Field III, F. R., and Nallicheri, N. V., 1989, Engine state-of-the-art a competitive assessment of steel, cost estimates and performance analysis, Research Report BR 89-1, Automotive Applications Committee, American Iron and Steel Institute. Cristinacce, M., Gardner, R., and Rogers, D., Spotlight in automotive engine, Powder Metal. Doege, E. and Bohnsack, R., 2000, Closed die technology for hot forging, Journal of Materials Processing Technology, Vol. 98, pp. 165-170. Edser, C., 2001, Conrod convolution, Metal Powder Report. Ernst, E., Eilrich, U., and Weber, M., 1996, PM-connecting rods: Porosity versus performance an inevitable conflict?, SAE Technical Paper 960383, pp. 26-29. Folgar, F., Widrig, J. E., and Hunt, J. W., 1987, Design, fabrication and performance of fiber FP/Metal matrix composite connecting rods, SAE Technical Paper 870406.

235

Fukuda, S. and Eto, H., 2002, Development of fracture splitting connecting rod, SAE of Japan, 20024016, Vol. 23, pp. 101-104. Goenka, P. K. and Oh, K. P., 1986, An optimum connecting rod design study - A lubrication viewpoint, Journal of Tribology, Vol. 108, pp. 487-496. Gunyaev, G. M., Borovskaya, S. M., and Panin, V. I., 1994, Using CFRP for the design of a connecting rod for automobile applications, Journal of Automobile Engineering, Vol. 208, No. 1, pp. 33-36. Gupta, R. K., 1993, Recent developments in materials and processes for automotive connecting rods, SAE Technical Paper 930491. Hagiwara, M., 1984, A method to simplify the strength design of bolted joints case of connecting-rod bolts, Experimental Mechanics, pp. 28-32. Hippoliti, R., 1993, FEM method for the design & optimization of connecting rods for small two-stroke engines, Technical Papers/Small Engine Technology Conference, pp. 217-230. Hoag, P. Y. and Yeager, D. A., 1990, Making a fractured powder metal connecting rod, United States Patent, 4936163. Illia, E. and Chernenkoff, R. A., 2001, Impact of decarburization on the fatigue life of powder metal forged connecting rods, SAE Technical Paper 2001-01-0403, pp. 1-6. Imahashi, K., Tsumuki, C., and Nagare, I., 1984, Development of powder-forged connecting rods, SAE Technical Paper 841221, pp. 1-7. Jinka, A. G. K., Bellet, M., and Fourment, L., 1997, A new three-dimensional finite element model for the simulation of powder forging processes: Application of hot forming of P/M connecting rod, International Journal of Numerical Methods in Engineering, Vol. 40, pp. 3955-3978. Jinka, A. G. K. and Bellet, M., 1996, Hot forging of a P/M connecting rod: Three dimensional computer model, International Journal of Powder Metallurgy, Vol. 32, No. 3, pp. 255-258. Johnson, P. K., 1995, Powder metal gets parts in shape, Powder Metal Design News, pp. 89-90. Karhausen, K. and Kopp, R., 1995, Improvement of microstructure in forging of a connecting rod by means of finite element simulations, Steel Research, Vol. 66, No. 1, pp. 20-26.

236

Kimura, A., Nakamura, S., Isogawa, S., Matsubara, T., Kimura, K., and Sato, Y., 1991, A free machining titanium alloy for connecting rods, SAE Technical Paper 910425, pp. 9-16. Klink, U., 1985, Multi-stage honing of connecting rods, Industrial Diamond Review, Vol. 45, No. 5, pp. 289-293. Kuch, I. and Kummerien, P., 1991, Potential for lightweight design of a hybrid connecting rod for automobile applications, SAE Technical Paper 910424. Kuratomi, H., Takahashi, M., Houkita, T., Hori, K., Murakami, Y., and Tsuyuki, S., 1995, Development of a lightweight connecting rod made of a low-carbon martensite steel, Japan SAE, Vol. 16, No. 4, pp. 406-407. Kuratomi, H., Uchino, M., Kurebayashi, Y., Namiki, K., and Sugiura, S., 1990, Development of lightweight connecting rod based on fatigue resistance analysis of microalloyed steel, SAE Technical Paper 900454, pp. 57-61. Liu, F. and Hathaway, A., 2001, Post-delubrication peening of forged p/m conn rod, Proceedings, Advances in Powder Metallurgy and Particulate Materials. Lu, P. C., 1996, The shape optimization of a connecting rod with fatigue life constraint, International Journal of Materials and Product Technology, Vol. 11, No. 5/6, pp. 357370. Luong, M. P., 1997, Rapid determination of fatigue strength of connecting rods using infrared thermography, Mechatronics; Automotive Electronics, pp. 387-394. Majidpour, M., Shakeri, M., and Akhlaghi, M., 2002, Quasi-dynamic stress analysis of connecting rod and cyclic stress-time histories generations, The Second International Conference on Internal Combustion Engines. Marra, M. and William, R., 1989, Mechanical properties of a powder forged material for connecting rods, Advances in Powder Metallurgy: Proceedings of the Powder Metallurgy Conference & Exhibition, Vol. 1, pp. 313-320. Marra, M. P., Compton, W. A., and Skurka, J. C., 1992, Fatigue testing of a powder forged connecting rod, SAE Technical Paper 920218. Murray, C. J., 1988, Compliant connecting rod reduces bearing stresses, Automotive Design News, Vol. 44, pp. 278-279. Murikov, D., Pestov, V., Romaneko, L., and Belanov, G., 1983, Improved design of the cross head and connecting rod assembly in 6M 40 compressors, Chemical and Petroleum Engineering, Vol. 19, No. 10, pp. 423-424.

237

Nakamura, S., Mizuno, K., Matsubara, T., and Sato, Y., 1993, Development of high fatigue strength free machining microalloyed steel for connected rods, SAE Technical Paper 930619, No. 972, pp. 23-30. Niebel, B. W., Draper, A. B., and Wysk, R. A., Modern Manufacturing Process Engineering, McGraw-Hill, pp. 381-394. Nortan, R., 1994, Design of Machinery, McGrow Hill, pp. 583. OBrien, R. C., 1988, Fatigue properties of P/M materials, SAE Technical Paper 880165. Olaniran, M. A. and Stickels, C. A., 1993, Seperation of forged steel connecting rods and caps by fracture Splitting, SAE Technical Paper 930033, pp. 1-6. Paek, S. Y., Ryou, H. S., Oh, J. S., and Choo, K. M, 1997, Application of high performance powder metal connecting rod in the V6 engine, SAE Technical Paper 970427. Park, H., Ko, S. Y., Jung, S. C., Song, B. T., Jun, Y. H., Lee, B. C., and Lim, J. D., 2003, Development of fracture split steel connecting rods, SAE Technical Paper 2003-011309. Pillinger, I., Hartley, P., Sturgess, C. E. N., and Rowe, G. W., 1985, A threedimensional finite element analysis of the cold forging of a model aluminium connecting rod, Proceedings of the Institution of Mechanical Engineers, Vol. 199, No. C4, pp. 319324. Rabb, R., 1996, Fatigue failure of a connecting rod, Engineering Failure Analysis, Vol. 3, No. 1, pp. 13-28. Repgen, B., 1998, Trend of modern connecting rods, Technology and Services, pp. 1-4. Repgen, B., 1998, Optimized connecting rods to enable higher engine performance and cost reduction, SAE Technical Paper 980882, pp. 23-26. Richter, K., Hoffmann, E., Lipp, K., and Sonsino, C. M., 1994, Single-sintered PM connecting rods for passenger cars - an illusion?, Powder Metallurgy-Structural Applications, Vol. 1, pp. 39-42. Richter, K., Hoffmann, E., Lipp, K., and Sonsino, C. M., 1994, Single-sintered conrodsan illusion?, Metal Powder Report, Vol. 49, pp. 38-45. Saritas, S. and Davies, T. J., 1984, Fracture behaviour of powder forged steels, Modern Developments in Powder Metallurgy, Vol. 15, pp. 599-609.

238

Schreier, L., 1999, Tension and compression in connecting rods, http://emntserver.unl.edu/Mechanics-Pages/Luk-chreier/unzip/Tension%20and%20Compression% 20in%20Connecting%20Rods%20VI.htm. Shenoy, P., 2004, Quasi dynamic stress analysis and optimization of connecting rod, Master thesis, The University of Toledo. Shiao, Y. J. and Moskwa, J. J., 1993, An investigation of load force and dynamic error magnitude using the lumped mass connecting rod model, SAE Technical Paper 930617, No. 972, pp. 1-10. Siegert, K. and Ringhand, D., 1994, Flashless and precision forging of connecting rods from P/M aluminum alloys, Journal of Materials Processing Technology, Vol. 46, No. 1-2, pp. 157-167. Skoglund, P. and Bengtsson, S., 2001, Warm compaction cuts the costs of connecting rods, Metall., Vol. 55, No. 6, pp. 358-360. Sonsino, C. M., 1988, Fatigue design of sintered connecting rods, Metal Powder Report, Vol. 43, Issue: 5, pp. 332-337. Sonsino, C. M., 1990, Fatigue design of sintered connecting rods, Metal Powder Report, Vol. 45, Issue: 6, pp. 408-412. Sonsino, C. M., 1996, Fatigue design of high loaded PM parts, Metal Powder Report, Vol. 51, Issue: 6, pp. 36-45. Stephens, R. I., Fatemi, A., Stephens, R. R., and Fuchs, H. O., 2000, Metal Fatigue in Engineering, 2nd Edition, John Wiley & Sons, Inc. Sugita, J., Itoh, T., and Abe, T., 1990, Engine component design system using boundary element method, SAE Technical Paper 905206. Suzuki, S., Toyama, K., and Konda, N., 1988, Development of as-sintered P/M connecting rod for automobiles, SAE Technical Paper 880168, pp. 1-7. Takahashi, K., Hoshina, E., and Ogino, M., 1994, Development of high strength powder-forged connecting rod, SAE Technical Paper, Vol. 3, pp. 2227-2230. Takemasu, T., Vazquez, V., Painter, B., and Altan, T., 1996, Investigation of metal flow and preform optimization in flashless forging of a connecting rod, Journal of Materials Processing Technology, Vol. 59, No. 1-2, pp. 95-105. Tilbury, R. J., 1982, The prediction and measurement of axial forces, bending moments and accelerations in an engine connecting rod, Journal of Strain, Vol. 18, No. 2, pp. 5559.

239

Todd, R., Allen, D., and Alting, L., 1994, Manufacturing Process Reference Guide, Published by Industrial Press Inc. Vazquez, V. and Altan, T., 2000, Die design for flashless forging of complex parts, Journal of Materials Processing Technology, Vol. 98, pp. 81-89. Vijayaraghavan, D., Brewe, D. E., and Keith, T. G., 1993, Effect of out-ofroundness on the performance of a diesel engine connecting rod bearing, Journal of Tribology, Transactions of ASME, Vol. 115, pp. 538-543. Weber, M., 1990, Powder forged connecting rod state of the art and economics, International Symposium on Automotive Technology and Automation, Vol. 3, pp. 77-82. Weber, M., 1991, Cost effective finishing of powder forged connecting rods with the fracture-splitting-method, SAE Technical Paper 910157, pp. 1-3. Weber, M., 1991, Powder forging of automotive connecting rods, Congress Seminar Paper, Vol. 31. Weber, M., 1993, Comparison of advanced procedures and economics for production of connecting rods, Powder Metallurgy International, Vol. 25, No. 3, pp. 125-129. Webster, W. D., Coffell, R., and Alfaro, D., 1983, A three dimensional finite element analysis of a high speed diesel engine connecting rod, SAE Technical Paper 831322, pp. 83-96. Whittaker, D., 1991, Sintered materials - Can they operate under fatigue loads?, Materials and Design, Vol. 12, No. 3, pp. 167-169. Whittaker, D., 2001, The competition for automotive connecting rod markets, Metal Powder Report, Vol. 56, No. 5, pp. 32-37. Wright, D., 1993, Testing Automotive Materials and Components, SAE Publications, pp. 121-122. www.diecasting.org/techdatabase.asp, 2001 www.forging.org www.mpif.org www.upinc.net/excello.htm

240

Yang, R. J., Dewhirst, D. L., Allison, J. E., and Lee, A., 1992, Shape optimization of connecting rod pin end using a generic model, Finite Element in Analysis and Design, Vol. 11, No. 3, pp. 257-264. Yoo, Y. M., Haug, E. J., and Choi, K. K., 1984, Shape optimal design of an engine connecting rod, Journal of Mechanisms, Transmissions and Automation in Design, Vol. 106, pp. 415-419. Young, W. C., 1989, Roarks Formulas for Stress and Strain, 6th Edition, McGraw-Hill Inc. Zhu, Z. G. and Chen, Y., 1983, The stability of the motion of a connecting rod, Journal of Mechanisms, Transmissions and Automation in Design, Vol. 105, pp. 637-640.

241