«©SinterCast 2012 Introduction As the demand for high torque, low emissions and improved fuel economy continues to grow, engine ...»
The compression results show that the compressive yield strength of unalloyed pearlitic CGI is approximately 25% higher than that of the Cr-Mo alloyed grey iron. The data also show that the room temperature compressive yield strength increases linearly with increasing pearlite content and is apparently insensitive to the effects of chrome and molybdenum alloying. The role of chrome is to stabilize pearlite at elevated (400°C) temperatures while molybdenum improves creep and thermal fatigue performance. Finally, the compression tests show that the elastic modulus of CGI (145-155 GPa) is the same in both tension and compression.
Hardness The graphite growth and carbon diffusion behaviour during solidification of compacted graphite iron naturally favours the formation of ferritic rather than pearlitic matrices. Therefore, pearlite stabilizers such as copper and tin must be added to the base iron to ensure a predominantly pearlitic matrix. The solidification behaviour of compacted graphite iron results in 10 to 15% higher hardness than grey cast iron when compared at equal pearlite contents. While a typical Brinell hardness range for fully pearlitic grey iron cylinder blocks may be BHN 179-223, fully pearlitic CGI cylinder blocks may range from BHN 192-255.
Compacted graphite iron cylinder blocks containing approximately 70% pearlite have a similar hardness to fully pearlitic grey iron blocks. As shown in Figure 8, the hardness level of CGI increases linearly with increasing pearlite content. While this trend is certainly true, the slope and intercept of the linear correlation depend on the concentration of manganese, chromium, titanium and other trace elements in the raw materials as well as the casting shake-out time.
The influence of the graphite growth behaviour on hardness is evident from Figure 9. At a fixed (85-100%) pearlite content, the Brinell hardness is effectively constant from 0 to 90% nodularity. However, a stepreduction in hardness occurs as soon as flake patches begin to form. The rapid decrease in hardness between CGI and grey iron is due to the complex compacted graphite morphology which prevents slip and fracture in the matrix and at the graphite/matrix interface.
Wear Resistance Wear is a complex phenomenon that comprises several tribological mechanisms and has no universally accepted test procedure or quantitative criteria for establishing the suitability of a given material. In contrast to the clearly defined specification limits for tensile properties or hardness, the wear behavior of materials is typically evaluated relative to the performance of other candidate materials. Experimental procedures can vary from sophisticated real-load simulations to relatively simple pin-on-disk or abrasion tests.
In general, unalloyed pearlitic CGI incurs approximately one-half of the wear of unalloyed pearlitic grey cast irons when exposed to scarring or abrasion conditions. Hrusovsky  showed that the average scar width in LFW-1 pin-on-disc tests was 45% less for pearlitic CGI than for pearlitic grey iron. Similar results have been reported for abrasion tests where the weight loss of CGI specimens during abrasion tests is 40-55% less than that of grey iron [18,19]. Although the test techniques and results vary, the consistency shows that CGI has superior wear resistance to grey iron and is a viable material for cylinder liner applications.
In order to determine the suitability of CGI as a cylinder liner material, a series of scuffing tests were conducted to compare the wear resistance of various CGI microstructures and compositions with commercially available cylinder liner materials. The scuffing test  was conducted by pushing a test specimen against a rotating nitrided steel cylinder. The load range and the sliding velocity (5.7 cm/sec) were selected to simulate the conditions near dead-centre in the cylinder bore of a combustion engine. Lubricant was not used in order to exacerbate the wear conditions. As the scuffing test proceeds, the surface of the test material roughens and the friction between the test piece and the nitrided cylinder increases. The relative scuffing resistance of each material is determined by comparing the lowest friction coefficient recorded, and the normal load above which the friction coefficient first exceeds 0.3. These points are known to relate to friction and scuffing behavior in internal combustion engines .
The chemical analysis of the four CGI variants and the conventional liner materials studied in these tests is summarized in Table V. The phosphorous-alloyed grey iron specimens and the surface-treated aluminum and grey iron specimens were sectioned from commercially available centrifugally cast cylinder liners or from the top-dead-centre sliding surface of parent bore cylinder blocks. Calibrations were made to account for the different area of surface contact for curved and flat specimens.
The scuffing results plotted in Figure 10 show the minimum coefficient of friction for each material and the applied normal load at which the coefficient of friction first exceeds 0.3. Ideal wear materials should have a low coefficient of friction and simultaneously tolerate high loads without incurring wear and thus increasing friction. According to these criteria, and for the conditions of this test, the performance of unalloyed pearlitic CGI is superior to that of the Nicasil coated aluminum liner and the laser hardened and Cr-alloyed grey iron liner. The behavior of unalloyed pearlitic CGI is not statistically significantly different from the phosphorous and phosphorous-boron alloyed grey iron specimens. The best performance was realized by the Cr-Mo alloyed pearlitic CGI which simultaneously displayed the lowest friction and the highest scuffing load.
Comparative wear tests have also been conducted by Volvo Technological Development  to discern the difference between unalloyed pearlitic CGI and conventional P-alloyed grey iron cylinder liners for diesel application. The engine simulation test rig pushes a standard top compression ring against a honed liner specimen with a normal load of 320 N. The ring is then oscillated with an 8 mm amplitude reciprocating motion at a frequency of 10 Hz. The contact region is submerged in used 10W/30 engine oil and heated to 80°C to replicate the wear behavior in running engines.
Operationally, the test is interrupted after three hours and the wear volume is measured by 3D profilometry to determine the running-in wear. The specimen is then repositioned and tested for a further thirteen hours to determine the steady-state wear behavior. The results of the comparative test show that the running-in (0-3 hours) and steady-state (3-16 hours) wear behavior of unalloyed pearlitic CGI is not statistically significantly different from that of the standard phosphorous-alloyed grey iron reference liners.
Fatigue and Notch Sensitivity Similar to tensile properties, the fatigue strength of a material is influenced by its microstructure and alloying elements. Therefore, fatigue data should always be presented together with microstructure data and the monotonic tensile strength of the base material to serve as a reference point. Fatigue data in this section are
thus reported in terms of two dimensionless indices:
These indices allow the notched and unnotched fatigue limits of a material to be approximated directly from its tensile strength. A summary of rotating-bending fatigue data from four different literature sources is provided in Table VI while classical S-N plots for notched and unnotched CGI are shown in Figure 11.
The literature data show that the rotating-bending Endurance Ratio of CGI ranges from 0.44 to 0.58, although values as low as 0.37 have been reported . Pearlitic irons tend toward the lower end of the Endurance Limit range which means that ferritic CGI retains a larger portion of its monotonic strength in fatigue applications. Nonetheless, the absolute fatigue limit of pearlitic CGI is approximately 25% higher than that of ferritic CGI (Figure 11). The rotating-bending fatigue limit of pearlitic CGI is approximately double that of pearlitic grey iron and similar to that of ferritic ductile iron.
It is well known that the notch sensitivity of cast irons is a direct function of the geometry of the imposed notch. Reference  employed a relatively severe 45o V-notch with a 0.25 mm root radius and obtained Fatigue Strength Reduction Factors of 1.72-1.80 for CGI. In contrast, Reference  incorporated a relatively smooth ‘notch’ geometry (actually a 1.2 mm diameter hole drilled through the centre of the test specimen) and obtained Fatigue Strength Reduction Factors ranging from 1.36 to 1.58. Finally, Reference  used the intermediate notch geometry of a 0.5 mm diameter by 0.5 mm deep drill hole and obtained intermediate results (1.42-1.50). While it is clear that CGI has a higher notch sensitivity than grey iron (1.05-1.15) and a lower sensitivity than ductile iron (1.50-1.85), the most critical factor is the notch geometry. Indeed, Palmer  states that if the root radius of the notch was increased to 5 mm in a 10.6 mm diameter fatigue specimen, the notch effect would be eliminated.
In addition to the basic data provided by conventional rotating-bending tests, designers are also interested in the fatigue behavior of components subjected to uniaxial (tension-compression) loading. This is particularly true for components which are subjected to both an underlying tensile load and cyclical stress fluctuations which are imposed about the mean tensile stress level. Depending on the mean stress level and the magnitude of the fluctuations, the resulting fatigue cycle can be either tension-tension or tensioncompression. Tests of this nature are conducted with uniaxial hydraulic servo fatigue machines [26,27]. The results of such fatigue tests are presented in so-called Goodman diagrams as shown in Figure 12 [7,26] for uniaxial loading and in Figure 13  for three-point bending. The fatigue limits in the absence of any mean stress can be read directly from the intersection of the fatigue limit lines with the vertical axis, while the maximum permissible stress (at 107 cycles) at any mean stress level is represented by the distance between the 45o line and the respective fatigue line for the material. Again, the performance of pearlitic CGI is similar to that of ferritic ductile iron.
The behavior of CGI under fully reversed torsional fatigue conditions has been studied by Sumimoto et al . Although they only studied ferritic cast irons, their data (Table VII) show that the torsional fatigue limit of CGI is approximately the same as that of ductile iron, and thus approximately 40% higher than that of steels with similar tensile strengths. The values obtained for the Endurance Ratio of CGI are also in good agreement with that presented in Table VI for rotating-bending conditions.
The good fatigue performance of CGI stems directly from the compacted graphite morphology. The rounded edges of the graphite particles do not contribute to crack initiation and actually serve as crack arresters once cracks are formed. The complex coral-like morphology and irregular surfaces of the graphite particles thereafter result in good adhesion and thus present a more tortuous crack propagation path relative to the smooth flake surfaces in grey iron. These same morphological characteristics result in a two-to-four-fold increase in the thermal fatigue resistance of CGI relative to grey iron [29-31]. Despite that consistently superior thermal fatigue results have been reported for CGI under a variety of thermal cycle conditions, it must be stressed that such results may not be directly transferable to the performance of thermally loaded components such as cylinder heads. The difference in geometry, constrainment and the critical roles of thermal conductivity and elastic modulus in transferring heat and stress cannot be accurately simulated in uniaxial test pieces. Practical experience has shown that CGI can improve the service life of high performance direct injected diesel engine cylinder heads provided that the flame deck thickness is reduced to facilitate heat transfer and distribution of accumulated stress.
Damping Capacity The damping capacity of cylinder block materials is frequently referred to in consideration of the anticipated NVH (Noise, Vibration and Harshness) performance of a finished engine. The ultimate noise level of an operating engine is dependent upon many factors including the stiffness of the cylinder block (derived both from the design of the block and the elastic modulus of the chosen material), interaction between excitation frequencies and local frequency modes, the influence of ancillary components such as pumps, belts and gears, and the damping capacity of the block material.
Damping capacity results are reported in the literature for several different types (chemistry, microstructure) of cast iron. The test and measurement techniques vary between references but are all based on the amount of reduction of the vibration wave amplitude during successive wavelength cycles. Results can be reported in terms of the Loss Factor (), the logarithmic decrement () or the damping capacity (). Although these terms are interchangeable via the relationship : = 2 = 2, it is most convenient to normalize results and present the relative damping capacity data on a scale of 0 to 1. Thus, the data from the literature are summarized in Table VIII.