j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 ( 0 ) 328–338 journal homepage: www.elsevier.com/locate/jmatprotec Accuracy of hard turning ´ a , B Karpuschewski b , K Gyani a , V Bana c,∗ J Kundrak a b c Department of Production Technology, University of Miskolc, Miskolc, Hungary Institute for Production Technology and Quality Assurance, Otto-von-Guericke University, Magdeburg, Germany Advanced Technology Centre (ATC), Philips DAP bv., Drachten, The Netherlands a r t i c l e i n f o a b s t r a c t Article history: Nowadays, hard turning is frequently used to replace grinding The economic benefits of Received 16 October 2006 hard turning are obvious but for achievable accuracy the situation is somewhat ambiguous Received in revised form Although machine tool factories offer lathes with the same accuracy as grinding machines 28 August 2007 in some cases problems may arise in keeping the prescribed geometrical accuracy Inves- Accepted September 2007 tigations were performed in a working environment in order to determine the attainable size, form and positional accuracy obtained with hard turning Error sources of machining errors that occurred in hard turning and in grinding were taken into account, giving typical Keywords: differences between the two processes In the parts produced in series, size deviations were Hard turning measured as well as out-of-roundness, cylindricity error and parallelism error of the bore’s Accuracy generatrices The workpieces used for the investigation are disc-type parts with bores, i.e., Form error gears that are built into transmissions Our first measuring series evaluates the achievable Size error accuracy with hard turning while the second includes the comparison of grinding with hard turning The most important error sources are identified We present measures for keeping prescribed tolerances and propose methods for eliminating the means error source © 2007 Elsevier B.V All rights reserved Introduction The primary task of hard turning – as a finishing operation – is to ensure the quality and reliability of the parts The quality criteria of hard turning as a cutting operation can be found in the technical drawings The most important quantities are geometrical accuracy, surface topography, and the integrity of the subsurface layer Geometrical accuracy includes size errors, form errors and positional errors Surface topography covers the drawing, the roughness and the bearing curve of the surface Surface integrity describes changes in the physical properties of the material that result from machining (Barry and Byrne, 2002) For analysis of the geometrical accuracy, four typical characteristics of hard turning – as opposed to grinding – must be outlined ∗ Corresponding author E-mail address: viktoria.bana@hotmail.com (V Bana) 0924-0136/$ – see front matter © 2007 Elsevier B.V All rights reserved doi:10.1016/j.jmatprotec.2007.09.056 These are: (1) the significantly higher cutting force, (2) the omission of coolant, (3) the single point form generation, and (4) the minimum value of the depth of cut The cutting force occurring in hard turning is higher than that in conventional turning or grinding The passive force occurring in hard turning – the component perpendicular to the cutting speed – is a multiple of the main cutting force, while in traditional turning it is only a fraction of this value The extraordinarily high passive force, which contributes to the material removal, significantly loads the elements of the machining system, causing elastic deformation and deteriorating the machining accuracy The disadvantageous effect of the high passive force must be compensated for by an increase in machine tool rigidity Hard turning can be done in dry conditions at relatively high speed The relative high friction coefficient and the pas- j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 ( 0 ) 328–338 sive force cause a significant friction force which transforms into heat The other source of the generated heat is the high cutting speed This cannot be reduced, because PcBN is only effective at high temperatures The high temperature generated during material removal causes thermal expansion of the workpiece, which also deteriorates the machining accuracy The disadvantageous effect of the hot swarf falling on the elements of the machine tool must be compensated for with the increase of “heat rigidity” of the machine (Schmidt and Dyck, 2000) The surface generating element of hard turning is the single-point tool tip, which shapes the surface of the workpiece and is accompanied by significant force and heat effects Under such conditions the single-point tool tip reacts sensitively to any irregularities It abreacts the allowance distinctions, the hardness differences and the other heterogeneities of the material with the creation of machining errors On the other hand, grinding, where the surface generating element is the linear generatrix of the wheel and the forces are also lower, hardly detects any inhomogeneities in its path; it simply eliminates them and creates a more accurate surface The fourth specification influencing the accuracy is the depth of cut In hard turning this cannot be reduced arbitrarily, although this is possible in grinding Moreover, grinding can also be performed with zero depth of cut, called spark-out, to eliminate any deflections of the system by continuously reducing the forces Because of the necessity of a minimum depth of cut, hard turning is followed by higher forces than in grinding, even in the finest smoothing operations Due to the four typical characteristics of hard turning mentioned above, the tool tip is exposed to a substantially more intensive physical–mechanical load than a single grain of the grinding wheel The density of energy transmitted into the workpiece on the tool tip is much higher Therefore, the stationary state is more difficult to maintain and the accidental error sources arising in the material removal process are more difficult to handle than in grinding Most unfavourable effects can be eliminated by an increase in the robustness of the machining system The main element of the machining system is the machine tool, which must possess extraordinarily high static and dynamic stiffness and must also withstand the heating effect of the hot swarf falling upon it A second element is the clamping device for the workpiece It has a large influence on the geometrical accuracy in disc-type components, which sometimes possess thin walls Extremely rigid clamping devices and deliberate clamping forces systems must be applied The third element is the tool-clamping device together with the tool, which offers few possibilities to increase the rigidity The fourth element of the system is the workpiece, which is given and bears the errors issuing from ă the motions of system elements (Tonshoff et al., 1997) Unintended error sources occurring in hard turning can be divided into two classes: error sources depending on the load and error sources independent of the load Error sources depending on the load: Cutting force, which creates elastic deformations; Cutting heat, which causes distortions; 329 Tool wear, which increases the force and the heat; Insufficient static rigidity of the machining system; (a) The machine tool with insufficient stiffness becomes deformed from the force (b) The workpiece’s clamping device of inadequate rigidity is deformed (c) The weak tool holder and tool are bent (d) A workpiece with unsatisfactory stiffness is deformed by the clamping force Because of the inadequate dynamic stiffness of the machining system, low-frequency oscillations may cause form error Error sources independent the load: Uneven allowances cause force fluctuation and form error and release stresses Inhomogeneities of the workpiece material cause from error by the force fluctuation Manner of the surface generation: in turning a point, in grinding a line generates the surface Construction of the clamping device and the deforming effect of the clamping force may be a significant source of form errors Instead of concentrated force it is more suitable to clamp with distributed force Number of clampings: hard turning has an advantage because with the single point tool and one clamping various surfaces can be machined In grinding usually several clampings are necessary, which is a source of significant form and positional errors Experimental conditions 2.1 Type and main sizes of the components and accuracy prescriptions This investigation was performed in a working environment on gears with bores machined in batch The gears are built into the transmissions of motor vehicles, and are disc-type components of different sizes The material of the gears is case-hardened steel: 16MnCr5 (AISI 5115) with hardness 62 ± HRC Four out of 20 measured gears are presented with the measuring setup in Fig in order to show the sizes and types investigated The accuracy prescription for the gears applies to the size, form and positional accuracy In many cases the size accuracy of the bores is IT6, occasionally IT5 or IT7 For form errors, the out-of-roundness of the bore and the flatness error of the faces are prescribed Furthermore, the parallelism of the bore’s generatrices and the axial run-out of the faces has to be kept within certain limits The specifications of form and positional accuracy are set by the internal standards of the factory on the basis of functional conditions These are much stricter than the prescriptions in the general standards For instance, to the Hungarian standard MSZ ISO 2768-2:1991 permits outof-roundness equal to half of the diameter, which is 2–3 ␮m larger than the value prescribed by the factory Table summarizes the prescriptions of accuracy for hard turned and ground surfaces for the 10 gears chosen as exam- 330 j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 ( 0 ) 328–338 Fig – Main types and sizes of gears A, B, C: Measuring planes of out-of-roundness Table – Prescription of accuracy for hard turned/ground surfaces j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 ( 0 ) 328–338 331 Fig – Interpretation of measuring results (a) Out-of-roundness, (b) cylindricity, (c) parallelism of the bore’s generatrix, (d) measuring setup for determination of flatness and axial run-out, (e) flatness and (f) axial run-out ples The root diameter/bore diameter (droot /d1 ) and bore length/bore diameter (L/d1 ) ratios are also shown, since these play a very significant role in the formation of the measuring results Although the prescriptions of accuracy according to the internal standards of the factory not include cylindricity, cylindricity error was also measured and the alteration of its value is presented In positive cases the cylindricity error mirrors the parallelism of the generatrices As the investigation of parallelism is done in two-dimensional planes, the result does not provide any reliable information about the cylindricity error The finish hard turning was performed on an advanced machine tool suitable for the requirements of hard turning and was done in one clamping, in a hydraulic three-jaw chuck The special jaws centralize on the pitch circle of Fig – Out-of-roundness, cylindricity and parallelism in hard turning 332 j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 ( 0 ) 328–338 Fig – Generation of out-of-roundness in hard turning the gears and clamp with concentrated force The clamping force is not known; only the pressure of the system, which is 12 bar The finish grinding was done on several grinding machines, with several clampings; a three-jaw chuck was used 2.2 Machining conditions The experimental conditions for machining of gear bores are: (a)Hard turning Machine tool: PITTLER PVSL-2 lathe Cutting tool: PcBN CNGA 120408 BNC80 ◦ ◦ ◦ n = −6 , ˛n = , Är = 95 , εr = 80◦ , r␧ = 0.8 mm (CAPTO C5-PCLNL-17090-12) Technological data: vc = 120–228 m/min, f = 0.08–0.1 mm/rev, ap = 0.1 mm (b)Grinding Machine tool: SI-4/A internal grinder (VEB Berliner Werkzeugmachinenfabrik) Grinding wheels: CBN wheel 50 × 32 × 20 9A 80 K7 V22 Bay state CBN wheel 60 × 36 × 20 9A 80 K7 V22 Bay state Fig – Size accuracy of hard turned bores machined in sequence Fig – Generation of cylindricity error in hard turning j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 ( 0 ) 328–338 Technological data: Wheel speed: Speed ratio: Workpiece speed: Depth of cut: vc = 25 m/s q = 80 vw = 18 m/min ae = 0.020 mm/double stroke, roughing ae = 0.020 mm/double stroke, finishing Sparking out: Traverse speed: Feed rate: Coolant: Fig – Diagrams of hard turned and ground profiles 333 8–6 double stroke vf,L = 2200 mm/min, roughing (traverse speed) vf,L = 2000 mm/min, finishing f = 13.7 mm/workrev, roughing f = 12.0 mm/workrev, finishing half synthetic, Q = 50 l min−1 334 j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 ( 0 ) 328–338 Grinding of the gear bores presented in Table was performed with internal traverse grinding The grinding allowance is 0.4 mm in radial direction For each piece the dressing of the wheel is done four times with natural diamond (0.5 carat) The depth of cut for dressing is ad = 0.01 mm Measuring methods and devices The interpretation of the measuring data and review of the measuring methods can be seen in Fig The determination of the real geometrical forms was done by a mm diameter stylus head Out-of-roundness is either defined by the longest distance between the fitting circle and the given points of the real profile, or the instrument determines a so-called reference circle and out-of-roundness is determined as a value of the largest positive local roundness deviation added to the absolute value of the largest negative local roundness deviation (Fig 2a) The last method is more frequently used and it was applied in this study The reference circle is defined and written out by the measuring computer on the basis of the least square sum of the deviations Cylindricity error is given as value of the largest positive local cylindricity deviation added to the absolute value of the largest negative local cylindricity deviation (Fig 2b) The computer generates the real cylinder using the circumferential section method: the instrument determines the out-of-roundness diagrams in three or four planes perpendicular to the axis and from them it applies a cylinder superficies built up from straight lines This is the real cylinder form The parallelism error of the generatrices is the absolute difference in local diameter at the top and the local diameter at the bottom of the cylindrical future of the two associated lines fitted through the two generatrix profiles obtained from an intersection of a plane through the axis of the least squares reference cylinder and the cylindrical feature within the full extent of the feature (Fig 2c) The flatness error of the face and its run-out is determined during one scanning by analysing the writtenout profile in two different manners (Fig 2d) As the face can be a plane either with or without run-out, both errors may appear together In the measurement of flatness firstly a reference plane must be determined in the profile outstretched into a plane The trace line of the reference plane (MMZRP) is also defined by the least square sum method, by means of regression analysis (Fig 2e) After this two reference planes (OMZRP, IMZRP), parallel to the reference plane are fitted on the real surface The distance between these two planes determines the flatness error (FLTt) In the measurement of the axial run-out of the face the real surface is also touched with two planes that are perpendicular to the datum axis rather than parallel with the reference plane (Fig 2f) The distance between these two planes defines the axial run-out (R) The diameter of the bores was measured by the Derby Etalon 454 coordinate measuring machine The applied stylus was the Renishaw Brown Shape TP-ES This measuring machine measures the diameter of one circle with three touches We repeated it on four circles and their average value formed the real value for the diameter The measurement of out-of-roundness, parallelism, cylindricity, flatness and axial run-out was done with a Talyround 252e roundness measuring machine, as was the preparation of the diagrams presented in this article In the measurement of geometrical accuracy, electronic filters are important in order to ensure the range of suitable frequency transmissions, i.e., the number of undulations being set regularly on the circumference, which can be observed by the instrument during one revolution We used a Gauss-type filter with characteristic of 1–500 undulations/revolution This filter senses undulations up to 500, which are set regularly on the circumference Compared with Gauss-type filters of 1–50 u/r or 1–15 u/r, this filter can give a more-detailed idea of the form of the out-ofroundness It is allowed to compare the gained diagrams only if they were measured with the same filter Out-of-roundness, cylindricity error and parallelism error in hard turning The geometrical accuracy of hard turning was investigated in series production One series each of three gears was measured; the series contained 285, 60, and 200 pieces The gears differ in droot /d1 ratio Out-of-roundness, cylindricity and parallelism measurements were performed at regular intervals: after each 25th piece for the 285-piece series, each 5th for the 60-piece series, and each 10th for the 200-piece series Average values of these measurements are given in Fig The measuring results truly reflect the geometrical errors issuing from the deformation of the workpiece The “rigidity” of the workpiece, the weakest element of the system, is defined by the droot /d1 ratio With the decrease of the droot /d1 ratio, out-of-roundness and cylindricity error increase The form errors increase not only in absolute meaning but also its scattering significantly rises However, the parallelism error of the generatrices hardly changes, nor does its scatter; each measured value remains much lower than the prescribed 0.006 mm Out-of-roundness and cylindricity error are presented in Figs and 5, where every measuring result is shown For the gear with droot /d1 = 2.83 ratio, which can be regarded as quite rigid, the out-of-roundness of 0.006 mm prescribed for quality IT6 can be kept, but this cannot be guaranteed for the Fig – Summary of measured results j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 ( 0 ) 328–338 gear of droot /d1 = 1.34 ratio The reason for this is obviously the clamping chuck, because the clamping method and force are not suitable It is mechanical clamping with high pressure instead of magnetic clamping On the newly developed lathes, in addition besides 12 bar, two lower pressures are also 335 set with and bar As can be seen in Fig 5, the generation of the cylindricity errors also proves that the wall thickness (droot /d1 ) must be taken into account in the selection of the clamping force Although there are no prescriptions for cylindricity in the drawings, their magnitude must be controlled Fig – Diagrams of hard-turned and ground profiles 336 j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 ( 0 ) 328–338 because they significantly influence the function of the gears As can be observed in Fig 5, for the ratio of 2.83 the cylindricity error does not exceed 0.004 mm but for the ratio of 1.34 it is 3–4 times higher Size accuracy of hard turning The size accuracy of hard turning was measured for three gears produced in series The measuring series consisted of 339, 297 and 102 pieces The result of the measurements is presented in Fig 6.The magnitude of scattering does not exceed the whole tolerance range for the bores with accuracy IT6 This means that on the lathe the accuracy IT5 can be ensured easily Moreover, if the possibility of size correction is used maximally, it is possible to meet more restrictive tolerances as well, with proper rigidity and good maintenance of the machine tool Furthermore, nowadays machine tool factories produce lathes not only for disc-type parts but also for external, cylindrical, conical and shaped surfaces Their guaranteed machining accuracy is IT5 with out-of-roundness of ␮m and a cylindricity deviation of ␮m Formation of accuracy in comparison with grinding Many factories have already changed to hard turning and, therefore, in some cases grinding is only applied as a necessary solution However, we were able to measure gears produced with both hard turning and grinding Fig introduces the main data and measuring diagrams of the relatively large The out-of-roundness diagrams apply to the plane noted as C because, according to our experience, the highest out-of-roundness always appears here (Kundrak and Bana, 2003), due to the deforming effect of the three-jaw chuck The characteristic out-of-roundness is a three-lobe form for both processes Cylindricity error, like out-of-roundness, is higher for hard turning The reason for this is the higher clamping force and form generation with a single-point tool, whereas in grinding the bore is conical At the point where the grinding tool is first applied the diameter is larger, since on this side the wheel is more overtravelled than on the other side The parallelism is better in hard turning but on the entering side the diameter is smaller due to insufficient conduction of the intensive heat The parallelism error of grinding is higher; the expansion on the entering side can be explained by the running out of the end stroke of the wheel to a larger extent The final results of measurements, with the addition of flatness and the axial run-out of face, are presented in Fig Although the out-of-roundness is higher in hard turning, it still satisfies the quality prescribed for IT5 The parallelism of the generatrices is also suitable The cylindricity tolerance is not prescribed, therefore, its generation is less important The flatness is adequate for hard turning, but for grinding it is higher than the permitted limit The axial run-out is adequate in both hard turning and grinding but much higher in grinding, where it is about five times of the hard turned values These results show that the ground gear does not fulfil the require- ments prescribed in the drawing because of its large flatness error The following hard turned and ground pair (Fig 9) also possesses a bore with accuracy IT5, but their wall thickness (rigidity), at the droot /d1 ratio was 1.66, is significantly lower than the previous one of 2.32 The data for the gears, prescriptions for accuracy, and the diagrams of accuracy is given in Fig In comparison with the previous gear, the out-ofroundness is much higher and because of the three-jaw chuck, a three-lobe form can be observed, which appears here more clearly than before The cylindricity error is slightly higher in hard turning but in grinding the bore diameter is higher on the entering side The reason for the conicity is that wheel overtravel occurs to a higher degree here than on the chuck-hand side Here the parallelism of the generatrices for hard turning is also better than in grinding In hard turning, at entering a lower accumulation of heat can be observed, which increases at outgoing In addition, on the hard turned piece a softer mark was found on the scanned side that was abreacted by the single point tool by more material removal On the bore of the ground gear the result of unequal overtravel of the wheel also appears The results of all performed measurements are summarized in Fig The out-of-roundness measured in planes A, B, and C is notable It is always the highest in plane C because here the deforming effect of the clamping force is the highest Moreover, for this gear neither hard turning nor grinding meet out-of-roundness limits, especially in the case of hard turning The prescription for cylindricity is not given, therefore its shaping is neutral The parallelism of the bore’s generatrices is definitely acceptable Comparing Figs and 10, it is prominent that the flatness error and the axial run-out of the ground part face are 4–5 times higher than those of the hard turned components This arises because of the grinding process applied The face has to be machined on the internal grinding machine However, face grinding can be regarded as a necessary procedure on these machine tools The armed apparatus that clamps the grinding spindle can be brought into work position outside As a consequence of this we must work with a light apparatus with low rigidity, which leads to a higher degree of deformation from the grinding force Moreover, the working of the appa- Fig 10 – Summary of measured results j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 ( 0 ) 328–338 ratus that performs the dressing of the grinding wheel is not always precise, because sometimes the dressing of the small cup wheel is done by hand Because of the effort made for relatively small sizes, the grinding quill is also weak Therefore it is no wonder that the face plane is not precise 337 As for the axial run-out, the surface grinding of the face is performed on a separate machine in a different clamping While the clamping was done on the pitch circle in internal grinding, for face grinding the clamping is on the opposite plane surface The alteration of the clamping surface causes Fig 11 – Diagrams of hard-turned and ground profiles 322 M Jacobson et al / Journal of Materials Processing Technology 128 (2002) 318–323 shows no alteration to the surface However on the same test samples thin layer of changed structure could be found, see Fig This white layer was less then mm deep when using cutting speed at 50 m/min This type of transformed surface was approximately found on less then 5% of the surface The reason for evaluating the test sample from 50 m/min was that it showed the highest compressive stress values It is therefore believed that the lower speed gives less heat and less transformation of the surface layer The white layer normally gives tensile stress Discussion The surface generated by hard turning bainite steel depends on the machining conditions This study has shown how the cutting speed affect residual stresses This study is a small part of a bigger investigation on how different machining parameters affect the result generated in hard turning More then 100 residual stress profiles have been measured with different machining parameters This larger study support that the cutting speed has this effect on residual stress generated The amount of error on an individual sample deviates, depending on how well aligned the sample is with the X-ray diffraction equipment An average of Æ20 MPa on the measured value is the normal error The stylus probe measurement equipment used for roughness measurement showed stable results for roughness measurements The tests were conducted with machines and tools used for normal production The purpose was to test with conditions as close as possible to normal production Results presented in this study are valuable since they show that compressive residual stress can be generated when bainite steel is hard turned It is also clear that the amount of compressive stresses generated can be altered depending on the cutting speed When residual stress is generated there is an effect from both the heat generated and mechanical work going in to the surface and subsurface When increasing cutting speed one also increase the strain rate in the process which gives more mechanical work in to the process, leading to compressive stress When increasing the cutting speed more heat is introduced in the process, which has a tendency to produce tensile stress at the surface This could explain why the compressive stress first increase then later decrease while increasing the cutting speed The two opposing phenomena counteract to create the final result The amount of residual stress that is beneficial for improving bearing fatigue life differs with each application For a theoretical case Broszeit et al [9] concluded that the minimum equivalent stress maximum occurs when the residual stress is À0.21 of the maximum Hertzian contact pressure Schreiber et al [10] have a more practical approach In their study, they introduced different degrees of residual stress by using different time periods when coldworking bearing balls The cold-worked balls were then assembled into a 6206 deep groove ball bearing After assembly, a rolling contact fatigue test was performed The rig used a Hertzian contact pressure of 3500 MPa This more practical approach resulted in the conclusion that an optimum of residual stress is achieved at around À600 to À1000 MPa Liu and Mittal [4] also concluded that in order for a residual stress profile to have a favourable effect, at every point on the service stress profile the residual stress must be of the same sign as the service stress Also, the magnitude of the residual stress should be less than or equal to twice the magnitude of the service stress Service stress in rolling contact is primarily compressive in nature Therefore, for the initial residual stresses to be beneficial, they also must be compressive Bearing this in mind, it is clear that hard turning of bainite steel can be favourable in terms of improving fatigue life The measured values fall in the range mentioned above, and can be altered to suit different applications The equation for Rt (Eq (1)) only accounts for the geometric portion of the roughness The geometric roughness and secondary roughness have been well described by Capello et al [8] The real surface finish is a combination of the geometric roughness and the secondary roughness derived from the plastic flow of the material, the removal process, the tool wear, the tool, and workpiece chatter among others The equation is useful for medium and high levels of surface roughness, since the geometric portion dominates the roughness value For lower levels of roughness, the model systematically underestimates the real surface roughness, because in this case the secondary roughness becomes more significant than geometric roughness When analysing the surface it becomes clear that when the cutting speed is low, too low temperature is reached in the cutting zone and the material is hard to cut On the other hand if the cutting speed is to high the amount of heat generated makes the material soft enough to escape the cutting edge When cutting with 170 m/min the right temperature is reached and roughness depends only on the geometric roughness The surface roughness value Rt then fits Eq (1) well A white layer is a result of microstructure alteration It is called a white layer because it resists standard etchants and is white under an optical microscope In addition, white layers often produce tensile stresses at the surface An interesting question is however this very thin layer has a detrimental effect on fatigue life White layer produced in grinding and in hard turning is not the same thing, this is well described by Koă nig et al [11] With the information available concerning the surface integrity, there is indication that hard turning of bainite steel is a possible process An interesting next step would be to perform a fatigue test with hard turned components Mittal and Liu [3] have already shown that an extremely good surface finish can be obtained with hard turning The robustness of the process also must be documented Koă nig et al [12] showed how the surface of a 52 100 component at first is compressive, but soon turns to tensile when tool wear appears M Jacobson et al / Journal of Materials Processing Technology 128 (2002) 318–323 323 Conclusions References Tests were conducted to find the effect of hard turning on the surface integrity of bainite steel The outcomes were reported as residual stress, surface roughness and microstructure analysis From the analysis of experimental data and discussions, the following conclusions can be drawn: [1] M Field, J Kahles, Review of surface integrity of machined components, Ann CIRP 20 (2) (1971) 153–163 [2] E Brinksmeier, J.J Cammett, P Leskovar, J Peters, H.K Tonshoff, Residual stresses-measurements and causes in machining processes, Ann CIRP 31 (1982) 491–510 [3] S Mittal, C.R Liu, A method of modelling residual stresses in superfinish hard turning, Wear 218 (1998) 21–33 [4] C.R Liu, S Mittal, Optimal pre-stressing the surface of a component by superfinish hard turning for maximum fatigue life in rolling contact, Wear 219 (1998) 128–140 [5] Y Matsumoto, F Hashimoto, G Lahoti, Surface integrity generated by precision hard turning, Ann CIRP 48 (1) (1999) 59–62 [6] A.M Abra˜ o, D.K Aspinwall, The surface integrity of turned and ground hardened bearing steel, Wear 196 (1996) 279–284 [7] Y Fujimoto, O Ohhata, Providing Residual Stress by Machining after Heat-treatment, Elsevier, Amsterdam, 1992, pp 1350– 1355 [8] E Capello, P Davoli, G Bassanini, A Bisi, Residual stresses and surface roughness in turning, J Eng Mater Technol (1999) 346–351 [9] E Broszeit, Th Preussler, M Wagner, O Zweirlein, Stress hypotheses and material stresses in Hertzian contacts, Z Werkstofftech 17 (1986) 238–246 [10] E Schreiber, W Simon, H.-W Zoch, Work hardening of ball surfaces, ASTM Special Technical Publication, Symposium on Creative Use of Bearing Steels, no 1195 (1993) 8192 [11] W Koă nig, A Berktold, J Liermann, N Winands, Top-quality components not only by grinding, Ind Diam Rev 54 (562) (1994) 127–132 [12] W Koă nig, A Berktold, K.-F Koch, Turning versus grinding—a comparison of surface integrity aspects and attainable accuracies, Ann CIRP 42 (1993) 39–43  Hard turned bainite steel consistently shows the same residual stress profile along the depth The amount of compressive stress increases to a maximum of À775 MPa around 10 mm below the surface At a depth of 100 mm, the effect from hard turning on residual stress is small  The cutting speed clearly affects the amount of residual stress generated in hard turning Maximum compressive stress is generated at a cutting speed of 230 m/min  The level of compressive residual stress produced is, considering the literature, at the right level to have a positive effect on rolling fatigue life  The cutting speed affects the surface roughness At a speed of 170 m/min a minimum Rt value is found This correlates well with the surface roughness that can be predicted from Eq (1)  The microstructure analysis showed little or no alteration to the surface or subsurface Acknowledgements The authors would like to thank SECO and SKF The authors is also grateful for the funding received from Vinnova Surface Integrity Difference between Hard Turned and Ground Surfaces and Its Impact on Fatigue Life F Hashimoto1 (2), Y.B Guo2, A.W Warren2 The Timken Company, Canton, Ohio 44706, USA Dept of Mechanical Engineering, the University of Alabama, Tuscaloosa, AL 35487, USA Abstract Surface integrity is the key for process selection of hard turning or grinding in industry This study identifies the fundamental differences in the integrity of hard turned and ground surfaces and the subsequent impact on rolling contact fatigue life The significant findings are: (a) mechanical deformation plays a larger role during hard turning than grinding, while the size effect in grinding induces higher surface hardness; and (b) a hard turned surface may have a more than 100% longer fatigue life than a ground one with an equivalent surface finish (Ra 0.07 µm) due to the very different characteristics of surface integrity Keywords: Surface, Fatigue, Hard Machining INTRODUCTION Hard turning and grinding [1,2,3,4,5,6,7] are competitive finishing processes for the manufacture of precision mechanical components However, their very different process characteristics make process selection a great challenge in industrial application Fundamental process knowledge is the key for process selection and design The different geometrical features of cutting edges and grains used in hard turning and grinding create different surface structures For example, a turned surface shows much wider and more regular feed marks than those of a ground one Grinding can achieve very smooth surfaces of less than 0.05 µm Ra Hard turning may achieve equivalent or better 2D surface roughness under certain cutting conditions The peaks and valleys of the turned and ground surfaces also differ in depth and occurrence Both surface types show negative skewness [8] which gives good lubricant retention capability With the tool and grinding wheel wear, surface structure will be changed significantly and a 3D surface map will be necessary to evaluate its effect on component performance However, very little research has been done in this regard The most significant difference in residual stress by fresh hard turning and grinding is that hard turning may induce a relatively deep maximum compressive residual stress in the subsurface [4], while grinding produces maximum compressive residual stress at the surface A worn tool or grinding wheel induces tensile residual stresses on the surface [9] A compressive residual stress induced by hard turning and grinding was found to improve rolling contact fatigue (RCF) life [4,10] Furthermore, deep compressive residual stresses are much more beneficial to bearing fatigue life than shallower stresses of greater magnitude Recent studies [11,12] have shown that distinct residual stress patterns hardly affect either the magnitudes or the locations of peak stresses and strains in the subsurface They do, however, have a significant influence on surface deformations The slope and depth of a compressive residual stress profile are key factors for RCF damage Nevertheless, it is very challenging to Annals of the CIRP Vol 55/1/2006 isolate the effect of process induced residual stress profiles on fatigue The combined effects of severe surface deformations, high local temperatures, and rapid quenching rates cause the machined surface to undergo both physical and metallurgical changes such as a sandwiched surface structure including white and dark layers [1] Tensile residual stresses and increased hardness are often associated with a white layer Increased surface hardness is often encountered at gentle machining conditions [13] White layer properties such as hardness, retained austenite, residual stress, and chemical compositions are significantly different between hard turning and grinding [1,14,15,16] For example, grinding tends to produces a much thicker white layer than turning [6,16] Component performance such as RCF life [10] can be significantly reduced by the presence of a white layer by hard machining Since white layer is a generic description of structure appearance under a microscope, the properties and effects of a white layer must be analyzed for a specific process and operational loading However, whether hard turning or grinding at best practice produces similar surface integrity (surface structure, micro/nanohardness, microstructure, etc.) is not clear yet Furthermore, the effect of a turned or ground surface free of the white layer on RCF life has not been clarified Since process induced surface integrity is of great importance for component performance in service, the objective of this study is two-fold: (1) to reveal the basic differences of surface integrity by turning and grinding hardened AISI 52100 steel at best practice; and (2) identify the subsequent impacts of surface integrity on RCF life MACHINING TESTS Work samples of AISI 52100 steel were cut from a 76.2 mm diameter solid bar at 19.05 mm thickness The test specimens were then machined to ensure parallelism and perpendicularity before heat treatment Heat treatment consisted of austenizing at a temperature of 815ºC for hours, followed by quenching in an oil bath for at 65ºC for 15 minutes Finally, tempering was conducted at 176ºC for hours which produced a final hardness of 61-62 HRC The test specimens were then machined by both turning and grinding processes as shown in Tables and The machining parameters were selected to fall within the gentle machining range so as to ensure that there would be no phase transformations on the machined surface The face turning operation was conducted using a CNC BRIDGEPORT lathe which was able to maintain a constant cutting velocity as the cutting tool moved from the outer edge to the center of the workpiece A fresh round CBN cutting tool insert was used for the facing operation and was rotated a certain degree after each sample was completed in order to use a fresh cutting edge in each cutting test A face grinding operation was performed using a vitrified Al2O3 wheel which was dressed prior to machining and ample coolant was used to prevent excessive heat at the machined surface After machining, test samples were cross sectioned with an abrasive cutter, mounted in epoxy, and polished with alumina polishing compound until a mirror finish was obtained Surface integrity of the machined specimens was then characterized by surface finish, microstructure, and micro/nanohardness a result of the shallow depth of cut and small feed rates used in the turning process, the as turned surface roughness is better (≈ 30%) than that of the as ground surface The surface roughness of the samples after polishing is within ±15% of each other The roughness Ra between 0.06 µm ~ 0.07 µm of the polished surfaces is smooth enough to eliminate or at least minimize the roughness effects on following micro/nanoindentation tests (a) as ground 0.1 mm 0.1 mm (c) ground & polished Work material Velocity AISI 52100 Steel (62 HRC) CBN, GE BZN 8100 (0.015/15° chamfer & radius of 6.35 mm) 1.98 m/s Feed DoC 0.051 mm/rev 0.254 mm Sample Type Coolant type # of samples Dry turning Ground Turned Cutting tool AISI 52100 Steel (62 HRC) Grinding wheel Vitrified Al2O3 (dia 254 mm) Wheel speed 24.0 m/s Table speed 15.2 m/min Cross feed Down feed (rough) 1.1 mm/pass 12.7 µm/pass # Passes (rough) Down feed (finish) 5.1 μm/pass # Passes (finish) Coolant type Water soluble # of samples 0.1 mm (d) turned & polished Surface Roughness Ra (µm) as machined as polished 0.181 0.071 0.138 0.061 Table 3: Surface roughness 3.2 Subsurface microstructure To view subsurface microstructure of the specimens after machining, the samples were polished to an acceptable mirror finish and then etched with a 2% nital solution Optical microscope images of the cross-sections of the test specimen are shown in Figure 2(a),(b) Hardened zone (4 µm) Table 2: Grinding conditions SURFACE INTEGRITY CHARACTERIZATION 3.1 Surface structure Surface structure was characterized by both an optical microscope and by stylus measurement Optical images of the as turned and as ground surfaces as well as the subsequently polished surfaces are shown in Figure 1(ad) The ground and feed marks are clearly seen on both surfaces and each is characterized by the corresponding machining method The grinding wheel is composed of many random irregular bonded abrasives and therefore the machined surface has a random distribution of grinding marks In contrast, hard turning is performed by a geometrically defined cutting edge and, therefore, the feed marks have a uniform spacing determined by the prescribed feed rate Surface profile measurement was carried out using a stylus profilometer The average surface roughness, Ra, is calculated for all cases and is summarized in Table As 0.1 mm Figure 1: Optical image of the machined surface Table 1: Hard turning conditions Work material (b) as turned Thermal affected zone (14 µm) (a) Hardened zone (6 µm) Thermal affected zone (8 µm) (b) Figure 2: Optical Images of subsurface microstructure Both turned and ground surfaces are free of thermal damage, while grinding temperature has a much deeper excursion in the subsurface It shows that the subsurface has two different zones characterized by a strain hardened zone in near surface and a thermally affected zone in the subsurface (not softened compared with bulk hardness as seen in the following section) It should be pointed out the strain hardened zone is not a white layer despite its appearance in comparison to the bulk material, and the thermally affected zone is not the dark layer encountered in hard turning The different appearance of the two zones is just due to their different resistance to etching as a result of different grain deformation and size Surface Type Ground Turned Surface Hardness Microhardness, 25 gf Nanohardness, GPa Machined Polished Polished 1183 1148 8.8 929 1046 7.0 1100 900 ROLLING CONTACT FATIGUE (RCF) TESTS 4.1 RCF testing conditions RCF tests of the polished turned and ground samples were conducted at peak Hertz pressure of 4.5 GPa (above the shakedown limit) and spindle speed of 2600 rpm After polishing, surface roughness of all the turned and ground samples is equivalent Eight chrome steel ball bearings of diameter 5.56 mm were used The samples and balls are First points measured in subsurface 700 Hard Turned 500 30 60 Ground 90 120 150 Depth below surface, um µm Table 4: Surface hardness Figure 3: Microhardness in subsurface Nanohardness, GPa Subsurface measurements were performed as a 10 × matrix with row and column spacing of µm and 20 µm, respectively to avoid interference with each other The subsurface microhardness is shown in Figure The ground specimens had a higher hardness not only on the top surface but also throughout a depth of 100 µm when compared to the turned ones At depths greater than 100 µm the hardness of both specimens become uniform It was observed that the hardness was lowest at the first positions in near surface The apparent “softening” is not due to thermal effects, but rather is caused by edge effect induced by the indent size 3.4 Surface and subsurface nanohardness Since indent size in a hardness test is critical to evaluate material property at small scales, nanoindentation was performed to avoid the “edge effect” since the nanoindent size is substantially smaller than that of a microindent [16] Nanohardness was measured using a Hysitron Triboindenter The measurements were made with a diamond Berkovich indenter with a maximum load of mN Nanohardness of the machined surface was determined using a × matrix of indentations directly on the top surface The average nanohardness was in Table and the trend agrees with the measured microhardness data Subsurface nanohardness in Figure was determined using a × matrix of indentations The surface nanohardness for the ground specimen is about 25 % higher than that of the turned one It is clear that the apparent softening associated with edge effect observed in microindentation is not present in nanohardness test The nanohardness for the as ground specimen remains slightly higher than that of the as turned one to a depth of about 30 µm The nanohardness profiles agree in trend with the measured microhardness data in Figure However, the different depth of hardness variation may be influenced by the different interactions between material second phase particles and indenters The basic mechanism for the more hardened ground surface/subsurface is most likely due to the size effect induced by severe strain gradient in grinding The smaller down feed in grinding induces a severe strain gradient in near surface, while the relatively larger depth of cut in turning may substantially reduce the size effect Measured on very top surface 1300 HK (25 gf) 3.3 Surface and subsurface microhardness Microhardness of the as machined and polished (0.07 µm Ra) samples was measured with a Knoop indenter at a load of 25 gf The microhardness of the ground surface is about 27% and 10% higher than those of the as turned and as polished surfaces respectively as shown in Table 10 Measured on very top surface First points measured in subsurface Hard Turned Ground 10 15 20 25 30 µm Depth below surface, um Figure 4: Nanohardness in subsurface well lubricated during the fatigue tests A load cell was used to real-time monitor the applied load An acoustic emission (AE) sensor was mounted to the specimen to on-line monitor fatigue damage process AE signals of counts, amplitude, RMS, and energy were then sent into a PC for signal processing The pre-amplifier gain was set to 40 dB with a threshold of 45 dB The monitoring continued until increased AE signals were detected because of surface pitting and spalling Based on the fatigue testing design, the RCF life (# Cycles) can be calculated as # Cycles = nN t (1) where n is the number of balls, N is the spindle speed, and t is the total testing time 4.2 Life comparison of turned vs ground surfaces Figures and show the sensitivity of AE amplitude to fatigue (spalling) The shoots of AE amplitude correspond to surface spalling AE amplitude is the most sensitive signals among the other signals [10] Subsurface fatigue cracks propagate along ≈ 30°-45° with regard to the rolling direction To evaluate the variation of fatigue life, three samples of each surface type were tested The average RCF life and deviations are shown in Figure It shows that the turned and polished surfaces have an average life 315.8 (± 76.8) million cycles, while the ground and polished surfaces have an average life 157.8 (± 21.3) million cycles It demonstrates that a superfinished turned surface may have more than 100% fatigue life than a ground one with equivalent surface finish The fundamental mechanisms that contribute to the fatigue difference are the distinct surface structure and properties generated in the processes However, further isolating the effect of individual surface integrity factors such as residual stress on fatigue will be a future research subject Amplitude (dB) vs time (sec) ACKNOWLEDGMENT This research is based upon work supported by the National Science Foundation under Grant No DMI0447452 Amplitude shoot 100 80 Rolling direction 60 40 Subsurface cracks 20 0 200000 34.4 500000 86.0 800000 137.6 1000000 Time (sec) 172 (106 cycles) Amplitude (dB) vs time (sec) Figure 5: AE signal and the fatigued ground sample 100 Amplitude shoot 80 Rolling direction 60 40 Subsurface cracks 20 Time (sec) 0 500000 1000000 1500000 86 172 258 2000000 2500000 344 (10 cycles) RCF life (10^6 cycles) Figure 6: AE signal and the fatigued turned sample 400 300 200 HT 100 G Surface type Figure 7: Life comparison of turned vs ground samples SUMMARY The new findings of surface integrity and fatigue resistance for hard turned and ground surfaces can be summarized as follows: Hard turning can achieve very smooth surfaces (0.14 µm Ra) being equivalent to grinding (0.18 µm Ra) The turned surfaces have much wider and more regular feed marks than those of the ground ones Mechanical deformation plays a larger role during turning, whereas surface temperatures penetrate deeper into the subsurface in grinding The size effect is more prominent in grinding than turning, which is demonstrated by the higher hardness in the ground surface/subsurface The apparent “softening” measured by microindentation is due to the surface edge rather than thermal effect The effect of a turned or ground surface free of the white layer on fatigue life was clarified A superfinished turned surface may have a fatigue life twice as long as a superfinished ground one with equivalent surface finish REFERENCES [1] Töenshoff, H.K., Wobker, H.G., Brandt, D., 1995, Hard Turning-Influences on the Workpiece Properties, NAMRI/SME, 23: 215-220 [2] Elbestawi, M.A., Srivastava, A.K., El-Wardany, T.I., 1996, A Model for Chip Formation during Machining of Hardened Steel, 45/1:71-74 [3] Poulachon, G., Moisan, A., 1998, A Contribution to the Study of the Cutting Mechanisms during High Speed Machining of Hardened Steel, Ann CIRP, 47/1:73-76 [4] Matsumoto, Y., Hashimoto, F., Lahoti, G., Surface Integrity Generated by Precision Hard Turning, Ann CIRP, 48/1:59-62 [5] Shaw, M.C., Vyas, A., 1998, The Mechanism of Chip Formation when Hard Turning Steel, 47/1:7780 [6] Malkin, S., Guo, C., Wu, Y., Varghese, V., 1999, Temperature and Energy Partition for Grinding with Vitrified CBN Wheels, Ann CIRP, 48/1:247-250 [7] Klocke, F., Kratz, H., 2005, Advanced Tool Ege Geometry for High Precision hard Turning, Ann CIRP, 54/1:47-50 [8] Kundrak, J., Bana, V., 2003, Microgeometry of Bore th Holes after Hard Machining, Int Res Confer., TMT 2003, Lioret de Mar, Barcelona, Spain [9] Konig, W., Klinger, M., Link, R., 1990, Machining Hard Materials with Geometrically Defined Cutting Edges-Field of Applications and Limitations, Ann CIRP, 39:61-64 [10] Schwach, D.W., Guo, Y.B., 2005, A Fundamental Study on the Impact of Surface Integrity by Hard Turning on Rolling Contact Fatigue, Trans of NAMRI/SME, 33:541-548 [11] Guo, Y.B., Barkey, M.E., 2004, FE-Simulation of the Effects of Machining-induced Residual Stress Profile on Rolling Contact of Hard Machined Components, Int J of Mechanical Sciences, 46/3:371-388 [12] Guo, Y.B., Barkey, M.E., 2004, Modeling of Rolling Contact Fatigue for Hard Machined Components with Process-Induced Residual Stress, Int J of Fatigue, 26:605-613 [13] Guo, Y.B and A.W Warren, A.W., 2004, Microscale Mechanical Behavior of the Subsurface by Finishing Processes, ASME J Manuf Sci Eng., 127:333-338 [14] Barry, J., Byrne, G., 2002, TEM Study on the Surface White Layer in Two Turned Hardened Steels Mat Sci Eng., A325: 356-364 [15] Guo, Y.B., Janowski, G.M., 2004, Microstructural Characterization of White Layers by Hard Turning and Grinding, Trans NAMRI/SME, 32:367-374 [16] Guo, Y.B., Sahni, J., 2004, A Comparative Study of the White Layer by Hard Turning versus Grinding, Int J Machine Tools and Manufacture, 44 :135-145 Journal of Materials Processing Technology 153–154 (2004) 746–750 Identification of cutting errors in precision hard turning process J.M Zhou∗ , M Andersson, J.E Ståhl Department of Mechanical Engineering Lund University, Division of Production and Materials Engineering, Lund, Sweden Abstract During past 10 years, the availability of polycrystalline cubic boron nitride (PCBN) tools has made application of hard turning increase remarkably in a range of industrial areas, especially in automotive, roller bearing and hydraulic industry Among the applications of CBN cutting tools, precision hard turning attracts great interests since its environmentally friendly and flexibility, which ultimately has potential to replace grinding Using today’s high precision hard turning with high precision motion control, high static and dynamic stiffness and thermal stability CNC machine, the dimensional accuracy can reach IT5 However, this accuracy still does not meet the demands from most industries In addition, precision hard turning has been hampered by uncertainties with respect to part quality and process reliability Many factors affect the machining accuracy and process stability The presented work analysis the possible error driver factor and error sources in a precision hard turning and the strategy of on-line compensation of the dimensional errors in the process, based on the monitoring of tool wear and prediction of thermal expansion Test results indicate that within the certain range of flank wear the developed method can improve the geometric accuracy significantly © 2004 Elsevier B.V All rights reserved Keywords: Hard turning; PCBN; Monitoring; Cutting Introduction Hard turning, using superhard cutting materials such as polycrystalline cubic boron nitride (PCBN), is an environmentally friendly and flexible manufacturing technique The potential advantages of precision hard turning include low production cost, high productivity, flexible tooling and workpiece quality In many cases, single point cutting tools can complete the entire machining process in a single fixture, thereby enabling reduced setup times and lower costs The use of standard tool shapes and the generation of part geometry via CNC systems guarantee the enormous flexibility of this process Today, hard turning technology has been widely used in various industrial areas, such as roller bearings, automotive and hydraulic parts [1] However, in the area of precision hard turning, due to demands of geometric accuracy of a few microns, its application is limited by uncertainties with respect to part quality and process reliability Although the high precision hard turning has enormous potential for the substitution of grinding operation in the manufacture of precision parts, which will be subjected to high stress levels, it must, however, be bear in mind that the kinematics of a turning operation differ greatly from those of grinding process In turning operations, each point on the ∗ Correpoding author Tel.: +46 46 2228601; fax: +46 46 2224529 E-mail address: jinming.zhou@mtov.lth.se (J.M Zhou) 0924-0136/$ – see front matter © 2004 Elsevier B.V All rights reserved doi:10.1016/j.jmatprotec.2004.04.331 workpiece is produced by the cutting edge at a precisely defined moment Any relative displacement between workpiece at the machining point, results directly in form and dimensional errors [2] It is vital, therefore, to ensure prior to the machining operations, that one is familiar with the type of cutting errors which are likely to occur during an operation It is then possible to enhance the machining outcome via control based compensation strategies or to minimise errors by ensure optimal design [3] In presented work, the possible error driver factors and error sources in precision hard turning were analysed and a strategy was proposed for cess, tooling, machine structure and fixture devices Major error drive factors, error sources and their relationships are schematically shown in Fig J.M Zhou et al / Journal of Materials Processing Technology 153–154 (2004) 746–750 747 cutting zone becomes so viscous that it fills the grooves and flows in a uniform and homogeneous way to the side of the cutting tool forming high ridges [2] There are many causes of geometric form errors that can be attributed to the manufacturing process sequence Form errors are often due to errors in machining and slide way error, but they can also be the result of sagging of the workpiece under its own weight, thermal effects produced during machining, deflections of the workpiece or tool, or stress relief after machining 2.2 Thermal energy Fig Major error driver factors and error sources in precision hard turning The effects of tool wear, cutting forces and process temperature are superposed and lead to displacement between cutting edge and surface of the workpiece, thus resulting in form and dimension errors in final products 2.1 Tool wear Tool wear could have significantly effects on dimensional, form and surface roughness errors As a tool becomes worn, the geometry of the tool tip is changed The wear of the tool tip on the clearance side will result in loss of the effective depth of cut, which can generate both dimensional and form errors of the workpiece by change of alignment between tool and workpiece Fig shows the development of diameter differences during hard turning of 30 rings Experiment shows that for a CBN tool with 20◦ in chamfer angle and 10 ␮m of edge radial, when flank wear reach 0.2 mm, the errors in radial/axial direction will grow up to 25 ␮m Increase of thrust force, due to tool wear, will also have a great effect on the dimensional and form accuracy Experimental results show that when a tool changes from sharp to 0.2 mm in flank wear (VB), thrust force level will increase more than 200%, which can have substantial impact on the accuracy by the induced deformation in machine, tooling and chuck Worn tool will also deteriorate the surface topography Due to the thermal effect of tool wear, the material in the Thermal energy is one of the greatest factors influencing the accuracy in precision hard turning by inducing thermal expansion of tool and workpiece The effect of thermal influence on precision hard turning is one of major error source affecting the dimensional and form accuracy of part According to McKeown et al [4], the thermal effects can contribute more than 50% to overall error of the machined workpiece The thermal expansion includes expansion of the workpiece and cutting tool, owing to heat generation and transfer during the cutting process Fig exhibits the results of thermal expansion on both workpiece and tool based on FE calculation Results from experiment and numerical calculation indicate that in finishing hard turning thermal expansion on the tool tip and workpiece can reach up to 10 and 15 ␮m, respectively [5,6], as shown in Fig 3(b) The thermal expansion of both tool and workpiece changes the effective depth of cut of the tool, which leads to cutting error of the workpiece During precision hard turning the cutting zone temperature has to be at least 800 ◦ C to soften workpiece for cutting This leads to an area of high temperature around the actual cutting zone in the tool nose and the workpiece The temperature changes were detected to be in between and 15 K in tool shaft and workpiece, depending on the process parameters [4] The thermal expansion of tool shaft and workpiece contribute the major form deviation in workpiece in the dry cut With coolant, the peak values of tool and workpiece temperature could be reduced essentially by more than 70% and form deviation come down to 50% of the error comparing without cooling [6] 2.3 Clamping Fig Effects of tool wear on dimensional accuracy Chuck is crucial part to maintain the high form accuracy and dimensional tolerance in precision machining The possible machining errors may be induced by different type of chucking, chucking forces, chucking rigidity and accuracy In rolling bear machining, the maximum deformation in radial/axial direction may reach up to ␮m by using ordinary three jaws chuck due to the deformation of the ring under the chucking forces Deformation may vary with different workpiece and clamping force The chucks with even distribu- 748 J.M Zhou et al / Journal of Materials Processing Technology 153–154 (2004) 746–750 Fig Calculation of thermal expansion on workpiece and cutting tool by FE [5] (a) Thermal expansion on workpiece under different tool conditions (b) Thermal expansion on cutting tool tion of chucking forces, such as six jaws chuck or magnetic chuck are recommended in this case Even magnetic chuck is not prefect, which will meet the problem of chip entanglement during the cutting due to the magnetic force and centre of the workpiece The trend of form errors shows higher in chuck side than non-chuck side in precision hard turning of rolling bear due effect of the magnetic force Fig represents errors in roundness when using three different chucks Table Properties of different tool holder materials Young’s modulus (GPa) Thermal expansion (10−6 /K) Thermal conductivity (W/m/K) Density (g/cm3 ) Hardness (HV) Tensile strength (MPa) Carbide Steel 320–340 6.0 80 17 280–330 700–950 206 11.5 45 7.8 450 1450 2.4 Tooling The geometric and kinematical errors can also come mostly from inaccuracy of the tool system, such as, insert, tool holder and insert clamping device Investigation has suggested that after re-clamping insert, the repeatability errors at the tip of the insert can reach up to several microns, and the displacement of the tool tip under the cutting load can also reach several microns [7,8] These errors are concerned with the quasi-static errors and can easily be calibrated and compensated In finishing hard turning, cutting forces are much higher comparing with conventional turning Higher forces may cause elastic deformation on tool holder and workpiece and change the position of tooling and machine stiffness In fact, in many cases tool holder reinforced with high density metal or even solid high density metal tool holder, such as carbide tool holder, has been recommended in precision hard turning because of its better thermal conductivity, damping properties and strength Compared with conventional tool holder, the superior mechanical properties of carbide tool holder can be seen from Table 2.5 Machine Precision hard turning demands higher precision machine to eliminate the form and dimensional errors induced by moving component of the machine, such as spindle, slide bed and tail stock The accuracy demands on these machine tools are comparable with grinding machine tools concerning static, dynamic and thermal stiffness, as well as accuracy of the spindle system and slides The application of conventional machine tools, with proper modification in head stock and slides, enable to reach workpiece qualities in the range of IT6-7 in hard turning Error compensation Fig The errors caused by chucks Error compensation could be divided into two categories depending on the extent of the repeatability of the system, according to Ramesh et al [3] One method is ‘pre-calibrated J.M Zhou et al / Journal of Materials Processing Technology 153–154 (2004) 746–750 error compensation’ wherein the error is measured either before or after a machining process and the same is used to alter or calibrate the process during subsequent operation The other method is ‘active error compensation’ wherein the error as monitored during the machining process is used to alter the process during the same operation Since the accuracy of a machining process is affected by the overall effect of the various error sources mentioned before, the error compensation system should take into consideration the interaction between these sources rather than consider individual error separately In order to account for errors on a continuous basis in such a way that the interaction between the various error components could be considered, real time error compensation system need to be employed Typically thermal induced errors are suitable to be compensated by pre-calibrated error compensation method due to the complication of temperature measurement and thermal expansion calculation during machining However, tool wear induced error may suitable to be compensated with active error compensation because it is relatively easer to monitoring tool wear during the cuttings The thermal expansion on workpiece and cutting tool can be predicted before machining by means of numerical calculations After the error is predicted, it can be compensated for in the CNC tool path The numerical methods used for this purpose include finite element method, finite difference method and boundary element method The accuracy of the results by these predictive methods, however, is very much dependent on how accuracy of the input for the numerical calculation and how close the boundary condition to the practical situation A knowledge database can be built for the results from numerical calculation for different cutting conditions Experimental result show that through error compensation the radial form error of the workpiece can be reduced to ␮m from 15 ␮m under condition of small flank wear (VB < 0.1 mm) [5] For tools with high flank wear the correction is less good, the actual cutting error seems to fluctuate during the fine cut The instability of the fine cut using a very worn-out tool can be attributed to the instable tool forces, which has been observed in force monitoring for hard turning When the flank wear of the CBN insert is above 0.1 mm, the stability of the cutting process is gradually deteriorated and the cutting force, especially the passive force, fluctuate within a large range which makes it difficult to identify the tool flank wear and the thermal expansion [8] On one hand, the fluctuation of the cutting force influences the cutting power, and thus changes the rate of heat flowing into the workpiece and the tool On the other hand, the fluctuated, large passive force induces significant deflection of the workpiece, which was not incorporated in the compensation of errors in the current cutting test Therefore, it is not sufficient to compensate only the error from thermal expansion if very a worn-out tool is used for the fine cut of hard turning Tool force induced deflection of the workpiece should also be incorporated in error compensation 749 Fig Method of dimensional error compensation in precision hard turning To minimise the error induced by tool wear and thermal expansion in precision hard turning, it is essential to identify the error in the different phases of the cut and then to compensate for it during later processing A method combining pre-calibrated error compensation and active error compensation was proposed as shown in Fig Tool wear can be identified during the process based on the relationship between passive force and tool wear in a tool wear monitoring system, as shown in Fig The thermal expansion error under different cutting conditions can be predicted off-line Based on the tool wear from a monitoring system and correspondent thermal expansion of workpiece and tool predicted from FE calculation, error compensation is made through calculation of a compensated contour on the workpiece and generation new NC code The NC code is transferred to the CNC system through the communication port between the monitoring system and the CNC’s control system Conclusions and discussions Error prediction and compensation is essential issue in super finishing hard turning Three error-drive factors, including tool wear, cutting forces and thermal expansions, play an important role These factors directly drive the error sources in process, tooling, machine structure and fixture in the process of error formation on workpiece during hard turning To predict and compensate the cutting errors, effectively monitoring of tool wear, cutting forces and cutting temperature is necessary In this article, two methods of error compensation were proposed One is pre-calibrated error compensation only based thermal expansion prediction with use of developed FE model Another was proposed for prediction and compensation of the total cutting errors by combination of active error compensation and pre-calibrated error compensation In this method, an in-process monitor- 750 J.M Zhou et al / Journal of Materials Processing Technology 153–154 (2004) 746–750 ing system was used for monitoring of tool wear and cutting forces to predict the tool wear and cutting forces induced errors Thermal induced errors are pre-calculated by using developed FE model, based pre-measured test values under different cutting conditions These pre-predicted thermal induced error are stored a knowledge database Active error compensations are achieved by synthesising the results from in-process monitoring system and knowledge database Test results suggested that it is possible to achieve the desired high accuracy of the workpiece by means of error compensation from the FE model based error prediction, especially for cutting conditions in which the tool wear for the fine cut is slight If the tool wear is severe in fine cut, the cutting becomes instable, owing to fluctuation of relatively large tool forces In this case, compensation of the error only from thermal expansion is not sufficient Tool wear monitoring becomes critical It can be concluded that the FE prediction-based error compensation is promising and shows a great potential in improving the quality of hard turned parts But more sufficient and robust cutting error compensation is to integrate tool wear and cutting force monitoring system with FE model based pre-error compensation Acknowledgements Authors would like to thank Dr Jochmann and Dr Lai for valuable contributions in the experiment, calculations and fruitful discussions in the project The authors would also like to thank European Community for supporting this project References [1] W Koenig, A Berktold, K.F Koch, Turning versus grinding A comparison of surface integrity aspects and attainable accuracy, Annal CIRP 42/1 (1993) [2] F Klocke, S Jochmann, Development of models to demonstrate the influence exerted on form and dimensional accuracy in high precision hard turning operations, in: Proceedings of the International Conference on Ultraprecision Manufacturing Engineering, 1997 [3] R Rsmesh, M.A Mannan, A.N Poo, Error compensation in machine tools – a review part I: geometric, Int J Machine Tool Manuf 40 (2000) 1235–1256 [4] P.A Mckrowen, M Weck, R Bonse, Reduction and compensation of machines, Annal CIRP 44 (2) (1995) 589–595 [5] J Lai, FE prediction of cutting error from thermal expansion of workpiece and tool in hard turning, MicroHard Project Report, SKF Engineering and Research Centre, B.V Nieuwegein, The Netherland, 2001 [6] J Bryan, International status of thermal error research, Annal CIRP 39 (2) (1990) 645–656 [7] J.M Zhou, M Andersson, J.E Stahl, Development of tool wear and cutting force monitoring system, Int J Adv Manuf 23 (5–6) (2003) [8] M Weck, J Luderich, A Wieners, High precision turning of hardened steel – a challenge for a new generation of machine, in: Proceeding of the 7th annual meeting of American Society of Precision Engineering, 1992 Surface Integrity Generated by Precision Hard Turning Y Matsumoto, F Hashimoto (21, G Lahoti (1) Timken Research, The Timken Company, Canton, USA Received on January 7,1999 Abstract Rolling contact fatigue tests were conducted to find the effect of precision hard turning The tests showed that hard turning provides as good a fatigue performance as grinding Hard turning produces compressive residual stresses in a deep subsurface, which contribute to a long fatigue life The effect of cutting parameters on residual stress was investigated in order to find why deep residual stresses are created It was determined that the tool edge geometry is the dominant factor deciding the residual stress profile Keywords: Cutting, Residual stress, Hard Turning INTRODUCTION Hard turning applications have increased drastically in the auto industry for the last 15 years, however, not many new developments have been seen in recent years One large but difficult area for a new expansion is in precision hard turning It demands geometry accuracy of a few microns Nonetheless, it is possible to meet the specificationswith a high precision machine tool and in a very tight process control environment as it was suggested in[l] Rolling contact surfaces require such accuracy There are increased interests to reduce costs of machining these surfaces The significant cost benefit comes from reduced labor, capital investment, tooling and set up time Prior to fully develop this application for production, the process needs to produce a surface that warranties good product performance A wide area of the surface integrity of machined surfaces was investigated by Metcut Research Associates[2] in the early 70s In particular, they showed a high correlation between subsurface residual stresses and the fatigue limit for high strength steels There was a similar work on grinding AlSl 52100 steel[3] In hard turning, one of the author's previous work[4] showed that the subsurface compressive residual stresses help increase fatigue life in tension None of these investigations involved rolling contact tests and their applied stress field is different from that of a rolling contact surface On the effect of surface integrity on rolling contact fatigue life, there exist several literatures A compressive residual stress produced by shot peening was found to improve rolling contact fatigue life(51 Scott et al.[6] showed that the compressive residual stress produced by grinding is beneficial to fatigue life of ball bearings Specifically, a compressive residual stress in a deep layer was more Annals of the CIRP Vol 48/7/7999 beneficial than a shallower residual compressive stress of greater magnitude From this background information, it was determined that the study needed to first find whether hard turned surfaces had any detrimental effect on rolling contact fatigue life or not Then, it had to examine how the surface integrity affects the fatigue life Finally the investigation was led to generate a preferred residual stress profile by modifying tool edge geometry BEARING FATIGUE LIFE TESTS Two fatigue tests were conducted using small taper roller bearing assemblies of a 120mm outside diameter for the first test and large taper roller bearings of a 450mm outside diameter for the second test Bearing materials for both tests were a case carburized steel with hardness of HRc 58-62 Two groups of bearings were prepared for each test For small bearing assemblies, the only difference between the two groups was the machining process for the bearing inner ring race One group of bearings had an inner ring race ground and supertinished This group represents the bearings machined to our best surface finish specificationand it was used as a baseline The other group had an inner ring race hard turned and superfinished Hard turning was done using a high precision CNC lathe to create a bearing race profile A CBN cutting tool with a 10 degree and 2mm chamfer was used No tool edge honing was applied It will be clarified later that the tool edge preparation is important for the surface integrity Both groups of bearings had a surface finish of less than 0.1 micron Ra after superfinishing Figure l a exhibits the fatigue life result for the small bearings The bar in the middle of the bands is the best estimate of the time by when 15.9% of bearings failed 59 The best estimates are exhibited in terms of relative lives instead of actual revolutions The bands indicate that 65% of the bearing's relative lives fall within this range The hard turned and superfinished bearing group gave 1.11 times more life with 90% confidence than the life of the ground and superfinished group For the large bearings, two groups were also prepared using similar processes to those used for small bearings Due to their large size it was more difficult to maintain geometrical accuracy This time all the surfaces of both inner and outer rings of one group of bearings were machined by hard turning The other group of bearings was ground All the race surfaces were superfinished Fatigue lives for both groups are shown in Figure 1b The fatigue life of hard turned and superfinished bearings was at least as long as that of ground and superfinished bearings with 90% confidence These were both very encouraging results EFFECT OF RESIDUAL STRESS In order to find a cause for the fatigue life difference between ground bearing and hard turned bearing, surface and subsurface conditions of bearings before fatigue tests were reviewed Since both groups in the above tests used a superfinish process for the final finish, the surface roughness and topographies were the same and they can be eliminated from the cause The roundness of hard turned bearings was better than that of ground bearings because of the difference between the accuracy of the machine tools used Both roundness were within the product tolerance and they not affect the fatigue life There was no microstructure alteration on the surface after grinding as well as hard turning as bearings were made with proper conditions For hard turning, a fresh insert was used to machine these bearings and the insert was replaced well before any significant amount of tool wear was developed An independent test showed that hard turning could produce untempered-martensite and overtempered-martensite near the surface using a worn tool The depth of the damaged layer was within microns In order to create the damage, these tools had to be used twice as long as the time used to prepare life test bearings The superfinish did not generate enough heat or plastic deformation near the surface to create microstructure alterations The most significant difference between the ground bearings and the hard turned bearings after superfinishing was the residualstress profile I 0; Ground and Superfinished Hard Turned and Superfinished Figure 1b: Effect of finish processes on fatigue life(large bearings) Figure compares the residual stress profile in the ground and then superfinished surface to that in the hard turned and superfinished surface These profiles were taken in the circumferential direction of the bearing race The profiles in the axial direction had similar characteristics to that in the circumferentialdirection After grinding and superfinishing a high compressive residual stress was generated near the surface, but diminished quickly as the distance from the machined surface increased On the other hand, after hard turning and superfinishing a residual stress reached deeper in the machined surface Residual stress profiles taken after only hard turning also showed deep compressive residual stresses The following superfinishingmerely changed the residual stresses near the surface Considering the difference between the plastic deformation taken place in the ground surface and that in the hard turned surface, it is reasonable to expect that the compressive residual stress generated by hard turning is deeper h m n -200 w= -400 L K -600 I I +Hard ((I O ln -800 I ~ -1000 I Turned & I Superfinished I +Ground& I Superfinished I I 0.02 0.04 0.06 0.08 0.1 0.12 Depth below Surface(mrn) Figure : Residual stress profiles , Ground and Superfinished Hard Turned and Superfinished Figure 1a : Effect of finish process on fatigue life(smal1 bearing) 60 There was no significant difference between ground and hard turned surfaces other than the depth of compressive residual stresses Also, the high compressive residual stresses on the surface not seem to affect the fatigue life This is consistent with the result obtain by Scott et a1.[6] for grinding, mentioned previously The examination of microstructures of the fatigue tested inner rings revealedthat the "butterflies"in the hard turned rinqs were located twice as deep as those in the ground rings Butterflies are the microstructure changes created by a severe local plastic deformation around a microcrack during fatigue tests A clear relationshipwas found in the post test analysis between the depth at which butterflies were found and the fatigue life, namely the deeper the butterflies, the longer the life From these evidences, it was concluded that the deep compressive residual stress created by hard turning is beneficial MAINTAIN DEEP COMPRESSIVE RESIDUAL STRESSES 4.1 Turning with sharp tool The above findings led this investigation to find a way to maintain the deep compressive residual stress penetration In order to reduce experimental factors, first a sharp tool without a chamfer or an edge hone was used for hard turning to generate residual stress The rake angle was degrees A cutting condition was selected to create an accurate geometry profile for a long duration Figure shows the stress profile created by a sharp tool The surface residual stress in the circumferential direction was tensile and that in the axial direction was compressive but low There was no microstructure change in the machined surface Therefore, the tensile residual stress on the surface was created mechanically and not thermally The compressive residual stresses are still laying deeper than those of ground surfaces 800 I 4.2 Effect of depth of cut and feed rate As an attempt to increase the penetration depth of residual stresses, depth of cut and feed rate were varied The ranges of variables were chosen so that the tool life and the geometry accuracy for precision hard turning could be maintained Figure shows the effect of depth of cut on the residual stress in the circumferential direction No significant difference can be observed in this depth of cut range Figure also shows the effect of feed rate on the residual stress in the circumferential direction The residual stresses near the surface shifted towards tension as feed rate was increased However, in a depth below 0.04mm, there was no change in the residual stress profiles z 300 $ : a = 0.025 400 n 0.051 200 100 -400 0.102 -I I 0.02 0.04 0.06 0.08 0.1 0.12 Distance From Surface(mm) Figure 5: Effect of depth of cut 800 600 400 -800 200 0.02 0.04 0.06 0.08 0.1 Depth below Surface(mm) 0.12 -200 Figure : Residual stress generated by a sharp tool -400 -600 Now superfinishing was applied to the hard turned surface created by a sharp tool Superfinishing removed approximately microns of the surface layer The residual stresses near the surface turned to more compressive as shown in Figure Superfinishing insures the removal of tensile residual stresses near the machined surface 0.12 Figure 6: Effect of feed rate 4.3 , 0.02 0.04 0.06 0.08 0.1 Distance from surface(mm) Effect of tool edge geometry As the depth of cut and feed rate did not significantly affect the residual stresses in the deep subsurface, the size of a primary cutting zone seemed to be of secondary importance On the other hand, the size of the plastic deformation taken place near the tool edge was thought to play a major role In order to increasethe plastic deformation size a tool edge honing of 0.02mm was applied to a sharp tool Figure shows the residual stress profiles in the machined surface created by a sharp tool and a 0.02mm honed tool The honing dearly creates a deeper compressive residual stress Most of the commercially available CBN tools are sold with a minor honing of less than 0.013mm Therefore, compressive residual stresses are usually created on the surface Further attempts to find the effect of the honing size was not successful since honing could not reliably repeat its 61 size This is due to the lack of control in the material removal of honing As an alternative to honing, a secondary chamfer of 30 degree and 03mm long was added to a major chamfer of 15 degree and 12mm long on a tool edge as shown in Figure 8.The angle of the secondary chamfer to the rake face is 45" It was kept very short as it can increase thrust force As this chamfer can be ground on the tool edge, its size control was easier than honing This double chamfer tool was tried again against a sharp tool Figure 9.Shows the comparison of residual stress profiles With the double chamfer tool, the compressive residual stress again was larger and deeper than that created by a sharp tool Either by tool honing or by adding a secondary chamfer, this investigation revealed the dominant effect of tool edge geometry on residual stress magnitude and depth Further study on the double chamfer size effect is being conducted , 200 I h m z - g! -200 c fn -400 '5-600 w + -800 02 mm Honing - 0.02 0.04 0.06 0.08 0.1 0.12 Distance from surface(mm) Figure 7:Tool edge honing effect 30" Figure 8: Double chamfer geometry(mm) 0.02 0.04 0.06 0.08 0.1 0.12 astancehwnsurface(m) Figure 9: Effect of double chamfer 62 CONCLUSION Tests were conducted to find the effect of precision hard turning on rolling contact fatigue life The differences in the surface integrity between hard turning and grinding were examined The causes for creating deep residual stresses by hard turning were analyzed Special tools were preparedto successfully create residual stresses in deep subsurfaces The investigationshowed that: Hard turned and superfinishedbearings have at least as long fatigue life as ground and supertinished bearings In one test, it was better than ground and superfinished bearings The depth of compressive residual stresses is the major difference between hard turned and ground surfaces Depth of cut does not produce a significant effect on residual stress for precision hard turning The primary deformation zone has a secondary effect on residual stresses Feed rate only changes residual stress near the surface but not in a deep layer Tool edge geometry is the dominant factor determining the residual stress profile ACKNOWLEDGEMENTS We would like to thank Sumitomo Electric Co., specially Mr Fukaya Mr Matsumoto and Dr Nakai for advising the tool design and providing CBN tools for the experiments We also appreciate the assistance of many staff members in the Timken Company; especially Mr D Gang for Xray residual stress work and Mr T Glaser for conducting hard turning Finally, We are grateful for the support of this investigation by the Timken company REFERENCES Koenig, W, Berktold A., and Koch, K F., 1993, Turning versus Grinding, A Comparison of Surface Integrity Aspects and Attainable Accuracies, Annals of the %IRP 4211;39-43 Field, M and Kahles, J F 1971,Review of Surface Integrity of Machined Components, Annals of the CIRP, 20/2;153-163 Tarasov, L P., Hyler, W S and Letner, H R., 1958, Effect of Grinding Direction and of Abrasive Tumbling on the Fatigue Limint of Hardened Steel, Proceedings of American society of Testing Materials, 57:601-622 Matsumoto, Y Magda, D., Hoeppner D W and Kim, T Y 1991,Effect of Machining Processes on the Fatigue Strength of Hardened AlSl4340 Steel, Journal of Engineeringfor Industry,Transaction of the ASME, 113.154-159 Townsend, D P and Zaretsky E V., 1988,Effect of Shot Peening on Surface Fatigue Life of Carburized and Hardened AlSl 9310 Spur Gears, SAE Technical Paper Series 881291 Scott, R L., Kepple, R K and Miller, M H., 1962, The Effect of Processing-induced near-surface Residual Stress on Ball Bearing Fatigue, Proceedings of a Symposium on Rolling Contact Phenomena, Elsevier PublishingCo 301-316