Application of solid modeling techniques for geometric simulation of surface topography in boring operations Authors: Mohammad Mehrabinasab, Behnam Moetakef-imani, Mohsen Fallah Published: 2026 ABSTRACT AbstractThe quality of the machined surface and the dynamic stability of the cutting process are directly correlated to the dynamic cutting force, which itself depends on the geometry of engagement between the cutting tool and workpiece that defines the dynamic chip thickness. The surface topography is one of the most important figures of merit for the evaluation of performance in machining processes, which is influenced by the geometry of the cutting edge, the kinematics of the cutting operation, the flexibility of the machine tool structure, and the resulting structural vibrations during the chip formation process. In order to define the instantaneous engagement between the cutting tool and workpiece precisely, all these factors should be taken into account. In this paper, the mechanics, dynamics, and geometry of boring operations are considered for the development of a virtual simulation model by using the solid modeling techniques. The dynamic parameters of the boring bar are defined by modal analysis experiments, and the cutting force coefficients are experimentally identified by conducting mechanistic cutting tests. The experimental cutting tests are conducted in both absolutely stable and unstable cutting conditions. In order to validate the developed model, firstly, the simulated cutting forces are compared with the corresponding experimental results. Secondly, the simulated topography of the machined surface in both stable and unstable cutting conditions is compared with the SEM images from real cut surfaces. The presented geometric simulation model shows a remarkable potential for exact simulation of boring operations in stable and unstable conditions.Keywords: Geometric simulation, Boring bar, ACIS solid modeler, Surface topography, ChatterSubject terms: Engineering, Materials science, Mathematics and computing Introduction The term “virtual machining” means to use digital computers for the purpose of simulating the geometry, mechanics, and dynamics of machining processes, in order to improve the performance and stability of cutting operations during the production of industrial products. Various researchers have presented machining simulation models that predict the behavior of the cutting process in different conditions. The accurate estimation of dynamic cutting force is one of the main objectives in these models, which helps to optimize the performance of cutting processes. The cutting force can be computed by analytical, numerical, or mechanistic approaches1. The mechanistic force model has remarkable advantages over other methods for modeling the mechanics of cutting processes in various cutting operations with different cutting edge geometries2. The cutting coefficients and the dynamic chip thickness are the two main variables that should be properly computed for the estimation of dynamic cutting forces. The cutting force in machining processes should be appropriately calculated to study the regenerative chatter phenomenon and its impact on the topography of the cut surface. The dynamic cutting force is influenced by not only the cutting parameters such as spindle speed, feed rate, depth of cut, and geometric parameters of the cutting edge, but also it highly depends on the cutter workpiece engagements (CWE), which is affected by the structural response of the cutting tool. In fact, the CWE engagement is directly influenced by the structural vibrations imposed on the tool’s tip. The instantaneous engagement geometry will change due to the cutting vibrations, which change the instantaneous dynamic chip thickness accordingly. Various geometric simulation models are proposed for the exact computation of CWE, which can be classified as solid modeling approaches, discrete methods, and analytical approaches. The solid modeling method was first developed almost six decades ago, when the CAD/CAM systems started to represent models with geometric dimensions3. Today, the acceptable solid modeling approaches are constructive solid geometry (CGS)45 and boundary representative (B-rep)6. In the B-rep method, the vertices, edges, and faces are used in order to describe the boundaries of solid geometries7, in which various parametric representations such as Bezier, B-spline, and NURBS can be used in order to model the curves, surfaces, and volumes. It uses parametric representation and topological information to determine the connectivity and adjacency between the geometric elements8. The B-rep technique provides an exact and continuous model for the cutter swept volume during the CWE, but computing the intersections between surfaces is rather time-consuming9. The idea behind the CGS method is to combine simple geometric volumes, such as a cube or a sphere, by using Boolean operations, in order to provide a simple representation of more complex solid geometries. This method was used for the computation of CWE and the simulation of cutting forces in machining processes45. However, the B-rep method has better capabilities for modeling of CWE in multi-axis machining processes of complex sculptured surfaces, and most of the recent research in this field was conducted by using the B-rep method. By integrating cutting edge geometry and trochoidal motion within the ACIS environment, researchers were able to simulate semi-roughing operations6. Another study concluded that theoretical surface roughness models could be successfully validated against real micro-milling experiments, proving that solid modeling is a robust tool for predicting finish quality10. Further developments expanded these capabilities to 2.5-axis milling and utilized Boolean operators to update workpiece volumes dynamically.11. There are some other studyies in litrature1213, have used a B-rep-based model to compute the chip volume in other machining processes. Further research have used the triangular mesh model as a simpler B-rep based method to model the geometry of machining processes1415.The disadvantage was computation of intersections that required the updating of a NURBS-based B-rep model, which was computationally demanding. In another study16, the application of B-rep solid modeling was suggested without continuous updating of the workpiece surface, but it was only performed once after the completion of each tool path.While these investigations confirmed the high geometric fidelity of B-rep models, they also highlighted a persistent challenge: the heavy computational load required to calculate surface intersections in real-time. Consequently, discrete methods like Z-mapping and Z-buffer became the preferred choice for defining engagement in sculptured surface machining due to their speed. Z-map is the most adopted approach in which the workpiece boundary is shown with the intersection points of the Z direction vectors (ZDV) with the upper surface of the part17. But, only one intersection between the model and ZDV is allowed. In order to solve this issue, later research allowed the storage of multiple intersections and gaps between the intersection points. This method was suitable for shapes that do not have overhangs. The dexcel representation was based on the Z-map approach that was specifically developed for workpieces with overhang18. Nie and Feng introduce Adaptive voxel-update schemes, which are a uniform voxel framework with a direct voxel-tracing algorithm19. In order to generate a “ functional cutter surface,” they trace the tool geometry, then, perform Boolean subtraction. The advantage is avoiding scanning the whole workpiece, and the CWE map is calculated by voxelizing cross-sectional circles of the cutter. The discrete methods have lower accuracy compared to solid modeling techniques due to the rasterization problem, which diminishes by sharpening the resolution at the cost of higher computational cost. In order to overcome this issue, different analytical approaches were suggested by the researchers. By using the B-spline theory20, the geometry of cutting edge was modeled in boring operations as well as the kinematics and dynamics of the cutting tool. They used a geometric algorithm for exact computation of the boundaries of the removed chip, the contact length of the cutting edge as well as the dynamic chip load distribution on the cutting edge.The advantages of the method was considering the tool tip radious in computing the CWE engagement. Recent research combined analytic geometry and discrete models to achieve high accuracy and efficiency8.Nie and Zhao suggest a tri-level grid (tri-dexel) method21. In fact, the discrete voxel output is optimized with analytic envelop computation, reaching high precision. With the advent of 4D geometric models, researchers represent the continuous evolution of the tool and workpiece in space-time. Zhang et al.22 to compute the instantaneous engagement and perform Boolean set operations in 4D construct two 4D meshes: one for the time-dependent simulation (WOR) and one for the tool’s swept volume (TOR) in 4D. Building these models is computationally heavy;This model is able to capture CWE with a continuous spatio-temporal representation. In the machining era, CWE modeling in turning and boring is simpler than milling because the engagement area remains continuous. By contrast, in milling due to rotating multi-edge cutters and complex multi-axis toolpath calculating, CWE are far more challenging. Prior researchers have studied the engagement geometry and tool vibration in boring operations separately. This study bridges the gap, calculating the CWE by considering machining vibration in a solid modeling environment. Exact simulation of the cut surface in machining operations is a very complicated task, especially in unstable cutting conditions, in which the surface topography is affected by the vibration marks left on the cut surface during the previous revolutions of the spindle. In this article, a virtual simulation model of the boring process is developed within the ACIS solid modeling environment to predict surface topography in both stable and unstable cutting conditions Henceforth, the article is organized as follows: First, the dynamic parameters of the boring bar and the mechanistic cutting force coefficients are experimentally identified through modal analysis and mechanistic cutting tests. Next, the kinematics of the boring operation and the exact geometry of the cutting edge are modeled using a boundary representation (B-rep) approach. The model is then extended to include the dynamics of the process by solving the equations of motion for the flexible tool in the radial direction and adding the resulting instantaneous displacements to the kinematic motion. The proposed model is validated by comparing simulated cutting forces and acceleration signals with experimental results in stable and unstable conditions. Finally, the simulated topography of the machined surface is compared with SEM images of real cut surfaces to evaluate the model’s accuracy in predicting feed marks and chatter grooves. Experimental setup The cutting tests are conducted on a general-purpose lathe machine tool. The boring bar is the most flexible component in the machine tool structure. It is highly susceptible to cutting vibrations in the radial direction. The dynamics of boring bar are directly affected by the length-to-diameter ratio (LD) of the cutting tool. In order to measure the tool’s vibrations, IEPE accelerometer sensors are installed near the cutting head. The experimental setup for identification of cutting force coefficients is shown in Fig. 1. The Kistler 9255-B dynamometer is used as the clamping support of the boring bar, in order to measure the cutting forces in three mutually perpendicular directions. The experimental setup for modal analysis tests is also illustrated in Fig. 2. The experimental vibration data was gathered using Tenlee IEPE accelerometers mounted at the tool tip. The primary sensor used was the TL122A100 model, which provides a frequency range of 0.5 to 10 kHz and a shock limit of ±2000g peak. These sensors facilitate measurements in both single-axis and triaxial configurations. For the models employed in this study, the acceleration measurement range is rated at 100g, ensuring high-fidelity signal capture within a linear response zone below 5 kHz. The experimental setup for identification of cutting force coefficients. The experimental setup for modal tests. Parameters of boring operations The chip formation process in boring operations is a three-dimensional, geometrically intricate phenomenon that is affected by various parameters. For instance, it is influenced by geometric parameters (such as tool working angles and cutting edge radius), mechanical parameters (the cutting coefficient of workpiece material), dynamical parameters (the model parameters of the cutting tool), and the cutting parameters (cutting speed, cutting depth, and feed rate). The influential parameters of the boring process are briefly introduced in this section. Geometric parameters of boring bar The cutting head of boring bar is selected from Sandvik Coromant products as shown in Fig. 3 left. The tool cutting edge angle is 91 degrees. The cutting insert is TNMG 160408-PM 432523. The geometric parameters of insert are inscribed circle diameter (IC), cutting edge effective length (LE), corner radius (RE), and insert thickness (S), which are 9.525 mm, 15.6978 mm, 0.7938 mm, and 4.763 mm, respectively Fig. 3 right. The workpiece is made of 6063 aluminum alloy, with an outer diameter of 120 mm, an inner diameter of 92 mm (following preparation), and a length of 100 mm Fig. 4. (Left) Cutting head of the boring bar24, (Right) Generic representation of insert. Workpiece during the process. Mechanical parameters The mechanistic cutting force model can be described by the following equations25: , , and are the cutting coefficients, and , , and are the edge coefficients of the force model.The terms b and h represent the width of cut and the uncut chip thickness, respectively. In order to extract the cutting/edge coefficients for boring of Aluminum alloy 6063-T6, 18 cutting experiments are conducted with a spindle speed of 500 rpm (cutting speed is 144.51 m/min), depth of cut 1-3 millimeters, and feed rates 0.05-0.32 millimeters per revolution. The average value of cutting force is computed for each of the force’s components, and the variation of average force is plotted versus feed rate as illustrated in Fig. 5. The cutting and edge force coefficients are obtained by linear regression analysis, and are reported in Table 1. The resultant cutting force coefficient is 964 MPa, as reported in previous research26. Identification of cutting and edge force coefficients by using the mechanistic cutting force modeling approach. Table 1The cutting and edge force coefficients for boring of Aluminum 6063-T6.Force directionEdge coefficient (N/m)Cutting coefficient (N/m)Depth of cut (X)5.15215.26Feed (Y)56.14408.25Cutting speed (Z)37.38846.68 The cutting and edge force coefficients for boring of Aluminum 6063-T6. Dynamic parameters of boring bar Experimental modal analysis is a well-adopted approach for obtaining the response of flexible structures in the frequency domain. The impact applied by the instrumented hammer excites a broad range of the system’s frequencies, and the frequency response function (FRF) of the system is computed by dividing the discrete fourier transform (DFT) of the output signal (displacement, velocity, or acceleration) by the DFT of the input force signal. The modal instruments used in this study are shown in Fig. 3. The modal tests are conducted for two different LD of the steel boring bar. The experimental and analytical FRFs of boring bar in the radial direction for LD 4 and 8 are illustrated in Fig. 6 and Fig. 7, respectively. The dynamics of the boring bar is only studied in the radial direction, as the radial deflections have a dominant effect on the dynamic chip thickness and dimensional accuracy compared to the tangential direction27. The peak picking method is used to extract the modal parameters of boring bar structure around the dominant fundamental frequency. The obtained parameters are reported in Table 2. The differences between the experimental data and the fitted FRF are not significant, particularly considering that the FRF amplitude is plotted on a dB scale, which visually amplifies the discrepancies between the two curves compared with a linear scale representation. Nevertheless, as shown in Figs. 6 and 7, the experimental data and the fitted FRF exhibit very good agreement near the fundamental frequency, where the two curves almost completely overlap. Therefore, the modal parameters extracted from the fitted curves accurately represent the dynamic behavior of the boring bar in the vicinity of the fundamental frequency. According to the reported values, the natural frequency and damping of boring bar structure with LD 4 are much higher than the boring bar with LD 8. In fact, in real cutting tests, it is observed that the cutting process is highly stable with the cutting tool that has the LD 4. But, due to a lack of dynamic stiffness, the cutting process is unstable in almost every cutting condition by using the boring bar with LD 8. The experimental and analytical FRFs of boring bar in radial direction for LD 4. The experimental and analytical FRFs of boring bar in radial direction for LD 8. Table 2Modal parameters of boring bar for the first mode of vibration.ParameterUnitLD 4LD 8Natural frequency[Hz]669.5192.5Modal mass[kg]2.453.6Modal stiffness[N/m]434448055269191Modal damping[Ns/m]578.479.2 Modal parameters of boring bar for the first mode of vibration. Cutting conditions The cutting conditions for the final experiments used in this study are given in Table 3. Test A is performed by a stiff boring bar with LD 4, and represents the stable cutting conditions, while Test B is conducted by a flexible boring bar with LD 8, which represents the unstable cutting conditions. Table 3Cutting conditions.ParameterUnitTest ATest BDepth of cut[mm]1 and 22Feed rate[mm/rev]0.160.16Spindle speed[rpm]500500Cutting speed[m/min]144.51144.51LD–48 Cutting conditions. Geometric simulation The boring operation is simulated inside the environment of ACIS. All geometric dimensions, kinematic parameters, and dynamic factors are defined in the programming module of ACIS. Examples of geometric parameters include workpiece dimensions, cutting edge geometry, and tool path. The kinematics of motion is defined by feed rate, depth of cut, and spindle speed. The dynamic parameters include the natural frequency, modal damping, and static stiffness for the fundamental mode of the boring bar. Kinematics of cutting process In this section, the kinematics and geometry of the cutting process are used to formulate the motion of the cutting edge and present the surface generation algorithm for boring operations. The geometry of the cutting edge, including the cutting edge radius, side cutting edge angle () and back cutting edge angle (), is clearly depicted in Fig. 8. In boring operations, the cutting edge constantly removes chips from the workpiece surface. The spindle rotation is required for the continuation of the chip formation process, while the cutting tool moves into the workpiece hole. The kinematics of boring operation directly affect the geometry, accuracy, and quality of the machined surface topography. The profile of cutting edge at initial condition (-) and ready for cutting operation () by considering the machining angles. The kinematics of boring operation are implemented based on the theory of homogeneous transformation matrix (HTM). The incremental movements of the cutting edge and workpiece can be represented by the combination of individual motions, i.e., rotation around the workpiece axis and translation along the tangential direction. The profile of cutting edge is transformed by linear transformation matrices in order to generate the envelope of the machined surface. In addition to the kinematics of motion, the radial vibrations of cutting tool are also considered in the proposed model. Two different coordinate systems are defined in order to describe the surface generation mechanism in boring operations, which are shown in Fig. 9. Coordinates system required for kinematic simulation. The Global coordinate system (CGS) is a fixed coordinate system located at the center of the workpiece, which is used to store all the geometries at their initial position inside the ACIS environment. It is represented by XYZ. The tool center coordinate system (TCCS) is located at the center of tool’s tip corner as shown in Fig. 9. The coordinate system is shown by at the instant of time. The axis is in the direction of workpiece radius, and the axis is in the direction of feed motion (or the workpiece center axis). The tool path of boring operation has a helical geometry as illustrated in Fig. 10.. The vertical distance between two adjacent tool paths is defined by the value of feed rate. Illustration of the translational and rotational motions of the tool’s tip along the tool path. The position of TCCS is updated with respect to the GCS, such that the new cutting edge can be modeled inside the updated TCCS. Thus, a matrix transformation between TCCS and GCS is required. During the generation of the cut surface, the previous location of cutting edge is mapped to the present location by rotating the cutting edge around the workpiece axis and translating the cutting edge along the tangential direction. According to Fig. 10, the cutting edge is firstly rotated by an increment of degrees around axis, and is then translated by an increment of along the axis. The rotation operator is defined by: represents the incremental rotation of the cutting edge around the Z axis of the TCCS. The translation operator is defined as follows: is the incremental translation of the cutting edge along the Y axis of TCCS. To define the next position of cutting edge from its previous position with one transformation, the resultant transformation matrix is obtained by: In this equation, represents the resultant transformation matrix mapping the cutting edge position between two subsequent time steps, where and denote the incremental rotation operator around the axis and the incremental translation operator along the axis of the tool center coordinate system (TCCS), respectively. The cutting edge can be also be analytically mapped from its initial location to the location corresponding to the instant of time by: As shown in Eq. (7), represents the cumulative transformation matrix that maps the cutting edge from its initial location to the position at the instant by calculating the product of all incremental rotation and translation operators from the start of the simulation. Consider that the position of a point on the cutting edge is defined in the coordinate system by . Similarly, the position of the same point is defined in the coordinate system by . Under the following transformation, each point is transformed from its initial position to its final position: In Eq. (8), denotes the position vector of a point P on the cutting edge profile at the instant, obtained by applying the cumulative transformation matrix to its initial position vector . Dynamics of cutting process Regenerative chatter is an unwanted vibration phenomenon that occurs in most of metal cutting processes, and if not properly controlled, it will lead to deterioration of surface quality, dimensional accuracy, and generation of noise or damage to the machine tool structure. The regeneration of waviness and coupling of structural modes during the cutting process are the primary mechanisms for generation of chatter vibrations. It is a very complicated task to define the individual effect of each the two mechanisms on the process stability, as these two phenomena are inherently correlated with each other, especially in rough cutting conditions28. This paper only includes the regenerative chatter phenomenon by considering the dynamics of the cutting tool in the radial direction. The regeneration effect is due to the presence of phase shift between the subsequent vibrations. If the ratio of spindle frequency to chatter frequency is a constant integer, then chatter may not occur. The instantaneous depth of cut is defined by: where a(t) represents the instantaneous depth of cut, is the nominal depth of cut, and y(t) is the dynamic deflection of the cutting edge in the radial direction. The trajectory of cutting tool’s tip motion is shown in Fig. 11 by assuming that the chatter vibrations can be modeled by a sinusoidal wave. In general, the chatter vibrations reach a fully developed stage by passing through a transient phase in which the cutting tool’s structure constantly absorbs energy from the chip formation process, and the amplitude of tool’s tip oscillations start to grow out of bounds. Schematic trajectory of the cutting edge during regenerative chatter. The relation between spindle speed and chatter frequency is defined by29: is the chatter frequency (Hz) and n is the spindle speed (rev/min). N defines the integer number of vibration waves and defines a portion of the vibration wave. The most intensive chatter vibrations occur when the regenerative phase shift is equal to 180 degrees. By computing the ratio of chatter frequency to spindle speed in Eq. 10, it is observed that complete waves and a fraction of the 22th wave are formed during each revolution of the spindle. The phase shift is therefore 216 degrees, which means that after five revolutions, the phase difference between the first wave and the sixth wave becomes zero. This defines that the chatter groove is periodically repeated after five revolutions of the spindle. The five succeeding vibration waves corresponding to five subsequent revolutions of the spindle are shown in Fig. 12. Five succession vibration waves with phase shift of 216 degrees. ACIS simulation environment 3D ACIS® Modeler provides a powerful environment for modeling of complicated geometries. ACIS is a geometric molder developed based upon the B-rep technique with Open C++ object oriented architecture. It can be used to model complex analytical geometries as well as modeling highly intricate free-form curves and surfaces by using the theories of B-spline and NURBS. This software is utilized by a verity of software developers in a diverse range of engineering software packages including computer-aided-design and manufacturing (CAD/CAM), computer-aided-engineering (CAE), architecture engineering18. In this paper, ACIS environment is used for the purpose of simulating the boring process with a geometric approach in order to extract the exact geometry of the machined surface topography. The “scheme” language is used in order to perform the mathematical computations on arrays, lists and vectors as well as creating analytical geometries in the ACIS environment. It allows for the development of a variety of platforms through the development of script programs, optimization compilers, graphic packages and virtual reality systems as well as supporting the object-oriented programming approach11. Simulation of boring process in ACIS environment The kinematics of boring bar is included in the first simulation stage and the geometry of machined surface is generated without including the dynamic vibrations at tool’s tip. In this stage, only the geometry of cutting edge and workpiece is included as well as the kinematic parameters of the cutting process. The boring simulation is executed by using the appropriate coordinate transformations. In the next stage, the instantaneous dynamic vibrations of the boring bar during the chip formation process is calculated and also added to the geometric simulation algorithm in order to model the surface topography in the presence of dynamic vibrations Kinematic simulation in scheme The basic idea of geometric simulation is based upon the fact that the cutting edge sweeps a volume during its kinematic motion relative to the workpiece. The surface of workpiece is formed by the subtraction of this swept volume form the workpiece volume. The engagement volume also defines the removed chip geometry. Thus, the cutting edge is incrementally rotated and translated with respect to the workpiece at each time step. The swept volume at each increment is defined by specifying the instantaneous sweep direction. During each increment, the center of the cutting edge is transformed by a fraction of the feed. The position of the cutting edge also transforms due to the incremental rotation of cutting edge relative to workpiece. In order to detect the cutting edge sweep volume, the algorithmic module developed in Section "Geometric parameters of boring bar" is used. The removed chip is modeled by knowing the two subsequent positions of the cutting edge. At each position, the vectors tangent to the trajectory of cutting edge should be properly modeled. Two subsequent position of the cutting edge profile are shown as and in Fig. 13. The tangent direction vectors along the spiral tool path of the cutting edge in boring operation are shown in Fig. Fig. 14. The tangent direction vector is obtained by a 90 degree-rotation of the vector that connects the center of the workpiece to the instantaneous center of the cutting edge. The tangent and normal direction vectors are depicted in Fig. 14 by blue and red, respectively. The swept volume of each time increment is therefore defined by the geometry of cutting edge profiles, and , at the beginning and end of the time increment as well as the tangent direction vectors, and , which defines the starting and ending slope of the sweep path. The swept volume is the subtracted from the workpiece volume in order to update the surface geometry as shown in Fig. 15. The algorithmic flow chart for the geometric simulation of the boring process by including the kinematics of process is illustrated in Fig. 16. Illustration of incremental swept volume due to the kinematic motion of cutting edge relative to workpiece. Illustration of the tangent direction vectors during the kinematic motion of the cutting edge along the helical tool path. Incremental updating of workpiece surface during the cutting process. The algorithmic flow chart for simulation of surface topography by considering the kinematic motion of the cutting tool relative to the workpiece. Dynamic simulation in scheme Realistic simulation of dynamic cutting forces in machining processes has a crucial effect on the dynamic behavior of the system. The magnitude of the dynamic cutting force at each time step is directly related to the instantaneous chip area removed. The exact computation of instantaneous chip area is therefore one of the most important steps during the geometric simulation. In contrast to the static component of the cutting force, the magnitude of dynamic cutting force is time varying due to the oscillations in the dynamic chip thickness, which is generally computed by: where is the resultant cutting coefficient of the workpiece material, obtained from the mechanistic approach, is the approximation of chip area that is obtained by multiplying the width of cut with the dynamic chip thickness . The value of dynamic chip thickness is influenced by the regeneration of waviness mechanism25, which depends on the present and past vibrations of the cutting tool. In this paper, the instantaneous chip area is directly extracted from the removed chip geometry and is used for dynamic simulation of cutting force in boring operations. So the cutting force value is estimated based upon a solid modeling approach by using the following formula: The instantaneous vibration of the cutting edge in the radial direction is obtained by solving the equations presented in29. The equation of motion for the boring bar with a single degree of freedom in the radial direction is defined by: is defined as the instantaneous radial force, and n refers to the discrete time step of the simulation. The radial acceleration at each time step is therefore calculated as: The radial velocity and displacement are then computed through numerical integration using the Euler method: The schematic representation of the dynamic simulation module is shown in Fig. 17, which is implemented in the scheme programming environment. The schematic representation of the dynamic simulation module that includes the regeneration mechanism. It should be noted that the planar trajectory of the tool’s tip is circular, as shown in Fig. 14. However, with the introduction of dynamic deflections in the radial direction, the trajectory path deviates from the circular path, as shown in Fig. 18. As a result, the tangent direction vectors of the sweep volume should also be modified according to the instantaneous deflection of the tool’s tip. The instantaneous swept volume is generated by the modified direction vectors, similar to the kinematic simulation stage, and is then subtracted from the workpiece volume. The algorithmic flow chart for the geometric simulation of the boring process, including the dynamics of the process, is illustrated in Fig. 19. The instantaneous directions of the vectors tangent to the trajectory of the tool’s tip that is used as the sweep direction for the cutting edge’s sweep volume. The algorithmic flow chart for simulation of surface topography by considering the dynamic motion of the cutting tool relative to the workpiece. In the conventional time-domain approach, the dynamic chip thickness is estimated via the following time-dependent equation [25]: In this expression, h(t) represents the instantaneous chip thickness at time t, which is determined by the nominal mean chip thickness and the difference between the tool’s radial position in the previous revolution and its current radial position x(t). In the numerical simulation model, the dynamic chip thickness is computed by considering the current location of the cutting edge as well as three previous positions of the cutting edge, as shown in Fig. 20, using Eq. (17). In Eq. (17), the dynamic chip thickness is calculated by considering the difference between the current radial position of the tool X(N) and the minimum position recorded from previous revolutions , where f represents the resulting cutting force calculated using the identified cutting () and edge () coefficients . Computation of dynamic chip thickness from the current and previous positions of the cutting edge. The geometric simulation model proposed in this study includes the geometry, mechanics, kinematics, and dynamics of cutting process. Instead of estimating the dynamics chip thickness from the present and previous vibrations of the cutting edge by using approximate mathematical formulations, the instantaneous chip area is exactly extracted from the model. The approximate and realistic geometry of the chip area is shown in Fig. 21. The geometric simulation is able to model the boring operation with high accuracy, especially in unstable conditions where the deflections of tool’s tip are relatively high, as it can compute the exact geometry of chip area by considering the cutting edge radius into account. The realistic and approximate geometry of the chip area with and without considering the cutting edge radius. Results Two cutting tests are conducted by the boring bar with LD 4 (Test A) and LD 8 (Test B), in order to investigate the dynamic behavior of cutting process in stable and unstable conditions. Test A The simulations are conducted with the parameters given in Table 4; the mean values of radial force magnitude and chip area are compared for the Simulink and ACIS simulations in the stable cutting tests with depth of cut a=1 mm and a=2 mm. The variation of dynamic cutting force and instantaneous chip area are compared for both models during the first three revolutions of the spindle, as shown in Figs. 22 and 23. There is a good agreement between the results obtained from the Simulink model and the ACIS model. This comparison shows that in the stable conditions, the geometric model is strongly able to present the results of the time domain model in the Simulink environment. Table 4Cutting conditions of Test A.ParameterUnitTest A-1Test A-2Depth of cut[mm]12Feed rate[mm/rev]0.160.16Spindle Speed[rpm]500500Cutting Speed[m/min]144.51144.51 Cutting conditions of Test A. The variation of instantaneous chip area during the first revolutions of the spindle for (a) Test A-1 and (b) Test A-2. The variation of dynamic cutting force during the first revolutions of the spindle for (a) Test A-1 and (b) Test A-2. As illustrated in Fig. 24 and Fig. 25, the Discrete Fourier Transform (DFT) plot of the cutting acceleration signal is compared for the ACIS model versus the experimental results measured from tests 1 and 2, respectively. Since the cutting process is stable, the dominant natural frequency of the boring bar is slightly excited during the cutting operation. The DFT plots are reasonably similar in terms of the dominant frequency and maximum amplitude of vibrations. In addition, the time-domain signature of radial acceleration obtained from ACIS simulation is compared with the experimental test 2 with depth of cut a=2 mm, as illustrated for Fig. 26. The comparison of the DFT plots for the radial acceleration signal obtained from Test A-1 (a) ACIS simulation and (b) Experiment. The comparison of the DFT plots for the radial acceleration signal obtained from Test A-2 (a) ACIS simulation and (b) Experiment. The comparison of the radial acceleration signals obtained from Test A-2 (a) ACIS simulation and (b) Experiment. The machined surface topography obtained from the experimental test and ACIS simulations for depth of cut 2 mm are shown in Figs. 27 and 28, respectively. The feed marks are the primary feature of the cut surface in the stable cutting test. According to the selected feed rate, the distance between the succeeding feed marks is 0.16 mm, as shown on Fig. 27. There is a reasonable agreement between the topography of the simulated and experimental surfaces. The topography of machined surface for depth of cut 1 mm is similar to the depth of cut 2 mm. The experimental surface topography for boring operation (Test A-2) with (a) 110x zoom and (b) 500x zoom. The simulated surface topography for boring operations (Test A-2). The comparison between SEM images and the ACIS simulated results Figs. 27 and 28 confirms the validity of the ACIS simulation model. To add more confidence, the surface roughness of these experiments are also studied. Table 5 shows the surface roughness values in terms of Ra parameter. The ACIS simulations predict the surface roughness of tests (with depth of cut 1 and 2 mm) to be and respectively. The experimental results reported in the table, shows that the measured and predicted surface roughness are in good agreement, which confirms the validity of the ACIS simulation and theoretical model. Table 5Surface roughness Ra.Depth of cut (mm)Ra value ACISRa value ExperimentsTest A-112 m2.23 mTest A-225.1 m4.89 m Surface roughness Ra. Test B The simulations are conducted with the parameters given in Table 6. The simulated acceleration is compared with the experimental result as depicted in Fig. 29. The proposed geometric simulation model can successfully estimate the dynamics of boring bar in the unstable cutting conditions. The delay in the measured signal’s growth can be attributed to the physical time required for the regenerative effect to build up sufficient energy in the actual machine-tool system, whereas the simulation responds more rapidly. The eventual surpassing of the simulated signal by the measured one highlights the non-linear nature of real-world chatter. As the vibration amplitude increases, factors such as tool-flank interference (process damping) and non-linearities in the cutting force–which are simplified in the linear model–can lead to a more rapid escalation of instability in the experimental setup. Table 6Cutting conditions of Test B.ParameterUnitTest BDepth of cut[mm]2Feed rate[mm/rev]0.16Spindle Speed[rpm]500Cutting Speed[m/min]144.51LD–8 Cutting conditions of Test B. Comparison of the experimental and simulated acceleration in Test B. The simulated and experimental surface topography are also compared in Fig. 30 for the unstable cutting Test B, which was conducted by a boring bar with LD of 8. As previously discussed, after five subsequent revolutions of the spindle a chatter groove is formed that includes five succeeding feed marks. The simulated surface topography is in good agreement with the experimental topography, which shows the remarkable capability of the developed geometric simulation model. By computing the ratio of chatter frequency to spindle speed in Eq. (10), it is observed that N = 21 complete waves and a fraction of the 22th wave are formed during each revolution of the spindle. The phase shift is therefore 216 degree, which means that after five revolutions, the phase difference between the first wave and the sixth wave becomes zero. This describes that the chatter groove is periodically repeated after five revolutions of the spindle. The surface topography: simulation (a – two chatter grooves) versus experiment (b – one chatter groove). The dynamic stiffness of the boring bar greatly reduces in the radial direction with an increase of LD from 4 to 8. The topography of machined surface radically changes due to the transition of cutting dynamics from stable state to unstable one. The geometry of chatter grooves is directly correlated with the chatter frequency, spindle speed and regenerative phase shift. By using the radial deflections of cutting tool as well as the cutting geometry and mechanics, the presented model can simulate the topography of cut surface is both stable and unstable cutting conditions. Conclusions The topography of machined surface in boring operations is affected by the kinematics, geometry and dynamics of the cutting process. In this article, a geometric simulation model is developed in the ACIS solid modeler environment for the prediction of surface topography in stable and unstable cutting conditions. The solid modeling approach is the more exact and precise method for computation of the CWE in machining processes. The proposed model includes the dynamics of boring bar into the geometric simulation model as opposed to the previous studies. The topography of machined surface in stable cutting conditions is characterized by the feed marks which are generated by the tip of cutting edge with finite radius. On the other hand, the dynamics of boring bar in mutual interaction with the dynamics of cutting process plays a primary role on the topography of the machined surface in unstable cutting conditions. The chatter grooves or channels are thus the main feature of cut surface in unstable cutting conditions as shown by the results of this research. A comparison between Figs. 30 and 27 indicates that feed marks constitute the dominant surface feature in Fig. 27, whereas surface tearing becomes the dominant characteristic in Fig. 30. The tearing marks are localized and clearly observable at the bottom of the grooves. Fig. 29 corresponds to the conditions presented in Fig. 26. As shown in Fig. 26, under stable cutting conditions, the vibrations induced by the cutting process exhibit a damped response. In contrast, under unstable cutting conditions, the acceleration amplitude remains nearly constant and undamped due to the regenerative chatter mechanism. In the stable regime, an impulsive vibration with a relatively large amplitude occurs once per spindle revolution, which subsequently decays over time. However, in the unstable regime, the tool acceleration amplitude does not decay and instead maintains a significant and nearly constant magnitude. The proposed model can reasonably predict the surface topography of workpiece in both stable and unstable cutting conditions. Because it includes the exact geometry of cutting edge, as well as the modal parameters of the boring bar related to its fundamental mode and the cutting force coefficients of the workpiece material. The presented dynamic geometric simulation model can be extended to other cutting processes as well in order to provide a deeper understanding of the impact of cutting mechanics and dynamics on the topography of the cut surface, especially in unstable cutting conditions. The spindle frequency in this study is equal to 8.33 Hz which is far from the fundamental structural frequency of the boring bar with LD4 (in stable cutting condition with 600 Hz) as well as boring bar with LD8 (in unstable cutting condition with 200 Hz). The root cause of sever vibrations with LD8 is the regenerative chatter mechanism, which is a self-excited vibration phenomenon rather than forced structural vibration phenomenon, caused by matching of the spindle frequency with the fundamental structural resonance of the boring bar. In boring, the tool is a cantilever beam. Its stiffness decreases with the cube of the length (L3). Even a slightly flexible workpiece is often 10–100 times stiffer than a boring bar with LD8. If the workpiece frequency is much higher than the tool frequency, it doesn’t participate significantly in the vibration. The Kistler dynamometer used in our setup provides a very stiff and stable base, supporting the “rigid clamping” assumption for this specific experiment. While this study focuses on the radial vibrations of the flexible boring bar, future developments could incorporate the structural dynamics of the workpiece and clamping system. In cases with low workpiece stiffness, the interaction between the natural frequencies of the tool and the part would further influence the global dynamic behavior and the resulting cutter-workpiece engagement. Author contributions Mehrabinasab and Fallah conducted the experiments and wrote the main manuscript text. Mehrabinasab performed the simulation work. Professor Moetakef-Imani conceived the main idea and managed the project. All authors reviewed and approved the final manuscript. Funding This project is financially supported by Ferdowsi University of Mashhad (research and technology grant ID: 3/40663). Data availability The datasets generated during and analysed during the current study are available from the corresponding author on reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally to this work: Behnam Moetakef-imani and Mohsen Fallah. References Associated Data Data Availability Statement The datasets generated during and analysed during the current study are available from the corresponding author on reasonable request.