The Effect of Tool Geometry and the Cutting Force System in Turning
 
 

D. A. Gustafson | UIUC | MS | 1990

ABSTRACT:

In this thesis, a model for the turning process has been presented which describes the force systems which arise for sharp tools and for tools on which flank wear has evolved. To be consistent with previous metal cutting research, sharp tool force modeling is performed in the external coordinate system. Force model refinements focus on improving the prediction of the direction of the friction force in the plane normal to the cutting velocity, defined by the effective lead angle, and the manner in which the effect of tool geometry is captured in the model coefficients, Kt, and Kf. The results of the sharp tool force modeling in this thesis may be summarized as follows:

1. Stabler's Chip Flow Rule has been extended to a three-dimensional tool with a finite rose radius. This method implies that the friction force at any point on the cutting edge acts in the direction given by the local inclination angle. Summation of the discrete friction forces along the cutting edge yields the effective lead direction for a given tool geometry and set of cutting conditions.

2. A numerical approach has been taken to calculate the chip flow direction, effective reke angle, chip thickness, and chip area at discrete points along the cutting edge. accordingly, the average chip thickness for a given tool geometry and set of cutting conditions is defined as the average of the local chip thickness measured in the local chip flow direction. By searching normal to the cutting edge, the chip cross-sectional area in the plane normal to the cutting velocity may be obtained.

3. The term used to describe the effect of tool insert geometry in the models for the coefficients, Kt and Kf, is the unweightd effective rake angle, alpha (sub eu). This formulation improves the predictive ability of the model for kf, resulting in more accurate radial and longitudinal force predictions.

4. Cutting tests were performed across a range of cutting conditions to verify the accuracy of the sharp tool force model. The average error in the force predictions for these sixteen verification tests was 4.0% for tangential force, 7.2% for the radial force, and 8.0% for the longitudinal force.

The refined turning force model for sharp tools has been augmented by a model which predicts the force system arising on a tool with flank wear. The inputs required for the wear model are the tool geometry, cutting conditions, workpiece material hardness, and the wear land width. Wear forces are attributed to interference between the tool flank and the workpiece; they have been assumed to be proportional to the area of flank-workpiece contact and the workpiece material hardness. The wear land has been modeled as a plane surface parallel to the cutting velocity. wear forces acting on the tool flank are assumed to not affect the manner in which cutting takes place in the shear zone.The wear forces are modeled in the external coordinate system and may be added to the predicted forces for a sharp tool to obtain worn tool force predictions.

The following results were obtained during the validation of the flank wear force model:

1. Based on the results of the three tool life tests which were performed using stainless steel workpieces, tool life test results have been shown to be repeatable.

2. An additional tool life test run on cast iron using a coated carbide insert validated the assumptions made in the wear force model regarding wear geometry and the rate at which forces increase with flank wear. Significantly, the effective lead angle was found to be unaffected by the presence of flank wear on the cutting insert. The flank-workpiece coefficient of friction was lower than the 0.3 value assumed initially, resulting in overprediction of the tangential wear force. Assuming a friction coefficient of 0.2 produced accurate tangential wear force predictions (13% error) over the range of wear land width observed during this test.

3. Cutting tests using inserts with wear lands precision ground onto them were performed to verify the relationshipbetween wear geometry and worn tool cutting forces. Total force predictions contained 22% and 33% error in the longitudinal and radial directions, although the tangential and resultant force predictions were excellent (8% and 6% error, respectively). The prediction error associated with the longitudinal and radial forces was primarily due to error in the wear force contribution to the total force predictions. This error may have been a result of the discontinuous profile of the ground cutting edge, which affected the resultant direction of the longitudinal and radial wear forces. In addition, the transition point on the nose where the wear pattern ends may have caused a dynamic element to be present in the measured force signal, particularly in the radial direction.

The proposed turning force model predicts the increase in cutting forces for a given wear land width. These force predictions could then be used to predict system deflections, surface error, and power consumption. Alternatively, the amount of flank wear on a tool may be inferred from the increase in the cutting forces that is measured over a tool's life. Force data may, therefore, be analyzed on-line to detect tool condition. If the wear model is used as a component in a feed control algorithm, then corrective action may be taken to avoid potentially serious consequences.

If you have any comments or suggestions please e-mail jwsuther@mtu.edu.