Professional Documents
Culture Documents
CIRP - Paper - Simulation-Based Twist Drill Design and Geometry Optimization
CIRP - Paper - Simulation-Based Twist Drill Design and Geometry Optimization
Methodology
For the geometry model (Fig. 2), drill margin, body diameter and the
transition between them are modeled as radiuses. The flute from
For
, nodal displacements at margin in two reference planes
located with distance to loads and constraints are evaluated (Fig.
5).
is calculated from nodal displacement difference
,
margin radius
and reference plane distance
.
,
,
and
max
cutting edge is split into primary and secondary cutting edge. The
secondary also incorporates the indentation zone around the dead
center. The large variation of cutting conditions requires a
discretization of cutting edges into ECTs (Fig. 6). For this modeling
technique it is assumed, that all forces acting on the cutting edge
can be expressed by two components, friction force
in the rake
face plane and normal force normal to the rake face.
Fig. 7. Iterative best-fit of the maximum inscribed ellipse to chip flute cross-section
&
The coefficients
,,
and
,,
are specific for tool and
workpiece material and determined from calibration experiments
with multivariate regression. For the calibration, spot drilling
experiments are used with different cutting speeds and feed rates
and full factorial design. When the drill engages the workpiece, the
width of cut continuously increases and increments from thrust and
torque signal are assigned to each ECT.
and
are calculated
for each ECT by transformation. In a reverse direction,
and
can be calculated for arbitrary drill geometries from geometric data
and coefficients
,,
and
, , . Verification experiments
show that prediction accuracy is increased if two separate models
for primary and secondary cutting edge are used. Calibration
experiments were carried out with six different twist drill geometries.
The variance of the coefficients , ,
and , ,
from these
experiments was very low. This qualifies the cutting force model for
prediction of and
of new twist drill geometries. A weak point is
the lack of a reliable cutting force model for thrust force from the
indentation zone. Currently, this force is included in the secondary
cutting egde force model. A significant extension of the indentation
zone would result in thrust force estimation errors. To prevent this,
a separate constraint function limits the indentation zone size.
3.3 Chip Evacuation Capability
An appropriate chip evacuation capability (CEC) is essential for
twist drills. CEC is considered as good if chip removal does not lead
to a significant torque increase while drilling depth increases. In
practice, multiple parameters are known to influence CEC.
Systematic investigations are difficult because of parameter
interactions. Some parameters, such as exact chip forming, are still
not predictable with adequate accuracy. A simple method for CEC
prediction is the evaluation of flute cross-sectional area [8, 9, 10,
11]. Hands-on experience does acknowledge its relevance on CEC.
However, this criterion neglects the influence of the flute shape. A
more sophisticated method is the calculation of the maximum
inscribed ellipse size in the flute cross-section [14]. An ellipse rather
than a circle is used as spiral chip representation because of the
drill twist. The cross-sectional shape of spiral chips is roughly
circular in a projection normal to chip flow direction. Due to the helix
angle of chip flow, it becomes an ellipse in a drill cross-sectional
view (Fig. 7).
In this assumption, the ellipses minor axis is aligned with the drill
center point and contacts the hole wall at one and the flute at two
points. Both criteria seem eligible for computer-aided geometry
optimization and do well in fist trials. However, they do only a very
inaccurate and abstract CEC estimation [11]. Experimental
investigations have been carried out to find a universal relation of
CEC and a model-based criterion. Even though the general
tendency that very small flutes are prone to chip clogging can be
acknowledged, no general relation between the flute size and CEC
could be found. As a result, the flute enlargement through reverse
web taper has the most significant effect on CEC. Spiral chips,
formed in a flute region close to the drill point are in a state of radial
compression. A comparing of the spiral chip diameter with the bestfit ellipse diameter shows this phenomenon. A flute enlargement
reduces mechanical transmission due to reduced friction between
chips and the hole wall. Thus, it reduces torque increase. Even
though also chip formation and chip breakage influence chip
clogging, the chip clogging tendency is noticeably reduced. For an
improved CEC criterion, chip expansibility due to flute enlargement
is taken into account. In experiments short breaking spiral chips
from thick web twist drills deviate from a circular shape. A better
chip shape prediction was attained with a new chip forming model
(Fig. 8). The actual chip shape depends on chip flute shape in chip
flow direction. It is partially a copy of the chip flute curvature but has
also smaller radiuses.
Fig. 10. Coolant channel outlet gap, shape calculation and simplified model
, ,
refer to the flow speeds before intake, in the channel and
after outlet (Fig. 9). Because
, it is considered that
0.
For the flow speed after blow-off, a reference distance of 10 d is
used. Here
equals the (negligible) atmospheric pressure and
0.625 [16].
is the total pressure drop.
0.3164
0.6:
2:
0.5
1052.9
51.986
1,408.5
706.56
404.36
1,240.3
422.66
121
64
121
64
is
Fig. 11. Actual outlet gap width calculation and linear averaging
0
0
1
0.3164
0
0
1
121
64
value. Drilling torque and thrust are computed from geometry data
using a semi-empiric cutting force model. For the prognosis of
coolant supply, the coolant flow resistance of drill is calculated.
Twist drill geometry solutions are generated and optimized
automatically by application of heuristic algorithms, such as Genetic
Algorithm, MOGA and NSGA-II. The validity of the drill geometry
solution is verified by constraint functions including a chip flute
grindability verification. This methodology gives the tool designer
extensive computer assistance. However, the automatic geometry
optimization has limitations that require further research work.
Fig. 19. Two-stage meander-shaped search path for valid grinding parameters
References
[1]
Chen, W.-C., Fuh K.-H., Wu C.-F., Chang, B.-R., 1996, Design optimization of
a split-point drill by force analysis, ASME J. Mat. Proc. Tech., 58:314-322.
Paul, A., Kapoor, S.G., DeVor, R.E., 2005, Chisel edge and cutting lip shape
optimization for improved twist drill point design, ASME Int. J. Mach. Tool.
Manuf., 45:421-431.
Paul, A., Kapoor, S.G., DeVor, R.E., 2005, A Chisel Edge Model for Arbitrary
Drill Point Geometry, ASME J. Manuf. Sci. eng., 127:23-32.
Audy, J., 2008, A study of computer-assisted analysis of effects of drill
geometry and surface coating on forces and power in drilling, ASME J. Mat.
Proc. Tech., 204:130-138.
Treanor, G.M., Hinds B.K., 2000, Analysis of stresses in micro-drills using the
finite element method, Int. J. Mach. Tool. Manuf., 40:1443-1456.
Jang, D.Y., Kuo, C.-H., 1997, Geometric Parameter Optimization for Optimum
Stress in a Twist Drill using FEM, DOE, and Genetic Algorithm, Society of
Manufacturing Engineers, MR97-182.
Chen, Y.R., Ni, J., 1999, Analysis and Optimization of Drill Cross-Sectional
Geometry, Society of Manufacturing Engineers, MR99-162.
Schulz, H., Emrich, A.K., 2000, Optimization of the Chip flute of Drilling tools
using the Principe of Genetic algorithms, Intelligent Computation in Manuf.
Engineering, Vol. 2:371-376.
Schulz, H., Emrich, A.K., 2001, Using the principle of Genetic Algorithms for
the Optimization of the Chip Flute of Drilling Tools, Production Engineering,
Vol. VIII/2:107-110.
Abele, E., Schulz, H., Stroh, C., Emrich, A.K., 2002, Optimization of the Chip
flute of drilling tools with Methods of genetic algorithms, Int. Sci. Conf. Prod.
Eng. 8, Brijuni, pp. III:057-068
Emrich, A.K., 2001, Optimierung des Spanraumes bei Bohrwerkzeugen mit
Hilfe Genetischer Algorithmen, Dissertation Technische Universitt Darmstadt.
Goldberg, D.E., 1998, Genetic Algorithms in Search, Optimization and
Machine Learning, Addison-Wesley Publishing Company.
Chandrasekharan, V., 1996, A model to predict the three-dimensional cutting
force system for drilling with arbitrary point geometry, PhD Thesis, University
of Illinois at Urbana-Champaign.
Leadbetter, R.E., 1993, The influence of twist drill design on rigidity and chip
disposal, Thesis (MPhil), Aston University, Birmingham.
Blasius, H., 1908, Grenzschichten in Flssigkeiten mit kleiner Reibung,
Zeitschrift fr Mathematik und Physik, Vol. 56.
Rinaldi, P., 2003, ber das Verhalten turbulenter Freistrahlen in begrenzten
Rumen, Dissertation, Technischen Universitt Mnchen.
Dubbel Taschenbuch fr den Maschinenbauer, 20. Ausgabe 2007, Springer
Verlag.
Spurk, J.H., Nuri, A., Strmungslehre - Einfhrung in die Theorie der
Strmungen, 6. Auflage 2006, Springer Verlag.
Lin, C., Kang, S.K., Ehmann, K.F., 1996, A CAD approach to helical groove
machining - Part 1: Mathematical model and model solution, ASME Int. J.
Mach. Tool. Manuf., 36:141-153.
Lin, C., Kang, S.K., Ehmann, K.F., 1997, A CAD approach to helical groove
machining - Part 2: Numerical evaluation and sensitivity analysis, ASME Int. J.
Mach. Tool. Manuf., 37:101-117.
Hsieh, J.-F., 2005, Mathematical model for helical drill point, ASME Int. J.
Mach. Tool. Manuf., 45:967-977.
Poritsky, H., Dudley, D.W., 1943, On cutting and hobbing gears and worms,
ASME J. Appl. Mech., pp. 139-146.
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
Fig. 20. Optimization progress and exemplary fitness comparison of drill solutions
[14]
[15]
[19]
[16]
[17]
[18]
[20]
[21]
[22]