Available online at www.sciencedirect.

com

ScienceDirect
Procedia CIRP 14 (2014) 581 – 586

6th CIRP International Conference on High Performance Cutting, HPC2014

A comparative study of CNC part programming addressing energy

consumption and productivity
Gustavo M. Minquiza,*, Vicente Borjaa, Marcelo López-Parraa,
Alejandro C. Ramírez-Reivicha, Miguel A. Domínguezb, Alejandro Alcaideb
a
Faculty of Engineering, Universidad Nacional Autónoma de México, Av. Universidad 3000, México, D.F., 04510, México
b
Instituto Tecnológico de Puebla. Av. Tecnológico num. 420 Col. Maravillas Puebla Pue. CP 72220, México
b
Universidad Tecnológica de Puebla. Antiguo Camino a la Resurrección 1002 - A, Zona Industrial, 72300 Puebla, México

* Corresponding author. Tel.: 52 55 56 22 30 04; fax: +0-000-000-0000. E-mail address: [email protected]

Abstract

This paper reports on a comparative study realized to contrast between the part code generated by two different CAM systems and one part
program generated by the authors. This comparison addresses in particular tool path, but also includes number of lines, cycle time and energy
consumption. The comparison is carried out through a case study: a mechanical component was designed so as to produce a number of CNC
part programs generated with two different CAM systems; one more program was directly programmed by the authors in the machine tool´s
controller. Different sample parts were produced using the same machine tool, tooling and process parameters; and measurements of energy
consumption and cycle time were gathered. The results show differences between the tool paths generated by each approach and are used by the
authors to further define strategies to produce trajectories that could reduce cycle time and energy consumption.
© 2014 Published by Elsevier B.V. Open access under CC BY-NC-ND license.
© 2014 The Authors. Published by Elsevier B.V.
Selection and peer-review
Selection and peer-reviewunder
underresponsibility
responsibility of the
of the International
International Scientific
Scientific Committee
Committee of theof
6ththe 6th International
CIRP CIRP International Conference
Conference on High
on High Performance
Performance Cutting. Cutting

"Keywords: Trajectories, Energy consumption, CNC machines, Computer aided manufacturing."

tools to find out the time required to generate the NC program

1. Introduction
for a particular part. The study revealed that NX 7.5 required
11.43 minutes, followed by PRO/EW5 with 15.11 minutes and
Manufacturers of machined parts use computer aided
Work NC with 19.48 minutes. Soori et al. [3] used Catia in
manufacture to increase their productivity; CAM systems help
order to find the relation between geometric errors and their
them to achieve high precision tolerance levels. CAM
correlation with accuracy and repeatability of complex
software developers and engineers use different strategies and
manufactured workpieces.
methods to generate CNC part programs; the particular
Daneshmand, et al. [4] investigated the time required by
differences, advantages and disadvantages of these approaches
different tool trajectories i.e. zig-zag and helices employing
are discussed in this article. Kannan et al. [1], in 2008, studied
vertical and spherical cutters; to machine a part. The
several CAM software and proposed criteria to select design
trajectories were generated by Master CAM V9 and Catia
programs; the data base produced by the authors takes into
V5R18. In this study, Catia required less time for vertical
account functionality, performance, efficiency,
cutter’s back and forth movements, and for spherical cutter’s
communication, support, price and some other software
helical trajectories. Daneshmand et al. [4] also discovered that
distinctive features. Kannan’s study ranks as most
Master CAM was faster when using a zig-zag trajectory with
advantageous software Catia, PRO-E, Inventor and AutoCAD.
vertical and spherical cutters and constant spiral trajectories.
In 2012 Michalik et al. [2] carried out a study on 3 software

2212-8271 © 2014 Published by Elsevier B.V. Open access under CC BY-NC-ND license.
Selection and peer-review under responsibility of the International Scientific Committee of the 6th CIRP International Conference
on High Performance Cutting
doi:10.1016/j.procir.2014.03.009
582 Gustavo M. Minquiz et al. / Procedia CIRP 14 (2014) 581 – 586

The paper presented herein reports an experiment aimed at Sandvik ® coated carbide cutting tool with 4 flutes model
investigating the behavior of cycle time, trajectories and RA216.24-3250BAK24P 1620 was selected. This tool can be
energy consumption when using different methods to generate used for rough and finish machining in dry and wet
CNC code. Machining variables are introduced first outside operations. The basic parameters to produce the part were
the process and then inside the process as described by Borgia initially specified considering this type of cutting tool, Table
et al. [5]. The paper compares the outcomes of part programs 1.
generated by two commercial software tools, i.e. Catia and
Master CAM, and a program generated by the authors. Table 1. Machining parameters for cutting tool RA216.24-3250BAK24P1620

Nomenclature Characteristics Parameter

Cutting Speed (m/min) 200
CAM Computer Aided Manufacturing Feed Rate (mm/min) 782
Depth of Cut (mm) 1.0
CAD Computer Aided Design
Spindle speed (RPM) 5013
NC Numerical Control
STEP Standard for the Exchange of Product model data
Other process parameters were calculated using
information provided by Sandvik ®, the tool supplier.
2. Methodology
2.3 Machining process design
Based on the work reported by Rangarajan and Dornfeld,
[6] which states that the workpiece orientation and the length As mentioned before, two commercial software, i.e. Catia
of the NC code segments influence the reduction of V5 R21 and Master CAM X7, were used to generate the NC
machining time, the experiment reported in this paper uses code for the case study part. A third part program was
three different options to generate NC part programs. generated by the authors directly on the controller of the
Six main activities were carried out during this experiment. machine tool used for this experiment.
These are presented below.
Table 2. Process parameters
2.1 Case study part design
Tool paths Strategy Parameters
A particular part was designed for the experiment. Its Facing - A Zig-Zag Across overlap Along Overlap Max. Step
general dimensions are 90 mm length, 75 mm width and 30 10% 105% Over -90%
mm thickness. The case study part’s geometry Fig. 1 required Profile - B Entry and Multi passes: Compensation Start and
different machining operations: Flat bottom holes, pockets, Exit: Tangent Rough type: Computer end contour:
contours, and a flat surface. The purpose of using different Extend
operations was to compare diverse strategies when generating Profile – C Entry and Multi passes: Compensation Start and
the NC part programs. The material selected for the part was Exit: Tangent Rough type: Computer end contour:
steel 1018. Perpendicular Extend
Hole Roughing Machining Entry motion Max. Step
D-E-F Parallel spiral Direction Climb Helix Over -90%
Hole Roughing Machining Entry motion Max. Step
D-E-F Constant Direction Climb Helix Over -90%
Overlap spiral
Pocket Roughing Machining Entry motion Max. Step
Circle - I Constant Direction Climb Helix Over -90%
Overlap spiral
Pocket Roughing Machining Entry motion Max. Step
Circle - J Parallel spiral Direction Climb Helix Over -90%
Pocket Roughing Machining Entry motion Max. Step
Square -K Constant Direction Climb Helix Over -90%
Fig.1. Case study part Overlap spiral
Pocket Roughing Machining Entry motion Max. Step
2.2 Cutting tool selection Square - L Parallel spiral Direction Climb Helix Over -90%
Pocket Roughing Machining Entry motion Max. Step
The energy required by a process using a cutting tool with Angle - M Zig Zag Direction Climb Helix Over -90%
4 flutes is less than the one required by the same process but Surface-N Zig Zag Max Step over Machining Tolerance
employing a 2 flute tool [7]. This is due to the fact that the 0.05mm Angle – 90° 0.005mm.
tool with more flutes requires less power. Based on this, a
 Gustavo M. Minquiz et al. / Procedia CIRP 14 (2014) 581 – 586 583

Table 2 presents the strategies and process parameters 2.6 Machined case study part
selected for the different tool paths required by the case study
part. These parameters were used to generate all of the NC Three part samples were machined using similar process
part programs. strategies and parameters. Samples were measured at the end
of the experiment to verify that final dimensions matched the
2.4 Machine tool selection original design.
The codes produced by the two software tools and the code
The case study part was manufactured in a HAAS – VF3 programmed by the authors were compared taking into
machining center. This machine has the following account the differences between the cycle time, energy
characteristics: X axis 1016 mm, Y axis 508 mm, Z axis 635 consumption and the number of lines of their corresponding
mm, spindle maximum speed 8100 RPM. The machine machining processes. Surface finish and tool life were
operates at 7.0 bar pressure and uses a Fanuc controller. considered out of the scope of this investigation. The authors’
approach studied only three output variables i.e. cycle time,
2.5 Machine center and measurement equipment wiring energy consumption and number of lines from the many
comprehended in a machining process, as suggested by Li and
Circutor ® brand name, Portable Network Analyzer AR6, Kara [8]. Yet, authors believe that researching these three
was used for the experiment, Table 3. variables helped in achieving some important energy
consumption conclusions. As mentioned above, for the three
Table 3. Characteristics of the Circutor ® Analyzer AR6 NC programs, the same type of cutting tool and machining
strategies were used.
Variable Resolution Accuracy
Current 0.1 mA ± 0.5% 3. Generation of the NC code
Voltage 0.01 V ± 0.5%
Power 0.1 W ± 0.5% In order to increase productivity, it is necessary to consider
the final surface quality of the desired part when designing the
Fig. 2 shows the wiring realized to carry out the machining process [9]. Productivity is associated to the use of
experiment. Fig. 2b illustrates the machining center, i.e. a the standard ISO 10303- AP238 [10], which is related to the
three-phase power Hass center. Fig. 2a shows the alligator data communication between CAM systems and NC
clips and current clamps, which are connected to Circutor ® machining centers. As stated by Garrido et al. [10] the STEP-
analyzer AR6 for data acquisition Fig. 2c. The measurement NC language provides high level information that facilitates
equipment took data every second, sending information to the the development of a range of large capacity machine
computer shown by Fig. 2d. controllers. As mentioned above, the experiment reported in
this paper studies the code generated by three different
sources to manufacture the same part.
a b To generate the part program by Master Cam X7, first the
case study part was modelled using a solid modeler and then
exported to IGES. The IGES file was imported to Master
Cam. The different machining strategies included in the CAM
software were used, the tool supplier’s process parameters
were employed and tool trajectories were planned including
the required air cut movements. This air cut movements were
considered since they contribute to energy consumption of the
machining process, as mentioned by Diaz et al. [11].
c d The strategies to generate tool paths offered by the
software were selected based on the least machining time
criterion. With this in mind, two strategies stand out: Parallel
Spiral and Constant Overlap Spiral. In the simulation of the
machining process of the case study component, the part
could be produced in 16.9526 minutes.
In the machining process modelled using Catia the
Prismatic Machining section was used. Strategies similar to
the ones employed with Master Cam were also tested; Inward
Fig. 2. (a) Clamp Voltage-Current; (b) Machine tool; (c) Circutor ®
Spiral Morphing and Outward Spiral Morphing were selected.
Analyzer; (d) Computer
The Multi Axis Sweeping tool was used for surface N Table
2, which exhibits similar characteristics to the ones of Master
Cam X7. The simulation of the machining process indicated a
total machining time of 20.5333 minutes.
584 Gustavo M. Minquiz et al. / Procedia CIRP 14 (2014) 581 – 586

The third NC program of the part was written by the Table 4 shows the cycle times and the number of code
authors on the machine tool controller, Fig. 3. Each part lines of the three part programs studied in this experiment.
trajectory was programmed using a drawing that includes all Many of the codes of the part program produced with Catia
the distances, radii and intersection points required by the NC were G1, so the generated trajectories are based on point
program. To generate a NC part program in the machine coordinates. There were differences between the number of
controller, the authors looked for the easiest way, as lines corresponding to cutting and surfacing operations in the
technicians normally do. Nested cycles with sub-routines that code generated by Master-Cam and by Catia. The last one had
include the M97 code were used. The most complex geometry more lines. The program produced by the authors had the least
was the surface N with an angular inclination, and in order to number of lines. This program also had the least number of
machine it, miscellaneous and preparatory codes were trajectory segments. The trajectory segments of a part
required. program determine the machining time, the more segments,
the more time [6].
The energy consumption of the machining process derived
from the three NC programs is presented in Table 5. The rows
of the table present each geometry feature of the case study
part, which are identified with letters as presented in Fig. 1
and Table 5. For example, facing operation is identified by the
letter A. As shown by Table 5, the total cycle time of the
program produced by Catia was 1.733 minutes, having an
energy consumption of 2.3 kW. The MasterCam program
produced a time of 4.017 minutes with a total energy
consumption of 2.3 kW. Finally, the code written by the
authors required 0.95 minute to produce the part with an
energy consumption of 2.1 kW.
Fig. 4 is based on Dahmus and Gutowski [12] to analyze
the power required by the machining center. The energy
consumed by the machine is divided into fixed, variable and
the one actually required to machine the part. This approach
easily identifies the energy required to machine the part. Fig.
4 shows that the energy consumption before machining is
86.95% of the total. Fixed energy is mainly consumed by the
panels, the lighting and computers. Variable energy is used by
the axes, the spindle and during tool changes. So, 13.04% of
the total energy used by the machine is employed to actually
machine the part and is determined by the cutting parameters,
Fig. 3. Machine tool controller interface the tool path and by the NC code to generate the part.

Each part program was tested in the simulation software of

the machine controller to verify their functionality and
trajectories. This allowed the detection of possible program Machining

mistakes and if extra G codes had to be added in the part

program.
86.95 %

Variable Power
4. Experimental results

The simulations carried out with the Catia and Master – Constant Power
Cam software, produced an accuracy of the 96% and 97%
respectively, when compared against the real machined cycle Fig. 4. Energy consumption breakdown of the machine tool used in the
time. experiment

Table 4. Results of the experiment Table 5 shows that longer times correspond to the main
part profile, section B Fig. 5, and to the surface, section N Fig.
Catia Master-Cam Made by Author 5. This table also shows that the average energy consumption
Real Cycle Time (min.) 20.7 16.9 11.3 is 2.2 kW, but it varies with the machining strategy, the depth
Simulated (min.) 20.3 16.5 ----------- of cut and the machining operation. This way, as
Cycle Time demonstrated by Daneshmand et al. [4] the machining
Line of the NC code 915 693 207 strategy affects the overall machining time.
 Gustavo M. Minquiz et al. / Procedia CIRP 14 (2014) 581 – 586 585

Guo et al. [9] showed that an increase of the depth of cut Fig. 5 presents one of the parts machined in the
influences the energy consumption. Based on this, a depth of experiment, showing the machining operations used to
cut of 2 mm was used in the experiment, to decrease the cycle produce it as indicated in Table 5. Table 5 presents that the
time and the energy consumption. This was tested in the tool path “D” to machine a hole Fig. 5, has a cycle time of
simulations and the relation stated by Guo et al. [9] was 0.050 min with Catia, 0.117 min. with Master Cam and 0.017
confirmed. min. with the code made by the authors.

Table 5. Cycle times and power consumptions measured in the experiment 25.0
20.7
Catia MasterCam Made by Author 20.0
16.9
Toolpath Cycle Power Cycle Power Cycle Power
Time Time Time 15.0

Minutes
(min.) (kW) (min.) (kW) (min.) (kW) 11.3

A 1.733 2.300 4.017 2.3 0.950 2.1 10.0

Idle 0.117 2.100 0.300 1.9 0.133 1.9
B 3.667 2.900 3.600 2.7 3.200 2.9 5.0
Idle 0.217 2.000 0.083 2.3 0.017 2.4
C 0.750 2.200 0.733 2.3 0.500 2.2 0.0
Idle 0.117 2.000 0.033 2.2 0.050 2.7 Catia MasterCam Made by Author

D 0.050 4.300 0.117 3.8 0.017 6.2

Fig. 6. Cycle times of the NC programs used in the experiment to machine the
Idle 0.033 2.000 0.050 2.2 0.017 2.7
case study part
E 0.050 4.300 0.050 3.8 0.017 6.6
Idle 0.033 2.000 0.050 2.2 0.017 3.9
F 0.050 4.300 0.050 3.8 0.017 5.4 3.5
3.2
Idle 0.033 2.000 0.033 2.1 0.017 2.7 3.0
G 0.050 4.300 0.067 3.7 0.017 6.5 2.5 2.5
Idle 0.033 2.000 0.033 2.1 0.017 3.8 2.5

H 0.050 4.300 0.050 3.9 0.017 5.2

KiloWatts

2.0
Idle 0.033 2.000 0.017 2 0.017 2.2
1.5
I 0.150 3.300 0.150 3.1 0.083 4.8
Idle 0.000 0.000 0.033 2.1 0.050 2.2 1.0
J 0.217 2.000 0.067 3 0.100 2.2
0.5
Idle 0.033 2.000 0.000 0 0.017 2.2
K 0.700 2.600 0.433 2.8 0.333 3.1 0.0
Idle 0.133 2.000 0.050 2.1 0.017 2.6 Catia MasterCam Made by Author
L 0.283 2.800 0.250 2.9 0.400 2.3
Fig. 7. Power consumption in the case study part
Idle 0.133 2.100 0.067 2.3 0.017 2.3
M 1.083 2.7 0.083 2.8 0.933 2.6
Idle 0.650 1.8 0.117 1.7 0.033 0.6 Table 5 also shows that the cycle time and the power
N 10.283 2 6.367 2.2 4.283 1.4 consumption required for each one of the tool paths generated
by the three programs used in the experiment are different.
The total cycle times and energy required to machine the case
study part with each NC program are presented in Fig. 6 and
Fig. 7 respectively. By analyzing the code of each one of the
part programs, the authors confirmed that the difference in the
cycle times is associated with strategies to generate “safe”
tool paths that avoid collisions [13]. This, approach generates
more code lines, Table 4, and therefore longer cycle times.
On the other hand, the authors did not produced code with
a “safe” strategy, when programming directly in the CNC
machine controller. This produced a part program with lesser
lines and therefore shorter cycle time.
In the same way, the differences on energy consumption
are also due to the additional lines produced when using
“safe” strategies to generate part programs. The code
Fig. 5. Machined part
generated by the authors is just based on the desired
geometry, so trajectories did not consider incremental depth
586 Gustavo M. Minquiz et al. / Procedia CIRP 14 (2014) 581 – 586

of cuts to avoid tool or workpiece damage as commercial experiments in its facilities. The research reported in this
software did. This, yet again, causes more code lines, Table 4. paper was sponsored by the Universidad Nacional Autónoma
As a result, the tool paths generated by the authors produced de México (UNAM-DGAPA-PAPIIT IT103012) and the
energy spikes that are reflected in Table 5, e.g. tool paths D, E Consejo Nacional de Ciencia y Tecnología (Beca 355272 y
and F. Proyecto SEP-CONACYT-Ciencia Básica No. 83239).

5. Conclusions References

The experiment reported in this paper showed that the two [1] Kannan, G. , Vinay, V.P. Multi-criteria decision making for the selection
CAM software used produced different NC part programs for of CAD/CAM system. Int J Interact Des Manuf 2008; 2: pp. 151–159.
[2] Michalik, P. , Zajac, J. , Duplák, J. , Pivovarnik, A. CAM Software
the case study part. The two commercial software generated Products for Creation of Programs for CNC Machining. Future
different process parameters and machining strategies. As Communication, Computing, Control and Management 2012; LNEE 141,
concluded by Díaz et al. [7] the energy consumption is pp. 421–425.
influenced by the cutting tool, its number of flutes and the [3] Soori, M. , Arezoo, B. , Habibi, M. Dimensional and geometrical errors of
feed rate. three-axis CNC milling machines in a virtual machining system.
Computer-Aided Design 2013; 45: 1306–1313.
As demonstrated by the experiment, the CAM software [4] Daneshmand, S. , Mirabdolhosayni, M. , Aghanajafi, C. Sifting Through
selection affects the cycle time to produce the part. The two the Optimal Strategies of Time-Based Tools Path Machining in Software
simulations of the software tools used in the experiment CAD-CAM, Middle-East Journal of Scientific Research 2013; 13 (7): pp.
correlate accurately with the real machining time of the case 844-849.
study part. The analysis of the part programs generated by the [5] Borgia, S. , Leonesio, M. , Pellegrinelli, S. , Valente, A. Energy driven
process planning and machine tool dynamic behavior assessment.
software tools indicated that the cycle time of a simple part Procedia CIRP 9 2013; pp. 91 – 96.
could be reduced by using nested cycles and reducing the [6] Rangarajan, A. , Dornfeld, D. Efficient Tool Paths and Part Orientation
lines of the part program via the use of loops. This reduces the for Face Milling, CIRP Annals - Manufacturing Technology 2004;
response time between the controller and the servomotors of Volume 53, Issue 1, pp. 73-76.
the machine tool. [7] Diaz, N. , Helu. M. , Jarvis, A. , Tönissen, S. , Dornfeld, D. , Schlosser, R.
Strategies for Minimum Energy Operation for Precision Machining,
To generate NC programs directly on the machine’s Project Report from the Dept of Mech Eng at University of California,
controller consumes much time due to the fact that it is Berkeley 2009.
necessary to calculate the required intersection points of the [8] Li, W. , Kara, S. An empirical model for predicting energy consumption
part’s geometries. Besides, to generate the code for a part with of manufacturing processes: a case of turning process. Proceedings of the
complex surfaces is not feasible [14]. Institution of Mechanical Engineers, Part B: J Eng Manufac 2011; 225:
1636.
The energy consumption registered in Table 5, shows that [9] Guo, Y. , Loenders, J. , Duflou, J. , Lauwers, B. Optimization of energy
a faster machining process was produced by the program consumption and surface quality in finish turning, Procedia CIRP 1 2012;
generated by the authors. During the analysis of the part pp. 551–556.
programs, it was found that the idle cuts defined by the CAM [10] Garrido, J. , Marín, R. , Ignacio, J. , Sáez, J. Extensión del estándar ISO
software tools consume time but make part programs safe and 10303 AP-238 para la automatización del acceso a datos de procesos
CNC, Revista Iberoamericana de Automática e Informatica Industrial
reliable. 2009; Vol. 6, Num. 1 pp.105-116.
Table 5 also shows that in the program made by the [11] Diaz, N. , Redelsheimer, E. , Dornfeld, D. Energy Consumption
authors, the cycle time of each machined hole, is considerably Characterization and Reduction Strategies for Milling Machine Tool Use,
low, but the energy consumption is higher, up to 50%, that the Proc of the 18th CIRP Intl Conf onLife Cycle Engineering 2011; 263-7.
one required by the operations produced by the software tools. [12] Dahmus, J. , Gutowski, T. An environmental analysis of machining.
Proc. IMECE2004.
At this point of the investigation, the authors consider that the [13] Senatore, J. , Segonds, S. , Rubio, W. , Dessein, G. Correlation between
above mentioned energy consumption could be caused by the machining direction, cutter geometry and step-over distance in 3-axis
higher number of lines of the part program generated by the milling: Application to milling by zones. Computer-Aided Design 44
commercial software. Further experimentation is being carried 2012; pp. 1151–1160.
out now in order to correlate these results with tool wear [14] López de Lacalle, L. , Lamikiz, A. , Sánchez, J. , Salgado, M. Toolpath
selection based on the minimum deflection cutting forces in the
measurements. programming of complex surfaces milling. International Journal of
The experiment also showed that programs with nested Machine Tools & Manufacture 47 2007; pp. 388–400.
cycles process information faster than the ones with more [15] Xu, X. , Nee, A. Advanced Design and Manufacturing Based on STEP.
code lines using G1 and G2 codes [15]. British Library Cataloguing in Publication Data. Springer-Verlag London
Limited 2009; pp. 80-90.
Acknowledgements

The authors thank the support provided by the company

Ramsa Maquinados, S.A. de C.V., to carry out the