Journal of Cleaner Production 41 (2013) 179e186

Contents lists available at SciVerse ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Modelling of direct energy requirements in mechanical machining processes

Vincent Aizebeoje Balogun*, Paul Tarisai Mativenga
School of Mechanical, Aerospace and Civil Engineering, The University of Manchester, Manchester M13 9PL, United Kingdom

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this research was to contribute towards the development of a new mathematical model and
Received 7 July 2012 logic for predicting direct electrical energy requirements in machining toolpaths. This model will track
Received in revised form the visibility and process dependence of energy and hence carbon footprint in machining process. This
24 September 2012
study includes a critical review of similar existing models and their limitations. The effect that machine
Accepted 12 October 2012
modules, auxiliary units and machine codes have on power and energy consumption during machining
Available online 22 October 2012
was studied and the electrical current consumption measured. A mathematical model for electrical
energy use in machining was developed addressing the limitations of existing models and validated on
Keywords:
Machine tools
a milling tool path. The paper provides valuable information on the impact of machine modules, spindles,
Energy models auxiliary units and motion states on the electrical energy demand budget for a machine tool resource.
Toolpaths This knowledge is fundamentally important in evaluating toolpaths and re-designing machine tools to
make them more energy efficient, to reduce electricity costs and associated carbon footprints.
Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction 1.1. Research aim and motivation

In the year 2010, 27% of electrical energy consumption in the At present the research community and industry do not have
United Kingdom (UK) was attributable to industrial sectors (DUKES, a robust method of calculating the energy that is going to be used in
2010). In most cases this electricity is predominantly generated machining a given product. Existing models are focussed on specific
through the use of fossil fuel. The use of carbon rich electricity energy or treating the machine tool as a black box. There is need to
generation sources is of critical global concern as these processes relate the machining numerical control (NC) commands to the
produce CO2 emissions. This implies that the higher the energy requirements in machining for process planners to be able
consumption of electricity in manufacturing industries, the higher to select minimum energy machining strategies. The motivation for
the carbon footprints left by such end products. As a result, the UK this work was to contribute towards an improvement in the
government and other worldwide leaders are making an increasing modelling capability for energy requirements in mechanical
demand for energy efficiency. machining and to develop a new model for explicit modelling of the
A machine tool plays a major role in manufacturing. The Euro- machine tool states, workpiece machinability and the impact of
pean Commission has cited it as being in a top three priority for cutting variables. It is essential to produce such a model and to raise
inclusion into the product categories regulated through the eco- the integrity of energy prediction models so that they can be used
design directive (EPTA, 2007). As such there is urgent need for in eco-friendly process planning to estimate the electricity
manufacturing sectors, particularly machine shops, to reduce footprint for machined products.
energy use per product manufactured, to help meet eco-design
directives and CO2 emission targets. Some relevant targets were
set by Kyoto protocol (1997) (Kyoto protocol, 2007). This calls for 1.2. Machine tool states and proposed improvements
machine tool designers to increase their understanding of energy
use by different design features of a machine tool. There is also The Cooperative Effort in Process Emission (CO2PE!) proposed
a need for the manufacturing industry to understand the impact of a unified taxonomy (Ostaeyen, 2010) and methodology (Kellens
machine tool motions and toolpaths on energy requirements for et al., 2012) so that in manufacturing, energy data collection can
machining. be standardised and presented in a globally compatible approach.
CO2PE! classified machine tool states into two categories: ‘Basic
State’ and ‘Cutting State.’ The states are based on operational
* Corresponding author. characteristics of the processes. In the ‘basic state’, electrical energy
E-mail address: [email protected] (V.A. Balogun). is needed to activate required machine components and ensure the

0959-6526/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jclepro.2012.10.015
180 V.A. Balogun, P.T. Mativenga / Journal of Cleaner Production 41 (2013) 179e186

Table 1 clearly shows that the machine tool is the dominant

consumer of electrical energy in machining compared to the actual
chip forming process. It is important to further understand how this
electrical energy requirement by the machine is distributed. This can
then help to develop more accurate models of energy demanded by
machine tools which could be used by machine tool designers and
process planners. A good basis for analysing energy use in machining
is through Gutowski et al. (2006) mathematical model for direct
energy requirement in machining as shown in Equation (1).

_
E ¼ ðP0 þ kvÞt (1)

Where, E is the direct energy in J or Ws required in machining

processes, P0 is the power in W, consumed by the machine before it
starts cutting, k is the specific energy requirement in Ws/mm3 for
machining a particular workpiece material, v_ is the material
removal rate in mm3/s, while t is the cutting time in s.
Considering both Equation (1) and Table 1, it can be noted that P0
dominates the direct energy requirement in machining. This
demonstrates how the selection of machine tools can have signif-
icant impact on direct energy requirement in machining. Equation
Fig. 1. Machine tool electrical energy consumption estimation model.
(1) also enables modelling of the machine tool energy requirements
to be done distinctly from the energy required for the chip
formation process. This mathematical model is very valuable for
operational readiness of the machine tool. In the ‘Cutting State’ the supporting process planning as it enables comparison and selection
energy is demanded at the tool tip to remove workpiece material as of machine tools and workpiece materials.
well as for modes of energy loss e.g. through machine noise, friction After this early work by the MIT group, a number of mathe-
etc. matical models for energy in machining have been proposed and
While COP2E! sets the framework, it does not clarify the exis- these are summarised in Table 2.
tence of a transitional state between the ‘Basic State’ and ‘Cutting From Table 2a, the modelling approach used by Mori et al. (2011)
State.’ In this paper the authors define a third and intermediate has some resemblance to Gutowski et al’s (2006) but Mori et al., split
state called the ‘Ready State’. This additional state is required to P0 into basic and idle power. This is in line with the proposal put
clarify the process that takes place after the machine is started. This forward by this paper in Fig. 1, assuming that the idle power can be
stage of the process requires more energy for the drives and spindle described as the ‘Ready State’ power. Calling it idle power suggests an
movement to bring the tool and workpiece to the correct, about to unnecessary step, while the term ready state clarifies the need to
cut position and to set-up the necessary cutting velocity. Examples bring the system to an about tool to engage position. In the current
for such activities could include ‘G00’, ‘S’ and ‘T’ (rapid, spindle paper, authors further propose, that the cutting power demand P3 in
speed and tool change machine features respectively). Fig. 1 shows the model developed by Mori et al. (2011), can be expanded to take
the extended machine tool electrical energy states proposed by the into account the specific cutting energy coefficient as introduced by
authors. Gutowski et al. (2006). Diaz et al. (2011) produced an interesting
In order to establish focus for electrical energy improvement, it approach acknowledging that during the machining process the tool
is important to understand how the electrical energy use or power engage and disengage with the workpiece and that the total cycle
demanded is distributed during machine use. A small number of time is not devoted to the actual material removal. Thus, modelling
researchers have explored the critical aspects of electrical energy the air cutting time reduces the chance of over estimation of the
use in machine tools. Some of these results are summarised in energy demand. Ultimately there is need to quantify the air cutting
Table 1. time for toolpath in order to estimate its impact on total energy

Table 1
Machine tool contribution to electrical energy demand.

Authors Observations
Kordonowy (2001) The constant energy on a Cincinnati Milacron 7VC Automated milling machine was 51.9% of the total energy
requirements in machining.
Dahmus and Gutowski (2004) and The energy consumption of machine tools during actual cutting processes was 85.2% for ‘Idle’ or machine
Gutowski et al. (2006) tool auxiliary function and 14.8% for ‘Cutting’.
Devoldere et al. (2007) The idle or stand-by mode consumed 1.7 kW energy on a five-axis milling machine and the biggest
consumer was the hydraulic pump, using nearly 0.9 kW of energy.
Diaz et al. (2009) Among the auxiliary units on a Mori Seiki NV1500DCG milling machine, the servo and the spindle consumed
the most energy in the basic and idle states.
Vijayaraghavan and Dornfeld (2010) For a Mori Seiki NV1500DCG milling machine, energy consumption was dominated by start-up and idle states.
Rajemi et al. (2010) In an MHP CNC Lathe during dry cutting, the machine module and idle power consumption was 61e69% of the
total power, in the cutting speed range of 500 to 300 m/min.
Anderberg et al. (2010) Reported that energy demand by machine tool auxiliary units dominated the cost components and CO2 footprint
of a manufactured product.
Avram and Xirouchakis (2011) Reported that the machine tool system consumed 45% and 55% of energy at low and high spindle speeds respectively
when in a non-cutting mode on executing a part program on a C.B. Ferrari A152 machining centre.
 V.A. Balogun, P.T. Mativenga / Journal of Cleaner Production 41 (2013) 179e186 181

Table 2
a: A summary of other models for direct energy requirements in machining. b: A summary of specific energy models in machining

Author(s) Direct energy in machining model

Mori et al. (2011)
E ¼ P1 ðT1 þ T2 Þ þ P2 ðT2 Þ þ P3 ðT3 Þ (2)
Where, E is the total direct energy requirements, P1, P2, P3 represent basic, idle and cutting power demand in
Whr and T1, T2, T3 are the corresponding time.
Diaz et al. (2011)
E ¼ Pavg *Dt ¼ ðPcut þ Pair Þ*Dt (3)
Where, E is the total direct energy requirements, Pavg is power demand, Dt processing time, Pcut is the cutting
power and Pair is air cutting.
He et al. (2012)
Etotal ¼ Espindle þ Efeed þ Etool þ Ecool þ Efix (4)
This expanded to
Ztms Ztcs m Z
X
tfe
 
Etotal ¼ Pm dt þ Pc dt þ Pi dt þ Ptool ttool þ Pcool ðtcoe  tcos Þ þ Pservo þ Pfan ðte  ts Þ (5)
tme tce i¼1 t
fs

Where, Etotal is the total direct energy requirements, Pm is the power for enabling the operating state of the spindle
transmission module, Pc is the power for material removal from the workpiece, tms and tme are respectively the starting
time and the ending time for spindle running, and tcs and tce are respectively the starting time and the ending time
for cutting, Pi, tfei, and tfsi are, respectively, the power, the starting time, and the ending time of the ith-axis feed motor,
Ptool is the power of the tool change motor, and ttool is the turret rotation time, Pcool is the power of the coolant pump
motors, and (tcoetcos) represents the running time of the coolant pump motor, Pservo and Pfan are the power of the
servos system and fan motors, respectively and (tets) denotes the running time of the machine tool throughout the
entire NC file.
Author(s) Energy in machining model
Diaz et al. (2011)
1
ecut ¼ k* þb (6)
MRR
Where, ecut is the specific cutting energy, k is a constant and has units of power and b represents the steady-state
specific energy.
Draganescu et al. (2003)
Pc
Ecs ¼ (7)
60hz
Where, Ecs is the specific energy consumption, Pc is the cutting power, h is machine tool efficiency and z is the
material removal rate.
Li and Kara (2011)
C1
SEC ¼ C0 þ (8)
MRR
Where SEC in kW/cm3 is the specific energy consumption; MRR is the material removal rate; C0, and C1 are
empirical coefficients and are not the same as the specific cutting energy and idle power because the empirical
approach considered the machine tool to be a single holistic system. Unfortunately this hides vital information
about the machine tool design and workpiece machinability.

demand. However, Diaz et al’s model was not focused on process axis have to be engaged to take the spindle from its current location
planning and does not explicitly model the impact of machine tools, on the workspace to the tool magazine pick up position and back
workpiece materials and cutting variables. which requires energy. Ultimately, the model does not incorporate
Unlike Gutowski et al. (2006), He et al. (2012) used cutting cutting conditions of cutting speed, feed and depth of cut which
forces instead of specific energy to model the energy required for prevents it from being an information source for process planning.
the chip formation process. The utility of using specific energy is There are also other approaches to modelling of energy
better because it is a simple concept to apply to a range of consumption in machining as shown in Table 2b. These are based
machining processes. It enables an assessment of the energy effi- on different methods used to measure efficiency. This can be total
ciency of materials based on their machinability. The limitation of energy normalised by the volume machined. While these may be
the model developed by He et al. is that modelling the fixed energy interesting as a benchmarking measure, Diaz et al. (2011);
simply based on power for servo drives does not present a complete Draganescu et al. (2003); and Li and Kara (2011) specific energy
picture as other equipment features are required to support the models do not directly give the energy footprint for a machined
machining process. For example this model ignores energy demand component nor do they render themselves supportive to process
for the computer used by the machine, the lights, lubrication of the planning which needs cutting speeds, feeds and depth of cut
machine, swarf conveyors and chillers. As such, this model is not information to be modelled explicitly.
generic for all machine tools and needs to be improved. The energy
used by machine equipment features in the unloaded state, need to 2. New improved model for direct electrical energy
be characterised according to the machine design and energy los- requirement in machining
ses. Additionally, He et al., 2012’s part of the energy model for tool
change does not consider the number of tool changes required to Considering Equation (1) and the new classification in Fig. 1, the
finish a machining job and the fact that turret indexing can be done model for direct energy requirements in machining can further be
using the shortest route. Additionally, on a milling machine, the improved into Equation (9a)
182 V.A. Balogun, P.T. Mativenga / Journal of Cleaner Production 41 (2013) 179e186

_ c
Et ¼ Pb tb þ ðPb þ Pr Þtr þ Pair tair þ ðPb þ Pr þ Pcool þ kvÞt (9a) Fig. 2 shows the distribution of power consumption of the
machine modules, auxiliary units and essential motions based on
Where, Et is the direct total energy requirement, Pb, Pr and Pcool in machine tool states of ‘Basic’ and ‘Ready’. The tip or cutting state is
W are the basic and ready state power (power increment above not shown since this is a study for a non-cutting operation. The
basic power to bring the machine to the about to cut position) and results show that the power demand of the ‘Basic states’ is 53%, 72%
coolant pumping power requirements respectively, tb and tr in and 63% for CNC MHP lathe, Takisawa CNC milling machine and
s are the basic and ready time respectively and k with units of Mikron HSM 400 High Speed Milling Centre respectively, of the
kJ/cm3 is the specific cutting energy which is closely related to the total power requirements for a machine operating at no cutting
workpiece machinability and the specifics of the cutting load. The ‘Ready states’ power budget is 47%, 28% and 37% respec-
mechanics; v_ in cm3/s is the rate of material processing; and tc is tively. This shows that the intermediate actions of getting the
the cutting time in s. Taking into account Diaz et al’s (2011) machine ready have a significant power demand though lower than
approach, Pair represents the average power requirements for start-up.
a non cutting approach and retract moves over the component Therefore, it is important that the power demand of the ‘Ready
and tair represents the total time duration in s of these non-cutting state’ is included in the estimation methodology of the total energy
moves. Obviously in machining the objective is to keep the non- demand for machine tool system as shown Equation (9). Hence, the
cutting time as short as possible in order to improve machine total energy demand of machine tool system could be estimated
actual cutting utilisation. using Equation (9).
Equation (9a) can further be re-organised into Equation (9b). The power requirements for individual aspects of the CNC MHP
 : lathe, MAC-V2 Takisawa Milling Machine, and Mikron HSM 400
Et ¼ Pb ðtb þ tr þ tc Þ þ Pr ðtr Þ þ Pair tair þ Pr þ Pcool þ kv tc (9b) high speed machining centre are shown in Figs. 3e5. For example
taking the MHP CNC lathe, it is noted that the machine start-up
3. Experimental investigation (24.04%), spindle running (24.43%), servo home location (16.27%),
fluid pumping (14.85%) and main switch (8.35%) dominate (>80%)
To validate the mathematical approach suggested by Equation the power demand of the ‘Basic’ and ‘Ready’ states of the machine
(9), cutting tests were done in milling to characterise energy tool under investigation. These are the key areas for improvement
requirements and further develop the model according to observed for eco-design of this type of machine tool for machine utilisation
electrical energy demand patterns. This was extended into the and optimisation.
application of the model for facing off a xey plane surface on In the cutting state, other auxiliary modules are activated
a component on vertical axis milling machines. The machines used which also consume electricity. This includes the tool change
were a MHP lathe, Takisawa CNC milling machine and a Mikron system, spindle speed acceleration or deceleration and coolant
HSM 400 High Speed Milling Centre. pumps.
A Fluke 345 Power Quality Clamp Meter was clamped on the
power bus at the back of the machine tool system under investi- 4.2. Tool change and spindle speedepower characteristics
gation and used for current measurement. Fluke 345 has in-built
functions for power measurement, Oscilloscope and Data Logger The tool change process accesses the tool magazine for tool
in a single, hand-held tool. True-rms ac and dc current measure- selection processes based on the programmed NC codes. In the
ments up to 2000 A can be measured. event of a machining task, as the machine tool is switched ‘ON’,
electrical current flows through the system to activate the machine
4. Results and discussions modules to get to the basic state as previously described. Just before
actual cutting starts, there will be a tool change action (could be
4.1. Energy consumption for machine modules and auxiliary units null in some machine tools system as in the case of fixed spindle
without cutting machine tool, vertical or universal milling machines), at which the
machine tool now completes the rapid axis movement to a point
A direct assessment of the energy demand of machine
modules was undertaken to identify the energy demand of the
machine modules and to understand the dominant energy
consumers. A CNC MHP lathe with Open MDSI architecture, Basic Machine Tool State Ready Machine Tool State
Takisawa CNC milling machine and Mikron HSM 400 high speed
machining centre were tested. To measure the electrical current 100
drawn by the machines using the Fluke 435 power clamp meter,
28%
current flow was recorded when the machine was switched ON 80 37%
47%
Power demand (%)

and then individual auxiliary units were identified through the

electrical circuitry. To characterise the electrical energy require-
60
ments by the machine, the current readings were recorded
without any cutting operation. The servos and spindle were then
manually indexed through the jog mode to measure the current 40
72%
during home positioning and tool change. 63%
53%
The power needed for switching on the machine modules of 20
the CNC MHP lathe with Open MDSI architecture was 1229 W.
The machine start-up consumed 3537 W of power. This was due
to the fact that at start-up, most of the auxiliary units are 0
powered. The rapid movement to home location (axes jog) MHP CNC Lathe MAC-V2 Takisawa Mikron HSM 400
Milling Machine High Speed Milling
required 2394 W. Rotating the spindle without cutting (idle
Centre
condition with spindle on) at a speed 1000 rpm required
3594 W. Fig. 2. Basic and ready states power relationship.
 V.A. Balogun, P.T. Mativenga / Journal of Cleaner Production 41 (2013) 179e186 183

Tool Servos idle Cooling Lubricating

Change 0.98% Fans(2x) pumps
4.84% 0.83% 0.59%
Computer
Unit
5.42%
Main Machine
Switch Start-up
8.35% 24.04%

Hydraulic
Pumps
14.26%
Spindle
speed
Servos
24.43%
Home
Fig. 5. Mikron HSM 400 high speed machining, auxiliary units power demand.
Location
16.27%

4.2.1. Effect of spindle speed on energy required by a DC motor

Fig. 3. Non-cutting power consumption distribution of the MHP MDSI CNC open lathe
driven MAC-V2 Takisawa Milling Machine
machine.
Current consumption at different spindle speeds was recorded
at no cutting on the MAC-V2 Takisawa Milling Machine using the
where the machine tool is in a Ready state. The energy demand for Fluke 435 power clamp meter. A tool holder diameter of diameter
tool change task Etc, can be estimated as shown in Equation (10). 50 mm with four uncoated carbide tool inserts, TPKN1603PPTR-
P30, and model number Bristol Erickson 10-527-008-1.P5030
with overhang of 105.16 mm. The material and process parameters
used are as shown in Table 3.
During the analysis, it was observed that the spindle exhibits
Etc ¼ Ptc ttc (10)
three different characteristics when running in non-cutting mode.
These can be related to the power spindle characteristics curve as
Where, Ptc and ttc, represents power demand in W and time in s
shown in Fig. 6. The zones were identified as zone A, B and C. The
respectively for tool change.
rate at which the spindle power required rises with increase in
The next energy consuming unit of a machine tool system is the
spindle speed depends on the spindle design and spindle power
spindle. Its analysis is complex (Hu et al., 2012). However, direct
characteristics as shown in Figs. 6e9. The influence of spindle
measurement and/or statistical modelling of the spindle power
speed on spindle power demand was evaluated and a regression
demand characteristics can be estimated with simplifications such
equation with R-squared of between 97 and 100% obtained. It is
as neglecting the power loss due to friction, vibration of the bearing
therefore, possible to estimate the power demand of the spindle at
units, heat, and viscosity of the spindle lubricant. The assembled
each zone using the power equation as shown in Equations (11)e
spindle of a machine tool generally consist of drives, motor and
(13). The spindle power consumption equation to be used
mechanical transmissions. The energy efficiency of drive compo-
depends on the spindle speed selected during machining processes.
nent and power output characteristics depends on the ratio of
From Figs. 6 and 7, for the power-spindle speed characteristics
delivered power to consumed power and it is therefore the effi-
for spindle speed ranges between 600 and 1500 rpm, the power
ciency of the system. The spindle does accelerations and deceler-
model in Equation (11) should be used for MAC-V2 Takisawa CNC
ation during machining processes. This characteristic affects power
milling machine.
demand.

Ps ¼ 0:8518N  345:26 (11)

Where, Ps is the spindle power and N, is the spindle speed.

For Zone B, spindle speeds ranges 2000e5000 rpm,

Table 3
Workpiece type and process parameters used.

Takisawa Milling Machine Mikron HSM 400

(with DC servo motor machining centre
model 20M, spindle (with HVC140-SB-
A06B-0652-B) 10-15/42-3F-HSK-E40
spindle)
Workpiece Stainless steel T316L Stainless steel T316L
Hardness 220 Vickers 220 Vickers
Spindle speed 650 RPM 650 RPM
Feed rate 75 mm/min 500 mm/min
Cutting depth 0.5 mm 0.5 mm
Cutting fluid Blasocut BC25 Blasocut BC25
Tool holder diameter 50 mm 8 mm
Fig. 4. MAC-V2 Takisawa Milling Machine auxiliary units power demand.
184 V.A. Balogun, P.T. Mativenga / Journal of Cleaner Production 41 (2013) 179e186

Power Linear (Power)

4500

y = 1.181x - 1682.5
R² = 0.9774
3150

Power (W)
1800

Fig. 6. Power-speed characteristics of a MAC-V2 Takisawa Milling Machine tool and 3

zones for energy profile. 450
1600 2800 4000 5200
Spindle Speed (RPM)

Ps ¼ 1:181N  1682:5 (12) Fig. 8. MAC-V2 Takisawa Milling Machine no load power-spindle speed characteristic
in zone B to 5000 rpm.

Likewise, for Zone C, spindle speeds ranges 4800e5600 rpm,

Based on the analysis of the three states of machine tools as
depicted in Fig. 10, the energy demand of each state can be summed
Ps ¼ 1:5513N þ 11423 (13)
and the total energy demand of a machine tool predicted.
The tool life is an important characteristic in machining
It is therefore clear that the spindle power consumption equa- processes. It needs to be used to effect a tool change so that surface
finish and product precision is not compromised. Thus incorpo-
tion to be used depends on the spindle speed selected during
machining processes as shown in Equations (11)e(13). Hence, rating tool life into the energy equation:
Total energy demand equation can therefore be re-written thus:
a generic model was formulated for the spindle speed power
demand as shown in Equation (14).   
t
Etotal ¼ Eb þ Er þ Ptc ttc INT 2 þ 1 þ Pair tair þ ðPs þ Pcool
T
Ps ¼ mN þ C (14)
_ c
þ kvÞt
Where Ps, is the spindle power demand, m, represent the spindle (15a)
speed coefficient and N, represent the spindle speed in revolution
Incorporating spindle power demand, Ps from Equation (14)
per minute and C, a constant.
then;
4.2.2. Development of an improved and new energy model for   
t
milling processes Etotal ¼ Eb þ Er þ Ptc ttc INT 2 þ 1 þ Pair tair þ ðmN þ C
T
The workpiece type and process parameter in Table 3 was used
to undertake a face cleaning cutting toolpath and the generated _ c
þ Pcool þ kvÞt
poweretime graph shown is shown in Fig. 10. The area under the (15b)
graph equates to the total energy demand of machining the
workpiece which was categorised into three zones thus; ‘Basic’,
‘Ready’ and ‘Cutting’ energy states as previously explained. Power Linear (Power)
4200
Power (Watt) Linear (Power (Watt))
1050
Power (W)

y = 0.8518x - 345.26
R² = 0.955
3400
Power (W)

700

y = -1.5513x + 11423
R² = 1

2600
350 4600 5100 5600
900 1200 1500
Spindle Speed (RPM) Spindle Speed (RPM)

Fig. 7. MAC-V2 Takisawa Milling Machine no load power-spindle speed characteristic Fig. 9. MAC-V2 Takisawa Milling Machine no load power-spindle speed characteristic
in zone A to 1500 rpm. in zone C to 5500 rpm.
 V.A. Balogun, P.T. Mativenga / Journal of Cleaner Production 41 (2013) 179e186 185

measured values of power and cycle time which led to an area

under the graph of 391 Whr and 402 Whr respectively for the
4000 Takisawa and Mikron HSM 400 CNC machines. The deviation of the
Cutting state zone
prediction from the energy calculated from the experimental
measurement of current demand was only þ2% and þ3% for the
Ready state zone
3000 Takisawa and the Mikron CNC milling machine respectively. These
values further prove that the energy model as stated in Equation
Power (W)

(15) can be used as a generic and robust estimate of the energy

2000 requirements in machining.
Basic state zone

1000
6. Conclusions

0 The electrical energy requirements for a machining process

1 25 50 75 100 125 150 175 200 225
needs to be modelled in order to account for and optimise the
Time (s) monetary and environmental impact of electricity usage in
manufacturing. This paper classified three categories for the energy
Fig. 10. Total power consumption trend for machining toolpaths. states of machine tools. In addition, to the start-up and tip (cutting)
energy an intermediate step of the ‘Ready state’ was proposed. The
ready state brings the cutting tool and workpiece to a proximity
state or an about to cut state. Current measurements were then
Where Ptc, ttc represent tool change power and tool change time done on an MHP CNC lathe, MAC-V2 Takisawa Milling Machine and
respectively. Mikron HSM 400 Milling centre and some conclusions can be
The cutting time, tc and Tool life T and material removal rate can drawn from the study.
be modelled for turning and milling as a function of cutting velocity
variables thus enabling the use of the equation in process planning. 1. There is growing evidence from literature that the tool tip
energy is typically lower than the energy required by a machine
tool operating at no load. For this reason it is important to
5. Validation of direct energy model during milling processes
further understand what constitutes the power requirements
and hence energy usage for a machine tool. The study shows
In order to validate the energy model in Equation (15),
that machine tools should not be left in a no-cutting mode for
machining trials were conducted on the Takisawa Milling Machine.
a longer time otherwise its energy footprint is significantly
A Bristol Erickson 10-527-008-1.P5030 tool holder of diameter of
increased.
50 mm with overhang of 105.16 mm was used with a four uncoated
2. On a CNC MHP lathe machine, the power requirement of the
carbide tool inserts, TPKN1603PPTR-P30. The testpiece was then
basic machine state (the machine start-up state) was 53% of the
also machined on a Mikron HSM 400 High Speed Machining Centre.
total power requirement for a machine running at no load. At
An SHR-161-6160K 8 mm diameter carbide mill end cutter and an
47% the ready state power is smaller but significant and hence
HSK40E-VC13-90 tool holder were used. The material and process
should be modelled more explicitly and accurately. The
parameters are stated in Table 3. Since the spindle speed is
significance of the Ready state was also confirmed for the
a dominant parameter for energy consumption, tests on both
Takisawa and Mikron HSM CNC machine tools.
machines were done at the same spindle speed. The power
3. The case study on the MHP CNC lathe machine show an
consumption of the corresponding power states of the machine
interesting fact that in a no-cutting mode, the bulk of the
tools system were measured with the Fluke 435 power meter and
power demand arises from machine start-up (45%), spindle
the result shown in Table 4.
power (15%), servo home location (10%), hydraulic pumps
The total energy demand on MAC-V2 Takisawa Milling Machine
(8.9%) and coolant pumps (8.2%). These are the key areas of
and Mikron HSM 400 machining centre calculated using Equation
focus on the redesign of the MHP lathe to target a lower energy
(15) was 399 Whr and 415 Whr. The Fluke 435 Clamp meter gave
footprint resource. Fluid pumping is a major energy consumer
as it required 17.1% of the total power. The design of more
Table 4 energy efficient pumps should be a target.
Power and total energy demand estimation of machine tools under investigation. 4. Total energy demand can be estimated using the generic model
Takisawa Mikron HSM 400 presented. The model was developed to consolidate the
CNC machine machine following key machine tool energy trends:
Basic state power (W) 2760 2904 a. In addition, to the Basic and Cutting States, explicitly
Ready state power (W) 648 401 modelling the energy required to take a machine tool from
Tool change power (W) 0a 920 the Basic State to a state where the axis and tool is ready for
Air cutting power (W) 2955 2917
Coolant power (W) 776 1790
action and about to cut. This has been named the Ready
Spindle no load power (W) 184 184 State.
Specific energy power (W) 125 333 b. Modelling of energy requirements for spindles based on
Total energy calculated from 399 415 spindle speed used and machine tool spindle e power
new model (Whr)
characteristic zones.
Total energy from measured 391 402
current and time (Whr) c. Accounting for the number of tool changes required and
% Difference between model þ2 þ3 associated energy for tool change by incorporating the tool
predicted and measured energy life.
a
Zero represents ‘do need’ for event i.e. tool already in spindle and single pass d. Modelling energy demand for air cutting during toolpath
tested. execution to account for repositioning the cutting tool.
186 V.A. Balogun, P.T. Mativenga / Journal of Cleaner Production 41 (2013) 179e186

e. Modelling energy with an explicit consideration and Draganescu, F., Gheorghe, M., Doicin, C.V., 2003. Models of machine tool efficiency
and specific consumed energy. J. Mater. Process. Tech. 141 (1), 9e15.
incorporation of cutting speeds, feed and depth of cut to
EPTA, 2007. Study for Preparing the First Working Plan of the Eco-design Directive.
support process planning. Report for tender No.: ENTR/06/026. ec.europa.eu/enterprise/policies/
f. Acknowledging that there are differences in number and sustainable-business/files/workingplan_finalreport_en.pdf (accessed March
design of machine tool accessories/modules. 2012).
Gutowski, T., Dahmus, J., Thiriez, A., 2006. Electrical energy requirements for
5. Further work is required to compare the data presented here a manufacturing process. In: Proceedings of 13th CIRP International Conference
with other machine tools and to model the energy consumed on Life Cycle Eng, Leuven, 623.
by machine axis and its dependence on G01, G02 and G03 axis He, Y., Liu, F., Wu, T., Zhong, F.P., Peng, B., 2012. Analysis and estimation of energy
consumption for numerical control machining. Proc. Inst. Mech. Eng. Part B
as well as plane of interpolation. J. Eng. Manuf. 226 (B2), 255e266.
Hu, S., Liu, F., He, Y., Hu, T., 2012. An on-line approach for energy efficiency moni-
toring of machine tools. J. Clean. Prod. 27, 133e140.
References Kellens, K., Dewulf, W., Overcash, M., Hauschild, M.Z., Duflou, J.R., 2012. Method-
ology for systematic analysis and improvement of manufacturing unit process
Anderberg, S.E., Kara, S., Beno, T., 2010. Impact of energy efficiency on computer life cycle inventory (UPLCI) CO2PE! initiative (cooperative effort on process
numerically controlled machining. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. emissions in manufacturing), Part 2: case studies. Int. J. Life Cycle Ass. 17 (2),
224 (B4), 531e541. 242e251.
Avram, O.I., Xirouchakis, P., 2011. Evaluating the use phase energy requirements of Kordonowy, D.N., 2001. A Power Assessment of Machining Tools, Massachusetts
a machine tool system. J. Clean. Prod. 19, 699e711. Institute of Technology, B.S. Thesis, Department of Mech. Eng., Cambridge, MA,
Dahmus, J., Gutowski, T., 2004. An environmental analysis of machining. Proc. ASME USA.
Inter. Mech. Eng. Congr. R&D Expos., 13e19. Li, W., Kara, S., 2011. An empirical model for predicting energy consumption of
Devoldere, T., Dewulf, W., Deprez, W., Willems, B., Duflou, J., 2007. Improvement manufacturing processes: a case of turning process. Proc. Inst. Mech. Eng. Part B
potential for energy consumption in discrete part production machines. In: J. Eng. Manuf. 225 (B9), 1636e1646.
Takata, S., Umeda, Y. (Eds.), Advances in Life Cycle Engineering for Sustainable Mori, M., Fujishima, M., Inamasu, Y., Oda, Y., 2011. A study on energy efficiency
Manufacturing Businesses. Springer, London, pp. 311e316. improvement for machine tools. CIRP Ann. Manuf. Tech. 60 (1), 145e148.
Diaz, N., Helu, M., Jarvis, A., Tonissen, S., Dornfeld, D., Schlosser, R., 2009. Strategies Ostaeyen, J.V., 2010. CO2PE! (Cooperative Effort on Process Emissions in
for Minimum Energy Operation for Precision Machining. In: Proc. of MTTRF Manufacturing) e Taxonomy. www.mech.kuleuven.be/co2pe!/taxonomy.php
2009 Annual Meeting. PRC, Shanghai. (accessed January 2012).
Diaz, N., Redelsheimer, E., Dornfeld, D., 2011. Energy consumption characterization Protocol, K., 1997. United Nations Framework Convention on Climate Change. Kyoto
and reduction strategies for milling machine tool use. In: Hesselbach, J., Protocol, Kyoto.
Herrmann, C. (Eds.), Glocalized Solutions for Sustainability in Manufacturing. Rajemi, M.F., Mativenga, P.T., Aramcharoen, A., 2010. Sustainable machining:
Springer, Berlin, Heidelberg, pp. 263e267. selection of optimum turning conditions based on minimum energy consider-
Digest of United Kingdom energy statistics (DUKES), 2010. Department of Energy ations. J. Clean. Prod. 18 (10e11), 1059e1065.
and Climate Change. TSO. www.decc.gov.uk/assets/decc/11/stats/publications/ Vijayaraghavan, A., Dornfeld, D., 2010. Automated energy monitoring of machine
dukes/2303-dukes-2011-chapter-1-energy.pdf (accessed November 2011). tools. CIRP Ann. Manuf. Tech. 59 (1), 21e24.