Professional Documents
Culture Documents
CM Fasc.2 2010
CM Fasc.2 2010
INSTITUTULUI
POLITEHNIC
DIN IAI
Publicat de
UNIVERSITATEA TEHNIC GHEORGHE ASACHI, IAI
Secia
CONSTRUCII DE MAINI
2010
Editorial Board
President: Prof. dr. eng. Ion Giurm, Member of the Academy of Agricultural
Sciences and Forest, Rector of theGheorghe Asachi Technical University of Iai
Editor-in-Chief: Prof. dr. eng. Carmen Teodosiu, Vice-Rector of the
Gheorghe Asachi Technical University of Iai
Honorary Editors of the Bulletin: Prof. dr. eng. Alfred Braier,
Prof. dr. eng. Hugo Rosman,
Prof. dr. eng. Mihail Voicu, Corresponding Member of the Romanian Academy,
President of the Gheorghe Asachi Technical University of Iai
Editors in Chief of the
Prof. dr. eng. Radu Ibnescu, Assoc. Prof. dr. eng. Aristotel Popescu
Honorary Editors: Prof. dr. eng. Gheorghe Nag, Prof. dr. eng. Cezar Oprian
Associated Editor: Prof. dr. eng. Eugen Axinte
Papers presented at
THE INTERNATIONAL CONFERENCE on
DESIGN, TECNOLOGIES & MANAGEMENT
IN MANUFACTURING
Iai, May 14th 16th, 2010
organized by the
EDITORIAL BOARD
MACHINE CONSTRUCTION
Fascicle 2
Conf.univ.dr.ing. Irina Cozmnc
Prof.univ.dr.ing. Radu Ibnescu
Conf.univ.dr.ing. Vasile V. Merticaru
2010
Secia
CONSTRUCII DE MAINI
SUMAR
Pag.
ANTNIO M. GONALVES-COELHO, GABRIELA NETIAN i
ANTNIO MOURO, Modelul matricei de proiectare n cazul
soluiilor redundante de proiectare (engl., rez. rom.)........................
21
31
41
49
57
65
75
83
91
97
BIRGIT KJRSIDE STORM, Lipirea cu adezivi a aluminiului tratat
superficial (engl., rez. rom.)...................................................................... 105
IOANA PETRE, DAN PETRE, CRISTINA FILIP i LAVINIA NEAGOE,
Aplicaii industriale ale muchilor pneumatici (engl., rez. rom.) ............. 117
MIHI HORODINC, Noi resurse ale cercetrii experimentale asistate
de calculator a puterii electrice absorbite n sistemele de fabricaie
(engl., rez. rom.)........................................................................................ 125
ION BOSTAN, VALERIU DULGHERU i ANATOL SOCHIREANU,
Dezvoltarea integrat CAE a transmisiilor precesionale utiliznd
platforma Autodesk Inventor (engl., rez. rom.) ....................................... 135
ILEANA FULGA i EUGEN STRJESCU, Propuneri de optimizare a
funcionrii morilor fluidice cu jeturi n spiral (engl., rez. rom.)............ 143
DORU CLRAU, IRINA TIA, DAN SCURTU i BOGDAN
CIOBANU, Analiza dinamic a mecanismului mecano-hidraulic de
protecie a turbinelor eoliene cu ax orizontal de mic putere prin
basculare n plan vertical (engl., rez. rom.)............................................... 153
CONSTANTIN CHIRI, ADRIAN HANGANU i DANIEL CALFA,
Cercetri privind sistemele hidraulice pentru deplasarea maselor mari
pe distante mici, cu frecven redus (engl., rez. rom.)............................ 161
2010
Section
MACHINE CONSTRUCTION
CONTENTS
Pag.
ANTNIO M. GONALVES-COELHO, GABRIELA NETIAN and
ANTNIO MOURO, On the Pattern of the Design Matrix in
Redundant Design Solutions (English, Romanian summary). .............
NIKOLAOS TAPOGLOU and ARISTOMENIS ANTONIADIS, CADBased Calculation of Cutting Force Components in Gear Hobbing
(English, Romanian summary)..................................................................
21
31
41
49
57
65
75
83
91
97
1. Introduction
Axiomatic Design (AD) was created in the late 1970s by Nam P. Suh with
the aim of building a systematic model for engineering education and practice
under the initial hypothesis that there are fundamental principles that govern
good design practice [1].
According to AD, any design object being it a product, a process or any
other technical system can be described by a vector in each one of four
design domains (see Fig. 1). The design process starts at the customer domain
with the definition of the customer needs (CNs). Mapping between the customer
and the functional domains allows finding the functional requirements (FRs).
Once this is done, another mapping translates the FRs into design parameters
(DPs), i.e. the set of properties that physically describe the design object.
Finally, mapping from the physical to the process domain leads to the process
variables (PVs), which outline how to make the design object [1].
{Y } = [A ]{X };
Aij =
Yi
; i = 1, ..., m;
X j
j = 1,..., n ,
where {Y} is a vector that represents the set of m requirements that should be
accomplished, {X} is a vector representing the set of n parameters of the design
object that is expected to fulfil the requirements, and [A] is the design matrix.
Usually, any prospective design equation is bounded by constraints [1].
Eq. (1) is not unique, and different {X} vectors would represent different
design solutions that are characterized by distinct design matrices, which
patterns would make the difference between good and poor design. The
good or the poor quality of any design solution is ruled by the Independence
Axiom, which states that, in good design, the selected parameters {X} should be
such that the requirements {Y} are fulfilled independently. As a result, the ideal
design solution should have the same number of requirements and parameters
(m = n) and the design matrix should be diagonal, case of which the design
solution is called uncoupled [1]. A triangular design matrix is acceptable as
well and corresponds to a decoupled design [1]. Any other pattern of square
design matrix corresponds to a coupled design, which should be recognized as
poor and as such should be avoided [1].
For any design solution where m > n, ADs Theorem 1 states that either the
design is coupled or some of its FRs can never be fulfilled [1]. An example of
such a design can be found elsewhere [2]. In the case of m < n, ADs Theorem 3
states that the design is either redundant or coupled [1]. At last, the specific case
of a design with a single requirement (m = 1) is worth to mention. In this case,
the design would be either uncoupled (if n = 1), or redundant (if n > 1). In fact,
the decoupled and the coupled conditions are impossible to attain in any singlerequirement design because there is no different FRs to couple.
The present work contains an analysis of the pattern of the design matrix of
redundant designs with more than one requirement (m > 1), a matter to which
the researchers have not paid sufficient attention so far.
2. The Key Characteristics of Redundant Designs
Fig. 2 depicts the design of a simple clamping device with one only
customer need: the clamping action, which can be attained by adjusting one
only FR the distance d.
(2)
{d}=
A11
.
A12
A more general case is the Eq. (3) that represents a design with arbitrary
numbers of requirements m and parameters n, with m < n. The equation relates
to a redundant design, and its quality depends on the pattern of the design
matrix, as per Theorem 3.
To better understand how the pattern of the design matrix of Eq. (3) could
characterize a good or a poor design, let us consider the three coexisting designs
represented by Eq. (4), Eq. (5) and Eq. (6), where Akj denotes the possible nonzero elements of the related design matrices.
(3)
(4)
(5)
FR1 A11
FR2 = A21
FR A
3
31
A12
A13
A22
A23
A32
A33
FR1 0 0 0 A14
FR2 = 0 0 0 A24
FR 0 0 0 A
3
34
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
DP1
DP2
DP3
DP4
DP5 ,
DP6
DP7
DP8
0 0
0 0
0 0
DP1
DP2
DP3
DP4
DP5 ,
DP6
DP7
DP8
A15
A16
A25
A26
A35
A36
(6)
FR1 0 0 0 0 0 0 A17
FR2 = 0 0 0 0 0 0 A27
FR 0 0 0 0 0 0 A
3
37
A18
A28
A38
DP1
DP2
DP3
DP4 .
DP5
DP6
DP7
DP8
One can see that DP1, DP2 and DP3 are the only possible contributing
parameters in Eq. (4). As for the design of Eq. (5), just DP4, DP5 and DP6 are
significant. At last, in what concerns to Eq. (6), only DP7 and DP8 contribute. In
other words, each one of the above-defined coexisting designs can fulfil all the
FRs of Eq. (3) using entirely different subsets of the DPs included in the latter
equation, in such a manner that all the DPs are taken into account.
Now, one can figure out that the design of Eq. (3) could be achieved by
merging the designs of Eq. (4), Eq. (5) and Eq. (6).
The design matrices of Eq. (4) and Eq. (5) could be reduced to block
matrices of size m x m by eliminating the zero elements of their design matrices.
Therefore, the surrogate condensed design equations could be obtained
from those equations by eliminating the non-relevant DPs, and by using the all
the existing FRs and the non-zero block matrices. For example, Eq. (7) is the
condensed equation that is obtained from Eq. (5).
The same treatment could be done to Eq. (6), but in this case we would
obtain a non-square block matrix of size (n mod m) x m.
FR1 A14 A15 A16 DP4
FR A
3
34 A35 A36 DP6
4. Conclusion
The key conclusion of this paper can be summarized as a new theorem: Let
us suppose a design with m requirements and n parameters, with m < n. Its
design matrix can be partitioned in (n div m) square block matrices of size
m x m, and an extra non-square block matrix of size (n mod m) x m, in such a
manner that the block matrices have not common elements. Such a design is
acceptable if each block matrix describes either an uncoupled or a decoupled
design. Otherwise, the design is of the coupled type.
Received: January, 30, 2010
REFERENCES
1. S u h N. P., The Principles of Design, Oxford Univ. Press, N. Y., 1990.
2. G o n a l v e s C o e l h o A. M., M o u r o A. J. F., Axiomatic Design: The
Meaning of the First Axiom. In: A. Toca, O. Pruteanu (Eds.), Tehnologii
Moderne, Calitate, Restructurare: Culegere de Lucrri tiinifice, 3, pp. 341-344,
Universitatea Tehnic a Moldovei, Chiinu, 2003.
3. F r a d i n h o J., G o n a l v e s C o e l h o A. M., M o u r o A. J. F., An
axiomatic design approach for the cost optimisation of industrial coupled
designs: A case study. In: Gonalves-Coelho A.M. (Ed.), Proc. 5th International
Conference on Axiomatic Design - ICAD 2009, pp. 201-207, Campus de
Caparica, 2009.
acceptabile iar soluiile cuplate nu sunt satisfctoare din punct de vedere tehnic i ar
trebui evitate. Matricele diagonale corespund soluiilor necuplate, matricele
triunghiulare corespund soluiilor decuplate iar toate celelalte tipuri de matrice ptratice
corespund soluiilor cuplate. n cazul n care numrul cerinelor funcionale este mai
mare dect numrul parametrilor de proiectare, unele cerine nu pot fi ndeplinite
niciodat i soluiile sunt de tip cuplat. Proiectarea redundant creeaz o clas specific
de soluii de proiectare n care numrul cerinelor funcionale este mai mic dect
numrul de parametri de proiectare.
Aceast lucrare prezint modaliti de populare ale matricei de proiectare, astfel
nct un proiect de tip redundant s poat deveni de tip necuplat sau decuplat.
In concluzie putem formula o nou teorem: Presupunnd c avem un produs sau un
proces a crui matrice de proiectare are m cerine i n parametri, unde m < n, atunci
matricea de proiectare poate fi descompus n (n div m) submatrici ptratice de ordinul
m x m i o submatrice neptratic de ordinul (n mod m) x m, astfel nct submatricile s
nu aib elemente comune. O astfel de soluie este acceptabil dac fiecare bloc descrie o
soluie necuplat sau decuplat, dac nu, atunci soluia este de tip cuplat.
1. Introduction
Gear hobbing is one of the most efficient generating processes for cutting
external cylindrical gears. Although, hobbing cutters are still quite expensive
due to their complex geometry. Thus their extensive exploitation is very
important for industry. In gear hobbing the chip formation varies for each
cutting tooth due to the fact that every tooth always cuts the same generating
position. As a result, different wear laws are developed leading to uneven wear
distribution on the hob teeth. Thus, the need to adopt an effective wear
prediction model in gear hobbing arises.
In this paper 3D chip determination is studied in order to feed more accurate
data to an existing wear prediction model. Moreover total wear distribution on
the hob teeth can be calculated. This information can be used to optimize
tangential shifting conditions in order to maximize tool life and prolong the
time interval before hob cutter resharpening.
Figure 1 presents the basic nomenclature of the hob cutter that has been
used in the wear simulation program developed in the present work. As it can be
seen in the upper part of the figure, three distinct motions are required by this
cutting method those being the workpiece revolution, the tool rotation, and the
10
11
because the chip characteristics can be calculated with the accuracy provided by
a CAD system.
3. Hob Wear Simulation Process
In order to achieve uniform wear distribution in the majority of the cutting
teeth, a hob tangential shifting is required. In the left part of Figure 2, simulated
chips and the corresponding gaps formed at the indicated generating positions,
are illustrated. These chips are categorized into groups according to their shape
and the cutting direction of the process. In the right part of Figure 2 the wear
laws of several teeth after tangential shifts can be observed.
12
In the first column the wear laws of teeth number 1, 2, 3 and 4 and the total
wear distribution on the hob after a certain number of cuts are presented. The
second column shows the wear laws after the first tangential displacement equal
to the tool axial pitch . After this sifting the cutting tooth i is placed at the
generating position of the tooth i+1. Thus the tooth number 1 that was cutting in
the generating position number 1 quits, whereas the tooth number 5 cuts for the
first time in the generating position 4.
For example, tooth number 4 was cutting initially in the generating position
4. After AS* number of cuts, tooth 4 reaches a certain flank wear depth. In the
first tangential shift, tooth 4 cuts in the generating position 3, obeying the wear
law that governs that generating position. In this position, flank wear increases
with a different rate for another AS* number of cuts. Finally, in the last
tangential shift, tooth 4 cuts in generating position number 2, obeying a
different wear law.
As a result, in the bottom side of figure 2, the normalized flank wear
distribution on the hob after two tangential displacements is presented.
Consequently the wear distribution becomes uniform and the tool exploitation is
enhanced.
Fig. 3 The effect of chip geometry and shape on the tool wear development.
13
The wear behaviour on the hob teeth is primarily influenced by two sets of
parameters. The first set refers to machining data and the geometry of hob and
14
gear. The second set refers to the machine tool, the cutting and working material
combination and the used cooling lubricant. Every possible combination of the
above parameters results in different coefficients in the model. Wear
coefficients are experimentally defined for a variety of materials and geometries
widely used in gear industry.
The middle column of the figure shows the flow chart for wear calculation
and optimal tangential displacement determination. The program has the ability
to calculate wear progress at all generating positions including transient areas.
In order to shift the hob cutter, all the group gears have to be processed. At the
bottom of the figure the outputs of the program are presented. The output
includes wear progress graphs for all generating positions, and total wear
distribution graphs for all tangential displacements. The developed program also
calculates the maximum flank wear and the corresponding number of cuts.
3.2. 3D Chips Determination
As illustrated on figure 5, this paper introduces a novel way of calculating
equivalent chip dimensions. The equivalent chip thickness calculation is more
accurate due to the fact that is measured on the 3D chips cross-sections. Detail
A in the upper right part of the figure, shows the chip cross-section in plane 1.
)
The calculation of thickness in tooths head and in arc distance a f is hereby
determined. The equation predicting the wear progress in the individual
generating positions and the equations of equivalent chip dimensions introduced
in [2] are presented in the bottom side of the figure.
15
The wear determination model includes two stages. The first stage involves
the employment of hobbing data in order to calculate the chip thickness, length
and group for every cutting position of all generating positions. Categorization
of chips in groups is determined by the program FRSWEAR [3].
Fig. 6 Calculations of the equivalent chip thickness and length based on 3D chips.
The measurement of the total length of the chip and the length compressed
on the heads corner are quite precise. As shown in the left bottom side of figure
6, the first step for measuring the chips length compressed on the tooths
heads corner, is to cut the section of the chip from the left flank to the tool head
center H, as the hobs profile is revolving. In this cross-section we can easily
16
and precisely measure the chips thickness and length in head h H and l H,TF in
CAD environment. This data is fed to the equation illustrated in figure 5 in
)
)
order to calculate the arc length a f . Determination of a f is very important due
to the fact that maximum wear depth has been observed by researchers in this
region [2], [3], [7]-[9]. The calculation method of chip thickness h f is similar to
h H . After h s and l H are determined for every generating position, data can be fed
to the wear prediction model. All the calculations above can be simplified by
)
the use of parametric design in CAD environment. In the first step the arc a f is
set as parameter. In the next step the section of the chip from point K to F as the
hob profile revolves is cut. In the chip produced h H and l H,TF can be easily
)
measure and the arc a f calculated precisely. The parameter is set at the
calculated value and the chips regenerated automatically. Thus, chip thickness
h f can be easily measured. The final step includes calculation of the equivalent
thickness h s . The upper part of figure 6 presents the variation of the chip
thickness for all the revolving planes in several generating positions. These
diagrams have been generated from the sequence of the chip thickness results
h H as mentioned above.
3.3. Development of the Wear Prediction Program
The gear flank wear determination program is implemented in MATLAB
high-level matrix array language.
Fig. 7 The Graphical User Interface of the wear simulation program HOBWEAR.
17
The user has the ability to select gear and hob material from a variety of the
most popular ones in industry. This selection changes the wear coefficients used
to determine the flank wear. An automatic report generator has been
incorporated for better organization of the program results.
4. Simulation Results
Wear graphs for each individual generating position is the first output of the
program developed, as presented in figure 8. As it can be seen, teeth number 10,
11 and 12 have almost reached the wear limit of 0,6 mm. For wear width below
0,2 mm the relation between number of cuts and flank wear is linear.
The normalization progress of total wear distribution on the hob cutter from
tangential shifting number 5 to 8 is illustrated in Figure 9. Optimization of
tangential shifting in gear hobbing has been studied by [10]. The credibility of
the above wear prediction model has been verified with the aid of a wide variety
of cutting experiments [2], [3].
Figure 10 presents the wear distribution on the cutting teeth of a hob cutter
for two cases. In both cases, 18 gears of the same width have been cut. In the
first case, 3 tangential shifts took place with 1 displacement per shift. In the
second case, 2 tangential shifts took place with 2 displacement per shift. It is
evident that in the second case the maximum wear depth on the hob cutter is
less than in the first one.
18
19
5. Conclusion
This paper suggests a model which simulates the wear progress on the
hobbing cutter. The end-user has the ability to optimize tangential shifting
conditions, owing to this model, in order to minimize the total gear
manufacturing cost.
Acknowledgements. The authors wish to thank the Research Committee of the
Technical University of Crete for their financial support (via basic research project
2009).
Received:March 10, 2010
REFERENCES
1. D I N 3972, Bezugsprofile von Verzahnwerkzeugen fuer Evolventenverzahnungen
Nach DIN 867, 1981.
2. B o u z a k i s K., K o m p o g i a n n i s S., A n t o n i a d i s A., V i d a k i s N., Gear
Hobbing Cutting Process Simulation and Tool Wear Prediction Models. ASME
Journal of Manufacturing Science and Engineering, 124, 1, 42-51 (2002).
3. B o u z a k i s K. D., Konzept und technologishe Grundlagen zur automatisierten
Erstellung optimaler Bearbeitungsdaten beim Waelzfraesen, Habilitation, TH
Aachen, 1980.
4. F r i d e r i k o s O., Simulation of Chip Formation and Flow in Gear Hobbing Using
the Finite Element Method, Ph.D. Thesis, Aristoteles University of Thessaloniki,
Greece, 2008.
5. D i m i t r i o u V., V i d a k i s, N., A n t o n i a d i s A., Advanced Computer Aided
Design Simulation of Gear Hobbing by Means of 3-Dimensional Kinematics
Modeling, ASME Journal of Manufacturing Science and Engineering, 129, 911918., (2007).
6. D i m i t r i o u V., A n t o n i a d i s A., CAD-based Simulation of the Hobbing
Process for the Manufacturing of Spur and Helical Gears, International Journal
of Advanced Manufacturing Technology, 41, 3-4, 347-357 (2008).
7. A n t o n i a d i s A., Determination of the Impact Tool Stresses During Gear
Hobbing and Determination of Cutting Forces During Hobbing of Hardened
Gears, Ph.D. Thesis, Aristoteles University of Thessaloniki, 1989.
8. A n t o n i a d i s A., V i d a k i s N., B i l a l i s, N., Fatique Fracture Investigation of
Cemented Carbide Tools in Gear Hobbing. Part 1: FEM Modeling of Fly
20
1. Introduction
One of the key components of any torque transmission system is high
precision involute gears. Gears can be constructed with a wide variety of
methods; Gear hobbing is the one mainly used in the modern industry. Gear
hobbing kinematics consists of three relative motions between the cutting tool
and the workpiece. This makes it difficult to simulate with analytical models.
HOB3D is a novel simulation program based on a commercial CAD
environment which can simulate the cutting process and provide results
including 3D solid chips and gaps as well as predicting the developing cutting
forces.
2. State of the Art
The research conducted in the area of gear hobbing can be divided into two
categories: gear hobbing process simulation and wear prediction. In the first
field a series of simulation models have been developed using CAD [1],[2],
FEA [3] or analytical [4]-[7] based models. Experiment based models [8], [9]
22
are used in the field of wear prediction order to calculate the wear of the cutting
tool and optimize the cutting so as to obtain uniform wear along the cutting tool.
Cutting forces prediction is an area of great interest also. Research conducted in
this area is based on Kienzle-Victors equations and depend on geometry of
chips.
3. Gear Hobbing and HOB3D Simulation Process
Gear hobbing kinematics is based on three relative motions between the
cutting tool and the workgear. These motions must be synchronised in order to
produce high quality helical and spur gears. As presented in Fig. 1 the work
gear rotates round its axis while at the same time the hob rotates round its own
axis and moves parallel to the gear axis. The hob is positioned in an angle
relative to the gear. The magnitude of this angle is relative to the hob helix
angle and the helix angle to the gear produced correspondingly.
23
The new simulation code called HOB3D uses three coordinate systems in
order to calculate the results required. The first (1) is positioned on the
examined cutting edge and has the x axis parallel to the hobs axis, the y axis
perpendicular to x axis and finally z axis perpendicular to the prior two.
Coordinate systems (2) and (3) have axis z running through the workgears axis.
In coordinate system (2), x axis is always rotating in order to point to the gap,
while the axes of coordinate system (3) are fixed.
24
25
26
27
The cutting forces components for the above chip are presented in the next
Fig. 6 which are calculated in accordance with the system 1.
28
5. Verification
The verification of the force calculation module was conducted in two
phases. First, the cutting forces simulated on specific teeth were compared to
the ones measured by B o u z a k i s [4] while other cutting forces calculated by
HOB3D were compared to the ones measured by G u t m a n n [5]. Fig. 7
illustrates the results of the first phase of the verification. As it can be seen, each
column of the figure illustrates the measured and calculated cutting forces on
one generating position measured on the coordinate system of the 3D gap (2). In
most of the cases the simulation code predicts not only the form but also the
magnitude of the cutting forces.
29
The second step of the verification is the comparison between measured and
calculated forces for all the cutting teeth simultaneously. Fig. 8 presents the
comparison between the measured and the calculated cutting forces in all three
directions of the coordinate system, as illustrated on the top left side of the
figure.
4. Conclusions
1. A novel simulation model for gear hobbing, developed in a commercial
CAD environment is presented in this paper. The model can simulate the
manufacturing of helical as well as spur gears
2. The developed model can predict the cutting forces components with
great accuracy. The calculated forces have been verified with the aid of cutting
experiments.
30
REFERENCES
1. D i m i t r i o u V., V i d a k i s N., A n t o n i a d i s A., Advanced Computer Aided
Design Simulation of Gear Hobbing by Means of 3-Dimensional Kinematics
Modeling, ASME J. of Manuf. Science and Eng., 129, 911-918., (2007).
2. D i m i t r i o u V., A n t o n i a d i s A., CAD-based Simulation of the Hobbing
Process for the Manufacturing of Spur and Helical Gears, Int. J. of Advanced
Manuf. Technology, 41, 3-4, 347-357 (2008).
3. F r i d e r i k o s O., Simulation of Chip Formation and Flow in Gear Hobbing Using
the Finite Element Method, Ph.D. Thesis, Aristoteles University of Thessaloniki,
Greece 2008.
4. B o u z a k i s K. D., Konzept und technologishe Grundlagen zur automatisierten
Erstellung optimaler Bearbeitungsdaten beim Waelzfraese, Habilitation, TH
Aachen 1980.
5. G u t m a n n P., Zerspankraftberechnung beim Waelzfraesen, Ph.d. thesis, TH
Aachen, 1988.
6. A n t o n i a d i s A., Determination of the Impact Tool Stresses During Gear
Hobbing and Determination of Cutting Forces During Hobbing of Hardened
Gears, Ph.d. thesis, Aristoteles University of Thessaloniki, 1989.
7. A n t o n i a d i s A., V i d a k i s N., B i l a l i s N., Fatique Fracture Investigation of
Cemented Carbide Tools in Gear Hobbing. Part 1: FEM Modeling of Fly
Hobbing and Computational Interpretation of Experimental Results. ASME J. of
Manuf. Science and Engineering, 124, 4, 784-791, 2002.
8. A n t o n i a d i s A., V i d a k i s N., B i l a l i s N., Fatique Fracture Investigation of
Cemented Carbide Tools in Gear Hobbing. Part 2: The Effect of Cutting
Parameters on the Level of Tool Stresses A Quantitative Parametric Analysis,
ASME J. of Manufacturing Science and Engineering, 124, 4, 792-798, 2002.
9. B o u z a k i s K., K o m p o g i a n n i s S., A n t o n i a d i s A., V i d a k i s N., Gear
Hobbing Cutting Process Simulation and Tool Wear Prediction Models. ASME
J.of Manuf. Science and Engineering, 124, 1, 42-51 2002.
1. Introduction
Preoccupations concerning the increase of the standard, standard type and
special cutting tools production and preoccupations concerning new tools with
superior constructive functional performances represent a major tendency in the
manufacturing of machines. The software resources implicates the existence in
the system of a collection of organized information in order to assure not only
the complete design of the cutting tool but also the analysis, the syntheses,
comparison, modeling, simulation optimization, visualization, the optional
presentation of the partial results etc. So the modern design is made in order to
obtain a solid 3D model that is later automatically detailed in sections and views
2D. The existence of the solid model signify the spectacular increase of the
information's number permitting practically the total knowledge of the
positions of any point or surfaces from the elements construction and of the
angles between directions or planes. This demarche ameliorates the tackling of
the cutting tools' design by means of the concurrent engineering.
32
The modern means for the assisted design (Catia, Solid Works, Ideas, Solid
Edge) are capable to make automatically sections in any point and after any
direction, pointing automatically distances between points or angles values.
In this way we can avoid the situations in which we are choosing for the
cutting tool a value for an angle at a peak, but in other sections or points the real
values are outside the domain of availability. Starting from these observations,
we want to develop a new method for the control of the manufactured cutting
tool, starting from the idea of the obtention of a solid model.
This demarche is extremely utile in the admission of the concurrent
engineering.
2. Using the Solid Model for the Cutting Tools' Control
2.1. Possibilities to obtain the 3D Models of the Real Pieces
2.1.1. The getting of the 3D model of the cutting tool by photography. The
software D Sculptor presents a relative new in the large landscape of the CAD
software and the software for treatment of the images and corps 2D and 3D. It
permits the creation on the computer photorealistic models for a large game of
objects using habitual photos, easy and fast. We do not need special hardware
elements, but only a normal
computer and a photo camera.
A digital camera is required,
but it is possible to use
scanned photos. The main
screen of the program is
presented in the fig. 1.
The basic processes for
the models construction are
presented down:
it is positioned the object to
acquire the model on the
calibration plane;
it is photographed the object
from many angles;
Fig. 1 The main screen of D Sculptor software.
the photos are imported in
the software D Sculptor;
D Sculptor detects automa-ticaly the model;
D Sculptor is calculating the three-dimensional model.
The obtained three-dimensional model (3D) (fig. 2) can be used with software. In
time D Sculptor 2.0 was improved at a technical level and from the point of
view of the interface.
33
It is possible to create models faster than before, and the last version of D
Sculptor 2.0 Professional has a superior accuracy.
2.1.2. Generation of the 3D model by scanning. A 3D scanner is a modern
device, intense developed during the last years, device that analyses a real
object or an average in order to collect data concerning the form, the texture and
the color of the object.
The acquired data can be used to digitally construct 3D models, utile for a
large game of applications. In the most situations, a single scanning do not
produces a complete model of the subject. Multiple scanning, some time
hundreds, from many directions are necessary usually to obtain information
about all the sides of the object.
These scans must be introduced in a common reference system, processes
named usually alignment or recording. The scans will fusion in order to create a
complete model.
Now the scanners are of two types:
scanners 3D with contact and without
contact. The scanners non contact can be
active or passive. It exists a large gamut
of technologies for the two categories. In
techniques are imposed the non contact
scanners, with laser.
The essential problem of the getting
of a 3D model by scanning is represented
by the great number of necessaries scans
and by the possible necessity to complete
the acquired model using programs for
Fig. 2. The solid model resulted
assisted design (Catia, SolidWorks etc.).
after the images treatment.
This last activity is more necessary if the
form of the scanned tool is more complex.
For these reasons we can affirm that for the simple tools the proposed
method is not efficient, because the completion of the model need too much
time for routine measurements, but becomes very efficient for the complex
cutting tools, measurable with difficulties, at which we obtain the geometric
parameters baffling to measure (and in the main cases with a low accuracy).
3. The Control of the Geometry Based on the 3D Model
After the getting of the 3D model using the software D Sculptor 2.0 or by
scanning, the model is exported in the Solid Works software, resulting the up
presented (fig. 2).
34
On the obtained model are choose points on the main edge and on the
secondary one and there are given command for the construction of the planes
in which are measured the requested geometric elements.
3.1. The Determination of the Fellows' Cutters' Geometry.
The fellows' cutters are tools destined to the manufacturing by mortising the
internal and external gear of the spur gears with right, inclined of "in V" teeth,
representing from that reason a high degree of universality, as consequence of
the cutting edges' access in areas inaccessible for other kind of tools (for
example, side mills, hobbing cutters, etc.). Also, the fellows' cutters are the
single cutting tools that can machine cylindrical gear wheels with interior
denture, by rolling.
3.1.1. Geometric parameters. The clearance angle at the tooth' peak presents
a big importance, because from that value depends the bigness of the profile
correction in time after successive sharpening.
Also, by the value of this angle depends the size of the lateral clearance
angle on the two flanks of the tooth that result much more little. So, it is
considered the lateral clearance angle at the level of the divided circ,
respectively if the tooth is cut after the division cylinder, the intersection curve
between this cylinder and the involute clearance angle of the tooth will be a
helix. Because the disposition on the involute flanks of the points in which the
values of the angles , , and are determined, it results that in any point we
obtain different values that must very between convenient limits.
The actual methods do not permit the determination of these angles, but
20
3031'
35
only of the values from the peak of the tooth (Fig. 3). The minimal values (that
must have a specified value) are determined on the base of mathematical
relations, but the real angles' variations not controlled.
If the intersection of the tooth with planes parallel with a medial plane of
the tooth is realized (posterior planes, Fig. 4), we obtain different values for the
angles p and p (the plane Pp
- Pp) in the points of
intersection between these
planes and the cutting edge
from the involute flank. The
points
in
which
the
determination is made can be
however dense (from example
0.01 mm), but such frequency
is not necessary, because the
variation limits and the
variation mode can be
observed in only some points,
and that fact simplify the
determination operations.
The Solid Works compuFig. 4 The intersection of the tooth
ter program, by the facilities
with paralel planes.
that are offered, give directly
under the form of a quote the
values obtained for the angles
determined in these planes.
Using a program that permits
the drawing of the mentioned
planes
at
determined
distances,
with
a
pre
established step, becomes
possible to obtain the desired
values in every point on the
edge, and with these values it
is possible to draw a chart
(Fig. 6).
Similarly, it is possible to
construct normal planes at the
Fig. 5 Intersections with orthogonal planes.
cutting edge's tangent in the
anterior determined points on
that edge, or orthogonal planes (Fig. 5). Obviously, in these planes are
determined the angles n and n, or the angles O and O. The determination
conditions are similar with the before precise conditions.
36
4.5
4.27
4
3.5
3.31
3.31
3.31
3.31
3
2.5
p
2.11
1.5
1
0.5
0
M1
M2
M3
M4
2.11
3.31
4.27
3.31
3.31
3.31
3.31
Fig. 6 Variation chart of the angles p and p along the cutting edge
on the involute flank of the fellows' cutter.
3.5
3.31
2.91
2.52
2.5
2.24
2
n
n
1.5
1
0.94
0.95
0.97
0.5
0
M1
M2
M3
M4
0.94
2.52
2.24
3.31
2.91
0.95
0.97
Fig. 7 Variation chart of the angles O, O (noted in the figure with n, n).
Excepted the presented determinations, that are essential for a good service
of the fellows' cutter, it is possible to determine another angles, e,g. in sections
with quidam planes, as in Fig. 8, or with front planes (Pf - Pf) as in Fig. 10.
In this case too, the angles are directly posted by the Solid Works computer
program. The values for the angles for the determinations from the Fig. 8 and
Fig. 10 are presented in the charts from the Fig. 9 and Fig. 11.
37
Fig. 8 Sections with certain parallel planes with the intersections' presentation
7
6.44
6
5
4.44
4
3.36
3
2
3.38
2.91
3.52
3.31
0
0
2.02
1
0
M1
M2
M3
M4
2.02
2.91
4.44
6.44
3.36
3.38
3.31
3.52
We can observe the anomaly of the variation of the normal back rake angle
O. A lot of verification proves that the model is correct, so it is to research the
real situation.
A proposal for a complete control of cutting tools made on the basis on the
getting of a 3D solid model is new and brings the first advantage that it is
possible to obtain all the angles searched at any point of the cutting edges or
active planes.
38
2.5
2.32
2.05
2.01
1.93
1.5
f
f
1
0.5
0
0.74
0.62
0.56
0.48
M1
M2
M3
M4
2.32
2.05
2.01
1.93
0.48
0.56
0.62
0.74
39
4. Conclusion
The method is especially useful for complex tools, small, with the active
surfaces and curved cutting edges, to which access control with current
instruments is very slowness or impossible and where the definition of
theoretical planes control angles is difficult to apply. The other method of
obtaining the 3D model through photography has its limits. Further research
will develop methodologies for control of complex tools for very large or very
small, with curved surfaces, methodologies able to change the angles of the long
edges or surfaces with steps as small.
Received: March 25, 2010
1
"POLITEHNICA" University from Bucharest ,
Department of Machines and Production Systems
e-mail: eugen_strajescu@yahoo.com
2
S.C. MUNPLAST S.A.
e-mail: ostefu@yahoo.com
3
"VALAHIA" University from Targoviste
e-mail: ddumitru@yahoo.com
REFERENCES
.
1. M i n c i u C . , S t r j e s c u E . , .a., Scule achietoare, ndrumar de proiectare.
Editura Tehnic, Bucureti, 1995.
2. S t r j e s c u , E . , Proiectarea sculelor aschietoare, Litografia I.P.Bucureti, l984.
3. E n a c h e , t . , S t r j e s c u , E . , M i n c i u , C . , Metode i programe pentru
proiectarea asistat a sculelor achietoare. Litografia I.P.Bucureti, 1988.
4. S t r a j e s c u E . , P a v l o v O . , Metodologie de control asistat de calculator al
sculelor achietoare pe baza unor metode neconvenionale de obinere a
modelelor 3D. Conferina ICEEMS, Braov, 2005.
5. S t r j e s c u E . , P a v l o v O . , D u m i t r u , D . , C ontributions Concerning the
Informatic Control of the Cutting Tools, International Conference on
Manufacturing Science and Education - MSE Sibiu, 2009.
CONTRIBUII PRIVIND CONTROLUL INFORMATIZAT
AL CUITELOR ROAT
(Rezumat)
n lucrare se prezint bazele unei metodologii pentru controlul informatizat al tuturor
elementelor geometrice i constructive ale sculelor achietoare n general, i al cuitelor roat
pentru mortezat roi dinate cilindricve n special. Metodologia are la baz obinerea unui model
solid, iar acest lucru se poate face fie prin fotografiere, folosindu-se programe potrivite, fie prin
scanare 3D, fie n faza de proiectare. Metoda propus permite controlul absolut al tuturor
parametrilor geometrici i constructivi, n orice punct i n orice plan, precum i trasarea
graficelor de variaie. Se poate imagina o metod prin care un program de proiectare asistat de
tipul Solid Concept, parametrizat, s modifice unghiurile sculei pornind de la valoarea admisibil
a unui anume unghi ntr-un anume plan.
1. Introduction
The profiling of tools which generated by enveloping by the rolling method
rack gear tool and gear shaped tool may be make by some methods:
42
Silviu Berbinschi et al
The two rolling centrodes, C1 circle, associated with the profiles ordered
curls and C2 straight line, associated with the rack-gear tool, are in rolling
movement, so, is respected the condition:
(1)
43
= Rrp 1 ,
x = +a ,
where:
(3)
x
= ; x = y
z
represent the matrix of the current points in the space , respectively xyz;
(4)
a =
Rrp
is the matrix formed with the coordinates of point O1, in the global reference
system, with instantaneous velocity in the translation movement of C1
centrode and Rrp the value of circular centrode C1 (rolling radius).
Also, the revolution movement of C1 centrode is described by transformation
44
Silviu Berbinschi et al
x
X
T
y = 1 (1 ) Y
z
Z
(5)
X
where Y is the matrix of the current point in space XYZ, and
Z
1
0
0
1 (1 ) = 0 cos (1 ) sin (1 )
0 sin ( ) cos ( )
1
1
(6)
(7)
0
0 X 0
1
= 0 cos (1 ) sin (1 ) Y
0 sin ( ) cos ( ) Z R
1
1
rp
while the profile, belongs to the profiles ordered curl, associated with the C1
centrode, in form,
0
= Y (u )
Z (u )
(8)
( u ,1 ) = 0;
(9)
The enveloping of the profile family (9) represents the rack-gear tools
profile.
45
Often, the profile (8) may be replaced by the equations of one surface,
cylindrical or cylindrical helical, of which belongs the profile (8), as crossing
in-plane profile section of surface (the section plane is a perpendicularly plane
on X axis), surface which is described by equations on form:
X = X ( u, t ) ;
(10)
Y = Y ( u, t ) ;
Z = Z ( u, t ) ,
46
Silviu Berbinschi et al
4. Conclusions
47
REFERENCES
1. L i t v i n F.L., Theory of Gearing Reference Publication 1212, NASA. Scientific and
Technical Information Division, Washington D.C., 1984.
2. O a n c e a N., Generarea suprafeelor prin nfurare, Vol. I, Teoreme
fundamentale, Editura Fundaiei Universitare Dunrea de Jos, Galai, 2004,
ISBN 973-627-106-4.
3. O a n c e a N., Generarea suprafeelor prin nfurare, Vol. II, Teoreme
complementare, Editura Fundaiei Universitare Dunrea de Jos, Galai, 2004,
ISBN 973-627-106-4;
4. T e o d o r V., O a n c e a N., D i m a M., Profilarea sculelor prin metode analitice,
Editura Fundaiei Universitare Dunrea de Jos, Galai, 2004, ISBN (10)
973-627-333-4;
5. M i n c i u C., C r o i t o r u S., I l i e S., Aspects regarding generation of non-involute
gear profiles, Proceedings of the International Conference on Manufacturing
Systems, ICMaS 2006, Ed. Academiei Romne, Bucuresti, ISSN 1842-3183, pp.
311-314;
6. B a i c u I., O a n c e a N., Profilarea sculelor prin modelare solid, Ed.
Tehnica-Info, Chiinu, 2002, ISBN 9975-63-172-X;
7. D i m a M., O a n c e a N., T e o d o r V., Modelarea schemelor de achiere la
danturare, Ed. Cermi, Iai, 2007, ISBN 978-973-667-270-5;
8. M o h a n, L. V., S h u n m u g a m, M. S., An orthogonal array based optimization
algorithm for computer-aided measurement of worm surface, Int. J. Adv. Manuf.
Technol. (2006) 30: 434443;
9. P o t t m a n n H., W i e n, T. R a n d r u p, O d e n s e, Rotational and Helical
Surface Approximation for Reverse Engineering, Computing, 60, 307-322
(1998);
10. I v a n o v V., N a n k o v G., K i r o v V., CAD orientated mathematical model for
determination of profile helical surfaces, International Journal of Machine Tools
& Manufacture, Elsevier Science, Pergamon, 38, 8, pp.1001-1015, UK, (1998).
48
Silviu Berbinschi et al
1. Introduction
It was present the general algorithm for profiling the tools generating by
enveloping, by the rolling method, in the CATIA design environment.
In following are presented an application of the proposed algorithm for
rack-gear tools profiling generating ordered curl profiles, for a hexagonal shaft,
as so as, the comparison for the profile determination of the rack-gear tool with
results obtained by analytical methods, based on specialised software, realised
in Java [6], [7].
In the paper entitled Contributions to the Elaborations of a Graphical Method for
Profiling of Tools Which Generate by Enveloping. I. Algorithm is proposed a new
50
Silviu Berbinschi et al
51
52
Silviu Berbinschi et al
0
1
2
3
4
5
6
7
53
Table 1
Points coordinates on tools profile in CATIA program
Nr. Crt. [mm] [mm] Nr. Crt. [mm] [mm] Nr. Crt. [mm] [mm]
1
170
17.07
-5.96
340
36.18
-5.81
17
1.41
-0.78
187
18.96
-6.24
357
38.04
-5.44
34
2.98
-1.57
204
20.86
-6.45
374
39.88
-5
51
4.59
-2.31
221
22.78
-6.6
391
41.69
-4.51
68
6.26
-3.01
238
24.71
-6.68
408
43.46
-3.96
85
7.98
-3.64
255
26.63
-6.7
425
45.2
-3.35
102
9.73
-4.23
272
28.56
-6.65
442
46.89
-2.69
119
11.52
-4.75
289
30.48
-6.54
459
48.53
-1.97
136
13.34
-5.21
306
32.39
-6.36
476
50.13
-1.21
153
15.19
-5.62
323
34.29
-6.11
501
52.36
(1)
X = 0;
Y = u;
Z = a,
54
Silviu Berbinschi et al
with u variable parameter, measured along the profile and a constant value
depending the profile form.
(2)
= 0;
( ) = u cos a sin + Rrp ;
= u sin + a cos Rrp ,
or
(4)
u = Rrp sin .
The equations assembly (2) and (4), for u variable between limits
umin = 0.5 Rrp ; umax = 0.5 Rrp , for a hexagonal shaft, represent the rack-gear
tools profile reciprocally enwrapping with the shafts profile.
In Table 2 are presented the tools profile coordinates determined by the
analytical method [6], [7].
In Figure 5 are presented the profiles form and the errors obtained at
profiling, regarding the coordinates transformation:
55
CATIA = JAVA ;
CATIA = JAVA + 26.1799.
(5)
Table 2
Points coordinates on tools profilenormal method, in Java program
Crt. no. [mm] [mm] Crt. no. [mm] [mm] Crt. no. [mm] [mm]
1
1.40E-06 -26.1799 248 6.698025 -0.28286 496 0.199296 25.83104
2
0.05004 -26.093 249 6.698477 -0.1695 497 0.149679 25.9186
3
0.099941 -26.0059 250 6.698703 -0.05562 498 0.099923 26.00594
4
0.149712 -25.9185 251 6.698701 0.057729 499 0.050028 26.09306
5
0.19934 -25.831 252 6.698471 0.171609 500 8.88E-07 26.17994
4. Conclusion
56
Silviu Berbinschi et al
REFERENCES
1. L i t v i n F. L., Theory of Gearing Reference Publication 1212, NASA. Scientific
and Technical Information Division, Washington D.C., 1984;
2. O a n c e a N., Generarea suprafeelor prin nfurare, Vol. I, Teoreme
fundamentale, Editura Fundaiei Universitare Dunrea de Jos, Galai, 2004,
ISBN 973-627-106-4;
3. O a n c e a N., Generarea suprafeelor prin nfurare, Vol. II, Teoreme
complementare, Editura Fundaiei Universitare Dunrea de Jos, Galai, 2004,
ISBN 973-627-106-4;
4. T e o d o r V., O a n c e a N., D i m a M., Profilarea sculelor prin metode analitice,
Editura Fundaiei Universitare Dunrea de Jos, Galai, 2004, ISBN (10)
973-627-333-4;
5. D i m a M., O a n c e a N., T e o d o r V., Modelarea schemelor de achiere la
danturare, Editura Cermi, Iai, 2007, ISBN 978-973-667-270-5;
6. C u c u M., O a n c e a N., T e o d o r V., Discretly Known Reciprocally Enwrapping
Surfaces Representation Model - Softwares Descriptions, Buletinul Institutului
Politehnic din Iai, Publicat de Universitatea TehnicaGh.Asachi Iai, LII
(LVI), Fasc.5A, Secia Construcii de Maini (2006), pp. 229-232;
7. C u c u M., O a n c e a N.,T e o d o r V., Metoda tangentelor profilarea
sculei-cremalier pentru profiluri circulare, Tehnologii Moderne Calitate
Restructurare (2007), Universitatea Tehnic a Moldovei, 2, Chiinu, pp. 86-89,
ISBN 978-975-45-034-8, ISBN 978-9975-45-035-2.
8. * * * CATIA Version 5.19. DS.- Documentation
1. Introduction
The turning process is probably the most important process of all cutting
processes. It is therefore very important to analyze the process for to
optimization it.
Cutting force has a value and direction of action that depend on the quality
of processed material, on the cutting regime elements, geometric parameters of
cutting tools, cooling-lubricating liquids [4].
The resultant cutting force is defined by the relation:
(1)
where:
Fz=(0.20.3)Fx,
Fy=(0.250.4)Fx.
58
n
piece
Fx
t
Fy
Fz
chip
tool
feed rate
) ( )
p , & , T = g p (& ) (T ) ,
(2)
( )
) (
p , J1 , &, T = G p , J1 (& ) (T ) ,
(3)
( )
(4)
(5)
p
= 0 1 + cut
0p
( )
( )
1n
p
= 0 1 + p
0
59
p
, if p < cut
;
1n
p
, if p cut
,
where: 0 is the initial yield stress, p - the plastic strain, 0p - the reference
p
plastic strain, cut
- the cut-off strain, n - the strain hardening exponent.
The rate sensitivity function (& ) for the Drucker-Prager model is defined
as
(6)
& m1
(& ) = 1 + , if
&0
1
(7)
& m2 &
(& ) = 1 + 1 + cut
&0
&0
& &cut ;
1
1
m2
m1
(T ) = (Tcut )
T Tcut
if T Tcut ,
Tmelt Tcut
The turning simulations were carried out on cylindrical extruded bar with 50
mm diameter and 80mm length for AISI 1060. Length of cut L=20mm. The
cutting tests were conducted without coolant.
Order to study the simulation results, we set a plan of experiments as Table
1. We set the tools parameters (tools angle and material), process parameters
(feed rate, f, cutting speed, v and depth of cut, t). The tools material is carbide
plates which had the parameters according to the Table 1.
60
Table 1
The plan of the experiments
PROCESS PARAMETERS
TOOL
feed, f,
speed, v,
depth, t,
PARAMETERS
mm/rev
rev/min
mm
Clearance angle, ,
0.083
-5
0.208
Lead angle, ,
45
0.416
Nose radius, r, mm
0.4
0.5
78.54
157.08
314.16
For meshing were used triangular elements with 3 nodes, both for piece and
for the tool. AdvantEdgeTM program automatically generates the finite element
network after that specified a maximum length of finite element, mean that the
smallest and most finite element of edge length [5].
The Druker-Prager model was used to simulate the process of chip breaking.
61
Fig. 5 The finite element network for the whole piece-tool-chip with continuous chips
of plastic strain for AISI 1060, f=0.208 mm/rev; v=157.08 m/min.
62
The value of tangential force increases when feed rate increase as shows in Fig.
no. 7.
F [N]
1400
1200
1000
v1
800
v2
600
v3
400
200
f [mm/rev]
0
0
0.1
0.2
0.3
0.4
0.5
0.6
Table 4
The values of the temperature T, C
Cutting speed, v, m/min
314,16
157,08
571.408
480.065
699.709
589.149
714.676
611.191
729.953
622.045
78,54
427.476
496.942
699.709
729.953
63
4. Conclusion
REFERENCES
1. G r z e s i k W., N i e s l o n y P., FEMbased thermal modelling of the cutting
process using power law temperature dependent concept. Archives of Materials
Science and Engineering, 29, 2 (2008), pp. 105-108.
2. z e l T., Modeling of hard part machining: effect of insert edge preparation in
CBN cutting tool. Journal of Materials Processing Technology, 141, (2003) pp.
284-293.
3. D a v i m J. P., F a r i a P., M a r a n h o C., C a r l o s C. A., Finite element
simulation of precision machining on AISI 1045, 2008, 5 Congresso LusoMoambicano de Engenharia, Maputo, Moambique, pp.1-7(ref:25A003).
4. D a v i m J. P., M a r a n h o C. A study of plastic strain and plastic strain rate in
machining of steel AISI 1045 using FEM analysis, Material and Design., 30,
2009, pp. 160-165.
64
5. *** Third Wave Systems AdvantEdgeTM FEM. Users manual, Minneapolis, 2008.
6. G r z e s i k W., B a r t o s z u k M., N i e s o n y P., Finite element modelling of
temperature distribution in the cutting zone in turning processes with differently
coated tools, 13th International Scientific Conference on Achievements in
Mechanical and Materials Engineering, 2005.
7. H a g l u n d A. J., K i s h a w y H. A., R o g e r s R. J., , An exploration of friction
models for the chiptool interface using an Arbitrary LagrangianEulerian finite
element model, Wear. 2008, 265, pp. 542-460.
1. Introduction
The previous papers [1]-[4] were presented the new calculus models for
cutting force evaluation (Eqs. 1), that could replace the traditional equations for
Fz, Fx, Fy components. The theoretical quantities Cz, Cx, Cy in Eqs. (1) result
using the Eqs.(2) and (3), if the chips deviation angle =0 [5], and with Eqs.(4)
if 0.
Fz = C z o t s C d1+ mz ,
(1)
1+ mx
Fx = C x o t s C d
1+ m y
Fy = C y o t s C d
,
.
66
C z = k1 cos ,
(2)
C x = k 2 sin K + k1 k 3 ,
C y = k 3 cos K k1 k 4 .
(
)
k 2 = cos N ( tg N ),
k1 = cos N0 1 + tg N0 ,
(3)
k 3 = sin cos K ,
k 4 = sin sin K .
In Eqs.(3), the rake angle is measured in the plane normal to the active
cutting edge and in Eqs.(4) this angle is measured in the plane including the
chips real flow direction (0). In the same equations, represents the friction
coefficient between the chips and the tooths rake face, considering =0.5...0.6
in designing activities [5].
Using the models in Eqs.(1) ask for the quantities mx, mz, my, that are
amplifying the cutting forces level through the chips contraction coefficient.
Therefore, experimental tests are necessary for the forces Fz, Fx, Fy and also for
the chips contraction coefficient Cd, when cutting with a single tooth.
2. Experimental Tests to Validate the Models for Fz, Fx, Fy
In order to validate the models in Eqs.(1) for the force components Fz, Fx, Fy
on the cutting tooth, a Kistler dynamometer is used to measure the main force
component Fz when cutting four different types of steel. In the same time, the Cd
values are determining by measurement of the chips lengths. These chips are
resulted from certain cutting lengths limited by four longitudinal and equidistant
channels executed on the workpiece and refilled with brass wedges, in order to
ensure the continuity of the cutting process [5].
The experimental values for Cd result using the model in Eq.(5), where the
cutting lenth L = 0.25dc and the chips length, La, result by measuring a
number of n chips. The quantity c represents the channel width, measured on
the workpiece circumference of diameter D.
Cd =
(5)
67
n
L
, La = Lai n.
La
i =1
The experimental tests were conducted using five lathe tools (Co...C4)
having different values of the main constructive angles, as presented in Table 1.
This table includes also the calculated values for k1... k4 and Cz, Cx, Cy, based on
Eqs.(2) and (3).
Table 1
Lathe tools constructive geometry
Lathe
tool
Co
C1
C2
C3
C4
k1
k2
k3
k4
Cz
Cx
Cy
[]
[] [mm]
[]
[]
0
0
70
10
0
1
0.5
0
0
1
0.47 0.17
-10
0
70
10
0
0.898 0.67
0
0
0.898 0.63 0.23
0
-10
70
10
0
1
0.5 -0.06 -0.16 0.985 0.41 0.33
0
10
70
10
0
1
0.5
0.06 0.16 0.985 0.53 0.01
10
0
70
10
0
1.07 0.32
0
0
1.07 0.3 0.01
The specific working conditions and the experimental results (Fz, Cd, mz)
are presented in Tables 2 6. The values of mz were obtained using Eq.(6),
where Cz is according to Table 1 and o is a steel type characteristic.
Fzexp = C z o t s C d1+ mz .
(6)
Table 2
OL44 The influences of tools geometry and cutting feed
o
Cut
material/
Tools type
[daN/
mm2]
OL44/ Co
29
OL44/ C1
29
OL44/ C2
29
OL44/ C3
29
OL44/ C4
29
t
[mm]
s
[mm/
rev]
0.1
0.25
0.1
0.25
0.1
0.25
0.1
0.25
0.1
0.25
n
[rev/
min]
630
630
630
630
630
630
630
630
630
630
L
[mm]
La
[mm]
D
[mm]
Cd
Fz
[daN]
mz
42.5
42.5
38.5
38.5
41.5
41.5
41.5
41.5
41.5
41.5
16.52
17.38
13.03
14.18
14.25
15.11
15.22
16.18
16.47
17.33
58.5
58.5
53.5
53.5
57.3
57.3
57.3
57.3
57.3
57.3
2.57
2.44
2.91
2.72
2.91
2.75
2.73
2.57
2.52
2.39
54.9
113.5
63.4
125.8
58.16
123.6
47.9
118.0
47.65
118.0
1.38
1.30
1.34
1.27
1.17
1.14
1.12
1.24
1.21
1.32
68
Table 3
40C10 The influences of tools geometry and cutting feed
o
Cut
material/
Tools type
[daN/
mm2]
40C10/ Co
78
40C10/ C1
78
40C10/ C2
78
40C10/ C3
78
40C10/ C4
78
t
[mm]
s
[mm/
rev]
0.1
0.25
0.1
0.25
0.1
0.25
0.1
0.25
0.1
0.25
n
[rev/
min]
630
630
630
630
630
630
630
630
630
630
L
[mm]
La
[mm]
D
[mm]
Cd
Fz
[daN]
mz
36.56
36.56
37.03
37.03
36.56
36.56
36.56
36.56
36.56
36.56
19.86
20.74
17.49
19.38
20.32
22.84
23.8
26.36
21.14
22.62
51.0
51.0
51.6
51.6
51.0
51.0
51.0
51.0
51.0
51.0
1.84
1.76
2.12
1.91
1.80
1.60
1.54
1.39
1.73
1.62
52.5
110.0
59.0
117.3
55.0
113.0
54.7
104.7
40.4
103.4
0.98
0.82
0.91
0.86
1.16
1.29
1.95
2.05
0.60
0.88
Table 4
MoCM14 The influences of tools geometry and cutting feed
Cut
material/
Tools type
MoCM14/
Co
MoCM14/
C1
MoCM14/
C2
MoCM14/
C3
MoCM14/
C4
[daN/
mm2]
t
[mm]
74
74
74
74
74
s
[mm/
rev]
0.1
0.25
0.1
0.25
0.1
0.25
0.1
0.25
0.1
0.25
n
[rev/
min]
400
400
400
400
400
400
400
400
400
400
L
[mm]
La
[mm]
D
[mm]
Cd
Fz
[daN]
mz
51.09
51.09
50.54
50.54
50.14
50.14
50.14
50.14
50.14
50.14
30.53
32.16
23.47
25.12
28.11
29.96
18.08
20.14
21.84
24.13
69.5
69.5
68.8
68.8
68.3
68.3
68.3
68.3
68.3
68.3
1.67
1.59
2.15
2.01
1.78
1.67
2.77
2.49
2.30
2.10
57.9
116.4
60.9
134.3
58.9
119.5
57.1
118.0
50.3
102.6
2.99
2.96
1.88
1.98
2.60
2.64
1.01
1.04
1.22
1.24
Table 5
OSC10 The influences of tools geometry and cutting feed
s
o
t
Cut material/
[mm/
[daN/
[mm]
Tools type
rev]
mm2]
0.1
OSC10/ Co
52
2
0.25
0.1
OSC10/ C1
52
2
0.25
0.1
OSC10/ C2
52
2
0.25
0.1
OSC10/ C3
52
2
0.25
0.1
OSC10/ C4
52
2
0.25
n
[rev/
min]
500
500
500
500
500
500
500
500
500
500
L
[mm]
La
[mm]
D
[mm]
Cd
Fz
[daN]
mz
57.9
57.9
57.0
57.0
57.0
57.0
57.0
57.0
57.9
57.9
25.6
28.3
22.5
24.3
23.9
25.9
27.6
29.9
30.0
32.8
78.2
78.2
77.0
77.0
77.0
77.0
77.0
77.0
78.2
78.2
2.26
2.05
2.53
2.35
2.38
2.20
2.06
1.90
1.93
1.76
59.4
119.2
65.3
137.2
61.2
128.6
57.8
113.9
54.4
106.4
1.13
1.12
1.09
1.07
1.05
1.04
1.39
1.32
1.41
1.36
69
Table 6
The influence of main cutting speed
Cut material/
Tools type
[daN/
mm2]
t
[mm]
s
[mm/
rev]
OL44/ Co
29
0.1
OSC10/ Co
52
0.1
MoCM14/ Co
74
0.1
40C10/ C3
78
0.1
v
[m/
min]
85
115
165
80
122.5
164
87
120
160
80
102
150
Cd
Fz
[daN]
mz
2.62
2.57
2.50
2-35
2.26
2.20
1.67
1.52
1.48
1.90
1.86
1.80
60
55
52
68
63
61
30
27
25
65
63
58
1.43
1.38
1.37
1.19
1.21
1.24
1.72
2.08
2.09
1.21
1.24
1.22
Constant parameters:
K=70; =10; =0; =0; wear h=0; cutting in air
For the most ductile steel type (OL44) and the most resistant one (40C10),
the experimental results regarding the influences of the constructive angles,
cutting feed and main cutting speed are illustrated in the diagrams in Figs. 1-4.
OL44
=0
s=0.1mm/rev
70
OL44
=0
s=0.1mm/rev
40C10
=0
s=0.1mm/rev
71
OL44
=0
=0
72
40C10
=0;=0
s=0.1mm/rev
From the diagrams in Figs.1 4 result that Fz and Cd are decreasing when
the angles , and cutting speed values are growing. When the cutting feed is
increasing, the force Fz grows and the coefficient Cd decrease. For OL44 steel,
the Fz values are smaller and the Cd are larger than those resulted for 40C10
steel.
For OL44 steel, the mz exponent is relatively constant in the 1.2 1.4 range.
For 40C10 steel, the mz values are variable in a larger domain, 0.8 1.22. For
=0 and =0 the mz exponent may achieve even the value of 1.95 (Fig.2b).
This is the reason why mz it is necessary to be reached for every material type.
The chemical composition, the nature and the structural micro constituents
distribution of the cut material determine the plastic deformation capability.
Consequently, the cutting forces level, the values of Cd and the mz values
through which Cd amplify the cutting forces values are indirectly influenced.
4. Conclusions
73
REFERENCES
1. C o z m n c M., C o z m n c I., P o e n a r u S., A new Model for Estimating
the Force Components Fz, Fx, Fy when Cutting Metals with Single Tooth Tools,
Buletinul Inst. Polit. Iai, editat de Univ. Tehnic Gheorghe Asachi, LV(LIX),
Construcii de maini, 1, 2009, pp.1-7.
2. C o z m n c M., C o z m n c I., P o e n a r u S., Variation of the Ratios
between Forces Fz, Fx, Fy and the Plastic Strain Force in Single Tooth Cutting.
Buletinul Inst. Polit. Iai, editat de Univ. Tehnic Gheorghe Asachi, LV(LIX),
Construcii de maini, 1, 2009, pp.8-14.
3. P o e n a r u S., C o z m n c M., Model for the Deformation Force at Metals
Cutting with Single Tooth Tools. Buletinul Inst. Polit. Iai, editat de Univ.
Tehnic Gheorghe Asachi, LV(LIX), Construcii de maini, 2, 2009, pp.1-7.
4. C o z m n c M., C o n s t a n t i n e s c u C., Bazele achierii. Ed Gh. Asachi,
Iai, 1995.
1. Introduction
The cutting forces developed during the milling process are affected by a
series of parameters, including cutting feed, the milling depth and width, tooth
number, tooth geometry, the relative position of cutting teeth and the material
being cut, and the hardness of the material being cut. Therefore, the most
calculus models of face milling forces are including these parameters, and also
some specific elements of the tools circular motion [1].
This paper presents new theoretical models for the evaluation of cutting
force components at face milling, considering besides the parameters described
above, also the specific elements for each variant of face milling process:
complete (full), symmetrical incomplete or unsymmetrical incomplete, and the
relative position of cutting teeth and the material being cut: cut-down milling
and cut-up milling.
The face milling forces are also depending on the number of teeth that
simultaneously cut (odd or even number of teeth) [2].
76
zs =
arcsin
t
at incomplete face milling [2].
D
FZ = Fz + 2
( z s / 2 ) 1
Fz cos(
1
(1)
FX = Fx + 2
( z s / 2 ) 1
Fx cos(
1
( z s / 2 ) 1
2
2
zi ) = Fz 1 + 2 Fz cos( zi ) ,
z
z
1
( z s / 2 ) 1
2
2
zi ) = Fx 1 + 2 Fx cos(
zi ) ,
z
z
1
FY = Fy z s .
77
b)
FZ = 2 Fz cos(
1
(2)
zs / 2
FX = 2 Fx cos(
1
zs / 2
2
2
zi ) = 2 Fz cos(
zi ) ,
z
z
1
zs / 2
2
2
zi ) = 2 Fx cos( zi ) ,
z
z
1
FY = Fy zs .
From Eqs. (1) and (2) result that in symmetrical face milling the forces FZ,
FX and FY depend on the components Fz, Fx and Fy developed on a single-tooth
level, the number of teeth that simultaneously cut (zs) and the cutting tooth
position beside the XYZ coordinates system of the tool. The axial component FY
doesnt depend on the relative position of the cutting tooth.
2.2. Theoretical Models of Cutting Force Components in
Unsymmetrical Face Milling
In this case, the values of cutting force components FZ, FX, FY are influenced
by the relative position of the tool and the material being cut, resulting different
equations for cut-up (conventional) milling and cut-down (climb) milling.
Fig. 2 shows the geometrical models of unsymmetrical cut-up milling and its
three possible variants:
a) Unsymmetrical face milling with t' = D/2, = 90, and z s1 = z / 4 ;
78
b) Unsymmetrical
z s2 =
face
milling
with
t"
<
D/2,
<
90
and
z
2t
arccos(1 ) ;
D
2
c) Unsymmetrical face milling with t > D/2 , where t = t' + t"' and t"' = t -D/2,
> 90 and z s3 = z s1 + z s4 , where z s4 =
z
2t
arcsin( 1) .
2
D
a) z s1 = z / 4
z/4
FZ = Fz sin(
0
(3)
z/4
FX = Fx sin(
0
z/4
2
2
zi ) Fx cos( zi ) ,
z
z
0
z/4
2
2
zi ) + Fz cos( zi ) ,
z
z
0
FY = Fy z s1 .
b) z s2 =
2t
z
arccos(1 )
2
D
z s2
s2
2
2
FZ = Fz sin( zi ) Fx cos( zi ) ,
z
z
0
0
(4)
z s2
FX = Fx sin(
0
s2
2
2
zi ) + Fz cos( zi ) ,
z
z
0
FY = Fy z s2 .
79
c) For the unsymmetrical milling with t > D/2, we suppose that for a
number of teeth that simultaneously cut z s3 = z s1 + z s4 , the values of cutting
force components will be obtained as the sum of the values of cutting force
components corresponding to a tooth z s1 , Eqs. (3), and a tooth zs 4 , Eqs. (5).
2
zi ) +
z
z s4 1
2
FX = Fx cos( zi )
z
1
z s4 1
FZ =
z s4 1
Fz cos(
1
z s4 1
(5)
F sin(
x
F sin(
z
2
zi ) ,
z
2
zi ) ,
z
FY = Fy z s4 .
The values of cutting force components corresponding to the total number
of teeth that simultaneously cut ( z s3 ) result using the Eqs. (6).
2
FZ = Fz sin(
zi ) +
z
0
z/4
z s 4 1
cos(
z/4
2
2
z i ) Fx cos(
zi )
z
z
0
z s 4 1
2
+ Fx sin(
zi )
z
1
(6)
z/4
FX = Fx sin(
0
2
zi ) +
z
z s 4 1
Fx cos(
1
z/4
2
2
z i ) + Fz cos(
zi )
z
z
0
z s 4 1
2
Fz sin(
zi )
z
1
FY = Fy z s3 .
It can be observed that in case of cut-up face milling with cu t' = D/2, and t''
< D/2, the same relations for the cutting force components FZ, FX, FY are
obtained, distinguishing only by the number of teeth that simultaneously cut.
Fig. 3 is illustrating the geometrical models of unsymmetrical cut-down
milling and the same three possible variants:
80
For all variants of unsymmetrical cut-down face milling, the cutting force
components FZ, FX, FY are given by the Eqs.(7) (8).
a) z s1 = z / 4
z/4
FZ = Fz cos(
1
(7)
z/4
FX = Fx cos(
1
b) z s2 =
z/4
2
2
zi ) + Fx sin( zi ) ,
z
z
1
z/4
2
2
zi ) Fz sin(
zi ) ,
z
z
1
FY = Fy z s1 .
z
2t
arccos(1 )
2
D
z s2
s2
2
2
FZ = Fz cos( zi ) + Fx sin( zi ) ,
z
z
0
0
(8)
z s2
FX = Fx cos(
0
s2
2
2
zi ) Fz sin(
zi ) ,
z
z
0
FY = Fy z s2 .
c) Similarly to cut-up milling, for the unsymmetrical cut-down milling
with t > D/2, we use the same method for determining the values of components
FZ , FX , FY . First, will be determined the values of cutting force components
corresponding to a number of teeth that simultaneously cut z s4 , Eqs. (9).
81
2
FZ = Fz sin( zi )
z
0
z s4 1
2
FX = Fx sin( zi ) +
z
0
(9)
z s4 1
F cos(
x
z s4 1
F cos(
z
2
zi )
z
2
zi )
z
FY = Fy z s4
The values of cutting force components corresponding to a total number of
teeth that simultaneously cut ( zs3 ) result using the Eqs.(10).
2
FZ = Fz cos(
zi ) +
z
1
z/4
z s 4 1
Fz sin(
0
z/4
2
2
z i ) + Fx sin(
zi )
z
z
1
z s 4 1
2
Fx cos(
zi )
z
0
(10)
2
FX = Fx cos(
zi ) +
z
1
z/4
z s 4 1
Fx sin(
0
z/4
2
2
z i ) Fz sin(
zi )
z
z
1
z s 4 1
2
+ Fz cos(
zi )
z
0
FY = Fy z s3 .
Similarly to cut-up milling, it can be observed that in case of cut-down face
milling with cu t' = D/2, and t'' < D/2, the same relations for the cutting force
components FZ, FX , FY are obtained, distinguishing only by the number of teeth
that simultaneously cut.
3. Conclusions
1. The theoretical models presented in the paper are concentrating the main
elements that influence the level of cutting force at face milling. These specific
elements refer to the tooth geometry (, K, ), standard compression yield point
of the material being cut, the cross-sectional area of the theoretical chip, the
milling width, the contraction chips coefficient, the constructive geometry of the
tooth (, K, ), the number of teeth that simultaneously cut, the cutter diameter
and teeth number.
82
2. Using those models to evaluate the cutting force provides extra accuracy
because we are considering the forces on a tooth, the number of teeth that
simultaneously cut, and also the relative position of the tool and material being
cut for all possible variants of the face milling process.
Received: January 29, 2010
REFERENCES
1. C o z m n c M., C o n s t a n t i n e s c u C, Bazele achierii, Ed. Gh. Asachi,
Iai, 1995.
2. C o z m n c M. et al., About the cutting forces at face milling, Buletinul
Institutului Politehnic din Iai, 2, Construcii de Maini, (2009).
3. C o z m n c M. et al., A new model for estimating the force components Fz, Fx and
Fy when cutting metals with single tooth tools, Buletinul Institutului Politehnic
din Iai, 1, Construcii de Maini, (2009).
4. C o z m n c M. et al., Variation of the ratios between forces Fz, Fx and Fy and the
plastic strain force in single tooth cutting , Buletinul Institutului Politehnic din
Iai, 1, Construcii de Maini, (2009).
1. Introduction
The nesting of the blanks is a problem of special importance in the area of
cold plastic forming processes. By nesting the blanks we understand the optimal
cutting of metal sheets into individual blanks or strips. We understand as well
the judiciously positioning or placing on metal sheets or strips the plan
components as structured technological forms used to detach the pieces in order
to get a minimum amount of waste scrap and an optimum coefficient of material
utilization [1].
84
85
a)
b)
c)
Fig. 2 Nesting variants.
In the present case, for cutting the 6.6 m2 of the metal sheet, it was chosen
the RADAN software of designing, modeling and simulating with the AMADA
ARIES 245 pressing center. Following there are presented some sequences of
the program that calculated and cut the part from Fig. 1.
86
87
88
There are presented a part of the steps of the processing program. The
generated program processes a number of 98 pieces from a 3000 x 1500 mm
sheet metal. At the end, as it is noticeable, the program shows exactly the
number of program blocks, the number of tool changes, the programmed stops
and the run time.
Line43:
Line44:
Line45:
Line46:
Line47:
Line48:
Line49:
Line50:
Line51:
Line52:
Line53
Line54:
Line56:
Line57:
Line58:
Line59:
Line64:
Line65:
B2
N43X 2.5Y119.5T12
N44Y40.5
N45X202.5
N46Y119.5
N47X407.5
N48Y40.5
N49X612.5
N50Y119.5
N51X817.5
N52Y40.5
N53G98X17.5Y59.98I205.J0.P3K0
N55G98X201.5Y162.52T1 called from Line 61
N56G72X 2.Y40.02 called from Line 61
N57G66I204.J0.P 85.Q 5. called from Line 61
N58M00 called from Line 61
N63G98X17.5Y100.I0.J0.P0.K0
N64G50
65
25 544
0
3
4
0
1813 secs = 30.21 min.
In the figures below are presented the steps that must be covered in order to
cut the part. Therefore, in Fig. 3, there are presented the processing centres that
use the RADAN software. From these ones, it was chosen for the part to be cut
using the AMADA ARIES 245, using as well the specified material with 1 mm
thickness.
In figure 4, based on the part geometry and dimensions, are presented four
types of standard metal sheets that may be processed by the machine. There are
also presented the usage coefficients of the material and the exact number of
89
parts that may result from the process and the number of metal sheets that are
needed to cover the approximately 6.6 m2 .
It is noticeable that the coefficient of utilization of the material varies based
on the dimensions of the sheet and the number of parts increases proportionally
with the sheet dimensions. In Fig. 5 is presented the placing of the part on the
metal sheet, the anchorage of the metal sheet on the pressing centre and the
distances between the pieces that are about to be cut. In the last figure, Fig. 6 is
presented the technological flow of cutting and the modality in which the tool
approaches and clips step by step the part.
4. Conclusions
1. The CNC pressing centres are much more efficient than the classical
pressing methods, the evidence of this fact being the calculation made for the
two types of processing of blanks and the coefficient of utilization of the
material.
2. For the CNC pressing centres the coefficient of utilization varies
depending on the dimensions of the metal sheets while, for the classical
pressing, if it would have been used a longer metal strip the usage coefficient
would have been the same. I this case it is more appropriate to use a CNC
pressing centre.
3. The number of parts obtained from CNC pressing centres is much
higher than the number of parts obtained with the classical presses.
4. The CNC pressing centres are characterised by the easiness of
calculation at changing the setting of a parameter that has direct influence over
the processing of the blanks, due to the presence of the CAD/CAM system
presented.
5. An important advantage of the CNC pressing centres, unlike the
classical presses, at the processing of simple configuration parts (like the one
presented above) is the geometry and the fabrication costs of the tools used in
production.
90
REFERENCES
1. B r a h a V., N a g i t G h., N e g o e s c u F., Tehnologia Presarii la Rece, Editura
Tehnica, Stiintifica si Didactica CERMI, Iasi, 2003, pp 138-163.
2. P a n M., R a o Y., An integrated knowledge based system for sheet metal cutting
punching combination processing, Knowledge-Based Systems 22, 2009, 368
375.
3. R a o Y., H u a n g G., P e i g e n L., S h a o X., D a o y u a n Y., An integrated
manufacturing information system for mass sheet metal cutting, International
Journal Advanced Manufacturing Technology 33, 2007, 436 448.
4. E n d o J., O h b a S.,A n z a i T., Virtual manufacturing for sheet metal processing,
Journal of Materials Processing Technology, 60 (1996, 191196.
5. C h a t u r v e d i S., A l l a d a V., Integrated Manufacturing System for Precision
Press Tooling, International Journal Advanced Manufacturing Technology, 15,
(1999) 356365.
1. Introduction
The Ultrasonic Testing Method of materials is practiced by more than 50
years [1], [3].Since the first examination in detecting defects using ultrasonic
oscillations of different materials, it became a classic test method based on
measurements. That method takes in consideration all those factors that have a
certain impact [4]. It is expected that in our days the ultrasonic testing to give
results with minimum tolerance, if that method is sustained by a large variety of
instruments and by one advance technical execution [4].
The structure transformations of the material underlayers by manufacturing
process lead to changes for the workpiece general properties [2]. After the type
of defects are detecting, that will be compared with contractual standard. The
contractual standard would decide if the detected defect could be accepted, not
accepted, or remediable [2]. In many cases some of defects could be accepted if
their values respect standard limits. Once detected the defects types, they must
92
2. Method of Investigation
2.1. The Ultrasonic Testing Method
Performing the non-destructive control with ultrasound effectuation consist
principally in transmitting ultrasound signals (waves), produced by a waves
generator, through the examined pieces [2], [3]. The ultrasounds signals are
reflected by any surface and by any defect from the interior of the piece [1]-[5]
that was exposed to testing, see fig.1.
Fig. 1 The ultrasonic testing method [5] Fig. 2 The geometry examination [2]
Device
USM
35X
USLT
2000
Table 1
The necessary instruments of control US
Transducer
Piece
Reference
model: B2S
Aluminum bare
90 x 318 mm
Axle OL47
900 x 1500mm
Coupled
Oil
93
of ultrasonic waves from transducer through the examined piece, the reflection
modalities of the waves and the way displaying on the screen of the device.
When the diameter of transducer is comparable to the width of the piece, due to
the divergence and partial reflection of the beam from the lateral surfaces of the
piece, along the back wall echo and additional signal is formed, more or less
distinct, named divergence marginal echo [2]. The necessary condition for
eliminated this effect is:
(1)
B Dp
2l
tg arcsin 1, 22
,
Dp
3. Experimental Results
The experimental studies were performed with the devices type
Krautkramer (USM 35 X and USLT 2000) in order to check possible errors in
measurement. The pieces subjected to ultrasonic testing are: an aluminum bar
with dimensions 90 x 318 mm and one axle in OL47, forged and heat treated.
By conducting the experiments we achieved the following results.
3.1. First case. In this case the back wall echo can be confused with an
echo of a defect (See fig. 3). These echoes can be called divergent effects.
94
This measurement error is due to amplification (61 dB) that distorts the back
wall echo base and to the transducer which is positioned near the edge of the
sample.
3.2. Second case. It may be noted that in this case the effect of divergence
and misinterpretation can be eliminated by positioning the transducer on the
axial direction of the sample. In this manner, the repetition rate of the flaw
(PRF) does not influence the accuracy of the signal and the interpretation is
objective. Bottom echo is very clear. (See fig. 5).
3.3. Third case. Analyzing the measurements made on both samples we can
conclude that the divergence effect can be eliminated by reducing the repetition
rate of the flaw.
a)
b)
As we can see in Fig. 7a), parasites echoes are present and at a PRF9 (See
Fig.7b)) these additional effects disappear. In this kind of situations it is
recommended to get more measurements at a minimum level of PRF.
95
These cases study presented show how to determine the marginal effect of
divergence and to eliminate stray echoes in order to obtain effective results.
4. Conclusions
Behind the effectuated researches one can evidence the following
conclusions:
1. Considering the applicability of this method and its expansion
possibilities, it is necessary to analyze the factors that influence ultrasonic
nondestructive testing.
2. The position of transducer directly influences the results of testing. It is
preferable that measurements are effectuated in different points of the
investigated surface.
3. When the transducer is near the edge of the piece, the back wall echo
becomes smaller and it can be interpreted as a defect. This leads to
misinterpretation.
4. To eliminate stray echoes that appear on the display device is
necessary for the repetition rate of the flaw (PRF) is minimized
Acknowledgements. This paper was realised with the support of BRAIN Doctoral
scholarships as an investment in intelligence project, financed by the European Social
Found and Romanian Government.
Received:March 3, 2010
REFERENCES
1. B e r k e M., Nondestructive Material Testing with Ultrasonics, Introduction to the
Basic Principles, The e-Journal of Nondestructive Testing & Ultrasonics, 5,
2000.
2. V o i c u I. S a f t a, Defectoscopie nedistructiva industriala, Ed. Sudura, Timisoara,
2001.
3. K r a u t k r a m e r J. & H., Ultrasonic testing of materials, Springer Verlag Berlin
Heidelberg New York, 1983.
4. P e t c u l e s c u P., Ultrasounds Fundamental. Aplications, Ed. Univ. Ovidius,
Constanta, 2002.
96
1. Introduction
Accuracy is the degree of form areas of agreement between actual shape of
the areas resulting from the processing and documentation of performance
prescribed form. The causes of form deviations are multiple of which can be
mentioned:
elastic deformation of pieces during processing;
tighten the fixing the wrong parts in devices;
failure of machine tools guidance;
wear or deformation of system machine-tool - cutting tools elements.
The effects of shape irregularities are of the least desired, with negative
implications for economic and technical performance of pieces processed:
edit the sliding friction between the parts of joints,
reduce static and dynamic tightness of parts and joints,
edit character fits,
they increase wear and may lead to locking parts in contact
decrease the durability of parts in contact.
98
99
Transducer was chosen and the whole system was built base on
recommendations from the literature [3] [5].
Part of the measure is placed on two supports adjustable so that it can be
defined by two points equal height verification. This determines a straight line,
to which will determine the actual right - edge play. For measurement: using
an inductive transducer type Hottinger Baldwin Messtechnik WA T-10mm,
supplied with 20 V DC voltage from a source Hameg HM 8040-2. Output is
taken from a data acquisition card National Instruments USB 6009.
Measurement chain is completed with a PC. No other mode of signal
conditioning is required. Virtual instrument was developed programming
environment LabVIEW 8.2 Student Edition.
The block diagram of the application is presented in Fig. 2 [6] and involves
the DAQ Assistant Express VI, the Waveform Chart for displaying the
acquired signal and also the Build Table and the Write to Measurement File
blocks for writing the acquired signal values in a table and in a file for further
analysis.
100
0.02
d [mm]
0.01
-0.01
-0.02
0
50
100
150
l [mm]
101
moving average is 5 and in this case the first few elements of yf are
calculated as follows:
yf(1) = y(1)
yf(2) = (y(1) + y(2) + y(3))/3
yf(3) = (y(1) + y(2) + y(3) + y(4) + y(5))/5
yf(4) = (y(2) + y(3) + y(4) + y(5) + y(6))/5 and so on.
The diagram of the whole acquired data filtered for a span equal with
3 is shown in Fig. 4.
0.02
d [mm]
0.01
-0.01
-0.02
0
50
100
150
l [mm]
102
d [mm]
0.01
-0.01
-0.02
0
10
20
30
40
50
l [mm]
Fig. 5 The filtered and unfiltered values for a span equal with 3.
0.02
d [mm]
0.01
-0.01
-0.02
0
10
20
30
40
l [mm]
Fig. 6 The filtered and unfiltered values for a span equal with 5.
50
103
4. Conclusions
REFERENCES
1. K e j k P., K l u s e r C., B i s c h o f b e r g e r R., P o p o v i c R. S. A low-cost
inductive proximity sensor for industrial applications, Sensors and Actuators A:
Physical, 110, 1-3 (2004) pp. 93-97.
2. M a r i M., A new inductive displacement transducer, Sensors and Actuators A:
Physical, 70, 3, 30 (1998), pp. 223-237.
3. P a r k J., M c k a y S., Practical data acquisition for instrumentation and control
systems, Newnes, Elsevier 2003.
4. R i p k a P., T i p e k A., Modern Sensors Handbook, ISTE Ltd, London, 2007
5. S i n c l a i r I.R., Sensors and transducers, Newnes, Butterwork Heinemann,
Oxford, 2001.
6. U n g u r e a n u C., I b a n e s c u R., C o z m i n c a I., Virtual Instrument for
Linear Position Measurement, Buletinul Institutului Politehnic din Iasi, publicat
de Universitatea Tehnica Gh. Asachi Iasi, LV(LIX), 4, 2009, pag.54-62.
7. www.mathworks.com
104
1. Introduction
Adhesive Bonding of aluminium and aluminium alloys can be carried out so
a strong bonding is the result. The perfect bonding between two aluminium
106
parts will be a bonding, where the material will break in the aluminium part,
because in that case the bonding is stronger than the aluminium.
Anodized aluminium can - as aluminium with other surface treatments - be
bonded by an adhesive bonding. For obtaining a strong bonding, which means a
bonding which has a least the same strength as the aluminium alloy, there need
to be a good adhesion between the anodized aluminium surface and the
adhesive. For adhesive bonding there are various theories. No single theory
explains adhesion in general and in reality it is probably a combination between
the different theoretical explanations.
There are six theories of adhesion. They are physical adsorption theory,
chemical bonding theory, diffusion theory, electrostatic theory, mechanical
interlocking theory and weak boundary layer theory.
For anodized aluminium the physical adsorption theory and mechanical
interlocking are the main methods for making the bonding, but the chemical
bonding theory can occur if the surface is treated in a special way.
The physical adsorption need to be present, if a bonding shall be strong
enough. For creating the physical adsorption it is necessary to have a sufficient
wetting of the surface.
The physical adsorption gives the adsorption forces in the interphase
between the metal or the anodized layer and the adhesive. The larger interphase
between the two interfaces the higher strength will be obtained in the adhesive
bonding.
The mechanical interlocking will always be possible and in many cases, it is
the only connection between the metal/anodized layer and the adhesive. This is
the case, if the adhesive is not able to wet the surface.
The mechanical interlocking is good in many ways, because if the surface is
larger the area over which the physical adsorption can occur is larger. If the
surface is rough the forces created in the interphase will give forces in different
directions, which also will give a better physical adhesion.
The chemical bonding in the interphase is the absolute most attractive force
to obtain, if it can be possible. In some cases it is possible to choose an
adhesive, which can react chemical with the surface and especially this can be
possible in a fresh made anodized aluminium surface, where reactive groups
still are present.
107
of a surface and the determination of the surface tension by the droplet method
is sketched. The surface tension of as well the metal/anodized metal and the
adhesives depends on the secondary bonding forces in the material. For
aluminium and anodized aluminium the surface will react with CO2. This
reaction causes a low surface tension of the surface, which gives a bad wetting
with the adhesive.
The surface tension can be measured or it can be calculated from knowledge
about the solubility parameters. The solubility parameters can be measured
indirectly or they can be calculated.
tot2 = Evap/Vl,
tot is the total is the total solubility, Evap the heat evaporation and Vl is the
molar volume.
The one dimensional solubility parameter gives not the total answer for the
secondary bonding forces. Therefore Charles Hansen defined the Hansen
Solubility Parameters, HSP, which are three dimensional solubility parameters.
(2)
tot2 = d2 + p2 + h2,
d is the contribution from the dispersion forces, p is the contribution from the
polar forces and h is the contribution from the hydrogen forces.
The three contributions will be put into a three dimensional coordinate
system, where the dispersion forces will be printed with the double value of the
two others. Each material, chemical or solvent will besides a three dimensional
solubility parameter, which will be the centre in a sphere, also have an action
radius, R0. The action radius is the area around the centre where the
material/chemical is able to be dissolve or in other way to interact with other
materials. The HSP can be set up in a three dimensional diagram and the
wetting possibility can in that way directly be observed by studying if there are
overlap by the spheres. In Fig. 2, two spheres are set up besides each other.
108
They are not able to cover each other and the materials behind them are not able
to wet each other.
RED = Ra/R0,
Ra is the distance between the centre of the two spheres and R0 is the centre of
the material, which shall be compared with the other sphere. If the RED number
is <1, there is compatibility/solubility between the two compared materials. If
the RED number is >1, there is no compatibility/solubility between the two
compared materials.
As mentioned above the surface tension, , can be measured or be calculated
from the HSP. Different researchers have found the correlation between the
surface tension and the HSP. The most used is
(4)
109
110
A fresh anodized surface does not need any treatment if it has not been
touch before bonding.
5. Adhesive Bonding
An adhesive bonding shall be constructed so the forces on the bonding is the
best possible for the bonding. An adhesive bonding like a shear stress and
dislike a direct tension. In Fig. 3 the two types of forces are sketched.
Fig. 3 The figure shows to the left, the shear force and to the right, the tension [3].
6. Structural Adhesives
Adhesives used for aluminium will be a structural adhesive. With a
structural adhesive means an adhesive with a high strength. The perfect bonding
will be a bonding where the bonding will have a strength which is the same as
the aluminium. The bonding shall not break in interphase between the
aluminium and the adhesive but in the aluminium or in the adhesive layer. If the
break will be in the interphase, the wetting has normally not been good enough,
or the cleaning of the surface has not been sufficient, or the forces in the surface
of either aluminium or the adhesive have not been sufficient.
Structural adhesives can be many different polymers. All of them are
polymers which will cure up under cross linking. All of them will have an
opening time. The opening time the time from the adhesive has been added to
the product until it will start the cross linking - depends on the curing system
and can be regulated to be short or long depending on the application of the
product. There are of course limitations for how short and how long the opening
time for the adhesive can be. The structural adhesives have no green strength,
which means that for several applications it is necessary to use a tool to fix the
products until the adhesive has obtained the strength. The structural adhesives
are epoxy adhesives, polyurethane adhesives, acrylic adhesives or other
adhesives which can cross link. In Fig. 4 the curing reaction of an epoxy
111
adhesive is shown. There are many different types of curing reactions for epoxy
and the shown is only an example. In Fig. 5 the reaction of a polyurethane
adhesive is shown.
112
rate of crosslinking the higher strength will be obtained. Opposite will a low
density of reactive groups for crosslinking gives a lower strength and a higher
flexibility of the adhesive. Bonding with a high strength and stiffness is
necessary under a shear stress and especially if a peeling is a risk. If the load
will be a dynamical load the bonding need to have some flexibility as well as
strength and the adhesive need certain flexibility.
The epoxy adhesives can general obtain the highest strength, because I most
epoxy adhesives the density of aromatic groups is high. Most of the epoxy
resins are made on basis of bis-phenol A, which contains two aromatic groups.
In Fig. 6 the formation of the epoxy resion bisphenol-A-diglycidylether
(DGEBA) is shown. Aliphatic groups can be added and they will change the
strength and the flexibility of the cured adhesive. Adding aliphatic groups can
also be used for changing the opening time of the adhesive, because a longer
aliphatic chain will decrease the density of reactive epoxide groups and there
fore change the reaction time. Epoxy adhesives are normal two component
systems. With mixing the two reactants the curing process will start. Increasing
temperature will decrease the reaction time. With changing in reaction
temperature the curing process the cross linking process can change and the
properties of the adhesive can change.
The polyurethane adhesives can as well obtain a good strength. Most of the
polyurethane adhesives can obtain a good flexibility, especially if they are made
on basis of an aliphatic isocyanate. The polyurethane adhesives can be one- or
two component systems. Two component systems will start the reaction, when
the isocyanate and the polyol the two active components are mixed with
each other. Again increased temperature will decrease the reaction time and
increased temperature can change the properties of the adhesives. One
component polyurethane adhesives will start the reaction under influence of
humidity in the atmosphere. The humidity will react with one of the
components in the adhesive and make it active for reacting with the other
component. If the humidity is low the reaction will not take place as expected.
113
The acrylic adhesives cure up with a radical reaction over a double bond. For
initiating the reaction an initiator system shall be reacted. This can happen under
influence of adding the initiator or under adding energy f.ex. with UV irradiation
for starting the reaction. The acrylics can be one or two component adhesives.
The rate of cross linking depends on the curing system. The rate of the cross
linking depends also on the temperature and on the time in which the system will
react.
Fig. 7 The HSP plot of DP460 from 3M. DP460 is an epoxy based adhesive.
114
The HSP can be measured for the different adhesives. In Fig. 7 the HSP plot
of the epoxy based adhesive DP 460 from 3M is shown and in Fig. 8 the HSP plot
of the polyurethane adhesive Sikaflex from Sika Industries is shown.
From comparing the HSP for an adhesive with the HSP for a surface
aluminium or anodized aluminium it can be foreseen if it can be possible to
wet the surface with the adhesive and if it can be possible to make a reaction
between the adhesive and the surface. Fig. 9 shows the HSP plot for an unsealed
anodized aluminium surface.
7. Results
For several systems combined with anodized aluminium and a structural
adhesive the strength and the flexibility has been measured. The adhesives used
in the experiments have been commercial adhesives and the anodizing process
has been carried out either in a laboratory or in a commercial system. For the
laboratory made anodizing the process has been made with parameters and
chemicals as near the commercial systems as possible.
115
The strength of the bonding depends of the adhesive and of the surfaces. In
a group of experiments where the aluminium alloy 6063 with a tensile strength
of 100 MPa was used the strength of the bonding on plate of 20 mm and the
bonding area of 20 x 20 mm2 have for anodized unsealed aluminium been
measured to app. 7400 N with DP460 and to 3700 N for the flexible Sikaflex 360HC. The strength of bonding with DP460 (epoxy resin) can be up to 10000
N on the area 20x20 mm2.
8. Conclusions
It is possible to bond aluminium with an adhesive and obtain a good
strength and flexibility. The adhesive bonding is best for a shear stress and
sensitive for especially a peeling. The adhesive used will always be a structural
adhesive. Epoxy adhesives, polyurethane adhesives and acrylics adhesives are
the most common used adhesives for aluminium.
For obtaining a good bonding the surface of the aluminium needs to be
wetted by the adhesive. If a surface shall be wetted by an adhesive the surface
tension of the surface need to be higher than the surface tension of the adhesive.
For estimating the possibility of wetting of the surface with an adhesive the
HSP can be used as a tool.
Adhesive bonding give a better strength for anodized aluminium if the
bonding is made on unsealed anodized aluminium. The unsealed anodized
aluminium is porous and will give a better mechanical interlocking together
with a physical adsorption in the interphase between the anodized layer and the
adhesive. Use of an adhesive as a primer can increase the strength in the
interphase. If it is possible a primer or an adhesive, which can give a chemical
reaction between the unsealed anodized aluminium, shall be used for obtaining
the highest strength of the bonding.
Aalborg University,
Esbjerg Institute of Technology
Esbjerg, Denmark
e-mail: bks@bio.aau.dk
REFERENCES
1. H a n s e n Ch. M., Handbook of Solubility Parameters A Users Handbook, CRC
Press, 1st edition, 2000.
2. O e r t e l G., Polyurethane Handbook, Hanser Publishers, 2nd edition, 1994.
3. P e t r i e E. D., Handbook of adhesives and sealants, McGraw-Hill, 2000.
116
INDUSTRIAL APPLICATIONS OF
THE PNEUMATIC MUSCLES
BY
1. Introduction
Although the pneumatic muscles have been conceived since 1930 by S.
Garasiev, a Russian inventor, only for a few years, they have been used in
different applications. Lately, pneumatic muscles became a better choice than
present day electric or other drives. They are usually used in factory floor
automation and nowadays, in industrial robotics as a main motion power source.
The current paper will focus on the pneumatic muscle with its applicability
in the industry. These actuators are usually cylindrical in shape and they are
composed by an interior inflatable tube typically made by neoprene rubber
wrapped in a multilayer tissue typically made by nylon. The tube, under the
action of compressed air, increases its diameter and decreases its length; the
stroke resulted is in direct relation with the pressure of the compressed air
passed into the muscle.
118
The utilization of the pneumatics offers many advantages, the most important
being the low weight and the inherent compliant behavior of its actuators.
Compliance ensures a soft touch and safe interaction. In contrast with
pneumatic muscle actuator, hydraulic and electric drives have a very rigid
behavior and can only be made to act in a compliant manner through the use of
relatively complex feedback control strategies. [1]
2. Pneumatic Muscle
2.1. History
Pneumatic muscle is an actuator system based on an inflatable and flexible
membrane operated by pressurized air (Fig. 1).
119
2.2. Generalities
The pneumatic muscles have many characteristics that make them being
easy to use and with great performances. Some of these characteristics are:
shock - absorbing, adjustability, simulating capability, storage capable,
safeness, lightweight, natural compliance and shock resistance.
Favourable response to commands, known as compliance is in direct
relation to air compressibility, and hence the pneumatic muscle can be
influenced by controlling/adjusting the command pressure.
The function principle consists in the fact that, under the action of
compressed air, the pneumatic muscle, which is blocked at one end, shortens its
lengths and expands its diameter. As the volume of the internal tube increases
due to the increase in pressure, the actuator shortens with a certain stroke.
Pneumatic muscle construction is based on an interior tube, made from
neoprene rubber wrapped in braided sleeves made of nylon with strengthening
and protecting role. The braided sleeves act to constrain the expansion for
maintaining the cylindrical shape. As we can see in the figure 1, the angle of the
enveloping tissue, denoted by , is one in relaxed state and differs in contracted
state. It has the value of 25.4 in the relaxed state of the muscle and of 54.7 at
maximum contraction. [4]
In Fig. 2 is presented the working principle of a pneumatic muscle.
F = p
3 cos 2 1
d2
2
4
1 cos
,
where p is the working pressure and d the interior diameter of the pneumatic
muscle. Upon completion of the maximum stroke the developed force is equal
to zero. Equation (1) allows plotting of the graph featuring the force developed
120
by a pneumatic muscle versus the enveloping angle and feed pressure (Fig. 3)
[5].
The advantages of the pneumatic muscles utilization are: power to weight
ratios in excess of 1 kW/kg; a varying force-displacement relation at constant
gas pressure, an adjustable compliance; the absence of friction and hysteresis;
the ability to operate at a wide range of gas pressures, and thus to develop both
very low and very high pulling forces; the possibility of direct connection to a
robotic joint; cheaper to buy and install than other actuators and pneumatic
cylinders; smooth and natural movement; fast -full contraction.
Disadvantages are: the force which can be applied is only tensile in nature;
its total displacement is only about 20% to 30% of its initial length; friction
between the netting and the tube leads to a substantial hysteresis in the forcelength characteristics; rubber is often needed to avoid the tube from bursting;
rubber deformation will lower the force output of this type of muscle up to 60%
[8].
a
b
Fig. 4 a WAP-1(source www.androidworld.com) ; b - Humanoid robot manufactured
by Shadow robotic company (source www.shadowrobot.com)
121
a
b
Fig. 6a Artificial limb developed at the bio robotics Lab, University of Washington.
b Shadow Leg (source www.shadowrobot.com)
122
Another application for the pneumatic muscle is the leg made for Shadow
Robot Company by David Buckley from North Carolina A & T University.
(Fig. 6b)
The dexterous hand was developed by the Shadow robotic company. The
hands operate just like human hands with five fingers. It contains an integrated
bank of 40 Air Muscles which make it move. The figure 7 illustrates this fact.
Caldwell used 18 small McKibben Muscles to power a dexterous fourfingered manipulator. Hannaford built an anthropomorphic arm, having fifteen
McKibben Muscles. The Soft Arm, developed by Bridgestone Co. has a
shoulder, an upper arm, a lower arm and wrist, and a useful payload of
maximum 3 kg. Yoshinada used hydraulically actuated McKibben Muscles to
power an underwater manipulator [1].
New Application Areas are: Simulator Technology, High Speed cutting
processes, Aeronautical Technology, Wood-working, Metal-working, Medical/
Biomedical, Mobile Applications, Building / Construction, Mining, Process/
Water, Amusement/ Recreation, Medical / Surgical / Hospital
4. Conclusions
1. Even though pneumatic muscles are not capable of offering an
extremely wide range of operations, in the case of humanoid robots they offer a
wide range of possibilities. Their low assembly weight and high power-toweight ratio make the pneumatic muscles to be considered for use in mobile
robotics [1].
123
REFERENCES
1. D a e r d e n F., L e f e b e r D., The concept and design of pleated pneumatic
artificial muscles, International Journal of Fluid Power, 2, 3, pp. 4150, 2001.
2. M a r c i n i n J., P a l k o A., Negative pressure artificial muscleAn
unconventional drive of robotic and handling systems, Transactions of the
University of Koice, pp. 350 354, Riecansky Science Publishing Co, Slovak
Republic, 1993.
3. B a l d w i n H. A., Realizable models of muscle function, Proceedings of the First
Rock Biomechanics Symposium, pp. 139148, New York, 1969.
4. S c h u l t e H. F., The characteristics of the McKibben Artificial Muscle, The
Application of External Power in Prosthetics and Orthotics, pp. 94115,
National Academy of SciencesNational Research Council, Publication 874,
Lake Arrowhead, 1961.
5. D a v i s S.T., C a l d w e l l D.G., The biomimetic design of a robot primate using
pneumatic muscle actuators, 4th International Conference on Climbing and
Walking Robots CLAWAR, 2001.
6. K e n n e t h, K u K.K., B r a d b e e r R., Static Model of the Shadow Muscle under
Pneumatic Testing, 2006
http://www.ee.cityu.edu.hk/~rtbrad/muscles%20riupeeec%202006.pdf
7. D e a c o n e s c u A. , D e a c o n e s c u T., Contribution to the Behavioural Study
of Pneumatically Actuated Artificial Muscles. 6th International Conference of
DAAAM Baltic Industrial Engineering, Tallinn, Estonia, Vol. 1, pag. 215-220,
2008
8. * * * Air Muscles 2008 [Online] Available at:
http:// www.techalone.com 23/01/2010] [Accessed 27.12.2009]
http:// www.festo.com [Accessed 27.12.2009]
http://www.shadowrobot.com [Accessed 27.12.2009]
124
MIHI HORODINC
Abstract. This paper presents some theoretical considerations and experimental
results on computer aided research of manufacturing systems (e.g. machine-tools)
using the evolution of the absorbed electric power (EP) of the driving motors. Based
on an original computer aided data acquisition and processing method, the electric
power evolution monitoring (and its components, especially the EP real part) should
be now a very useful tool in experimental research (kinematic chains condition,
working processes loading, and so on). In a new approach the numerical describing
with high resolution of real EP evolution can be used also to evaluate the
performances in dynamic conditions, especially using the evolution in frequency
domain (by power spectral density analysis). Some phenomena in kinematic chains
was observed with a certain scientific priority, using the real EP monitoring
capabilities, such as the behaviour of driving belts and the behaviour of the elastic
system of the rotor on the electric driving motor. The real EP monitoring can be
used to supervise any other electric actuated equipment.
Key words: manufacturing systems, experimental research, real electric power,
computer aided monitoring, data processing.
1. Introduction
The working processes (WP) on manufacturing systems (MS) or any other
working equipment needs mechanical energy, usually delivered by an electric
driving motor (EDM) and transported by a kinematic chain. By mechanical
loading point of view the behaviour of MS is mirrored in the real part of electric
power (EP) evolution (active electric power, AEP). The EDM seems to be a
very appropriate sensor of mechanical loading, useful in MS and WP
monitoring and diagnosis or even in WP active control. The state of the art [2],
[3] indicates that generally for monitoring it is used just the electric current
126
Mihi Horodinc
absorbed by the EDM, but for MS driven with alternating voltage EDM, the
best characterization of loading is done using AEP evolution. In some papers
[4], [5] the AEP it is taken into account, but the research resources are not
completely exploited. This paper presents some experimental research results
based on a new approach on computer aided AEP monitoring for MS powered
by EDM.
2. EP Monitoring Principle on EDM and Experimental Conditions
Fig. 1 describes a computer assisted monitoring bench test developed for an
asynchronous EDM supplied on a three phase 50 Hz frequency sinusoidal
voltage, symmetrical network (3380V). A single phase is used for computer
data acquisition (instantaneous current i(t) IC and voltage u(t) IV in time
P=
1T
3 T
p(t ) dt = u (t ) i (t ) dt = 3 U I cos ,
T0
T 0
where: T=20 ms is the period (reciprocal of the frequency f), U - the effective
(root mean square, RMS value) voltage, I - the effective current, - the angle of
127
phase (AP) between current ant voltage. Using the numerical values u[tk] and
i[tk] of the voltage and the current, the numerical evolution of the AEP can be
described by calculus (based on Eq. (1), as average value on the period T) with:
(2)
P[tl ]
3 n(l +1)
3 n(l +1)
(u[tk ] i[t k ] t ) = (u[tk ] i[t k ]) ,
T k =nl +1
n k =nl +1
where: P[tl] is the numerical value of the AEP (tl=lT, l=1,2.m, m values for a
registration of mT total duration), n=T/t, t is the reciprocal of the sampling
rate for u[tk] and i[tk], tk=kt. If t0 then in Eq (2) the symbol becomes =.
The sampling rate of AEP is equal with frequency f value (50 s-1). Using the
effective values U[tl] and I[tl] (the amplitudes of the voltage and current divided
by square root of 2), the AP and REP (reactive electric power Q) are given by:
(3)
[tl ] arccos
P[t l ]
3 U [tl ] I [tl ]
and
The experimental research was done on a Romanian lathe SNA 360 (used as
MS, see Fig. 1) with the kinematics of the main shaft gearbox described in Fig.
Fig. 2 Overview on the kinematics of the SNA 360 lathe main shaft
gearbox used as MS in experimental research.
2. An asynchronous EDM with 5.5 KW and 1440 rpm (speed in rotations per
minute) is used to drive it. Two different configurations of the gearbox were
used (see Kd1 and Kd2), each one is able to actuate the main spindle (shaft) Ms
on 1000 rpm or 1600 rpm.
128
Mihi Horodinc
129
There is an interesting behavior of the EDM mirrored in AEP evolution, see the
regions I and II. There are two free responses on 5.2 and 10.8 Hz frequency, due
to the elastic system associated to the rotor (mechanical inertia, magnetic field
stiffness and air friction damping).
A second experiment describes (Fig. 4) the mirroring of the clutching
dynamics of the gearbox (on Kd1). In A the MC3 clutch is engaged, the gearbox
start the motion and the AEP consumption increases during a transitory regime
B (with a 4.16 Wh average electrical energy stored as kinetic energy). The
peak B1 indicates that the BT1s driving belt start the sliding (the dynamic
torque is bigger than the friction torque between the belt and the pulleys). In B2
the sliding disappears and immediately after (in C) starts the steady-state regime
on 1000 rpm. Here the absorbed AEP is used to cover the waste of mechanical
energy (dry and viscous friction).
The AEP evolution seems to be noisily, but it will be proved later (Fig.7)
that it is very useful for gearbox diagnosis. In D the clutch is disengaged (by
switching-off the electrical supply) so the AEP suddenly decreases. The level of
power in E is for a short time lower than in A, the difference (183 W) describes
the viscous friction in MC3 clutch.
130
Mihi Horodinc
available kinetic energy from gearbox in electrical energy (with a total amount
of 0.464 Wh) delivered on the electrical network. After that the AEP increases
with an overshoot in C, the generator becomes again electric motor. In D the
steady-state on 1000 rpm is installed. In E the MC3 clutch is switched-off, so
just the shaft I is driven. The AEP consumption in steady-state idle regime is
smaller than in Fig.4 because of the gearbox heating (the viscosity of the
lubricant so the viscous friction decreases).
131
132
Mihi Horodinc
behaviour of the driving belt from BT1. The belt generates a very strong
variable component of AEP, with the fundamental frequency on 5.4 Hz (first
peak on Fig. 7) and 6 harmonics (see the peaks 4, 6, 9, 11, 14 and 18). It is
generated by the belts stiffness variation (the belt is a little bit damaged, it has a
small tear). The fundamentals frequency fFB is given by:
(4)
f FB = f EDM
D1
Lc
= 24.84 Hz
125mm
1812mm
= 5.38 Hz .
where: fEDM is the rotation frequency of the EDMs rotor, D1- the diameter of
the belt pulley of the EDMs rotor, Lc- the belts length. A new experiment was
done, with PSD analysis on IEP evolution during the steady-state regime, with
all the clutches disengaged, see Fig. 8. The BT1s belt behaviour is better
indicated with the harmonic F and the harmonics H1,H2H6. Because the
mechanical loading of the EDM is smaller, the speed of rotation is bigger so the
frequency in Fig. 8 increases. The first harmonic is bigger than the fundamental
because of the EDMs rotor elastic system resonance.
5. Conclusions
133
REFERENCES
1. B i r a n A., B r e i n e r M., Matlab for Engineers, Addison-Wesley Pub. (SD), May
1995, ISBN-13: 978-0201565249.
2. D i m l a E., D i m l a., Sensor signals for tool-wear monitoring in metal cutting
operations a review of methods, Int. J.of Machine Tools & Manufacture, 40, 8,
(2000), pp. 1073-1098, ISSN:0890-6955.
3. D o n g-W o o, C., S a n g J. L., C h o n g N a m C., The state of machining process
monitoring research in Korea, Int. J. of Machine Tools & Manufacture, 39, 11,
November, 1999, pp. 1697-1715.
4. H w a-Y o u n g K., J u n g-H w a n A., Chip disposal state monitoring in drilling
using neural network based spindle motor power sensing, Int. J. of Machine
Tools & Manufacture, 42, 10 (2002), pp 1113-1119.
5. M i n g L., T e t Y., S a e e d R., Z h i x i n H., Fuzzy control of spindle power in
end milling process, International Journal of Machine Tools & Manufacture, 42,
14 (2002), pp. 1487-1496, ISSN:0890-6955.
6. H o r o d i n c M., Utilizarea parametrilor energetici n monitorizarea, diagnoza i
conducerea sistemelor de prelucrare, Ed. Performantica, Iai, 2004.
1. Introduction
The engineering methods based on computer permitted to develop a new
type of processional transmissions with multi-couple meshed teeth, which, from
the technological point of view, can be manufactured by means of a new
method of processing conical teeth with convex-concave profile.
It appeared the necessity of elaboration of new profiles adequate to the
spheroid-spatial motion of the gears, which would ensure high performances to
the processional transmission.
Considering the necessity of achieving the transfer function continuity and
gear multiplicity some objectives were taken into account. One of them is the
integrated method of design, modelling and simulation using powerful means of
136
Ion Bostan et al
2. General Information
The ever-growing requirements, especially, concerning the bearing capacity,
the kinematical accuracy and the kinematical possibilities, impose the necessity to
develop a new type of planetary gearing with distinct performances. The gearing
improving is one solution of the problem. Novicov - Wildhaber, Symarc and other
gearings have increased considerably the bearing capacity of gear. Another
direction of developing gears is the design of new types of mechanical gears.
The creative search of designers has crowned with the elaboration of a new
type of gearing harmonic gear. W. Musser, an American engineer, patented the
action principle of the harmonic gear in 1959. Starting that year W. Musser
patented a big number of diverse constructive diagrams for harmonic gears (teeth,
friction and tapped gears) and couplings, and demonstrated the possibilities of the
new construction principle of mechanical gears. Thus, in 1961, the harmonic
drive was produced at one of the American companies, for the first time at
industrial scale. Harmonic drive are compact and possess increased bearing
capacity; they provide high kinematical accuracy and possibility to transmit
motion in sealed mediums, which is one of the basic advantages of harmonic
drive. As their disadvantages we can mention reduced reliability of the flexible
element (and, thus, of the gear, on the whole), reduced working order at high
speeds, and also some technological difficulties.
By the end of the 70s Prof. Ion Bostan designed a new type of gears, new in
principle, - processional planetary gears with multiple gearing (B o s t a n, 1991).
Over 20 years, research was carried out comprising the total range of problems
from the idea to the implementation: fundamental theory of the precessional gearengineering calculation procedures-Know-How manufacturing technologiesapplications. The research results have been published in over 450 scientific
papers, in about 150 patents, in 3 monographs and in one design guidebook.
The absolute multiplicity of the processional gear (up to 100% of teeth pairs,
geared simultaneously, compared to 5-7% in classical gears) provides increased
bearing capacity and kinematical accuracy, small dimensions and mass. In
addition to the above said, extended kinematical possibilities (83600 compared
to 79300 in sinusoidal gears), reduced acoustic emission and solution of all
technological problems, as advantages, open new perspectives for precessional
planetary gears utilisation in various fields of mechanical engineering (B o s t a n,
1991). On the whole, processional planetary gears can be divided in two basic
groups:
- power processional planetary gears;
- kinematical processional planetary gears.
137
2 7 4 5
(1) i =
Zg1 Za
Zb Zg 2 Zg1 Za
where:
Z g1 , Z g 2
are
number of a roller the
crown satellite g 1 and g 2 ;
Z a and Z b is represented
number of a teethes the
cog-wheels a and b .
138
Ion Bostan et al
(2)
m
Z1E
=
(k1md1m k2md2m )
k1m2 + k2m2 +1
2
m2
m2
(k1md1m k2md2m )2 + (k1m2 + km2
2 + 1) (RD d1 d2 )
Z =
,
k1m2 + k2m2 +1
m
1E
Where:
m m
m
m m
m2
m
X1D
X
X
Y
Y
Z
X
+
+
m
1D 1D 1D 1D 1D 1D
k1mY1Dm + Z1D
(
);
m
m
k1 =
; k2 =
m
X1D
m m m
m
m
Z1D
X1D Y1D Y1D X1D
d =
m
1
m
RD2 cos X1D
m m m m
X1D Y1D X1D Y1D
m
2
; d
( R cos + d Y ) .
=
2
D
m m
1 1D
m
X1D
The 3D model of the central wheel (fig. 3, b) was designed by using CAD
Autodesk Inventor (5) (B o s t a n et al, 2007).
a.
139
b.
The dynamic model has been created on the basis of rigid model. As the
initial data has been specified speed on input shaft [deg/sec] and the moment of
torsion on the output shaft [Nm].
Kinematic joints fig. 5, have been enclosed according to movements in
gearing. The crankshaft and the case of a reducer by means of cylindrical roller
bearings are connected by a cylindrical joint to an opportunity of a selfcentering. A crankshaft and the block satellite by means of two tapered bearings
with pair rotation in points of the appendix of loading on bearings (point-line),
pair rotation (revolution) rollers on an axis of the block satellite and 3D contact
of rollers to a mobile and fixed wheel gear.
140
Ion Bostan et al
At the following stage has been executed the kinetostatics analysis with
calculation and simulation of total loadings in gearings.
141
b.
a.
The vertical line on the diagram (Fig. 7) shows the precession phase
reported symbolically to the gearing time of the pinion tooth with the central
wheel tooth (as reference the time was taken, because as CAE input parameters
the angular velocity was introduced). On the diagrams one can see the
distribution of load among the teeth which corresponds to the respective
precession phase (blue colour distribution of load on the active profile of
central wheel tooth, red colour per one precession cycle). It is possible to state
from the diagrams that despite the precession phase, the load among teeth is
uniform.
Due to load transmission from the input shaft to the output shaft by a big
number of teeth couples geared simultaneously = ( Z 4 1 ) the normal
forces at teeth contact are much smaller as in the classical transmissions with in
volute transmission.
4. Conclusions
1. Taking into account the fact that in processional planetary gearings the
( Z4 1 )
teeth couples transmit the load simultaneously we can conclude that
2
the bearing capacity of the processional gear is much bigger than that of the
classical in volute gear (in which only 5...7 % of wheel teeth gear
simultaneously).
142
Ion Bostan et al
REFERENCES
1. B o s t a n I. Precessionnye eredaci s Mnogoparnym Zacepleniem.(Ed.) tiina, 356p.
1991. ISBN 5-376-01005-8, Chiinu.
2. B o s t a n I., D u l g h e r u V., G r i g o r a ., Planetary, precessional and
harmonic transmissions, Bucureti - Chiinu, 1997, 198 p.
3. B o s t a n I., I o n e s c u F l., D u l g h e r u V., A. S o c h i r e a n u., Kinetostatic
Analysis of the Sphere - Spatial Mechanisms by using 3D-Models and Simulations. Meridian Ingineresc, Revue of the Technical University of Moldova
and Moldavian Society of Engineers, 2, ISSN 1683-853-X, (2004). pp. 59-61.
4. http://www.solid-dynamics.com (Motion Inventor)
5. www.autodesk.com (Autodesk Inventor).
1. Introduction
The fluidic mills are the single equipments that permit the dry grinding and
the get of an ultra fine granulometry (1-25 m), indifferent by the product
resistance and by the hardness of the product's crystals.
The grinding is produced by the collision between the particles that are
accelerated at very high speeds in flows and gas jets, or by their impact with
solid surfaces (fixed or mobile) from the interior of the milling room.
We consider important to signalize that the real or as a rough guide speeds of
the entrainment and grinding particles' jets are considered secret and can not be
found in the literature. At this moment exist in the world 7 fabricants that
produce fluidic mills. In fact, there are many societies that commercialize
fluidic mills, but they are procuring the equipments offered from one of the
mentioned fabricants.
All the fluidic mills fabricants consider strictly confidential any data
concerning the sizing and the calculus of these equipments from the point of
view of the fluidic mechanics, refusing to give any theoretical or experimental
value concerning the geometry of the fluidic nozzle the jets' speeds and
particles' speeds or values bound of the fluidic grinding process.
144
3. Practical Applications
3.1. The Optimization of the StarJet and CosmoJet Mills
The APTM Enterprise realized a great number of researches in order to be
able to optimize the activity of the produced spiral jetmills, presented in a great
number of scientific papers [1] - [10] and in a doctoral dissertation.
145
146
Fig. 2 Example of the optimized technological flux for the fabrication of the
cosmetic micronized powders (make-up, cheek powder).
practice in the enterprise APTM, for a new jetmill family, named "StarJetMillennium" (Fig. 1). An optimized technological process is shown in Fig. 3)
147
The first three typo dimension of this family (4" - 8" and 12") applied an
important number of the optimization algorithms of the milling process that
form the research object of this research. It results implicitly the fact that these
algorithms (fig. 2) where already verified in practice for moreover micronized
powders categories (that are partly confidential).
Fig. 3The example of the spiral jetmills' optimization algorithm applied for
the definition of the parameters recomended in industry and for the qualitative
differentiation of the cosmethic powders.
148
149
We can observe the fact that after the micronization, all the materials
are grinded at granulometry about 15-30 m, with a remarkable
uniformity. We make the assignation that in this paper where presented
only the non confidential elements that show the optimization.
We present (fig. 5) an example of a technological process for the getting of
cosmetic powders micronized with optimization algorithms. In fig. 6 and 7 we
present micronized powders realized with optimization algorithms
(visualization on microscope with phase contrast). We can affirm that the
150
Mills from the family StarJet optimized with algorithms similar with the
presented algorithms were delivered and perfectly work from the year 2002, in
many enterprises for decorative cosmetic products of the societies Avon (USA),
Natura (Brazilia), Belcorp (Columbia), EverBilena (Taiwan), Swan Cosmetics
(Mexic), Cosmetica (Canada), GammaCroma (Italia), Hellenica (Grecia),
GiPiccos (Italia), etc.
The up presented algorithms, partially used in production, ask for verifications and ulterior confirmations, forming the object of some future studies.
4. Conclusions
1. The research from what doctoral dissertation introduced an original threedimensional model of the fluidic running in the spiral jetmills that describes
very veridical the complex jet in this kind of mills.
2. The results are obtained as suit of a very large gamut of theoretical and
experimental researches for the determination of the jets' speed.
3. The optimization algorithms were verified on models and in realty on 37
mills having different typo dimensions, in many fields (cosmetic,
pharmaceutics, and metallurgies) in above 20 countries.
151
APTM Enterprises,
Lugano, Swiss
e-mail: i@fulga.com
2
POLITEHNICA University of Bucharest,
Department of Machines and Production Systems
Bucharest, Romania
e-mail: eugen_strajescu@yahoo.com
REFERENCES
1. F u l g a I . , H u d d l e s t o n K . , Cosmetics cross an even finer line, Powder
Reporter, vol. 1, UK, 1998.
2. F u l g a I . , S t r j e s c u E . , S a n d u I . G h . , Experimental Model of the fluidic
Mills with Spiral Jets, Academic Journal of Manufacturing Engineering, Editura
Politehnica, Timioara, Romnia, 53 - 57, ISSN 1583 7904 (2006).
3. F u l g a I . , S t r j e s c u E . , Theoretical and Experimental Contributions
Concerning the Logistics of the Research Systems of the Fluidic Mills with Spiral
Jets, Buletinul Institutului Politehnic, Iai, Romnia, Construcii de Maini, LII
(LVII) , 5, 313 - 320, ISSN 1011 - 2855 (2007).
4. S t r j e s c u E . , F u l g a I . , The Visualization of the Fluidic Fluxes in the Mills
with Jets, Industrial Engineering Journal RECENT, 8, 3b (21b), 361 - 366, ISSN
1582 - 0246. Braov, Romnia (2007).
152
1. Introduction
The control system a wind turbine is equipped with is designed to reduce
dynamic load on its blades, to increase turbine reliability and operational safety.
This system is one of the most important developments meant to consistently
improve low power horizontal axis wind turbines.
Further to the analysis of control and protection system design models used
for low power horizontal axis wind turbines (HAWT), the best constructive
solution was chosen, which consists of the vertical tilting of the wind turbine.
The tilting begins when the wind speed exceeds the speed limit beyond which
the turbine safety may be jeopardized. This solution meets the system
requirements: it is reliable, it does not include electrical drive components, it
has no built-in electronic devices and it does not require the existence of a
hydraulic drive system.
154
2. Constructive-Operational Description
A wind power plant turns wind energy into mechanical power. The chosen
solution refers to very low power (500 1000 W) wind turbines.
When the wind speed does not exceed a certain speed limit, the rotor
assembly does not tilt. The propeller rotates at an angular speed, which depends
on the wind speed and on the mechanical output load.
When the wind speed reaches or exceeds the speed limit, the rotor assembly
tilts and the propeller changes its rotation plane.
Damping
device
HAWT
Protection
mechanism
Balancing
device
Tilting
device
155
Fig. 2 shows the Simulink block diagram of the vertical tilting protection
mechanism.
Weld2
Revolute 1
Body2
Body1
B
CS2 CS1
CS4
CS5
CS1
CS1
CS3
CS2
Scope 3
Body Actuator1
Link 2
Revolute 3
Body Actuator
Env
B
Ground 2
Machine
Environment
360
Constant
Scope 4
Body
Joint Sensor 2
CS1
-K -
Abs
Product 7
Scope 2
|u|
Gain
Scope
B
Ground 1
Product 2
Product 3
[1x3]
Constant 1
[A]
Goto
Scope 6
sin
AxRo/2
Product
Trigonometric
Function 1
0.3
Scope 5
Constant 4
Signal 1
Signal 2
Signal 3
Signal Builder 1
Product 4
-C-
Scope 1
Integrator
16
lambda /\r
Product 5
Joint Actuator
1
s
Product 1
Product 6
-C3r Constant 3
65
Constant 2
0
[A]
From
cos
u
Trigonometric
Math
Function
Function
Fig. 2 The Simulink block diagram of the vertical tilting protection mechanism.
156
157
Fig. 6 shows the set of signals used to simulate wind speed variation.
The second signal is employed to achieve static balance, where the wind
speed is thought to have the same value as the maximum admissible rating
speed.
Using the ramp signal (signal 3), the system response is tested, that is the
rotor tilting, when the wind speed exceeds the maximum admissible rating
speed. Fig. 7 shows the axis inclination angle variation at the 3rd signal.
158
159
REFERENCES
1. B a n s a l R. C., B h a t t i T. S., K o t h a r i D. P., On some of the design aspects
of wind energy conversion systems, Energy Conversion and Management, 43, 16,
November 2002, pp. 2175-2187
2. B i a n c h i F. D., D e B a t t i s t a H., M a n t z R. J., Wind Turbine Control
Systems, Principles, Modelling and Gain Scheduling Design. Springer Verlag,
Heidelberg, 2007.
3. B i a n c h i F., M a n t z R., C h r i s t i a n s e n C., Power regulation in pitchcontrolled variable-speed WECS above rated wind speed. Renewable Energy
29(11), 19111922, 2004.
4. G r i m b l e M. J., J o h n s o n M. A., Advances in industrial control, Springer
Verlag London Limited
5. M o l e n a a r D. P., Cost-effective design and operation of variable speed wind
turbines, Delft University Press, 2003.
6. N o v a c P., E k e l u n d T., J o v i k I., S c h m i d b a u e r B., Modeling and
control of variable-speed wind-turbine drive-system dynamics. IEEE Control
Systems, pp. 28 - 38, August, 1995.
7. * * * 160 kW Wind turbine with Hydraulic Transmission at Schiedam Ref: 16013 or
34127
8. * * * Low-wind turbine with hydraulic blade control. Measuring program included
Ref: 8700106
9. * * * Wind turbine with hydraulic transmission, Patent number WO03098037.
10. * * * Water current turbine, Patent number WO2005103484.
11. * * * MATLAB Simulink http://www.mathworks.com/products/simulink/
12. * * * Contract no. 21-047/2007 financed by National Centre for Programme
Management from Romania.
160
1. Introduction
To move heavy masses on small distances with lover frequencies the author
recommends one of the following hydraulic charts, see Fig. 1.
In both variants the working tool is the linear hydraulic engine CH, with the
active surface S and volume Vo, which moves the mass M.
For the first variant the step of the move is programmed at the level of the
pump, through the volume/course V. We will admit that for a complete course
the pump send to the hydraulic engine the oil volume V. The oil rich the pump
through the orifices A and B.
In the case of the second variant the flow regulator RD controlled the oil
flow, independent to the working pressure for a specific time, when the
distributing valve is on. The quantity of oil which enters in to the linear
hydraulic engine is controlled in the time by the time stepping system C (t).
162
a)
b)
Fig. 1 Hydraulic charts for lower frequencies and small flow generators.
The mathematical models presented in the following rows are equal satisfied
for both variants.
2. Mathematical Models for the Stationary Regime
For the engine CH which works on the strength F, when a small quantity of
oil V, ideal liquid, incompressible, theoretical we obtain a theoretical movement
of the mass, xT, with the mathematical expression:
(1)
xT =
V
S
For a real liquid, with the elasticity module E, we obtain a real movement, in
static regime, xP.
(2)
xR = xT 1 .
E
p=
F
.
S
163
Q = S v + a p +
M
V0 dp
.
E dt
dv
+ bv + F = pS .
dt
(6)
V = Q dt .
(7)
xRD = v dt .
In the last relations we have the following supplementary parameters: vinstant speed of the CH engine, a- linear coefficient of the lost oil which are in
direct ratio with pressure, t-time, xRD-real movement in dynamic regime.
To find the real movement, in dynamic regime, we must simulate the
equation on a computer program. We consider the same data like in the case of
the static regime, in advance we know M = 200 kg and the values of a and b
coefficients. When the V quantity of oil is inserted we presume that the system
is in balance. The time which the system is in charge with the V oil quantity is
established as well as we obtain a movement theoretical equal with that we have
on the previous step (t = 2 s). The simulating chart in the simplified way is
presented in Fig. 2.
164
The V part generates the oil flow which is introduced in the linear hydraulic
oil CH. We can look on the evolution of the following parameters: pressure p,
speed v, movement x.
We can observe that the speed is stable around the value of 0,003 mm/ min
and the final value of XRD is 0,995 mm.
Fig. 4 Piston movement after a period of nr. of steps generated by the pump.
165
Regarding the system is permanent under the pressure his dynamics is very
well. The system is a stabile one first order system.
The dynamic simulation, are we expected, is the most precise one, regarding
the fact that he take care of the specific oil gaps.
4. Conclusions
1. Regarding the system is permanent under the pressure his dynamics is
very well. The system is a stabile one first order system.
2. The aim is to show the realistic part of the theoretical simulation of one
system designed to move a heavy owen, 200 t, on a short distances, 50 mm,
with low frequency, one complete cycle in 24 hours for Mittal Steel - Galati.
166
1. General considerations
Drying is an effective, economic and ecological method for preserving
agricultural products, plants, fruits and vegetables. An optimal drying process
led to products with high nutritional potential, obtained with reduced specific
energy consumption.
For dryers with a capacity smaller than 20 m2 surface of tray, the price of
automatic driving system can double the cost of installation and thus leads to
reduced automatic management functions, with impact on the quality of final
products. These dryers can be easily transported to the place where is performed
168
harvest of products which require drying; for this reason it is necessary a high
level of energy independence and high reliability. [ 5] - [7], [9], [10].
For optimizing the drying process is still necessary to measure the mass of
water extracted during the drying process. In order to reduce production costs of
the dryer, the measurement is performed for a sample represented by one or
more boxes of dry suspended on a force transducer.
To simplify design, transducer should be arranged in the drying chamber at
high temperatures and high humidity. In papers [5], [6], [9] have been examined
these issues and therefore was developed a pneumatic force transducer,
unconventional, which is adapted to operating conditions of convective dryers
and requires only 63mW electrical power and 5mW power air power. [12] - [14]
Optimal management of drying processes requires online measurement of
the variation of water extracted from the dried bodies during a drying.
Constructive solution for measuring the change in weight of a column of tray is
technically complicated and expensive. Hence, it is used a weight measurement
of the change in middle column boxes, specifying that the drying process has a
similar evolution throughout the column height. It should be noted that the
temperature in the drying is in the range 50...80C, therefore, was decided to
use pneumatic sensors for measuring forces, because measured size, pressure, is
not influenced by changes of temperature. [4] - [6]
Convective dryers consume more thermal power and about 8% electrical
power, required for operating the fan and automation system. [8], [10]
This paper presents and analyzes the operation and performance of a
scheme for measuring on-line weight of boxes loaded with dried material using
a pneumatic force transducer working under sampled. The solution presented is
remarkable due to very low energy consumption and a reduced price. The
measurement system is coupled to dedicated leadership PLC automatic dryers.
They use basic concepts of low cost automation, minimize energy
consumption and not least are extremely efficient economically.
2. Pneumatic Force Transducer
Applying concepts of minimizing energy consumption, energy
independence and low cost automation system was designed an unconventional
pneumatic transducer which does not require special source of compressed air
consumption is very low, which simplifies construction, and hence the cost
transducer, and the total energy consumed is very small.
It was chosen a pneumatic system because of the pressure signal, that is
proportional to the measured weight, which is not influenced by environment
temperature of measurement, and accuracy is high.
The main element of the force transducer is pneumatic load cell, which
converts measured force Fmas in a force proportional pressure pmas.
169
Figure 1 shows a functional diagram for pneumatic load cell for the drying
process, which can be used for proving that force change occurs slowly and
processes take place in a single direction.
Force measuring Fmas(t) is applied on top of a rod (4) attached to the rigid
center (3) of an embossed flexible membrane (2) with effective diameter Def
and constant effective area Sef, mounted on a body (1). On the rod (4) is
attached a nozzle (5), with the diameter dd, which is based on a ball (6) with
diameter db, closing the chamber as air access to the outside through holes in the
membrane, rigid centre and rod.
The measurement chamber can be connected in parallel with an air capacity
Vad. Pneumatic circuit supply is done by RP variable air resistance and DP
distributor type 2x2. Pressure source must be pal 1,5 pmas max.
Pressure measured pmas (t) is applied to a converter p / U, which gives the
output voltage yF[1, 3] Vdc.
The steady, the balance of forces, pressure pmas (t) in the enclosure is:
(1)
p mas (t ) =
Fmas (t ) + Gem
,
S ef
170
(2)
ma (t ) RaTa
.
pmas (t )
The position of dynamic equilibrium of cell phone equipment load, if the air
distributor DP is closed, the volume of gas is constant Vm. Therefore,
thermodynamic equilibrium equation is:
(3)
dt
and
dma (t ) Vm + Vad
=
S RT
dt
ef a a
(5)
dFmas
dt .
(p
mas
171
(8)
1
M red (t ) = (Fmas (t ) + Gem ) ) .
g
Dmev exhaust mass flow (t) depends on the distance nozzle-ball h(t), the
pmas(t) and external pressure patm. Sev exhaust section (h, db, dd) is a truncated
cone with the tip in the centre of the ball and the base on the nozzle [3], where
Rh is hydraulic radius [3]:
(9)
2
2
2
Rh = 0.5 h + h d b d d + 0.25 d b 0.5d b
dma (t )
= Dmi (t ) Dmev (t ) .
dt
172
When the condition (6) is accomplished, the nozzle starts to rise and there is
an exhaust flow Dmev (t)> 0, which reduces ma(t) and pmas (t) begins to decrease.
Once it finds a stabilizing amount pmas (t), driving algorithm keeps ui = 1 still
400 ms, then ui = 0 and stops the air supply Dmi (t) = 0. In this period Dmev(t)> 0
until h(t) = 0.
173
4.
174
175
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
176
1. Introduction
In order to achieve sustainable development, which requires the use of
renewable energy, primarily to maintain the existing energy (fossil fuels) to a
suitable level, but also for environmental conservation, in the last 20 years has
178
179
180
181
those areas which are prone to failure. It can be stated that the method enables
the so-called "defect management", the monitoring of the infrared radiation
emitted by a technical system during normal operation often being the only
imaging method that enables the actual visualization of the notions of "fault"
and "inclination to failure".
Thermography is very often the only quick on-site investigation of a
facility. The areas which are prone to failure can be observed through a mere
scanning of the frontage of the device.
2.1.2. Ultrasonic technology. Ultrasound is applied differently, depending
on the nature, geometry and destination of the product, taking into accounts the
technical performance of the Probes Flaw system. Steel products are the ones
which are most commonly examined by means of the ultrasound technique.
Ultrasonic control can be differentiated and customized according to the
methods specific to steel products: metal, forgings, castings, welded structures
and products made of austenitic steel.
The ultrasound check reveals all the types of internal defects of welded
joints. The ultrasonic method can be used to determine wall thickness and the
number of deposited layers. This technology is applicable to all metals and
nonmetallic materials. Having a great penetrating force, the ultrasound enables
the verification of large sections. The application is only limited by the rough
structures which are highly heterogeneous. The devices are lightweight,
portable and autonomous, and the technology will ensure good results even in
on-site working conditions. The verification techniques, particularly the
immersion control, lend themselves to mechanization and automation.
The result is safe and immediate and it can accurately predict the location,
size and depth of the defects. From an economic standpoint, the ultrasound
method is much cheaper and more productive than the radiation technique in
those cases where the number of defects exceeds a certain limit [14].
2.1.3. X-ray technology. X-rays can penetrate a large number of different
materials, including composites. Usually, the images are obtained as shadow
variations along the X-ray propagation. Under normal operation conditions, the
X-rays are not able to reveal the cracks which have a parallel orientation with
respect to the rays.
The energy requirement is low and the safety prerequisites can be fulfilled
from a remote position. The X-ray sources can be small, air-cooled and easily
implemented for various applications. This method is able to detect the lack of
binder between the metal, as well as cracks and holes in the metal. Some
sources enable the detection of defects at less than 10 micrometres. This method
can be used during real-time inspection for quality control or shortly after the
blade of the wind turbine is set in motion. [13].
182
2.2. Discussion
The advantage of the thermographic method is that this technology is able
to produce an overall picture of the whole wind farm. A quick assessment can
be made by a user with some experience. The main problem with this method is
that the thermal images are limited to some electrical components that produce
excessive heat during operation. This can also be applied to defective
components if they move excessively, causing friction and heat.
The ultrasonic methods will not detect single fibre breaks or composite
materials. These methods do not work for complex structures (only the blades,
the tower and the nacelle can be monitored).
An advantage of the X-ray technology is that the images are obtained in
parallel, not through scan as is the case with the thermographic approach and,
therefore it is faster. However, problems can appear when the X-ray image is
interpreted. It should be noted that the sources emitting the X-rays are small and
air-cooled, thus fulfilling the safety prerequisites and they can also be controlled
remotely.
3. The Damage of the Structural Integrity of the Wind Power
Impairment can occur in any component or part of the turbine and it can
take any form, ranging from a crack in the concrete foundation to a split blade.
Various cases of structural damage are reported from time to time in several
countries such as Wales, Scotland, Spain, Germany, France, Denmark, Japan
and New Zealand [8]. In Germany, in 2002, a blade broke away, and parts were
found scattered throughout the area [9]. In another case, a blade flew to a
distance of 8 km and entered the window of a house. A detailed documentary of
this accident is available http://www.caithnesswindfarms.co.uk. According to a
survey performed in Germany [10] the frequency of damage is almost equal for
all the mechanical systems and structures. The failure varies from 4% for the
structural parts and the gear to 7% for the rotor blades.
Although structural damage may occur in any component, the most
common type of damage is encountered in rotor blades and in the tower [11].
Special attention is given to the blades, as they are the key elements of a wind
power generation system and as their cost can average 15-20% of the total cost
of the turbine. It was noted that the damage to the blades is the most expensive
type of repair and also the one which lasts the longest [12].
Additionally, even minor damage on the blades can lead to serious
secondary damage to the entire wind turbine system when action to repair it is
not taken immediately, leading to the collapse of the entire set [8].
183
The main construction elements of a turbine blade are shown in Fig. 2. The
blades are made of a mixture of fiberglass and composite materials. They are
designed to capture wind energy and transfer it to the turbine rotor and their
profile is the result of complex aerodynamic studies, on which turbine
efficiency is dependant. The development in size of the blades also caused an
increase in the size of the rotor.
Damage to blades may occur in different ways. The most common types of
damage is listed below [5] and a sketch of these is available in Fig. 3.
Typical damage description:
1. Damage formation and growth in the adhesive layer joining the skin and
the main spar flanges (skin/adhesive debonding and/or main
spar/adhesive layer debonding);
2. Damage formation and growth in the adhesive layer joining the up and
downwind skins along leading and/or trailing edges (adhesive joint
failure between skins);
3. Damage formation and growth at the interface between face and core in
sandwich panels in skins and main spar web (sandwich panel face/core
debonding);
4. Internal damage formation and growth in laminates in skin and/or main
spar flanges, under a tensile or compression load (delamination driven
by a tensional or a buckling load);
5. Splitting and fracture of separate fibres in laminates of the skin and main
spar (fibre failure in tension; laminate failure in compression);
6. Buckling of the skin due to damage formation and growth in the bond
between skin and main spar under compressive load (skin/adhesive
debonding induced by buckling, a specific type 1 case);
7. Formation and growth of cracks in the gel-coat; debonding of the gel-coat
from the skin (gel-coat cracking and gel-coat/skin debonding).
184
There are numerous causes which can lead to damage in the wind power
systems. Some of the reasons could be improper installation, operation under
severe conditions or low quality components.
Lightning can cause severe structural damage and the destruction of several
towers. Fire and strong wind can also damage a turbine. The most critical task is
probably when the turbine is stopped due to high wind.
The greatest danger occurs when, during a period of strong wind, the brake
system fails, and the turbine cannot be controlled. The brake system of a turbine
rotor is designed to halt if the wind is too strong.
When the brakes do not work, the turbine gets out of hand. In Germany, on
several occasions during 1999, 2000 and 2003, the wind turbine brakes failed to
stop the rotor, and blades were loose, pieces of which are found more than
500m away from the pillar.
185
4. Conclusions
1. Since the addition of the MSR monitoring system to a group of wind
turbine can negatively affect performance, the number and location of the
sensors is an important problem, which was addressed to a significant extent in
the current literature. The methods to be implemented for wind turbines should
demonstrate that they can perform well despite the small number of
measurement points and under the constraint that these points will be located in
areas prone to damage.
2. The technologies for monitoring and damage detection can be selected to
obtain comprehensive information. Detection methods are based on a network
of sensors adapted for early warning at the onset of damage. The methods can
detect the types of damages of less than 1 micron and the location of the
damage can be accurately determined.
3. Some of the limitations are the number of sensors they call for, as well as
the noise produced during the operation of the turbine. In addition, the
maintenance of a wired and wireless network between rotating parts, blades and
nacelle is still difficult. The most promising methods are the ultrasound and the
thermographic techniques. The X-ray methods, which were discussed here, can
also be applied for non-destructive testing, where in situ results must be fully
verified.
4. Finally, since the environmental benefits are constantly emphasized, the
generation of electric power with the help of wind systems is a necessary
innovation. Therefore, related industries require monitoring systems that can
provide effective maintenance programs, which provide the earliest possible
detection of defects.
5. The most important component of a wind turbine which requires
monitoring is the blade, so we must implement a reliable system of low-cost
sensors. This system should be integrated into the turbine blades in the
manufacturing phase, which will signal any potential failure in the turbine.
Therefore, the wind turbine should be built intelligently.
Acknowledgements. The authors would like to thank HCI Environment SRL for the
support and constructive comments.
POLITEHNICA University,
Faculty of Energetics
Bucuresti, Romania
e-mail: mihai.manescu@expert-mediu.eu
186
REFERENCES
1. F a r r a r C. R., S o h n H., Pattern recognition for structural health monitoring
Workshop on Mitigation of Earthquake Disaster by Advanced Technologies,
Las Vegas, NV, USA, 30 Nov.1 Dec. 2000.
2. G h o s h a l A., S u n d a r e s a n M. J., S c h u l z M. J., P a i P F., Structural health
monitoring techniques for wind turbine blades. J. Wind Eng. Ind. Aerodyn., 85
(2008) 30924
3. H a m e e d Z., H o n g Y. S., C h o Y. M., A h n S. H., S o n g C. K., Condition
monitoring and fault detection of wind turbines and related algorithms: a review
Renew. Sustain. Energy Rev. , 2007.
4. K i m H., M e l h e m H., Damage detection of structures by wavelet analysis. Eng.
Struct. 26, 34762, 2004.
5. S r e n s e n B. F., J r g e n s e n E., D e b e l C. P., J e n s e n F. M., J e n s e n H.
M., J a c o b s e n T. K., H a l l i n g K. M., Improved design of large wind
turbine blade of fibre composites based on studies of scale effects (Phase 1)
Summary Report (Ris-R Report) Ris National Laboratory, Denmark, 2004.
6. C a s el i t z P., G i e b h a r d t J., M e v e n k a m p M., On-line fault detection and
prediction in wind energy converters Proc. EWEC, (Thessaloniki, Greece, 1994,
pp 6237
7. G i e b h a r d t J., R o u v i l l a i n J., L y r n e r T., B u s s l e r C., G u t t S., H i n r i
c h s H., G r a m-H a n s e s K., W o l t e r N., G i e b e l G., Predictive condition
monitoring for offshore wind energy converters with respect to the IEC61400-25
standard Germany Wind Energy Conf. DEWEK, Wilhelmshaven, Germany,
2004.
8. R o s e n b l o o m E., A Problem with Wind Power www.aweo.org, 2006.
9. A s h l e y F., C i p r i a n o R. J., B r e c k e n r i d g e S., B r i g g s G. A., G r o s s
L. E., H i n k s o n J., L e w i s P. A., Bethany Wind Turbine Study Committee
Report www.townofbethany.com, 2007.
10. H a h n B., D u r s t e w i t z M., R o h r i g K., Wind Energy (Reliability of Wind
Turbines Experiences of 15 Years with 1,500 W) ed J Peinke et al, Berlin:
Springer, 2007, pp 32932.
11. C a i t h n e s s W i n d f a r m Information Forum 2005 Wind Turbine Accident
Data to December 31st 2005 http://www.caithnesswindfarms.co.uk/
12. F l e m m i n g M. L., T r o e l s S., New lightning qualification test procedure for
large wind turbine blades, Int. Conf. Lightning and Static Electricity,Blackpool,
UK, 2003, pp 36.110
13. C h i a C h e n C i a n g, J u n g-R y u l L e e, H y u n g-J o o n B a ng Structural
health monitoring for a wind turbine system: a review of damage detection
methods (IOP PUBLISHING) 2008.
187
188
Conform Farrar i Sohn [1] sistemul de monitorizare este definit ca un proces care
detecteaz defeciunile aprute n construciile nalte. Defeciunile sunt definite aici ca
modificri ale materialelor i / sau proprietile geometrice ale acestor sisteme, inclusiv
schimbri n condiiile limit i conectivitatea sistemului, care afecteaz n mod negativ
performana sistemului. Exista mai multe cauze de avariere a structurii cum ar fi
absorbia de umiditate, oboseala, rafale de vnt [2], stres termic, coroziune, foc i chiar
trsnet [3]. Palele turbinelor care nu au un sistem de protecie contra fulgerelor sunt
adesea lovite de acestea.
n general, dezvoltarea cu succes ale metodelor de MSR depinde de doi factoricheie: tehnologia de detectare i analiz a semnalului i algoritmul de interpretare [4].
Componentele MSR sunt formate din sistemul de achiziie de date, filtrare de date,
recunoaterea de model i luarea deciziilor. Fiecare dintre aceste componente este la fel
de important n determinarea strii siguran a unei structuri [5].
Tehnologiile de monitorizare sau de detectare a avariilor pot fi selectate pentru a
obine informaii globale. Metodele de detectare au la baz o reea de senzori adaptat
pentru o avertizare timpurie de la debutul avariei. Metodele pot detecta tipuri de daune
de sub 1 micron i se pot localiza deteriorrile sau impactul destul de precis.
Limitarea dintre aceste tehnoligii este dat de numrul de senzori solicitai, precum
i de zgomotul produs n timpul funcionrii turbinei. Metodele cele mai promitoare
sunt cele cu ultrasunete i termografia. Metodele cu raze X, discutate, pot de asemenea,
s fie aplicate pentru control nedistructiv, atunci cnd rezultatele n situ trebuie s fie
complet verificate.
n cele din urm, dup ct de mult este pus accentul pe mediu, beneficiile sistemelor
eoliene n generarea de energie electric, sunt mari, dar i costisitoare cu construcia i
ntreinerea. Prin urmare, industriile conexe necesit sisteme de monitorizare care s
poat oferi eficien programelor de ntreinere, care s ofere detectarea defectelor ct
mai devreme posibil.
1. Introduction
Many systems for which a centrifugal pump is otherwise suitable may,
however, have a variable demand in which case, a certain loss of efficiency may
have to be accepted from part of the head or part of the capacity used for control
purposes, using either discharge throttling or bypass control. Both methods will
inevitably result in power loss, so if economic regulation is of primary
importance, discharge regulation by speed control should be investigated first
since this is less wasteful of power and there is usually a considerably smaller
loss of pump efficiency. Speed control is now a particularly attractive
proposition with the increasing availability of variable frequency power units,
[1]. The measures for the improvement of the economic performances are
established from the systems dispatcher. The criterions used are the maximum
190
output of the pumping equipages and the optimum diameter of the piping. The
replacement of the existent equipment, that is obsolete from physical and
technological point of view, must be done with new equipments with
performances that will meet the requirements of an optimum operation from
both energetic and economic perspectives, [2].
2. Methodology
This analysis is more likely to be of academic rather than practical interest,
however, since the main requirements with regulated flow are:
a) To achieve the required variation in delivery to meet demand variations.
b) To realise good pumping efficiency at all demand levels, [3].
The connections analyze between the functional parameters of pumping
station and the quality parameters of drinkable water are:
a) The connections estimation between the pumps load, debit and the waters
turbulence.
b) The values analyze of chlorine and ammonium concentration depending on
debit.
c) The water conductivity analyzes depending on debit.
The best power and economical performances will correspond to the
pumping solution which ensures the covering of the request area (Q, H) with the
best output. The pumping efficiency is established by studying technical
implications of modernization measures of the power station. It is imperative to
use the automatic systems for the water treatment. Energy efficiency and
economic efficiency for the pumping supply system are tightly connected to the
proper choice of pumping device and appropriate operation of the hydraulic
system, [4].
The evaluation of connections between the functional parameters of
hydraulic systems and the quality indicators of water, depending on time is
obtained with the help of an original automatic calculation program, elaborated
by authors. Allowing for the actual situation of the adduction and the
effectuations necessity of the repair workings capital or the pipes replacement
in some sectors, respective of the pumping aggregates, for to assure a powereconomical evolution favorable of the system, one are determined: the optimum
nominal diameter of the pipes and the maximum theoretical outturn of the
pumps what must be used and the changes convenience of the existing
equipment, [3].
Depending for the installations total outturn and for the water volume W
circulated in the period of reference, the consumption of power for the waters
pumping Eso is independent at stops duration at top:
191
(1)
Eso =
In Eq (1) has used the following notations: Ntm (kW) - power average
necessary for waters transport:
(2)
Ntm (Q, D, L) =
K N .Q.H t K N .Q.L.J
=
.
p
p
W (m3) -volume of water brought in T (hours) period; J (-) - slope metric of the
water transport; D (m) - nominal pipes diameter; Q (l/s) - debit of water; L (m)
- pipes length; p (%) - pumps outturn; KN , Kj - coefficients; Ht (m) - loss of
charge:
(3)
Q
H t = L.J (Q, D ) , J = K j . D (Q, J ) = K j .
J
D ,
1/
The investment in the pipes network and the investment in the pumping
station are calculated with the fallowing equations:
(5)
Q
I c ( D, L) = L.a. K j .
1,15.K N .Q.L.J
; I p ( N ti ) = I po .
(6)
Q
C A = A + C AI = ac .L.a. K j .
1,15.K N .Q.L.J
+ a p .I po .
192
The annual medium marginal cost is calculated with the fallowing equation:
CM
(7)
C
L.a
Q
= A=
. K j .
W
W
J
L.a
A* =
. Kj
W
( )
I po .a p 1,15.K N .Q.L.J
+
.
p
W
= A *.
Q /
J
+ B.Q
I po .a p 1,15.K N .L
;B =
.
p
W
.J
;C =
+ C.Q.J ;
pE .(1 + krE ) .K N .L
3600. p
(8)
CM
Q
= 0 A. .
J
1+
J
+ B. p .Q
.J
p 1
+ C.Q = 0
The available data for choosing the proper pumping device require a careful
analysis to determine the head characteristics of the pressurized supplies for
each specific configuration and to evaluate the proper operational regimes of the
pumping devices. The calculation method is applied in the exoneration case of
pumping station CUG Iasi for drinkable water from Iasi city.
Before of exoneration the pumping station CUG Iasi has been equipped
with two 8 NDS pumps; after exoneration the pumping station CUG is equipped
with two Wilo NPG 200-500-160/4/499 pumps.
The measure and control system of drinkable waters quality, of functional
parameters of the pipes network from CUG pumping station has been complete
automated in August December 2008 with performance installations by
Endress Hauser concern. All system has automated calculus program that
permits of the dispatcher from S. C. APAVITAL S. A. Iasi industry to pursue
the parameters network in 17 demurrages from the supply system with
drinkable water of Iasi town, (Fig. 1). The registering is permanent, 24 hours.
The water samples are analyzed automatic at each 2 5 minutes. Drinkable
water is treated with ammonium and chlorine in accordance with the properties
of the analyzed water.
193
194
Fig. 2 Diagram for the functional parameters of the CUG Iai pumping station:
conductivity, chlorine, pressure, flow and turbulence for the perioad 5.07.2009
6.07.2009.
1000
900
800
700
AIc ( D )
600
AP ( k , Q , D ) 500
A ( k , Q , D ) 400
300
200
100
0.3
0.4
0.5
0.6
0.7
0.8
Fig. 3 The variation of annual cost associated to the investment in pipes AIc, annual
total cost AP for pumping and the annual average cost associated to the investments A,
depending on diameter D for the CUG pumping station.
The results of the data working relative at either sector of the system,
effectuated with MathCAD program, adequate to the supply with power under
average and low tension are systematized in Fig. 3. It is present the variation of
the following parameters: AIc annual cost associated to the investment in
195
pipes, AP annual total cost for pumping and A - the annual average cost
associated to the investments in the usages conditions of the following
parameters: ac = 0,075; a = 6000; = 2; k pipes equivalent absolute rugosity
= 0,8 mm; Kj = 0,00145; = 2; = 5,143+0,131.k; p = 72 %; T = 6200
hours/year; p = 0,46; Ipo = 65; Q = 0,15 m3/s; D = 0,3; 0,4; ...; 0,6 m.
4. Conclusions
REFERENCES
1. A l e x a n d r e s c u A., Statii de pompare, Ed. Politehnium, Iasi, 2008.
2. A l e x a n d r e s c u A., Masini si echipamente hidraulice, Ed. Politehnium, Iasi,
2008.
196
1. Introduction
It is a well-known fact that during of the turbine running (rotational
motion), under the effect of: hydraulic forces and moments developed on the
blades, runner masses (with or without oil) and vibration, the turbine shaft is
submitted to specific variable stresses (tension, bending and torsion). Taking
into account the results obtained in [1], in the present work, the durability of the
horizontal axial hydraulic turbines shaft was evaluated using the professional
programs ANSYS and AFGROW. In order to obtain also numerical examples
were used the bulb turbines in function at the Power Plants Iron Gates II and
Gogosu, for which we know the necessary data. The obtained conclusions allow
avoiding dangerous troubles during the turbine operation.
198
Fig. 1 The 3D shaft model, obtained with the INVENTOR program [initiation].
b) Cracks network
Fig.2 Cracks in the joining zone between the runner flange and the main shaft body.
199
Usually, together with the increase of stress level is enhanced also the
fatigue crack propagation speed. As a result, the propagation of the crack speed
can be correlated with the variation of the stress intensity factor
K,
da
= f (K ) , Fig. 4.
dN
200
In conformity with Fig. 4, after the crack initiation, the maximum danger is
represented by the zone III, characterized by great propagation velocities
leading to instable increases of the failure. In this zone, the velocity of crack
propagation is correlated with the variation of the crack intensity factor, through
the relation proposed by Forman [2]:
da
C (K )
=
= f (K, R),
dN (1 R )K C K
n
(1)
Nf
a cr
Nd
ad
dN = N f N d = N r =
da
.
f (K , R )
K I = ( t + H b )
a
F (a, c, Di , De , ) ,
Q
where t - represents the axial stress applied to the shaft, b - represents the
bending stress, H and F depends on the crack geometry (crack depths and
length), the shaft thickness and the frontal position of the crack, a is the depth
and c the half-length of the crack,
(4)
a
Q = 1 + 1,464
c
for
201
a
1.
c
202
(5)
da
= C1 K max (1 R) m 1
dN
da
= C1 K max (1 R)1m
dN
n1
n1
for R 0,
for R < 0
are C1 = 1,447 x 10-12; n1 = 3,6; m = 1; the values for the breaking tenacity
corresponding to a plane state of stresses are KC = 110 MPa m1/2 , respectively
for a plane state of deformation KIC = 77 MPa m1/2, the yielding point C = 262
MPa, the elastic modulus E = 206843 MPa and Poissons ratio = 0.3, the limit
value of the stress intensity factor under which the crack does not propagate K
= 1,5 MPa m1/2.
In Fig. 6 is presented the crack evolution for some time intervals under a
load having a symmetric alternative bending cycle.
a) a = 1mm, c = 2mm
N = 80370 hours (initiation)
b) a = 16 mm, c = 32 mm
N = 153243 hours (propagation)
The simulation begun with the following initial value of the crack a = 1 mm
and c = 2 mm, values which are reached after 80370 running hours. The number
203
of hours, for the crack propagation is added to the initial time (80370). The
obtained results show that after 15937 running hours the crack penetrates
completely the annular wall of the shaft.
3. Conclusions
The studies of the fatigue crack propagation lead to the following results:
1. In the joint zone between the flange and the shaft body, the crack occurs
after approximately 80370 running hours.
2. The propagation of the fatigue cracks were obtained using the Paris law
with the completion of Walker, and show that the crack reach the dimensions a
= 16 mm (depth) i 2c = 64 mm (length) after 153243 running hours. After that
the propagation speed increases to a high degree and after only 159737 running
hours the breakdown occur (a=150 mm and 2c= 320 mm).
3. The obtained results can be used by the supervising staff in order to
prescribe the intervals of periodical inspections and to avoid major damages.
Acknowledgements. The present work has been supported from the Contract No.
RU 177/10.10.2008, BC 146/13.10.2008 (Analiz privind soluia de fiabilizare a
arborelui turbinelor aplicat cu ocazia retehnologizrii hidroagregatelor din CHE Porile
de Fier II. Propuneri de metodologie de urmrire n timp a strii arborilor turbinelor din
CHE Porile de Fier II i CHE Gogosu)
REFERENCES
1. B o r d e a u I., P o p o v i c i u M.O., N o v a c D., Fatigue Studies Upon
Horizontal Hydraulic Turbines Shaft and Estimation of Crack Initiation Machine
Design, University of Novi Sad, Faculty of Technical Sciences, 2009, pp 183186.
2. F o r m a n R.G., H e a r n e y V.E., E n g l e R.M., Numerical analysis of crack
propagation in cyclic-loaded structures, Journal of Basic Engineering, Trans.
ASME, Vol. 89, (1967).
3. H a r t n e r J. A., AFGROW Users Guide and Technical Manual, Wright-Patterson
Air Force BASE, Ohio, 2008.
204
1. Introduction
Through scope of investment and energy consumption, adductions have a
significant share in the systems of water supply, and their rational design
behaves more optimization processes, among which an important place held
optimize their route.
In current practice of design, setting the optimal solution is, usually, by
analytical study of two or three variants selected from many possible intuitive
206
Teodor Milos et al
( )
( )
(1)
0 for i = j
The optimal route is the best way of graph , with the total value:
(2)
( ) =
(u ij ) min .
u j
( )
Vi = ni ,
( i = 0, n ).
hence:
(4)
Vn = 0 ,
207
Vi = min V j + mij ,
j i
( i = 0, n 1 ; j = 0, n )
and Vn = 0 .
(6)
Vn0 = 0 .
Calculate:
(7)
and then:
(8)
Ordinal iteration of k expressed by relations (8) gives values only for the
finite length of roads at most k + 1 arriving at xn , choosing between them is the
minimum.
From iteration to the next:
(9)
Vi k Vi k 1 ,
j .
208
(10)
Teodor Milos et al
Vi k = mij + V jk 1 = mij + V jk .
(11)
( )
( )
( )
( )
u ij if arc xi ,x j exist
n
0 for i = j
Are introduced as data entry, graph order and its associated matrix, on lines,
whose elements are considered equal to u ij , if there arc xi , x j , or equal to
( )
209
210
Teodor Milos et al
(12)
M=
Vi0
Vi1
Vi 2
Vi3
42 53
197 197
73 73
3
4
92 61 82
0 61
61
92
92
92
92
92
82
82
82
82
82
Route with minimum total cost is given by way of minimum value in this
graph, which is determined using Bellman-Kalaba algorithm.
For each Vi k (k = 0,1,...) is added to previous matrix a column in which
values are inserted properly. It follows successively:
a) Calculate the values Vi0 = mi 7 (i = 1, ..., 7), which pass into the column Vi0 .
b) Calculate values Vi1 for i =1,...,7 (j =1,...,7), which is passing in that
column:
(13)
(
(
(
(
(
(
V11 = min V j0 + m1 j
j 1
1
V j0 + m2 j
V2 = min
j 2
V31 = min V j0 + m3 j
j 3
1
0
V4 = min V j + m4 j
j 4
V 1 = min V 0 + m
j
5j
5
j 5
1
0
V6 = min V j + m6 j
j 6
V 1 = 0
7
) = min(V
) = min(V
) = min(V
) = min(V
) = min(V
) = min(V
)
+ m ) = 153
+ m ) = 153
+ m ) = 61 .
+ m ) = 92
+ m ) = 82
0
2
0
1
0
1
0
1
0
1
0
1
27
37
47
57
67
211
(14)
V12
2
V2
V32
2
V4
V 2
5
2
V6
V 2
7
(
= min (V
= min (V
= min (V
= min (V
= min (V
) (
) = min(V
) = min(V
) = min(V
) = min(V
) = min(V
)
,..., V + m ) = 153
,..., V + m ) = 153
,..., V + m ) = 61 .
,..., V + m ) = 92
,..., V + m ) = 82
= min V 1j + m1 j = min V21 + m12 , V31 + m13 ,..., V71 + m17 = 197
j 1
j2
j 3
j4
j 5
j 6
1
j
+ m2 j
1
j
+ m3 j
1
j
+ m4 j
1
j
+ m5 j
1
j
+ m6 j
1
1
1
7
27
1
1
1
7
37
1
1
1
7
47
1
1
1
7
57
1
1
1
7
67
=0
Since Vi 2 = Vi3 (i = 1, ..., 7), algorithm stops and the road is the minimum of
V12 = V13 = 197 . This value is reached on the way (1, 2, 6, 7), thus resulting in
optimal route of adduction as: S1, A, C, and L. To resolve this problem the
computer program BEL_KAL was used.
(15)
(
(
(
(
(
(
V13 = min V j2 + m1 j
j 1
3
V j2 + m2 j
V2 = min
j2
V33 = min V j2 + m3 j
j 3
3
2
V4 = min V j + m4 j
j4
V 3 = min V 2 + m
j
5j
5
j 5
3
2
V6 = min V j + m6 j
j 6
V 3 = 0
7
) = 197
) = 153
) = 153
) = 61 .
) = 92
) = 82
212
Teodor Milos et al
5. Conclusions
1. This study put highlights how the economic fact of a technical problem
can be optimized using mathematic-informatics structure type graph.
2. Mathematical model is easily programmable in an evolved language,
obtaining immediate results. The only problem is populating the matrix attached
graph.
3. Bellman-Kalaba algorithm was originally designed for the economy, but
the adaptation was possible because the optimal path the economy is a virtual
way and here is a real way.
Acknowledgements. The present work has been supported by the Romanian
Government Ministry of Education, Research and Innovation, The National Centre
for Programs Management (CNMP) through, CNMP project no. 21-036/2007 and
CNMP project no. 21-41/2007.
Received:
REFERENCES
1. A l e x a n d r e s c u , A. Concerning optimizations working of the pumping station
for water feedings, 18th International Conference on Hydraulics and Pneumatics,
Prague, Czech Rep., Sbornik, ISBN 80-02-01567-3, 2003, pp. 263-268.
2. A n t o n L. E., B a y a A., M i l o s T., R e s i g a R., Experimental Fluid
Mechanics, Vol. 1, Ed. Academic Horizons, Timisoara, Romania, ISBN 9738391-72-5, 2002.
3. B e l l m a n , R.; K a l a b a , R. Dynamic Programming and Modern Control
Theory, McGraw-Hill, New York, 1965.
4. N a y y a r M.L. Piping Handbook, 7th Edition, McGrawHill, ISBN: 978-0-07047106-1, New York, San Francisco, Tokyo, Toronto, 2000.
5. S r b u , I. Energetically Optimization of Water Distribution Systems, Ed.
Academiei, ISBN: 973-27-0575-2, Bucuresti, Romania, 1996.
213
1. Introduction
The transmission of the movement of rotation and torque between co-axial
shafts can be achieved through couplings. From the category of permanent fixed
couplings in this paper the sleeve rigid coupling with cylindrical pins will be
analyzed. The geometric characteristics of the sleeve rigid coupling with
cylindrical pins are shown in Fig.1. The cylindrical pins can be placed in the
same plane (Fig. 1a), or more often in two different planes at 90 (Fig. 1b).
The shafts diameter can be obtained by the preliminary calculation so as to
ensure the transmission of a mechanical power P = 8 kW at the rotational speed
of n = 200 rpm.
With the calculated values for the angular rate = 2n 60 = 20.944 rad/s;
for the torque M t = P = 381.9719 Nm and for the safety torque
M tc = k M M t = 800 Nm (the safety coefficient k M shall be adopted within
216
1.53, the adopted value is 2.0944), the shafts diameter can be calculated using
the relationship:
d =3
(1)
the value d
16 M tc
at
t =
16 DM tc
D4 d 4
).
(3)
f =
4 M tc
dd s2
217
The obtained values are within the safety range of the torsional strength and
the shearing strength of the material. Even if as a result of these calculations a
good dimensioning of the coupling will be obtained, to answer a series of
additional issues (the torque asymmetry on the coupling parts, the distribution
of critical sections on each coupling part, etc.) should be further analyzed by
advanced numerical methods.
2. The Structural Analysis of the Sleeve Coupling
with Pins in the same Plane
The main steps of the procedure for structural analysis of sleeve couplings
with cylindrical pins in the same plane using CATIA/Generative Structural
Analysis environment are:
1. Creating the 3D model. Shall be carried out separately the five elements
of the coupling using the Part workbench. The appropriate materials are
assigned to each element. The assembly of the five coupling elements is
carried out using the Assembly workbench, Fig. 2a.
2. Configuring the mesh. Shall be carried out separately for each element,
by changing the characteristic parameters Size and Sag, Fig. 2b.
3. Applying the restraints. On the free extremity of the driven shaft a
clamp condition must be applied, Fig. 2c.
4. Applying the loads. On the free extremity of the driving shaft an 800
Nm torque around the shaft axis will be applied, Fig. 2d.
5. Applying the conditions for the interaction between the five elements of
the sleeve coupling. First, the interaction condition of General Analysis
Connection type will be applied between the driving shaft and the first
pin, between the first pin and the sleeve coupling, between the sleeve
coupling and the second pin and finally, between the second pin and the
driven shaft. Second, for all the active areas in terms of the torque
transmission the shearing strain condition must be specified by setting
the just property of interaction between the active coupling elements
(Bolt Tightening Connection Property), Fig. 2e.
6. Launching the solver, running the numerical analysis and results
visualization. The von Misses stress will be presented for both the
driving and the driven shafts in Fig. 2f (scale factor 500), for both the
cylindrical pins in Fig. 2g (scale factor 700) and, finally for the sleeve
coupling in Fig. 2h (scale factor 500).
218
For the sleeve coupling with pins in two different planes at 90 the analysis
procedures takes the same steps: creating the 3D model (Fig. 3a), configuring
the mesh (Fig. 3b), applying the restraints (Fig. 3c) and the loads (Fig. 3d),
applying the interaction conditions (Fig. 3e), running the analysis and results
visualization for both shafts (Fig. 3f scale factor 500), for both pins (Fig. 3g
scale factor 700) and for the sleeve coupling (Fig. 3h scale factor 500).
219
Fig. 3 Analysis procedure steps for sleeve couplings with cylindrical pins
in two different planes at 90 :
a creating the 3D model; b configuring the mesh; c applying the restraints; d
applying the loads; e applying the interaction conditions; f von Misses stress for
shafts; g von Misses stress for pins; h von Misses stress for sleeve coupling.
220
4. Conclusions
REFERENCES
1. D e m i a n T., Elemente constructive de mecanic fin. Ed. Didactic i Pedagogic,
Bucureti, 1980.
2. D r g h i c i I., et al., Calculul i construcia cuplajelor. Ed. Tehnic, Bucureti,
1978.
STRUCTURAL ANALYSIS OF A
BIMETALLIC STRIP THERMOSTAT
BY
1. Introduction
The bimetallic springs used as sensitive elements of the thermostats are
made by two beams of rectangular cross section bounded together. The two
beams are made of materials with different coefficients of thermal expansion,
a > p . Under a uniform temperature field T the bimetallic strip will curve
toward the passive material (the material with lower coefficient of thermal
expansion). The sensitivity of the bimetallic strip will be better as long the
difference a p will be higher. In this paper, the materials used are: titanium
as passive layer and zinc, aluminum, bronze, brass, copper and steel as passive
layer. Characteristics of materials relevant for the analysis are shown in Table 1.
The main geometric characteristics of the bimetallic strip are shown in Fig. 1.
222
Passive
layer
Titanium
E 1011, [Pa]
1.14
-6
-1
9.5
10 , [C ]
825
02 ,[MPa]
Table 1
Material characteristics
Active layer
Zinc
0.97
31.2
140
Aluminum Bronze
0.7
1.1
23.6
17.8
95
520
Brass
1.31
16.7
350
Copper
1.1
16.5
290
Steel
2
11.7
250
The main steps of the procedure for structural analysis of the bimetallic strip
using CATIA/Generative Structural Analysis environment are:
1. Creating the 3D model. Shall be carried out separately the two
rectangular cross section beams using the Part workbench. The
appropriate materials are assigned to each element. The assembly of the
two beams is carried out using the Assembly workbench, Fig. 2a.
2. Configuring the mesh. Shall be carried out separately for each beam, by
changing the characteristic parameters Size and Sag, Fig. 2b.
3. Applying the restraints. At one side of the bimetallic strip a clamp
condition must be applied, Fig. 2c.
4. Applying the loads. On the both beams a thermal load with T = 30 C
will be applied, Fig. 2d.
5. Applying the conditions for the interaction between the two beams of
the bimetallic strip. First, the interaction condition of General Analysis
Connection type will be applied between the active layer and the
passive layer. Second, for the active area in terms of the stress/strain
223
224
3. Comparative Analysis
After a preliminary analysis, the clamped end of the bimetallic strip has
been identified as being critical in terms of stresses. In order to improve the
computational process the local changing of the mesh parameters at the clamped
end is required. Thus, for the critical area the Size mesh parameter will be
0.25. The global mesh parameters are Size=1 and Sag=0.5.
The first set of comparative analyses is determining how the width of the
beams influences the behavior of the bimetallic strip. Thus, for every
combination of materials changing the width of the beams and maintaining
constant the other parameters were obtained the numerical values for maximum
deflection (Fig. 3a) and for the von Misses stress on the active layer (Fig. 3b),
as well as for the passive layer (Fig. 3c).
The second set of comparative analyses is determining how the length of the
beams influences the behavior of the bimetallic strip. Thus, for every
combination of materials changing the length of the beams and maintaining
constant the other parameters were obtained the numerical values for maximum
225
deflection (Fig. 4a) and for the von Misses stress on the active layer (Fig. 4b),
as well as for the passive layer (Fig. 4c)
4. Conclusions
226
REFERENCES
1. D e m i a n T., Elemente constructive de mecanic fin. Ed. Didactic i Pedagogic,
Bucureti, 1980.
2. Z a m a n i N.G., CATIA V5 FEA Tutorials. SDC Publications, 2005.
1. Introduction
The transfer from industrial to information and knowledge societies
radically transformed the markets and increased customer expectations.
228
229
order
placed
Customer
(online configurator)
decide
custom
request
Sales
preliminary offering
offering
record
request
(configurator, ERP,...)
Design
(CAD, FEM, ERP, PDM, MBS,...)
request
analyze
request
response
characteristic
curves
request
response
design
part
consultation
request
response
Purchasing
(ERP,...)
clarification
schedule
delivery
date
Production
(CAM, NX-Daten,...)
Legend:
Department
(used software)
timeline
document
action
230
In order to shorten the offering generation time, both the client request
clarification and the product design process have to be accelerated and the
quality of the transferred information has to be improved. The solution is a
software system that integrates the product data with existing virtual methods in
order to partially automate the configuration. If a customized product is needed,
a completely automated configuration system is impossible. The
algorithmization of the design process is elaborately discussed by F r a n k e in
[6]. However, it is possible to accelerate and optimize the configuration chain
through better requirement definition, product parameter visualization and goal
conflict analysis.
Fig. 3 presents the implementation solution the main software tool is the
configuration module, which is used by both sales and engineering design
departments in order to configure a custom solution. The output of both
departments is the customized offer information (such as prices, conditions,
related product pictures and site plans) and the customer specific technical
information (such as exact technical description and parametric 3D-CADmodels as starting point for a complete and precise product configuration. The
product specific knowledge base is the fundamental information source for the
whole configuration process. Its requirements and structure are presented in the
next section.
231
Data
structured
isolated
context independent
symbols
Information
Knowledge
mixed
connected
context dependent
cognitive pattern of action
232
The context for the interpretation of the data has a strategic role in the
knowledge output. The discussed company specific knowledge base for the
product configuration should therefore consider both involved sales and design
departments context and semantic. It is therefore important to recognize the
different views, as shown in Table 2. Nevertheless, the common goal is to
configure as fast as possible a high quality customized product, with minimum
resources, minimum costs and minimum losses.
Keeping this in mind, the following structure for a common knowledge base
is proposed:
233
Table 2
Key aspects for the sales and design departments
Departments:
Specialization:
Modules:
Software:
Goals:
Degree of content:
Sales
Marketing
Functional modules
Configuration tool
Sell the product
Sale probability
Engineering design
Engineering
Product parts and assemblies
Virtualization and calculation tools
Design with minimum costs
Technical feasibility
6. Conclusions
1. The need for customized products has increased due to market
globalization, internationalization and customer awareness. For staying
competitive, companies have to provide custom solutions for low prices, high
product quality and short lead times.
2. The knowledge necessary for configuring customized products in early
design phases is often scattered through numerous software tools and separated
234
REFERENCES
1. S h e h a b E., B o u i n - P o r t e t M., H o l e R., F o w l e r C., Enhancement of
Digital Design Data Availability in the Aerospace Industry. In Proceedings of
the 19th CIRP Design Conference, p. 589-590, 2009.
2. H i l m a n J., P a a s M., H a e n s c h k e A., V i e t o r T., Automatic Concept
Model Generation for Optimization and Robust Design of Passenger cars.
Advances in Engineering Software, 38: 795-801, 2007.
3. K r a u s e F. -L., The Future of Product Development, Proceedings of the 17th CIRP
Design Conference, Springer, 2007.
4. B i n n e r H. F., Ganzheitiches Wissenskonzept - Wissensframework zur Gestaltung
und Implementierung einer wissensbasierten Organisation. In ZWF Zeitschrift
fr wirtschaftlichen Fabrikbetrieb, Carl-Hanser Verlag, Mnchen, 103 (2008),
p. 540-543.
5. A l e x a n d r e s c u I., F r a n k e H. -J., Fast Offer and Optimized Design for
Complex Custom Products. 1st Symposium on Multidisciplinary Studies of
Design in Mechanical Engineering, p. 11-12, 2008.
6. F r a n k e H. -J., Untersuchungen zur Algorithmisierbarkeit des Konstruktionsprozesses. VDI Verlag, 1976.
7. B o d e n d o r f F., Daten- und Wissensmanagement. Springer, p. 2, 2003.
235
1. Introduction
In the literature, the word innovation is more favored than the concept of
creativity [1], [6]. Many definitions show the value, the importance and the
time intensity of innovation [8], [12].
Innovation is often described as the value-adding stage of the creativity
process, suggesting a higher sense of value for innovation. Creativity is a precondition from which innovation develops, is a new realization in practice
supported by creative thinking [8].
From the analysis of the information from literature, the innovation concept
can be concerned in terms of five basic elements: process, product, human
factors, domain and socio-organizational framework. If it could be demystified,
described and modeled the creative process, it would be able to enhance
individual innovative potential and facilitate the innovative process. When the
238
239
240
241
approach. It was determined that the most efficient innovative solutions grow up
from contradictions. TRIZ Matrix operates with the contradictions principles.
This is designed on the basis of 39 factors organized on rows and columns,
represented in the Fig. 4, and at intersection between rows and columns can be
identified the contradictions. For each contradiction it is presented a set of
principles, from a total of 40, in order to solve it. If the problem is structured on
the contradictions principle, the TRIZ matrix offers sets of innovative
principles. The essential contribution of the users matrix is to match the
standard principles to that innovative problem [10].
242
system. Thereby a resource is anything in the system which is not being used to
its maximum potential. Discovery of such resources then reveals opportunities
through which the design of a system may be improved [7].
3. Case Analysis Results
From the TRIZ analogy exploration, it can be combined the Blooms model
presented in Fig. 2 with the self-managed presented in Fig. 3. The role of this
analogical approach is to define a model to stimulate the innovative potential.
In a first stage it is built a matrix structure having as columns the pillars
from the model of self-managed team and as rows the cognitive domains from
Blooms Taxonomy. In the cells resulted at the intersection of rows and
columns, are inserted the Blooms domain verbs from the relative line. So, it is
obtained a matrix that offers the necessary stimulants for innovative activity.
The matrix is presented in Figure 5 as the Innovative Acting Matrix. The
importance of this matrix is that, those who access it, being in situations marked
by one of the innovative pillars, have the opportunity to explore the innovative
problem in an acting style. The exploration in an acting manner of the
innovative problem is achieved gradually from the base level - knowledge to a
superior level - evaluation.
In the second stage, on the same matrix structure described above, in the
cells from the intersection of rows and columns, are inserted the nouns of the
Blooms domain from the relative line. It is obtain a matrix that provides the
necessary support for innovative activity. The matrix is presented in Figure 6 as
the Innovative Support Matrix. The importance of this matrix is that, those who
access it, being in situations marked by one of the innovative pillars, have the
opportunity to receive the necessary resources to explore the innovative
problem.
243
Conclusions
REFERENCES
1. A m a b i l e T. M., From individual creativity to organizational innovation. In
Innovation: A cross-disciplinary perspective. (Gronhaug K., Kaufmann G.,
Eds.), Scandinavian University Press, Oslo-Norway, pp. 139-166, 1988.
2. B e l b i n R. M., Management Teams. Butterworth Heinemann Publisher, LondonUK, 1981.
244
1. Introduction
One of the main domains of interest in rapid product development is
represented by machine manufacturing, where each machine contains thousands
of parts which have to be made as quick and cheap as possible, in order to
achieve the prescribed quality [1].
But, by economical means and, of course, due to progress and innovation,
the products do improve, and this leads to major or minor modifications brought
to the machine, thus, many other internal parts have to be changed. The
available time for this changes, has become increasable shorter due to
246
Client specifications
Physical model
(sample or prototype)
Digitizing system
3D geometry
Conventional
approach
Points cloud
3D coordinates
3D CAD
system
Triangular model
(file format *.stl)
Surface model 3D
reconstruction
Conceptual model
CAE system
No
Design
requirements
satisfied?
Unconventional
approach
No
Yes
CAM system
Final product
(part or tool)
247
248
CNC milling
machine with
touching probe
CMM
with laser
system
CMM
with
touching
probe
CNC
milling
machine with
laser system
CT
Computer
tomography
Digitizing
software
Points cloud
(3D coordinates)
249
group, we can also use CMMs or CNC-milling machines but equipped with
laser beam probes (ex. Renishaw, Metris LC Mitutoyo), or associated optical
sensors (ex. CCD cameras) [2].
3. Advantages and Weak Points of Different Scanning Equipments
Scanning
equipments
Laser
Advantages
Week points
CCD
cameras
Fast
It is possible to use two or three cameras
simultaneously
Insensible to part color
Contact
(classic)
Contact
digitizers
(Renishaw)
250
5. Conclusions
1. Product development (parts/moulds/tools) via integrated reverse
engineering represents a fairly new method that is in research and development
phase.
2. This paper aims to reveal some of the highlights and weak points of
existing scanning equipments. We see that as far as accuracy goes, the best
results are obtained by means of contact sensors. Nevertheless, in case of fragile
or soft materials, laser or optical systems, without contact, provide better
results. The ideal scanning device should be the combination of a camera, laser
and contact probe, all, supported by appropriate controller.
Received: March, 12, 2010
REFERENCES
1. H e r b e r t s o n T., Reverse engineering, in: Fourth International Conference on
Industrial Tools ICIT 2003, April 8th12th, Bled, Slovenia, pp. 419422.
2. N. N., Scanning Systems for Reverse Engineering, Renishaw Apply Innovation, H2000-3120-04-B, Renishaw plc, UK, consulted at 18 November 2009,
http://www.renishaw.com/en/cmm-probes-software-and-retrofits--6329
3. H o n g w e i L., Adaptive patch-based mesh fitting for reverse engineering, State
Key Lab of CAD&CG, Zhejiang University, Hangzhou 310027, China, Received
13 May 2007; accepted 1 October 2007, pp. 3-10.
4. Z e x i a o X., Complete 3D measurement in reverse engineering using a multi-probe
system, Engineering college, Ocean University of China, Qingdao 266071,
Peoples Republic of China, Received 7 October 2004; accepted 20 January
2005, Available online 7 March 2005, pp. 1135-1139.
1. Introduction
There are many studies directed to determine the influence of the cutting
parameters in bone drilling and the production of heat and the increase in the
bone temperature due to this heat. In those studies the temperature was
measured with thermocouples, the most usual way, or through other variables.
There are less studies in the same area that use a more advanced method for the
register of the temperature in bone cutting, the thermology. But there are even
less studies that use a more dynamic method to obtain the temperature and be
able to relate it directly to other variables such as pressure, speed of cutting, etc.
Hereby our study tries to make a contribution in the use of the infrared
thermology. Infrared radiation is a form of electromagnetic radiation like: radio
waves, microwaves, ultraviolet rays, gamma rays, visible light, etc. All of them
emit energy in the form of electromagnetic waves and they travel at the speed of
light. All matter above absolute zero (0 K, -273C) emits infrared energy.
Thermographic cameras detect invisible infrared radiation emitted by objects
and they transform it in an image inside the visible spectrum in which the color
252
scale (or greys) shows the different intensities. The intensity of the infrared
radiation depends on the temperature and on the surface characteristics of the
object, the color and the kind of material. Thermographic cameras give a value
of the temperature for every single point, without taking into account that, for
the same temperature, two different materials can emit infrared energy with
very different intensities.
With this new tool we will be able to determine the local heating
experienced by the bone during the process of cutting (drilling) for dental
implants. This way, we can see, frame by frame, the evolution of the
temperature in every single point in the area of study, so every frame
corresponds to a thermography with the data of the temperatures of every single
point of the image. With this data we can study data from heating and cooling
patterns, zone and/or specific peaks and lows.
253
Surgical motor
Surgic XT Motor System (with a built-in irrigation system). Technical
data: Power 210 W; motor speed: 200 40.000 rpm; torque: 50 Ncm. We used
254
a handpiece for the drilling tests to keep it as centred as possible from the load
transducer from which it will hang.
To hold the handpiece we built a device, made from aluminum, which could
both, hold the handpiece and also attach it to the machine that generates the
force, as seen in figure 2.
Testing Machine
For the movement generating we used a Shimadzu Precision Universal
Tester machine, model AG-100KNIS-MO. It is a floor model with two
columns, able to work with speed tests of 0.5 m at 1000 mm/min, according to
the manufacturer. The setting up can be seen in Figure 1.
Drills
Tool used for drilling the bone is a cobalt drill bit for metals. Such a drill bit
is not very different in shape from the ones used in dental surgery and dental
implants. Also the material it is made of is very similar to the materials used for
the dental drill bits. The difference in temperatures between the drill bits we
used and the dental drill bits should not be very different. The point of this test
is to measure the temperature of bone drilling using thermology, so for that
matter, the drill bits used for the tests should be as good as others. The drill bit
used in the tests is 2.22 mm in diameter. It is cylindrical straight and has two
lips. Drills can be seen in figure 3.
Bone
For the tests we have used cow bone, specifically the femur. It is a fresh
bone bought the previous day in a local market. The cow bone characteristics
255
are very similar to the ones of the human bone, for that reason cow bones have
been used for the scientific community in multiple tests of this type.
4. Methodology
For the test we used a 2.22 mm in diameter drill bit turning at a speed of
2,000 rpm attached to a handpiece from the surgical motor. For the set up of the
surgical motor we choose the maximum power to avoid a fall in the turning
speed. The handpiece was attached to the tester that made it move down at a
speed of 1.25 mm/second. The drill was made in a single movement until we
reached a depth of 8 mm. This is not the usual way for implantology, which
uses a discontinuous drilling of one or two seconds. This way we could get a
higher temperature in the drill bit and check the efficiency of thermology in
bone drilling.
The thermographic camera was placed in front of the force generating
machine, as seen in figure 1, which gave us a general view of the cutting area.
Over the thermographic camera was placed a macro lens in order to increase the
resolution of the camera in the area of contact between the drill bit and the bone
sample. This setting allowed us to register the temperature in the drilling zone,
which included the temperature of the drill bit and the superficial temperature of
the bone in the critical area, beside the tool. We made several tests where we
256
register the temperatures from the beginning of the drilling till the exit of the
drill bit once reached the 8 mm depth.
6. Conclusions
These are some of the most relevant results of the thermographies during the
measurement of temperature in bone drilling. These results could be applicable
to the analysis of temperature using a thermographic camera in other tests, for
example, in the drilling of other materials (wood, metals, resins, alloys, new
materials). During the tests it is important to keep the bone at body temperature
(37 C) or adjusted to the temperature of the tool to isolate the increase of
temperature during the drilling process. With the use of the thermographic
camera, we can isolate the average temperature of the elements in the test, as
well as the peaks and lows, the highest and lowest puntual temperature, the
highest and lowest temperature of the area around the tool before the drilling, of
the area of the material, etc.
The continuous register of temperature allowed us study the cooling of the
surface following the heating of the surface due to the drilling. We did a linear
regression of the data according to the time getting this adjustment:
(1)
T = 14.1566 0.112166*Time,
with a cooling rate of 0.11/s. The cooling is due to the deepest layers in the
bone which cool the outer layers heated during previous drillings.
The P-value of the analysis of variance (p<0.0001) confirms the relation,
statistically significant, between temperature and time with a probability of
95.0%. The adjusted model only explains 46.97% of the variability of data, but
it is evident the tendency of data to the cooling in the surface layer of the bone.
257
Since the first contact between tool and bone, the recorded temperature starts to
increase, and in a following study of data, a few points can be made.
The starting temperature in the drilling point of the bone is 14.415 C. The
temporal sequence of the temperature in the drilling point is shown in this chart:
Table 1
Temperature in the drilling point
0
14.42
1
16.96
2
18.40
3
21.81
4
27.61
5
26.49
6
32.11
7
46.18
8
51.25
The drilling lasted 8.5 seconds. This is the chart for the increase of
temperature according to the time.
There is a tendency in the increase of temperature as the drill bit drills the
bone. Some adjustments were made for the analysis of data, and the closest one
was:
(2)
T = 16.5913 + 0.5518*Time^2.
258
more precise. Finally, we can concluse that the applicability of the temporal
sequence of thermograms for dynamic study of the temperature in drilling
operations to implant dentistry, is shown to be very high. While there are
several aspects to consider when conducting research like this.
Received:23.02.2010
1
University of Valladolid-Spain,
Department of Materials Science and
Metallurgical Engineering-Mechanical Engineering
Valladolid-Spain
e-mail: roblop@eis.uva.es
2
Gheorghe AsachiTechnical University,
Department of Machine Manufacturing Technology
Iai, Romania
e-mail: negoescu@tcm.tuiasi.ro
1. Introduction
Sustainable Development permanently represents the crucial goal of
mankind and under this imperative, all the human activities, both social and
economical and of course the productive ones must able to assure the harmony
between profit, people and Earth. It is widely accepted that from engineering
point of view, industrial society in general will reach Sustainable Development
only via Sustainable Manufacturing and via Sustainable Design. In this sense,
industrial development activities starting from plant engineering and design
must be compulsory considered in relation to advanced concepts such as Life
Cycle Analysis (LCA) and Extended Producer Responsibility (EPR). The
260
On-site recycle
Raw material
extraction
Product
manufacturing
Product
Reuse
or Recycle
Raw material
processing
Cradle to cradle
Product
use
Product
Packaging
LCA
Product
marketing
Product
transportation
and storage
261
Waste
Reduction
Product
Modification
Process Change
Better
housekeeping
Process
control
RECYCLING
OFF-SITE
Recycling
On-site
Recycling
GLASS
RECYCLING
Technology
Raw material
change
change
262
Fig. 3 Statistic data on glass recycling in Spain from 2001 until 2008 [2].
263
Fig. 4 Evolution of glass recycling rate in Spain from 2001 until 2008.
264
265
4. Conclusions
1. The presented study and the obtained synthesis results have proved the
adequacy of implementing a new glass recycling factory in the selected location
and they are useful for efficient establishment that plant for glass recycling.
2. A manufacturing process design, together with a study of the technical,
human and economic necessary for the plant will complete the best design of
the plan to get a finished product with the required quality and the lowest
possible cost.
266
Universidad de Valladolid,
Valladolid, Spain
e-mail: olga.marina.montes@gmail.com
2
Gheorghe Asachi Technical University,
Department of Machine Manufacturing Technology
Iasi, Romania
e-mail: merticaru@tcm.tuiasi.ro
REFERENCES
1. C a s a l s-C a s a n o v a, M., Complejos Industriales. Editions UPC, Catalua
Polytechnic University, 2005.
2. Ecovidrio - Informe anual 2008, Available from: - http://www.ecovidrio.es/app/
GeneraPaginas.asp?seccion=../app/WebEcovidrioNet/wEstadisiticasRecicladoTo
talNacional.aspx, Accessed: 02/02/2010.
3. E l-H a g g a r, S. M., Sustainable Industrial Design and Waste Management: Cradleto-cradle for Sustainable Development. ISBN-10: 0123736234, ISBN-13:
9780123736239, 406 p, Publisher: Academic Press, 2007. Available from:
http://www.engineeringvillage.com. Accessed: 2008-02-02.
4. G a l i s M. et al., Digital product development for the entire product life cycle.
Academic Journal of Manufacturing Engineering, 6, 3 (2008), Timisoara,
Romania, p.55-60
5. M e r t i c a r u V.jr., M u s c a G., A x i n t e E., PLM in Relation to SCM and CRM,
for Integrating Manufacturing with Sustainable Industrial Design, Proceedings
of ICOVACS 2008: International Conference On Value Chain Sustainability,
Izmir, Turkey, November 12-14, 2008, Izmir University of Economics
Publication No: IEU-026, ISBN 978-975-8789-25-2, pp.109-118.
1. Introduction
The problem of thermal protection in hostile environments is certainly not
new, but it does not loose its interest due to the fact that there are large range of
activities requiring such protective equipment, from winter and extreme sports,
cycling, motorcycling to working in harsh winter conditions. There are a lot of
studies regarding developments of textile fabrics and materials to improve
thermal insulation. Still, the fluid flow around the garments in dynamic
conditions is less investigated and the problem of designing textile structures
with improved behaviour in relation to air flow remains open to debate.
268
In order to formulate solutions for the fabric structure one must understand
the nature of the problem. The conditions characterising a hostile environment
include among others strong winds. The flow of these air currents is different,
according to the type of surface they encounter. Larger surfaces, with a planar
distribution present a laminar flow of the air currents, while complex 3D
surfaces are characterised by the formation of pressure fields and turbulences
that affect the thermal behaviour of the exterior layer in the garment by
reducing its insulation. Areas like the arms, hips or shoulders, with smaller 3D
surface present problems in relation to the laminar air flow. Previous studies
regarding the air flow around a human body show that these are the critical
zones in a protective garment. Due to its specific nature, the shape of these areas
in a protective garment cannot be modified. Therefore, the improvement of air
flow must be based on modifying the fabric surface in the critical areas so that
the pressure fields are diminished. The fabric surface has to be designed so that
the air currents are channeled toward the exterior, avoiding turbulences. From
this point of view, the 3D knitted fabrics represent the best solution. Apart from
the shape of their cross section, it is important to define the optimum paths for
these channels and their position in the protective garment.
2. Definition of Air Flow at Garment Level
An initial analysis was carried out in order to identify the air flow critical
areas on the human body. Its purpose was to determine the air pressure and
velocity distribution within the fluid. The model represented therefore a fluid
mesh and not one of the solid body. It was created with the ALGOR v12 FEA
software package, using a number of 491 2D elements (558 nodes), defined in
Fig. 1. The FEA model is presented in Fig. 2. The body is considered an
obstacle, while the environment is meshed using bidimensional finite elements.
269
t = 15 s
t = 80 s
t=3s
t=3s
t =80 s
Fig. 5 Velocity fields distribution
t = 70 s
Fig. 6 Turbulences distribution
The images presented show that the turbulences in the air flow, as well as
higher velocities are characteristic to some parts of human body the shoulders,
270
the hips and the chest. Such critical areas are susceptible to generate discomfort
in windy conditions.
A solution must be found to diminish these turbulences and the pressure
fields encountered. From the fabric point of view, the air flow at surface level
can be controlled through: the type of raw material, fabric structure and
structural parameters. The most important factor is the fabric structure, mainly
the surface geometry. Weft knitted fabrics offer a large range of structural
possibilities and are characterised by good elasticity and high formability. In
comparison with other types of fabrics, especially woven, the weft knitted
fabrics can be produced with complex 3D architectures that are specific only to
them and do not require costly technological changes on the machines.
3. Models For The Air Flow At Shoulder Level
The models were transferred to the ALGOR v20 FEA software package to
be analyzed. The model was meshed automatically resulting a number of 96,584
3D elements (bricks) and a number of 108,512 nodes. The fluid considered is
the air, its flow velocity chosen to be 50 km/h. This simulates an extremely
hostile environment.
271
The fluid properties were selected from the Material Library Manager and
are presented in Fig. 9. The meshed model for the shoulder is illustrated in
Fig.10, while the meshed model for the shoulder and 3D fabric is presented in
Figure 11.
The models were analyzed using finite elements, based on steady fluid flow
processor. The parameters considered are illustrated in Figs. 12 and 13.
272
The use of a 3D fabric with rectangular channels changes the air flow. The
velocity distribution is different in comparison to the situation described for the
shoulder without fabric. The areas characterized by high air velocity are
distanced from the shoulder areas suggested by Figs. 15, 16 and 17.
273
Fig. 17 Air velocity distribution for shoulder and 3D fabric top partial view
From Figs. 15-17 it results that the air flow is improved, diminishing the air
velocity at shoulder level and implicitly the air turbulences.
4. Conclusions
1. FEA studies concerning protective equipment for cold environments have
shown that the air flow at body level presents critical areas where the flow is
characterized by pressure fields and turbulences, causing heat loss and
discomfort. It is the case of hips, chest and shoulders.
2. One solution for the improvement of the air flow is using fabrics with 3D
structure or 3D surface effects. These knitted fabrics are produced on flat
knitting machines without any adaptation required. The paper presents two
structural variants with 3D effects and 3D architecture rib fabric with relief
effects and sandwich fabrics with complex shapes. The fabrics are characterized
and their 3D structure is discussed, emphasizing their advantages in obtaining
surface effects that have the potential to control the air flow.
274
3. Two FEA studies were carried out, one at 2D level, in order to identify
the air flow critical areas. The 2D analysis does not ensure complete data
regarding the in depth air flow so a further 3D finite element analysis was
required. This analysis indicated the flow phenomenon dimensions at shoulder
level and allowed a comparison between air flow at shoulder level and the air
flow when a fabric is added.
4. The study took into consideration two models of the shoulder: without
and with fabric. The fabric selected was the sandwich fabric with rectangular
channels with an uniform distribution.
5. The study shown that the use of this fabric brought an improvement in air
flow when used at shoulder level. The study requires additional work taking
into account the followings: fabrics with other 3D effects (differently shaped
channels circular) and non-uniform distribution of these channels (patterned
distribution of the 3D effect). One important target is to determine the shape of
the channel path. These studies will lead to the optimization of garments used
for protection against cold.
Received: March 20, 2010
REFERENCES
1.
2.
3.
4.
5.
275
Index autori
A
Alexandrescu Adina 189
Alexandrescu Adrian 189
Alexandrescu Irne 227
Alexandrescu Aurora 189
Antoniadis Aristomenis 9,21
B
Baciu Lupacu Radu 91
Bloiu Victor 197
Bran Marian 197
Brglzan Mircea 205
Bdru Rodica 205
Belis Taxiarchis 9
Berbinschi Silviu 41,49
Bordeau Ilare 197
Bostan Ion 135
C
Calfa Daniel 161
Ciobanu Luminia 267
Clrau Doru 153
Chiri Constantin 161
Ciobanu Bogdan 153
Cozmnc Irina 97
Cozmnc Mircea 65
Croitoru Cristian 65
D
Dobnd Eugen 205
Dogaru Constantin 31
Dulgheru Valeriu 135
Dumitra Ctlin 267
Dumitracu Nicolae 41,49
Dumitrescu Ctlin 167
Dumitrescu Liliana 167
Dumitru Dumitru 31
Dua Petru 91,237
F
Fetecu Ctlin 57
Filip Cristina 117
Franke Hans Joachim 227
Fulga Ileana 143
G
Gonalves Coelho Antonio 1
H
Hanganu Adrian 161
Haraga Georgeta 167
Horodinc Mihi 125
I
Ibnescu Radu 97
L
Loghin Carmen 267
Lopez Roberto 251
M
Manole Iolanda 83
Murad Erol 167
Mnescu Mihai 177
Marsavina Liviu 197
Milo Teodor 205
Manea Adriana 205
Mihalache Andrei 245
Martn Oscar 251
Montes Marina Olga 259
Merticaru V. Vasile 259
Matei Ana Maria 75
Milea Marius Nicolae 75
Mocanu Costel 57
Mourao Antonio 1
N
Nag Gheorghe 83,245
Neagoe Lavinia 117
Novac Drago 197
Negru Radu 197
Negoescu Florin 251
Netian Gabriela 1
O
Oancea Nicolae 41,49
P
Panaitescu Valeriu 167
Pavlov Olimpia 31
Petre Dan 117
Petre Ioana 117
Popoviciu Mircea 197
R
Rpanu Marius Ionu 83,245
Rotman Iustina 91
S
Sahin Mehmet 267
San Juan Manuel 251
Santos Francisco 251
Scurtu Dan 153
Sochireanu Anatol 135
V
Vietor Thomas 227
Vlad Daniel Viorel 57
Vod Mircea 197
W
Weingold Andrei 83
Y
Yaldz Suleyman 267
Z
Zahariea Dnu 215, 221