CM Fasc.2 2010

BULETINUL
INSTITUTULUI
POLITEHNIC
DIN IAI
Publicat de
UNIVERSITATEA TEHNIC GHEORGHE ASACHI, IAI
Tomul LVI (LX)

Fasc. 2
Secia
CONSTRUCII DE MAINI
2010
BULETINUL INSTITUTULUI POLITEHNIC DIN IAI

PUBLISHED BY
GHEORGHE ASACHI TECHNICAL UNIVERSITY OF IAI

Editorial Office: Bd. D. Mangeron 63, 700050, Iai, ROMANIA
Tel. 40-232-278683; Fax: 40-232-237666; e-mail: polytech@mail.tuiasi.ro
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
MACHINE CONSTRUCTION Section
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
Editorial Advisory Board

Prof. dr. eng. Nicuor Amariei, Gheorghe Asachi
Technical University of Iai, Romania
Assoc.Prof.dr.eng. Aristomenis Antoniadis, Technical
University of Crete, Greece
Prof. dr. eng. Virgil Atanasiu, Gheorghe Asachi
Prof. dr. eng. Manuel San Juan Blanco, University of
Valladolid, Spain
Prof. dr. eng. Petru Berce, Technical University of Cluj
Napoca, Romania
Prof. dr. eng. Ion Bostan, Technical University of
Chiinu, Repablic of Moldova
Prof. dr. eng. Walter Calles, Hochschule fr Technik und
Wirtschaft des Saarlandes, Saarbrcken, Germany
Prof. dr. eng. Doru Clrau, Gheorghe Asachi
Prof. dr. eng. Francisco Chinesta, Ecole Centrale de
Nantes, France
Assoc.Prof.dr.eng. Conalves Coelho, University Nova of
Lisbon, Portugal
Assoc.Prof.dr.eng. Mircea Cozmnc, Gheorghe Asachi
Prof. dr. eng. Spiridon Creu, Gheorghe Asachi
Prof. dr. eng. Gheorghe Dumitracu, Gheorghe Asachi
Prof. dr. eng. Ctlin Fetecu, University Dunrea de
Jos of Galai, Romania
Prof. dr. eng. Mihai Gafitanu, Gheorghe Asachi
Prof. dr. eng. Radu Gaiginschi, Gheorghe Asachi
Prof.dr.ir.Dirk Lefeber, Vrije Universiteit Brussels, Belgium
Prof. dr. eng.Dorel Leon, Gheorghe Asachi Technical
University of Iai, Romania
Prof. dr. eng. James A. Liburdy, Oregon State University,
Corvallis, Oregon, SUA
Prof. dr. eng. dr. H.C. Peter Lorenz, Hochschule fr
Technik und Wirtschaft, Saarbrcken, Germany
Prof. dr. eng. Noura-Barbu Lupulescu, University

Transilvania of Braov, Romania
Prof. dr. eng. Francisco Javier Santos Martin,
University of Valladolid, Spain
Prof. dr. eng. Fabio Miani, University of Udine, Italy
Prof. dr. eng. Mircea Mihailide, Gheorghe Asachi
Prof. dr. eng. Sevasti Mitsi, Aristotle University of
Thessaloniki Salonic, Greece
Prof. dr. eng. Gheorghe Nag, Gheorghe Asachi
Prof. dr. eng. Vasile Neculiasa, Gheorghe Asachi
Prof. dr. eng. Dumitru Olaru, Gheorghe Asachi
Prof. dr. eng. Cezar Oprian, Gheorghe Asachi
Prof. dr. eng. Juan Pablo Contreras Samper,
University of Cadiz, Spain
Prof. dr. eng. Loredana Santo, University Tor Vergata,
Rome, Italy
Prof. dr. eng. Cristina Siligardi, University of Modena,
Italy
Prof. dr. eng .Fernando Jos Neto da Silva, University
of Aveiro, Portugal
Prof. dr. eng. Filipe Silva, University of Minho, Portugal
Prof. dr. eng. Laureniu Sltineanu, Technical
University of Iai, Romania
Lecturer dr.eng. Birgit Kjrside Storm, Aalborg
Universitet Esbjerg, Denmark
Prof. dr. eng. Ezio Spessa, Politecnico di Torino, Italy
Prof. dr. eng. Alexei Toca, Technical University of
Chiinu, Repablic of Moldova
Prof. dr. eng. Roberto Teti, University Federico II,
Naples, Italy
Prof. dr. eng. Hans-Bernhard Woyand, Bergische
University Wuppertal, Germany
Papers presented at
THE INTERNATIONAL CONFERENCE on
DESIGN, TECNOLOGIES & MANAGEMENT
IN MANUFACTURING
Iai, May 14th 16th, 2010
organized by the
FACULTY OF MACHINE MANUFACTURING &

INDUSTRIAL MANAGEMENT
Papers published with the support of

NATIONAL AUTHORITY for SCIENTIFIC RESEARCHERS
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

BULLETIN OF THE POLYTECHNIC INSTITUTE OF IAI
Publicat de
UNIVERSITATEA TEHNIC GHEORGHE ASACHI DIN IAI
Tomul LVI(LX), Fasc. 2
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.)........................
TAXIARCHIS BELIS i ARISTOMENIS ANTONIADIS, Model pentru

evaluarea uzurii dinilor frezelor de danturat pe baza determinrii
achiilor 3D (engl., rez. rom.). ..................................................................
NIKOLAOS TAPOGLOU i ARISTOMENIS ANTONIADIS, Determinarea

prin simulare CAD a componentelor forei de achiere la frezarea
danturilor (engl., rez. rom.)........................................................................
21
EUGEN STRJESCU, CONSTANTIN DOGARIU, OLIMPIA PAVLOV i

DUMITRU DUMITRU, Contribuii privind controlul informatizat al
cuitelor roat (engl., rez. rom.).................................................................
31
SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRACU i

NICOLAE OANCEA, Contribuii la elaborarea unei metode grafice
pentru profilarea sculelor care genereaz prin nfurare. I. Algoritm.
(engl., rez. rom.) .......................................................................................
41
SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRACU i

NICOLAE OANCEA, Contribuii la elaborarea unei metode grafice
pentru profilarea sculelor care genereaz prin nfurare. II. Aplicaie
pentru profilarea sculei-cremalier (engl., rez. rom.)................................
49
CTLIN FETECU, DANIEL-VIOREL VLAD i COSTEL MOCANU,

Simularea procesului de strunjire folosind analiza cu element finit
(engl., rez. rom.)........................................................................................
57
MIRCEA COZMNC, CRISTIAN CROITORU i CTLIN

UNGUREANU, Cercetri experimentale pentru validarea unei noi
metode de evaluare a forelor de achiere (engl., rez. rom.) .....................
65
ANA-MARIA MATEI i MARIUS NICOLAE MILEA, Forele de achiere

la frezarea frontal n funcie de forele dezvoltate la nivelul unui dinte
(engl., rez. rom.)........................................................................................
75
MARIUS-IONU RPANU, GHEORGHE NAG, IOLANDA-ELENA

MANOLE i ANDREI WEINGOLD, Aspecte comparative privind
croirea la operaia de decupare-perforare pe prese clasice i centre de
presare prevzute cu comand numeric (engl., rez. rom.).......................
83
IUSTINA ELENA ROTMAN, PETRU DUA i RADU ADRIAN BACIU

LUPACU, Consideraii teoretice i experimentale privind
determinarea efectului de divergen (engl., rez. rom.) ...........................
91
CTLIN UNGUREANU, RADU IBNESCU i IRINA COZMNC,

Sistem de msurare computerizat (engl., rez. rom.).................................
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
EROL MURAD, CTALIN DUMITRESCU, GEORGETA HARAGA i

LILIANA DUMITRESCU, Sistem de msurare pneumatic a masei de
ap extras n procesele de uscare convectiv (engl., rez. rom.)............... 167
MIHAI FLORIN MNESCU i VALERIU PANAITESCU, Tehnologii
folosite pentru monitorizarea defeciunilor structurale aprute n
funcionarea grupurilor eoliene (engl., rez. rom.)...................................... 177
AURORA ALEXANDRESCU, ADINA SIMONA ALEXANDRESCU i
ADRIAN CONSTANTIN ALEXANDRESCU, Reabilitarea staiei de
pompare pentru alimentare cu ap (engl., rez. rom.)................................. 189
ILARE BORDEAU, MIRCEA OCTAVIAN POPOVICIU, DRAGO
NOVAC, LIVIU MARSAVINA, RADU NEGRU, MIRCEA VOD,
VICTOR BLOIU i MARIAN BRAN, Contribuii n evaluarea
durabilitii arborilor hidroagregatelor axiale orizontale (engl., rez.
rom.) ......................................................................................................... 197
TEODOR MILO, MIRCEA BRGLZAN, EUGEN DOBND,
ADRIANA MANEA, RODICA BDRU i DANIEL STROI,
Traseul optim al unei conducte de aduciune utiliznd algoritmul
Bellman-Kalaba (engl., rez. rom.) ............................................................ 205
DNU ZAHARIEA i MIHAELA TUDORACHE, Analiza structural a
cuplajelor fixe de tip manon cu tifturi cilindrice (engl., rez. rom.) ....... 215
DNU ZAHARIEA i MARIUS STACHIE, Analiza structural a arcurilor
bimetalice lamelare (engl., rez. rom.) ....................................................... 221
IRNE ALEXANDRESCU, HANS-JOACHIM FRANKE i THOMAS
VIETOR, Managementul cunoaterii n configurarea produselor
complexe personalizate n fazele incipiente ale dezvoltrii acestora
(engl., rez. rom.)........................................................................................ 227
PETRU DUA i IULIANA LAURA TARANOVSCHI, Cercetri cu privire
la activitatea de inovare din mediul tehnic (engl., rez.
rom.).......................................................................................................... 237
ANDREI MIHALACHE, GHEORGHE NAG i MARIUS-IONU
RPANU, Avantaje i puncte slabe ale diferitelor tehnici de inginerie
invers (engl., rez. rom.) ........................................................................... 245
ROBERTO LOPEZ, MANUEL SAN JUAN, FRANCISCO SANTOS,
OSCAR MARTN i FLORIN NEGOESCU, Termografie aplicat n
frezarea osoas (engl., rez. rom.)............................................................... 251
OLGA MARINA MONTES i VASILE V. MERTICARU, Studiu asupra
oportunitii unei noi fabrici de reciclare a sticlei, pentru o dezvoltare
regional durabil (engl., rez. rom.).......................................................... 259
CTLIN DUMITRA, CARMEN LOGHIN, SULEYMAN YALDIZ,
MEHMET SAHIN i LUMINIA CIOBANU, Modelarea 3D a
suprafeelor textile cu destinaia optimizrii curgerii fluidelor (engl.,
rez. rom.).................................................................................................... 267

BULLETIN OF THE POLYTECHNIC INSTITUTE OF IAI
Published by the
GHEORGHE ASACHI TECHNICAL UNIVERSITY OF IAI
Tomul LVI(LX), Fasc. 2
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). .............
TAXIARCHIS BELIS and ARISTOMENIS ANTONIADIS, Hobbing Wear

Prediction Model Based on 3D Chips Determination (English,
Romanian summary).................................................................................
NIKOLAOS TAPOGLOU and ARISTOMENIS ANTONIADIS, CADBased Calculation of Cutting Force Components in Gear Hobbing
(English, Romanian summary)..................................................................
21
EUGEN STRJESCU, CONSTANTIN DOGARIU, OLIMPIA PAVLOV

and DUMITRU DUMITRU, Contributions Concerning The Computer
Aided Control of the Fellows' Cutter (English, Romanian
summary)...................................................................................................
31
SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRACU

and NICOLAE OANCEA, Contributions to the Elaborations of a
Graphical Method for Profiling of Tools which Generate by
Enveloping. I. Algorithms (English, Romanian summary).......................
41
SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRACU

and NICOLAE OANCEA, Contributions to the Elaborations of a
Graphical Method for Profiling of Tools which Generate by
Enveloping. II. Application for Rack-Gear Tools Profiling (English,
49
CTLIN FETECU, DANIEL-VIOREL VLAD and COSTEL

MOCANU, The Numerical Simulation of Turning Process using Finite
Element Modeling (English, Romanian summary)...................................
57
MIRCEA COZMNC, CRISTIAN CROITORU and CTLIN

UNGUREANU, Experimental Researches Regarding a New Method
for Cutting Forces Evaluation (English, Romanian summary)................
65
ANA-MARIA MATEI and MARIUS NICOLAE MILEA, Face Milling

Forces Depending on the Forces Developed on a Single-Tooth (English,
75
MARIUS-IONU RPANU, GHEORGHE NAG, IOLANDA-ELENA

MANOLE and ANDREI WEINGOLD, Comparative Aspects
Regarding the Nesting for Blanking-Punching Operation on Classical
Presses and Numerical Commanded Pressing Centers (English,
83
IUSTINA ELENA ROTMAN, PETRU DUA and RADU ADRIAN BACIU

LUPACU, Theoretical and Experimental Considerations on
Determining the Effect of Divergence (English, Romanian
summary)...................................................................................................
91
CTLIN UNGUREANU, RADU IBNESCU and IRINA COZMNC,

Computerized Measurement System (English, Romanian summary).....
97
BIRGIT KJRSIDE STORM, Adhesive Bonding of Surface Treated

Aluminium (English, Romanian summary).............................................. 105
IOANA PETRE, DAN PETRE, CRISTINA FILIP, and LAVINIA
NEAGOE, Industrial Applications of the Pneumatic Muscles (English,
Romanian summary)................................................................................. 117
MIHI HORODINC, Some New Resources on Computer Assisted
Experimental Research of the Absorbed Electric Power in
Manufacturing Systems (English, Romanian summary)........................... 125
ION BOSTAN, VALERIU DULGHERU and ANATOL SOCHIREANU,
Integrated CAE Development of Precessional Drives using Autodesk
Inventor Platform (English, Romanian summary).................................... 135
ILEANA FULGA and EUGEN STRJESCU, Proposals for the
Improvement of the Fluidic Spiral Jetmills' Activity (English,
DORU CLRAU, IRINA TIA, DAN SCURTU and BOGDAN
CIOBANU, Dynamic Analysis of Mechanical-Hydraulic Protection
Mechanisms of Low Power Horizontal Axis Wind Turbines through
Vertical Tilting (English, Romanian summary)........................................ 153
CONSTANTIN CHIRI, ADRIAN HANGANU and DANIEL CALFA,

Research on Hydraulically Systems which Move Heavy Masses on
Small Distances with Lower Frequencies (English, Romanian
summary)................................................................................................... 161
EROL MURAD, CTALIN DUMITRESCU, GEORGETA HARAGA and
LILIANA DUMITRESCU, Pneumatic Metering System for Amount of
Water Extracted in Convective Drying Processes (English, Romanian
summary)................................................................................................... 167
MIHAI FLORIN MNESCU and VALERIU PANAITESCU, Technologies
for Monitoring Structural Damages Arising in the Functioning of Wind
Turbines (English, Romanian summary)................................................... 177
AURORA ALEXANDRESCU, ADINA SIMONA ALEXANDRESCU and
ADRIAN CONSTANTIN ALEXANDRESCU, Pumping Station
Exoneration for Water Supply (English, Romanian summary)................. 189
ILARE BORDEASU, MIRCEA OCTAVIAN POPOVICIU, DRAGOS
NOVAC, LIVIU MARSAVINA, RADU NEGRU, MIRCEA VODA,
VICTOR BALASOIU and MARIAN BRAN, Contributions regarding
Durability Evaluation of Horizontal Axial Hydraulic Turbines Shafts
(English, Romanian summary).................................................................. 197
ADRIANA MANEA, RODICA BDRU and DANIEL STROI,
Optimal Routes of Pipeline Supply using the Bellman-Kalaba
Algorithm (English, Romanian summary)................................................ 205
DNU ZAHARIEA and MIHAELA TUDORACHE, Structural Analysis
of a Sleeve Rigid Coupling with Cylindrical Pins (English, Romanian
summary)................................................................................................... 215
DNU ZAHARIEA and MARIUS STACHIE, Structural Analysis of a
Bimetallic Strip Thermostat (English, Romanian summary).................... 221
IRNE ALEXANDRESCU, HANS-JOACHIM FRANKE and THOMAS
VIETOR, Knowledge Management for the Configuration in Early
Phases of Complex Custom Products (English, Romanian summary)...
227
PETRU DUA and IULIANA LAURA TARANOVSCHI, Researches
Regarding to Innovative Activity in Technical Environment (English,
ANDREI MIHALACHE, GHEORGHE NAG and MARIUS-IONU
RPANU, Advantages and Weak Points of Different Reverse
Engineering (RE) Techniques (English, Romanian summary)................. 245
ROBERTO LOPEZ, MANUEL SAN JUAN, FRANCISCO SANTOS,
OSCAR MARTN and FLORIN NEGOESCU, Thermography Applied
to Bone Drilling (English, Romanian summary)....................................... 251
OLGA MARINA MONTES and VASILE V. MERTICARU, Study on the
Opportunity of a New Glass Recycling Factory for Regional
Sustainable Development (English, Romanian summary)........................ 259
CTLIN DUMITRA, CARMEN LOGHIN, SULEYMAN YALDIZ,

MEHMET SAHIN i LUMINIA CIOBANU, Modeling 3D Surface of
Textile Structures for Fluid Flow Improvement (English, Romanian
summary)................................................................................................... 267

Publicat de
Universitatea Tehnic Gheorghe Asachi din Iai
Tomul LVI (LX), Fasc. 2, 2010
Secia
CONSTRUCII DE MAINI
ON THE PATTERN OF THE DESIGN MATRIX IN

REDUNDANT DESIGN SOLUTIONS
BY
ANTNIO M. GONALVES-COELHO1, GABRIELA NETIAN2,

and ANTNIO MOURO1
Abstract. Axiomatic Design was created with the aim of building a
systematic model for engineering education and practice, taking into
account the initial hypothesis that there are fundamental principles that
govern good design practice. According to this design theory, the design
solutions can be classified as uncoupled, decoupled or coupled, depending
on the way their design matrices are populated. Uncoupled solutions are the
best, decoupled solutions are acceptable, and coupled solutions are poor
design and should be avoided. Redundant designs make up a specific class
of design solutions, in which the number of functional requirements is
lesser than the number of design parameters. This paper discusses how the
design matrix could be populated, so that a redundant design could be either
uncoupled or decoupled.
Keywords: Axiomatic Design, Design Matrix, Redundant Design, Ideal
Design.
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
Antonio M. Gonalves-Coelho et al.
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].
Fig. 1 The Design Domains [1].
The left-to-right mapping between any two contiguous domains can be

represented by a design equation of the form:
(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,
Bul. Inst. Polit. Iai, t. LVI (LX), f. 2, 2010
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.
Fig. 2 Example of a simple redundant design.
The adjustment of d can be achieved by suitably setting the values of two

design parameters: the position of the cam in the end of the hand lever (see Fig.
1), which is denoted by angle , and the angular position of the threaded rod,
which is represented by angle . Thus, the design equation for this redundant
design with one requirement and two parameters is:
(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
FR2 = A24 A25 A26 DP5 .

(7)
FR A
3
34 A35 A36 DP6
As a result, the redundant design of Eq. (3) could be considered suitable

design if the condensed equations obtained from Eq. (4), Eq. (5) and Eq. (6)
are not coupled. The coupled condition is excluded in Eq. (4) and Eq. (5) if the
relevant DPs are chosen so that their non-zero block matrices are either
triangular or diagonal.
As for Eq. (6), the coupled condition is excluded if its non-zero block
matrix is populated in such a manner that the condensed equations correspond
to uncoupled or decoupled designs. If this is not the case, one can selectively
freeze as many DPs as required, so that the condensed design become
uncoupled or decoupled, as exemplified elsewhere [3].
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
UNIDEMI, Universidade Nova de Lisboa,

Faculdade de Cincias e Tecnologia
Departamento de Engenharia Mecnica e Industrial
Monte de Caparica, Portugal
e-mail: goncalves.coelho@fct.unl.pt
ajfm@fct.unl.pt
2
Ministerul Comunicaiilor i Societii Informaionale,
Bucureti, Romnia
e-mail: gabriela.nestian@mcsi.ro
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.
MODELUL MATRICEI DE PROIECTARE IN CAZUL SOLUTIILOR

REDUNDANTE DE PROIECTARE
(Rezumat)
Proiectarea axiomatic a aprut din necesitatea crerii unui model sistematic de
educare i practic inginereasc, pornind de la principiile fundamentale de bune practici
ale proiectrii. Conform acestei teorii, n funcie de modul de populare al matricelor de
proiectare, soluiile de proiectare pot fi clasificate n soluii necuplate, decuplate sau
cuplate. Soluiile de tip necuplat sunt cele mai bune, soluiile de tip decuplat sunt
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.

Publicat de
Secia
CONSTRUCII DE MAINI
HOBBING WEAR PREDICTION MODEL

BASED ON 3D CHIPS DETERMINATION
BY
TAXIARCHIS BELIS and ARISTOMENIS ANTONIADIS

Abstract. Gear hobbing is a machining process widely used in the industry
for massive production of external gears. In this process the variant chip formation
on each generating position causes uneven wear distribution on the hob teeth. This
paper presents a new method for the determination of wear parameters based on
3D chips. The 3D chip characteristics are fed to an existing wear model in order to
calculate the hob wear distribution more precisely.
Key words: gear hobbing, tool wear, simulation
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
Taxiarchis Belis and Aristomenis Antodiadis
tool tangential displacement. Considering the above kinematic, two different

hobbing types may be applied according to the direction of the axial feed, the
climb and the up-cut hobbing respectively. In the present paper the tooths
profile generation complies with DIN 3972 [1].
Fig. 1 Basic kinematics and essential parameters of gear hobbing.
2. State of the Art

As mentioned before, wear prediction in gear hobbing is a challenging task
because of the large amount of wear influencing data and the complex
kinematics of the process. Many studies have been published, introducing
different methods for wear determination. These studies can fall under two
categories. The first includes methods based on wear calculation in individual
generating positions [2], [3].These methods are based on the simulated chip
geometry and cutting conditions. Lately, FEM-based methods for wear
determination have been introduced [4]. These methods take into account
occurring stresses in the tool-chip contact areas. Nevertheless a FEM-based
simulation needs a considerable amount of time before it can be realised.
The wear model presented in [2] is used in the present paper,. The
contribution of the proposed method is the precise 3D chips determination
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.
Fig. 2 Determination of the flank wear at individual hob teeth

considering the shift data.
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
3.1. HOBWEAR Formulation

Figure 3 presents wear laws for different chip groups. There are five chip
groups concerning the chip flow obstruction intensity, introduced by [3]. The
determination of the chip group involves prior knowledge of the cutting
direction and the examined gear flank. As it can be easily noticed, the wear rate
in group I is more intense. On the other hand, cutting with chip group 0 causes
less flank tooth wear and the achieved number of cuts increases.
This can be explained by the fact that in group I there is intense chip flow
obstruction. In the region of the tooth head chip flow obstruction phenomena
are the most intense due to collision of chip distinct sections. In the bottom side
of the figure the included equivalent chip thickness and length are further
explained.
Figure 4 presents the general structure of the program developed. The left
part of the figure shows the categorized data input. Data such as chip group,
equivalent thickness and length are loaded into the program from an external
source. The undeformed chip geometries are calculated in various generating
and revolving positions with the aid of HOB3D [2], [3].
Fig. 4 HOBWEAR simulation algorithm.
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.
Fig. 5 Determination of equivalent chip dimensions.
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.
Fig. 8 Wear progress in individual generating positions.
18
Fig. 9 Total wear distribution on the hob cutter after shifting.
As it can be seen in the second case, the wear distribution is more

uniform due to the fact that more teeth are involved in the cut. Authors in [10]
have studied the optimal selection of the shift displacement and number of
shifts and introduced a nomogram for various shifting conditions.
Fig. 10 Cutting of 18 gears with different tangential displacement parameters.
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
Technical University of Crete,

Department of Production Engineering & Management
Chania, Crete, Greece
e-mail: tbelis@isc.tuc.gr
antoniadis@dpem.tuc.gr
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
Hobbing and Computational Interpretation of Experimental Results. ASME

Journal of Manufacturing Science and Engineering, 124, 4, 784-791, (2002).
9. 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 Journal of Manufacturing Science and Engineering, 124, 4, 792-798,
(2002).
10. B o u z a k i s K., A n t o n i a d i s A., Optimizing of Tangential Tool Shift in Gear
Hobbing, Annals of the CIRP, 44, 1 (1995).
MODEL PENTRU EVALUAREA UZURII DINILOR FREZELOR DE

DANTURAT PE BAZA DETERMINRII ACHIILOR 3D
(Rezumat)
Frezarea danturii este un procedeu de prelucrare larg rspndit n industrie, n
special pentru producia de serie a danturilor exterioare. n acest proces, modul diferit
de formare a achiei n fiecare poziie de generare a profilului duce la uzura neuniform
a dinilor frezei. Lucrarea prezint o nou metod de determinare a parametrilor uzurii,
frezelor pe baza achiilor 3D. Caracteristicile achiilor 3D au fost stabilite pornind de la
un model existent de uzur, n vederea determinrii cu mai mult acuratee a distribuiei
uzurii la nivelul dinilor frezei.

Publicat de
Secia
CONSTRUCII DE MAINI
CAD-BASED CALCULATION OF CUTTING FORCE

COMPONENTS IN GEAR HOBBING
BY
NIKOLAOS TAPOGLOU and ARISTOMENIS ANTONIADIS

Abstract. One of the most commonly used gear manufacturing process
especially for external is gear hobbing. The optimisation of the process of gear
hobbing is of great importance in the modern industry. This paper presents a novel
simulation program called HOB3D that can simulate the cutting process in a
commercial CAD environment, thus producing results with the optimal precision.
The outputs of the program include the 3D chip and gap geometry as well as the
developing cutting forces which were validated using experiments.
Key words: gear hobbing, cutting forces, simulation.
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
Nikolaos Tapoglou and Aristomenis Antoniadis
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.
Fig. 1 Gear hobbing.
Gear hobbing process is affected by a series of parameters which be divided

in three categories: hob, gear and process parameters. The first include module
(m), external diameter (dh), number of origins (z1) and number of columns (ni)
of the hob. Gear parameters are number of teeth (z2), helix angle (ha) and gear
width. Finally, process parameters include axial feed (fa) and cutting speed (v).
A series of parameters can be calculated from those above mentioned, those
being distance e, the helix angle of the hob (), gear diameter (dg) and depth of
cut (t).
In order to simulate the process of gear hobbing, new software has been
developed. The proposed simulation model has been embedded in a commercial
CAD program thus taking advantage of its accuracy resulting in more detailed
calculations.
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.
Fig. 2 HOB3D flowchart.
24
Fig. 2 presents the flowchart of this simulation model. As it is illustrated,

after all the input data, shown on the left side of the figure, has been defined, the
initial workpiece is developed in the CAD environment and the number of the
effective teeth is determined accordingly. The gear hobbing simulation process
is executed for all these teeth. The first step in the simulation process is the
positioning of the cutting edge on the 3D space.
This positioning is repeated along the path that is covered by the cutting
edge. The profiles that arise are combined in order to form a 3D surface. The
next step is the creation of an assembly that includes the workgear and the 3D
surface. Afterwards, Boolean operations are used in order to generate the 3D
chip and the 3D gap. The above described process is repeated for all active teeth
in every rotation of the cutting tool. Finally, after the end of the simulation
process the cutting forces on every one of the active teeth of the hob are
calculated. These forces are added up and the total cutting forces are calculated.
4. HOB3D Formulation
4.1. Simulation Formulation
In HOB3D all the movements involved in the process are transferred to the
hob cutting tooth motion. Furthermore, the simulation process is carried out on
one gap of the gear thus reducing the simulation time. In order to identify the
cutting teeth, those are numbered. The tooth which has the local axis Y1 parallel
to local axis X2 when it passes through the center of the gap is named Tooth 0,
the tooth that passes after that is named Tooth 1 and the tooth previous to that is
named Tooth-1 and so on.
The simulation process is illustrated on Fig. 3. As it is presented, the hob
profile according to DIN3972 [10] must be designed first. This profile is
presented on the top left frame of Fig. 3 and positioned on the 3D space, as
illustrated on the next frame of fig. 3. The positioning is repeated for a series of
times until the tooth is out of the cut.
After this step, there is a series of profiles correctly positioned on the 3D
space. Every profile represents a revolving position of the hob and its
positioning includes hob rotation, workpiece rotation and feed rate. All these
profiles are combined in order to form a 3D surface like the one seen in the last
frame of the first line of fig. 3. Then, the 3D surface is assembled with the 3D
gap produced from the previous tooth. The final stage of the simulation process
is the extraction of the final gap geometry and the non-deformed chip geometry.
This is achieved with the aid of Boolean operations, as supported by the used
CAD environment.
25
Fig. 3 HOB3D simulation process.
4.2. Force Component Calculation Formulation

After the simulation is completed, the cutting forces can be calculated. The
force calculation code is performed in accordance with Kienzle-Victors
equations. The calculation process is based on sections made on every one of
the solid chips.
A series of sections are made for every chip. Each section is made in the
plane of the hobs cutting edge. The planes are visible on the top left frame of
Fig. 4. After the creation of the section on the 3D chip, the crossection is
discritised along the cutting edge. This discritisation is presented on the top
right frame of fig. 4.
The three force components are calculated according to the elementary chip
width and thickness for every elementary chip section. The three force
components are rotated in order to match the local coordinate system (1) and
then added up in order to produce the total cutting forces on every crossection.
After all the sections are calculated the total forces on the tooth are obtained, as
presented on the bottom of Fig. 4.
26
Fig. 4 Force calculation algorithm.
4.3. Simulation Results

HOB3D has been used in order to produce the 3D solid chips in the full cut
phase of gear hobbing. These chips were analysed so that the cutting forces are
calculated. The crossections made on the chip are presented on the right side of
Fig. 5 whereas the chip thickness on six of the crossections is illustrated on the
left side of the same figure.
27
Fig. 5 Chip thickness of the solid chip.
The cutting forces components for the above chip are presented in the next
Fig. 6 which are calculated in accordance with the system 1.
Fig. 6 Cutting forces on one generating position.
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.
Fig. 7 Comparison between calculated and measured 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.
Fig. 8 Comparison between calculated and measured cutting forces.
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.
Received: March 10, 2010
Technical University of Crete,

Department of Production Engineering&Management,
Chania, Crete, Greece,
e-mail: ntapoglou@isc.tuc.gr
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.
10. D I N 3972, Bezugsprofile von Verzahnwerkzeugen fuer Evolventenverzahnungen

Nach DIN 867, 1981.
DETERMINAREA PRIN SIMULARE CAD A COMPONENTELOR
FOREI DE ACHIERE LA FREZAREA DANTURILOR
(Rezumat)
Unul dintre cele mai uzuale procedee de prelucrare a danturilor, n special a celor exterioare,
este frezarea. Optimizarea acestui tip de proces de danturare prezint la ora actual o importan
deosebit n industrie. Lucrarea prezint un nou program de simulare, numit HOB3D, care
permite simularea procesului de achiere ntr-un mediu CAD comercial, cu rezultate optime n
ceea ce privete precizia de prelucrare. Elementele rezultante ale programului de simulare se
refer att la achia 3D i la geometria golului dintre dinii frezei, ct i la nivelul forelor de
achiere dezvoltate, elemente care au fost validate experimental.

Publicat de
Secia
CONSTRUCII DE MAINI
CONTRIBUTIONS CONCERNING THE COMPUTER

AIDED CONTROL OF THE FELLOWS' CUTTER
BY
EUGEN STRJESCU1, CONSTANTIN DOGARIU1,

OLIMPIA PAVLOV2 and DUMITRU DUMITRU3
Abstract. In the paper are presented the bases of a methodology for the
determination of all the geometric and constructive elements of the cutting tools
for the fellows cutter, tools having curve edges, starting from the getting of a solid
model 3D of the measured tool. There are shown the possibilities to obtain a 3D
model and the mathematical model of the geometric parameters control.
Key words: fellows' cutter, control, cutting tools, 3D model, geometry,
curved edges.
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
Eugen Strjescu et al.
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'
Fig. 3 The determination of the angles v and v at the peak.
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
Fig. 9 The charts of the angles O and O
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
Fig. 10 Intersections with frontal planes in different points.
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
Fig. 11 Values for the angles f i f in frontal plans Pf - Pf.
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.
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.

Publicat de
Secia
CONSTRUCII DE MAINI
CONTRIBUTIONS TO THE ELABORATIONS OF A

GRAPHICAL METHOD FOR PROFILING OF TOOLS
WHICH GENERATE BY ENVELOPING
I. ALGORITHMS
BY
SILVIU BERBINSCHI, VIRGIL TEODOR,

NICOLAE DUMITRACU and NICOLAE OANCEA
Abstract. The generation of the ordered curls profile (surfaces) associated
with the centrodes of the rack-gear tool represent an fundamental problem for the
profiles reciprocally enveloping associated with a couple of rolling centrodes. The
knowing, in principle, of the generating rack-gear allow the construction of the
tool and, also, the determination of the axial profile of the worm mill for the
ordered curl profiles generation. They are known and used more solutions for the
solving of this problem: analytical solutions, based on the fundamentals theorems
of gearing; complementary theorems, obtained from the fundamentals theorem;
graphical methods based on graphical environments capabilities. In this paper, is
proposed an algorithm for the approach of issue of ordered curl profiles based on
the CATIA design environment capabilities. It is proposed a general algorithm
which allows making applications, ended with the numerical determination of the
rack-gear reciprocally enveloping with ordered curl profiles. They are solved
profiles known in discrete form, substituted by spline approximation. The method
allows drawing the gearing lines as so as the interference trajectories. The
obtained results are comparing with results obtained by analytical methods.
Key words: enveloping surfaces, rack-gear tool, graphical design method.
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
- analytical methods, based on fundamental methods of surfaces enwrapping

first Olivier theorem, Gohman theorem, normals method, W i l l i s [1], [2];
- complementary analytical methods minimum distance method, the
substitutive circles family method, the in-plane generating trajectories
method [3]-[5];
- graphical-analytical methods [6];
- graphical methods, using the capabilities of CAD software [7].
We mention that the methods proposed and used for the study of
reciprocally enveloping surfaces respect the enveloping fundamental theorem.
The proposed solutions leads to comparable results, in most cases
identically, for the crossing profile tools shape, which generated by rolling
ordered curls profiles associated with a couple of rolling centrodes.
Literature include multiples applications for various domains as methods of
tools profiling, revolution surfaces, for generating helical surfaces [8]; the
modelling of the cylindrical surfaces with disk tools [9]; generation of the
helical compressors rotors [10].
In this paper, is proposed a method for profiling of the rack-gear tools
reciprocally enveloping with ordered curl profiles, based on the capabilities of
the CATIA design environment.
2. Generation Kinematics
In principle, the tools profiling determination problem for rack-gear tool
reciprocally enveloping with an ordered curls profile, associated with a circular
centrode, assume to respect the kinematics of the generating process, see figure
1.
Fig. 1 Couple of rolling centrodes.
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 ,
where is the linear velocity in the translation of C2 centrode;

Rrp 1 the value of velocity in point O1the gearing pole, from C1
centrode, in the rotation movement around z axis;
1 variable angular parameter.
In the rotation of C1 centrode the translation movements along the C2
centrode and the rotation around Z axis are evenness.
They are defined the reference systems:
xyz is the global reference system, with z axis overlapped to the rotation axis
of the C1 centrode;
XYZ mobile reference system, initially overlapped to the global reference
system, joined with the ordered curls profile ;
mobile reference system, joined with C2 centrode of rack-gear tool,
with axis parallel and of the same sense with the global reference system xyz.
The kinematics of the rolling process of the two centrodes, C1 and C2,
tangents in point O1 gearing pole presume that the velocities of points
belongs to the two centrodes, temporally situated in point O1, to be equals.
In this way, the global motion of the reference system, joined with
centrode C2, is described by the transformation,
(2)
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
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)
is the rotation transformation matrix, around X axis, with angle 1

(counterclockwise rotation).
The assembly of equations (3) and (6), with the respect of rolling condition
(1), determine the relative motion,
(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)
with u variable parameter, describe a profiles family in the rack-gear space:
( u ,1 ) = 0;
(9)
( ) ( u,1 ) = Y ( u ) cos (1 ) Z ( u ) sin (1 ) + Rrp 1 ;

( u ,1 ) = Y ( u ) sin (1 ) + Z ( u ) cos (1 ) Rrp .
1
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 ) ,
with u and t independent variable parameters.

To the profiles family (9) was associated the enveloping condition
specifically for one of the enwrapping fundamental theorems (first Olivier
theorem, Gohman theorem, normals method [1], [2]) or one of the
complementary theorem (minimum distance, substitutive circles family,
in-plane generating trajectories) in order to determine the envelope of profile
family as generating rack-gear profile [3], [4].
Also, are known graphical solutions for profiling tools which generate by
enwrapping method, ordered curls profile associated with a rolling centrodes
couple (the regions method [6], [7]).
3. Kinematics Method in the CATIA Design Environment
Is proposed a new solution for the profiling of rack-gear tool reciprocally

enveloping with an ordered curl profiles associated with a couple of rolling
centrodes, using the capabilities of CATIA software, by making an kinematics
entity which will simulate the rolling movement of centrodes: C1 rolling
circle with radius Rrp, associated with the profiles curl; C2 straight line,
associated with the rack-gear.
The proposed solution has the advantages that use the capabilities of a very
versatile product, which may offer rigorous numerical results.
Also, being a graphical method, the rough errors, due firstly to the passing
curves, which may be erroneously considered as profiles zone, are easy to
identify and deleted from analysis.
The proposed solve is based on the capabilities of Part environment where
are synthesized the elements of a mechanism able to simulate the enveloping
condition, in this case, the normal condition. These elements, created in the Part
environment, are inserted in a file of Assembly environment, assuring the
position of the mechanism elements in the start position, in following in the
DMU Kinematics environment are defined the kinematics couples.
The mechanism motion is made by the command Simulation, establishing
the intermediate position number (Shots) and creating with the command Replay
46
a movie of the mechanism successive positions.

With command Trace is draw the trajectory of any point which belongs to
one element of the mechanism regarding any another element, including the
global reference system, determining the gearing line between the profile to be
generate and the profile of the rack-gear tool.
These trajectories represent spline curves, constructed by successive points
obtained by the mechanism rolling. The coordinates of these points may be
exported as text file or any spreadsheet, see figure 2.
Fig. 2 Generating algorithms in CATIA environment.
4. Conclusions
1. The rolling process kinematics for two centrodes becomes possible to be

described in the graphical design environment as CATIA.
2. The precision for these profile description, in this graphical environment,
the geometrical forms and the kinematics of these allow a rigorous analysis of
the reciprocally enveloping surfaces.
3. A software suit specifically for the graphical design environment was
synthesised in the end of paper, for the tools profiling reciprocally of an
ordered curl of surfaces (profiles). Specifically applications will allow
establishing the proposed method quality.
47
Acknowledgements. The authors gratefully acknowledge the financial support of the

Romanian Ministry of Education, Research and Innovation through grant
PN_II_ID_791/2008.
Received:January, 20, 2010
Dunrea de Jos University of Galai,

Manufacturing Science and Engineering Department
Galai, Romania,
e-mail: nicolae.oancea@ugal.ro
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
CONTRIBUII LA ELABORAREA UNEI METODE GRAFICE PENTRU

PROFILAREA SCULELOR CARE GENEREAZ PRIN NFURARE
I. ALGORITM
(Rezumat)
Generarea vrtejurilor ordonate de profiluri (suprafee), asociate unor centroide, cu
scule de tip cremalier, reprezint o problem fundamental pentru profiluri reciproc
nfurtoare asociate unui cuplu de centroide n rulare. Cunoaterea principial a
formei cremalierei generatoare permite construcia sculei-pieptene i, de asemenea,
determinarea profilului sculei melc pentru generarea vrtejurilor ordonate de profiluri.
Sunt cunoscute i utilizate multiple soluii pentru abordarea unei astfel de probleme:
soluiile analitice, bazate pe teoremele fundamentale ale angrenrii; teoremele
complementare, derivate din metodele fundamentale, metode grafice, bazate pe
posibilitile mediilor de proiectare grafic CAD.
n lucrare se propune un algoritm general pentru abordarea problematicii
profilurilor (suprafee) bazat pe facilitile mediului de proiectare CATIA. Se propune
un algoritm general, n baza cruia se realizeaz aplicaia, finalizat cu determinarea
numeric (grafic) a formei cremalierei reciproc nfurtoare vrtejului ordonat de
profiluri.
Sunt tratate profiluri cunoscute analitic sau sub form de coordonate discrete,
substituite prin forme de aproximare spline.
Metodica permite tratarea liniilor de angrenare precum i traiectoriile de
interferen a profilurilor. Se compar rezultatele obinute prin aceast metod cu
rezultatele obinute prin metode analitice.

Publicat de
Secia
CONSTRUCII DE MAINI
CONTRIBUTIONS TO THE ELABORATIONS OF A

GRAPHICAL METHOD FOR PROFILING OF TOOLS WHICH
GENERATE BY ENVELOPING
II. APLICATION FOR RACK-GEAR TOOLS PROFILING
BY
SILVIU BERBINSCHI, VIRGIL TEODOR,

NICOLAE DUMITRACU and NICOLAE OANCEA
Abstract. In this paper is presented an application of the general algorithm
regarding the tools profiling which generate by enveloping, by rolling method,
respectively the rack-gear tool generating an ordered curl of profiles. They are
presented the successive steps of this process and, also, numerical examples
regarding the crossing profile of the rack-gear tool. The method quality is proofed
by comparing the obtained results with results obtained for the same problem,
using one of the analytical methods, the Willis method, and software developed in
Java.
Key words: enveloping surfaces, rack-gear tool, graphical design method.
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
solution for the profiling of rack-gear tool reciprocally enveloping with an

ordered curl profiles associated with a couple of rolling centrodes, using the
capabilities of CATIA software, by making an kinematics entity which will
simulate the rolling movement of the two centrodes [8].
2. Application for Rack-Gear Profiling
In order to obtain the tools profile for various polygonal shafts, for various
side lengths, is need to follow two stages. The first stage is to choose the side
length and the diameter of the rolling circle as input data. This thing may be
done introducing in an text or Excel file the value, in following, the CATIA
software will automate modify the whole mechanism, with the new values. In
the same stage is created the mechanism, the rolling of this and the tools profile
determination. The coordinates of points belongs to profile will be exported in a
text or Excel file.
The second stage consist in the partial remake of the mechanism couples,
the rolling of this for the new tools profile, for the previous modified values
and the export of points in a new file.
2.1. First stage
2.1.1. Data input and mechanism elements construction. In this step are
created a folder with all files needed for the mechanism.
First of all for the input data is created a text of excel file with the names
and the values of the input parameters, namely the piece shaft radius Ra, the
polygonal side length L and the rolling circle radius of blank Rrp. These are the
parameters which may be modified by user. This file will be linked with the
parameters from the CATIA file.
In the following example is considered a hexagonal shaft with L=50 mm
and the including radius Ra=50 mm. In order to simplify the representation we
consider the rolling radius Rrp=50 mm. The shaft sketch and the associated
centrode are represented in figure 1. The mechanism elements are successive
created, in the Part environment, are suggestive named and inserted in an
assembly file, created in the Assembly environment, named GenCremaliera.
The Part type files are: baza; piesa; cremaliera and tachet.
The Baza file contains the standing part of the mechanism and represent
more lines needed to mount others elements, see figure 2.
From a point named OriginePiesa is construct a line named AxaArbore,
representing the pieces axis and next another line which will be as guide for the
rolling line named GhidajDreaptaRulare. The segment having the rolling radius
length and which contain the gearing pole is created in order to ease the
51
mechanism assembling. The point PolulAngrenarii is situated on the rolling

circle with radius Rrp.
Fig. 1 The shaft sketch.
This file contain a parameters table (Design Table) named DateIntrare

which link the line and points parameters from the Part file (from CATIA)
with the input data from the file DateIntrare.txt.
The file Piesa contain, firstly, the geometrical elements of the profile for
which is determined the tools profile, see figure 2. Another sketch from this
file is CercRulare which contain a circle with radius Rrp. The two values Ra and
Rrp are linked with the text file DateIntrare.txt. The file Cremaliera contain the
rolling straight line (DreaptaRulare) representing a circular pitch of the curl to
be generated and having the length as function of the rolling radius Rrp,
controlled by file DateIntrare.txt. The file Tachet have the straight lines which
are tangent and respectively normal at the pieces profile in the contact points.
Fig. 2 Assembly file with kinematics couples and tools profile.
52
2.1.2. Mechanism assembling and the creation of kinematics couples. In the

assembly file GenCremaliera are inserted all the above described elements,
with the command Insert Existing Component.
These have geometrical and dimensional constraints in order to be
positioned in accordance with the mechanism functionality. After inserting the
constraints, the GenCremaliera contain the whole mechanism in the initial
position. In the Assembly environment the mechanism elements were positioned
but doesnt exist yet any couple between them.
The kinematics couples are created in the DMU Kinematics environment. It
is created the mechanism named Generare.Scula.Cremaliera and is established
the standing element, Baza.
After the kinematics couples creation the mechanism function may be
simulated and look as in figure 2.
2.1.3. Mechanism rolling and determination of tools profile. In order to
rolling this mechanism is created a simulation with command Simulation and
are established the intermediate position number, Shot, number which will have
influence to the tools profile precision. For this example we considered 500
steps, so it will result 500 points. With this simulation is determined the tools
profile, see Figure 3.
0
1
2
3
4
5
6
7
Fig. 3 Tools profile in the rack-gear reference system.
2.1.4. Points coordinates extraction. After the determination of the tools

profile, these may be exported in a text file using the command Design Table,
from the Part environment.
This command will generate a text file PuncteProfil.txt, see Table 1.
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
2.2. Second stage

In the second stage if the user modify the DateIntrare.txt file saving the
modification of the input parameters, the CATIA software detect this and
warning by an Update message.
After this message the user have to update the assembly file and the
kinematics file in order to remake the kinematics couples which may be
modified.
3. Method Quality Verification
It is proposed in following the verification of this method by comparing the
numerical results obtained with graphical method regarding the results obtained
by one of the analytical methods: Gohman method; normals method;
minimum distance method etc [3].
For the polygonal shaft is defined the analytical form of the profile to be
generated, see figure 4:
(1)
X = 0;
Y = u;
Z = a,
54
with u variable parameter, measured along the profile and a constant value
depending the profile form.
Fig. 4 Straight lined profile.
The profile family is on form:
(2)
= 0;
( ) = u cos a sin + Rrp ;
= u sin + a cos Rrp ,
with variable parameter.

The specifically enveloping condition, in the in-plane generation trajectory
methods is:
(3)
cos u sin a cos + Rrp

,
=
sin
u cos a sin
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
Fig. 5 Errors obtained.
4. Conclusion
1. The proposed algorithm use the CATIA design environment capabilities.

2. The Tachet mechanism used in the problem solving is universal,
excepting the profiles with singular points.
3. The profiling precision of rack-gear tool is comparable with results
obtained by analytical methods, based on which was created dedicated software,
in present paper, in Java language.
Acknowledgements. The authors gratefully acknowledge the financial support of the
Romanian Ministry of Education, Research and Innovation through grant
PN_II_ID_656/2007.
Received:January, 20, 2010
Dunrea de Jos University of Galai,

Manufacturing Science and Engineering Department
Galai, Romania,
e-mail: nicolae.oancea@ugal.ro
56
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
CONTRIBUII LA ELABORAREA UNEI METODE GRAFICE PENTRU

PROFILAREA SCULELOR CARE GENEREAZ PRIN NFURARE
I. APLICATIE PENTRU PROFILAREA SCULEI-CREMALIERA
(Rezumat)
n lucrare este prezentat o aplicaie a algoritmului general prezentat, privind
profilarea sculelor care genereaz prin nfurare, prin metoda rulrii, n spe
scula-cremalier generatoare a unui vrtej ordonat de profiluri. Sunt prezentai paii
succesivi ai procesului de profilare a sculei cremalier, de asemenea, un exemplu
numeric privind forma profilului transversal al sculei cremalier.
Calitatea metodei propus este dovedit prin comparaia rezultatelor obinute, n
cadrul aceleiai probleme concrete de profilare, utiliznd ns una dintre metodele
analitice metoda Willis, i un produs soft dezvoltat n Java.

Publicat de
Secia
CONSTRUCII DE MAINI
THE NUMERICAL SIMULATION OF TURNING PROCESS

USING FINITE ELEMENT MODELING
BY
CATALIN FETECAU, DANIEL-VIOREL VLAD

and COSTEL MOCANU
Abstract. In this paper we present a study on numerical simulation of turning
process using finite element analysis program AdvantEdgeTM for AISI 1060
carbon steel material. The main objective is to determine the influence of different
cutting parameters (depth of cut, feed rate and cutting speed) on cutting force, von
Mises stress and temperature in the longitudinal turning process. The predict of
cutting forces by numerical simulation is important for the optimization of the
process. The simulations were conducted without coolant.
Key words: turning, numerical simulation, cutting forces, temperature.
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.
R = ( Fx2 + Fy2 + Fz2 ) (1.11.18)Fx,
58
Ctlin Fetecu et al.
n
piece
Fx
t
Fy
Fz
chip
tool
feed rate
Fig. 1 3D cutting force components according AdvantEdgeTM.
2. Numerical Simulation of Cutting Process

2.1. The Mathematical Model
Computer simulation of cutting process was achieved with AdvantEdgeTM

program. Elasto-plastic behavior of material during the cutting process is
characterized by a nonlinear relationship between stresses and strains. In the
mathematical modeling of surface flow movement, in software, are
implemented a series of mathematical models, where the most important being
model Power Law and model Drucker Prager [5].
Model Power Law defined by the relation
) ( )
p , & , T = g p (& ) (T ) ,
(2)
( )
where: g p is strain hardening, (& ) - strain rate sensitivity, (T ) - thermal

softening.
Model Drucker-Prager defined by the relation
) (
p , J1 , &, T = G p , J1 (& ) (T ) ,
(3)
where: G p , J1 is strain hardening plus hydrostatic pressure, (&) - strain rate
sensitivity, (T ) - thermal softening, p - the plastic strain.
( )
The function g p for the Drucker-Prager model is defined as
(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
, if & > &cut .
The thermal function (T ) for the Drucker-Prager model is defined as

(8)
(9)
(T ) = c0 + c1T + c2T 2 + c3T 3 + c4T 4 + c5T 5 . if T < Tcut ,
(T ) = (Tcut )
T Tcut
if T Tcut ,
Tmelt Tcut
where: c0 through c5 are coefficients for the polynomial fit, T - the

temperature, Tcut - the linear cut-off temperature, Tmelt - the melting
temperature.
2.2. Finite Element Machining Model
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
Ctlin Fetecu et al.
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
Side rake angle, ,
-5
0.208
Lead angle, ,
45
0.416
Nose radius, r, mm
0.4
0.5
78.54
157.08
314.16
The mechanical properties and chemical composition of work-piece

material, AISI 1060, are show in Table 2.
Table 2
The chemical composition and mechanical properties of AISI 1060
Chemical composition, %
Yield
Ultimate Tensile Hardness
Material
strenght,
Strenght
Cu
Mn
P
S
MPa
MPa
Bhn
AISI
0.605 0.75 0.04 0.05
370
625
179
1060
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].
Fig. 4 The finite element network for the whole piece-tool-chip.
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.
Fig. 6 Von Mises Stress f=0.208 mm/rev; v=157.08 m/min.
3. The results of numerical simulations
In according with the plan of experiments, it was made numerical

simulation for each cutting regime. For different values of the feed rate and the
cutting speed the average values of tangential force Fx in according with
AdvantEdgeTM are presented in Table 3.
Table 3
The average values of tangential force, Fx, N
Feed, f,
Cutting speed, v, m/min
mm/rev
314,16
157,08
78,54
0.083
306.177
308.298
315.186
0.208
544.703
557.041
544.703
0.416
880.12
950.191
1073.24
0.5
1015.5
1135.32
1205.5
62
Ctlin Fetecu et al.
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
Fig. 7 The effect of feed rate on cutting force.
For the simulation made, the average values of temperature of cutting

tool are presented in Table 4.
Feed, f,
mm/rev
0.083
0.208
0.416
0.5
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
Fig. 8 The temperature distribution in the cutting zone.
In the FEM software AdvantEdgeTM, the Johnson-Cook constitutive model

were used to predict the thermal behaviour in longitudinal turning process [1].
This software allows the temperature distribution, cutting forces and von
Misses stress to be predicted to appropriate measurements [2].
63
With AdvantEdgeTM is possible to determine optimum machining

parameters with lower costs and in short time without experimental processes
[3].
4. Conclusion
1. This paper presents a study on numerical simulation of turning process

using finite element analysis for AISI 1060 material. Simulation was done using
AdvantEdgeTM program. Based on numerical simulations we determine the
influence of cutting parameters (depth of cut, feed rate and cutting speed ) on
cutting force, von Mises stress and temperature from cutting zone.
2. Based on numerical simulations to determine the size variation of cutting
force and temperature parameters based on cutting regime.
3. It was also determined variation von Mises stress across-chip track-and to
investigate the formation of the chips.
4. Future work will be as based on results of the numerical simulation will
to make the experiments and will compare the results with those of simulation.
Acknowledgements. The work of Drd. Daniel-Viorel Vlad was supported by Project
SOP HRD EFICIENT 61445".
Dunarea de Jos University of Galati,

Department of Machine Manyfacturing
Galati, Romania,
e-mail: catalin.fetecau@ugal.ro
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
Ctlin Fetecu et al.
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.
SIMULAREA PROCESULUI DE STRUNJIRE FOLOSIND

ANALIZA CU ELEMENT FINIT
(Rezumat)
In aceast lucrare se prezint un studiu privind simularea numeric a procesului de
strunjire folosind programul de analiz cu elemente finite AdvantEdgeTM n cazul,
oelului OL60 (AISI 1060). Obiectivul principal este de a determina nfluena
parametrilor regimului de achiere (adncime de achiere, avans i vizeza de achiere)
asupra componentei tangeniale a forei de achiere, Fx, a tensiunii von Mises i a
temperaturii din zona de achiere.
Comportarea elasto-plastic a materialului n timpul procesului de achiere este
caracterizat de o relaie neliniar dintre tensiuni i deformaii.
n vederea modelrii matematice a deplasrii suprafeei de curgere, n programul
AdvantEdgeTM sunt implementate o serie de modele matematice, cele mai importante
fiind: Model Power Low i Druker Prager Model. Pentru simulrile efectuate , am
folosit modelul Druker Prager.
Forma achiei reprezint un indicator al condiiilor de achiere, artnd gradul de
deformare plastic suferit de stratul de material detaat. Forma achiei depinde de natura
materialului prelucrat, geometria sculei, regimul de achiere etc. Se deosebesc achii de
rupere i achii de deforrnare plastic.
Rezultatele obinute vor sta la baza realizrii experimentelor fizice, astfel nct s
putem confrunta rezultatele obinute din simulare cu cele din experiment.

Publicat de
Secia
CONSTRUCII DE MAINI
EXPERIMENTAL RESEARCHES REGARDING A NEW

METHOD FOR CUTTING FORCES EVALUATION
BY
MIRCEA COZMNC, CRISTIAN CROITORU

and CTLIN UNGUREANU
Abstract. The paper presents some preliminary experimental researches
regarding the cutting forces level when cutting four types of steel. The aim of
these tests is to finalize the calculus models for cutting force components Fz, Fx,
Fy on a cutting tooth. The new models are considering the area of the chips
theoretical cross section (txs), the compression yield point of the cut material and
the chips contraction coefficient, as presented in our previous papers. The chips
contraction coefficient includes all the working parameters influences and also
the interdependencies between them. Beside those elements, the calculus models
contain a theoretical component, depending on the constructive angles of the tool,
, K and and the friction coefficient between chips and the cutting tools rake
face.
Key words: model, chips contraction, cutting force, experimental.
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
Mircea Cozmnc et al.
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 .
C z = cos cos( + )[1 + tg ],

(4)
C x = cos sin ( + ) cos K [1 + tg ] + cos sin (K )[ tg ],
C y = cos cos(K )[ tg ] cos K sin ( + )sin K [1 + tg ].
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
Main constructive geometry

Values calculated with Eqs. (2) and (3)
K
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
Fig.1a The influence of angle for OL44 steel.
70
OL44
=0
s=0.1mm/rev
Fig.1b The influence of angle for OL44 steel.

40C10
=0
s=0.1mm/rev
Fig.2a The influence of angle for 40C10 steel.
40C10
=0
s=0.1mm/rev
Fig.2b The influence of angle for 40C10 steel.
71
OL44
=0
=0
Fig.3a The influence of feed values for OL44 steel.

OL44
=0;=0
s=0.1mm/rev
Fig.3b The influence of speed values for OL44 steel.

40C10
=0
=0
Fig.4a The influence of feed values for 40C10 steel.
72
40C10
=0;=0
s=0.1mm/rev
Fig.4b The influence of speed values for 40C10 steel
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
1. The experimental researches regarding the values of cutting force Fz and

coefficient Cd for the four steel types were stressing the known dependencies
from angles and , cutting feed and main cutting speed.
2. The level of influence of the constructive angles and , of cutting feed
and speed takes different values depending on the material being cut.
3. Experimental tests are necessary for each material, to determine the
values for mz, mx, my exponents in order to use the models in Eqs. (1)-(4) for
the force components Fz, Fx, Fy on a cutting tooth.

Received:
73
Gheorghe Asachi Technical University,

Department of Machine Tools
Iasi, Romania
e-mail: cozminca@tcm.tuiasi.ro
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.
CERCETRI EXPERIMENTALE PENTRU VALIDAREA UNEI

NOI METODE DE EVALUARE A FORELOR DE ACHIERE
(Rezumat)
n lucrare se prezint modul de realizare a unor cercetri experimentale preliminare
n vederea stabilirii mrimii forelor de achiere, la strunjirea a patru mrci de oel, n
scopul definitivrii modelelelor de calcul pentru componentele Fz, Fx, Fy ale forei de
achiere la nivelul uni dinte achietor. Noile modele de calcul au n vedere aria seciunii
transversale a achiei nedeformate (txs), limita de curgere a materialului achiat i
coeficientul de deformare plastic a achiilor. Coeficientul de deformare plastic a
achiilor include toate influenele parametrilor de lucru, precum i interdependenele
dintre aceste influene. Pe lng aceste elemente, modelele de calcul propuse cuprind i
o component teoretic, n funcie de unghiurile constructive ale dintelui, , K, i
coeficientul de frecare dintre achii i suprafaa de degajare a dintelui achietor.

Publicat de
Secia
CONSTRUCII DE MAINI
FACE MILLING FORCES DEPENDING ON THE

FORCES DEVELOPED ON A SINGLE-TOOTH
BY
ANA-MARIA MATEI and MARIUS NICOLAE MILEA

Abstract. This paper presents some theoretical models for the evaluation of
cutting forces components in face milling, developed on a singletooth level. The
cutting forces at face milling depend on the every variant of face milling process
(teeth number which simultaneously cut) and the relative position of cutting teeth
and the material being cut (cut - down milling and cut - up milling). Regarding to
this, the determination of cutting force components at face milling FZ, FX, FY is
based on the forces for a single tooth, the cutting tooth position beside the XYZ
coordinates system of the tool and the number of teeth that simultaneously cut.
Key words: cutting, face milling, forces.
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
Ana-Maria Matei and Marius-Nicolae Milea
2. Theoretical Models of Cutting Force Components in Face Milling

depending on the Cutting Forces developed on a Single-Tooth
The face milling forces components acting on the tool FZ , FX si FY are
depending on the cutting forces components developed on a single tooth level
Fz , Fx si Fy , which are influenced by the cutting tooth geometry (, k, ), the
standard compression yield point of the material being cut, the cross - sectional
area of the chip, the chips friction coefficient , and the chips contraction
coefficient Cd [3,4].
Considering the specific elements of the symmetrical or unsymmetrical
milling, there are several possibilities of modelling the cutting force
components in face milling.
2.1. Theoretical Models of Cutting Force Components
in Symmetrical Face Milling
Fig. 1 shows the geometrical models of complete and incomplete face
milling. The values of specific elements for each milling variant result as
follows: the contact angle = 180, t = D, the number of teeth that
simultaneously cut zs = z/2 at complete face milling and < 180, t < D,
zs =
arcsin
t
at incomplete face milling [2].
D
The theoretical models for the evaluation of cutting force components in

face milling are developed in the light of two simplifying assumptions, as
follows:
a) The cutting force components on a tooth, respectively Fz the main cutting
force (the tangential force), Fx the radial force and Fy the axial force, take
values depending on the cross-sectional area of undeformed chip;
b) The tooth z1 is the one to which the tooth coordinates system xyz
corresponds to the mill coordinates system (FZ = Fz, FX = Fx, FY = Fy), and the
other teeth that simultaneously cut are placed equidistant ( = 2/z).
The values of cutting force components in symmetrical face milling depend
on zs and Fz, Fx , Fy , as follows:
a) For symmetrical milling with odd zs the Eq. (1) results.
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
Fig.1 Forces model in symmetrical face milling.
b)
For symmetrical milling with even zs the Eq. (2) results.

zs / 2
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
Fig. 2 Forces model in unsymmetrical cut-up face milling.
According to the geometrical model of face milling process shown in Fig. 2,

the cutting force components FZ, FX and FY result using the Eqs.(3).
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
Fig. 3 Forces model in unsymmetrical cut-down face milling.
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

z s4 1
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
Gheorghe Asachi Technical University of Iai

e-mail: anca_reea@yahoo.com
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).
FORELE DE ACHIERE LA FREZAREA FRONTAL N FUNCIE DE

FORELE DEZVOLTATE LA NIVELUL UNUI DINTE
(Rezumat)
n lucrare sunt prezentate modele teoretice de evaluare a componentelor forei
de achiere la frezarea frontal n funcie de forele dezvoltate la nivelul unui dinte.
Forele la frezarea frontal depind att de varianta de frezare, poziia relativ a dinilor
achietori fa de semifabricat. poziia dintelui achietor n raport cu sistemul de
coordonate XYZ al sculei ct i numrul de dini care achiaz simultan.

Publicat de
Secia
CONSTRUCII DE MAINI
COMPARATIVE ASPECTS REGARDING THE NESTING

FOR BLANKING-PUNCHING OPERATION ON
CLASSICAL PRESSES AND NUMERICAL
COMANDED PRESSING CENTERS
BY
MARIUS - IONU RPANU1 GHEORGHE NAG1,

IOLANDA - ELENA MANOLE1 and ANDREI WEINGOLD2
Abstract. The nesting operation represents an important chapter in the design
of stamping technology of parts or blanks in the area of cold plastic forming
processes, with a large applicability in industry. The type of nesting has a very
important role in determining the material consumption, the number of blanks cut
concomitant, in determining the type and the construction of the stamping device
or of the punch. The goal of this paper is to elaborate a comparative study
between nesting for the classical presses stamping and for the CNC pressing
centres. In order to obtain this goal, in this paper a case study, along with the
results obtained and the associated conclusions are presented.
Key words: punching, blanking, stamping press, CNC, nesting.
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
Marius Ionu Rpanu et al.
The objective of this paper is to elaborate a comparative study upon the

nesting for the classical presses and nesting for the numerical commanded
centres in order to emphasize the advantages and the disadvantages of both
operations and, not the last, based on the effectuated researches, to obtain high
quality and high precision parts at the smallest cost possible.
2. Particularities of the Nesting Operation
The cold plastic forming researches and especially the nesting for numerical
controlled pressing centres researches made by J a c k s o n and M i t t a l [2] are
based on blanking-punching using a very well planned algorithm, method that
leads to the automatical generation of the CNC program. R a g g e n b a s s and R
e i s s n e r [2] have researched the connection between pressing and laser, with
applicability on the processing centres with CNC. In Twente University was
studied and developed the processing of the metal sheets domain through
planning and managing the factors that influence the machinability of metal
sheets using NC [2], [3]. Some of the most important CAD/CAM programs that
are used to perform this technological process are: RADAN, WICAM, Nesting
Software, EditCNC, CNCezPROTM, and some of the pressing centre with NC
are TRUMPF, AMADA ARIES, Finn-Power, Mazak, Bystronic, etc.
The nesting of the blanks for classical presses takes into account certain
coefficients that depend on the nature of the material, the placing of the part
and the nesting style of it (with waste, without waste, with little waste).
During the nesting of the parts for the NC pressing centres it must be taken
into consideration the type of the material of the metal sheet. The software used
for modelling, simulating and generation of the numerical control scheme will
consider all the factors and parameters that may influence the process
productivity.
3. Comparative Study
For this study was chosen a pinchbeck part made of CuZn15, SR EN
1653:2003, having the configuration from Fig. 1 and the thickness of 1 mm.
The processing of this part was followed both for the classical presses and the
CNC pressing centres. For the classical nesting, three variants of placing were
considered, as in Figure 2. For those variants were calculated the coefficients
with values of 53.77%, 54.16 % and 30.75%.
In case of using the NC cutting centres, CAD/CAM software were used in
order to realise and integrate the designing, modelling, simulating, planning and
development functions of the CNC program for processing the metal sheet,
choosing the optimum cutting variant with minimum waste.
85
Fig. 1 The processed part.
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
Fig. 3 The choice of the pressing centre for the processing.
Fig.4 Standard sizes of metal sheets with their coefficients

of utilization and the number of cut parts per sheet.
Fig.5 Position of the parts on 850 x 300 mm sheet metal.
Fig.6 The cutting path of the parts.
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
Amada Aries 245(RA1)

Verification Analysis
Total number of program blocks
Total number of program characters
Number of sub programs
Number of tool changes
Number of programmed stops
Number of repositions
Run time
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.
Received: March, 22, 2010

Department of Machine Manufacturing Technology,
e-mail: ripanumariusionut@yahoo.com
2
S.C. BMTech S.R.L of Vladeni,
Iasi Romania,
e-mail: andrei.weingold@bmtech.ro
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.
ASPECTE COMPARATIVE PRIVIND CROIREA

LA OPERAIA DE DECUPARE-PERFORARE PE PRESE CLASICE
I CENTRE DE PRESARE PREVZUTE CU COMAND NUMERIC
(Rezumat)
Operaia de croire reprezint un capitol important n proiectarea tehnologiei de
tantare a pieselor sau semifabricatelor din domeniul deformrii plastice la rece avnd o
aplicabilitate foarte mare n industrie. Croirea joac un rol important n determinarea
consumului de material, numrului de semifabricate tantate concomitent, n
determinarea tipului i construciei tanei sau matriei. Scopul acestei lucrri este de a
elabora un studiu comparativ ntre croirea pieselor pe presele clasice i croirea acestora
pe centrele de comand prevzute cu comand numeric, de a scoate n evidena
avantajele i dezavantajele acestei operaii i nu n ultimul rnd, pe baza cercetrilor
fcute, s putem obine piese de o calitate superiaor i precizie ridicat la un cost cat
mai redus. In acest sens, in lucrare este prezentat un studiu de caz, impreuna cu
rezultatele obtinute si concluziile aferente.

Publicat de
Secia
CONSTRUCII DE MAINI
THEORETICAL AND EXPERIMENTAL CONSIDERATIONS

ON DETERMINING THE EFFECT OF DIVERGENCE
BY
IUSTINA ELENA ROTMAN, PETRU DUSA

and RADU ADRIAN BACIU LUPASCU
Abstract. This paper presents the experimental studies conducted on the basis
of ultrasound examination of the parts with different geometries and structures.
The main objective of this work is to eliminate disturbances in order to perform an
examination and a correct interpretation in case of the A-scan representation. The
experimental results and conclusions drawn follow the analysis of the factors
involved in the process, which negatively influences the interpretation of signals
from the investigation display device.
Key words: transducer, device, back wall, divergence, discontinuity.
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
Iustina Elena Rotman et al
be confronted to the standards or to the contractual technical documentations to

establish if they are admissible, inadmissible or remediable [2]. In many cases
some influences can be neglected without to exceeding the measure limits
prescribed in the technical documentation [2], [4]. This presume the exactly
knowledge of the influence factors and of the results interpretation mode.
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]
The ultrasonic equipment used in this research is presented in the table

below:
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
2.1.1. The analysis of the investigation. Often in industrial practice there

are problems of interpreting the image obtained on the screen investigation
device. This paper presents an experimental study over the determination of the
divergence marginal effect and the elimination of disturbances in order to obtain
a correct examination and interpretation. Its objective is to highlight the journey
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
where: B the diameter of the piece; D p the diameter of the transducer; l

the length of the piece; the divergence beam of fascicle. Also the back wall
echo can be confused with a fault echo. These confusions occur due to the piece
geometry, to the transducer diameter which is comparable to the diameter of the
piece and to the differences in length. To eliminate these interpretation errors of
and disturbances we will present several different cases.
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.
Fig. 3 Ultrasonic testing display.
Fig. 4 The divergent effect.
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).
Fig. 5 Ultrasonic testing display.
Fig. 6 The ultrasonic beam shape .
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)
Fig. 7 Ultrasonic testing display:

a) The parasite echo; b) The additional echo disappears
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.
Gheorghe AsachiTechnical University,

Department of Machine Manufacturing Technology
Iasi, Romania
e-mail: yustinikrotman@yahoo.com
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
5. * * * Ultrasound and Ultrasonic Testing. NDT Resource Center. Available from:

http://www.ndt-ed.org/EducationResources/HighSchool/Sound/ultrasound.htm,
Accessed: 07/02/2010.
CONSIDERATII TEORETICE SI EXPERIMENTALE PRIVIND

DETERMINAREA EFECTULUI DE DIVERGENTA
(Rezumat)
Aceasta lucrare prezinta un studiu experimental ce urmareste evidentierea
parcursului ultrasunetelor de la traductor catre piesa de examinat, modul de reflexie al
acestora si de afisare pe ecranul instrumentului de investigat cu scopul de a obtine
rezultate eficiente. Obiectivul principal al acestei lucrari este determinarea efectului
marginal de divergenta si eliminarea factorilor perturbatori ce pot influenta negativ
rezultatele testarii cu ultrasunete. Studiile de caz prezentate si concluziile finale
indeplinesc obiectivele propuse si pot contribui la dezvoltarea bazei de cunostinte.

Publicat de
Secia
CONSTRUCII DE MAINI
COMPUTERIZED MEASUREMENT SYSTEM

BY
CTLIN UNGUREANU1, RADU IBNESCU2

and IRINA COZMNC1
Abstract. All technological machining systems parts made real surfaces
certain deviations from the theoretical and nominal areas. The role of quality
control is to decide whether or not these deviations fall within the allowable
accuracy classes that imposed parts. The paper presents a computerized system for
determining and recording deviations in shape of the real parts
Key words: measurements, inductive transducers, data acquisition.
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
Ctlin Ungureanu et al.
Deviations in shape are defined as deviations of actual surface shape of the

surface adjacent to, or deviation from the actual profile shape as the adjacent
profile. Profile is adjacent to the same profile as the profile geometry, the outer
tangential to the actual profile, and placed so that the distance between it and
the actual profile to be minimum. Maximum permissible deviation form as is
tolerance.
Traditional means of measuring the deviations in shape are instrumentality
mechanical comparators, scale value of 0.01 mm or 0.005 mm. They have the
advantages of simplicity and low cost, but recording and graphic representation
of measurement data for a production series is difficult. These disadvantages are
removed if for control is used a measuring system equipped with a displacement
transducer, data acquisition card and computer [1], [2].
2. Experimentals
Experimental facility is shown in Figure 1.
Fig.1 Experimental facility.
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.
Fig. 2 The block diagram of the application [6].
The DAQ Assistant may be founded in the Express Functions palette of

the block diagram and it is needed to create and configure a task that reads a
voltage level in a specified range from a DAQ device. After the DAQ Assistant
is placed on the block diagram, a dialog box window is launched in order to
configure the data acquisition system. The dialog box displays a list of channels
of the installed DAQ device and from this list the physical channel to which the
instrument connects the signal must be selected. The others configurations are
related to terminal configuration (differential or single ended), acquisition mode
(continuous or not), signal input range and rate of acquisition. The Write to
Measurement File Express VI is used for storing the information about the data
100
VI generates in a specific file (*.lvm) in the default LabVIEW Data folder

installed by LabVIEW in the default file folder of the operating system.
Experimental data were represented graphic with a Matlab application.
Calibration curve was presented in [6]. Diagram thus obtained is shown in
Fig.3.
0.02
d [mm]
0.01
-0.01
-0.02
0
50
100
150
l [mm]
Fig. 3 Experimental diagram.
In these diagrams l represents the length of the measured piece in

mm and d the form deviation also in mm.
Trough numerous measurement data recorded it is possible that the
chart should be influenced by measurement noise. To remove these
noises, was made a filtration of data with a Matlab application using
smooth function [7]. The default syntax of smooth function is
yf = smooth (y,span),
where y is a column vector containing the acquired data and yf is the
result column vector after a moving average filter is applied. The smooth
function implements a lowpass filter with filter coefficients equal to the
reciprocal of the span. The span must be odd but the default span for the
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]
Fig. 4 The filtered values for a span equal with 3.
It is noted that the chart is much more precise and accurate

assessment can be made on the actual shape of the part to be measured.
The filtration effect is observed more clearly if the graph is made for
filtered and the unfiltered data only for first portion of the controlled
piece. In Fig. 5 is represented by the dotted line unfiltered data and
filtered data with continuous line for a span equal with 3 and in Fig.6 for
a span equal with 5.
102

0.02
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
1. The computerized system for measuring deviations of form

enables the rapid and effective control of the geometric quality of parts in
machine building.
2. It can measure the absolute value of deviation and may draw a
diagram of the actual shape of the part to control.
3. However, because of noise measurement is required filtering
acquired data. You can use low-pass filter easily facilitated by MatLab
software package. It remains to be studied in future what is the best level
of filtering.
Received: February 20, 2010
Technical University Gh. Asachi,

Iai, Romania,
e-mail: cungurea@yahoo.com
2
Department of Theoretical Mechanics
e-mail: ribanesc@yahoo.com
,
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
SISTEM DE MSURARE COMPUTERIZAT

(Rezumat)
n lucrare este prezentat un sistem de msurare a abaterilor de form geometric.
Piesa de msurat este amplasat pe dou reazeme reglabile, astfel nct dreapta
adiacent s fie paralel cu placa de control pe care se efecueaz msurarea. Pentru
msurare se utilizeaz un traductor inductiv, alimentat de la o surs de curent continuu.
Se utilizeaz o cartel de achiziie de date, un PC i mediul de programare LabVIEW.
Cu datele astfel obinute se pot trasa diagrame ale formei reale a piesei. Datorit
zgomotelor de msurare, forma real a piesei este mai greu de observat. Pentru a
nltura influena zgomotelor de msurare se recomand o filtrare de tip trece jos a
profilului obinut. Aceasta se poate face cu ajutorul funciei smooth a programului
Matlab, cu nivele diferite de filtrare.

Publicat de
Secia
CONSTRUCII DE MAINI
ADHESIVE BONDING OF SURFACE TREATED ALUMINIUM

BY
BIRGIT KJRSIDE STORM

Abstract. Adhesive Bonding of aluminium and aluminium alloys can be carried
out so a strong bonding is the result. The perfect bonding between two aluminium
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. For making a strong bonding
a structural adhesive shall be used. A structural adhesive can be made of epoxy,
polyurethane, acrylate, cyanoacrylate and a few other types. In some cases an
anaerobe adhesive can cure up with aluminium. There is interaction between the
adhesive, the aluminium alloy and the surface treatment. Aluminium creates in
atmospheric air immediately an oxide layer on the surface. Anodized aluminium
reacts with contaminations and CO2, which gives a surface with a low surface tension,
though aluminium will have a high surface tension. The low surface tension of the
reacted surface cause a bad wetting of the surface with the adhesive, and an
insufficient wetting causes bad adhesion between the aluminium and the adhesive. For
making a good bonding a good wetting is necessary. The sufficient wetting can be
made on the aluminium by making a surface treatment. The surface treatment can be
an anodizing. For obtaining the best adhesion on anodized surface the adhesive shall
be carried out before the sealing immediately after the anodizing. The fresh anodized
aluminium has a bigger surface and open pores, which cause a mechanical bonding in
addition to the adhesion. The non sealed aluminium has reactive groups, which in
some cases can react with reactive groups in the adhesive and can cause a chemical
bonding as well. The non-sealed anodized aluminium can be treated with a primer,
which can react with the active groups and can run into the pores and fill out the
porousity. Before the primer has cured up totally the adhesive shall be added to the
system for making a chemical reaction between the primer and the adhesive. Using
such a system can give a higher strength in the bonding.
Key words: aluminium, anodizing, adhesive, adhesive bonding, surface
treatment, wetting
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
Birgit Kjrside Storm
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.
2. Wetting and HSP

The surface shall be able to be wetted by the adhesive, if a strong bonding
shall be created. If it is possible to obtain a good wetting, the possibility to
increase the physical adsorption in the interphase is present. For obtaining a
good wetting the metal/anodized metal surface must have a surface tension
which is higher than the surface tension of the adhesive. In figure 1 the wetting
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.
Fig. 1 Surface tension and determination of surface tension [5].
The solubility parameters can be a one-dimensional parameter or a three

dimensional parameter. The solubility parameter is defined as
(1)
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.
Fig. 2 Two spheres in a HSP diagram.

The two spheres are not touching each other. The two materials behind are not compatible.
If wetting of a material with another material shall be possible, the solubility

spheres for the two materials shall at least touch each other. To calculate if
wetting of a material with another material for example an adhesive is possible
the RED relative energy density value can be calculated. The RED value is
defined as:
(3)
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)
= 0,0688V1/3 [D2 + k (P2 + H2)]
is the surface tension and k is constant depending on the liquids involved.

Other similar correlations are used.
109
3. The surface of Anodized Aluminium

Aluminium alloys and anodized aluminium will theoretically have a high
surface tension, so wetting of the surface with an adhesive would not be a
problem, because the adhesives will have a surface tension much lower.
Aluminium and anodized aluminium will react with CO2 and possible
contaminations in the air and will there fore in practical have a surface tension,
which is so low that wetting with the adhesives is not possible, because the
actually surface tension is lower for aluminium surface than for the adhesive.
If the surface is new created the surface tension will still be high and wetting
will not be a problem. For solving the problem with not anodized aluminium the
aluminium surface is often chromatized or phosphatized to obtain a good wetting.
The surface of anodized aluminium itself is possible to wet with an adhesive
especially if it is new treated and that will say that as soon after the anodizing the
wetting will take place the better adhesion will be obtained. The bigger surface
the better wetting is possible and that means also that if the anodized surface is
used for bonding before sealing. The fresh anodized aluminium surface is the best
surface for adhesive bonding, because the surface there has reactive groups,
which will give possibility for chemical bonding in the interphase. The reactive
groups give a bigger contribution to the surface tension because of more hydrogen
bonds and the physical adsorption in the interphase will increase. The mechanical
interlocking will also increase because of the porousity in the anodized surface.
After sealing the surface is like un-anodized aluminium and the advantages
obtained by the anodizing will disappear for the bonding.
4. Surface Treatments
Before bonding the surface need to be clean. The surface shall as mentioned
above be with so many reactive groups as possible and the surface tension shall
be as big as possible. The possibility for mechanical interlocking shall be
present if possible.
If the surface is contaminated with oil or grease from production or cutting
process the surface tension will be the same as the surface tension of oil and
grease, which will say very low, because those materials will almost have a
contribution form the dispersion forces, very few if any from the polar forces
and no from the hydrogen forces. Such a surface is impossible to bond on. The
grease and oil shall be removed with an organic solvent as f.ex. isopropyl
alcohol or CO2. The surface can if it is non anodized aluminium or is a sealed
anodized aluminium be treated with a Scotch Brite for obtaining a bigger
surface for mechanical interlocking. The treatment with Scotch Brite shall take
place before the cleaning for removing dust from the surface.
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].
The adhesive bonding dislikes a peeling, but in practice it is difficult to

create a bonding, where peeling is not possible. The peeling is the weakest point
for an adhesive bonding and is often the limitation for making the bonding. If it
is possible the bonding shall be constructed so the forces are shear forces and
peeling cannot occur.
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.
Fig. 4 Curing of epoxy with a primary amine [4].
Fig. 5 Formation of biuret from diisocyanat and urea [2].
A structural adhesive can also be a cyanoacrylate, but for working with a

quick curing adhesive can give problems in a production. On the market there
are many different adhesives of all the mentioned types. They are different in
opening time and in strength and flexibility. They can have different colours as
well, and often the colours are used as an indicator for seeing if the adhesive is
placed all over where it shall be.
The properties of the adhesive depends on the polymer and the cross linking
system in the adhesive. Aromatic groups in the adhesive give higher strength
and aliphatic groups give higher flexibility. The more aromatic groups and the
higher density of the aromatic groups the higher strength will be obtained. The
more aliphatic groups there are in the polymer, the better flexibility there will be
possible to obtain. The density of the reactive groups in the adhesive will also
give possibility for strength and flexibility. The more reactive groups for cross
linking there are, the higher rate of cross linking can be obtained and the higher
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.
Fig. 6 Formation of DGEBA from bisphenol A and epichlorhydrin [4].
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.
Fig. 8 The HSP plot of Sikaflex-360HC. Sikaflex-360HC is a polyurethane 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.
Fig. 9 - The HSP plot of an unsealed anodized aluminium.

The anodizing has been made in the laboratory
For an unsealed anodized surface it can with some polyurethane adhesives

and some epoxy adhesives be possible to make a chemical bonding in the
interphase. Such an adhesive can be used as a primer if it can be possible to find
another adhesive, which can give the strength of the bonding and can wet the
primer.
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.
Received:January, 30,l 2010
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
4. P i z z i A. et al., Handbook of Adhesive technology, Marcel Dekker, Inc, 2nd edition,

2003.
5. www.specialchem.adhesives.com
LIPIREA CU ADEZIVI A ALUMINIULUI TRATAT SUPERFICIAL

(Rezumat)
Lipirea cu adezivi a aluminiului i aliajelor de aluminium poate fi realizat astfel nct
s se obin mbinri rezistente. O lipire perfect a dou piese din aluminiu va fi acea
lipire la care ruperea mbinrii are loc n piesa de aluminiu, lipitura fiind mai rezistent ca
aluminiul. Pentru a obine o lipitur de mare rezisten, se vor folosi adezivi structurali,
care pot fi de tip epoxi, poliuretan, acrilat, cianoacrilat etc. n unele cazuri, un adeziv
anaerob poate interaciona cu aliajul de aluminiu i cu tratamentul de suprafa.
Aluminiul, n aer atmosferic, creeaz imediat un strat de oxid la suprafa. Aluminiul
anodizat reacioneaz cu contaminatorii i cu CO2 rezultnd o suprafa cu tensiuni
superficiale reduse, dei aluminiul are tensiuni superficiale ridicate. Asemenea tensiuni
superficiale reduse vor determina o umectare slab a suprafeei cu adeziv, o umectare
insuficient determinnd o adeziune necorespunztoare ntre aluminiu i adeziv. Pentru o
lipire de calitate este necesar o bun umectare. O umectare suficient poate fi obinut
prin aplicare de tratamente superficiale. Un asemenea tratament poate fi unul de
anodizare. Pentru a obine adeziune optim pe suprafaa anodizat, adezivul trebuie aplicat
imediat dup anodizare. Aluminiul proaspt anodizat are suprafa de contact mai mare i
pori deschii, rezultnd o lipire mecanic suplimentar adeziunii. Aluminiul poate avea
grupuri reactive care uneori pot reaciona cu grupuri reactive din adeziv, rezultnd i o
lipire chimic. Aluminiul anodizat poate fi tratat cu o amors, care poate reaciona cu
grupurile active i poate intra n pori i umple porozitile. n asemenea cazuri, adezivul se
aplic nainte de epuizarea amorsei, pentru a avea loc reacie chimic ntre amors i
adeziv. Asemenea soluii pot asigura rezisten sporit mbinrilor cu adezivi.

Publicat de
Secia
CONSTRUCII DE MAINI
INDUSTRIAL APPLICATIONS OF
THE PNEUMATIC MUSCLES
BY
IOANA PETRE1, DAN PETRE2, CRISTINA FILIP1,

and LAVINIA NEAGOE1
Abstract. Pneumatic muscles were, lately, used in many applications
especially in the field of industrial robots, because of the evolution of the
technology. The characteristics of the pneumatic muscles make them being easy
to use and with great performances. Pneumatic muscles have applications in
robotics, biorobotics, biomechanics, artificial limb replacement and industry. In
this paper I will present some industrial applications of the pneumatic muscles.
The utilization of the pneumatic muscle is soon to become a good alternative for
the present day electric or mechanically drives because of their characteristics and
because of the evolution of the industrial development.
Key words: pneumatic muscle, actuator, industrial application.
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
Ioana Petre et al.
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).
Fig. 1 Pneumatic muscles and their function principle

(source http://www.festo.com/hm2001/eng/1001.htm).
Pneumatic muscles were first conceived in 1930 by G a r a s i e v, a Russian

inventor [2]. According to B a l w in [3], J. L. McKibben introduced it to
motorize pneumatic arm orthotics for helping control handicapped hands: due to
the similarity in length-load curves between this artificial muscle and skeletal
muscle, it seemed an ideal choice for this purpose [4], [5].
The artificial muscle, which construction is simple, was made of a rubber
inner tube covered with a shell braided according to helical weaving. The
muscle was closed by two ends, one being the air input and the other the force
attachment point. When the inner tube was pressurized, the muscle inflated and
contracted [6].
The Bridgestone rubber company (Japan) proposed a redesigned and more
powerful version of the pneumatic muscle in the 1980s under the name of
Rubbertuators and used them to power an industrial use robot arm, Soft Arm.
At the present, McKibben-like muscles are being brought to the market by
Festo Ag. & Co.
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.
Fig. 2 Working principle of a

pneumatic muscle
Fig. 3 Dependence of the force on the

envelope angle and on the working pressure [7]
The force F developed by pneumatic muscle is given by Eq. (1) [4].

(1)
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
Ioana Petre et al.
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].
3. Industrial Applications of Pneumatic Muscles

Pneumatic muscles were, lately, used in many applications especially in the
field of industrial robots, because of the evolution of the technology.
The major application of pneumatic muscles is in the field of industrial
robots, usually in those that mimic human actions. Because of their analogy
with human skeletal muscles, they are able to perform many functions specific
to human hand.
As example in this area we can specify the stepping robot WAP-1 is the first
biped robot designed in 1969 by Ichiro Kato from Waseda University of Tokyo.
(Fig. 4a)
a
b
Fig. 4 a WAP-1(source www.androidworld.com) ; b - Humanoid robot manufactured
by Shadow robotic company (source www.shadowrobot.com)
121
The humanoid bipedal walking machine developed by Shadow Robot

Company is a robot that can function in a human environment to do various
repetitive things. (Fig. 4b)
The humanoid muscle-robot developed by Festo AG & Co. in collaboration
with the department of bionics and evolutionary technology of the Technical
University of Berlin (Fig. 5). This robot is a feasible extension and place holder
of men capable of operating in places either too dangerous or inaccessible for
human beings. It can operate from terrestrial and deep sea operations to tasks
carried out in outer space.
Fig. 5 Humanoid muscle-robot moving (source www.festo.com).
Pneumatic muscles have been used in construction of artificial limbs.

At the bio robotic lab of University of Washington the limb as shown figure
6a was developed. The major requirements of their research team were:
continuous operation, low weight, quieter operation, user satisfaction, no
maintenance.
a
b
Fig. 6a Artificial limb developed at the bio robotics Lab, University of Washington.
b Shadow Leg (source www.shadowrobot.com)
122
Ioana Petre et al.
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.
Fig. 7 The Shadow Hand (source www.shadowrobot.com).
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
2. The utilization of the pneumatic muscle is soon to become a good

alternative for the present day electric or mechanically drives because of their
characteristics and because of the evolution of the industrial development.
Transilvania University of Braov,

Technological Engineering Faculty,
Economic Engineering and Production Systems Department,
Braov, Romnia
e-mail: ioana.petre@unitbv.ro
2
Material Science and Engineering Faculty,
Materials Engineering and Welding Department,
e-mail: petredanbv@yahoo.com
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
Ioana Petre et al.
APLICAII INDUSTRIALE ALE MUCHILOR PNEUMATICI

(Rezumat)
Muchiul pneumatic este un sistem bazat pe o membran care se contract, i care,
sub aciunea aerului comprimat i mrete diametrul i i micoreaz lungimea.
Lungimea cursei depinde in mod direct de nivelul presiunii alimentate. Un muchi
pneumatic include un tub interior care este realizat dintr-un material elastic, de obicei
neopren. Acest tub este acoperit de un esut cu mai multe straturi realizat din nylon,
pentru a-i da rezisten i pentru a-l proteja de influenele din mediul de lucru.
Caracteristici precum capacitatea de a absorbi ocurile, greutatea redus, rezistena
la ocuri, gabarit si masa redusa pe unitatea de putere (1KW/kg), elasticitate
(comportare ca de arc) datorata pe de-o parte compresibilitii aerului si pe de alta
variaiei forei cu deplasarea, amortizarea ocurilor datorate impactului, posibiliti de
conectare uoara, sigurana (fr pericol de electrocutare sau incendiu) fac din muchii
pneumatici instrumente fezabile pentru utilizarea uoar in domenii diverse i cu
performane deosebite. Domeniile de aplicare ale muchilor pneumatici se refer la
robotic, biorobotic, biomecanic, dispozitive de protezare si sustinere a scheletului
osos i industrie.
n aceast lucrare se prezint cteva aplicaii industriale ale muchilor pneumatici. Dei
prima apariie a muchilor pneumatici a fost n 1930 fiind inventai S. Garasiev, iar apoi
n 1950 muchii cu membran mpletit introdui de J. L. McKibben acetia au cunoscut
o utilizare mai intensa abia in ultima perioad.
Cteva exemple in domeniul roboilor industriali sunt: braul acionat pneumatic
realizat de J.L.Mc Kibben, robotul WAP 1 realizat la Universitatea Waseda din Tokyo,
robotul humanoid conceput de Festo in colaborare cu Universitatea Tehnic din Berlin,
robotul humanoid realizat de Compania Shadow, proteza realizat de Universitatea din
Washington, piciorul i mna realizate pentru Compania Shadow, care efectueaz
micri similare cu cele efectuate de corpul uman. Desigur c aplicaii sunt multe si nu
au fost toate menionate in acest articol, nsa trebuie reinut c muchii pneumatici,
datorit caracteristicilor lor, sunt utilizai n special n domeniul roboilor mobili.
Acetia nlocuiesc cu succes motoarele electrice sau hidraulice fiind mai uor de folosit
i ndeplinind aceleai funcii.
Dezvoltarea continu a tehnologiei a dus la apariia unor noi arii de aplicaie pentru
muchii pneumatici, cum ar fi: tehnologia simulrii, procese de tiere cu vitez mare,
tehnologia aeronautic, lucrul cu lemnul, lucrul cu metalul, industria medical,
construcii, industria minier.

Publicat de
Secia
CONSTRUCII DE MAINI
SOME NEW RESOURCES ON COMPUTER ASSISTED

EXPERIMENTAL RESEARCH OF THE ABSORBED ELECTRIC
POWER IN MANUFACTURING SYSTEMS
BY
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
Fig. 1 Experimental features of the EP bench test.
evolutions [6] provided by two transformers CT and VT) via a numerical

oscilloscope (ADC 212/50, Pico Technology Limited, UK). The acquired data
are processed in a personal computer. The instantaneous electric power p(t)
evolution (IEP) is given by: p(t)=3u(t)i(t). The AEP (one value on each period
T) is given by:
(1)
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
Q[tl ] 3 U [tl ] I [tl ] sin([tl ]) .
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
3. Some Experimental Research Results on Monitoring
Let be first a simple experiment. The belt drive transmission BT1 is

switched-off (the belt was removed), no loading on EDMs rotor anymore.
Fig. 3 The AEP evolution during EDM starting procedure

(no loading on the EDMs rotor).
The evolution of AEP during the EDM starting procedure (star-triangle
Fig. 4 Gearbox start-stop clutching dynamics (on Kd1

configuration) mirrored in AEP evolution.
connection) is given in Fig. 3. A strong transitory regime (high power, short
129
time, sees Bs area) is observed, because of the dynamics of accelerated rotor

motion, it needs a big amount of AEP stored as kinetic energy. In D the star
connection is changed automatically in triangle connection, also a transitory
regime (marked as Bt) occurs.
Fig. 5 Numerical filtered (with p=10) evolutions for AEP

(given in Fig. 4) and REP.
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
For certain experimental purposes the AEP evolution can be numerical

filtered using a sliding average low pass filter described with the equation:
p x [i ]
xout [ j ] = in where: p is the filter parameter, xin[i] - the input signal, xout[j]
i =1 p
- the output signal, both in numerical format, j=i+p. Figure 5 shows the result
of filtering for AEP evolution given in Fig.4 (with p=10).
Also is presented the filtered evolution of REP in the same experimental
conditions. As it is well known from the electrical network theory, the variation
of REP (1,036 VAR peak to peak) is smaller than the AEP variation (4,805 W).
The evolution of AEP during a third experiment is described in Fig. 6.
There is a dynamic behaviour with negative absorbed AEP. Before A, the
gearbox moves in steady-state idle running regime on 1600 rpm (using Kd2
kinematic chain with MC2 clutch engaged). In A the clutch MC3 is engaged
(and MC2 is automatically switched-off), so Kd2 becomes active and Kd1
inactive. Here occurs a transitory regime B, with negative AEP (the angle >
/2, see Eq. (1)), for a short time the EDM works as a brake, it convert the
Fig. 6 Clutching dynamics with negative AEP.
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
4. Gearbox Diagnosis using the AEP Processing in Frequency Domain
There is also a very important experimental research resource useful in

gearbox diagnosis. The AEP or IEP evolution in steady-state it is a mixture of
many variable components. Each variable signals component describes the
behavior of a gearboxs part (belts, spindles, gears, et cetera). The identification
of these components can be done [1] by computer assisted AEP or IEP power
spectral density (PSD) analysis using Fast Fourier Transform (FFT). The
correlation between signals components and gearboxs parts can be done using
the frequency and the speed of rotation. Figure 7 presents a zoom on the PSD of
the AEP evolution in steady state on 1000 rpm (already described before; see on
Fig. 4 the encircled area).
On Fig. 7 there are a lot of different peaks, each one described by the frequency
and the amplitude (the higher peaks amplitude are not clearly indicated because
Fig. 7 The evolution of power spectral density of the

AEP signal absorbed on 1000 rpm (steady-state).
of zoom). The PSD amplitude is proportional with the amplitude of signal

component. Each peak has a number, for each one there is a short description
about the frequency and the correspondence with the gearbox parts (see Fig. 2).
The peak 5 is associated with the spindle II behaviour, the peak 7 with the
spindles III and V, the peak 9 with the spindle I and the peak 10 with the
EDMs rotor (24.84 Hz frequency, for 1490.4 rpm the real speed of rotation).
The peak 9 is generated by the driving belt pulley run-out error motion (65 m).
There are also some harmonics of these peaks. There is a very interesting
research result, with certain scientific priority, here is very well mirrored the
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 .
Fig. 8 The BT1s belt behaviour mirroring in PSD

analysis of the IEP signal evolution.
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
1. The paper proves that the computer-assisted monitoring of EP parameters

on manufacturing systems (or any other electric actuated equipment) can be an
available experimental research procedure.
2. The computer assisted EP monitoring uses very simple procedures of data
acquisition and processing. Using these research features, it is possible to detect
and study a lot of new static and dynamic phenomena from kinematic chain.
133
Acknowledgements. The author would like to thank Prof. Costache D r u t u. For, he

is the one who, long time ago, gave me the basic idea of some of my researches.
Received:
Gheorghe Asachi Technical University

Department of Machine-Tools
Iasi, Romania,
e-mail: horodinca@tuiasi.ro
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.
NOI RESURSE ALE CERCETRII EXPERIMENTALE

ASISTATE DE CALCULATOR A PUTERII ELECTRICE
ABSORBITE N SISTEMELE DE FABRICAIE
(Rezumat)
Lucrarea prezint o serie de noi oportuniti ale cercetrii experimentale a
sistemelor de fabricaie (monitorizare i diagnoz) pe baza analizei evoluiei temporale
i n domeniul frecven a parametrilor ncrcrii energetice (putere activ PA i
instantanee PI) ai electromotoarelor de acionare. Se prezint exemplificri legate de
monitorizarea proceselor tranzitorii i de utilizarea analizei spectrale a semnalelor PA i
PI, pentru cercetarea comportrii lanurilor cinematice ale micrii principale.

Publicat de
Secia
CONSTRUCII DE MAINI
INTEGRATED CAE DEVELOPMENT OF PRECESSIONAL

DRIVES USING AUTODESK INVENTOR PLATFORM
BY
ION BOSTAN, VALERIU DULGHERU

and ANATOL SOCHIREANU
Abstract: The paper presents the modelling and simulation of processional
drives designed in two variants capable of high transmission ratio and torque for one
stage compact construction. The constructions were designed in Inventor and also as
multi body systems in MotionInventor. The simulations of the drives provide
information concerning positions, velocities, accelerations, point trajectories, forces
and moments, energies, as well as contact forces at the contact between gear teeth
and satellite teeth and other data concerning the system. The bearings of the two
drives were modelled as multi body systems in SolidDynamics for defining an
approach for studying the noise emission during running through simulation with
possibility of extension to the contact between teeth.
Key words: precessional transmissions, Inventor, 3D multi body model
SolidDynamics
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
creation and management of parametrical models of the mechanical assemblies

on the basis of CAD-CAE.
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
As the planetary processional

transmission is the new transmission, in
the beginning of effectuation the
simulation
and
calculation
with
programs CAE it is necessary to
calculate theoretically base parameters.
On fig. 1 is shown kinematical scheme
of planetary processional transmission,
but on fig. 2 is shown the design of
planetary processional transmission. The
main elements of transmission make:
crankshaft 1, block satellite 2, fixed
wheel gear 3 and mobile wheel gear 4.
The important point this that, having
Fig. 1 Kinematical scheme of
such design (single-stage) is possible to
planetary processional transmission.
receive transfer rate up to 3600, using
designs of double-stage it is possible to receive transfer rate up to 14 million.
Obviously that, having such big transfer rate, there are big loadings on a tooth.
This problem is solved by that, all teeth it is participate in the gearing, and
loading is transferred by half from them. Meaning that, all teeth are in gearing at
the big loadings, the kinematic error is small.
Principle of functioning of the planetary processional transmission the
following: crankshaft 1 with inclined section to specify a block satellite 2
spatially spherical movement, block satellite by means of roller crown
interaction with fixed wheel gear 3 and mobile wheel gear 4 (having a special
structure generated by means of the equations), in turn a mobile wheel gear 4 it
is rigidly connected with output shaft of a reducer, transfers the moment and
speed of rotation. The direction of rotation of an input shaft and output shaft can
be in one or in different directions. It is visible and in calculation of transfer rate
if it has positive number we have an identical rotation. The transfer rate of a
planetary processional transmission is defined by the relation (1).
1 3 6
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 .
Fig. 2 Design of planetary processional transmission.
138
Ion Bostan et al
3. Calculation by CAE Simulation of Kinetostatics

Parameters of Processional Transmission
3.1. 3D Model elaboration of Planetary Processional Transmission
Calculation of planetary processional transmission by a simulation is carried

out using the simplified 3D model created in program Motion Inventor 2004+
but (4) from which it is possible to determine and check up dynamic loadings in
bearings which have been designed earlier.
Dynamic processes in planetary processional transmission derive, to a great
extent, from the interaction of conical rollers of the satellite crowns with
generating surfaces of central wheel teeth. The bearing capacity defined by gear
forces (static and dynamic), the noise emission and the transmission
vibroactivity, depend on the gear dynamic processes on the whole. With
account of these important factors in the elaboration of 3D model of the initial
processional gearing, the linear contour of central wheel teeth profile (fig. 3, a)
were designed applying parametric equations:
m
m
X1E
= k2mZ1E
+ dm2 ;
m
Y1Em = k1mZ1E
d1m ;
(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.
Fig. 3 3D linear contour of the central wheel teeth (a),

and the model 3D of the central wheel (b).
The 3D model for calculation is shown on fig. 4, it includes a crankshaft 1,

a fixed wheel gear 2, the block satellite 3, a mobile wheel gear 4 rigidly
connected with output shaft 5.
Fig. 4 3D Model of planetary processional transmission.
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
Fig. 5 Kinematics joints in the planetary precessional transmission.
3.2. Precessional Transmission Kinetostatic Simulation
The simulation of model processional gearings has been created in some

stage. At the first stage has been executed the kinematic analysis, from
definition of the following parameters: transfer ratio, absolute angular speed of
the block satellite, relative angular speed of the block satellite, angular speed on
the output shaft. The received results are shown on Fig. 6.
Fig. 6 Kinematic calculation of planetary processional transmission.
At the following stage has been executed the kinetostatics analysis with
calculation and simulation of total loadings in gearings.
141
An important advantage of the processional planetary transmission consists

in the insurance of absolute multiplicity (=100%) of the central wheel teeth
and pinion simultaneous gearing. This advantage provides small dimensions
and mass (reduced material consumption) and small cost of the final product.
b.
a.
Fig. 7 Normal forces among the teeth and

their distribution among simultaneously geared teeth.
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
2. The elaboration of the planetary processional transmission dynamic

sample (CAE) based on developed 3D model allows verification of kineticstatic parameters previously defined.
Received: January, 15, 2010
Technical University of Moldova

Department of Theory of Mechanisms and Machine Parts
e-mail: dulgheru@mail.utm.md
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).
DEZVOLTAREA INTEGRAT CAE A TRANSMISIILOR PRECESIONALE

UTILIZND PLATFORMA AUTODESK INVENTOR
(Rezumat)
n lucrare se prezint modelarea i simularea transmisiilor planetare precesionale
proiectate n dou variante cu raport de transmitere mare i capacitate portant ridicat
ntr-o singur treapt. Construciile transmisiilor precesionale au fost proiectate n softul
Inventor i simulate ca sisteme multicorp n MotionInventor. Simulrile transmisiilor au
permis obinerea informaiilor privind poziia, vitezele, acceleraiile, traiectoriile
punctelor, fore i momente, energii, de asemenea, forele de contact la contactul dintre
dinii roilor dinate i ai satelitului i alte date privind sistemul. Angrenajele a dou
transmisii au fost modelate ca sisteme multicorp n SolidDynamics pentru studiul
emisiei de zgomot n timpul lucrului prin simulare, cu posibilitatea extinderii la
contactul dintre dini.

Publicat de
Secia
CONSTRUCII DE MAINI
PROPOSALS FOR THE IMPROVEMENT OF

THE FLUIDIC SPIRAL JETMILLS' ACTIVITY
BY
ILEANA FULGA1 and EUGEN STRAJESCU2

Abstract. In this paper is presented a documentary study about the particle
size reduction. There are presented the authors' realizations concerning the
optimization of the spiral jetmills produced in the APTM Enterprises, based on a
new mathematical and experimental model of the flows, developed by the authors.
There are presented conclusions about the research directions to be attacked in the
future.
Key words: micronization, spiral jetmills, fluidic mills, optimization.
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
Ileana Fulga and Eugen Strajescu
2. Clarifications about the Possibilities to improve the Spiral Jetmills'

Activity, resulted from the Mathematical Model
2.1. The Supersonic Jet's Protection
The construction of the Venturi tubes by where the material is introduced
and of the nozzles from where the compressed air is blowing off represents an
important factor for a good activity of the fluidic spiral jetmill. The super sonic
nozzle (the one by where is insufflated fluid under pressure, realizing important
debits) must have a determined length. The prolongation of the active area of
the nozzle behind the peripheral wall, or nozzles with inclined termination, and
not perpendicular on the axis's direction of the nozzle could realize a
compressibility effect manifested in an opposite sense, because the boost
pressure decrease, the circulation is reduced and the jet is displaced in a
direction more nigh to the tangent.
Of course this kind of position must be considered as "off-set" beside the
desired equilibrium position, considered as "ideal". It could be imposed in that
case the "protection" of the protuberant peak of the nozzle by an aerodynamic
modeling of the peripheral wall of the milling room, or by the practice of some
clefts on the same wall.
2.2. The Geometry of the Supersonic Jet
It is proved that the employment of the convergent - divergent nozzles, at
necessary and effective pressures do not present great advantages in the great
majority of the practical applications.
It is recommended for the reduction of the losing, the construction of jets as
short as possible, with an alignment near the perfection, with the same radial
angle. Because deviations from the alignment between different jets although
minima, create important tourbillions, unequal, in the transversal flux,
discontinuities in the peripheral circulation of the particles and tourbillions in
the central classification vortex.
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
Fig. 1The analyzis of the evolution of the APTM spiral jetmills,

considering the application of the process' optimization.
The results presented in this doctoral dissertation were transposed in
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
The effect of the micronization about the dispersed pigments can be

observed the besting the case of the sample of Prussian Blue (fig. 4): being in a
non micronized stadium this pigment presents agglomerates at 100-150 m,
whiles after the micronization it is to observe a granulometry contained
between 20-25 m.
Fig. 4 Pigment (Prusia blue) nonmicronizes = 150 m (left) and

micronized = 20 m (right).
Fig. 5Example of cosmethic powders micronized with optimization

algorithms; visualization on microscope with phase contrast.
Even without the microscope or another magnifier, from the photos'

observation it is to note the intensification of the chromatic coverage of the
pigment (fig. 2), because after the micronization, the specific surface of the
product is majorized for 2000 - 3000 times, and particles are very uniform from
the point of view of the granulometry.
Another object of the optimization is the getting of a micronization with a
plain classification room, in order to not charge to frequent the mill (fig. 6).
149
Fig. 6 Example for an optimized production for

the filling up of the classificatory.
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
micronized powders are an especial good quality, representing a revolution in

the cosmetic and pharmaceutical industries.
Fig. 6 Non micronized cosmethic pigments,

with a granulometry of 120 - 350 m.
Fig. 7 Micronized cosmethic pigments, with a granulometry of 15 - 30 m.
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
4. From the experimental researches it was obtained important conclusions

that permit to optimize the spiral jetmills in at least two directions.
5. The first one is the micronization operation conditions and the best
geometry of the mill.
6. A second one is the optimization of the operation conditions in order to
realize the filling of the classification part of the mill. In this case, the
productivity of the mill increases with important values.
7. The researches for the optimization in the case of the spiral jet mills must
go on,, because we appreciate that it exists a lot of unexplored ways and
possibilities.
Acknowledgements. The authors would like to thank APTM Enterprises for the
support and constructive comments and also to the staff of the Laboratory of APTM.
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
5. F u l g a I . , S t r j e s c u E . , Contributions to the Fluid Dynamics in Jetmills by a

New Aerodinamic Model of Supersonic Flow, Buletinul tiinific U.P.B., Serie
D, 74, 1, ISSN 1454 - 2358, Romnia (2008).
6. F u l g a I . , G a d o n n a J . P . , R e m o l u e M . , F r o i d e v a u x L . , A j a n a
A . , C h a y v i a l l e L . , Improvements in MDI technology with total process
containment, European Aerosol Conference, EAC, Belgium, 2005.
7. F u l g a I . , S t r j e s c u E . , Probleme actuale i de perspectiv n construcia
utilajelor pentru mcinare i n tehno-logia mcinrii, A XI-a Conferin
tiinific cu participare internaional TEHNOMUS XIII, Universitatea "tefan
cel Mare" Suceava, Romnia, Facultatea de Inginerie Mecanic, pg. 94-103,
ISBN 973-666-154-7, 2005.
8. F u l g a I . , S t r j e s c u E . - The Present Stage and the Perspectives of Researches
in the Area of Fluid Grinding Operation Equipment with Spiral Jets,
International Conference on Manufacturing Systems, Buletinul Institutului
Politehnic Iai, Romnia, secia Construcii de maini, LI (LV), 5, ISSN 15826392 (2005).
9. F u l g a I . , S t r j e s c u E . - Contributions Concerning the Perfecting of the
Crushing Technologies of the Cosmetic Particles and of the Metallic Carbides in
the Wheels with Fluid Jets, International Conference Technologies and Quality
for Sustained Developement, pg. 357 - 360, ISBN 973-720-035-7, 2006.
10. F u l g a I . , S t r j e s c u E . - Experimental Model of the fluidic Mills with Spiral
Jets. (Long Abstract), The 3rd International Conference on Manufacturing
Science and Education, MSE, Sibiu, Romnia, pg. 79 - 80, ISSN 1843 - 2522,
2007.
PROPUNERI DE OPTIMIZARE A FUNCIONRII

MORILOR FLUIDICE CU JETURI N SPIRAL
(Rezumat)
n lucrare se prezint succint concluziile unui studiu anterior asupra realizrii unui
model matematic privind comportarea jeturilor supersonice ale morilor fluidice cu jeturi
n spiral. Pe baza lor ,se enun unele posibiliti de optimizare a funcionrii acestor
tipuri de mori, precum i rezultate comparative ale mcinrii unor pulberi cosmetice.
Rezult din lucrare i faptul c micronizarea produselor simplific substanial
tehnologiile de mcinare, ceea ce micoreaz costurile producerii pulberilor de orice fel.

Publicat de
Secia
CONSTRUCII DE MAINI
DYNAMIC ANALYSIS OF MECHANICAL-HYDRAULIC

PROTECTION MECHANISMS OF LOW POWER HORIZONTAL
AXIS WIND TURBINES THROUGH VERTICAL TILTING
BY
DORU CLRAU, IRINA TIA, DAN SCURTU

and BOGDAN CIOBANU
Abstract. Generally, 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. The paper present the dynamic analysis of a mechanicalhydraulic protection mechanism used for low power HAWT. The method consists
of the vertical tilting of the wind turbine. The analysis of the system responses
revealed a good dynamic performance when using the vertical tilting solution
chosen to protect low power HAWT.
Key words: HAWT, protection mechanism, vertical tilting.
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
Doru Calarasu et al.
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
Fig. 1 Constructive block diagram.
Thanks to its components (Fig. 1), this mechanical-hydraulic protection

mechanism also ensures the operation of the wind turbine at wind speeds higher
than the rating speed, having the following functions:
a) the rotor assembly is directed towards the wind stream using the
mechanisms lower bearing, by means of which the assembly is placed on the
support pillar;
b) the tilting device becomes active when the wind speed exceeds the safety
limit and it reduces the active turbine surface exposed to the wind;
c) the balancing device ensures stable balance to the wind turbine rotor at wind
speeds lower than the safety limit and unstable balance when the wind speed
reaches the safety limit;
d) the damping device protects the plant against wind gusts by delaying the
propellers resuming its initial position.
3. Digital Simulation Tests performed on the Conceptual Model of
Mechanical-Hydraulic Protection Mechanisms
The Matlab-Simulink programming environment was used for the
performance of the digital simulations.
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
Joint Initial Condition

Joint Spring & Damper
Weld
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.
Fig. 3 The rotors surface exposed to wind depending

on the turbine axis angle of inclination.
156
The mechanical-hydraulic protection mechanism includes the tilting device,

the balancing device of the wind turbines own weight and the two-way
movement damping device. The initial position (meaning a normal vertical
turbine operation) is triggered by a back stop. The model allows testing using a
set of signals to simulate the wind speed evolution.
Through vertical tilting, the rotors surface exposed to wind decreases as
shown in Fig. 3.
The arm of the component determining the tilting moment increases with
the increase of the turbine axis tilting angle (Fig. 4). Fig. 5 shows the variation
of the moment determined by the rotors tilting strength.
Fig. 4 Variation of the rotor tilting strength arm.
Fig. 5 Moment of the rotor tilting strength.
157
Fig. 6 shows the set of signals used to simulate wind speed variation.
Fig. 6 Signals for model testing.
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.
Fig. 7 Axis inclination angle at wind

ramp speed.
Fig. 8 Tilting strength at wind

ramp speed.
Figure 8 shows the tilting strength variation in the conditions described

above.
158
Fig. 9 Inclination angle of the turbine

rotor axis.
Fig. 10 Rotor tilting strength.
If the system constructive-operational parameters are changed (the damping

coefficient and the return strength decrease), figure 9 shows the evolution of the
inclination angle of the wind turbine rotor axis, while figure 10 depicts the
evolution of the tilting strength.
4. Conclusions
1. The simulation diagram allows both constructive and operational system
parameters and operation testing signals to be altered.
2. Digital simulation may be used to optimize constructive parameters
(balancing components weight, rotor axis section lengths, balancing weight,
damping system characteristics).
3. The analysis of the system responses revealed a good dynamic
performance when using the vertical tilting solution chosen to protect low
power horizontal axis wind turbines.
Acknowledgements. This work has been supported by the Romanian National Fund
for Science and Research through National Centre for Programme Management under
contract No. 21-047/2007.
Received: April 16, 2010

Department of Fluid Mechanics
Iasi, Romania
e-mail: dorucalarasu@yahoo.com
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.
ANALIZA DINAMIC A MECANISMULUI MECANO-HIDRAULIC DE

PROTECIE A TURBINELOR EOLIENE CU AX ORIZONTAL DE MIC PUTERE
PRIN BASCULARE N PLAN VERTICAL
(Rezumat)
Sistemul de control cu care este prevzut o turbin eolian are rolul de a diminua
ncrcrile dinamice pe pale, de a crete fiabilitatea turbinei i sigurana n exploatare.
Acest sistem reprezint una dintre direciile prioritare de aciune n vederea
rentabilizrii turbinelor eoliene cu ax orizontal de mic putere.
Lucrarea prezint analiza dinamic a unui mecanism mecano-hidraulic de protecie
a turbinelor eoliene cu ax orizontal (TEAO) de mic putere prin bascularea turbinei
eoliene n plan vertical. Bascularea ncepe atunci cnd viteza vntului depete viteza
160
considerat limit pentru sigurana turbinei. Aceast soluie ntrunete caracteristicile

solicitate sistemului: este fiabil, nu conine elemente de acionare electric, nu conine
electronic nglobat, nu presupune existena unei transmisii hidraulice.
Pentru viteze ale vntului mai mici de o limit maxim impus, ansamblul rotor nu
basculeaz. Elicea se rotete cu o vitez unghiular care depinde de viteza vntului i de
ncrcarea mecanic la ieire. Pentru viteze mai mari sau egale cu cea maxim impus,
ansamblul rotor basculeaz iar elicea i modific planul de rotaie.
Testarea prin simulri numerice a modelului conceptual de mecanism mecanohidraulic protecie a TEAO de mic putere prin basculare n plan vertical s-a efectuat
utiliznd mediul de programare Matlab-Simulink.
Din analiza rspunsurilor sistemului rezult performane dinamice bune pentru
soluia basculrii n plan vertical utilizat n scopul proteciei turbinelor eoliene cu ax
orizontal de mic putere.

Publicat de
Secia
CONSTRUCII DE MAINI
RESEARCH ON HYDRAULICALLY SYSTEMS WHICH

MOVE HEAVY MASSES ON SMALL DISTANCES
WITH LOWER FREQUENCIES
BY
CONSTANTIN CHIRI, ADRIAN HANGANU

and DANIEL CALFA
Abstract. In this paper the authors try to present the mathematical models,
static and dynamic, useful for designing of hydraulically systems which move
heavy masses on small distances with lover frequencies. In the last time
mathematical modeling represents an obligatory step in the designing process of
the new products. In this paper are presented the results of the theoretical
simulation in concordance with the experimental results. 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.
Key words: research, modeling, simulation, data acquisition.
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
Constantin Chirita et al.
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
In the relation (2) p is the balance pressure.

(3)
p=
F
.
S
For V = 8 cm3, S = 80 cm2, F= 32.000 N, E = 1,5e9 N/m2, the results are:

xT= 1 mm, xR= 0,999 mm.
163
3. Dynamic Mathematical Models

For both systems from Fig. 1 we can use the following model:
(4)
(5)
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.
Fig. 2.a The simulating chart.
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. 3 Data acquisition application in Lab View.
The feedback time of the system is very short approximately 0,003 s, as we

can see in Fig. 2. After the collecting data procedure from the experiment,
where was used a LabVIEW interface with NI DAQ 6024, for the speed and
movement we obtain the following characteristics, see Figs. 3, 4 and 5.
Fig. 4 Piston movement after a period of nr. of steps generated by the pump.
165
Fig. 5 The real speed of the piston in mm/minutes.
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.

DISAHP Department
Iasi, Romania
e-mail: disahp@gmail.com
166

REFERENCES
1. C h i r i C., D a m i a n L., H a n g a n u A.-C., C a l f a D., mpingtor de

translaie pentru cuptor rotativ de calcinare. Propunere de brevet de invenie, nr.
A00317/ 10.05, 2007.
2. C h i r i , C., H a n g a n u, A.-C., Modelation and Simulation of Hydraulically
Systems which move Heavy Masses on Small distances with Lower Frequencies,
International
Scientific-Technical
Conference
HYDRAULIC
AND
PNEUMATICS 2007, Wroclaw, Poland ,October, 10-12, 2007, pp. 404-408.
3. P r o d a n D., Hidraulica mainilor-unelte , Editura Printech, Bucureti, 2004.
4. P r. o d a n D., Maini-unelte, modelarea i simularea elementelor i sistemelor
hidrostatice, Editura Printech, Bucureti, 2006, pp. 210-218.
5. * * * LabVIEW 7.2, 2006.
6. * * * http://www.hydramold.com.
7. * * * http://www.ni.com.
CERCETRI PRIVIND SISTEMELE HIDRAULICE PENTRU DEPLASAREA

MASELOR MARI PE DISTANTE MICI, CU FRECVEN REDUS
(Rezumat)
n lucrare autorii prezint cercetrile efectuate n cadrul DISAHP pentru proiectarea
sistemelor hidraulice care deplaseaz mase grele pe distante mici, cu frecven redus.
Sunt evideniate rezultatele simulrii n concordan cu rezultatele experimentale la
deplasarea unei sarcini grele de 200 t, pe distane scurt 50 mm, cu frecven joas, ntrun un ciclu complet de 24 de ore, pentru beneficiarul Mittal Steel din Galai.

Publicat de
Secia
CONSTRUCII DE MAINI
PNEUMATIC METERING SYSTEM FOR AMOUNT

OF WATER EXTRACTED IN CONVECTIVE
DRYING PROCESSES
BY
EROL MURAD1, CTALIN DUMITRESCU2,

GEORGETA HARAGA1 and LILIANA DUMITRESCU2
Abstract. Increased decentralization of drying processes for agricultural
products require reliable and easy handling installation at low costs. The
economic efficiency of these plants depends on the use of optimal management
for processes that require measurement of the mass of water extracted, in the
thermal conditions of drying chambers. We developed a new scheme for
measuring on-line the weight of tray loaded with dry material using a pneumatic
force transducer working under sampled measurement. The solution presented is
remarkable due to very low energy consumption and low cost automation feature.
The measurement system is coupled to PLC for automatic management of driers.
Dynamic behaviour and energy consumption were determined by simulation
experiments performed with a model and a numerical simulation program,
developed in simulation environment MEDSIMFP10. Simulation experiments
have confirmed low energy consumption air and good measurement accuracy.
Keywords: pneumatic transducer, force, drying, energy consumption, lowcost
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
Erol Murad et al.
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.
Fig. 1 Functional diagram of pneumatic force transducer.
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
where: Gem is the weight of mobile equipment, which is constant.

For measuring a force with some variation, included in the measurement,
the supply with compressed gas is sampled with an u1 order form, short pulse,
which provides increased pressure pmas to Fmas balance; then the rigid centre,
membrane and nozzle amount of the ball with h(t) and excess flow gas Dev(t) is
discharged outside.
For measuring a force with a change in one respect, specific processes of
drying, the weight of bodies during drying decreases continuously, it is
necessary to supply compressed gas only at the beginning of drying burden.
During drying Fmas decreases continuously, pmas decreases continuously, and
additional gas is discharged outside through the space between the gasket and
valve, to maintain balance as it shows the Eq. (1).
170
Erol Murad et al.
This method of measurement consumes a very small amount of compressed

gas, which means little air power.
Pneumatic output signal of the enclosure pmas is converted into unified
electric signal using a converter P / U.
As the outdoor temperature is constant and the variation of force Fmas(t) is
slow, can be considered, in terms of thermodynamics, the measurement is an
isothermal process at constant volume. Variation of the air mass in the
enclosure depends on pressure variation pmas(t) produced by the variation of
force Fmas(t):
Vm + Vad =
(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)
pmas (t )(Vm + Vad ) = ma (t ) RaTa ,
hence the differential equation of the dynamic behaviour of load cell:

(4)
(Vm + Vad ) dpmas (t ) = RaTa dma (t ) ,

dt
dt
and
dma (t ) Vm + Vad
=
S RT
dt
ef a a
(5)
dFmas
dt .
When the force produced by the membrane exceeds the loading:

(6)
(p
mas
(t ) S ef Fmas (t ) Gem ) > 0 ,
is entering a phase characterized by acceleration a(t) that makes lifting the

nozzle from the ball to h(t):
(7)
a (t ) = ( p mas (t ) S ef Fmas (t ) Gem )/ M red (t ) .
Mred low mass value (t) at the transducer rod is:
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
The kev rate of contraction of the jet exhaust is calculated from a

relationship F(Re, Rh, Ra, Ta, pmas, patm) [3].
If Fmas(t) decreases continuously, so in drying processes, change dma(t)/dt
will be negative; over time, the gas is discharged outside for reaching a value
pmas(t) with which is balanced Fmas(t).
Under these conditions, mass balance for air in the enclosure is:
(10)
dma (t )
= Dmi (t ) Dmev (t ) .
dt
It was made a model and a simulation program of this type of transducer

operation in MEDSIMFP.10 simulation environment. Figure 2 presents
transducer operation for a sampling period starting sequence and in Figs 3 and 4
are shown different working arrangements, with a simulation of 300 s.
3. Measurement Algorithm
For reducing energy consumption and air power, it was designed an
algorithm to drive transducer operation under sampled with a period Tes (10
... 100) s.
Reading the transducer voltage output yg is made in PLC with a frequency
fy 100 Hz.
At the beginning of the sample, it is given a prompt ui = 1 which opens the
DP and the through RP entry pass flow Dmi (t), that results in increased air mass
ma (t) and default and pmas (t), see Figure 2.
At the beginning of the sample, it gives a command ui = 1 which opens the
DP and through the RP entry pass flow Dmi(t), which results in increased air
mass ma(t) and hence the pmas (t), see Figure 2.
172
Erol Murad et al.
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.
Fig. 2 Measurement scheme starting sequence.
Fig. 3 Experiment simulation pneumatic transducer Fmas (t) = const.
173
Fig. 4 Experiment simulation pneumatic transducer dFmas(t) / dt <0.
4.
Results of Simulation Experiments
It was modelled and simulated a pneumatic transducer with the following

constructive and functional characteristics: Def = 30 mm, db= 4 mm, dd = 2 mm,
Fmas (t) = 40 N const., Mem = 0.04kg, pal = 1,4 bar.
In figure 2 is given the power system measurement sequence for
determining dynamic performance and efficiency measurement algorithm.
From simulation results that the maximum acceleration of lifting the nozzle
is only 97.5 mm/s2 and maximum lifting Hmax = 0.272 mm. It is noted that
during movement of the nozzle is within 4 s, after which the pressure in the
enclosure remains constant and proportional to the force measured Fmas (t.).
Replacing the nozzle on the ball is amortized at a rate practically zero, which
does not produce shocks in operation and deformations in the contact nozzleball, ensuring high durability of a transducer.
DP distributor is open only 2 seconds, and the volum consumed in the
sequence of measurement only 2.257 Ncm3 air.
Figure 3 shows a sequence of measurement for 300 seconds, with a
sampling period of Tes = 60 s. It is found that air consumption is 3.672 Ncm3,
resulting in an average consumption over a sample of only 0.744 Ncm3, or
consumption of 44.64 Ncm3 per hour.
174
Erol Murad et al.
Figure 4 presents a simulated experiment where the measured force

decreases continuously and slowly. Transducer works well with an air
consumption similar to the previous experiment.
If it is considered that the opening between two orders of distributor DP,
pressure may decrease too much due leaks and wear, hourly air consumption
would be maxim 135.42 Ncm3 / h, which is a small volume. For example in
continuous operation for 24 hours would consume 3.25 Ndm3 air, which would
require a reservoir of 2 dm3 uploaded once a day to 3 bar.
5. Conclusions
1. We designed a pneumatic force transducer for particular process
consisting in a slow change of the measured force, typical for convective
drying.
2. Transducer operates with very low air energy consumption, is simple,
affordable, sustainable, low-gauge and weight.
3. They were applied mechatronics and automation principles at low cost
for transferring as many functions to PLC management of the technological
process.
4. For exhaust valve nozzle-ball version has been used to ensure normal
operation even in difficult assembly and operation, ensuring a good seal chosen
option and a relatively large exhaust section to small movements.
5. On the dynamic, by controlling the very low input flow , in a chamber
measuring small volume, for reaching a maximum speed of less than 100 mm/s2
for the mobile part, and a maximum displacement of 0,3 mm. These values
show that there are no shocks during the operation, and sustainability of nozzleball system is very high.
6. Simulation experiments show that a transducer with head scale of 50 N
can operate continuously 24 hours powered by a 2-liter tank loaded to 3 bar
once a day.
7. The results obtained in simulated experiments represent tests of the
experimental model, which pursue to prove both design and functional
optimized variants functional, through change values for pair ball diameter and
nozzle diameter.
POLITEHNICA University of Bucharest,

Faculty of Biotechnical Systems
Bucharest, Romania
e-mail: erolmurad@yahoo.com
2
INOE 2000-IHP
Bucharest, Romania
e-mail: dumitrescu.ihp@fluidas.ro
175
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
A v r a m M., Acionri hidraulice i pneumatice, Ed. Universitar, Bucureti,

2005.
F e i d t M. L., Termodinamica i optimizarea energetic a sistemelor i
proceselor, Ed. BREN, Bucureti, 2001.
D i m i t r i e v V. N., G r a d e k i i B.G., Osnov pnevmoavtomatiki,
Mainostroienie, Moskva, 1973.
M u r a d E., MEDSIMFP10, Mediu software simulare sisteme, Free Pascal-V2.1.4,
U.P.B., 2008.
M u r a d E., Msurarea parametrilor procesului de uscare a produselor ceramice
cu traductoare pneumatice neconvenionale, HERVEX 2007, Climneti 14-16
noiembrie 2007.
M u r a d E., C h e r c h e T., Traductoare pneumatice neconvenionale cu
consum redus de energie pentru msurarea forelor din instalaii agricole i n
industria alimentar, HERVEX 2008, Climneti 15-17 noiembrie 2008.
M u r a d E., H a r a g a G., B u l e a r c M., B d i l e a n u M., Convection
dryers under cogeneration, Conferina Internaional, ENERGIE - MEDIU
CIEM 2009, UPB, Bucureti 12-14 noiembrie 2009.
M u r a d E., P r e d e s c u C., R i z o i u G., S i m a C., H a r a g a G., Instalaii de
uscare mobile i ecologice destinate zonelor montane, HERVEX 2009,
Climneti 18-20 noiembrie 2009.
M u r a d E., M a i c a n E., M a r i n A., Du m i t r e s c u C., Dinamica
traductoarelor pneumatice cu consum redus de energie pentru msurarea
forelor n procese lente, HERVEX 2008, Climneti 18-20 noiembrie 2009.
M u r a d E., C h e re c h e s T., Optimizarea consumului de energie a sistemului de
conducere automat a proceselor de uscare convectiv, HERVEX 2008,
Climneti 18-20 noiembrie 2009.
R a d c e n c o V., A l e x a n d r e s c u N., I o n e s c u E., Calculul i proiectarea
elementelor i schemelor pneumatice, Editura Tehnic, Bucureti, 1985.
* * * Catalog SMC, 2009.
* * * Catalog FESTO, 2009.
* * * Catalog Motorola 2008.
SISTEM DE MSURARE PNEUMATIC A MASEI DE AP

EXTRAS N PROCESELE DE USCARE CONVECTIV
(Rezumat)
Pstrarea produselor agricole vegetale, fructe i legume, se poate face n condiii
foarte bune prin uscarea lor; aceast soluie, beneficiind de un condus optimal,
realizeaz produse cu potenial nutritiv ridicat, cu consumuri energetice specifice
reduse. n actualul context economic, derularea acestor procese n locaii
176
Erol Murad et al.
descentralizate, de dimensiuni mici i medii, necesit instalaii uor de deplasat, fiabile

i cu un pre redus. Eficiena economic a acestor instalaii depinde de utilizarea
conducerii optimale a proceselor de uscare care necesit msurarea, n condiiile termice
din camerele de uscare, a masei de ap extrase.
S-a dezvoltat o nou schem de msurare on-line a greutii unei casete ncrcat cu
material de uscat care utilizeaz un traductor pneumatic de for care lucreaz n regim
de msurare eantionat. Soluia prezentat se remarc prin consumuri foarte mici de
energie i este caracteristic pentru automatizarea cu cost redus. Sistemul de msurare
este cuplat la PLC-ul destinat conducerii automate a instalaiei de uscare.
Comportarea dinamic i consumurile energetice s-au determinat prin experimente
de simulare realizate cu un model i un program de simulare numeric, dezvoltat n
mediul de simulare MEDSIMFP10. Experimentele de simulare au confirmat consumuri
reduse de energie pneumatic (un rezervor de 2 l ncrcat la 3 bar odat pe zi), precum
i o precizie bun de msurare.

Publicat de
Secia
CONSTRUCII DE MAINI
TECHNOLOGIES FOR MONITORING STRUCTURAL

DAMAGES ARISING IN THE FUNCTIONING
OF WIND TURBINES
BY
MIHAI FLORIN MNESCU and VALERIU PANAITESCU

Abstract. The need for sustainable development requires the use of renewable
energy, primarily to maintain the existing energy (fossil fuels) to a suitable level,
but also for environmental conservation. Accordingly, renewable energy sources
began to be intensively studied and developed, mainly in the last ten years. In
Romania, the first wind groups were put into service in 2005 (Ploiesti Industrial
Complex - a wind turbine Vestas V52 - 660 kW, relocated), now the installed power
being around 14 MW. For the purposes of harnessing wind energy for electricity
production, because at heights exceeding 100 m, the wind is not influenced by
existing deficiencies ground (trees, buildings, hills, etc..) Wind groups have become
increasingly high and access to review and repair is difficult. As a result of this it is
necessary to monitor the time of damage that may occur during operation of a wind
farm to reduce the time the wind group is stopped and production losses are as small
as possible. To ensure proper function safely it is necessary to implement a
monitoring system for structural resistance (MSR) of wind turbine. This system and
some of the damage assessment methods will be presented in this paper. The MSR
has two branches: a network of sensors that collect information and an algorithm /
software to interpret data from the sensors. Some of the technologies that can be
used, methods of ultrasonic, thermal scanning, x-rays, etc. will be presented in this
paper. Likewise, the most frequent types of structural damage that can occur in a
wind group will also be discussed in this paper.
Key words: wind turbine, monitoring structure resistance, sensors and damage
detection.
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
Mihai Florin Mnescu and Valeriu Panaitescu
emphasized the efficiency of plants used as feedstock in electricity generation,

renewable sources. Unlike other renewable energy sources, wind energy is
limited because of its technological maturity, good infrastructure and relative
price competitiveness. To have an effective higher cost-benefit, wind generators
groups have become increasingly high.
Some of the problems arising from the operation of wind turbines are listed
below:
a) the areas where wind farms are developed wind regions and which are
usually difficult to reach, such as high mountain areas or areas of Dobrogea, due
to existing infrastructure;
b) the height of pillars (about 100 m), which makes revisions and access
for repairs difficult;
c) there have been accidents that led to the collapse of the entire wind
structure and consequently to enormous losses;
d) the tendency to have a high cost-benefit led to large structures;
e) in Romania there isnt a mature experience in the production of
electricity using wind power, the first wind turbine was put into service in 2005,
with a pillar height of 54 m. The wind farms that are now developed are usually
100 m high and, in case of damage, this fact can lead to serious accidents, since
action plans may not be effective.
Given the above, it becomes necessary to monitor the whole resistance
structure of wind turbines using a network of sensors.
The monitoring systems are safe and low-cost, and their implementation can
prevent energy losses during revisions or damage. These systems can efficiently
diagnose which parts of the wind turbine must be replaced and they can also
work to reduce duration of a scheduled review. Furthermore, these systems can
predict the life cycle of the plant whose working time averages10-30 years.
According to F a r r a r and S o h n [1] a monitoring system is defined as a
process that detects faults which occur in high buildings. A fault is defined here
as a alteration of the materials and / or of the geometric properties of these
systems, including alterations in bordering conditions and system connectivity,
which adversely affect system performance. There can be many causes of
structural damage such as moisture absorption, fatigue, wind gusts [2], thermal
stress, corrosion, fire and even lightning [3].
In general, the successful development of methods of MSR depends on two
key factors: the technology employed for detection and signal analysis and the
algorithm for interpretation [4]. MSR components consist of a data acquisition
system, data filtering, pattern recognition and decision making. Each of these
components is equally important in determining the safety status of a structure
[5].
The monitoring process involves well-timed observation, using data from a
series of sensors and statistical analysis in determining the status of the
structure. In the case of long-term process MSR, the sensors provide regularly
179
updated information on the ability of the structure to perform its intended

function, depending on the inevitable aging process and the degradation caused
by the operation conditions. Ensuing extreme events such as earthquakes, MSR
is used to obtain nearly real-time, reliable information on the integrity of the
structure [1].
The same as wind turbines, the MSR system can provide data on the state of
integrity of the assembly. The damage caused by wind turbines which can be
detected by MSR is listed below, depending on the functional characteristics of
the system. Early warning, with the following advantages and benefits:
a) Prevention of breakdown; better planning of the revisions; reducing the
cost of repairs; minimization of the reparation time.
Problem identification, with the following advantages and benefits:
a) Servicing at the appropriate time; reduction of the number of replaced
parts; solving problems while still in warranty; reduction of the costs; operation
at normal parameters.
Continuous monitoring, with the following advantages and benefits:
a) Constant update on the working parameters; security, less stress.
The benefits of having a detection system faults are: [6], [13].
a) The prevention of premature breakdown: prevention of disasters, failure
and secondary defects.
b) Lower maintenance costs.
c) Remote monitoring and remote diagnosis: turbines are usually built within
a long distance from the premises of the beneficiary.
d) Improvement of the capacity to warn against impending failure; the
duration of the repairing process can be reduced in the windy season and
consequently the production will not be affected.
e) Support for the further development of a turbine: the data which has been
gathered can be used to improve the models of the future turbine generation.
With a reliable MSR, one can successfully plan an efficient maintenance
and repair strategy of the wind turbines, especially offshore. Maintenance and
repair can be performed on demand, when the weather conditions are
favourable. Similarly, the mobilization to cover the costs of personnel, materials
and equipment can be optimized [7].
An ideal monitoring system consists of two main components: a network of
sensors for data collection and an algorithm for data analysis, which is used to
interpret the measurements and determine the physical condition of the
structures.
These methods, which are either already applicable or promise application
to the wind turbine systems in the near future, are presented below.
180
2. Damage Detection Technologies

2.1. Non-destructive Technologies for Monitoring
2.1.1. Thermal scanning technology. This technology is used to detect
anomalies, based on the differences in the temperature of the inspected areas,
such as the blades of the wind turbine, with the help of infrared sensors or heat
detecting cameras [6]. The differences in temperature are correlated with the
diffusion of heat and, as a result, they indicate abnormalities or damage.
Specialized equipment, called thermovision or thermographic camera and
similar in size and appearance to the very familiar video camera of everyday
life, is used to obtain thermal images. In this way, objects are viewed in terms
of the infrared radiation (IR) emitted by them and not in that of the visible
radiation, which can be effortlessly detected with the naked eye.
The thermographic method is extremely useful in practice because it
permits the achievement of the so-called "predictive maintenance", a term
which is applicable to all industries. The first sign of a defect or an operating
problem is often given by the increased heat in that particular area, therefore an
increase in the emission of infrared radiation. In other cases, the unjustified
lower temperatures of some areas or of some parts of the device may be a sign
of negative phenomena occurring at their level. After having been generated,
the termograma is digitally processed in order to locate the points where heat
stress is present and the defects. It should also be noted that the assessment of
the termograme obtained by scanning the device, system or facility during
proper functioning, can provide extremely valuable information on the normal
thermal map, which will serve as reference for future evaluation and will also
help with the remediation of any potential failure.
Fig. 1 Image of wind turbine produced by thermography.
Additionally, the thermographic method is a non-invasive technique for

measuring the infrared radiation emitted by the object under examination and
generates, after a very short thermal analysis performed in real time, a thermal
paper which may be sometimes of vital importance to the wind turbines. This
method enables the precise measurement of surface temperature; the high
resolution is a valuable advantage in detecting differences in the temperature of
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).
Fig. 2Components blade.
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.
Fig. 3A representations of some types damage in the blades.
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.
Fig. 4Examples of damaged turbines at 100%.
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.
Received April 15, 2010
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
14. L a d i n g L., M c G u g a n M., S e n d r u p P., R h e i n l a n d e r J., R u s b o r g

J., Fundamentals for remote structural health monitoring of wind turbine
bladesa preproject ANNEX B Ris-R-1341(EN) Report Ris National
Laboratory, Denmark, 2002.
TEHNOLOGII FOLOSITE PENTRU MONITORIZAREA

DEFECIUNILOR STRUCTURALE APRUTE N
FUNCIONAREA GRUPURILOR EOLIENE
(Rezumat)
Necesitatea dezvoltrii durabile impune utilizarea resurselor regenerabile de energie, n
principal pentru meninerea celor existente (combustibili fosili) la un nivel convenabil,

dar i pentru conservarea mediului nconjurtor. n aceste condiii sursele de energie
regenerabila au nceput s fie intens studiate i dezvoltate cu precdere n ultimii zece
ani. n Romnia primele grupuri eoliene au fost puse n funciune n anul 2005
(Complexul industrial Ploieti o centrala Vestas V52 660 kW, relocat), n prezent
existnd o putere instalata de circa 14 MW.
n sensul valorificrii energiei eoliene pentru producerea de energie electric,
datorit faptului c la nlimi de peste 100 m, vntul nu mai este influenat de
neregularitile existente la sol (arbori, cldiri, coline, etc.) grupurile eoliene au devenit
tot mai nalte, iar accesul pentru revizii i reparaii este dificil. Ca urmare a acestui fapt
este necesar o monitorizare din timp a avariilor ce pot s apar n timpul funcionrii
unui parc eolian, pentru a reduce timpul n care grupul eolian este oprit, iar pierderile de
producie s fie ct mai mici. Pentru a se asigura o bun funcionare n condiii de
siguran este necesar a se implementa un sistem de monitorizare a structurii de
rezisten (MSR) a ansamblului eolian. Acest sistem i cateva metode de evaluare a
avariilor vor fi prezentate n acest articol. MSR are n componenn dou ramuri: o reea
de senzori ce colecteaz informaiile i un algoritm/software pentru interpretarea datelor
provenite de la senzori. Cateva dintre tehnologiile ce pot fi folosite, metode cu
ultrasunete, scanare termica, raze X, etc. vor fi prezentate n articol. Tot n acest articol
vor fi prezentate cele mai frecvente avarii de structur ce pot aprea la un grup eolian.
Spre deosebire de alte surse regenerabile de energie, energia eolian are o limit
datorit maturitii sale tehnologice, infrastructur bun i o competitivitate relativ de
pre. Pentru a avea o eficien cost-beneficiu mai ridicat, grupurile generatoare eoliene
au devenit tot mai nalte.
Monitorizarea structurii de rezisten (MSR) este esenial n raport cu cellalte
sisteme de monitorizare pe care le are n componen un grup eolian, deoarece
prbuirea ntregii structurii are urmri grave asupra siguranei populaiei, a mediului i
din punct de vedere financiar al investitorului.
Sistemele de monitorizare sunt sigure, au un cost redus, iar prin implementarea lor
se pot reduce pierderile de energie din timpul reviziilor sau avariilor. Prin aceste sisteme
se pot diagnostica ce piese ale grupului generator trebuie nlocuite i reducerea timpului
din cadrul unei revizii programate. Totodat aceste sisteme pot face o predicie asupra
ciclului de via al centralei al crui timp de lucru proiectat este de 10-30 ani.
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.

Publicat de
Secia
CONSTRUCII DE MAINI
PUMPING STATION EXONERATION FOR WATER SUPPLY

BY
AURORA ALEXANDRESCU, ADINA SIMONA ALEXANDRESCU

and ADRIAN CONSTANTIN ALEXANDRESCU
Abstract. Profitability of water distribution activity depends largely on the
relationships between operational capability and service costs, related to
suppliers performance, volume of distributed water and effective operating costs.
The main variables that influence the total selling price are required investment
value, specific consumption of electrical energy for pumping power, unit price of
the electrical energy and total volume of monthly consumed water billed. The
selection of rehabilitation and modernization measures must rely on market
studies results that appropriately establish the quantities of water that may be
distributed and billed. Present and future water requirements will be determined
based on the analysis of actual operation data and on estimation of future trends in
water consumption on national and international levels.
Key words: adduction, conductivity, chlorine, pipe network, pumping station,
tank.
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
Aurora Alexandrescu et al.
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 =
Ntm .W K N .Q.H t .W K N .Q.L.J .W

=
=
.
3600
3600. p
3600. p
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/
It is calculated annual medium costs associated to the power consumption

and the annual average cost associated to the investments:
(4)
C AE = pE .(1 + krE ).E , A = AI c + AP = ac .I c + a p .I p
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 .
It is calculated the annual medium total costs:
(6)
Q
C A = A + C AI = ac .L.a. K j .
p .(1 + krE ) .K N .Q.L.J .W

.
+ E
3600. p
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
pE .(1 + krE ) .K N .Q.L.J

3600. p
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
The minimum marginal cost has the following configuration:
(8)
CM
Q
= 0 A. .
J
1+
J
+ B. p .Q
.J
p 1
+ C.Q = 0
3. Results and Discussion
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
The all measurement system of drinkable waters conductivity from pipes

network supplied of CUG Iasi pumping station includes the following elements:
temperature integrated sensors inductive and conductive sensors. The turbidity
and the solid suspensions quantity Tb from drinkable water system are
measured with the Liquisys M CUM 223/253 type system. The turbidity sensors
are Turbimax W CUS 31 type.
Fig. 1 The emplacement points of 17 automatic pursuit systems automate of the

supply network with drinkable water from Iasi city.
The conductivity, chlorine and turbidity concentration variations are

measured in same moments of 1.06.2009 day, (Fig. 2) in CUG pumping station.
The automized regulation system of chlorine and chlorine dioxide quantity from
CUG pumping station is Liquisys M CCM 223/253 type. The CSC 140/141
type sensors are used to visualize the automized measurement and processing of
the Cl2, HOCl, OCl- concentrations, depending on the drinkable water pH. The
measurement system of ammonium quantity from drinkable water is Stamolys
CA 71 AM type. The ammonium concentration variation in same time in
1.06.2009 day is presented in the CUG pumping station E1, in Fig. 3. The
digital transmission system of the pressure p and the debit Q from CUG
pumping station at the S. C. APAVITAL S. A. Iasi dispatcher is Cerebar M
HART, respectively Priline Promag 50 types.
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
1. It comes out that chlorine and ammonium concentrations grow up in the

maximum consumption period of network: Cl = (1,1 2,89) [mg/l] and Am =
(42,55 43,31) [mg/l]. In the same time, pressure and debit grow up because to
the increase of the active consumers number in the network: p = (3,01 4,09)
[bar] and Q = (0 1650,74) [m3/h].
2. The water turbidity grows up at (0,49 0,59) FNU values on starting
and stop period for each pumps. In short time from the starting and stop
moment, the turbidity declines at (0,26 0,30) FNU values. The waters
conductivity is constant (0,53 0,54) [mS/cm].
3. The flow and load variation from hydraulic system influence the water
quality and the substances quantity imperative for the drinkable water treatment:
in the analyse period 1.05.2009, 00:00 hour and 15.05.2009, 23:55 hour,
maximum chlorine is Clmax = 3,94 [mg/l] in 7.05.2009 day, 3:45 hour; at same
moment the hydraulics system has the following parameters: Tb = 0,42 FNU
turbidity, Q = 0,121 [m3/s] flow, H = 69,34 [m] load, Cv = 0,51 [mS/cm]
conductivity. The minimum chlorine is Clmin = 1,1 [mg/l] in 3,05.2009 day, 3:35
hour; the parameters of the CUG pumping station at same moment are: 3:35
turbidity, Q = 0 [m3/s] flow, H = 64,344 [m] load, Cv = 0,52 [mS/cm]
conductivity.
Acknowledgements. This work has been supported by the National Centre of
Management Programmers, Romania, under financial contract No. 21-041/2007.
Gh. AsachiTechnical University,

Department of Fluid Mechanics, Machines and Hydraulic Acting,
Iasi, Romania,
e-mail: auralexis@yahoo.com
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
3. A l e x a n d r e s c u A., A l e x a n d r e s c u S. A., A l e x a n d r e s c u C. A.,

Contributions concerning the power optimization of the pumping stations,
ASME Fluids Engineering Division Summer Conference, ISI Thomson
Proceedings of 11th International Symposium on Advances in Numerical
Modelling of Aerodynamics and Hydrodynamics in Turbo machinery, August
10-14, FEDSM2008-55007, Jacksonville, Florida, USA, 2008.
4. A l e x a n d r e s c u A., B e r t e a A., Economic efficiency of the investment for
pumping stations exoneration, ISI Thomson Proceedings in The 6th International
Conference Management of Technological Changes, Centre for Continuing
Education and Training CETEX, MTC, Alexandropoulos, Greece, Vol. I,
2009, pp. 437-441.
REABILITAREA STATIEI DE POMPARE PENTRU ALIMENTARE CU APA

(Rezumat)
Sistemul de msurare i de control a calitii apei potabile, a parametrilor
funcionali ai reelelor de conducte trebuie complet automatizat n ntregul municipiu
Iai i generalizat la nivelul ntregii ri. Dei investiia n sistemele moderne de
automatizare, control i msurare din reelele de alimentri cu ap potabil din marile
orae este mare, aceasta poate fi amortizat repede, prezentnd multiple avantaje:
- Posibilitatea optimizrii funcionarii ansamblului staie de pompare
- Obinerea unei caiti foarte bune a apei potabile, dozarea fiind efectuat automat,
n funcie de calitatea apei ce curge prin reelele de conducte.
- Obinerea unei concentraii optime a bulelor de gaze coninut n apa potabil de
(3 10) % pentru evitarea fenomenului de vortex n instalaii.
Turbiditatea apei potabile crete brusc la valori cuprinse ntre (0,49 0,59) FNU pe
toat perioada n care se pornesc i se opresc fiecare din pompe. n scurt timp de la
momentul nchiderii, respectiv a deschiderii pompelor, turbiditatea scade la valori
cuprinse ntre (0,26 0,30) FNU. Conductivitatea apei rmne aproximativ constant,
variind n intervalul (0,53 0,54) [mS/cm]. Clorinarea variaz n intervalul (1,1 2,89)
mg/l i crete odat cu creterea turbiditii apei potabile.
Variatia debitului si a sarcinii din sistemul hidraulic studiat influenteaza puternic
calitatea apei si implicit cantitatea de substante necesare tratarii apei pentru potabilizare:
in perioada de analizat, cuprins n intervalul 1.05.2009, ora 00:00 i 15.05.2009 ora
23:55, clorinarea maxim Clmax = 3,94 [mg/l] s-a nregistrat n ziua de 7.05.2009, ora
3:45; n acelai moment s-au nregistrat i urmtorii parametri: turbiditatea Tb = 0,42
FNU; debitul Q = 0,121 [m3/s]; sarcina H = 69,34 [m]; conductivitatea Cv = 0,51
[mS/cm]. n aceeai perioad, clorinarea minim Clmin = 1,1 [mg/l] a fost nregistrat n
ziua de 3,05.2009, ora 3:35. Ceilali parametri ai staiei de pompare CUG n acelai
moment au fost: Tb = 0,43 FNU; debitul Q = 0 [m3/s]; sarcina H = 64,344 [m];
conductivitatea Cv = 0,52 [mS/cm].

Publicat de
Secia
CONSTRUCII DE MAINI
CONTRIBUTIONS REGARDING DURABILITY EVALUATION

OF HORIZONTAL AXIAL HYDRAULIC TURBINES SHAFTS
BY
ILARE BORDEASU1, MIRCEA OCTAVIAN POPOVICIU1, DRAGOS

NOVAC2, LIVIU MARSAVINA1, RADU NEGRU1, MIRCEA VODA1,
VICTOR BALASOIU1 and MARIAN BRAN1
Abstract. Using previous researches regarding cracks initiation, the present
paper analyzes the durability of horizontal bulb turbines shaft, especially in the
joining zone between the shaft main body and the runner flange. For
exemplifying, there were used the data of the turbines running in the Power Plants
Iron Gates II and Gogosu. The obtained conclusions can be used to avoid serious
failures of horizontal hydraulic turbines. The abstract of the paper is to be written
here. It contains the main ideas and original contributions and conclusions of the
authors research.
Key words: axial hydraulic turbines, cracks, stress alternation cycles, fracture
mechanics characteristics
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
Ilare Bordeasu et al.
Fig. 1 The 3D shaft model, obtained with the INVENTOR program [initiation].
2. Estimation of Fatigue Crack Propagation for Variable Stresses

with Constant Amplitudes
For the selected axial hydraulic turbine (Power plant Iron Gates II) the shaft
presents the following characteristics [1]:
a) the initial crack occurs in the joining zone between the shaft flange (for
coupling with the hydraulic runner) and the shaft body, fig. 2;
b) in order to obtain the needed data we choose the AISI 1020 steel [6] instead
of the actual used 20C steel [7]; both steels have almost the same mechanical
characteristics but for the AISI 1020 we found all the supplementary
characteristics needed for fatigue tests;
c) the shaft dimensions are: length 7572 mm, maximum diameter 2300 mm,
the main shaft body, between flanges, has an external diameter of 1200 mm and
an internal one of 600 mm;
d) according with the ANSIS program (Fig. 3), till crack initiation, the
minimum number of cycles was 3,0139 108, corresponding to 80,370 running
hours.
a) Deep circumferential cracks
b) Cracks network
Fig.2 Cracks in the joining zone between the runner flange and the main shaft body.
199
Fig. 3 Number of cycles till crack initiation.
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
Fig. 4 Variation of crack propagation speed as a function of stress intensity variation.
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)
where: C and n are constant values depending on the used material;

K = Kmax Kmin - represents the variation of the stress intensity factor; KC
- is the critical value of the stress intensity factor (breaking tenacity); R represents the asymmetry ratio of the stress cycle.
Expressed by Nr, the necessary number of cycles for the extension of the
crack, the lifetime can be obtained by solving the equation for dN.
Integrating both members it results:
(2)
Nf
a cr
Nd
ad
dN = N f N d = N r =
da
.
f (K , R )
With the equation (2) it is possible to calculate the number of cycles Nr

necessary for the extension of the crack from the detectable length ad (for which
corresponds the number of cycles Nd) till the critical length acr (which
corresponds to the Nf number of cycles).
For the lifetime computation (the final number of cycles Nr) it was used the
specialized program AFGROW, developed by H a r t n e r at WRIGHTPATTERSON AIR FORCE BASE [3] in order to estimate the lifetime of some
components for fighter planes. It was taken into consideration a ring section,
having an elliptical crack as can be seen in Fig. 5. For such geometry, the
intensity factor was proposed by Raju and Newman [4] as being:
(3)
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,

1.65
(4)
a
Q = 1 + 1,464
c
for
201
a
1.
c
Fig. 5Computation model for crack propagation.
As was previously mentioned, the mathematical simulation was done with

the program AFGROW, version 4.12.15 from 10.07.2008. The entrance data
were: the external diameter (Do): 1.20 m, the internal diameter (Di): 0.60 m, the
crack depth (A): 0.001 m, crack half length (C): 0.002 m
In conformity with the shaft disposition, for studying the crack propagation
there were taken into consideration two load situations with constant amplitude:
a) a bending loading with a symmetrical alternative cycle having max = - min
= 42 MPa, with an asymmetry coefficient of R = -1.
b) a composed load (bending plus elongation); considering that the static
elongation stress is superposed over the bending stress it result an approximate
pulsating cycle with max = 88,9 MPa, min = - 6,5 MPa , R = - 0,07 [1].
In order to introduce the material constants, characterizing the crack
propagation, the used steel was considered to be equivalent with the AISI 1020
for which the characteristic data are included in the data base of the AFGROW
program. So, the constants for the Walker law [5]:
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)
c) a=150 mm and 2c= 320 mm

N = 159737 hours (fracture, breakdown)
Fig. 6 Propagation of circumferential crack in the hollow shaft.
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)
Politehnica University of Timisoara,

e-mail: ilarica59@gmail.com
2
Hidroelectrica Iron Gates, Dr.Tr. Severin
e-mail: dragos.novac@hidroelectrica.ro
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
4. R a j u I. S., N e w m a n J. C., Stress Intensity Factors Circumferential Surface

Cracks in Pipes and Rods, Proc. of 17th National Symposium on Fracture
Mechanics, Albany, NY, 1984.
5. W a l k e r K., The effect of stress ratio during crack propagation and fatigue for
2024-T3 and 7075-T6, ASTM STP 462, ASTM, 1970.
6. * * * Fatigue Design Handbook, Second Edition, SAE, Warrendale, 1988,
7. * * * 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, Contract No. RU 177/10.10.2008, BC 146/13.10.2008
CONTRIBUII N EVALUAREA DURABILITII ARBORILOR

HIDROAGREGATELOR AXIALE ORIZONTALE
(Rezumat)
n cadrul lucrrii se evalueaz durabilitatea arborilor turbinelor axiale, n zona de
racord a flanei ce-l cupleaz cu rotorul turbinei, plecnd de la rezultatele obinute n
lucrrile anterioare privind iniierea fisurii. Ca model este folosit arborele de turbin
bulb, aflat n exploatare la CHE Porile de Fier II i CHE Gogou, pentru care
dispunem de datele necesare. Concluziile rezultate permit evitarea unei eventuale avarii
periculoase.

Publicat de
Secia
CONSTRUCII DE MAINI
OPTIMAL ROUTES OF PIPELINE SUPPLY USING

THE BELLMAN-KALABA ALGORITHM
BY

ADRIANA MANEA, RODICA BDRU and DANIEL STROI
Abstract. In this paper are presented the application of graphs theory for
determining the optimal route of a pipeline supply being at the great distance of
the target consumer (pipeline network of a city). It applies when the distance from
source to target, because the configuration of the land, there are several variants of
the route passing through some mandatory points. In this way the route has n
sections and on each section the total cost (investment plus operating for one year)
has a certain value. If it can browse the route by more than two then the method
becomes profitable. Implementation of the method is through a special program,
using the Borland Pascal programming and Bellman-Kalaba algorithm.
Mathematical resolving is by the matrix. Numbering the sections with 1 ... n, in
order to obtain the final optimal browsing, it is the range of selected sections. In
actual conditions when more and more sources of drinking water are becoming
more polluted, the feeding is justified to be from remote mountain areas of the
natural springs.
Key words: graph theory, pipeline, optimal route, matrix, Bellman-Kalaba
algorithm, and network.
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
decisions, whose error is inversely proportional to the degree of experience of

the designer.
Modern mathematical disciplines, by operational calculation, put at
specialists disposition a vast apparatus of scientific analysis in determining the
optimal decisions for the design of water supply. In this context, describes a
deterministic mathematical model optimization of the route of the water supply
adduction, based on the theory of graphs.
2. Calculation Algorithm
Modeling this problem is achieved through representation related directed
graph G=(X, U) consisting of the source as the origin, route as required arcs and
( )
points as vertices. For each arc u ij U is associated a number u ij 0 , in
conventional units, depending on the optimization criterion adopted. The route

is the best way to graph, the minimum value which is determined by applying
the algorithm Bellman-Kalaba.
Graph G= (X, U) is attached to a matrix M whose elements mij are:
( )
(1)
u ij the arc value from xi to x j
mij = if vertices xi and x j are not adjacent .
0 for i = j
The optimal route is the best way of graph , with the total value:
(2)
( ) =
(u ij ) min .
u j
If the notes with Vi the minimum value of the road ni , ( i = 0, n )

existing from the tip of xi to the tip of xn :
(3)
( )
Vi = ni ,
( i = 0, n ).
hence:
(4)
Vn = 0 ,
207
then, under the principle of optimality:

(5)
Vi = min V j + mij ,
j i
( i = 0, n 1 ; j = 0, n )
and Vn = 0 .
To solve system (5) shall be iterative, noting Vi k the value of Vi obtained at

iteration k, namely:
Vi0 = min ( i = 0, n 1 );
(6)
Vn0 = 0 .
Calculate:
(7)
Vi1 = min V j0 + mij , ( i = 0, n 1 ; j = 0, n ) and Vn1 = 0 .

j i
and then:
(8)
Vi k = min V jk 1 + mij , ( i = 0, n 1 ; j = 0, n ) and Vnk = 0 .

j i
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 .
Numbers Vi k ( i n ; k = 0, 1,... ) monotone decreasing pattern formed that

reach to the minimum necessary, after a finite number of iterations which not
exceeding n 1 . So algorithm stops when it reaches an iteration k, such that
Vi k = Vi k +1 , ( i = 0, n ), and the minimum between the peaks road x0 and xn is
V0k = V0 k +1 . To identify which roads have minimum values founded, are

derived from (8) that have them at the last iteration we have:
208
(10)
Teodor Milos et al
Vi k = mij + V jk 1 = mij + V jk .
Based on the algorithm described, computer program named BEL_KAL was

designed in Borland PASCAL language.
Were made following nomenclatures: N is the order of the graph, V (I , J ) is
the column vector built at each iteration k, X (I ) is the sequence of minimum
road; VAL - the minimum road graph, M (I , J ) - matrix associated with graph,
whose elements are:
(11)
( )
( )
( )
( )
u ij if arc xi ,x j exist
n
mij = u ij if arc xi ,x j not exist

i,j =1
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
( )
0, otherwise (MATR_EX1.dat file).

As the data output sequence result road peaks of minimum values and value
of the road.
3. Case Study
Apply dynamic programming method to solve the problem of site selection

and capture of a main route to supply water through a sequential optimization
where deterministic, discrete.
It is considered a line of supply for locality L, departing from two locations
of S1 and S2 capture (fig. 1). Possible routes through the required A, B, C, D, E,
forming three sectors.
Putting the problem of determining the route for which the total cost is
minimum is determined for each track investments and prepare partial sequence
graph in Figure 2, where each arc is associated a cost, in conventional units.
They noted with x1 , x2 and x3 the decision variables for each sector. These
variables will not take numeric values, but will be vertices of graph that you are
on the same alignment.
209
Fig. 1 Variants of the adduction route.
It exemplifies the application of Bellman-Kalaba algorithm to determine the

optimal bus route for L locality from the source S1 (fig. 1).
Fig.2 Graph of adduction routes.
Modeling problem is achieved through representation related directed graph

G = (X, U) of order n = 7, consisting of the source point of origin, route as
required arcs and points as vertices. For each arc u ij U is assigned a cost in
conventional units (fig. 2) and matrix M attached to graph G = (X, U) are the
210
Teodor Milos et al
( ) is the value attributed to
elements mi,j defined by relations (1), where u ij

edge u ij U .
(12)
M=
Vi0
Vi1
Vi 2
Vi3
42 53
197 197
73 73
153 153 153
3
4
92 61 82
0 61
61
153 153 153

.
61 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
+ m12 , V30 + m13 ,..., V70 + m17 =
0
1
+ m21 , V30 + m23 ,..., V70
0
1
+ m31 , V20 + m32 ,..., V70
0
1
+ m41 , V20 + m41 ,..., V70
0
1
+ m51 , V20 + m51 ,..., V70
0
1
+ m61 , V20 + m61 ,..., V70
27
37
47
57
67
211
c) Calculate values Vi 2 for i = 1, ..., 7 (j = 1, ..., 7)
(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
+ m21 , V31 + m23
1
7
27
1
1
+ m31 , V21 + m12
1
7
37
1
1
+ m41 , V21 + m42
1
7
47
1
1
+ m51 , V21 + m52
1
7
57
1
1
+ m61 , V21 + m62
1
7
67
=0
d) To calculate the values corresponding to column Vi3 for i = 1, ..., 7 (j = 1,

..., 7), an analogue, finally are obtained (15).
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:
Politehnica University of Timisoara,

Department of Hydraulic Machinery
Timisoara, Romania,
e-mail: teodor.milos@gmail.com
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
TRASEUL OPTIM AL UNEI CONDUCTE DE ADUCIUNE

UTILIZND ALGORITMUL BELLMAN-KALABA
(Rezumat)
Prin amploarea investiiei i a consumului de energie, aduciunile au o pondere
nsemnat n cadrul sistemelor de alimentare cu ap, iar proiectarea raional a acestora
comport mai multe procese de optimizare, n rndul crora un loc important l deine
optimizarea traseului acestora.
n practica actual de proiectare, stabilirea soluiei optime se face, de obicei, prin
studierea analitic a dou sau trei variante selectate din mulimea posibil prin decizii
intuite, a cror eroare este invers proporional cu gradul de experien al proiectantului.
Disciplinele matematice moderne prin calculul operaional pun la ndemna
specialistului un vast aparat de analiz tiinific n stabilirea deciziilor optime pentru
problemele proiectrii sistemelor de alimentare cu ap.
n aceast lucrare se prezint modalitatea de aplicare a teoriei grafurilor din
informatic pentru stabilirea traseului optim a unei conducte de alimentare aflat la
distan mare de obiectivul consumator (reea de conducte a unei localiti).
n acest context, se descrie un model matematic determinist de optimizare a traseului
unei magistrate de aduciune a apei, bazat pe teoria grafurilor. Traseul optim este dat de
drumul de valoare minim n graf, care se determin aplicnd algoritmul BellmanKalaba.
Se aplic n cazul cnd, pe distana de la surs la obiectiv, datorit configuraiei
terenului, exist mai multe variante de traseu care trec obligatoriu prin nite puncte
cheie. n acest fel traseul are n tronsoane, iar pe fiecare tronson costul total (investiie
plus exploatare timp de un an) are o anumit valoare. Dac posibilitile de parcurgere
ale traseului sunt mai mult de dou atunci aplicarea metodei devine rentabil.
Implementarea metodei se face prin intermediul unui program special, utiliznd
mediul de programare Borland Pascal, i algoritmul Bellman-Kalaba. Rezolvarea
propriu-zis este matriceal. Numerotnd tronsoanele cu 1n, n final se obine ordinea
optim de parcurgere ca ir al tronsoanelor selectate.
n condiiile actuale, cnd tot mai multe surse locale de ap potabil sunt poluate,
devine tot mai justificat alimentarea din surse ndeprtate, din zone montane cu izvoare
naturale.

Publicat de
Secia
CONSTRUCII DE MAINI
STRUCTURAL ANALYSIS OF A SLEEVE RIGID

COUPLING WITH CYLINDRICAL PINS
BY
DNU ZAHARIEA and MIHAELA TUDORACHE

Abstract. In this paper the displacements, as well as the von Misses stresses
of a sleeve rigid coupling with cylindrical pins will be analyzed using CATIA
Generative Structural Analysis workbench. The structural analysis procedure will
be presented. The visual representations of the numerical results will be analyzed
at both the assembly level and the part level. The assembly level analysis allows
identifying the most critical section of the entire assembly. The part level analysis
allows observing the results for the driving shaft, the driving pin, the sleeve
coupling, the driven pin and the driven shaft.
Key words: sleeve rigid coupling, cylindrical pins, structural analysis,
CATIA.
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
Dnu Zahariea and Mihaela Tudorache
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
0.065m being thus obtained. The material used for all
components is steel with yield strength 02 = 2.5 108 N/m2.
Fig. 1 Constructive types:

a with pins in the same plane; b with pins in two different planes at 90 .
The geometric characteristics of the sleeve rigid coupling with cylindrical

pins can be determined with the relationship: the outside diameter of the sleeve
coupling D = k D d = 0.11 m (the coefficient k D shall be adopted within
1.51.8, the adopted value is 1.6923); the length of the sleeve coupling
L = k L d = 0.2 m (the coefficient k L shall be adopted within 24, the adopted
value is 3.0769); the diameter of the pins d s = k s d = 0.028 m (the coefficient
k s shall be adopted within 0.250.44, the adopted value is 0.4308) [1], [2].
The sleeve coupling will be checked at the torsional strain with:
(2)
t =
16 DM tc
D4 d 4
).
The pins will be checked at the shearing strain with:
(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
3. The Structural Analysis of the Sleeve Coupling

with Pins in two different Planes at 90
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).
Fig. 2 Analysis procedure steps for sleeve couplings

with cylindrical pins in the same plane:
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.
219
Fig. 3 Analysis procedure steps for sleeve couplings with cylindrical pins
in two different planes at 90 :
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
1. Both constructive solutions comply with the requirements of the

resistance. Thus, the maximum equivalent stress is 2.33 10 7 Pa for the
coupling with pins in the same plane and 2.97 10 7 Pa for the other case. Both
the maximum stresses are in the shearing areas of the driving pin. For the two
shafts, the maximum stress areas are placed near the holes. For the sleeve
coupling the maximum stress areas are placed also near the holes, on the
cylindrical inside surface of the sleeve.
2. The two pins respond differently at the applied loads. Thus, the driving
pin takes a larger effort having a bigger strain. Moreover, the strain profile
along the pins longitudinal axis is different.
3. The mesh configuration parameters have a great influence on both the
hardware resource requirements and the numerical analysis errors.
Gh. Asachi Technical University

Department of Fluid Mechanics, Hydraulic Machines and Drives
Iai, Romnia
email: dzahariea@yahoo.com
email: mihaela_tudorache10@yahoo.com
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.
ANALIZA STRUCTURAL A CUPLAJELOR FIXE DE TIP

MANON CU TIFTURI CILINDRICE
(Rezumat)
n lucrare se prezint etapele analizei structurale i rezultatele unui studiu cu privire
la cuplajele fixe de tip manon cu tifturi cilindrice plasate n acelai plan, respectiv n
plane decalate la 90. Analiza efectuat n programul CATIA permite evidenierea
rapid a unor aspecte particulare care ar putea fi puse n eviden cu dificultate printr-un
calcul clasic de rezistena materialelor: asimetria distribuiei momentului, asimetria
strii de deformare, influena concentratorilor de tensiune, poziia seciunilor critice
pentru fiecare element al cuplajului. Modificarea factorului de scal i vizualizarea
animat a tensiunilor echivalente i a deformaiilor permit interpretri interesante cu
privire la evoluia n timp a strii de solicitare i a tendinei reale de deformaie a
cuplajului fix de tip manon cu tifturi cilindrice.

Publicat de
Secia
CONSTRUCII DE MAINI
STRUCTURAL ANALYSIS OF A
BIMETALLIC STRIP THERMOSTAT
BY
DNU ZAHARIEA and MARIUS STACHIE

Abstract. In this paper the structural analysis of a bimetallic strip will be
analyzed using CATIA Generative Structural Analysis workbench. The structural
analysis procedure will be presented. Six different bimetallic strips have been
considered with titanium as active layer and zinc, aluminum, bronze, brass, copper
and steel as passive layer. The numerical analysis has been performed under the
same global conditions (geometrical characteristics, restraints and thermal load)
using two variable parameters: the width and the length of the beams. For all these
study cases only the thermal load generated by an imposed temperature field has
been considered. Charts are presented for comparative analysis of the influence of
the material, the width and the length of the beams on the von Misses stress and
the maximum deflection.
Key words: thermostat, bimetallic strip, structural analysis, CATIA.
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
Dnu Zahariea and Marius Stachie
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
Fig. 1 Geometric characteristics.
In the paper two comparative analyses will be performed. First one is

studying the influence of the beams width b = {1,5,7.5} mm for the same beams
length L = 50 mm. The second one is studying the influence of the beams
length L = {50,100,150} mm for the same beams width b = 5 mm. The beams
thickness will be the same for all analyzed cases ha = h p = 1 mm.
2. The Structural Analysis Procedure
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
transmission (the bounded surface) the stress/strain condition must be

specified by setting the just property of interaction between the active
and the passive layers (Fastened Connection), Fig. 2e
6. Launching the solver, running the numerical analysis and results
visualization. The maximum deflection will be presented for the
bimetallic strip in Fig. 2f (scale factor 100). The von Misses stress will
be shown in Fig. 2g (scale factor 100) for active layer and in Fig. 2h
(scale factor 100) for passive layer.
Fig. 2 Steps of structural analysis procedure:

applying the loads; e applying the interaction conditions; f maximum deflection; g
von Misses stress for the active layer; h von Misses stress for the passive layer.
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).
Fig. 3 The influence of the width of the beams:

a maximum deflection; b von Misses stress for active layer;
c von Misses stress for passive layer.
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)
Fig. 4 The influence of the length of the beams:

a maximum deflection; b von Misses stress for active layer;
c von Misses stress for passive layer.
4. Conclusions
1. The width of the beams has a negligible influence on the maximum

deflection of the bimetallic strip. As far as the status of the stresses, it becomes
apparent that with width increasing, the von Misses stresses grow as well in
both the active and the passive layers. For the width of 7.5 mm, the von Misses
stresses are very close to the admissible values, even overcoming for a small
amount in the case of zinc material. Regardless of the value of the width of the
beams, for all combination of materials, the stress in the active layer is greater
than the passive layer.
2. The length of the beams is definitely affects the values of the maximum
deflection. The von Misses stresses are growing as well, but to a lesser extent
than the growth of the maximum deflection. And in this case, the von Misses
stress in the active layer is towards the passive layer.
226
3. The best combination of materials in terms of deflection is titan for

passive layer and zinc for the active layer. Due to the smaller mechanical
resistance of aluminum and zinc, increase the width of beams over a certain
limit becomes dangerous, especially in the case of accidental temperature
increases.
4. For all cases analyzed the maximum von Misses stress shall be recorded
in the clamped area. Another critical area is the interaction area between the two
beams of the bimetallic strip.
Gh. Asachi Technical University

Department of Fluid Mechanics, Hydraulic Machines and Drives
Iai, Romnia
email: dzahariea@yahoo.com
email: marius_stachie@yahoo.com
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.
ANALIZA STRUCTURAL A ARCURILOR

BIMETALICE LAMELARE
(Rezumat)
n lucrare se prezint analiza structural a unui arc bimetalic lamelar utiliznd
mediul de analiz CATIA Generative Structural Analysis. Sunt analizate ase arcuri
bimetalice lamelare avnd ca material pasiv titanul, iar ca material activ zincul,
aluminiul, bronzul, alama, cuprul i oelul. Analiza numeric a fost efectuat pentru
aceleai caracteristici globale (dimensiuni geometrice, condiii pe contur i sarcini)
folosind doi parametri de control: limea i grosimea arcurilor lamelare. Sunt
prezentate diagrame comparative pentru analiza influenei materialului, a limii i
grosimii arcurilor lamelare asupra deformaiei maxime i a strii de solicitare a arcurilor
bimetalice analizate. Principalele concluzii sunt: sensibilitatea maxim se nregistreaz
pentru combinaia de materiale titan-zinc; zonele puternic solicitate sunt ncastrarea i
interfaa dintre cele dou lamele; tensiunile echivalente mai mari se nregistreaz n
stratul activ; limea lamelelor influeneaz n principal starea de solicitare, n timp ce
lungimea acestora influeneaz cu precdere valorile sgeii maxime a arcului bimetalic
lamelar.

Publicat de
Secia
CONSTRUCII DE MAINI
KNOWLEDGE MANAGEMENT FOR THE CONFIGURATION

IN EARLY PHASES OF COMPLEX CUSTOM PRODUCTS
BY
IRNE ALEXANDRESCU, HANS-JOACHIM FRANKE

and THOMAS VIETOR
Abstract. Nowadays, large and saturated markets in developed countries enforce
the production of a broad palette of customized products, which satisfy the customers
individual needs. In addition, interesting emerging markets together with the constraints
of the current economic crisis lead to further differentiation in terms of product
functionality and allowable costs. Furthermore, global markets also imply many different
standards and norms, lately more strict due to the worldwide environmental concerns. In
order to be effective in producing custom products, companies in the field of mechanical
engineering integrate in their product lifecycle technologies the newest parametric
CAD/CAM systems, as well as different software for the sales and design departments,
such as configuration or calculation tools. However, producing high quality products with
relatively short delivery times requires that a common knowledge base exists between the
enterprise departments and clear processes and interfaces are defined, integrating all the
used systems. This is commonly not the case and therefore unnecessary iterations are
done in order to produce a technically correct offering based on the specific customer
request. This paper discusses the need and some ideas for a better methodology to faster
and better design complex custom products in the field of mechanical engineering. It
explains the need for a structured knowledge base prior to the beginning of the new
custom product design process. This knowledge base offers the necessary data for the
early design phases and connects the sales and design departments by ensuring a correct
and transparent information transfer between the customer requirements, the configuration
software, the used parametric CAD tools and the generated customer specific offers. The
result is a faster, optimized and technically correct basic configuration of custom products
in early design phases.
Key words: complex custom products, early design phases, product configuration,
knowledge management.
1. Introduction
The transfer from industrial to information and knowledge societies
radically transformed the markets and increased customer expectations.
228
Irne Alexandrescu et al.
Globalization and internationalization differentiated the spectrum of offered

goods and increased the need for custom solutions. There are many different
standards and norms coming from saturated markets or interesting emerging
ones, which have to be satisfied with consideration to the massive constraints
additionally imposed by the economic crisis and the environmental concerns.
Product specific knowledge management nowadays is one of the main
factors for the sustainable competitive advantage of a manufacturing company.
It is an important differentiator for the unique core competences of the
company, as well as the essence for developing successful, knowledge based,
market oriented products and services.
In order to achieve customer satisfaction, companies in the field of
mechanical engineering producing custom solutions have to master their
knowledge and generate fast, optimized and technically correct offers for their
products. This paper describes how this can be achieved by using virtualization
and automatization techniques on a methodically structured product knowledge
base.
2. Scattered Knowledge Base in Early Design Phases

In order to deal with the increasing requirements and the competitive
markets, companies use for the generation of the offering of complex custom
products several software tools, such as: Computer Aided Design and
Manufacturing (CAD/CAM) Software, Product Data/Lifecycle Management
(PDM/PLM) solutions, Enterprise Resource Planning (ERP) systems,
configuration tools and calculation programs. Common applications are in the
aircraft [1] or automotive industry [2].
However, through the variety of software, the complexity of the internal
processes is increased. A good overview to the state of the art is shown in the
CIRP Proceedings [3].
Unfortunately, in praxis, the product knowledge base is scattered through
these systems and normally in each development phase the product knowledge
is documented in separate specific systems. The existing PLM/PDM systems
are often not properly used and there is seldom a platform available to
interconnect or translate the different pieces of information, for a global review
or decision taking.
Fig. 1 shows a sequence diagram for the standard processes that take place
in the early development phases for the generation of a customized offer; the
involved departments and their individual software are also displayed.
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
Part Design (e.g. Hydraulic)

(CAD, FEM, calculation, PDM,...)
request
response
Purchasing
(ERP,...)
clarification
schedule
delivery
date
Production
(CAM, NX-Daten,...)
Legend:
Department
(used software)
timeline
document
action
Fig. 1 Sequence diagram of the standard processes

for the generation of a customized offer.
Currently the customer describes to the sales department the requirements

for the desired product. The sales department interprets the customer
requirements and communicates them further to the design department. There is
normally no sufficient common platform for communication and therefore
unnecessary iteration cycles are done. This can be improved through direct
interfaces between the customer and the two departments, by using a common
configuration tool, which generates department specific product information
and documentation. This solution will be discussed further in the next section.
Another common problem that occurs in the processes described in Fig. 1 is
that generally the project information and awareness remain only in a small
group of specialists. They are not reused for later following projects and are
unavailable for new employees. The company specific knowledge base would
help in solving this problem, by documenting existent intellectual property of
the employees into the pre-structured product specific knowledge.
A good example of a holistic knowledge framework to design and
implement a knowledge based organization is presented by B i n n e r in [4]. This
paper concentrates on the early design phases and can be seen as a methodology
for the previous steps taken in knowledge frameworks systems.
It is therefore not necessary to consider the entire companys knowledge
base, but rather only the product specific knowledge used in early design phases
to generate a correct first offer. Such comprehensive and integrated knowledge
management solutions are currently almost absent in practice.
230
3. Suggested Product Configuration Chain

As discussed in [5] by A le x a n d r e s c u and F r a n k e , the solution for a
faster and better generation of offers is a complete and integrated system that
sustains the whole configuration chain, from the client request analysis to the
configured model of the product. The suggested product configuration chain is
shown in Fig. 2.
Fig. 2 Methodology for faster and better design of high quality

complex customized products.
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
Fig. 3 The knowledge based configuration module, as

a main software component of the solution.
4. Knowledge Base Structure

Before defining the company specific knowledge base structure for the
configuration in early phases of a customized product, the following three
concepts need to be defined: data, information and knowledge. The common
interpretation is:
- Data is unprocessed facts and figures without any added interpretation
or analysis.
(e.g. "<dct:modified>")
- Information is data that has been interpreted so that it has meaning for
the user.
Information = data + semantic + context
(e.g.: "The document A was updated.")
- Knowledge is a combination of information, experience and insight that
may benefit the individual or the organization.
Knowledge = information + linking
(e.g.: "B was updated after A, so B contains the latest information.")
The above mentioned three different concepts are presented in Table 1,
according to the description made by B o d e n d o r f in [7]: information is just
a formal step to transforming the data into in the knowledge.
Table 1
The different concepts of data, information and knowledge according to [7]
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:
Fig. 4 Approach for the knowledge base.
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
In the center of the knowledge base is the abstract representation of the

product structure, from a neutral point of view. This representation is company
and product specific and it represents the basis for all the department specific
derived views. As exemplarily presented on the right side of the picture, the
module Mi in the main structure will contain its technical (e.g. for the design
department), as well as its functional (e.g. for the sales department) description.
The module has additional parameters and logic, for other derived views, as for
example for geometry calculation, validation of a module combination or
visualization of dependencies systems.
Further research is currently done to describe how at early stage goal
conflicts can be recognized and solved based on this structure.
5. Benefits of Using the Presented Knowledge Base

The main benefits of the presented solution and its knowledge base are:
- It reflects the company global view on its product structure definition;
- The collection and preservation of knowledge can be carried out without
strong efforts, as the configuration system is used on a daily basis;
- The sales and design departments are interacting better;
- The customized offers can be faster and technically more accurately
generated;
- There is a methodic, systematic and goal oriented approach towards a
common context for the configuration.
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
departments. Comprehensive and integrated knowledge base solutions rarely

exist in praxis.
3. A product configuration chain for an accelerated and technically correct
generation of offers was presented. The main modules of the implementations
are the configuration software and its knowledge base.
4. The product specific knowledge base structure for the configuration in
early phases was discussed.
5. The benefits of the presented knowledge base usage were listed.
Acknowledgements. The authors gratefully thank the German Federal Ministry of
Education and Research (BMBF) for supporting the Project KOMSOLV - 'Fast Offer
and Optimized Design for Complex Products with Conflicting Requirements'.
Received:
Technische Universitt Braunschweig,

Institute of Engineering Design
Braunschweig, Germany
e-mail: i.alexandrescu@tu-bs.de
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
MANAGEMENTUL CUNOATERII N CONFIGURAREA

PRODUSELOR COMPLEXE PERSONALIZATE N FAZELE
INCIPIENTE ALE DEZVOLTRII ACESTORA
(Rezumat)
n prezent, pieele mari i saturate ale rilor dezvoltate orienteaz fabricaia spre o
palet larg de produse personalizate, n vederea satisfacerii nevoilor clienilor
individuali. Pieele emergente interesate, confruntate cu constrngerile crizei economice
actuale, conduc n plus la o difereniere a funcionalitii produselor i a costurilor
admisibile. Pieele globale implic respectarea de standarde i de norme diferite, ns
acestea sunt din ce n ce mai stricte, datorit preocuprilor de mediu la nivel mondial.
Pentru a fi eficiente n fabricaia de produse personalizate, companiile din domeniul
ingineriei mecanice integreaz n tehnologiile ciclului de via al produsului nou sisteme
parametrice de tip CAD/CAM, precum i multiple aplicaii software pentru
departamentele de vnzri i de proiectare.
Cu toate acestea, fabricaia de produse de nalt calitate cu termen de livrare relativ
scurt impune existena unei baze de cunotine comune tuturor departamentelor
ntreprinderii, proceduri clare i interfee bine definite, care s integreze toate sistemele
utilizate. Acest aspect nu se regsete ns ntotdeauna n practic, prin urmare apar
iteraii inutile ocazionate de elaborarea unei oferte corecte din punct de vedere tehnic, n
baza unei cereri specifice a clientului.
n lucrarea noastr abordm problema unei noi metodologii, inovative i rapide, de
design personalizat al produselor complexe din domeniul ingineriei mecanice.
Abordarea noastr explic necesitatea structurrii unei baze de cunotine nc din
fazele incipiente ale procesului de design al unui nou produs. Aceast baz de
cunotine ofer datele necesare pentru fazele incipiente de proiectare i se conecteaz la
departamentele de vnzri i de proiectare prin asigurarea unui corect i transparent
transfer de informaii ntre cerinele clientului, software-ul de configurare, instrumentele
parametrice utilizate n CAD i oferta specific generat de client. Rezultatul cercetrii
const n creterea rapiditii, corectarea tehnic i optimizarea configuraiei de baz a
produselor personalizate nc din fazele incipiente ale dezvoltrii acestora.

Publicat de
Secia
CONSTRUCII DE MAINI
RESEARCHES REGARDING TO INNOVATIVE

ACTIVITY IN TECHNICAL ENVIRONMENT
BY
PETRU DUA and IULIANA LAURA TARANOVSCHI

Abstract. The innovation concept can be concerned in terms of five basic
elements: process, product, human factors, domain and socio-organizational
environment. The study is conceived to stimulate the innovative potential and
combine Blooms Cognitive Taxonomy with the creative model of self-managed
team, following an anallogy with Contradictions Matrix for solving the innovative
problems developed by Altschuller. The new model to stimulate the innovative
potential it is described by the Innovative Acting Matrix and by the Innovative
Support Matrix.
Key words: innovation, creative self-managed team model, Blooms
taxonomy, Contradictions Matrix, Innovative Acting Matrix, Innovative Support
Matrix.
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
Petru Dua and Iuliana Laura Taranovschi
innovative process is facilitated, the innovative products are developed. These

are, often, considered to be the result of collaborative teamwork. Increasing the
organizational effectiveness in a specific domain is a primary goal for selfmanaged work teams. Self-managed teams are so intensively focused on high
performance that often individual needs of the team members can be
overlooked. As a result of the integration of viewpoints, self-managed teams
offer multifunctional definitions and solutions to problems that generate
innovative products or services.
In this paper are explored the cognitive factors which influence the creative
thinking of individuals who work in an organizational team and in an
organization with high creative potential.
1. Conceptual Determination
In order to understand the manner in which the factors involved in the
innovative process operate, it is used as starting point Blooms Taxonomy for
educational objectives.
In 1956, Benjamin Bloom, an
educational psychologist, developed
taxonomy for Educational Objectives.
This taxonomy became a key tool in
structuring and understanding the
learning process. The taxonomy,
examines the cognitive domain,
which categorizes and orders thinking
skills and objectives. It is a
continuum
of
six
domains:
knowledge,
comprehension,
application,
analysis,
synthesis,
evaluation. The domains are arranged
in ascending order, each level is
Fig. 1 Blooms Taxonomy [3].
assimilated in the others presence,
starting from the basic level knowledge thought the higher one evaluation, as it is represented in the Fig. 1.
These depend one from another can not achieve one domain without other and
are arranged at three levels: cognitive domain, corresponding verbs and
cognitive activity support. Each cognitive domain, that is found in the center of
the model presented in the Fig. 2, are in correspondence with a set of verbs
located in the middle of the model and with a set of nouns situated at their
external level in this model. The set of verbs express how can be acted in order
to attain the corresponding domain. Nouns express the support that it is offered
to perform the activities for each domain.
239
Fig. 2 Blooms Taxonomy [4].
This paper aims to structure a model in which cognitive elements of

Blooms taxonomy can be integrated with another model which is specific for
activities of a team with high innovative potential.
2. Case Analysis
From previous studies in the area of the self-managed team concept, situated
in a research environment from industrial engineering domain, resulted a model
of creative self-managed team. To define this model, we used the following
tools: General Innovation Skills Aptitude Test - GISAT [5], Belbin test to
determine the role in a team [2] and Building Blocks focused on variables of
effectiveness teams [13]. The model was developed on the most significant
correlations between subjects scores to the three dimensions: innovative
profile, team roles and team variables.
The scores obtained for subject (S3) identified as Monitor-Evaluator and
subject (S15) identified as Plant from the perspective of the roles assumption
correlate with the scores obtained for first pillar - Generating and Assessing
Ideas from the innovative profile and with the team variables: Openness and
Confrontation, Support and Trust, Cooperation and Conflict (see on Fig. 3, the
quadrant for S15, S3). The scores obtained for subject (S11) identified as
Implementer and subject (S16) identified as Shaper from the perspective of the
roles assumption correlate with the scores obtained for second pillar - RiskTaking from the innovative profile and with the team variables: Judicious
Procedures, Appropriate Leadership and Regular Review (see on Fig. 3, the
240
quadrant for S11, S16).

The scores obtained for subject (S7) identified as Resource Investigator and
subject (S10) identified as Coordinator from the perspective of the roles
assumption correlate with the scores obtained for third pillar - Relationship
Building from the innovative profile and with the team variables: Individual
Development, Sound Inter-groups Relations and Good Communication (see on
Fig. 3, the quadrant for S7, S10). The scores obtained for subject (S4) identified
as Finisher and subject (S12) identified as Team Worker from the perspective of
the roles assumption correlate with the scores obtained for forth pillar
Implementing and Turning Ideas into Products, Processes and Services from the
innovative profile and with the team variables: Balanced Function, Clear
Objectives, Organizational Support (see on Fig. 3, the quadrant for S4, S12).
Fig. 3 The model of self-managed team [9].
In the literature, can be identified a method to stimulate the creativity

TRIZ The Theory of Inventing Problem Solving, comes from Russian
Language Teoriya Resheniya Izobretatelskikh Zadatch, developed by
Genrich Altshuller, Russian author. The method was developed after analysis of
almost two million of the worlds most successful patents. TRIZ is a method
which provides resources to access the best solutions to solve the innovative
problems. In this method can be identified four pillars that differentiate this
strategy from other innovative strategies for solving problems [7]:
Pillar 1: Contradictions many problem solvers try going directly
from problem to solution through trial and error. Looking at an analogous
standard problem and its associated standard solution is a more efficient
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].
Fig. 4 Extract from Contradiction Matrix [7].
Pillar 2: Ideality when TRIZ founder studied the patents database, he

discovered that the systems always tend to evolve towards increasing ideality.
The ideality part of TRIZ encourages problem solvers to break out of the
traditional start from the current situation type of thinking, and start instead
from what is described as the Ideal Final Result (IFR). The simple definition of
IFR is that the solution contains all of the benefits and none of the costs or
harms (environmental impact, adverse side-effects, etc) [7].
Pillar 3: Functionality this concept is developed by TRIZ through the
fact that it integrates knowledge from physics, chemistry, mathematics,
engineering, management and political science. It is successful largely because
that knowledge is in a form accessible to all users. In this way, the functionality
becomes the connection that makes possible sheering of knowledge between
widely differing industries [11].
Pillar 4: Use of resources in TRIZ terms relates to the unprecedented
emphasis placed on the maximization of use everything contained within a
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.
Fig. 5 Innovative Acting Matrix.
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
Fig. 6 Innovative Support Matrix .
The exploration of innovative problem is realized, by this time, using the

tools provided as support for each acting-innovative stage.
5
Conclusions
1. As a result of this scientific approach, it was designed a model to

stimulate the innovative potential with the support of two innovative matrices:
Innovative Acting Matrix; Innovative Support Matrix.
2. Innovative Acting Matrix represents a matrix that offers the necessary
stimulants for innovative acting (the verbs).
3. Innovative Support Matrix represents a matrix that offers the necessary
support for innovative results (the nouns).
4. Next research will focus on the applicability of these matrices in
situations that require the innovative potential.

Iasi, Romania,
e-mail: itaranovschi@tcm.tuiasi.ro
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
3. Blooms Taxonomy, Consulted on 11.12.2009 at:

http://blogs.wsd1.org/etr/files/blooms_taxonomy.jpg.
4. Blooms Taxonomy, Consulted on 15.12.2009 at:
http://mysciencelessons.files.wordpress.com/2009/07/bloomwheel3.gif.
5. C a m p b e l l A., W a t t D., Building innovation capacity one skill at a time. Proc.
of The Conference Board, Montreal-Canada, 2004.
6. C a v a l l u c c i D., TRIZ, the Altshullerian approach to solving innovation
problems. In Engineering design synthesis: Understanding approaces and tools.
(Chakrabarti A., Eds.), Springer Verlag, London, pp. 131-149, 2002.
7. CREAX Innovation Suite, Consulted on 03.12.2009 at :
http://www.creax.com/triztab/creaxsuite31/english/TRIZ/TRIZ.html.
8. L a n d r y C., The creative city: A toolkit for urban innovators, Earthscan
Publications, London, 2001.
9. T a r a n o v s c h i I. L., D u a P., N a g Gh., Exploring creativity in a research
environment. Proc. 14 th International Conference ModTech, Slnic MoldovaRomnia, 2010.
10. T e r n i n k o J., Z u s m a n A., Z l o t i n B., Systematic Innovation, An
Introduction to TRIZ. CRC Press LLC, USA, 1998.
11. V i n c e n t J., M a n n D., Systematic technology transfer from biology to
engineering. In Philosophical transactions - Royal Society. Mathematical,
physical and engineering sciences. Royal Society Publisher, London, 2002.
12. W e s t A. M., F a r r J. L., Innovation at work. In Innovation and creativity at
work: Psychological and Organizational Strategies, John Willey & Sons, New
York, pp. 3-13, 1990.
13. W o o d c o c k M., Team Development Manual. Gower Publisher, Aldershot, UK,
1989.
CERCETRI CU PRIVIRE LA ACTIVITATEA DE
INOVARE DIN MEDIUL TEHNIC
(Rezumat)
Din cercetrile anterioare asupra elementului inovativ, am ajuns la concluzia c
structura care faciliteaz dezvoltarea acestuia o regsim la nivelul echipei autoconduse.
Modelul de echip autocondus propus pentru stimularea potentialului inovativ este
susinut de patru piloni: Generare de Idei, Asumarea Riscului, Construire de Relaii i
Implementare. n momentul n care am privit aceste elemente din perspectiva
obiectivelor inovative i am considerat c inovarea se dezvolt cu ajutorul gandirii
creative, am utilizat taxonomia definit de Bloom pentru a elabora un model cognitiv
privind stimularea potenialului inovativ. Elementul strategic care a stat la baza definirii
acestui model este este reprezentat de Matricea Contradiciilor TRIZ elaborat de
Genrich Altshuller. n urma acestui demers au rezultat dou matrici: Matricea Aciunii
Inovative i Matricea Suportului Inovativ. Rolul acestora const n faptul c potentialul
inovativ poate fi stimulat daca sunt asimilate succesiv elementele taxonomice cognitive.

Publicat de
Secia
CONSTRUCII DE MAINI
ADVANTAGES AND WEAK POINTS OF DIFFERENT

REVERSE ENGINEERING (RE) TECHNIQUES
BY
ANDREI MIHALACHE, GHEORGHE NAG

and MARIUS-IONU RPANU
Abstract. The concept design, as part of the engineering process and as
support for CAD/CAE techniques, allows optimizing the product concept before
manufacturing, with CAM assistance. For some product development processes,
RE allows us to generate surface models by three-dimensional (3D) scanning
techniques, and as follows, this method allows different parts to be manufactured
in relatively short development period of time. The aim of this paper is to look
over RE techniques and to find possible advantages or weak points of different
scanning systems.
Keywords: Scanning (digitizing), reverse engineering, product design, rapid
prototyping, CAD/CAE/CAM.
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
Andrei Mihalache et al.
competiveness, and the requirements push everyone in rapid development. In

such case, Reverse Engineering (RE) techniques are useful. The most important
instruments found in the process, are different scanning systems, which
provides, in a short amount of time, dimensional description in digital concept.
RE has now been accepted as part of contemporary product design and
manufacturing processes. The RE technique is easy defined as the process
which results in creation of a mathematical model from a physical one.
Different cases have different RE requirements, from recovering the
mechanical design information to design based modifications. In case of
extracting the mechanical design information, we will be interested in tolerance,
as opposed to the case of design based extraction where the precedent will be
that of extracting the design intent.
So, the RE methods and techniques are essential because they allow to
capture and digitize an object surface geometry for later use with
CAD/CAE/CAM [1].
2. Product Development Approaches

Nowadays the management of product design can be achieved based on two
methods shown in (Fig. 1), a conventional approach and an unconventional
approach [1].
The conventional approach in product development using CAD/CAE/CAM
systems, normally starts with geometrical modeling using a CAD system. The
geometric model can be represented by wire frame or as surface or as solid
structure. Generated CAD data, can be exported to a standard format (IGES
points/STL binary, ASCII data, DXF poly line, VDA points or IGES/STL
surfaces) and then imported in the same data format to CAE systems (allowing
numerical model simulation) and/or CAM systems (allowing tools trajectory to
be generated). In a system with a unique database, the design information can
be divided to each application automatically, without manual transfer of data,
each time [3].
In the unconventional approach, we can see that the product development by
the conventional approach is not always applicable when we aim to re-engineer,
simulate or optimize parts/moulds/tools already existing, but without the CAD
data format. So, it will be necessary to apply one technique that allows to
capture part/mould/tool (or prototype) geometry and to generate a numerical
model to be used in CAE and CAM systems. This is called reverse engineering.
Client specifications
Physical model
(sample or prototype)
Digitizing system
3D geometry
Conventional
approach
Points cloud
3D coordinates
3D CAD
system
Reconstruction software for

the model
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)
Fig. 1 Sequences in manufacturing engineered products

(parts/moulds/tools)-adapted [1].
247
248

Physical model
(sample or prototype)
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)
Fig. 2 Digitizing techniques for 3D-geometries and generated data.
3. Differences between Digitizing and Scanning Processes

3D scanning (digitizing) is the process of gathering data from undefined
three-dimensional surface. During the scanning process, an analogue scanning
probe is commended to move back and forth (contact or non-contact) across the
unknown surface. During the process, the system records information about the
surface in form of numerical datas and generates a points cloud matrix (3D
coordinates). The terms of digitizing and scanning are often used to describe the
same process. Traditionally, digitizing refers to the process of taking discrete
points using a touch-sensible probe. However, with the introduction of new
technologies in data capture, such as laser or camera, digitizing term is now
used as general description for the process of data acquisition from undefined
surfaces [2] [4]. The digital points cloud can be captured using different
digitizing techniques (Fig. 2), classified into two major groups:
Mechanical techniques (by physical contact sensors);
Optical techniques (by non-contact with the object).
Related to the first group, usually they use a coordinate measuring machine
CMM, or a CNC-milling machine equipped with physical touching probe
sensors (ex. Retroscan or Renscan Renishaw, UK). Related to the second
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
Precise and fast scanning on Z-axis (0.001 mm)

Non-contact method
Equipment high price

It is not possible to scan reflective
materials
Scanning in X- and Y-axis is
inaccurate
(0.035 0.060 mm)
It is not possible to scan notches
areas or steep surfaces due to
additional reflections.
Sensibility to air dust
High price of equipments
Accuracy decreases linear with
camera distance
Scanning angle is equal regardless
the shape of parts surface; in case
of steep angles the measurement in
inaccurate
By camera scanning, a very sharp
picture is required; simultaneous
scanning of far and closer surfaces
demonstrates linear deviation of
results depending on focal distances
In case of oily or wet parts the
measurement in inaccurate
Dust causes bad scanning
It is possible to scan soft materials (or liquids)
CCD
cameras
Fast
It is possible to use two or three cameras
simultaneously
Insensible to part color
Non-contact method, it is possible to scan soft

materials
Contact
(classic)
Contact
digitizers
(Renishaw)
In case of special coaxial lightning it is possible to

scan small diameters and high depths on Z-axis
It is possible to scan very small areas: 1 mm2 (one
scan only needed) accuracy is above few
micrometers
Scanning on flat surfaces is very fast
Very precise on all axis (depends on scanning
equipment)
Fast scanning of known geometrical parts
Precise scanning of coins or similar geometries
Manual or automatic scanning possibility
Hand-held scanning equipment is useful for
scanning big products (airplanes, ships, big
machines or devices)
Precise on all axis (0.001 mm)
Relatively low price
It is possible to use different styluses for different
surfaces
Oil, liquid or dust do not interfere with the
scanning process
It is possible to scan unknown steep surfaces
Its inadequate for soft materials.

Scanning unknown surfaces is either
impossible or very slow and
inaccurate
Minimum diameter of stylus is 0.3

mm, scanning surface roughness is
not possible
It not possible to scan soft materials
Though material of stylus is wear
resistant, its surface wears in time
Scanning speed is lower (compared
to non-contact systems)
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
Gh. AsachiTechnical University,

Department of Machine Manufacturing
Iasi, Romania
e-mail: andrei.mihalache@yahoo.com
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.
AVANTAJE I PUNCTE SLABE ALE DIFERITELOR

TEHNICI DE INGINERIE INVERS
(Rezumat)
Dezvoltarea de produse (piese/ matrite/ scule) mpreun cu ingineria invers
integrat, este o metod recent, aflat n faza de cercetare i dezvoltare.
Aceasta lucrare prezinta unele posibilitati de utilizare si beneficii ale metodei n
procesul de productie, dar evideniaz i avantajele i punctele slabe ale diferitelor
echipamente de scanare utilizate n prezent.

Publicat de
Secia
CONSTRUCII DE MAINI
THERMOGRAPHY APPLIED TO BONE DRILLING

BY
ROBERTO LOPEZ1, MANUEL SAN JUAN1, FRANCISCO SANTOS1,

OSCAR MARTN1 and FLORIN NEGOESCU2
Abstract. Infrared radiation is a form of electromagnetic radiation. All matter
above absolute zero (0 K, -273C) emits infrared energy. Thermographic cameras
can be used for measure temperatures in a lot of applications. 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. 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 of this type.
Key words: thermography, bone, drilling.
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
Roberto Lopez et al.
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.
2. State of the Art

Due to the drilling process, the bone tissue heats up and if the temperature
around the drilling area exceeds the critical limit of 50 C, could result in
thermal necrosis, which is the death of the tissue exposed to that high
temperatures. Therefore, the importance of temperature control in this type of
operations results obvious.
Most researchers use conventional temperature measure methods, such as
the placing of thermo couples (A b o u z g i a & J a m e s, 1995), (B a c h u s et al.,
2001), (M a t t e w s et al., 1984). In this way, the temperature record can be
continuous; not on the contact area between the tool and the bone, but in the
gaps set for the thermo couples away from the drilling area.
The measurement of temperature (for example for fever) is very well known
in old and modern medicine. The infrared thermology is a contactless method
for measuring temperature and seems to be very useful in research and everyday
medicine. Everyday there are more and more studies that use thermography for
the measurement of temperature in different process. This way, (W i l d et al.,
2003), determined the temperature in bone drilling for facial human bones for
inplants. They obtain results for temperature peaks although they dont specify
how they record the thermography at the highest temperature.
Other researchers, like (U d i l j a c k et al., 2007) also use the thermographic
camera to measure the temperature of drilling while using different drill bits.
But neither in this case, nor in other cases like (W i l d et al., 2003), there is a
continuous record of the temperature to be processed. So it is understood that
the evaluation of the highest temperature is done objectively.
The advantages of the infrared thermography in continuous record to
measure the temperature are obvious:
253
There is no delay between the generation of heat and the perception of

the local heat.
It is possible to measure the temperature over a whole area.
Dynamic changes in temperature can be seen during the process.
The dynamic register of thermographies allows making a complete study of
the evolution of the temperature during the process.
3. Experimental Set-Up
In this study we have used infrared thermology to determine the cutting
temperature in bone drilling for dental implants. Figure 1 shows the
experimental set-up.
Fig. 1. Experimental set-up
Tools used during the tests are:

Thermographic camera
A camera Infratec of high-speed thermography was used (ImageIR 5300,
range from 700 to 1 m; with a high frame rates of up to 250 Hz). A lens with a
focal length of 100 mm was used with a macro lens of 500 mm. The infrared
camera thermography enabled the estimation of temperature increase and
temperature fields at bone drilling. The camera was set at an angle of 30 in
relation to the hole axis (drilling direction), and the temperatures were
determined (Fig. 1).
The results of the thermographic camera are images where each temperature
isotherm is denoted with a different colour, see Fig. 4.
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.
Fig. 2 Surgical motor attached to the force generating machine.
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.
Fig. 3 Drill used in test.
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.
Fig. 4 Drilling of bone and thermography.
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.
Fig. 5. Thermography sequence.
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.
Fig. 6 The fall of temperature.
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.
Fig. 7 The increase of temperature.
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.
Against a linear regression, the quadratic function explained up to 95.87% of

the total variability of data (5% more than the linear), making the adjustment
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
Iai, Romania
e-mail: negoescu@tcm.tuiasi.ro
Acknowledgments. This work has been developed thanks to DPI2006-15502-C02-02

Project from the National Program of Design and Industrial Production supported by
MEC (Spain) and FEDER.
REFERENCES
1. A b o u z i a M. B., J a m e s D. F., Measurements of shaft speed while drilling
through bone. J. Oral Maxillofac Surg, 53 (1995) 1308-1315.
2. B a c h u s K. N.; R o n d i n a M. T., H u t c h i n s o n D.T., The effects of drilling
force on cortical temperatures and their duration: and in vitro study. Medical
Engineering & Physics, 22 (2000) 685-691.
3. M a t t e w s L. S.; G r e e n C. A, G o l d s t e i n S. A., The thermal effects of skeletal
fixation-pin insertion in bone. Journal of Bone & Joint Surgery, 66A, 7 (1984)
1077-1083.
4. U d i l l j a k T.; C i g l a r D. & S k o r i c S., Investigation into bone drilling and
thermal bone necrosis. Advances in Production Engineering& Management, 2
(2007) 103-112.
5. W I l d W.; S c h t t e S. R.; P a u H. W.; K r a m p B., J u s t T., Infrared
thermography as a non invasive application for medical diagnostic. Proceedings
of XVII IMEKO World Congress, pp. 1744-1748, June 2227, 2003, Dubrovnik,
Croatia.
TERMOGRAFIE APLICAT N FREZAREA OSOAS
(Rezumat)
Radiaiile n infrarou sunt o form a radiaiei electromagnetice. Camere
termografice pot fi utilizate pentru msurarea temperaturii ntr-o mulime de aplicaii.
Cu acest instrument, este posibil s se determine zonele locale de nclzire n timpul
procesului de taiere pentru implanturi dentare. Aplicabilitatea termografiei pentru
studiul dinamic al temperaturii n operaiunile de tiere a implanturilor stomatologice, se
dovedete a fi foarte mare.

Publicat de
Secia
CONSTRUCII DE MAINI
STUDY ON THE OPPORTUNITY OF A NEW

GLASS RECYCLING FACTORY FOR REGIONAL
SUSTAINABLE DEVELOPMENT
BY
OLGA MARINA MONTES1 and VASILE V. MERTICARU2

Abstract. The paper presents a theoretical study and business justification
concerning the opportunity for the establishment of a new factory for glass
recycling, respectively for turning waste glass into raw material for manufacturing
new usable glass products. The presented study is part of a research approach
developed in the direction of establishing the economical and technical conditions
for establishing the discussed glass recycling factory in the province of
Valladolid, autonomous community of Castilla y Len from Spain, as possibility
to enlarge the number of similar enterprises, for regional sustainable development.
The state of development for the glass recycling industrial sector in Spain is
discussed and the adequacy of the location is analyzed in the paper, together with
the presentation of related conclusions.
Key words: glass recycling, sustainable development, business justification.
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
Olga Marina Montes and Vasile V. Merticaru
concepts of Reducing, Reusing, Recycling and Recovery of material resources

are also very important in providing Sustainable Industrial Development.
Of course, the cradle-to-cradle way in LCA is nowadays generally
accepted as the only viable solution for providing environmental protection and
avoiding the depletion of natural resources, as long as materials can be reused,
no waste gets produced or otherwise it can be recycled so that no damaging
impacts on the environment are generated within the closed loop of the product
lifecycle, [3], [5], as is shown in Fig. 1.
Off-site recycle
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
Fig. 1 LCA based on cradle-to-cradle concept.
On the other hand, as a policy measure which comes to outstand the

industrial producers crucial role in reducing the environmental impacts of their
products throughout their entire lifecycle, Extended Producer Responsibility
(EPR), encourages the reuse of products and packaging and, not less important,
the Design for Environment practices [5].
In relation to the above mentioned considerations, the paper presents a
theoretical study and business justification concerning the opportunity for the
establishment of a new factory for glass recycling, respectively for turning
waste glass into raw material for manufacturing new usable glass products. The
presented study is part of a research approach developed in the direction of
establishing the economical and technical conditions for establishing the
discussed glass recycling factory in the province of Valladolid, autonomous
community of Castilla y Len from Spain, as possibility to enlarge the number
of similar enterprises, for regional sustainable development.
261
Glass recycling, as industrial activity which comes to replace important

quantities of traditional raw material in new glass products industry with
processed waste glass, can be included under the umbrella of the Cleaner
Production techniques, [4], [5], covering the areas of off-site recycling and
process change, as it is shown in Fig. 2.
Cleaner Production Techniques
GREEN PRODUCTION
Waste
Reduction
Product
Modification
Process Change
Better
housekeeping
Process
control
RECYCLING
OFF-SITE
Recycling
On-site
Recycling
GLASS
RECYCLING
Technology
Raw material
change
change
Fig. 2 Area covered by glass recycling within Green Production.
In a plant engineering approach, it is important to analyze the industrial

sector in which the project will belong and to see if it makes sense to think
about the fact of beginning the studies to start it. In order to this, the state of
development for the glass recycling industrial sector in Spain is discussed and
the adequacy of the location is analyzed in the paper, together with the
presentation of related conclusions.
2. The Glass Recycling Sector in Spain

Since the end of the Second World War and particularly during the last
decade, there has been registered an increasing throw away tendency.
Statistics show that in the European countries the amount of household waste
has increased to more than 1 kg per person and per day, on average. The
composition of domestic waste varies from country to country, but the tendency
is the same everywhere: more paper, more plastic, more valuable metal and
glass, and less ash. Because of that phenomenon, the number of recycling plants
is also increasing, as response to balance the residuals proliferation.
262
According to the Spanish Ministry of Environment and Rural and Marine

Environment, each year in Spain approximately 592 kilograms of waste per
inhabitant and per year, [2], is produced. Much of this waste ends up in
landfills, occupying large areas of land and necessary addition to avoid soil
contaminating.
At the present time, in Spain there are a total of 15 plants dedicated to the
glass recycling and in the region of Castilla y Len there exist only one other
glass recycling plant.
Since the entry into force of the Packaging Act in 1998, all containers are
required to fund and implement a collection system being outlook openrecycling of packaging they put on the market. This legal obligation can be met
in two ways:
Individually, by a deposit-refund.
Collectively, through an integrated management system.
In this context and in order to provide an effective and economical solution
to all employers who package their products mostly in glass, Ecovidrio, [2],
arose as a nonprofit association whose main objective is to ensure the
traceability of glass packaging. In this way, every packaging that enters in the
recycling chain follows the appropriate process to become a completely new
one. Being the only integrated management system that specializes in glass, is
able to reduce costs and provide better service to all companies, large or small.
The 2008 financial year ended with 2455 companies adhered to Ecovidrio, with
107 more than in 2007, to contribute to sustainable development and
environmental care.
In Fig.3 and Fig. 4 there are summarized the most important data on glass
recycling in Spain from 2001 until 2008 [2].
Fig. 3 Statistic data on glass recycling in Spain from 2001 until 2008 [2].
263
Rate of recycling, [%]
Fig. 4 Evolution of glass recycling rate in Spain from 2001 until 2008.
3. Location Analysis and Justification

For making the decision to locate an industry, we need to know a number of
parameters that help defining the characteristics of possible locations and see
which one bestly fits the needs of industry to be implemented. The location is
an individual choice in which each industry values and ponders the most
appropriate option for its, [1].
The location parameters can be grouped into the following points:
Parameters concerning the nature of the industry and its classification;
Parameters concerning the analysis of the geographical;
Parameters analyzing the urban conditions;
Parameters relating to the study of environment;
Parameters relating to human capital and intellectual capital.
From the analysis of the above mentioned points there has been determined
to locate industrial facilities of the plant object of labor in the industrial area of
San Miguel del Arroyo, in the province of Valladolid, autonomous
community of Castilla y Len from Spain.
The following considerations, related with each of the above named
sections, provide a synthesis of the selected location.
3.1. Nature of the Industry and its Classification
The council enacted in 1982 the European Community Directive 82/501 on
the prevention of major accidents arising from certain industrial activities. This
standard classifies the considered facility as industrial activity and performing
dangerous quantities manufactured, processed or stored. Moreover, legalization
itself exists in each country and / or region and there may be legislation also
own by municipality. First, is taken into account that the legislation in force in
San Miguel del Arroyo allows the implementation of this industry to be built.
264
3.2. Analysis of geographical conditions

In this section it has been checked that the chosen area has all the natural
security conditions necessary for any deployment. There have been key issues
such as security of land, about flooding (unlikely), forest fires, erosion and
deforestation. With respect to climate and the presence or absence of surface
water, to implement the factory has no special requirements. With regard to the
existence of communications, industrial estate where the plant will be located
has good links. This is a site near Valladolid, capital of the Community of
Castilla y Leon and also is close to Madrid, capital of Spain, and near the
national road N1, the road which connects the center of the country to the
north. Communications are important both for the supply of waste glass and to
deliver the finished products.
In the community Castilla y Len there are important industries like
Renault, Michelin, Iveco, Pascual, Firestone, and the glass waste from these
industries should be the materials the plant will recycle. Also, Madrid is near of
the location, and the plant can recycle the waste from the industries of these
regions or from other regions of the country.
3.3. Analysis of urban conditions
Each plot of the considered area has planning conditions that define the
amount and location of the intended buildings. The following data provide some
of the characteristics of the industrial area of the town of San Miguel del
Arroyo.
General Information - Surfaces:
Total area: 86,810.27 m;
Endowment Surface: 5201.32 m;
Public open space (green areas): 40,672.98 m;
Maximum Plot: 2868.74 m; Minimum Plot: 177.18 m;
Minimum strippable: 175 m.
Other Data:
Number of plots: 44; Free Plots: 11;
Access roads: Yes, in concrete pavement;
Supplies: Yes, municipal network;
Sanitation: Yes, primary treatment;
Lighting: Yes; Electricity: Yes; Gas: Yes;
Telecommunications: Yes.
Rules planning: Municipal Town Planning Regulations;
Maximum building height: 7 m and 10 m ridge cornice;
Setback: 5.00 m and 3.00 m front edge - rear edge.
The previous planning conditions have been of paramount importance when
deciding the final location of the plant.
265
3.4. Environmental Study

Within this point, there have been considered the political and social
stability of the country or region where the possible location is. In addition, the
security of people working in and for industry should be maximal. With respect,
it must be said that Spain is a safe country without major political conflicts,
social, ethnic or religious, belonging to the European Community, which would
also facilitates free trade and unimpeded across Europe.
The industrial estates in the town of San Miguel del Arroyo is promoted by
an industrial Sodeva (Provincial Development Society of Valladolid), which
brings many advantages and subventions.
It should be noted that the town of San Miguel del Arroyo is close to the
University of Valladolid, which can determine a dynamic and availability of
their students as interns in research and development processes as well as in
other departments.
3.5. Human and Intellectual Capital
There has been taken into account that the area of localization has enough
human capital to meet the staffing needs of the company to implement. The
designed industrial plant requires very few employees, and most workers do not
require a very thorough qualification. Staff will be working in industry and must
have an adequate quality of life for their needs. They have the possibility of
living in a city, Valladolid, with all necessary services available to them, or live
in the town of San Miguel del Arroyo, where housing prices are lower, besides
having all the primary needs; schools, leisure areas, sports areas, shops. These
basic parameters can be framed within what would be called family welfare.
Besides human capital, there must be taken into account the intellectual
capital, i.e. preparation. As already mentioned above, the University of
Valladolid is relatively close to the point of location and can have enough well
prepared people.
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
Received: February, 20, 2010
Universidad de Valladolid,
Valladolid, Spain
e-mail: olga.marina.montes@gmail.com
2
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.
STUDIU ASUPRA OPORTUNITII UNEI NOI FABRICI

DE RECICLARE A STICLEI, PENTRU O DEZVOLTARE
REGIONAL DURABIL
(Rezumat)
n lucrare este prezentat un studiu teoretic i justificarea din punct de vedere

antreprenorial privind oportunitatea nfiinrii unei fabrici noi de reciclare a sticlei, respectiv
pentru transformarea deeurilor de sticl n materie prim pentru fabricaia de produse noi
din sticl. Studiul prezentat este parte a unui demers de cercetare dezvoltat n direcia
stabilirii condiiilor tehnice i economice pentru nfiinarea fabricii de reciclare a sticlei n
discuie, n provincia Valladolid din comunitatea autonom Castilla i Leon din Spania, ca
posibilitate de cretere a numrului de astfel de ntreprinderi similare, pentru o dezvoltare
regional durabil. n lucrare, se discut stadiul de dezvoltare a sectorului industrial de
reciclare a sticlei n Spania, se analizeaz adecvana localizrii investiiei i sunt prezentate,
de asemenea, concluziile aferente.

Publicat de
Secia
CONSTRUCII DE MAINI
MODELING 3D SURFACE OF TEXTILE STRUCTURES

FOR FLUID FLOW IMPROVEMENT
BY
CTLIN DUMITRA1, CARMEN LOGHIN1, SULEYMAN YALDIZ2,

MEHMET SAHIN2 and LUMINIA CIOBANU 1
Abstract. The efficiency of thermal insulation in a windy environment can be
increased by using adequate protective equipment. Following a FEA analysis
regarding the flow of fluids (air) at body surface, the presence of critical zones
with turbulences was detected. The critical zones are the ones with a change in
geometry, such as the shoulders, hips or hands. This generates the idea that
something has to be done to improve the fluid flow at surface level of the exterior
garment layer. The main problem is to ensure a certain path for the fluid
movement, so that the flow will become laminar. A simple and feasible solution is
creating a textile structure with an architecture that includes these paths. Different
variants of 3D geometries are proposed. Their dynamic behaviour is studied in
order to rank the proposed fabricse abstract of the paper is to be written here. It
contains the main ideas and original contributions and conclusions of the authors
research.
Key words: laminar air flow, weft knitted fabrics with relief effects, sandwich
fabrics, FEA, meshed models, fluid flow.
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
Ctlin Dumitra et al.
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.
Fig. 1 The type of FEA element used.
Fig. 2 Finite element based model.
It was presumed that a pressure field caused by winds up to 100 km/h is

applied to the model, corresponding to all human body. The selected properties
of the environment are illustrated in Fig. 3.
269
Fig. 3 Physical properties of the environment.
The analysis results are presented as pressure fields, velocity and

turbulences. The output variables were calculated for different values of the
input parameters. The wind was considered to blow with 90 km/h velocity for a
period of 90 seconds. The results are presented in Figs. 4 to 6.
t = 15 s
t = 80 s
t=3s
Fig. 4 Pressure fields

distribution
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
3.1. Definition of the Models

Two models were created using AutoCAD 2007, considering the human
shoulder. The first model represented the shoulder without a modified surface
and is illustrated in Fig. 7. This model continues the 2D analysis presented
above. The other model considers the surface modified by the presence of the
fabric. The fabrics considered had a rectangular shaped cross section as
discussed above and the resulting model is presented in Fig. 8.
The section height is 10 mm. Lower height values proved to be inefficient.
The rectangular channels are spaced also at 10 mm. The modelled shoulder was
included in a space volume in order to simulate accurately the air flow.
Fig.7 Model without 3D surface effect
Fig.8 Model with channels with

rectangular cross section
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.
Fig. 9 Selected air properties
Fig. 10 Meshed model
Fig. 11 Meshed model for shoulder and fabric
The models were analyzed using finite elements, based on steady fluid flow
processor. The parameters considered are illustrated in Figs. 12 and 13.
Fig. 12 Parameters for steady fluid flow
Fig. 13 Considered load curve
272
3.2. Results and Discussion

The following results were obtained based on the analysis presented above.
The pressure fields at shoulder level were identified for the shoulder without
fabric, as presented in Fig. 14. This constitutes a reference base for all models
containing the 3D fabrics. From Fig. 14 it results that there is higher pressure
field at the back of the shoulder, possibly caused by a suction effect.
Fig. 14 Distribution of pressure fields at shoulder level.
The distribution of air velocity at shoulder level is presented in Fig. 15 and

16. High velocity values are identified at the entire shoulder level, rest of the
human body these values are lower. That indicates that the turbulences are
higher in the shoulder area.
Fig. 13 Air velocity distribution,

general view
Fig. 14 Air velocity distribution
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.
Fig. 15 Air velocity distribution for

shoulder and 3D fabric general view
273
Fig. 16 Air velocity distribution for shoulder

and 3D fabric bottom view
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.
Technical University Gheorghe Asachi,

Iasi, Romania,
e-mail: dumitrascatae@yahoo.com
2
Selcuk University,
Department of Mechanics
Konya, Turkeyt
e-mail: syaldiz@yahoo.com
REFERENCES
1.
2.
3.
4.
5.
d e A r a u j o M., F a n g u e i r o R., H o n g H. Texteis technicos, volume I,

Braga, Portugal, 2000
H o l m e r I. Textiles for Protection Against Cold, in Textiles for Protection,
edited by R.A. Scott, Woodhead Publishing Ltd, Cambridge, UK, 2005, p. 378
397
L o g h i n C., C i o b a n u L., D u m i t r a s C. Study regarding the conventional
functions analyze and intelligent functions synthesis for the protective equipments
for aggressive environments research report, CEEX 105/2006
L o g h i n C., C i o b a n u L., D u m i t r a s C. Fundamentals of physical and
chemical processes at the interface level between aggressive environment and
protective equipment research report, CEEX 105/2006
Z i e n k i e w i c i O. C. The Finite Element Method for Fluid Dynamics,
Elsevier, Amsterdam Netherland, 2006
275
MODELAREA 3D A SUPRAFETELOR TEXTILE CU DESTINAIA

OPTIMIZRII CURGERII FLUIDELOR
(Rezumat)
Eficiena proteciei termice ntr-un mediu cu cureni puternici de aer (vnt) poate fi
mbuntit printr-un echipament de protecie adecvat. Ca urmare a unei analize cu
elemente finite a curgerii fluidelor (aerului) pe suprafaa corpului detectarea yonelor
critice turbionare poate fi realizat. Aceste zone au fost detectate n locurile n care se
modific geometria corpului cum ar fi umeri, brate, etc. Concluziile acestei analize au
condus la ideea c se poate obine o mbuntire a coeficientului de frecare a aerului
prin modificarea geometriei echipamentului textil. Se pune problema determinrii unei
geometrii convenabile a acestui echipament astfel nct sse micoreye sau s dispar
zonele de curgere turbionar. Se propun mai multe variante de structuri textile cu
geometrie 3D. Se realizeaz apoi un studiu destinat determinrii profilului optim al
structurii textile.
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
Stachie Marius 221

Storm Kjrside Birgit 105
Strjescu Eugen 31,143
Stroi Daniel 205
T
Tapoglou Nikolaos 21
Teodor Virgil 41,49
Tia Irina 153
Tudorache Mihaela 215
Taranovschi Iuliana 237
U
Ungureanu Ctlin 65,97
V
Vietor Thomas 227
Vlad Daniel Viorel 57
Vod Mircea 197
W
Weingold Andrei 83
Y
Yaldz Suleyman 267
Z
Zahariea Dnu 215, 221

CM Fasc.2 2010

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CM Fasc.2 2010

Uploaded by

Copyright:

Available Formats

BULETINUL

Tomul LVI (LX)

BULETINUL INSTITUTULUI POLITEHNIC DIN IAI

GHEORGHE ASACHI TECHNICAL UNIVERSITY OF IAI

MACHINE CONSTRUCTION Section

Editorial Advisory Board

Prof. dr. eng. Noura-Barbu Lupulescu, University

FACULTY OF MACHINE MANUFACTURING &

Papers published with the support of

BULETINUL INSTITUTULUI POLITEHNIC DIN IAI

TAXIARCHIS BELIS i ARISTOMENIS ANTONIADIS, Model pentru

NIKOLAOS TAPOGLOU i ARISTOMENIS ANTONIADIS, Determinarea

EUGEN STRJESCU, CONSTANTIN DOGARIU, OLIMPIA PAVLOV i

SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRACU i

SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRACU i

CTLIN FETECU, DANIEL-VIOREL VLAD i COSTEL MOCANU,

MIRCEA COZMNC, CRISTIAN CROITORU i CTLIN

ANA-MARIA MATEI i MARIUS NICOLAE MILEA, Forele de achiere

MARIUS-IONU RPANU, GHEORGHE NAG, IOLANDA-ELENA

IUSTINA ELENA ROTMAN, PETRU DUA i RADU ADRIAN BACIU

CTLIN UNGUREANU, RADU IBNESCU i IRINA COZMNC,

EROL MURAD, CTALIN DUMITRESCU, GEORGETA HARAGA i

BULETINUL INSTITUTULUI POLITEHNIC DIN IAI

TAXIARCHIS BELIS and ARISTOMENIS ANTONIADIS, Hobbing Wear

EUGEN STRJESCU, CONSTANTIN DOGARIU, OLIMPIA PAVLOV

SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRACU

SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRACU

CTLIN FETECU, DANIEL-VIOREL VLAD and COSTEL

MIRCEA COZMNC, CRISTIAN CROITORU and CTLIN

ANA-MARIA MATEI and MARIUS NICOLAE MILEA, Face Milling

MARIUS-IONU RPANU, GHEORGHE NAG, IOLANDA-ELENA

IUSTINA ELENA ROTMAN, PETRU DUA and RADU ADRIAN BACIU

CTLIN UNGUREANU, RADU IBNESCU and IRINA COZMNC,

BIRGIT KJRSIDE STORM, Adhesive Bonding of Surface Treated

CONSTANTIN CHIRI, ADRIAN HANGANU and DANIEL CALFA,

CTLIN DUMITRA, CARMEN LOGHIN, SULEYMAN YALDIZ,

BULETINUL INSTITUTULUI POLITEHNIC DIN IAI

ON THE PATTERN OF THE DESIGN MATRIX IN

ANTNIO M. GONALVES-COELHO1, GABRIELA NETIAN2,

Antonio M. Gonalves-Coelho et al.

Fig. 1 The Design Domains [1].

The left-to-right mapping between any two contiguous domains can be

Bul. Inst. Polit. Iai, t. LVI (LX), f. 2, 2010

Fig. 2 Example of a simple redundant design.

The adjustment of d can be achieved by suitably setting the values of two

Antonio M. Gonalves-Coelho et al.

Bul. Inst. Polit. Iai, t. LVI (LX), f. 2, 2010

FR2 = A24 A25 A26 DP5 .

As a result, the redundant design of Eq. (3) could be considered suitable

Antonio M. Gonalves-Coelho et al.

UNIDEMI, Universidade Nova de Lisboa,

MODELUL MATRICEI DE PROIECTARE IN CAZUL SOLUTIILOR

Bul. Inst. Polit. Iai, t. LVI (LX), f. 2, 2010

BULETINUL INSTITUTULUI POLITEHNIC DIN IAI

HOBBING WEAR PREDICTION MODEL

TAXIARCHIS BELIS and ARISTOMENIS ANTONIADIS

Taxiarchis Belis and Aristomenis Antodiadis

tool tangential displacement. Considering the above kinematic, two different

Fig. 1 Basic kinematics and essential parameters of gear hobbing.

2. State of the Art

Bul. Inst. Polit. Iai, t. LVI (LX), f. 2, 2010

Fig. 2 Determination of the flank wear at individual hob teeth