Journal of Mechanical Engineering / Strojniški vestnik 7-8-2010

55 YEARS

7-8 year 2010 volume 56 no.

Strojniški vestnik – Journal of Mechanical Engineering (SV-JME) Aim and Scope The international journal publishes original and (mini)review articles covering the concepts of materials science, mechanics, kinematics, thermodynamics, energy and environment, mechatronics and robotics, fluid mechanics, tribology, cybernetics, industrial engineering and structural analysis.

The journal follows new trends and progress proven practice in the mechanical engineering and also in the closely related sciences as are electrical, civil and process engineering, medicine, microbiology, ecology, agriculture, transport systems, aviation, and others, thus creating a unique forum for interdisciplinary or multidisciplinary dialogue.

The international conferences selected papers are welcome for publishing as a special issue of SV-JME with invited co-editor(s).

Editor in Chief Vincenc Butala University of Ljubljana Faculty of Mechanical Engineering, Slovenia Co-Editor Borut Buchmeister University of Maribor Faculty of Mechanical Engineering, Slovenia Technical Editor Pika Škraba University of Ljubljana Faculty of Mechanical Engineering, Slovenia Editorial Office University of Ljubljana (UL) Faculty of Mechanical Engineering SV-JME Aškerčeva 6, SI-1000 Ljubljana, Slovenia Phone: 386-(0)1-4771 137 Fax: 386-(0)1-2518 567 E-mail: info@sv-jme.eu http://www.sv-jme.eu Founders and Publishers University of Ljubljana (UL) Faculty of Mechanical Engineering, Slovenia University of Maribor (UM) Faculty of Mechanical Engineering, Slovenia Association of Mechanical Engineers of Slovenia Chamber of Commerce and Industry of Slovenia Metal Processing Industry Association Cover: Figure shows a process of positioning with resolution of 61 nm.

Fiberglas tube with diameter of 135µm is manipulated. A process of gripping is shown in the left column and a process of releasing is shown in the right column.

Image courtesy: Resistec UPR d.o.o. & Co. k.d.

Author of photography: Gregor Škorc

ISSN 0039-2480 © 2010 Strojniški vestnik - Journal of Mechanical Engineering. All rights reserved. SV-JME is indexed / abstracted in: SCI-Expanded, Compendex, Inspec, ProQuest-CSA, SCOPUS, TEMA. The list of the remaining bases, in which SV-JME is indexed, is available on the website. The journal is subsidized by Slovenian Book Agency.

President of Publishing Council Jože Duhovnik UL, Faculty of Mechanical Engineering, Slovenia International Editorial Board Koshi Adachi, Graduate School of Engineering,Tohoku University, Japan Bikramjit Basu, Indian Institute of Technology, Kanpur, India Anton Bergant, Litostroj Power, Slovenia Franci Čuš, UM, Faculty of Mech. Engineering, Slovenia Narendra B. Dahotre, University of Tennessee, Knoxville, USA Matija Fajdiga, UL, Faculty of Mech. Engineering, Slovenia Imre Felde, Bay Zoltan Inst. for Mater. Sci. and Techn., Hungary Jože Flašker, UM, Faculty of Mech. Engineering, Slovenia Bernard Franković, Faculty of Engineering Rijeka, Croatia Janez Grum, UL, Faculty of Mech. Engineering, Slovenia Imre Horvath, Delft University of Technology, Netherlands Julius Kaplunov, Brunel University, West London, UK Milan Kljajin, J.J. Strossmayer University of Osijek, Croatia Janez Kopač, UL, Faculty of Mech. Engineering, Slovenia Franc Kosel, UL, Faculty of Mech. Engineering, Slovenia Thomas Lübben, University of Bremen, Germany Janez Možina, UL, Faculty of Mech. Engineering, Slovenia Miroslav Plančak, University of Novi Sad, Serbia Brian Prasad, California Institute of Technology, Pasadena, USA Bernd Sauer, University of Kaiserlautern, Germany Brane Širok, UL, Faculty of Mech. Engineering, Slovenia Leopold Škerget, UM, Faculty of Mech. Engineering, Slovenia George E. Totten, Portland State University, USA Nikos C. Tsourveloudis, Technical University of Crete, Greece Toma Udiljak, University of Zagreb, Croatia Arkady Voloshin, Lehigh University, Bethlehem, USA Print LITTERA PICTA d.o.o., Barletova 4, 1215 Medvode, Slovenia General information Strojniški vestnik – The Journal of Mechanical Engineering is published in 11 issues per year (July and August is a double issue).

Institutional prices include print & online access: institutional subscription price €100,00, general public subscription €25,00, student subscription €10,00, foreign subscription €100,00 per year. The price of a single issue is €5,00. Prices are exclusive of tax.

Delivery is included in the price. The recipient is responsible for paying any import duties or taxes. Legal title passes to the customer on dispatch by our distributor.

Single issues from current and recent volumes are available at the current single-issue price.

To order the journal, please complete the form on our website. For submissions, subscriptions and all other information please visit: http://en.sv-jme.eu/ You can advertise on the inner and outer side of the back cover of the magazine.

We would like to thank the reviewers who have taken part in the peer-review process.

Strojniški vestnik - Journal of Mechanical Engineering is also available on http://www.sv-jme.eu, where you access also to papers’ supplements, such as simulations, etc.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8 Contents

Contents Strojniški vestnik - Journal of Mechanical Engineering volume 56, (2010), number 7-8 Ljubljana, August 2010 ISSN 0039-2480 Published monthly

Papers Gregor Škorc, Simon Zapušek, Jure Čas, Riko Šafarič: Virtual User Interface for the Remote Control of a Nano-Robotic Cell Using a Haptic-Device Zoran Stefanović, Ivan Kostić: Analysis of the Sailplane Final Approaches Performed by Cosine-Law Speed Variations Halil Demir, Abdulkadir Gullu, Ibrahim Ciftci, Ulvi Seker: An Investigation into the Influences of Grain Size and Grinding Parameters on Surface Roughness and Grinding Forces when Grinding Dragan Antić, Marko Milojković, Zoran Jovanović, Saša Nikolić: Optimal Design of the Fuzzy Sliding Mode Control for a DC Servo Drive Gašper Benedik, Brane Širok, Janez Rihtaršič, Marko Hočevar: Flow Characteristics of Bladeless Impeller Made of Open Cell Porous Material Slobodan Morača, Miodrag Hadžistević, Igor Drstvenšek, Nikola Radaković: Application of Group Technology in Complex Cluster Type Organizational Systems Miroslav S. Milićević: The Application of a New Formula of Nakaoka Coefficient in HF Inductive Welding Peter Fatur, Borut Likar: Statistical Analysis for Strategic Innovation Decisions in Slovenian Mechanical Industry Matija Javorski, Primož Čermelj, Miha Boltežar: Characterization of the Dynamic Behaviour of a Basketball Goal Mounted on a Ceiling Miodrag Zlokolica, Maja Čavić, Milan Kostić: Analytical Description of Polygonal Holes Boring - General Approach Nusa Fain, Niels Moes, Jože Duhovnik: The Role of the User and the Society in New Product Development Mitja Košir, Aleš Krainer, Mateja Dovjak, Rudolf Perdan, Živa Kristl: Alternative to the Conventional Heating and Cooling Systems in Public Buildings Instructions for Authors

423 436 447 455 464 477 491 497 505 511 521 531 541

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 423-435 UDC 621.398:004.8

Paper received: 23.04.2009 Paper accepted: 16.04.2010

Virtual User Interface for the Remote Control of a NanoRobotic Cell Using a Haptic-Device Gregor Škorc1,* _ Simon Zapušek2 - Jure Čas3 - Riko Šafarič2 RESISTEC UPR d.o.o. & Co. k.d., Kostanjevica na Krki, Slovenia 2 University of Maribor, Faculty for Electrical Engineering and Computer Science, Maribor, Slovenia 3 EM. TRONIC d.o.o., Maribor, Slovenia 1

This paper describes the development of a virtual user interface for the remote control of a nanorobotic production cell. The user interface combines two different software applications, built on two different software platforms. The first-host application is based on a LabView 8.5 software package and runs on a real-time target. It is used as a communication interface between the nano-robotic cell and a remote user interface. The remote application was created within a Microsoft Visual C 6.0 software package using C++ programming language. It is used for the virtual remote control of a nano-robotic cell. Depending on production demands, the remote user can choose between two different control techniques. The first one is a classical input algorithm where the user sets any move trajectory of the nano-robotic cell directly through a remote user interface. Each axis separately or all axes together can be moved in this way. Another control option supports acquiring movement trajectory using a hapticdevice. In this regime the user receives real-time force feedback information which makes remote control even more realistic. Both control regimes are supported by an animated, virtual, VRML model of the target application. This VRML model is used for off-line simulation or real-time monitoring of the target application movement. UDP protocol is used as a basic communication protocol between the host and remote applications. © 2010 Journal of Mechanical Engineering. All rights reserved.

Keywords: virtual remote control, nano-positioning, VRML, LabVIEW Real Time, MEMS assembly 0 INTRODUCTION Nano-technologies are actual and promising research fields nowadays. A lot of a work within nano-scale has already been published. Our work finds its place within a special part of nano-technologies namely automated nano-positioning and nano-assembly.

The sizes of manipulated parts are demanding especially clean working environment, therefore nano-applications are often built within vacuum chambers. These chambers usually include a semi-electronic microscope or special camera systems, for monitoring the processes. Authors have presented a solution for automated nanopositioning, based on SEM visual feedback in [1].

Another solution, where visual feedback was achieved using CCD camera and microscope is presented in [2]. A general review of scanning probe-based 2D nano-manipulation and gripperbased 3D nano-handling is given in [3]. Special user interfaces in combination with a haptic-

device, are used to upgrade such systems in order to simplify control [4-6]. Those haptic-devices are commonly connected directly to a target application. Live camera picture can be easily included in a user interface, for the purpose of monitoring of a nano-process [1-3]. Such a created application works fine as long as we control our application locally. As soon as we want to control an application remotely over the public network (e.g. using mobile internet connection from local mobile operator), live broadcasting of the camera picture can fulfill all of the communication capacities. We have a serious threat that transmission of the control’s information will be delayed. In the worst case can drastic delay of control information, lead to serious damages to the host application. It is known that a good internet connection is needed for the remote control of an application supported by live camera broadcasting. In those cases where we do not have it, is control of such an application very hard. In this paper we propose a

* Corr. Author's Address: Resistec UPR d.o.o. & Co. k.d., Krška cesta 8, 8311 Kostanjevica na Krki, Slovenia, gregor.skorc@resistec.si

423

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 421-433

solution for this problem, based on two different software applications (Fig. 1). The first runs on a host computer and can be used for local control of the target application. Because of the local character of the application, it supports a live camera picture. The second application is a remote application which remotely communicates with the host application, and gives full support to the haptic-device. In remote application we have replaced live camera picture with a virtual model, which is animated with real time positional information extracted from control data packets.

Broadcasting of live camera pictures was disabled (the lack of a live camera picture has been replaced by a corresponding virtual model) and the part of communication capacities was released. Because the model animation algorithm runs on the remote computer, and live camera picture is not transmitted anymore, is amount of transmitted data reduced to minimum. The virtual model is used for collision detection and

calculation of feedback forces for the hapticdevice, in continuation. The model gives us, in this case, even more opportunities for upgrading the application with newer functions than a camera picture gives. The developed application scales real operations within nano-scale, into virtual animation within macro scale. A communication link between the nano and macro worlds is established. In comparison to systems [1-3], where control of positioning device has a local character, our system gives an opportunity to overtake control remotely. Positioning is possible with accuracy of 61 nm and speeds of up to 35 mm/s. System is basically being developed for a purpose of MEMS assembly. Presented remote interface makes it useful even in a process of e-learning, within lectures of remote control engineering, at the University of Maribor, Faculty for Electrical Engineering and Computer Science, Slovenia.

Fig. 1. System components

424

Škorc, G. – Zapušek, S. – Čas, J. – Šafarič, R.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 421-433

Section 1 gives an overview of the used system components and presents there functionalities. A presentation of the host applications development, with description of the UDP (User Datagram Protocol) sender and receiver algorithm built within LabView software package is given in 2. Section 3 describes the development of the remote application, virtual model, and haptic device application programming interface (HDAPI) program built within C++ programming language. Section 4 shows the practical application and, finally, section 5 provides the conclusions of the work.

1 SYSTEM COMPONENTS The whole system can be divided on host and remote level as presented on Fig. 1. Host application is basically being developed for a purpose of MEMS assembly. Remote application extends functionality of the host application on that way that it simplifies programming of host application and allows remote programming (or monitoring) of it. Both application levels form closed application which can be used for different purposes. If such application is used as a production cell within some company, it eliminates need for “full-time” presence of expert engineer at the production line. Expert engineer can access host application remotely and remotely reprogram production process (or define production errors with monitor function). Small size of data packets, transmitted between host and remote application, make remote control possible even with mobile devices such as smart phone or PDA (under the condition that mobile device has access to public internet and preinstalled remote control software). Another possible use of presented system is in process of training of engineers or in the process of e-learning of students. Because remote interface includes functions for collision detection, real application cannot be harmed in the case of wrong preprogramming. Simulation functions built within remote interface make success of such training even better.

As already mentioned, the host level presents a real-time target application with nanorobotic cell (presented over a black line in Fig. 1).

It consists of a development computer marked as number 1, a real-time controller marked as number 2 and a target application marked as

number 3. Development computer is based on a Windows XP platform and supported by a LabView 8.5 software package. Primarily it is used for developing control algorithms and user interfaces, but can also be used for the execution of control algorithms. In this case we must take into account the limited frequencies of the program routines, which cannot achieve higher frequencies than 1 KHz. A better solution is the addition of a separate computer to the control system, which is then used only for the execution of control algorithms. Such a computer is presented in Fig. 1 and marked as number 2. A Real-time Desktop Target PC was built for this purpose. This dual processor PC is based on a LabView Real Time OS. The execution times of the control algorithms are drastically reduced, and execution frequencies of up to 1 MHz can be achieved. Our real-time target PC (Fig. 1, Number 2) is supported by a 7356 PCI motion controller card from National Instruments [7].

This card is capable to serve 6 independent axes (in our case we use 5 axes configuration), based on a positional feedback information’s from encoders and on-board PID controller. We have configured this card so that it acts as stepper driver, therefore, all axis outputs act as a STEP/DIRECTON outputs. An alternative configuration of this card, as servo driver, is also possible. On-board outputs and inputs are defined within TTL logical levels. Common robotic control functions (e.g. circular interpolation, linear interpolation, PTP move, SIN2 profiling and trajectory tracking) are supported in combination with LabView libraries.

The nano-robotic cell (Fig. 1, Number 3) is actuated using five linear piezo-motors produced

by PiezoMotor Upsala AB. According to the technical specification of the motor, movement in step lengths from 2 nm to 8 µm can be achieved.

With a factory delivered demo drive electronic motors can move with speeds of up to 12.5 mm/s (with our electronic of up to 35 mm/s) [8]. The motors are assembled so that two of them act as an X/Y manipulator and the other three as serving tables within the Z axis (Fig. 2). In closed loop control, positional feedbacks of the axes are achieved using linear encoders from NANOS instruments, which work on an electro-magnetic principle. Each encoder set consists of a magnetic scale and sensor electronics. The chosen system

Virtual User Interface for the Remote Control of a Nano-Robotic Cell Using a Haptic-Device

425

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 421-433

guarantees a resolution of 61 nm, with precision of  0.15% [9].

Fig. 2. Motors movement directions The built nano-robotic cell allows implementation of different micro and nano robotic-tools (Fig. 3). Figs. 3a and 3b present an influence of parasitic forces which can be used to handle micro – parts (with a use of special probe).

The problem in this case is to release the manipulated object. Temporally, we are using a special PZT gripper (Figs. 3C and 3D) which is used for gripping those objects of micro-scale size. It is placed at the top of the Y axis. This gripper was developed by the Fraunhofer-Institute of Reliability and Microintegration, Germany. It was fabricated by means of a UV-lithographic process and chemical wet etching technology from microstructurable photosensitive glass [10].

Using little changes in the basic form of the gripper, we achieved a gripper movement of approximately  10 µm (neutral opening of the gripper tip is 200 µm). [11]. Gripper is actuated with a piezo electric actuator and served by a PZT power module (DC voltages from -100V to +100V are used). Figs. 13e and 13f show manipulation with vacuum gripper similar to [12].

In this case the size of manipulated object is limited to the diameter of the used vacuum hose.

All manipulated objects presented in the Fig. 3 have varied in size from 80 to 200 µm.

Fig. 3. Different grippers

426

Škorc, G. – Zapušek, S. – Čas, J. – Šafarič, R.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 421-433

The lower level part of the application, presented under a black line in Fig. 1 is a remote application. Remote application can be used either for direct control of tasks on nano-robotic cell, or either for teaching of new tasks. It consists of a remote computer marked as number 4, remote application with included virtual model marked as number 5, and a haptic-device marked as number 6. A remote computer is the common notebook, based on a Windows XP platform, supported by a Microsoft Visual C software package, C++ programming language, and an OpenHaptics software tool from Sensable. It is used for developing and executing the remote application. The user datagram protocol (UDP), marked as number 7, is used as a basic communication protocol on the transport layer within internet protocol (IP) on the network layer.

The virtual VRML model marked as number 5 is also included in remote application. Because the remote user does not see the target application, it does not have a real feeling as to what exactly is happening to the target. There exists a risk of development hazardous. The model is used for animating the situation on the target application, which gives a realistic representative to the user who is remotely controlling the target. Model can be animated with online data acquired from the target application (encoder data extracted from transmitted data packet), offline data acquired from the simulation results (simulations are a function of the remote interface), or online data acquired from the haptic-device (built-in function of the remote interface). In order to provide a remote user with even more realistic representation about the situation on the target, the system uses a Phantom Omni haptic-device (Fig. 1, marked as number 6). It is the most basic for haptic-devices produced by Sensable Technologies Company. The chassis of the device is very compact and has two free programmable built-in buttons. The user takes over control of the device by moving the arm of the haptic-device over 6 degrees of freedom (X, Y, Z, roll, pitch and yaw). Three motors are built onto the X, Y and Z axes to give the user force-feedback information as a representation of friction or space limits. The positions of each axis are measured using encoders or potentiometers. It communicates with an IEEE-1394 interface. Both application levels can be used independent from each other or in combination with both of them.

2 DEVELOPMENT OF THE HOST APPLICATION As already mentioned, the host application uses a LabView 8.5 development platform, which simplifies programming of the target application.

The control algorithm and user interface were developed within this package. The control algorithm, which is based on adaptive bang-bang control, has already been presented in one of our previous papers [11]. This section will, therefore, concentrate just on those communication and user interface extensions added to the adaptive bangbang control method.

Fig. 4 represent a newly-developed target user interface which was built so, that it could be accessed over the internet. A real-time desktop target is supported by a web-server, which can be accesed by almost any web-browser, from almost any computer connected to the internet. The user interface gives detailed information about conditions on the nano-robotic cell. Data can be monitored as a current position in motor steps, encoder counts, a nano-meter move distance, a micro-meter move distance and a positioning error. Target positions for each axis separately can be set directly through web-application, or remotely acquired from a haptic-device. Selection between both trajectory input techniques can be done with a virtual button in the user interface. As soon as remote trajectory input is enabled, the user has to set communication parameters for the remote application (remote IP address and remote port) in order to establish connection between the haptic-device and the host. Microscopic objective in combination with CCD camera, installed at the target application, is used for monitoring the positioning processes on the nano-robotic cell.

The developed interface uses an UDP protocol for communication between host and remote application. In order to establish communication based on this protocol, both applications must have so-called UDP “receiver” and “sender” modules. Fig. 5 shows a diagram of the UDP receiver module installed on the host application.

The receiver has two modes. The diagram marked as letter A presents a situation when control using a haptic-device is disabled. The program algorithm takes over the target values from user interface and forwards them using those local variables, specific for each axis.

Virtual User Interface for the Remote Control of a Nano-Robotic Cell Using a Haptic-Device

427

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 421-433

Fig. 4. A Target user interface The diagram marked as B, shows a program state when control with a haptic-device is enabled. In this case a UDP connection is established and the first available data packet is received. Each received data packet consists of two separate sets of numbers, divided by a sepparation symbol. The first set of numbers gives information about axis number, and the second about a new target position for the same axis. In continuation of the program routine, this data packet is shifted twice, in order to extrapolate both data sets (Fig. 6).

Fig. 5. Receiver diagram

428

Fig. 6. Data set shift

Škorc, G. – Zapušek, S. – Čas, J. – Šafarič, R.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 421-433

The extrapolated axis number is used for controlling the case sentence, which orders a new target position for the proper axis variable. At the same time the axis number is forwarded to the sender module, where it is used as information about which axis current position shall be send to the remote application.

The diagram of the UDP sender module is presented in Fig. 7. It is built on the basis of recommendations from National instruments (LabView tutorial example). It has one mode of operation. The current position is acquired from the encoder as a number, saved in a local buffer, and sent to the user interface via an LabView indicator.

Fig. 7: Sender diagram

Virtual User Interface for the Remote Control of a Nano-Robotic Cell Using a Haptic-Device

429

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 421-433

The position number is converted into a string, because the UDP sender uses a string format for sent data packets. We have added case sentence to the basic algorithm which combines axis number and corresponding axis current position into the extended data string (oposite procedure as presented on Fig. 6). The extended data string corresponds to the form of received data packets on the following way. The same string is finally sent to the remote application using a UDP sender algorithm (UDP port is opened, data packet is sent, UDP port is closed).

Remote application receives positional information for one axis only at the same time. It is possible to configure the sender module so, that it sends information about all axes at the same time. We have constructed a test where the sent data package was formed so that positional information about the first axis was followed by positional information about four other axes, again separated by a separation symbol. In this case we had a problem of variation in position data package’s length, which is caused by changing of the single axis position data word’s length. Symbol which separates positional information for single axes was changing its position within the data package. Because our algorithm was always expecting the separation symbol at the same place within the data string, it has not referenced correct axis always. A more sophisticated method as shifting of the complete data package must be used for extracting single axis data, in this case. For this reason, we made a decision to transmit data packets for each axis, separately.

4. DEVELOPMENT OF THE REMOTE APPLICATION Our first goal when developing the remote application was to build a virtual model which would fully represent the real application. The main reason for building the virtual model was the wish to make remote connection between host and remote application more stable and faster. In Section 2 we have shown that the host application has the possibility to broadcast a live camera picture. This function works fine as long as we have a fast internet connection between host and remote application. In the case of a slow conenction (e.g. mobile internet connection from local mobile operator), broadcasted camera data

430

fulfills all available communication capacities, which provides a serious threat to transmission of control-data packets. Because we were using UMTS connection from our local mobile operator (using USB – UMTS modem ZTE MF636), we will focus only on mobile internet connections in continuation. Commonly three different mobile internet connections are used. The first one is called Global System for Mobile communications (GSM), which offers communication speeds of up to 14,4 Kbit/s. The second one is called General Packet Radio Service (GPRS) which offers communication speeds of up to 140,8 Kbit/s. The third one is called Universal Mobile Telecommunications System (UMTS) and is temporarily the most actual mobile connection.

Theoretically UMTS allows connections with speeds of up to 14 Mbit/s, but in practice are these speeds much lower (300 Kbit/s – 3,6 Mbit/s). In our experiments we have even noticed connections with speeds lower than 50Kbit/s. A huge amount of transmitted picture data can cause delay in transition of control data packets.

One of the easiest solutions to this problem is disabling the camera broadcasting, which certainly releases connection capacitates, but at the same time causes a lack of visual feedback information to the remote user. A few other already known solutions to this problem are reducing picture resolution (reduction in pixels number), reduction of picture color depth (e.g. from 24 bit to 8 bit), reduction of the number of acquired frames (e.g. from 50 Hz to 30 Hz) or use of a compression codec’s. We have defined minimum acceptable camera acquisition resolution as 640x480 pixels, minimum color depth as mono 8 bit mode and minimum refresh rate as 30 Hz. Size of the data packet was 1280 bytes in this case. For successful broadcasting of camera picture, stable connection with minimum speed of 307.2 Kbit/s is required. In comparison with control data packet which has in our case a size of 8 bytes, requires camera data packet much more connection capacities.

Because ratio between both packet sizes does not seems to be reasonable and all providers of mobile internet connections cannot guarantee stable connection within minimum requirements, we are proposing another possible solution. This is the use of a remote application with an implemented virtual model, which is animated using positional data received from the host

Škorc, G. – Zapušek, S. – Čas, J. – Šafarič, R.

StrojniĹĄki vestnik - Journal of Mechanical Engineering 56(2010)7-8, 421-433

application (control data packet). Internet connection is only used for transmission of positional control data packets (broadcasting of live camera picture was disabled, visualisation of production proces was done with animated model), and a fast internet connection is no longer needed. The implemented virtual model opens up a wide pallet of possible new functions that can be implemented in remote application. Some of them are collision detection, positional limitation, visualization, implementation of haptic-device etc. All these functions are executed locally on the remote computer, therefore, they do not require a good internet connection. We made a decision that we would use VRML language to develop the virtual model. Fig. 8 shows four different development stages. The model marked with the letter A was developed using a ProEngineer software package. The ProEngineer is one of the contemporary 3D modeling programs used for the development of new products. Because all of the mechanics for the nano-robotic cell was developed within this package, was the easiest way to create a VRML model, to use a built-in “export to VRML�

function. Tests on the so-created virtual model have shown that a lot of processor power is needed to animate the movement. A detailed look at the exported source code showed that the builtin function exports a model in the form of a separate object points. VRML language supports options to create a virtual model based on primitive shapes such as squares, cylinders and cones. The development of the model on the basis of primitive shapes requires less code.

Consequently, less processor power is needed to animate such a model. The virtual model programmed with primitive shapes is marked with letter B. Because the number of objects and with them connected mathematical functions was reduced, animations have run much faster on this model. From the beginning of the model development phase, was our goal to build as much as possible representative model of the host application. We decided to upgrade this model so, that basic objects of nano-robotic cell have been described with greater number of primitive objects and surface colors have been added to them. The result is presented under letter C.

Fig. 8. Model development phases

Virtual User Interface for the Remote Control of a Nano-Robotic Cell Using a Haptic-Device

431

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 421-433

Comparison with the real nano-robotic cell, as presented in Fig. 1 (marked as number 3), shows that a very good representative is achieved.

Unfortunately means increasing the number of primitive objects, a bigger need for processor power. To make remote control software useful even within mobile devices (e.g. PDA-s, smart phones) which have limited processor and memory capacities, we had to find a compromise between the realistic level of the model and processor usage. The model marked as letter C is an optimized model which is built as a remote application. It was tested on Windows based notebook presented in section 2 and HTC P3600 smart phone (haptic device was not used in this case).

The next step in the development phase was development of the remote user interface. It was developed under the Miscrosoft Visual C 6.0 software package, C++ programming language, and support of a MSDN library. The functionality of the haptic-device was established with a Phanteon v4.2.118 driver package and OpenHaptics v1.02.50 software tool. The VRML model was implanted and animated with the use of a VRaniML library. The whole process was done on a Windows XP based computer. Fig. 9 presents a screenshot of a user interface. The implemented VRML model can be seen in the

upper left hand corner. The rest of the interface is divided into 6 sections. These are a view section, a manual control section, a section for control using the haptic-device, a section for control within a PTP regime, a section for selection between on-line/off-line control, and a referenced positional section. The view section is used for selection of the most appropriate view aspect on the VRML model. Integrated virtual buttons allow us to translate or rotate the view aspect within the user’s given steps. ZOOM function is also available within this section. The section for manual control is used for moving separate axes.

Each axis has two virtual buttons. The first one is used for moving the axis one step forward and the other one for moving it one step backwards. The length of a motor step can be set individually. If the button is pressed continuously axis moves continuously in a give direction with a give step length. The section for control using the hapticdevice is used for enabling of the HID regime and selection of the axis, which will be controlled by the haptic-device. Axis selection function is also available directly on the haptic-device with the use of one of the buttons integrated on the hapticdevice. For safety reasons, HID regime still has to be enabled and confirmed within remote software. The number of the selected axis is shown within this section.

Fig. 9: User interface of the remote application

432

Škorc, G. – Zapušek, S. – Čas, J. – Šafarič, R.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 421-433

Moves are executed step by step as given in trajectory file. Any changes in position, within the described control sections are animated on the virtual model. The on-line/off-line section gives us a chance to simulate moves before they are transmitted to the real nano-robot cell. This option is given under the off-line regime, where any changes within any control section are executed as simulation on the virtual model. As soon as we enable an on-line regime, given changes are executed on the real application. This function is very helpful for testing some new trajectories. The last section is the reference position section, where we can set axes reference positions in certain specific cases, where default given references are unacceptable.

Because control user interface temporarly does not have ability to set user permisions (e.g. administrator or limited user), another user interface was built on the same basis, but only with the function of position monitoring of the target application. This interface uses C++ build receiver to animate the implemented VRML model.

All other fuctions available within the control user interface are disabled. This user interface is used only for a presentation purposes.

A complete source code for both user interfaces is very voluminous therefore, we focused only on that part which includes code for the haptic-device. Figure 10 shows an execution diagram of our program. The first program steps are initialization of the haptic-device, enabling the force generator, a schedule callback and the start of a scheduler. This is followed by initialization of the frame within which runs a socalled servo-loop. In order to ensure stable work of the control program loop, must this loop run with frequencies higher than 1 kHz. The first step within the servo-loop is acquiring position, which is followed by enabling integrated buttons. We check the state of the integrated positional button in continuation. When the positional button is pressed, a relative change of position is calculated. The servo-loop is closed by enabling a friction function and applying feedback force in case a collision has been detected. Finally we change the position of the nano-robot.

Fig. 10. Execution diagram of the remote application

Virtual User Interface for the Remote Control of a Nano-Robotic Cell Using a Haptic-Device

433

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 421-433

4 PRACTICAL USE OF A BUILT SYSTEM As we have already mentioned at the beginning of the paper we focused on the development of a nano-production cell. Until now we have prepared our application at a level which is capable of positioning of micro parts, with nano-meter accuracy. The final application is seen in the MEMS assembly process. For this purpose we still have to develop a suitable tools, which will allow us production of parts smaller than a few hundred micrometers. We are now developing a special gripper and special nozzle which will be used by first experiments of production. Paralely to the phase of developing the tools is our application used as an experimental object for the testing of new control techniques. Fig. 11 presents the temporary situation of our application.

Fig. 11. Temporary situation of the application Presented application was verified according to two different tests, performed with reference signals within nano and macro scale.

The first test was a step response test and the second was a trajectory tracking test. They have been performed with bang-bang, adaptive bangbang, fuzzy, adaptive fuzzy, polynomial and adaptive polynomial control techniques.

Verification was done on a base of the size of a position overshot and an average positioning error. The best result was achieved using polynomial control technique where the size of the overshot was within 61 nm, and the size of average positioning error within 3 nm. Detailed test results are given in [13]. Common to all tested techniques was, that they had produced always the same result either the system was controlled locally, either control was done remotely (using LAN or UMTS mobile network).

Because of the small size of the transmitted 434

control data packet (8 bytes) this result was expected.

5 CONCLUSION In this paper we have described the development of a remote application with an implemented VRML virtual model. The model was animated with real-time positional control data acquired from host application, and from this aspect used as a replacement for live camera pictures. The broadcasting of live camera pictures over the internet demands a good internet connection, which is not always available. The easiest way to establish stable remote control over the internet is to disable broadcasting of the camera picture. In this case the remote user loses visual presentation of a situation at the target application, which is very important feedback information in nano-technologies. We have solved this problem so, that we have developed a remote application where the lack of the live camera picture is replaced by an animated model, for which animation algorithms are executed directly on the remote computer. In this case visual feedback is established by the use of a processor on the remote computer and not on the host computer, as it is common. Our solution shows that the same model in continuation can be used for collision detection, off-line simulations, on-line monitoring, etc. In our work we have built an application using the most common software packages and protocols what makes it easy to reconfigure it for use on other applications. As basic protocol on network layer we have used IP and as a basic protocol on transport layer we have used UDP. This combination makes our application useful with different desktop computers and even with some mobile devices (e.g. PDA-s, smart phones). Small size of transmitted control data packet (8 bytes) allows stable remote control even with the slowest GSM internet connection. Fast internet connection is no longer needed. Although our project already gave good results, there are still open issues which will have to be solved in near future. As example, we have chosen UDP on transport layer because of its easy integration into different software packages, although it is known that it does not support delivery information about sent data packets. Any errors in data transmission cannot be recognized. In our experiment we haven’t

Škorc, G. – Zapušek, S. – Čas, J. – Šafarič, R.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 421-433

noticed problems connected with this weakness, but we must mention, that we have only one such system, communicating on a given communication port, reserved especially for it.

We suppose that this weakness would cause troubles in case of multiple systems communicating over the same port. Solution for this problem is implementation of a more secured TCP protocol on transport layer. One of the future tasks in our work will be to examine possibility of integration of this protocol. Parallel with that we are focused on development of different tools for MEMS production.

[6]

[7]

[8] 6 ACKNOWLEDGEMENT Operation part financed by the European Union, European Social Fund.

[9] [10]

7 REFERENCES [1]

[2]

[3]

[4]

[5]

Stolle, C., Fatikow, S. (2007). Control system of an automated nanohandling robot cell. 22nd International Symposium on Intelligent Control, 1-3 Oct., p. 664-669.

Fahlbusch, S., Fatikow, S. (2001).

Implementation of self-sensing SPM cantilevers for nano-force measurement in microrobotics. Ultramicroscopy, vol. 86, no.

1, p. 181-190, Zuobin, W., Fatikow, S., Shizhong, S., Ming, Y. (2007).

Robotic nanoassembly.

International Conference on Mechatronics and Automation, 5-8 Aug., p. 422-427, Fahlbusch, S., Shirinov, A., Fatikow, S. (2002). AFM-based micro force sensor and haptic interface for a nanohandling robot.

IEEE/RSJ International Conference on Intelligent Robots and System, vol. 2, p.

1772-1777.

Lim, T., Ritchie, J.M., Corney, J.R., Dewar, R.G., Schmidt, K., Bergsteiner, K. (year).

Assessment of a haptic virtual assembly system that uses physics-based interactions.

[11]

[12]

[13]

International Symposium on Assembly and Manufacturing.

Iglesias, R., Casado, S., Gutierrez, T., Garcia-Alonso, A., Yap, K.M., Yu, W., Marshall, A. (2006). A peer-to-peer architecture for collaborative haptic assembly. International Symposium on Distributed Simulation and Real-Time Applications, 2-4 Oct., p.25-34, National Instruments, Motion controller 7356 datasheet, Retrieved on 05.01.2009 from http://www.ni.com/pdf/ products/ us/735x.pdf Piezomotor Upsala AB (2003). PiezoLEGS data and user instructions, 3rd edition, p. 315, Retrieved on 05.01.2009 from Nanos instruments web page: http://www.nanosinstruments.de/ Keoschkerjan, R., Wurmus, H. (2002). A novel microgripper with parallel movement of gripping arms, Actuator, 8th International Conference on New Actuators, Bremen, June 10-12, p. 321-324, Škorc, G., Čas, J., Brezovnik, S., Šafarič, R. (2009). Position control with parameter adaptation for a nano-robotic cell, temporarily reviewed by Strojniški vestnik Journal of Mechanical Engineering, ???

Zesch, W., Brunner, M., Weber, A. (1997).

Vacuum tool for handling microobjects with a nanorobot, ICRA`97 IEEE Int. Conf. on Robotics & Automation, Albuquerque, USA, p. 1761-1766. Škorc, G., Čas, J., Šafarič, R. (2010).

Adaptive positioning of MEMS production system with nano - resolution, temporarily reviewed by Journal of Intelligent Automation and Soft Computing, ???

Virtual User Interface for the Remote Control of a Nano-Robotic Cell Using a Haptic-Device

435

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 436-446 UDC UDK 629.734.33:351.814.343

Paper received: 10.12.2007 Paper accepted: 02.07.2010

Analysis of the Sailplane Final Approaches Performed by Cosine-Law Speed Variations Zoran Stefanović - Ivan Kostić* University of Belgrade, Faculty of Mechanical Engineering, Serbia High lift-to-drag ratios of the contemporary sailplanes make them the most energy efficient flying vehicles. On the other hand, this capability may become their serious disadvantage during the landing, if their aerodynamic deceleration devices become inoperable in flight. Not being able to dissipate the excess energy quickly when close to the ground, they may fly over the available landing ground and finish up in front of the obstacles, with still too much energy to land and not enough to fly over them.

Beside the sideslipping flight in final, where energy is dissipated through the increased sideforce drag, another solution to this problem has been offered in a number of papers. By numerical analyses they have shown that landing distance in such cases could be minimized using rather complex oscillating flight paths in vertical plane. Although relevant distance reductions could be achieved through them, performing such paths in practice would require exceptional piloting skills. Instead of that, in this paper much simpler approach profiles have been analyzed, based on two types of cosine speed variations with constant periods and amplitudes, which could be flown by pilots of average flying experience. After establishing a quick convergence algorithm, numerical solutions for several typical cases, belonging to two general speed variation types, have been presented. The same initial and terminal reference energy states have been used. Although the distance reductions are smaller than obtained by distance-minimizing techniques, operational simplicity of presented techniques and some specific advantages prove them valuable within this category of problems. ©2010 Journal of Mechanical Engineering. All rights reserved.

Keywords: sailplane, final approach, inoperable spoilers, cosine speed variations 0 INTRODUCTION Modern sailplanes have very high glide ratios, and due to that they are able to fly very long distances without a power plant, loosing proportionally small height at the same time. This makes them extremely energy-efficient flying vehicles. Since the available landing grounds (runways, or sometimes just long enough countryside fields) are often limited by obstacles on both ends, sailplane pilots generally perform their final landing approaches at much steeper angles than during the gliding flight. This is normally done with spoilers extended, which increase drag, partially reduce lift and increase the dissipation of energy. Quick reduction of height is achieved without remarkable increase of the sailplane's speed. But the final approach can be one of the most critical phases of a sailplane flight if, for any technical reason, spoilers or other available aerodynamic deceleration devices become inoperable (cases which do not happen often, but are known in practice). With the nose

436

pointed rather steeply down to clear an obstacle and aim for the beginning of the landing ground, a sailplane will very rapidly gain too much speed and simply "refuse to land". Forcing it down to the ground at too high speed will make it bounceoff and, at worst, may lead to a crash landing. On the other hand, if patiently waiting for a sailplane to slowly decelerate above the ground, a pilot may fly over the available landing area and finish up facing the obstacles, neither being able to land in front of them, nor to fly over them.

One of the known operational techniques that can be used to face this problem is the sideslipping during the final approach phase.

During such an intentionally uncoordinated flight, additionally generated sideforce will increase the overall drag. Principally like with spoilers, this drag component will also dissipate additional quantity of energy and shorten the approach distance. But this technique requires a certain amount of skill. For example, in case of a not too experienced pilot forced to land on a narrow countryside field, improper estimation of the

* Corr. Author's Address: University of Belgrade, Faculty of Mechanical Engineering, Aeronautical Department,Kraljice Marije 16, 11120 Belgrade 35, Serbia, ikostic@mas.bg.ac.rs

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 436-446

actual flight direction while watching over his shoulder in this maneuver may finally place him in front of a wrong field, with no engine to help him go around and correct the error.

Beside this classical technique, in a certain number of papers the oscillating final approach patterns without sideslipping, performed in a vertical plane, have been considered as another potential option with an aim to minimize the landing distance in case of the aerodynamic decelerating devices failure. In order to emphasize a rather high complexity level of such kind of calculations, one of them performed for the Vuk-T sailplane [1] will be described very briefly. It treated the problem of minimizing the landing approach distance as an optimal control problem, where the initial and the terminal states were based on recommendations from [2]. In these papers the lift coefficient variation was established as a variable of the control function u (t ) according to [3] and the maximum lift coefficient value of 1.78 for the Vuk-T. Since the total time of the final approach, originally denoted as tk , is initially unknown, calculations were done in normalized time  , introducing another control parameter  , where t    , 0  t  tk , and 0    1 . Path for the minimum landing distance was obtained through an iterative calculation process, were the point was to determine such function u (t ) and an  that will minimize the so called performance index I , which is subjected to the dynamic, initial and

terminal state constraints. Index I included the integral interior penalty functions [4] for the minimum speed and the height constraints, combined by the empirical fixed constraint factors. The problem was solved using a gradient projection algorithm [5], which incorporated conjugate directions of search for a rapid convergence of the solution. Those calculations were done in Fortran 77 in double precision mode. Flight path of the so minimized approach distance and the appropriate lift coefficient are shown in Fig. 1.

Advantages of such an approach are the effective reduction of the approach distance and the fact that, during this phase of flight, the nose of a sailplane is permanently pointed in the direction of intended landing area. On the other hand, this obviously quite complex calculation gives as a result an unevenly oscillating flight profile. Such approach path could be rather difficult both for memorizing and for performing under the operational flight conditions. If several errors were accumulated during such final approach (and hoping that the sailplane would not be accidentally stalled in the final stages, flying at very small speeds in the vicinity of the ground), the sailplane could land further from the initially estimated touchdown point. If a pilot is forced to land on a rather short field, this might be equally critical as a potential directional misjudgment, mentioned in case of the sideslip landings on narrow fields.

Fig. 1. Example of a rather complex approach profile for the minimized approach distance and the corresponding lift coefficient, calculated for the Vuk-T sailplane (from [1])

Analysis of the Sailplane Final Approaches Performed by Cosine-Law Speed Variations

437

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 436-446

As a contrast to previously described method, the aim of this paper is to investigate very simple periodical flight path changes in the vertical plane without sideslipping, and the final approach distance reductions that could be achieved through them, also assuming that spoilers are inoperable. The primary goal is that such flight paths could sufficiently accurately and easily be flown by the pilots of average flying skills, using just two instruments that are always on board: a speed indicator and a stop or a wrist watch. The predefined cosine harmonic approach speed variations, divided in two general categories, all with constant amplitudes and periods, have been selected as a good mathematical resemblance of the pilot's natural control inputs in attempt to evenly "pump" the flight speed up and down. In order to preserve compatibility with the previous example, the Vuk-T sailplane has also been selected for actual calculations, subjected in all cases to the same initial conditions and input parameters as in [1].

Another very important aim of this paper is to keep the calculation model as simple as possible, but efficient and sufficiently accurate for the required purposes. It is clear that the potential end users of this calculation model would mostly likely be the amateur sailplane pilots - who are generally not experts in advanced programming, rather than highly trained engineers. Thus, if these calculations are kept simple enough, they could be incorporated in some of the available commercial software, like spreadsheet programs, which do not require highly sophisticated coding or recompiling. In that case, a pilot with only the general knowledge of informatics could perform even major program editing, for example, by experimenting with the different laws of speed variations, or making combinations of several simple paths in one approach, etc. Necessary input data could be obtained from the sailplane manufacturers, operation manuals supplied with the sailplanes, or from other available sources.

Such calculations should help pilots to define parameters of several possible final approach paths for the sailplane types which they fly, considering possible field position and length, general obstacle distribution, etc., and estimate in advance the approach distances in case that spoilers become inoperable. Memorizing the speed amplitudes, periods and expected approach

438

distance reductions, chances of making misjudgments considering the direction or distance in final approach should be much smaller. On the other hand, the price for increased safety using the techniques that will be presented in the following chapters is that the approach distance reductions will most probably be smaller than in case of the other two mentioned techniques.

1 ALGORITHM OF CALCULATIONS The sailplane configuration in final approach for these calculations is gear-down and spoilers-in. According to the flight test measurements performed on the Vuk-T sailplane prototype, at the Flight Test Center VOCBatajnica (which just slightly differed from the production sailplanes such as the one shown in Fig. 2), polar for this configuration is defined by equation:

CD  0.01756  0.0095 CL  0.021CL 2 ,

(1)

where CD and CL are drag and lift coefficients, respectively.

Fig. 2. Vuk-T of the Ljubljana aero-club As in [1], in this paper it will also be assumed that the nominal mass of the sailplane in flight is m  320 kg, and that the air density is   1.225 kg/m3. Aerodynamic wing area of this sailplane is S  12 m2. Using these values and Eq. (1), it can be easily calculated that maximum glide, or L / D ratio for this sailplane configuration is ( L / D) max  34.59 at the speed of V  77.99 km/h (values for gear-up configuration are different; also, this velocity is additionally influenced by the assumed mass). For the default

Stefanović, Z. – Kostić, I.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 436-446

glide regime, which will be used for comparison with approaches based on cosine velocity variations, the rounded value of V  80 km/h will be applied, for which L / D  34.52 .

Fig. 3. Forces and velocity components in descent Neglecting the rotation dynamics and assuming that the wind speed is equal to zero, equations of motion [6] for the purpose of these calculations (see also Fig. 3) can be written as: dV m X   D cos   L sin  (2) dt m

dW   mg  D sin   L cos  dt

(3)

dX  VX dt

(4)

dH W dt

(5)

W VX

(6)

tan  

where: 1 L  CL  V 2  S 2

(7)

1 D  CD  V 2  S (8) 2 In this paper, the speed variations in final approach will be assigned as inputs that can be divided in two general categories, depending on which of the following two equations is applied:  2 V  VAV  V cos   T

 t 

(9)

 2  V  VAV  V cos  t  T 

(10)

In these equations VAV represents the average speed, V is the half-amplitude, T is the period, while time t is the independent variable. For example, if we substitute VAV  85 km/h, V  5 km/h and T  17 s in (9), the initial speed at t  0 s will be V  80 km/h, at t  8.5 s speed will increase to V  90 km/h, while at the t  T  17 s it will be V  80 km/h again. In contrast to that, the application of (10) will lead to the initial speed decrease.

In order to achieve quick convergence of the solution, calculations have been performed in several iteration steps, with complexity and accuracy levels increasing gradually from one step to another. Also, in calculations aimed for engineering and practical purposes, small angle approximations can be applied for values of the glide path angle   10 o , but obtained solutions must be finally substituted into the full equations for the verification of the achieved accuracy.

Thus, with  expressed in radians, Eqs. (2) and (3) can be written as: dV m X  D  L   (11) dt dW  mg  D    L (12) dt It should be noted that all variables on the right hand-side of Eqs. (2) or (3), or (11) and (12), with the speed changing according to an assigned law, will also be time dependant (except the sailplane weight m  g ).

In the first iteration step, the lift coefficient variation with time along the flight path is initially estimated from the equation: 2 m  g CL (t )  (13)   V (t ) 2  S m

using Eq. (9) or (10) to define speed changes.

After that, the time dependant drag coefficient is calculated using Eq. (1). Both in this and the following iterations, a time step of t  0.1 s for the numerical analyses proved to be quite satisfactory. It should be noted that the Eq. (13) is actually obtained from (12), omitting the product D   and assuming that dW / dt  0 . In usual

Analysis of the Sailplane Final Approaches Performed by Cosine-Law Speed Variations

439

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 436-446

sailplane descents, products D  sin  are about 1000 times smaller than L  cos  , thus omitting D   does not affect the accuracy noticeably. On the other hand, for here applied cosine speed changes, the assumption dW / dt  0 is not true, but has been taken as an intentional "sacrifice" in the initial stage of the calculations.

Lift and drag forces are then calculated using Eqs. (7) and (8). To a first approximation, we can say that dVX / dt  dV / dt . Since the variation of V (t ) is a known differentiable function, dV / dt can be obtained both numerically and analytically (doing it both ways and comparing the results might be one of the verifications whether t is selected adequately).

The initial estimate of the flight path angle  can now be obtained directly from (11): m (dV / dt )  D(t )  (t )   (14) L(t ) Knowing these values, components are determined as: VX (t )  V (t )  cos  (t )

the

W (t )  V (t )  sin  (t )

velocity (15) (16)

In the sense of numerical calculations, their time derivatives at the ith time step are obtained as:  dVX  (Vi  Vi 1 )  (Vi 1  Vi )  dt   2  t  i

(17)

 dW  (Wi  Wi 1 )  (Wi 1  Wi ) (18)  dt   2  t  i In the second iteration step, the lift coefficient Eq. (13) is upgraded, this time including values obtained from (18): 2  m  ( g  dW / dt ) CL (t )  (19)   V (t ) 2  S

while Eq. (14) is upgraded using the values obtained by Eq. (17): m (dVX / dt )  D(t )  (t )   (20) L(t ) Lift, drag, and the velocity components with their derivatives are then recalculated applying the same algorithm as in the first iteration step, but including the refined values obtained using Eqs. (19) and (20).

440

In the third iteration step the whole procedure is repeated, this time using dVX/dt and dV/dt from the second step, etc; this calculation process can be repeated as many times as necessary, until the desired accuracy is achieved.

The X(t) and H(t) coordinates, which determine flight path profile, are calculated by numerical integration of the VX(t) and the W(t) from the last iteration step with respect to time (coming out from Eqs. (4) and (5)), using initial conditions X(0)=0 m and H(0)=50 m for all cases considered in this paper. The length of the flight path P(t) is obtained by numerical integration of the total velocity V(t).

To quantify the obtained accuracy, results from the last iteration step were substituted in full Eqs. (2) and (3). Differences between the left and the right-hand sides, calculated at each time step, were then compared with the calculated drag force in case of Eq. (2) and lift force in case of Eq. (3). Limit for so defined relative errors, which could be accepted for practical considerations, was established at the order of 1% or smaller. The presented algorithm has shown very high convergence rate, since practically all analyzed cases with the cosine speed variations have fulfilled this requirement after only three iteration steps. The only exception was case denoted as "I2" (see next chapter, Fig. 7), where the fourth step was introduced to reduce the maximum relative error from 2.2% to 1.2% in Eq. (2). Since the differences between the calculated X and H values in the third and fourth step in this case were of the order of centimeters, it has been assumed that any further accuracy improvements would not be necessary.

Terminal state is reached at the H  1 m and V  72 km/h. Thus, beside the final approach, the round-out phase and the hold-off phase (Fig. 4) also had to be calculated for the default case and case I-3 (next chapter, Figs. 5 and 8). For all other approaches the velocity amplitudes and the corresponding periods have been selected in a way that the round-out phase is an integral part of the final approach path.

For usual landings ((A) in Fig. 4), the round-out phase is often modeled as a circular arc i.e. R  const., through which the approach speed VAP changes are practically negligible. On the other hand, changes of the load factor n  L /(m  g ) are not.

Stefanović, Z. – Kostić, I.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 436-446

Fig. 4. Usual landings (A) consist of: (1) the final approach with  ≈ const., (2) the round-out phase, (3) the hold-off phase and (4) the landing run (not considered in this paper); in here analyzed cosine approaches (B), phase (2) is integral part of the phase (1) for which  ≠ const.

Radius of such modeled round-out phase can be determined from the equation: R

2

VAP 1  g n  cos 

(21)

where n is the load factor at the end of this phase. For both mentioned cases, n  1.05 has been assumed. Total variations of the height and the horizontal distance through the round-out phase are: H   R (1  cos  ) (22) X   R sin 

(23)

The hold-off phase has been modeled in the same way for all analyzed cases, through which the speed is gradually reduced to the intended touchdown value of VT  72 km/h. (For m  320 kg, the Vuk-T's stalling speed is Vstall  55.7 km/h; it should be noted that, for many sailplanes, decelerating to Vstall would lead to the tail-first touchdowns, which can cause damage to the structure). Although under operational conditions there is usually a small loss of height through this phase and strictly speaking vertical velocity component W  0 , for practical analyses the equation of level flight with center of gravity at a constant average height H  1 m can readily be used. Substituting   0o in Eq. (2), it becomes: dV dV m X  D  m (24) dt dt and thus:

  V 2  CD  S dV  2m dt

(25)

Initial condition is defined by V at the end of round-out phase (in the next chapter parameters at this point will be denoted using the symbol "(*)"), and for each consecutive time step speed reduction is calculated using (25). The new CL for the reduced speed is obtained from the equation of level flight, while the corresponding CD is calculated using Eq. (1). The calculation continues until VT  72 km/h is reached. Distance X flown in this phase is obtained by the integration of speed with respect to time, and for this phase horizontal distance is equal to the path length, X  P . The ground roll after touchdown has not been considered because, after the common terminal state parameters have been reached, ground roll for all analyzed cases would be the same for the same terrain categories and qualities.

2 SELECTED CASES AND RESULTS It is usual to assume that, before commencing a final approach, the sailplane would most probably fly at the speed close to that of the maximum glide ratio. As derived in the previous chapter, this value for the Vuk-T in the gear-down and spoilers-in configuration and m  320 kg is about 80 km/h. The final approaches are usually commenced at about H  50 m above the terrain, so these values, similarly as in [1], will define the initial (energy) state for all considered cases. Also, based on the

Analysis of the Sailplane Final Approaches Performed by Cosine-Law Speed Variations

441

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 436-446

sources mentioned in the introduction, the speed of 20 m/s, i.e. 72 km/h for the same sailplane mass, and the center of gravity height of 1 m above the ground, have been selected to define the terminal (energy) state for all cases.

Fig. 6. Case I-1

Fig. 5. Default path - steady glide As a reference, or a default case for all comparisons, a steady descent with linear approach path at a constant speed of 80 km/h has been selected (Fig. 5). On the other hand, it is well known that while flying at the speed of the best glide ratio, a sailplane will achieve the longest possible range from a certain height (which is, by the way, quite opposite from the pilot's intentions while attempting to land on a short field and the spoilers are not in function). At both higher and lower speeds, the range will always be smaller. So a question may arise - why select the best glide ratio speed for a reference, when all other constant speed approaches used as a reference would give steeper glide paths and more rigorous critics of here considered paths with cosine speed variations? Just partial, but hopefully sufficient answer could be that at the same height of H  50 m but different initial speeds, a sailplane will have the same potential energy, but different kinetic energies, or in other words, the different flight histories before reaching 50 m height. If we want to quantify the potentials of a certain flight profile to dissipate energy more effectively then some reference profile, both (1) their initial energy states, on one side, and (2) their terminal states on the other, must be the same for the comparison purposes.

While many other reasonable reference speed choices can satisfy the second requirement owing to the deceleration (as much as necessary) within the hold-off phase, a steady reference path with the approach speed of 80 km/h is the only one that satisfies the first requirement.

442

Fig. 7. Case I-2

Fig. 8. Case I-3 Several typical cases selected for the presentation and analyses in this paper are shown in Figs. 6. to 10. Those within the category "case I-..." are based on the application of Eq. (9), while those named as "case II-..." are obtained using Eq. (10). Numerical example from the previous chapter, with speed variations between 80 and 90 km/h and T  17 s, corresponds to the flight path shown in Fig. 6. The reference path is marked with "(A)", while all paths with cosine speed variations are denoted as "(B)". The tree symbols represent standard 15 meter obstacles.

The most important results, necessary for the discussions in the next chapter, are presented within the figures. Values X (*) and P (*) are the horizontal distance flown and the actual flight path length at the end of the round out phase,

Stefanović, Z. – Kostić, I.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 436-446

respectively; DAV is the distance-averaged drag force along the cosine final approach segment, not including hold-off; n shows the range of the load factor variations in cosine approaches;  denotes the extreme path angle variations, etc.

Parameters of the cosine law speed variations, influenced by the sailplane type and its actual mass in flight, have been carefully selected as easy-to-remember rounded numbers considering the speed variations and the periods. They also had to satisfy the terminal height conditions with the round number of cycles for the paths based on Eq. (10), and round number +1/2 cycles when (9) was applied. In cases II-1 and II-2, only for theoretical considerations, the periods of the order of T  20 were spread to the first decimal accuracy in order match the terminal 1 m height with some ± 5 cm accuracy obtained in other cases.

Fig. 9. Case II-1

spreadsheet version has also proven to be highly functional.

3 DISCUSSION OF THE RESULTS Let us first comment an issue which considers the oscillatory approach path profiles.

Looking at the Figs. 6 to 10, it may seem that the length along an oscillating path between the two points in final approach should be substantially longer than the straight-line distance between them. Thus, the first impression might be that just by shrinking a straight line into an oscillating path profile could, by itself, contribute to the decrease of the landing distance. On the other hand, relatively small differences even between the X (*) and the P (*) shown in these figures, suggest quite opposite. It should be noticed that, for the clarity of the presentation, scales for X and H had to be remarkably different (paths with  shown in figures would correspond to the L / D  1.6 ), thus the true appearances of the sailplane paths are largely distorted. That could be understood better by comparing Figs. 8. and 11, or by "manually" checking the default path length (neglecting the round-out phase curvature):

P(*)  17062  492  1706.7 m, while X (*)  1706.0 m ̶ see Fig. 5. Thus the oscillating path profiles, by themselves, can not cause any relevant X distance reductions in here presented cases (more detailed explanation, which can be generalized to all such flight paths, can be found in [7]), so attention should be focused primarily on the drag force.

Fig. 11. True appearance of the case I-3, with H and X coordinates drawn in the same scale Fig. 10. Case II-2 Although the calculations considering here presented subject have initially been done using a custom written Fortan code, a parallel effort has been made to obtain the results in one of the most commonly used spreadsheet programs, the MS Excel. Owing to a reasonable simplicity and high convergence rate of the applied algorithm, the

Dissipation of energy in sailplane descents is achieved through the work done by the drag force along the flight path. Since the prescribed initial and terminal energy states for all analyzed cases are exactly the same, an increase in the drag force with respect to a steady approach case must induce a proportional decrease of the flight path length P and consequently the decrease of the horizontal distance X between the initial and the terminal point.

Analysis of the Sailplane Final Approaches Performed by Cosine-Law Speed Variations

443

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 436-446

Table 1. Review of the most important landing profile parameters 1

Case

2

3

DAV [N]

X  X A  X B [m]

4

V [km/h]

5

6

( L / D)

T [s]

90.9 0 0 0 / Default case 91.5 26.4 - 10 - 0.9 (-2.6%) 19.9 Case II-1 93.1 56.7 + 10 - 1.8 (-5.2%) 17 Case I-1 94.1 78.9 + 10 - 1.8 (-5.2%) 7 Case I-2 95.2 96.0 - 20 - 5.3 (-15.4%) 20.6 Case II-2 96.6 (*103.9) 101.8 + 30 - 8.5 (-24.6%) 26 Case I-3 * Actually, the value denoted as DAV ' is the true generator of X for case I-3, since its second part is equivalent to the default path, just shifted to the left.

For all cases the varying drag force has been integrated along the final approach and the round-out phases with respect to the path length, and then divided by the total path length of these two landing segments. This way, the distanceaveraged drag forces DAV in approach have been obtained. The previously mentioned principle can be confirmed if these values are compared with the achieved distance reductions X , as shown in Table 1.

From Table 1 it is also obvious that the larger speed deviations V from the initial state velocity of 80 km/h generally correspond to higher X values. Remembering that the selected initial speed is practically the maximum glide ratio speed for the given configuration and mass, any diverging from it must cause the decrease of the glide ratio (Table 1, col. 5). This causes increased drag for the same amount of lift and, as a final consequence, shorter flight path. (If some other speed is selected for the initial state, then varying the velocity in the opposite "direction" from the ( L / D) MAX speed will lead to the decrease of the approach distance, and vice versa). Knowing that, a logical question might be - why the flight path should be oscillating at all, when a much simpler procedure, based on the continuous speed increase in final approach (let us say from 80 to 90 km/h), will decrease the L / D ratio and probably also lead to the decrease of the landing distance? Such an approach has been analyzed in [7], setting the period to T  120 s and using just half of the cycle based on Eq. (9). After the round-out and hold-off phases had been added, the X reduction of some 33 m was achieved, compared to the default path. On the other hand, in cases I-1 and I-2, where speed

444

variations were also between 80 and 90 km/h, the distance reductions were 56.7 m ( T  17 s) and 78.9 ( T  7 s), respectively. It implies that, for the same speed amplitude, shorter periods (more oscillations) generate larger landing distance reductions.

In an attempt to explain this particular phenomenon, let us compare load factor variations about the value n  1 in Figs. 6 and 7.

We can see that they are not symmetrical, i.e. the increase of n at the local path minimums is slightly larger than its decrease at the local path maximums (this applies for all analyzed cases).

Also, the shorter period of case I-2 induces larger overall variations of the n values then in case I1. Since larger load factors correspond to larger lift forces and consequently larger drag, and vice versa, the applied kind of speed variation generates the average drag increase through the asymmetrical load factor variations. Beside that, if the overall load factor variations are larger due to shorter periods, the average drag increase should also be higher, and thus case I-2 gives 22.2 m larger landing distance reduction than case I-1, although their speed amplitudes are the same.

A general conclusion might be that higher speed amplitudes (i.e. departing from the best glide ratio speed towards those that correspond to smaller L / D ratios) and shorter periods of speed variation are actually the two influence parameters that both contribute to the landing distance reduction. In here presented final approaches they inherently go together, and they must be combined carefully. Too large speed amplitudes with too short periods can be very unpleasant for the pilot, locally overstress some parts of the sailplane structure and may generally

Stefanović, Z. – Kostić, I.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 436-446

be very dangerous in the vicinity of the ground.

Such combinations should be based either on large amplitudes and long periods, small amplitudes and short periods or moderate amplitudes and moderate periods.

Let us now consider some other important practical aspects of here presented approach profiles. Cases II-1 and II-2, based on initial speed decrease, enable avoiding the 15 m obstacles at practically the same X distances of about 1200 m from the starting point, as in the default case. This is their advantage when obstacles are close to the beginning of the landing ground, since for all cases based on the initial speed increase (I-1, I-2 and I-3) this distance is some 100 ÷ 200 m smaller. On the other hand, the distance reduction obtained by case II-1 of only 26.4 m might be categorized as "too much trouble about nothing" (quite long period and very small departing from ( L / D) max ), so this case is of rather small practical significance. Quite opposite to that, the case II-2 with almost the same period, gives the second best landing distance reduction of 96 m, owing to the substantial drop of L / D at 60 km/h. It should be noticed that this speed is only some 4 km/h higher than the sailplane's stalling speed, and flying this approach would require caution.

Paths of the cases I-1 and I-2 give moderate X values of 56.7 m and 78.9 m, with respect to the maximum achieved X  101.8 m in case I-3. Although the ground roll phase is not explicitly analyzed in this paper, it should be mentioned that these two profiles have an advantage over the other path profiles considering this aspect. Namely, in case of the high emergency landings on short fields, one of the usual procedures is to try to force the sailplane to the ground at higher speed than nominal (72 km/h in our case) and then start using wheel brake, as a very efficient energy dissipating device. For the Vuk-T sailplane it has been estimated that the speed of 90 km/h could be acceptable top speed limit for such a procedure to be successfully performed. Due to their profiles, cases I-1 and I-2 could enable such forced landings at 90 km/h some 200 meters earlier the other presented cases, supposing that the 15 m obstacles are not further than approximately one kilometer from the starting point of the final approach.

The largest distance reduction of 101.8 m has been achieved in case I-3, which looks in a

way like a simplified version of the minimized flight path from [1], shown in Fig. 1. It combines a curved path generated by cosine speed variation with large speed amplitude and long period, and a straight path similar to the default case. It is clear that this is just one of many possible combinations, where in this paper the constant speed of 80 km/h for the second portion of the approach has been used intentionally to present the pure contribution of a single path oscillation with such speed amplitude on quite noticeable approach distance reduction. It natural that some other profile choices for the first and second part of the path could give even larger X values.

The distance reductions of the order of 70 ÷ 100 m, compared with the here considered total X distances of about 1.8 km, may not seem very spectacular at the first glance. On the other hand, it must be remembered that sailplane pilots sometimes have no other option but to land on a narrow and quite short countryside field surrounded by trees, power lines, telephone poles, houses, ditches, rivers, etc., and loosing the chance to extend spoilers in such situations makes the last minutes of the flight very critical. In such cases, using some of the relatively simple and quite safe procedures, such as examples given in this paper, to shorten the final approach for a distance which is close to or equal to a football field length, can make a substantial difference between the successful outcome and a disaster.

Also, a pilot must not forget to extend the landing gear - not only because it is normal to land a sailplane with the gear down (except on very rough terrains), but also because the extended gear on modern sailplanes causes a drag increase which is far from negligible. If a sailplane pilot can not use spoilers when the approach distance shortening is an imperative, any source of additional drag is extremely valuable.

4 CONCLUSIONS In this paper the two categories of simple cosine speed variations in final approach have been analyzed, as possible ways to reduce the approach distance in cases when spoilers become inoperable. The first category implies that the speed initially increases, and the second that it initially decreases from the reference staring value of 80 km/h at the height of 50 m. For actual calculations the Vuk-T sailplane has been

Analysis of the Sailplane Final Approaches Performed by Cosine-Law Speed Variations

445

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 436-446

selected. The input values for periods and speed variations have been chosen as rounded and easyto-remember numbers for practical use, and applied in different ranges within the two assigned cosine laws. Results were compared with the landing distance of the steady reference path flown at the same speed as at the initial state.

Presented examples have been carefully selected to show the influence of those input parameters on possible landing distance reductions, ranging from very small, of the order of 20 m, over moderate 50 ÷ 70 m reductions, to more than 100 m. For all presented cases the initial and the terminal energy states were the same. The oscillating path profiles, simply because they are curved, do not contribute remarkably to the landing distance reduction in any of the treated cases. The analyses have shown that the larger achieved landing distance reductions were actually proportional to the larger speed amplitudes and shorter periods of oscillations, both contributing to the increased energy dissipation. These factors must be combined carefully for operational conditions. Cases involving large speed amplitudes and short periods could be very unpleasant for the pilot and may cause local structural overloading, so this particular combination was not considered.

The so called distance-minimizing techniques, known in literature, are based on very complex oscillating paths in final approach, which would require exceptional piloting skills.

On the other hand, the goal of this paper was not to minimize the approach distance for any given sailplane, but to define general influential factors which could be combined within much simpler flying procedures, that can easily and quite safely be performed by pilots of average experience.

Although the approach distance reductions

446

obtained by here presented methods are smaller than obtained by distance-minimized approach for the Vuk-T sailplane, under operational conditions they can certainly make the difference between a successful landing and an undesired outcome. Presented principles can readily be applied to any other sailplane for which the required technical data are available.

5 REFERENCES [1] Stefanović, Z, Cvetković, D. (1996).

Minimum Landing-Approach Distance for a VUK-T Sailplane. 20th International Council of Aeronautical Sciences (ICAS) Congress Proceedings, p. 1061-1064, Sorrento, Italy.

[2] Metzger, D.E., Hedrik, J.K. (1975). Optimal Flight Paths for Soaring Flight. Journal of Aircraft, vol. 12, no. 11, p. 867-871.

[3] Pierson, B.L. (1977). Maximum Altitude Sailplane Winch-Launch Trajectories.

Aeronautical Quarterly, vol. 28, p. 75-84.

[4] Ladson, L.S., Waren, A.D., Rice, R.K. (1967). An Interior Penalty Method for Inequality Constrained Optimal Control Problems. IEEE Transactions on Automatic Control, vol.12, IV, p. 388-395.

[5] Pierson, B.L. (1975). Panel Flutter Optimization by Gradient Projection.

International Journal for Numerical Methods in Engineering, vol. 9, p. 271-296.

[6] Roskam, J., Chuen-Tau, E.L. (1997).

Airplane Aerodynamics and Performance.

DARcorporation, Kansas.

[7] Stefanović, Z, Kostić, I. (2007). Influence of Simple Harmonic Speed Variations on the Vuk-T Sailplane Approach Paths and Distances. FME Transactions, vol. 35, no. 2, p. 15-21.

Stefanović, Z. – Kostić, I.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 447-454 UDC 621.795:621.922.025

Paper received: 24.08.2009 Paper accepted: 24.05.2010

An Investigation into the Influences of Grain Size and Grinding Parameters on Surface Roughness and Grinding Forces when Grinding 1

Halil Demir1,* – Abdulkadir Gullu2 – Ibrahim Ciftci1 – Ulvi Seker2 Karabuk University, Technical Education Faculty, Karabuk, Turkey 2 Gazi University, Technical Education Faculty, Ankara, Turkey

This study was carried out to investigate the effects of grain size on workpiece surface roughness and grinding forces when surface grinding AISI 1050 steel. A previously designed and constructed dynamometer was used to measure and record the forces developed during grinding. Grinding tests were carried out using different grinding wheels of different grains. Ground surface roughness measurements were also carried out. The results showed that grain size significantly affected the grinding forces and surface roughness values. Increasing grain size and depth of cut increased the grinding forces and surface roughness values. For different grain sizes, depth of cuts of 0.01 and 0.02 mm did not result in any significant variations in the grinding forces but further increase in depth of cut led to variations of up to 50% in grinding forces. ©2010 Journal of Mechanical Engineering. All rights reserved.

Keywords: surface grinding, grinding forces, surface roughness, grinding wheel, grain size 0 INTRODUCTION Grinding is probably the oldest surface processing method. It has been utilised since the early days of civilisation. In these early days, it was observed that some natural materials scratched the others and resulted in wear in these other materials when they were slid against each other under pressure. These hard materials used by mechanic action were called “abrasives” and parallel to the developments in technology, these abrasives and abrasive processes also developed.

Abrasives were called with different names depending on their purpose of use and their properties [1]. Grinding is an important manufacturing process which shapes the workpieces with the required geometry, dimensions and tolerances. This process is especially used when the workpieces can not be shaped with the required accuracy and surface quality by the other processes such as turning and milling [2]. In grinding terminology, increasing grain number means decreasing grain size.

For manufactured products, surface integrity is important as it affects various properties of the manufactured parts like fracture toughness, corrosion rate, stress corrosion cracking, wear, magnetic properties and dimensional stability. Surface integrity covers

surface related aspects influencing surface quality. These are surface finish, metallurgical damage and residual stresses. Surface finish is related to the quality of processed surface [3] and [4]. Many machine parts like measuring devices, shafts, gears and rolls must have good surface properties. Grinding process is required to make the surfaces of these parts resistant to corrosion [5].

Some values obtained through engineering calculations do not coincide with those obtained through experimentally measured ones due to some unknown factors and stresses whose effect cannot be determined exactly. In engineering applications, theoretical calculations usually fail to produce accurate results and therefore, experimental measurements become unavoidable.

The accuracy of the empirical equations is also verified by experimental measurements. Wearing of abrasives and its detachment from grinding wheel is a factor which affects the grinding process. The researches on machine tools, cutting tools and workpiece materials necessitated knowledge of cutting forces developed during machining. Therefore, accurate measurement and analyses of the cutting forces are important.

Although much work has done for grinding in this area, the problems have not been solved completely [6-10]. Fastening of the workpieces in

*

Corr. Author's Address: Karabuk University, Tech. Edu. Faculty, 78050 Baliklar Kayasi,Karabuk, Turkey, hdemir@karabuk.edu.tr

447

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 447-454

grinding is one of the important steps and grinding related many errors are the result of unsuitable fastening and inadequate rigidity.

Accurate determination of grinding forces and taking precautions by taking into consideration these forces eliminates the likely problems.

Predetermination of the fastening forces is very important for automation of the design of workholding devices. When selecting standard workholding device components, accurate determination of cutting force dependent fastening forces is of crucial importance.

The aim of this investigation was to examine the influences of grinding wheel grain size and grinding parameters on ground surface roughness and grinding forces when grinding hardened AISI 1050 steel workpieces. The grinding wheels used had different grain sizes.

the abrasive grain and forms a groove. When the workpiece material can not resist the flow stress, chip is formed. The chip formation is called cutting stage. In this chip formation stage, energy is used most efficiently [12-15].

Vs

a

VW

Grain a

hm

448

Vs

Chip formaton

Ploughing

Friction

lk

1 MECHANICS OF CHIP FORMATION DURING GRINDING For grinding of a workpiece surface, ideal cutting can be obtained by many process combinations like ploughing due to lateral displacement, workpiece movement, grinding wheel movement, elasticity of the workpiece and vibration. Many parameters have effects on grinding process. Some of these parameters can be controlled while the others not [11] and [12].

Kinematic relation between grinding wheel and workpiece in grinding process is applied to each grain of the grinding wheel.

Previous work in this area was based on mechanics of mean single grain. Some faces of grain during grinding can be illustrated the geometrical relation between a single grain and workpiece. Non-deformed chip shape, tool path length of the abrasive grain (lk), maximum nondeformed depth of cut (hm) and chip geometry are shown schematically in Fig. 1.

Chip formation in grinding process can be divided into three successive stages: friction, ploughing and cutting. In up-cut grinding, grinding wheel grains rub on the workpiece surface rather than cutting due to the elastic deformation of the system. This is called friction stage. And then, plastic deformation takes place as the elastic limit is exceeded between the abrasive grain and workpiece. This is called ploughing stage. Workpiece material flows plastically through forward and sideward ahead of

Workpiece

Fig. 1. Three stages of chip formation in grinding [12] and [16] Grinding forces not only affect chip formation mechanics, grain wear and temperature distribution but also efficiency of the grinding operation. Therefore, grinding forces are among the most important factors affecting grinding quality.

2 MATERIALS AND METHOD In order to measure the grinding forces, a previously designed and constructed dynamometer was used [7]. The workpiece material was AISI 1050 steel in rectangular blocks of 15 x 15 x 100 mm. Chemical composition of AISI 1050 steel is given in Table 1. This material was supplied as rolled condition.

Before the tests, AISI 1050 specimens were heat treated and subsequently stress relief annealed.

Hardness measurements were also carried out on these specimens using an INSTRON WOLPERT DIA 7571 hardness measuring device. Hardness of this material was found to be as 50 HRc. In order to fix the specimens properly to the dynamometer and to eliminate any distortion due to the heat treatment and to remove defects or other impurities, the wider surfaces of the specimens were ground.

Demir, H. - Gullu, A.- Ciftci, I. - Seker, U.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 447-454

Table 1. Chemical composition of AISI 1050 steel (Weight %) Element % Element % C 0.510 Mo 0.042 Si 0.113 Sn 0.032 P 0.033 Pb 0.064 Mn 0.757 Cr 0.271 Co 0.020 Cu 0.227 V 0.020 S 0.044 The grinding tests were carried out using a TAKSAN TYT-400 surface grinder. Two components of the grinding forces were measured during a single pass. These grinding forces were recorded on a personal computer (PC) using the necessary hardware and software. Aluminium oxide grinding wheels of different grains were used as grinding wheel. Grain numbers of the grinding wheels were 36 (535), 46 (360), 60 (255) and 80 (180). All the grinding wheels were commercial products and produced by EGESAN, TR. Their grade and structure were M and 5, respectively according to ANSI standard. The grinding tests were carried out at depth of cuts (down feed) of 0.01, 0.02, 0.03, 0.04, 0.05 and 0.06 mm as grinding process is generally a finishing process. The wheel 350 mm in diameter was run at a constant revolution of 1596 rev/min and this resulted in a wheel surface speed of 1754 m/min. Table feed was 460 mm/s. Prior to each test, the grinding wheel was dressed and trued by diamond. After the grinding tests, surface roughness values (Ra) of all the ground surfaces were measured using a MITUTOYO Surftest 211 surface measuring device. For comparison purposes, pictures of these grinding wheels were also taken using a Vitec camera at x10, x20 and x30 magnifications.

3 RESULTS AND DISCUSSIONS There are many factors affecting the grinding forces and workpiece surface quality during grinding operations. In this study, the influences of grinding parameters (grinding wheel grain size and depth of cut) on surface roughness and grinding forces were examined. For the grinding tests, grinding wheel grain size and depth of cuts were varied while grinding wheel revolution, wheel dressing rate and flow rate of coolant were kept constant when grinding AISI 1050 steel workpieces. Variations of the

workpiece surface roughness values measured after the tests and the grinding forces were evaluated by constructed graphics in Microsoft Excel. Before the graphics were constructed, regression analyses of the data were carried out and the highest regression coefficient were determined and then the curve model was chosen.

Tangential force (Ft) and normal force (Fn) components were measured individually during grinding and their vector sum was regarded as the grinding force. This is given below:

F  Ft 2  Fn 2 .

(1)

3.1 Grinding Wheel Grain Size – Surface Roughness Relation Variation of mean surface roughness (Ra) values obtained after grinding with grinding wheels of 36, 46, 60 and 80 grain and at depth of cuts of 0.01-0.06 mm is given in Fig. 2. Surface roughness values (Ra) depending on the depth of cut after grinding with grinding wheels of 46, 50 and 80 grain are below 1 μm. However, grinding with grinding wheel of 36 grain results in surface roughness values (Ra) between 1.29 and 2.56 μm, increasing with increasing depth of cut.

Increasing depth of cut beyond 0.03 mm increases the difference between the surface roughness values obtained after grinding with grinding wheels of 46, 60 and 80 grain. Surface roughness values obtained after the tests carried out by changing grinding wheel grain size are as follows: 0.63-1.11 μm for grinding wheel of 46 grain, 0.46-0.80 μm for grinding wheel of 60 grain and 0.37-0.81 μm for grinding wheel of 80 grain. The lowest surface roughness value is obtained after grinding with grinding wheel of 80 grain. Fig. 2 shows that grinding wheel grain size has great influence on workpiece surface roughness. At 0.04-0.05 mm depth of cuts, the surface roughness obtained using grinding wheel of 80 grain was seen to be higher than those obtained by grinding wheels of 60, 46 and 36 grains by up to 30%, 80% and 400%, respectively. Increasing grinding wheel abrasive grain size led to increases in surface roughness values. The larger is the grinding wheel abrasive grain size the larger is the distance between the grains and also the larger is the removed chip cross-section [8]. In addition, roughness variation range becomes larger with increasing grinding

An Investigation into the Influences of Grain Size and Grinding Parameters on Surface Roughness and Grinding Forces when Grinding

449

StrojniĹĄki vestnik - Journal of Mechanical Engineering 56(2010)7-8, 447-454

3,5

36M5

3

46M5

2,5

) m u ( a2 R ,s s e n h g 1,5 u o R e ca fr 1 u S

60M5

y = 21,08 6x + 1 ,1 05 R2 = 0,8812

y = 8,7143x + 0,513 R2 = 0, 8831 y = 6,1714x + 0,367 R2 = 0,8848 y = 4,0286x + 0,347 R2 = 0,9137

2

36 1,6

1,2

46

0,8

60 0,4

80

0

36

80M5

46 60 Grinding wheel grain

80

Fig. 3. Grinding wheel grain size and surface roughness relation

0,5

Fig. 2. Grinding wheel grain size, depth of cut and surface roughness relation In this study, in order to show the influence of grinding wheel grain size on surface roughness well, arithmetical average of the surface roughness values obtained using the grinding wheels of same grain were obtained as follows: Raavr=[(Ra(td=0.01 mm))+(Ra(td=0.02 mm))+ +‌+(Ra(td=0.06 mm))]/6 .

(2) By using these arithmetical average values, the graph was plotted (Fig. 3). Surface roughness values were seen to increase significantly with increasing grinding wheel grain size. For example, with decreasing grinding wheel grain size, surface roughness values increased by 1.7, 2.29 and 5 times for grinding wheels of 60, 46 and 36 grains when compared to grinding wheel of 80 grain. These results show that a small increase in grain size leads to considerable increases in ground surface roughness values.

When the surface pictures of the ground materials are compared, it is seen that increasing depth of cut and grinding wheel grain size increases the depth and width of the grooves (Fig.

450

4). Grinding with grinding wheel of 46 grain at 0.02 mm depth of cut produced larger grooves (Fig. 4a) while decreasing grain size led to smaller grooves (Fig. 4b and c).

Surface Roughness, Ra (um)

wheel grain size and this, in turn, makes the control of surface quality difficult (Fig. 2).

Decreasing grinding wheel grain size value both results in diminishing roughness variation range and decreasing surface roughness value. The surface quality obtained using grinding wheel of 36 grain is not representative of a finishing surface and can be regarded a cleaning and a large volume of chip removal operation.

Small dirt and crater wear due to detached grains were observed after grinding with grinding wheel of small grain size (80 grain) at 0.03 mm depth of cut (Fig. 5a). Grinding wheel grains were found to penetrate the ground surface at higher depth of cuts of 0.04, 0.05 and 0.06 mm, Figs. 5b and 5c. Therefore, from Figs. 5b and c, the harmful effects of higher depth of cuts, which depend on the grain size, are seen. On the other hand, Figs. 6a and 6b show small parts of workpieces filling the porosities on the grinding wheel. In Figs. 6a and 6b, the white areas belong to a needle indicating the filled porosities on the grinding wheel for comparison purpose. Filling of the grinding wheel porosities results in deterioration in surface roughness. Detachment of the grinding wheel grains during grinding reduces the diameter of the wheel and this, in turn, leads to deviation from the required part dimensions [17]. Therefore, the grains embedded into the workpiece give rise to various problems like scratches on the workpiece and grain detachment.

These sorts of problems, in turn, cause undesirable results during service life of the parts.

It was observed at all stages of this study that surface roughness values significantly increased with increasing depth of cut. Fig. 7 shows that increasing depth of cut deepened and widened the grinding marks as grooves on the ground surfaces.

Demir, H. - Gullu, A.- Ciftci, I. - Seker, U.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 447-454

50 m a) 46M5

50 m

50 m

b) 60M5 c) 80M5 Fig. 4. Workpiece surface appearance depending on grinding wheel grain size

20 m

20 m a) Grinding wheel: 80M5 Depth of cut: 0.02 mm

b) Grinding wheel: 80M5 Depth of cut: 0.04 mm

c) Grinding wheel: 80M5 Depth of cut: 0.04 mm

Fig. 5. Workpiece surface appearance depending on grinding wheel grain size

0.5 mm 0.5 mm a) b) Fig. 6. Wheel pictures showing the filled porosities with workpiece material at 0.04 mm depth of cut for the grinding wheel of 80 grain

20 m

20 m

20 m

a) Depth of cut: 0.01 mm b) Depth of cut: 0.02 mm c) Depth of cut: 0.03 mm Fig. 7. The influence of depth of cut on surface structure for the grinding wheel of 60 grain 3.2 Grinding Wheel Grain Size – Grinding Force Relation The influence of grinding wheel grain size on grinding forces at 0.01-0.06 mm depth of cuts is given in Fig. 8. Grinding forces were recorded as 1.93-16.8 N for 46M5, 1.9-13.82 N for 60M5

and 1.7-9.18 N for 80M5 grinding wheels. These lower forces can be explained by the very small depth of cuts, very high cutting speed and small workpiece width. It was determined that grinding forces increased with increasing grinding wheel grain size at 0.03 mm depth of cut for the three grinding wheels of different grain sizes.

An Investigation into the Influences of Grain Size and Grinding Parameters on Surface Roughness and Grinding Forces when Grinding

451

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 447-454

16

46M5 ) 12 N ( e rc o f g n i8 d n ri G

60M5 80M5

4

Fig. 8. Depth of cut – grinding force relation depending on grain size Grinding with grinding wheel of 80 grain showed that the wheel was subjected to excessive forces at higher depth of cut (0.06 mm). Some burns and cracks were observed on the ground surface by the naked eye as the results these high forces.

As the grain size of the grinding wheel of 80 (180 µm) grain has smaller grains than those of the grinding wheels of 60 (255 µm) and 46 (360 µm) grains, this wheel results in larger contact area between the wheel and ground workpiece during grinding. This larger contact area increases the friction and workpiece temperature during grinding. As the result of increased temperature, residual stresses occur on the workpice [8] and [18]. This was observed as the residual stress related small cracks on the ground surface. The small cracks on the

0.05 mm

workpiece were inferred from the chips accumulated along the cracks due to magnetic field of the workpiece table. As grinder spindle rotates at a constant wheel speed, the number of the grains on the wheel which removes chip from the workpiece varies depending on the grain size.

Increasing grain number increases the grains which removes chip (Fig. 9). If grinding process is likened to milling operation with a multi-tooth cutter, the cross-section of chip per cutter tooth increases with decreasing number of cutter tooth when milling at a constant cutting speed and feed rate. This also increases the forces developed during milling [19]. Fig. 9 shows porosity and grain size.

Fig. 10a gives the variation of surface roughness values depending on grinding wheel grain size and depth of cut while Fig. 10b gives the variation of grinding force depending on the same parameters. It is seen from Fig. 10a that surface roughness is significantly affected by the grinding wheel grain size while the depth of cut has little influence on it. This can be explained by the different nature of the grinding operations when compared to other machining operations. In turning, milling and other machining operations, the chip is usually removed at one pass. The workpiece surface is formed dependent on feed rate, cutting tool geometry and other parameters.

Unlike turning and milling operations, the chip in grinding is removed at several passes. Therefore, the grinding operation is carried out at much lower depth of cuts.

0.05 mm

a) 46M5 b) 60M5 c) 80M5 Fig. 9. Depth of cut - grinding force relation depending on grain size

452

Demir, H. - Gullu, A.- Ciftci, I. - Seker, U.

0.05 mm

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 447-454

Surface roughness , Ra (m)

1,6 1,4

25

1

20

0,8

15

0,6

10

0,4 0,2 0 0,06 0,05 46 M 5

5

46M5 60M5

0,04 0,03

6 0 M5 80M5 0 01

0,02

80M5

Depth of cut (mm)

0,01

0,02

0,03

0,04

0,05

Grinding force (N)

1,2

0 0,06

Depth of cut (mm)

a) b) Fig. 10. Workpiece surface appearance depending on grinding wheel grain size, a) missing figure description, b) missing figure description This situation significantly alleviates the influence of depth of cut on surface roughness.

The grinding forces developed at 0.01-0.02 mm depth of cuts did not vary much. However, a 50% increase was seen when the depth of cut was increased further (Fig. 10b). The highest cutting force with increasing depth of cut was obtained when grinding with grinding wheel of 46 grain.

4 CONCLUSIONS The following conclusions can be drawn from the present study investigating the effects of grain size on workpiece surface roughness and grinding forces when surface grinding AISI 1050 steel:  Grinding wheel grain size was found to have great influence on surface roughness and grinding force values. Increasing grinding wheel grain size increased the surface roughness values and the grinding forces.  At 0.04-0.05 mm depth of cuts, the surface quality (in terms of Ra) obtained by 80 grain wheel was better by 30%, 80% and 400% than those for 60, 46 and 36 grain wheels, respectively.  Grinding wheel grains were found to penetrate the ground surface at higher depth of cuts of 0.04, 0.05 and 0.06 mm when the grinding wheel of 80 grain was used.

 At higher depth of cuts, some burns and cracks were observed on the ground surface by the naked eye for the grinding wheel of 80 grain.  Grooves, burns and waviness can be decreased if depth of cut is reduced when using grinding wheels with small grains.

5 REFERENCES [1] [2] [3]

[4]

[5]

[6]

Gullu, A., Poyrazoglu, O. (2000). The effect of superfinishing process on the surface quality. Technology, vol. 1, no. 1, p. 49-58.

Kalpakjian, S. (1991). Manufacturing process for engineering materials, 2nd ed., Addison-Wesley, New York.

Gondi, P. et al., please indicate all other authors (1993). Structural characteristics at surface and Barkhausen noise in AISI 4340 steel after grinding. Nondestructive Testing and Evaluation, vol. 10, p. 255-267.

Shaw, M.C. (1994). A production engineering approach to grinding temperatures.

Journal of Materials Processing Technology, vol. 44, p. 59-69.

Demir, H., Gullu, A. (1999). An investigation of relationship between grinding ratio and surface roughness in cylindrical grinding. Technology, vol. 1-2, p.

151-167.

Demir, H., Gullu, A. (2006). Design and construction of a dynamometer for measurement of grinding forces during

An Investigation into the Influences of Grain Size and Grinding Parameters on Surface Roughness and Grinding Forces When Grinding

453

StrojniĹĄki vestnik - Journal of Mechanical Engineering 56(2010)7-8, 447-454

[7]

[8] [9]

[10]

[11]

[12]

[13]

454

surface grinding operation. Technology, vol.

9, p. 111-118.

Demir, H. (2003). Investigation of the influences of grinding parameters on grinding forces and surface quality in surface grinding. Ph.D. Thesis, Gazi University Institute of Science and Technology, Ankara, Turkey.

Demir, H., Gullu, A. (2001). The effect of paramaters in the grinding. Journal of Engineering Sciences, vol. 7, p. 189-198.

Seker, U. et al. (2002). Design and construction of a dynamometer for measurement of cutting forces during machining with linear motion. Materials and Design, vol. 23, p. 355-360.

Gunay, M. (2003).

Experimental investigation of the influence of cutting tool rake angle on forces during metal cutting.

MSc Thesis, Gazi University Institute of Science and Technology, Ankara, Turkey.

Srivastava, A.K. et al. (1992). Surface finish in robotic disk grinding. International Journal of Machine Tools & Manufacture, vol. 32, p. 269-297.

Chen, X., Brian, W. (1996). Analysis and simulation of the grinding process, Part II: Mechanics of grinding. International Journal of Machine Tools & Manufacture, vol. 36, p. 883-896.

Ramseh, N. et al. (1980). Investigations on laser dressing of grinding wheels - Part I.

[14]

[15]

[16]

[17]

[18]

[19]

Preliminary study. ASME Journal of Engineering for Industry, vol. 102, p. 244251.

Rajmohan, B., Radhakrishan, V. (1994). On the possibility of process monitoring in the grinding by spark intensity measurements.

ASME Journal of Engineering for Industry, vol. 116, p. 124-129.

Srihari, G., Lal, G.K. (1994). Mechanics of vertical surface grinding. Journal of Materials Processing Technology, vol. 44, p.

14-28.

Huang, L. et al. (1999). Effect of tool/chip contact length on orthogonal turning performance.

Journal of Industrial Technology, vol. 15, p. 88-91.

Cander, P. (1975).

The intrinsic characteristics of ground surface. The Proceedings of The Abrasive Engineering Society’s International Technical Conference, missing date of the conference, missing city.

Matsumoto, Y. (1986). The effect of hardness on the surface integrity of AISI 4340 steel. ASME Journal of Engineering for Industry, vol. 108, p. 175-196.

Verkerk, J. (1971). Final report concerning CIRP cooperative work on the characterization of grinding wheel topography. Annals of the CIRP, vol. 26, p.

385-395.

Demir, H. - Gullu, A.- Ciftci, I. - Seker, U.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 455-463 UDC 621.3:681.5:004.94

Paper received: 16.01.2009 Paper accepted: 07.07.2010

Optimal Design of the Fuzzy Sliding Mode Control for a DC Servo Drive Dragan Antić – Marko Milojković* – Zoran Jovanović – Saša Nikolić University of Niš, Faculty of Electronic Engineering, Serbia A new controller for the optimal speed control of a DC drive is presented in this paper. The controller employs a variant of fuzzy sliding mode, optimized by a genetic algorithm. Proposed controller has many advantages, such as satisfactory control performance under a wide range of operating conditions, a faster response than conventional controllers and suppressed chattering phenomenon. The simulations, experimental results, and comparative analysis verify the efficiency, excellent performance, and robustness of such a control in the case of a DC servomotor. © 2010 Journal of Mechanical Engineering. All rights reserved.

Keywords: sliding mode control, fuzzy control, genetic algorithm, DC motor 0 INTRODUCTION One of the possible approaches to the robust control of the uncertain systems has been found in variable structure systems and sliding mode control [1]. The principal goal of the sliding mode control technique is to force a system state to a certain prescribed manifold, known as the sliding hyper surface. Once the manifold is reached, the system is forced to remain on it thereafter. When in the sliding mode, the system is equivalent to an unforced system of lower order, which is insensitive to both parametric uncertainty and unknown disturbances that satisfy the matching condition. The main drawback of the sliding mode control is the requirement of a discontinuous control law across the sliding manifold. In practical systems, this leads to a phenomenon termed chattering [2]. Chattering involves high-frequency control switching and may lead to excitation of unmodeled highfrequency system dynamics. Chattering also cause high heat losses in electronic systems and undue wear in mechanical systems. Smoothing techniques such as boundary layer [3] have been employed in order to prevent chattering.

However, such an approach leads to a loss of asymptotic stability and a controller that can guarantee final tracking accuracy only to within a certain vicinity of the demand.

Over the last few years, the apparent similarities between the sliding mode and fuzzy controllers [4] and [5] in diagonal form have been noticed. This fact has subsequently motivated considerable research effort in combining the two

topologies [6] in a manner that serves to reduce the limitations of the sliding mode, while still maintaining the guarantees of global uniform ultimately bounded stability and invariance to matched disturbance [7] and [8]. The main difficulty in designing fuzzy controller is the acquisition of the controller parameters that are usually determined by human expert’s knowledge or trail and error method. It is a difficult problem to find optimal parameters of the controller in order to achieve maximum performance.

Genetic algorithms [9] and [10] are optimization technique based on simulation of the phenomena taking place in the evolution of species. They have demonstrated very good performances as global optimizers in many types of control applications [11]. They are good optimizers for fuzzy controllers, also [12] and [13]. In this paper, the genetic algorithm is applied to determine optimal values of the key parameters required for the fuzzy sliding controller design [14] to [18].

To verify the efficiency of the proposed control method, simulation and experimental results of such control for a DC servo drive and comparative analysis with the other types of controllers are given in this paper, also.

1 SLIDING MODE CONTROL Consider the system of the following form:

x n   f  x, t   g x, t u  d .



where x  x, x ,..., x  n 1



T

(1)

is the state vector, d

represent disturbances, u is the control input and

* Corr. Author's Address: Faculty of Electronic Engineering, A. Medvedeva 14, 18000 Niš, Serbia, marko.milojkovic@elfak.ni.ac.rs

455

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 455-463

f, g are nonlinear functions of the state vector and the time.

The tracking control problem is to find a control law such that given a desired trajectory xd, the tracking error x-xd tends to zero despite the presence of the model uncertainties, unmodeled frequencies and disturbances [2]. With the tracking error defined as:

e  x  xd  e, e,..., e n1  , T

(2) a sliding surface (sliding line for second order systems) s ( x , t )  0 is determined by:

s ( x , t )  (d dt   ) n 1 e,   0 .

(3)

In order to derive control law such that the state vector remains on the sliding surface, we define a Lyapunov function [2] and [3]:

V

1 2 s .

2

(4)

Sufficient condition for the stability of the system (1) is:

1 d 2 s    s ,   0 .

V  2 dt

(5)

From (5) we can obtain so-called reaching condition that enables the system to reach the sliding surface in finite time interval:

s sgn s    .

(6)

For a second order system we have:

s   e  e , s   e  e   e   x   xd .

2 FUZZY SLIDING MODE CONTROL Instead of the sliding mode controller with boundary layer, fuzzy controller can be used [6].

The controller evaluates the variable control gain and the control signal that depends on the distance of the state vector from the sliding surface. The parameter values (scaling factors, membership functions, rule term base) of the fuzzy sliding mode controller can be chosen in such a way to obtain the best system behavior with respect to specific criteria. With the fuzzy sliding mode controller, different values for gain K in (9) can be selected. A large control gain is applied only when the state vector is far away from the sliding manifold, so that system moves towards the sliding manifold fast, but when the trajectory of the system is near the sliding manifold, the control signal will change (decrease) smoothly.

Each fuzzy rule can be defined in such a way that it includes the distance s between the state vector e and the sliding surface and fuzzy control value u corresponding to that distance.

With the increasing of distance s , absolute value of the control signal

  s

(7)

.

if s  NS then u  PS , if s  ZE then u  ZE , if s  PS then u  NS ,



.

u  g 1 fˆ  e  K ( x, t )sgn  s  , K ( x, t )  0.

(9)

To avoid large changes of the control signal, boundary layer [3] can be introduced, which leads to the control law:   s  u  g 1  fˆ  e  K ( x, t )sat    ,       0, K ( x, t )  0.

456

(10)

(11)

if s  PB then u  NB ,

(8)

Following control law can be used for establishing the sliding mode with respect to (8):

should be

increasing, too. Therefore, the complete rule base of this type of fuzzy sliding controller could take the following form: if s  NB then u  PB ,

From the reaching condition (7) we obtain:

ss  s ( e  f ( x , t )  g ( x , t )u  d   xd )

u

with the labels: NB is negative big, NS is negative small, ZE is about zero, PS is positive small, and PB is positive big.

One choice for normalized fuzzy sets of the linguistic variables s   NB, NS, ZE, PS, PB and u   NB, NS , NU , PS , PB are shown in Fig. 1.

The control law formed in such a way can be represented with:

u   K fuzzy  s  sgn  s  .

(12)

Calculation of the fuzzy variable gain Kfuzzy depends on the number and the shapes of

Antić, D. – Milojković, M. – Jovanović, Z. - Nikolić, S.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 455-463

the membership functions as well as the normalization coefficients.

Fig. 1. Fuzzy sets s and u This type of fuzzy sliding controller is similar to the boundary layer concept, which can be noticed from the Fig. 2.

Fig. 2. Fuzzy sliding hypersurface for the second order system 3 OPTIMIZATION OF THE FUZZY SLIDING MODE CONTROL Further improvements can be achieved by optimization of the fuzzy controller with the rule base (11) and membership functions given in Fig. 3. Fuzzy sets of the input variable s are trapezoid (generalized sets from Fig. 1), symmetrical and normalized. This choice for fuzzy sets enables flexible and smooth control actions. Parameters which can be optimized are labeled with a, b and c (see Fig. 3). These parameters represent characteristic points of normalized input variable and completely describe it. Output variable u has singleton symmetrical membership functions and the range -20 V to 20 V. Only parameter left to optimize is d (points +d and –d denote positions of membership functions PS and NS respectively).

Fig. 3. Fuzzy sets for fuzzy controllers a) input; b) output variables Fig. 4 represents the used simulation model in Simulink. There are two more system parameters accessible to optimization: sliding line slope f and scaling coefficient for the input variable, labeled with h.

The genetic algorithm is applied to determine optimal values of the adjustable parameters required for the fuzzy sliding mode controller design, so the overall conceptual simulation block diagram is given in Fig. 5.

Genetic algorithm used in simulation has the following parameters: initial population of 200, number of generations 100, stochastic uniform selection, reproduction with 8 elite individuals, Gaussian mutation with shrinking and scattered crossover. Chromosome has a structure that consists of six parameters encoded as real numbers: abcdfh. The goal of the simulation was to make a tracking error as small as possible. So, the following fitness function for genetic algorithm was chosen: T

F   e 2  t  dt ,

(13)

0

where T represents the simulation time.

4 SIMULATION RESULTS 4.1. Problem Statement DC motor represents the simplest DC drive system. Fig. 6 represents the structure of the electrical circuit of a DC motor.

Optimal Design of the Fuzzy Sliding Mode Control for a DC Servo Drive

457

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 455-463

Fig. 4. Simulation model

Fig. 5. Simulation block diagram of the DC motor control J = 0.001 kgm2, km = 0.001 Vsrad-1, ke = 0.008 NmA-1, B = 0.01 Nmsrad-1. The demand is that the shaft angular velocity should track a desired trajectory.

Let    t  represents a desired velocity and e    tracking error. If we choose the variables x1  e and x2  e as state space variables, then the system can be described with the following equations:

x1  x2

Fig. 6. Electrical model of a DC motor The motor dynamics are governed by two coupled first-order equations with respect to armature current and shaft speed:

di  u  Ri  ke , dt d J  kmi  B , dt

L

(14)

where L represents the armature inductance, R the armature resistance, i the armature current, ke the back EMF constant, ω the shaft angular velocity, J the moment of inertia, B the coefficient of viscous friction, km the motor torque constant, u the terminal armature voltage.

The plant used in simulations is a DC motor given with (14) and the following parameter values: L = 0.001 mH, R = 0.5  ,

458

k k B k R x2  m e x1  e u  r  L JL JL k k B R   r  m e r .

L JL

x2  

(15)

For the second order system, we can define switching function as:

s   x1  x2   x1  x1 ,

(16)

where λ represents a positive constant.

4.2 Sliding Mode Control First simulation employs sliding mode control in the reley control form: u  U 0 sgn  s  .

Sliding mode existance conditions can be fullfilled with the choosen values: U0 = 20 V and λ = 100. Fig. 7 shows the simulation results for the square reference input.

Antić, D. – Milojković, M. – Jovanović, Z. - Nikolić, S.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 455-463

The results demonstrate high-frequency control switching and the chattering problem.

Improvement can be made by using the quasirelay sliding mode control with the control law: u  U 0 x1 sgn  s  . Quasi-relay control unifies

the MATLAB environment in the solution of servomechanism control problems.

linear action proportional to the error and relay action that commutates linear control (Fig. 8).

4.3 Fuzzy Sliding Mode Control Second simulation was performed with the fuzzy-sliding mode control based on fuzzy controller with rule base (11) and the membership functions given in Fig. 1. Simulation parameters are the same as in the first case and the simulation results are given in Fig. 9. Tracking precision is about the same but the real improvement can be noticed on the control signal discontinuities and levels. Chattering is suppressed, also.

4.4 Fuzzy Sliding Mode Control Optimized with Genetic Algorithm Final simulation was performed with fuzzy sliding mode controller with rule base (11) and membership functions shown in Fig. 3. Genetic algorithm was used to find adjustable parameters as it is described in section 3. The genetic algorithm founded following optimal parameters of the fuzzy sliding controller: a = 0.02, b = 0.51, c = 0.54 (parameters describing membership functions of input variable s), d = 11.56 (parameter describing membership functions of output variable u), g = 918.67 (sliding line slope), h = 0.038 (scaling coefficient of input variable).

With these values, the results of the simulation are given in the Fig. 10.

It can be noticed that optimized control gives better results and further improvements in terms of lower tracking error, faster response and reaching stationary state, lower control signal discontinuities and chattering elimination.

5 EXPERIMENTAL RESULTS For the purposes of the practical verification of proposed control method, modular servo drive is considered. The modular servo system [19] consists of the “Inteco” servomechanism and open-architecture software environment for real-time control experiments.

The servo system supports the real-time design and implementation of advanced control methods using MATLAB and Simulink tools and extends

Fig. 7. Sliding mode control: a) reference and actual velocity; b) control signal; c) sliding variable; d) system trajectory in x1x2 plane

Optimal Design of the Fuzzy Sliding Mode Control for a DC Servo Drive

459

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 455-463

motor and generates a voltage signal proportional to the angular velocity.

Fig. 8. Quasi-relay control: a) reference and actual velocity; b) control signal The integrated software supports: on-line process identification, control system modeling, design and simulation and real-time implementation of control algorithms. The modular servo system uses standard PC hardware platforms and Microsoft Windows operating systems.

The servo system setup consists of several modules mounted at the metal rail and coupled with small clutches. The modules are arranged in the chain. The DC motor together with tachogenerator opens the chain. The gearbox with the output disk closes the chain. The potentiometer module is located outside the chain. The DC motor can drive the following modules: inertia, backlash, encoder module, magnetic brake, and the gearbox with the output disk. The rotation angle of the DC motor shaft is measured using an incremental encoder. A tachogenerator is connected directly to the DC

460

Fig. 9. Fuzzy sliding mode control: a) reference and actual velocity; b) control signal; c) system trajectory in x1x2 plane The servomechanism is connected to a computer where a control algorithm is realized based on measurements of angle and angular

Antić, D. – Milojković, M. – Jovanović, Z. - Nikolić, S.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 455-463

velocity. All functions of the board are accessed from the Modular Servo Toolbox, which operates directly in the MATLAB/Simulink environment.

Modular servo drive is nonlinear due to some nonlinear static characteristics such as hysteresis and saturation, which may occur in the following devices: operational amplifiers, actuators, finite word length in A/D and D/A converters.

Fig.11. Sliding mode control: a) reference and actual velocity; b) control signal

Fig. 12. Optimized fuzzy sliding mode control: a) reference and actual velocity; b) control signal Fig. 10. Optimized fuzzy sliding mode control: a) reference and actual velocity; b) control signal; c) system trajectory in x1x2 plane

This servo motor has been used in our experiments with described control methods employed. Figure 11 shows motor velocity and control signals for the sliding mode control with

Optimal Design of the Fuzzy Sliding Mode Control for a DC Servo Drive

461

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 455-463

the pure relay control form (starting method) described in section 4.2 while the Fig. 12 shows the same signals for the optimized fuzzy-sliding mode control (final method) described in section 4.4.

Given experimental results are similar to those obtained by simulations (Figs. 7, 9 and 10) with two notes in mind. Control signal in all experiments is limited to  5 V and high variations are results of previously described system nonlinearities. Performed experiments verify obtained both simulation results and mathematical background. Although our control method gives a little bit slower response comparing to the classical bang-bang control, its strength is obvious from the given control signals.

Optimized fuzzy sliding mode control has relatively low levels, and discontinuities of the control signal which results in lower heat losses and undue system wear.

[3] [4] [5] [6] [7]

[8]

[9]

6 CONCLUSION [10] A new method for systems control based on genetic algorithms applied on fuzzy sliding mode is realized in this paper. Illustration and verification of this new method is done by simulation and experiments on a DC servo drive system. Similar method can also be applied to AC machines. The performance of the proposed control method is compared with the other control approaches. The simulation results of the proposed controller, compared to the traditional variable structure controller and fuzzy sliding mode control verify the effectiveness of the introduction of fuzzy logic and genetic algorithms in the variable structure control design. Fast systems response, relatively low levels, and discontinuities of the control signal as well as suppressed chattering are achieved.

7 REFERENCES [1]

[2]

462

Utkin, V.I. (1977). Variable structure systems with sliding modes. IEEE Transactions on Automatic Control, vol. 22, no. 2, p. 212-221.

Utkin, V.I., Guldner, J., Shi, J. (1999).

Sliding mode control in electromechanical systems. CRC Press.

[11]

[12] [13]

[14]

[15]

[16]

[17]

Perruquetti, W., Barbot, J.P. (2002). Sliding mode control in engineering. Marcel Dekker Inc. Zadeh, L.A. (1965). Fuzzy sets. Information and Control, vol. 8, p. 338-353.

Passino, K.M, Yurkovich, S. (1998). Fuzzy control, Addison-Wesley. .

Palm, R., Driankov, D., Hellendoorn, H. (1996). Model based fuzzy control. SpringerVerlag, Berlin.

Antić, D., Dimitrijević, S. (1998). Nonminimum phase plant control using fuzzy sliding mode. Electronic Letters, vol. 34, no.

11, p. 1156-1158.

Antić, D., Milojković, M., Nikolić, S., Perić, S. (2010). Optimal fuzzy sliding mode control with a time-varying sliding surface.

ICCC-CONTI 2010, Timisoara, May 27.-29., p. 149-153.

Holland, J.H. (1975). Adaptation in natural and artificial systems. University of Michigan Press, Ann Arbor.

Mitchel, M. (1996). An Introduction to genetic algorithms, MIT Press, Cambridge.

Danković, B., Antić, D., Jovanović, Z., Milojković, M. (2007). Genetic algorithms applied in parameter optimization of casacade connected systems. ICEST, Ohrid, Macedonia, June 24.-27., p. 557-560.

Cordon, O., Herrera, F., Hoffmann, F., Magdalena, L. (2001). Genetic fuzzy systems, World Scientific, Singapore.

Hsieh, W.M., Leu, Z.G., Yang, H.C., Lin, J.Y. (2010). Tracking control of uncertain DC server motors using genetic fuzzy system, Lecture Notes in Computer Scinece, Springer, Berlin.

Wong, C., Huang, B., Lai, H. (2001).

Genetic-based sliding mode fuzzy controller design. Tamkang Journal of Science and Engineering, vol. 4, no. 3, p. 165-172.

Chieh-Li, C., Ming-Hui, C. (1998). Optimal design of fuzzy sliding-mode control: A comparative study. Fuzzy Sets and Systems, vol. 93, p. 37-48.

Choi, H.H. (2009). Adaptive controller design for uncertain fuzzy systems using variable structure control approach.

Automatica, vol. 45, no. 11, p. 2646-2650.

Antić, D., Milojković, M., Mitić, D. (2007).

An improvement of fuzzy sliding mode control of nonminimum phase plants by

Antić, D. – Milojković, M. – Jovanović, Z. - Nikolić, S.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 455-463

[18]

using genetic algorithms. SAUM, Niš, November 22.-23., p. 129-132.

Antić, D., Milojković, M. (2007). Nonlinear system control by using genetic algorithms

[19]

and fuzzy sliding mode. TEHNIKA, Elektrotehnika, vol. 56, no. 3, p. 9-16.

Inteco, Modular Servo System-User’s Manual, (2008).

Available at www.inteco.com.pl

Optimal Design of the Fuzzy Sliding Mode Control for a DC Servo Drive

463

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476 UDC 532.57:621.63

Paper received: 28.08.2009 Paper accepted: 29.04.2010

Flow Characteristics of Bladeless Impeller Made of Open Cell Porous Material 1

Gašper Benedik1,*, Brane Širok2, Janez Rihtaršič3, Marko Hočevar2 Domel d.d., Železniki and University of Ljubljana, Faculty of Mechanical Engineering, Slovenia 2 University of Ljubljana, Faculty of Mechanical Engineering, Slovenia 3 Domel d.d., Železniki, Slovenia

This article describes a bladeless turbo machine impeller that transfers energy from rotor to fluid over a structure of porous material. Impeller construction is described and theoretical description of fluid flow through rotating porous material is given. Pressure drops for impellers made of different materials and with different design parameters in dependence on air volume flow rate were measured.

Measurements of local air flow velocity field close to impeller circumference were performed on a stationary impeller with one component hot-wire anemometer. Air flow velocity dependence on local material structure and volume flow rate was analysed. Integral characteristics measurements for different impellers made of different materials and with different design parameters are presented. Local measurements of local radial and tangential velocities close to rotating impeller circumference are shown. ©2010 Journal of Mechanical Engineering. All rights reserved.

Keywords: bladeless impeller, turbo machine, porous open cell material, velocity field, hot-wire anemometer 0 INTRODUCTON In a common type of centrifugal turbo machine energy is transferred from motor to fluid over impeller blades. Such machines are widely and successfully used, however, they have some limitations. Because blade design is optimised only for optimal working point, unwanted phenomena are common, such as flow separation, reverse flow, stall and surge outside optimal working point [1]. As a consequence efficiency outside of working point decreases and noise level increases [2].

This paper presents a centrifugal turbo machine in which is the energy from the rotor to the fluid is transmitted over the structure of porous open cell material and not over impeller blades, as in common turbo machinery [3, 4]. It is expected that in this type of bladeless impeller made of open cell porous material all earlier mentioned unwanted phenomena should be absent and due to variable fluid flow streamlines this impeller should reach relatively uniform efficiency in a wide working regime. Its weakness is pressure drop which occurs during the fluid flow in porous structure and consequently reduction in efficiency that is expected to be larger in high volume flow regime because of 464

*

higher relative fluid flow velocities in porous material [4]. Consequently we expect the impeller is more suitable for use at low and medium volume flow rate working regime. We are particularly interested in flow regime up to the volume flow rate of 8 dm3/s, where the impeller of selected size will be preferably used.

This research will investigate pressure drop in dependence on air volume flow rate passing the stationary bladeless impeller.

Different variants of rotating impellers will also be investigated. The influence of material choice and changes of impeller dimensions will be analysed. We focus on low volume flow rate regime, where unwanted phenomena in conventional types of impellers are present. Paper should be considered as a study aimed at confirming the operation of bladeless impeller and revealing its integral and local characteristics.

Hot-wire anemometry will be used for local characteristics measurements. The research will be limited to impeller rotational frequency of 10,500 min-1. However, limited tensile strength of used porous material [5] enables rotational frequency of up to 20,000 min-1.

Tesla has developed a bladeless turbo machine composed of flat parallel co-rotating discs, spaced along a shaft [6]. A flow-through of

Corr. Author's Address: Domel d.d., Otoki 21, SI-4228, Železniki, Slovenia, gasper.benedik@domel.si

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476

fluid between the discs results in momentum exchange between the fluid and the discs, hence shaft torque and power [7]. Tesla turbo machine is interesting because of its simplicity, nearly white noise spectra and its possibility of using non-Newtonian, very viscous fluids or even abrasive two-phase flow mixtures [7]. However, due to low aerodynamic efficiency, which reaches 60% at most and drops rapidly outside the optimal working point, the device has not achieved wide industrial application [7]. Norbert [8] describes a patented device, where a disc of open cell porous material is rotated around the central axis. However, the flow direction is axial and the function of porous material is dust filtration and not fluid flow generation. There are a number of articles which investigate fluid flow through stationary metal foam. Dukhan [9] experimentally investigates pressure drop in dependence on volume flow for different materials. Boomsma [10] uses numerical simulation to investigate fluid flow through open cell foam on micro scale. Auriault [11] and Sawicki [12] theoretically describe laminar fluid flow through rotating porous media. There are no available references for experimental research done on bladeless impeller similar to the one described in this paper.

0.1 Impeller Design The design of bladeless impeller is shown in Fig. 1. Upper and lower wall and porous structure were manufactured to the selected size.

The porous material with the height of 12.5 mm or 8.5 mm was inserted between both walls and the structure was assembled with six rivets. Fig. 2 shows two samples of bladeless impellers. Booth

walls were manufactured with a CNC milling machine and consist of aluminium. Impeller is working on a common shaft with electric motor and rotates around the axis. Radial inducer, shown in Fig. 2 b) was also added for certain measurements. Impeller assembly outer radius was 62.5 mm and its height was 15 mm or 11 mm, depending on porous material used. 12.5 mm or 8.5 mm in height respectively. Each impeller was statically balanced by the removal of upper wall material.

Fig. 1. Impeller design with assembly parts [3]: (1) radial direction of fluid velocity at the exit, (2) axial direction of fluid at the entrance, (3) impeller upper wall, (4) open cell porous material, (5) axis of rotation, (6) impeller lower wall 0.2 Choice of Porous Material An example of used open cell porous material is shown in Fig. 3. It consists of open cell metal foam, produced by ERG materials, made of aluminium with homogenous structure and 88% material porosity. Porosity is evaluated as:



Vempty Vempty  Vsolid

%

(1)

Fig. 2 a). Bladeless impeller without and b) bladeless impeller with radial inducer .

Flow Characteristics of Bladeless Impeller Made of Open Cell Porous Material

465

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476

Vempty represents volume without solid material and Vsolid volume of solid material. 88% porosity was chosen due to relatively low fluid flow resistance, low density and sufficient mechanical strength [5, 9]. Computer tomography shows that real 3D structure differs from theoretical equilibrium models [13]. Individual sizes of pores and cells in the same material differ significantly, so we can only talk about structure's average properties. We use common expression PPI (pores per inch) which describes the average number of pores per inch [5]. In experiment 10, 20 and 40 PPI materials will be used.

volume of fluid. Consequently elementary volume moves in radial direction out of impeller.

The phenomena is named forced vortex present in contrast to free vortex which occurs in inlet and outlet impeller region.

Fig. 4. Forces acting on elementary volume of fluid in rotating porous material in absolute coordinate system with elementary volume of fluid in rotating field [9]

Fig. 3. Magnified photo of used porous material showing pore and cell diameter [9] 1 THEORETICAL BACKGROUND This section considers fluid flow passing through a stationary and rotating bladeless impeller, and is necessary for understanding the operation of device and understanding the later presented experimental results. Fig. 4 represents forces acting on elementary volume of fluid in rotating porous material. Dotted line represents flow trajectory and v elementary volume velocity in absolute coordinate system. Fd represents drag force, which acts in the opposite direction of particle relative velocity. Fc is centrifugal force and is a consequence of rotation with angular velocity ω. Fp is pressure force which is a consequence of pressure difference in different sides of elementary volume acting on elementary volume.

Radial equilibrium theory [14] describes the passing of fluid through impeller without considering the shape of the blade, so it can be used to describe phenomena in a bladeless impeller. Basic assumption is that due to rotation, centrifugal force Fc acts on the elementary 466

Another approach to theoretically describe the combination of laminar and turbulent fluid flow in rotating material with high porosity present can be carried out with the upgraded Darcy equations, called also called DarcyForchheimer equations [15]. It can be deducted from Navier-Stokes equations [16, 17]:

 K

v  C  v  v  p  f

v  0

(2) (3)

Vector v represents fluid velocity, p pressure, f forces acting on the fluid (centrifugal, Coriolis forces, we neglect gravitation), μ dynamics viscosity, matrix K permeability, matrix C drag. The simplified system of equations is called Darcy-Forchheimer equations. In case of homogenous porous material and radial 2D fluid flow passing through the impeller, equations could be modified. Eq. (3) represents continuity equation for incompressible flow. We use coefficient of permeability k, coefficient of drag c, dynamics viscosity µ, and height of porous material in impeller h, and pressure difference dp to describe pressure drop in dependence on impeller radius:

Benedik, G. – Širok, B. – Rihtaršič, J. – Hočevar, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476

dp  Q Q2   c dr k 2    r  h 4  2  r 2  h2

(4)

We have assumed that pressure difference dp varies only with change of impeller radius r.

The integration of impeller inner radius R1 to impeller outer radius R2 gives pressure drop Δp:

p R1 , R2   R2

(

R1



Q Q2 c )dr k 2   r  h 4  2  r2  h2

(5)

selected measuring points, 3 mm away from impeller perimeter. Impeller was slowly rotated with frequency 0.1 s-1, while anemometer was in fix position. Change in velocity field on account of slow rotation was corrected. Pressure differences were measured with pressure gauges.

Sealing between impeller and housing was done by two O-rings, other connections were sealed with silicon kit. Air flow straightener is used to assure laminar air flow at the entrance of impeller.

Our goal is low pressure drop Δp and thus the use of material with a high coefficient of permeability k and low coefficient of drag c.

Decrease in pressure drop Δp could be achieved by increasing the height of impeller h and inner diameter R1 or decreasing outer diameter R2. The first part of equation is dominant at laminar flow and the second part at turbulent flow.

2 EXPERIMENT, STATIONARY IMPELLERS The experiment consists of two parts, both carried out at the same measuring station, measurements on stationary impellers, described in this section and measurements on rotational impellers described in section 3. The experimental procedure included the calibration of measuring station and hot-wire anemometer, measurement of integral pressure drop for different stationary impellers and local measurements with hot-wire anemometer.

Pressure drop measurements were performed at variable volume flow rates that were generated by an external flow generator. Measurements of local velocities were performed such that flow velocities were measured close to impeller perimeter using hot-wire anemometer. Impeller angular position was simultaneously measured with a potentiometer.

2.1 Experiment Description Pressure drops were measured in dependence on air volume flow rate with measuring station shown in Fig. 5. It consists of compressed air supply, volume flow rate measuring device, pressure gauge, stationary impeller and potentiometer.

Hot-wire anemometer was used to measure velocity in

Fig. 5. Measuring station: (1) hot-wire anemometer, (2) measuring points, (3) pressure gauge, (4) air volume flow rate measurement device, (5) compressed air supply, (6) axis, (7) housing, (8) air flow straightener, (9) O-ring sealing, (10) bladeless impeller, (11) potentiometer, (12) personal computer with A/D converter, (13) signal conditioning unit, (14) potentiometer controller, (15) hot-wire amplifier, (16) positioning table Table 1 shows six different impellers which were investigated with the following parameters: inner diameter R1, outer diameter R2, height of porous material h, PPI, average pore diameter and presence of inducer. Average pore diameter Dpore is giving the same information as PPI and is added only for easier comparison of different materials. Radial inducer, used with impeller 5 and shown in Fig. 2 b), is used to pre rotate the air before it enters porous material and thus reduce aerodynamic loses in the inlet region of impeller. It has 18 blades with inlet angle of 10°, outlet angle of 90°, inner radius of 23.5 mm and outer radius of 33.75 mm. It was manufactured by rapid prototyping technology Fused Deposition Modelling made of ABSmaterial. Darcy number Da for different materials is also presented in Table 1. It is defined as:

Flow Characteristics of Bladeless Impeller Made of Open Cell Porous Material

467

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476

nr. impeller 1 impeller 2 impeller 3 impeller 4 impeller 5 impeller 6

Table 1. Parameters of measured impellers R2 [mm] h [mm] PPI Dpore [mm] 62.5 8.5 40 0.625 62.5 8.5 20 1.250 62.5 8.5 10 2.500 62.5 12.5 40 0.625 62.5 12.5 40 0.625 62.5 12.5 40 0.625

R1 [mm] 23.75 23.75 23.75 23.75 33.75 33.75 Da 

k , 2 D pore

(6)

where k represents coefficient of permeability estimated with Ergun empirical equation [18].

2.2 Integral measurements of Pressure Drop Air flow rate is calculated with an air flow measurement device on the basis of pressure difference at the orifice k according to ISO 5167 [19]. Air volume flow rate Q was calculated:

Q k

p z

Z

,

(7)

where Δpz is pressure difference at the orifice and ρz density of air calculated according to air pressure, temperature and humidity. Integral measurement uncertainty relates to the uncertainty of pressure measurements with pressure gauges, measurement of air conditions (temperature, relative humidity and air pressure) and the uncertainty of volume flow rate measurements.

Common measurement uncertainty of air pressure Δp in dependence on air flow Q is estimated with relative uncertainty of 3%.

2.3 Local Measurements of Exit Velocity Field Because preliminary measurements have shown the dominance of radial velocity and presence of local velocity fluctuations, presence of turbulence and requirement to measure velocity field with high spatial resolution, the choice of velocity meter was important. It was found, that a five-hole probe is inadequate and that hot-wire anemometer is suitable for this type of meassurement [20]. 1D anemometer Dantec Mini CTA with sensor Dantec 55P11 was choosen. The sensor wire was 5 µm thick and 1.25 mm long. Sampling frequency was 5 kHz and sampling interval was 10 s. During the entire 468

inducer no no no no yes no

Da [/] 0.316 0.316 0.316 0.316 0.316 0.316

sampling interval impeller was rotated slowly and measurements of velocity and angle were performed simultaneously. The temperature of anemometer wire was set to 250°C. The distance between anemometer and porous material was 3 mm. Calibration and meassurement were carried out according to Bruun [21] and Jørgenson [22].

Measurements were performed at the distance of 3 mm from the impeller circumference.

Measuring points were located along impeler height h with steps of 0.65 mm, where h = 0 mm matched wiht impeller lower wall and h = 12.5 mm matched with impeller upper wall.

Measurement uncertainty of velocity field measurement is a result of anemometer calibration, measurement of environmental conditions, limited measurement time, anemometer position and potentiometer angular position measurement, and measurement of air volume flow rate. Common measuring uncertainty velocity field measurement is estimated with relative uncertainty of 2.8% [23].

2.4 Results of Integral Pressure Drop Measurements This section presents pressure drop measurements in dependence on volume flow rate for six different impellers, described in Table 1.

Fig. 6 shows pressure drop in dependence on air volume flow rate for different impellers. All measurement results can be accurately approximated with second order polynomial approximation in accordance with Eq. (5).

Dukhan [9] made similar conclusions doing experiments on 1D air flow through metal foam square. Pressure drop for impellers made of material with smaller pore sizes (comparing impeller 1, 2 and 3 with variable PPI, h = 8.5 mm, R1 = 23.75 mm) is slightly larger in entire volume flow rates than for impellers made of material with larger average pore diameter.

Benedik, G. – Širok, B. – Rihtaršič, J. – Hočevar, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476

Fig. 6. Pressure drop in dependence on air volume flow rate for different impellers Larger difference was expected according to measurements presented by Dukhan [9]. Smaller average pore and cell size diameter means more obstacles for fluid flow through material and larger boundary layer surface. As a consequence higher drag coefficient c and lower coefficient of permeability k appear, which reflects in higher pressure drop. With the increase of impeller height h from 8.5 mm to 12.5 mm (impeller 1 and 4) pressure drop is decreased by app. 50% due to the decrease of relative velocity according to Eq. (5). With the increase of impeller inner radius from 23.75 mm to 33.75 mm (impeller 4 and impeller 6) pressure drop is decreased by 50% due to shorter air flow path through porous material and reduction of highest relative velocities in impeller inlet region. By comparing impeller with and without radial inducer (impeller 5 and 6) we notice higher pressure drop at volume flow rates higher that 2 dm3/s when inducer is used. Pressure drop in inducer occurs as a consequence of small inducer inlet angle (10°) and high number of blades (18) that represent an obstacle for air flow in radial direction (Fig. 2 b).

parameters given in Table 1. Fig. 7 shows radial exit velocity in dependence on angular position of hot-wire close to disc perimeter at different volume flow rates Q. Signal was captured from time dependent hot-wire velocity signal and simultaneously angular position was measured.

Time averaging of signal was performed for the angle of 0.5°.

Radial exit velocity fluctuations shown in Fig. 7 are connected with structure of open cell porous material. Local distribution of pores and ligaments causes stochastic velocity fluctuations.

Angular positions of local maxima and minima are preserved in tolerances at different air volume flow rates, which confirms connection with material structure.

A histogram of radial velocity distribution for different volume flow rates is shown in Fig. 8.

Local standard deviations of radial velocity σ increase with the increase of volume flow rate Q from 0.11 to 0.34. Meanwhile, local degree of turbulence defined with equation:

2.5 Results of Local Exit Velocity Field Measurements

increases with the increase of air volume flow rate from 22% to 30%. It can be concluded that fluctuating properties of velocity field on macro scale are axially symmetrical and homogenous, which is also valid for geometric properties of open cell porous structure.

Radial velocity field around impeller perimeter will be presented. The measurement procedure is described in section 0. Only measurement on impeller 5 will be presented with





v rad

,

Flow Characteristics of Bladeless Impeller Made of Open Cell Porous Material

(8)

469

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476

Fig. 7. Time averaged exit radial velocities for different volume flow rate Q in dependence on angular position on stationary impeller 5 clear connection between local material structure and radial velocity field. As expected areas with low velocity are close to lower and upper wall of impeller (h = 0 mm and h = 12.5 mm). Results of local exit velocity field measurements have revealed flow properties and will be used in further research on rotating bladeless impeller.

Fig. 8. Radial velocity histogram at different volume flow rates Q with given local standard deviations σ and turbulence intensity χ on stationary impeller 5 2D radial velocity field on a segment of impeller’s outer diameter with the height of porous material at 12.5 mm and the angle of 35° is shown on Fig. 9. Surfer 8 software was used for the generation of Fig. 9 b) and c) using kriging method. A match between the location of pores on the surface and areas with higher radial velocity is obvious (Fig. 9 c). Non-matching can be assigned to several influences. Firstly, material structure under the surface which is not visible in the photo also has an influence on velocity field.

Secondly, shift of high radial velocity area is caused also by a non-radial component of velocity. Thirdly, there is an influence of measurement uncertainty as described in section 0. Considering the mentioned influences there is 470

Fig. 9. a) Photo of measured segment, b) Graphical presentation of 2D measured radial velocity field, c) Combination of photo and 1.5 m/s velocity contour 3 EXPERIMENT, ROTATING IMPELLERS This section includes measurements of integral characteristics and local measurements of exit velocity field close to impeller’s outer

Benedik, G. – Širok, B. – Rihtaršič, J. – Hočevar, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476

perimeter. Table 1 shows parameters of measured impellers.

3.1 Integral Measurements of Characteristics Integral measurements of characteristics include the measurement of impeller aerodynamics efficiency and static pressure in dependence on air volume flow rate. The measurements were performed in Domel d.d. at the measuring device in accordance with IEC 60312 [24]. Impellers were driven by an electronically commutated electric motor with constant voltage supply. Static pressure measurements were converted to constant rotational speed 10,000 min-1. Aerodynamic efficiency ηaero is the efficiency of aerodynamic part of turbo machine, therefore, we will simply name it impeller aerodynamics efficiency. It includes the influence of impeller used, sealing between impeller and cover, bladeless diffuser and exit openings in the cover. However, sealing between impeller and cover, bladeless diffuser and exit openings in the cover were not changed when measuring with different impellers.

Aerodynamic efficiency ηaero is given as a ratio of air power Pair and input electrical power Pel to motor divided by motor efficiency ηm:

aero 

Pair Q  p ,  Pel m Pel m

(9)

Q is measured air volume flow rate and Δp is the measured pressure difference. Efficiency ηm of used electronic commutated motor was measured in the entire working regime. The motor shaft was loaded with torque from dynamometer Magtrol type 2WB43-HS. Electric power, torque and rotational frequency were measured. Torque was changed from 0 to 26 Ncm in steps of 2 Ncm. Results were presented with normalized aerodynamic efficiency, given as:

aeron 

aero aero max

,

(10)

where ηaeromax represents maximal aerodynamic efficiency measured with impellers (Table 1).

Dimensionless air volume flow rate φ and dimensionless static pressure ψ were also used when presenting results: Q ,  2  3 (11) ( 2  R2 )  n 4



p

2

( 2  R2 ) 2  n 2

.,

(12)

2 The uncertainty of integral measurements of characteristics refers to the uncertainty of power measurement, change in motor efficiency as a consequence of heating, motor efficiency measurements, pressure measurements, measuring device calibration and measurement of environmental conditions. Common measuring uncertainty is estimated with relative uncertainty of 3.0% [4, 20, 24].

3.2 Local Measurements of Exit Velocity Field Local measurements of exit velocity field were performed in Laboratory for Water and Turbine Machines at the Faculty of Mechanical Engineering in Ljubljana with calibrated 2D hotwire anemometer Dantec Mini CTA with sensor 55P62. Sensor wires are 5 µm thick and 1.25 mm and 1.25 mm long. Sampling frequency is 50 kHz and sampling interval is 2 s. Other settings are identical to the ones, described in section 0.

Measuring station is presented in Fig. 10.

The electric motor consists of rotor, stator, housing and cover. Impeller is attached to the shaft of the motor. The orifice for the regulation of air volume flow rate and hot-wire anemometer with positioning table are also shown.

Measurements were carried out along the height h1 in 20 measuring points in 1 mm steps. At h1 = 4 mm the height of a measuring point matched with the lower wall and at h1 = 16 mm the height of measuring point matched with the upper impeller wall. Anemometer position was changed with positioning table. The air enters through the orifice, travels through the impeller and exits through the openings in the cover. Open loop regulation rotational frequency was between 8,000 min-1 and 10,500 min-1.

Measurement uncertainty of velocity field measurement on the rotating impeller consists of anemometer calibration, temperature measurement, measurement of environmental conditions, limited measurement time, anemometer position and potentiometer angular position, working point setting, distance between sensor wires, volume flow rate measurement and decrease in generated air volume flow rate due to motor heating. Common measuring uncertainty is

Flow Characteristics of Bladeless Impeller Made of Open Cell Porous Material

471

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476

estimated with relative uncertainty of 3.2% [23, 25].

Fig. 10. Schematic presentation of the measuring station for local measurements. (1) hot-wire anemometer, (2) measuring points, (3) electric motor stator, (4) electric motor rotor, (5) power electronics, (6) constant voltage supply, (7) housing, (8) exit openings in the cover, (9) impeller, (10) cover, (11) orifice, (12) personal computer with A/D converter, (13) signal conditioning unit, (14) hot-wire amplifier, (15) positioning table 3.3 Results of Integral Characteristics’ Measurements Diagram in Fig. 11 shows normalized aerodynamics efficiencies of different impellers in dependence on dimensionless air volume flow rate. We compare impellers 1, 2 and 3 with same geometric parameters, but made of porous materials with different average pore diameter.

Impeller 1 with 40 PPI and h = 8.5 at volume flow rate 0.06 achieves approximately 6% higher normalized efficiency than impellers 2 and 3 with 20 PPI and 10 PPI and the same height h. The difference between impellers with 20 and 10 PPI at h = 8.5 mm is small and lies within measurement uncertainty. At volume flow rate of 0.023 we notice that impeller 3 with 10 PPI has the best efficiency.

We focus on the height h of impeller (Impeller 1 and 4). We notice better aerodynamic efficiency of impeller 4 with h = 12.5 mm for flow rates 0.005 and more. Difference increases with the increase of air volume flow rate, at air volume flow rate 0.025 the difference is 30%.

If we compare impellers with different inner radius R1 (Impeller 4 with R1 = 23.75 mm and impeller 6 with R1 = 33.75 mm) we notice 472

better aerodynamic efficiency of impeller 4 for volume flow rate of up to 0.021 and worse aerodynamic efficiency for volume flow rates of 0.022 and more.

When comparing impellers with and without an inducer (Impeller 5 and 6), we can observe that inducer significantly improves aerodynamic efficiency in the entire working region. At volume flow rate 0.025 inducer improves aerodynamic efficiency by 30%.

Dimensionless pressure differences of different impellers in dependence on the dimensionless air volume flow rate are presented in Fig. 12. We compare impellers 1, 2 and 3 with same geometric parameters, but made of porous material with different average pore diameters.

Static pressures turn out to be the same (within the measurement uncertainty) for volume flow rate of up to 0.19. At higher volume flow rates impeller 3 made of 10 PPI material generates slightly higher static pressure.

We focus on height h of impeller (impeller 1 and 4). We notice higher static pressure when impeller 4 with higher height h is used. The difference increases from 3% at zero volume flow rate to 22% at volume flow rate of 0.026.

Comparing impellers with different inner radii (impeller 4 with 23.75 mm and impeller 6 with 33.75 mm) we notice higher static pressure of impeller 4 for volume flow rate of up to 0.018 and lower at volume flow rates exceeding 0.018.

Comparing impellers with and without an inducer (impeller 5 and 6), we see that inducer significantly increases static pressure in the entire working regime.

For the explanation of results we can compare Fig. 11 and Fig. 12 with Fig. 6. By comparing impeller 1, 2 and 3 we can explain better integral characteristic of impeller 3 at higher volume flow rates by slightly lower pressure drop if material with larger average pore diameter is used (10 PPI). Explanation of better integral characteristics of impeller with higher height (impeller 1 and 4) and impeller with larger inner radius (impeller 4 with 23.75 mm and impeller 6 with 33.75 mm) is also due to lower pressure drop as seen in Fig. 6.

Maximal non-normalized aerodynamic efficiency is 35%. That equals to normalized aerodynamic efficiency 1.0 in Fig. 11 when impeller 5 was used. Existing bladed impeller has maximal non-normalized efficiency of 48%.

Benedik, G. – Širok, B. – Rihtaršič, J. – Hočevar, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476

Fig. 11. Diagram of normalized efficiency of impellers in dependence on dimensionless air volume flow rate

Fig. 12. Dimensionless static pressures of different impellers in dependence on air dimensionless volume flow rate Integral characteristics of bladeless impellers could be improved. The most important disadvantage of porous impeller is flow condition at the entrance of airflow into the impeller. The geometry of porous structure does not allow efficient air flow into porous impeller. This

disadvantage could be to large extent eliminated by using an inducer. This can be confirmed by the significant improvement of integral characteristic when radial inducer was used (impeller 5 with and 6 without inducer). Optimisation of inducer should additionally reduce loses at the entrance of

Flow Characteristics of Bladeless Impeller Made of Open Cell Porous Material

473

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476

impeller. Several types of inducers are possible, from solely axial, radial-axial to solely radial inducer.

The second possibility for optimisation is to change the geometry in flow tract. As shown in Fig. 6 and Fig. 12 pressure drop in porous material is responsible for changes in static pressure characteristics. As pressure drop is high in regions where air flow velocity through porous material is high; that is in the entrance part of impeller, we propose an increase of impeller height h in this particular region.

3.4 Results of Local Measurements of Exit Velocity Field Local velocity measurement results of radial and axial velocity field will be presented in the following section. Measurements were performed according to the experimental procedure with impeller 5, presented in section 0.

Fig. 13 shows exit velocity field profiles close to impeller perimeter in dependence on height h1. Fig. 13 a) presents radial and b) tangential velocity field profile. Both radial and tangential velocities decrease as air volume flow rate decreases. In Fig. 13 a) at zero flow rate we notice a phenomena of recirculation (Q = 0 dm3/s). Air exits the upper part of impeller (9 mm < h1 < 18 mm) and enters the lower part of impeller (4 mm < h1 < 9 mm). Negative radial flow velocity at height h1 of more than 17 mm at all measured volume flow rates shows air circulation between the impeller upper plate and cover. In order to track fluid flow trajectories, not only at exit from impeller, but also through metal

a)

474

foam, a visualisation method should be considered. Method similar to the one used by Rihtaršič [26] or Eberlinc [27] should be applied.

4 CONCLUSIONS This paper presents a bladeless impeller, in which the energy is transferred over a structure of porous material. We provide a theoretical background of fluid flow passing through rotating and stationary bladeless impeller. Models of different impellers are described and manufactured. The analysis of fluid flow through stationary impellers has shown that in order to decrease pressure drop impeller inlet diameter and impeller height should be increased. The choice of material average pore diameter also has an influence on pressure drop, but its influence is limited. Radial velocity field close to impeller perimeter was measured and velocity fluctuations have shown its connection with local material structure. Radial velocity above the location of pores was evidently higher than radial velocity above the ligaments.

Integral measurements of rotating bladeless impeller have shown static pressure and air flow generation as assumed in the theoretical part. The impeller with increased height, increased inner diameter and with radial inducer has shown the best integral characteristics. It was found that the inducer significantly improves flow conditions at the entrance of the impeller. We assume there is a possibility to improve integral characteristics of impeller by optimizing the inducer design and with variable height of impeller.

b) Fig. 13. Exit velocity field profiles in dependence on height h1, a) radial and b) tangential velocity

Benedik, G. – Širok, B. – Rihtaršič, J. – Hočevar, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476

ACKNOWLEDGEMENT Research was funded by European social fund and Ministry of Higher Education, Science and Technology. Research was supported by the Faculty of Mechanical Engineering, University of Ljubljana and company Domel d.d.

REFERENCES [1] Lakshminarayana, B. (1996). Fluid dynamics and heat transfer of turbo machinery.

Aerospace Engineering, Pennsylvania, USA. [2] Čudina, M., Prezelj, J. (2007). Noise generation by vacuum cleaner suction units, Part 1, Noise generating mechanisms.

Applied acoustics, vol. 68, p. 491–502.

[3] Širok, B., Benedik, G., Močnik, A. (2008).

Rotor centrifugalnega turbostroja. Patent announcement, P-200800232, Ljubljana.

[4] Benedik, G., Širok, B., Hočevar, M., Močnik, A. (2009). Research on bladeless impeller from porous open cell material.

Conference Kuhljevi dnevi 2009, Brnik.

[5] Liu, P. S. (2004). Tensile fracture behaviour of foamed metallic materials, Materials science and engineering, vol. 384, p. 352– 354.

[6] Tesla, N. (1913). Fluid propulsion P1061142. Patent announcement, New York.

[7] Rice, W. (1991). Tesla turbo machinery, International Nikola Tesla symposium, Arizona, USA. [8] Norbert, M. (2000). Patent DE 19838265.

Patent announcement, Kunzelsau, DE. [9] Dukhan, N. (2006). Correlations for the pressure drop for flow through metal foam.

Exp Fluids, vol. 41, p. 665–672.

[10] Boomsma, K., Poulikakos, D., Ventikos, Y. (2003). Simulation of flow through open cell metal foam using an idealized periodic cell structure, International Journal of Heat and fluid flow, vol. 24, p. 825–834.

[11] Auriault, J. L., Geindreau, C., Royer, P. (2002). Coriolis effects on filtration law in

rotating porous media. Transport in Porous Media, vol. 48: p. 315–330.

[12] Sawicki, E., Geindreau, C., Auriault, J. L. (2005). Coriolis effects during fluid flow through rotating granular porous media, Studia Geotechnica et Mechanica, vol. 27, no. 1–2, p. 143-154.

[13] Borovinšek, M., Vesenjak, M., Matela, J., Ren, Z. (2008).

Computational reconstruction of scanned aluminum foams for virtual testing. Journal of the Serbian Society for Computational Mechanics, vol. 2, no. 2, p. 16-28.

[14] Turton., R. K. (1984). Radial equilibrium theory, Principles of turbo machinery. USA. [15] Kramer, J., Jelc, R., Škerget, L. (2009).

Modeling of turbulent flow in porous media.

Conference Kuhljevi dnevi, Brnik.

[16] Jecl, R., Škerget, L., Kramer, J. (2005).

Comparison between the Forcheimer and the Brinkman model for convective flow in porous cavity with boundary domain integral method. Acta hydrotech., vol. 38, p. 1-17.

[17] Vafai, K., Hadim, A., H. (2000). Handbook of porous media, Section 9: Flow and thermal convection in rotating porous media, Marcel Dekker, New York.

[18] Dukhan, N., Patel, P. (2008). Equivalent particle diameter and length scale for pressure drop in porous metals, Experimental and thermal fluid science, Detroit, USA, vol.

32, p. 1059–1067.

[19] EN ISO 5167-1 (1995). Durchflussmessung von Fluiden mit Drosselgeräten, Fluid, Drosselgerät, Leitung, Kreisquerschnitt.

[20] Eberlinc, M., Širok, B., Hočevar, M., Dular, M. (2009). Numerical and experimental investigation of axial fan with trailing edge self-induced blowing. Forsch Ingenieurwes, Springer-Verlag, vol. 73, p. 129–138.

[21] Bruun, H.H. (1995). Hot-wire anemometry Principles and signal analysis, Oxford university press, New York.

[22] Jørgensen, F.E. (2005). How to measure turbulence with hot-wire anemometers, A practical guide, Dantec Dynamics.

Flow Characteristics of Bladeless Impeller Made of Open Cell Porous Material

475

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 464-476

[23] Eberlinc, M., Širok, B., Hočevar, M. (2009).

Experimental investigation of the interaction of two flows on the axial fan hollow blades by flow visualization and hot-wire anemometry. Experimental thermal and fluid science, vol. 33, p. 929–937.

[24] IEC 60312: 1998+A1:2000+A2:2004 (2005) Vacuum cleaners for household use – Methods of measuring performance.

[25] Benedik, G., Markič, I., Močnik, A., Širok, B., Hočevar, M., Rihtaršič, J. (2007).

Optimizacija rotacijskega izločevalnika za vodne sesalnike. Ventil, Ljubljana, vol. 13 no. 4, p. 250-256.

476

[26] Rihtaršič, J., Šubelj, M., Hočevar, M., Duhovnik, J. (2008). Flow analysis through the centrifugal impeller of a vacuum cleaner unit. Strojniški vestnik - Journal of mechanical engineering, vol. 54, no. 2, p.

81-93.

[27] Eberlinc, M., Dular, M., Širok, B., Lapajna, B. (2007). Influence of blade deformation on integral characteristic of axial flow fan.

Strojniški vestnik - Journal of mechanical engineering, vol. 54, no. 3, p. 159-169.

Benedik, G. – Širok, B. – Rihtaršič, J. – Hočevar, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490 UDC 005.72: 005.74

Paper received: 17.03.2009 Paper accepted: 25.01.2010

Application of Group Technology in Complex Cluster Type Organizational Systems Slobodan Morača1,* - Miodrag Hadžistević1 - Igor Drstvenšek2 - Nikola Radaković 1 1 University of Novi Sad, Faculty of Technical Sciences, Serbia 2 University of Maribor, Faculty of Mechanical Engineering, Slovenija The aim of this research was to contribute to the development of structural design procedures of complex - cluster type organizational systems. Industrial clusters can help companies to improve their own market positions, effectiveness, productivity and product quality. Organization of the production process in a company is an extremely complex process itself, and when it is transferred to the cluster level, the result is a complex task which is difficult to solve. For that purpose, this paper analyses the conditions and possibilities that would enable those structures to adapt to changes in the surroundings flexibility and management adequacy of production and organizational structures - by lowering the degree of complexity.

For the time being, no simple models which would enable an increase of process effectiveness in complex organizational units like clusters have been developed. One of the possible solutions which would decrease the complexity of flows and increase process effectiveness within an industrial cluster is the application of Group approach. ©2010 Journal of Mechanical Engineering. All rights reserved.

Keywords: industrial clusters, group technology, planning, work cells, complexity, flexibility 0 INTRODUCTION Modern concepts of increasing the effectiveness of production are based on the processes of automation, the application of modern materials and IT technology. They significantly reduce production costs, increase productivity and reduce the need for labour.

However, despite the revolutionary application of modern technology, the end of the 20th century and the beginning of the 21st century is further characterized by increased mobility of investments and recession, which is visible in the most developed countries, where the modern technology is most applied. All that has resulted in a constant decrease of production, which directly caused a decrease of employment rate, an increase of company debt and reduced possibilities of investments in new development projects.

In a competitive environment success of an organization is a function of industry attractiveness, its relative position in the industry, and the activities (strategy) it undertakes to remain ahead of others ([1] and [2]). Mintzberg explained that strategy is an evolutionary and organic process which is unpredictable; [3] explained that core competence gives an

organization competitive capability and remains central to its strategy planning process. Small and medium organizations encounter different kinds of problems such as resource limitations (especially human and financial resources), and market information [4], they face competition within and between large organizations [5].

Analyses have showed that the reasons for these problems are not only the inability of companies or their production or service systems.

Changes occur apart from how a company is capable to independently decrease its production costs or to increase the range of products.

Changes often depend on other economic and non-economic entities, geo-political factors and changes on the global market.

COMPANY

Unreliable Vendors

WC Scrap

Capacity Imbalances

Fig. 1. The need for increased accumulation

*

Corr. Author's Address: University of Novi Sad, Faculty of Technical Sciences, Trg Dositeja Obradovica 6, Novi Sad, Serbia, moraca@uns.ac.rs

477

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490

For a company to create specific conditions for the business, it has to provide sufficient working capital (WC) and the accumulation big enough to be utilized in case of an unpredictable situation - see Fig. 1.

The first problem is obvious: increased accumulation affects the increase of production costs, decreases turnaround coefficients of capital and flexibility. On the other hand, the users’ demands increase on a daily basis. They look for a variety of products, tailored to their requirements, which must be produced in small series adjusted to a small number of users.

Customer demands are contradictory to each other: the period of development of a new product should be as short as possible and the price of a new product should be either the same or even lower than the previous one. It brings us to the second problem: how to establish a system big enough to withstand the demands of the global market related to diversity, range and the amount of products and to be flexible enough to adapt to the changing demands of users.

To make this happen a continuous financing of development needs to be provided, an increase of coefficient of capital turnover, volume of production which will provide an optimal use of production capacities, possibility for flexible specialization, possibility to enter the global market, well trained and capable experts, as well as trust and connections on a very important local level.

It is clear that without the above mentioned elements the survival of companies is not possible, but it raises a question of how to achieve this in today's or tomorrow's business conditions? Investments in development are limited, so companies mainly have to find their development paths on their own, as well as their positions on the global market. One of the important development strategies which also provides competitive development, especially of Small and Medium-sized Enterprises (SMEs) and Regions, is to associate and develop complex organizational structures – clusters and business networks. Large enterprises merge and become even larger, and the best example is automotive industry. Small companies can survive on the market only if they associate with each other into systems which simulate a large enterprise, but maintain their flexibility.

478

Although this form of associating provides a lot of advantages, which will be mentioned in the Chapter 1, there are also many problems in functioning of complex systems. One of these problems is how to organize and manage such complex systems and establish effective production process? In chapters 2 and 3, we will describe a group approach as a possible solution.

1 CLUSTER AS A FORM FOR COMPANIES TO ASSOCIATE Companies are constantly asked to improve performances in order to get the chance to maintain or to improve their own market positions and financial situation. Clusters have the possibility to develop their own specific mixture of competitive advantages which is created on the basis of locally-developed knowledge as a result of mutual relations, cultural heritage and local characteristics. This is evident in the focus on clusters as an important concept in understanding growth and in thinking about development policy [6].

The idea of localized economies of scale in geographic agglomerations has a long history in economics, going back to Adam Smith’s early observations of labour specialization and to [7] explanations of why companies continue to localize in the same areas. Clusters arise in the presence of Marshallian externalities, which signify that companies benefit from the production and innovation activities of neighbouring companies in the same and related industries. There is abundant evidence that such externalities exist and lead to industry-level agglomeration [8].

Development of clusters is an effective way to improve business operations and bring it to a higher level. Modern business is based on fast response, quality, flexibility, innovation, connections and building the critical mass of capital and production / service potential. This relatively new style of doing business requires a team approach on the local level - cluster approach.

Clusters represent complex organizational systems that are flexible and can quickly be adjusted to oscillatory changes at the sale and purchase markets, generate employment, help the diversification of economic activities and make a significant contribution to exports and trade. Clusters also play an important role in

Morača, S. - Hadžistević, M. - Drstvenšek, I. - Radaković, N.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490

innovation and businesses where there is a need for application of modern technology. Thanks to their innovative flexibility, many of them become more productive and efficient than some large international corporations. In this process, emphasis should be put on creating a friendly business environment where the transformation of society towards a market economy shall take its place.

Cluster differs from other forms of associations within its geographical boundaries, involvement and utilization of funds, ways of exchange of products and partially finished products, information management - knowledge chains, and the importance of how they are connected. Clusters can be best understood and used as regional systems. According to Porter [9] they represent, "Geographic concentrations of mutually connected companies, specialized suppliers, service providers, companies from similar industries and institutions tied to them (i.e. universities, standardization agencies, trade unions), who compete, but also cooperate".

Basic characteristics of clusters are:  Clusters are based on systemic connections among companies; ties can be built on common or complementary products, production processes, essential technologies, needs for natural resources, demands for certain qualifications and/or distribution channels;  Clusters are geographically limited, defined mainly by distance and time that people are willing to take because of employment which job makers and company owners consider reasonable for meeting and creating business relationships; geographical range is under strong influence of travel and traffic systems, but also of cultural identity, personal priorities, and family and social conditions;  Clusters represent natural connection of companies. It must be emphasized that clusters do not operate as an imposed agglomeration, or forced association for any reason; clusters nourish unique attributes of companies and make it possible for them to choose levels and types of cooperation within a cluster, and to define what part of its capacities they will bring into clusters, and what part will remain "freelance", taking into consideration common needs, but also their

own benefits as a member of cluster association.  Associating into a cluster can bring a broad range of benefits to all partners as well as to the economy in general. Some of the benefits are the following:  Increased level of expertise; associating gives companies better knowledge about supply chain and makes it possible for companies to learn from each other and to cooperate;  Capability of companies to join complementary strengths and contract new works of larger scope for which, individually, they would not be able to bid in a public tender procedure;  Potential for large scale production (economy of scale), which can only be realized via specialized production in each of the companies, through joint purchase of supplies with large discounts or through joint marketing;  Strengthening of social and other informal connections, which leads towards creation of new ideas and new companies;  Better information flow within a cluster, e.g. making it possible for investors to identify good entrepreneurs, and for business people to find good service providers;  Enabling development of services’ infrastructure: legal, financial and other specialized business services.

As a result of the functioning of clusters, there are effects shown in Fig. 2:

COMPANY

Unreliable Vendors

WC Scrap

Capacity Imbalances

Fig. 2. Reducing inventory reveals problems so they can be solved This paper focuses on the establishment of organizational and managerial mechanisms within a cluster, which will enable an increase of production processes' effectiveness to the level of a cluster as a whole. That is why one of more important segments is to determine the levels of

Application of Group Technology in Complex Cluster Type Organizational Systems

479

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490

specialization in companies – participants in a cluster, and what desirable levels of specialization for more effective business are in case of specialization or in other words, economic diversity. Research shows that traditional production sectors are inclined to do better business when densely concentrated in one geographical area. Contrary to this, newer, hightech and service sectors are more comfortable with economic diversity environment.

General opinion is that specialization means lack of economic diversity and vice versa.

If that is the case, then improving industrial clusters bears a risk of creating highly specialized local economies. If local economies are specialized in only one industrial sector or a couple of them, then they are indeed much more sensitive to cyclic falls in those sectors. However, other opinions suggest that specialization and diversification do not necessarily exclude each other. Malizia and Feser [10] define economic diversity as "existence of multiple specializations". This means that is possible for local economies to be highly specialized in certain sectors and, at the same time, to have sound combination of economic activities. So we come to the concept of flexible specialization, which represents the possibility of companies to do what they do best, and cluster has the obligation to provide optimal utilization of capacities.

The establishment of organizational and managerial structures in complex organizational systems like cluster represents a big challenge because of diversity of clusters and characteristics of member companies. One of the possible models whose application shall enable the optimal use of the potentials of clusters is Group approach which is described in more details in Chapter 2.

2 THE GROUP APPROACH IN DESIGNING MATERIAL FLOWS The concept of Group Technology [11] is based on the simplification and standardization process, which originated at the beginning of 20th century. It emerged as a single machine concept that was created to reduce setup times [12].

Group approach in the design flows of material in the production system based on the idea of group technology which, since the work of Mitrofanov

480

[13], never stopped being up-to-date in scientific and expert circles.

This concept was further extended by collecting machine parts with similar requirements, completely processing them within a machine group or cell [14]. The ideas for Group approach came from the fact that there is similarity in objects which enter the production process of any company and that in real conditions there is a limited number of forms of these objects. In the core of Group technology set up by Mitrofanov is a unification of objects with similar characteristics into families. This creates conditions for an "increase" of the number of objects in a series and thus levels of series, along with gaining a range of effects during preparation of production and production itself:  Orientation of process engineers toward a narrow area when solving problems of designing technological procedures – there are small differences between objects within a group in regard to shape, measures, quality, materials used, etc.  Application of standardized – typical technological procedures for all objects within a group: same production flow, work places, same or similar tools for positioning of objects (possibly group accessories), same tools, same or similar processing modes, etc.  Simplification of preparation of work places when transferring from one object to another and reduction of time for preparations and finalization.

Based on ideas of Group technology of Mitrofanov, as well as the results of the research made by Burbidge [11], the new approach in production was developed: Group approach to design of effective industrial structures.

Using this approach, based on a classification of objects within the production process, groups of geometrically and technologically similar objects are created – operational groups (families), which represent the basis for Group approach in the planning of production technology. However, matters have been taken further here, by merging individual operational groups which have mutually similar technologies (using the same work places) into larger groups. By assigning all the necessary work places into a created large group, we create a so called working unit (production cell, work cell), capable for the production of all objects.

Morača, S. - Hadžistević, M. - Drstvenšek, I. - Radaković, N.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490

Fig. 3. Working Unit (WU) – the basic changes in approaches for production structures designing [15] Working Unit has all the characteristics of Production Cell but besides its executive (production) independence it has to have an organizational and controlling independence too, which means its total responsibility for quantity, quality, and delivery terms of similar working objects, and also for organizing and managing of processes [15]. The final result is, as shown in Fig. 3 that the entire production program is divided in parts of the program - a group of objects, and the whole production system into independent operating units in which some parts of the program are made. At the same time, each part of the production program consists of previously shaped operational groups of mutually very similar objects.

Apart from work places for production, as shown in Fig.4, other resources join the composition of a working unit (technological preparation, operational preparation, distribution of materials and tools, process QA, operational maintenance), which gives independent (autonomous) unit – a part of production process which is fully capable to produce one separate

component of production program. This approach in designing material flows in a production system provides a range of advantages, including the following most important ones:  Significant simplification of material flows – with shorter transport paths (instead of one flow of all production program objects, several flows with a smaller number of objects is realized in the entire production system through smaller sets – parts of production system),  Simplified production management (each working unit is managed independently, where the number of launched objects and work places which are managed is significantly smaller in comparison to management of the entire production system),  Production related problems, management, quality control, maintenance, etc, are located in much smaller parts of the production processes – work units, which has positive effects when it comes to responsibility for work and motivation of participants in the work process.

Application of Group Technology in Complex Cluster Type Organizational Systems

481

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490



Working unit 5

Working unit 4

Working unit 3

Working unit 2

Working unit 1

PRODUCTION SYSTEM

Foreman

Plastic parts

Tool issuer

Entry of material

Rotational parts Jagged parts

Scheduler

WORKING UNIT

Prismatic parts

Tin parts

PRODUCTION PROGRAM

Output of product Quality control

Operational maintenance

Fig. 4. Production system designed on the basis of Group approach Presented Group approach in designing material flows has been applied in a large number of companies and described in details in [16] and [17], and the process of clustering and the formation of business units has been supported by a computer program system APOPS-08 [18].

The major advantages of Group technology have been reported in literature as reduction in setup time, reduction in throughput time, reduction in work-in-process inventories, and reduction in material handling costs, better quality and production control, increment in flexibility, etc., [19], [20] and [21].

3 ADVANTAGES OF THE APPLICATION OF GROUP APPROACH IN CLUSTER ORGANIZATIONS Productivity and productivity growth determine prosperity. Innovation is a key driver of productivity growth. Clustering supports both productivity and innovation. Porter's Diamond theory provides a useful concept that can help businesses, government and other institutions to explore improvements in the productivity environment. Various models and solutions have been extensively studied in literature. These

482

models can be divided into the following categories:  Integration of production planning at the level of industrial clusters  Integration of production planning at the level of companies within the industrial cluster  Integration of production planning and distribution on the spot of procurement of raw materials, transport and distribution of semi or finished products to customers.

The aim of this paper is to present the application of the Group approach as a model of optimization of planning and programming production processes in complex organizational structures like clusters. The application of group technology in cluster produces savings and benefits in almost every area of the business:  It combines tasks, equipment, gages, tooling and schedules into larger groups of similar elements for similar solutions.  Purchasing can group similar parts and achieve quantity discounts. For non-standard purchased parts, grouping helps suppliers achieve savings and reduce price.  Accounting in industrial cluster is simpler in a group technology - costs are collected by cell and family rather than individual part.

Morača, S. - Hadžistević, M. - Drstvenšek, I. - Radaković, N.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490

Cluster production program can be diversified and composed of all products which are made by the member companies. Disparity in regional economic development is strongly influenced by the proportion of trade, local industries, resources and mix of organizations present in the cluster [22]. Participating companies can enter a cluster with only one part of their production program, and produce or distribute other products on their own, or in cooperation with companies which are not in their cluster. It is necessary to define basic products which are offered by a cluster, and adjustments of organizational and managerial cluster structures is done in regard to these products. Production program is further divided into structures and sub- structures, where individual requests towards cluster companies are defined for processing and assembling. Possibilities for process control and the shortening of production cycle depend on organization of a cluster.

Organization of the production process in a company is extremely complex process itself, and when we transfer it to the cluster level, we

get a complex task which is difficult to solve. For the time being, there are no simple models developed which would enable an increase of process effectiveness in a complex organizational units like clusters. In that regard, this paper makes a pioneering attempt. One of the possible solutions which would decrease the complexity of flows and increase process effectiveness within a cluster is application of Group approach.

By applying a Group approach in complex cluster type organizational systems, the role of work units from the Fig. 4 is replaced by cluster member companies, as shown in Fig. 5.

Previously, we stated that one of the significant characteristics of clusters is flexible specialization of companies for processing and assembling of structures and sub- structures from cluster production program. It enables the processing of structures and sub- structures with minimum costs and minimum time required.

In accordance with the Group approach the parts for processing are grouped according to two criteria: similarity of parts and potential of production system.

Fig. 5. Production realization within a cluster in accordance with the Group approach

Application of Group Technology in Complex Cluster Type Organizational Systems

483

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490

Application of the Group Technology on complex Cluster type organizational systems represents a new approach in creating effective production systems. Given approach is based on concepts of flexible specialization and Working Units with extended flexibility. Flexible specialization, as one of the basic advantages of Clusters, provides companies in Cluster to work on what they do best, for what they have trained labor force or technical-technological capacities, and still to have enough volume of work.

Companies, Cluster members, considered from the aspect of flexible specialization represent Working Units of extended flexibility.

Having in mind that Companies participating in a Cluster can choose which part of the Cluster production program or production capacity they will be part of, then the same applies for branches of the Companies as well.

Application of the Group Technology covers many issues. On the basis of the Analyses of the methods applied in designing technological procedures and designing the organization of work processes in Clusters on the territories of Serbia, Croatia, Slovenia and Italy, there have been determined the basic processes of Application of Group Approach on the level of an Industrial Cluster:  Harmonizing a common Cluster production program,  Classification of objects of work: o Adjusting the Systems of Classifications of objects of work according to the increase of performances of technicaltechnological systems of Work Units with extended flexibility, o Defining the Systems of Classifications of companies participating in Clusters and companies cooperating with a Cluster from the aspect of performances of technical-technological systems, organizational and managerial structures, o Defining correlations between the above mentioned Systems of Classifications.

Adjusting organizational and managerial structures of Clusters and member companies which will provide both, a more efficient information flow amongst the companies in a Cluster and an increased quality in controlling working processes.

484

3.1 Harmonizing a Production Programe

Common

Cluster

Companies in a Cluster have to harmonize which products, assembles, subassemblies and parts are important on the Cluster level from the aspect of requests coming from the environment and from the aspect of companies participating in producing them. In that way, two basic goals are accomplished: directing activities towards fulfilling customer demands and creating the synergy effect amongst the companies participating in the production. Research carried out in the period 2007 - 2009 by the Center for Competitiveness and Cluster Development both individually and also participating in GIFIP1, and UNIDO projects supporting development of the Cluster AC Serbia, demonstrate that without the existence of the above mentioned elements it is very difficult to accomplish effective functioning of Clusters.

3.2 Classification of Objects of Work Production program of a Cluster can comprise a huge number of different elements assembles, subassemblies or parts (Fig. 6 on the left). These elements can differ in regard to shape, material, technical-technological specifics etc.

Also, these elements are an integral part of different products which can be produced in different companies. For each of the individual elements produced in a Cluster, it is necessary to define the technological procedure starting from geometrical and technological characteristics of an element which, in case of a huge number of elements, requires a considerable waste of time.

The Group approach has in its basis the procedure of grouping of objects according to their similarities. In order, from non-homogenous group of elements (Fig. 6 on the left), to make a homogenous group of elements (Fig. 6 on the right), it is necessary to have the existence of Unique System for Classification which is applied on the level of the whole Cluster. When the homogenous groups of elements are generated, then designing technological procedure for a Group is carried out.

1

Bilateral cooperation programme Italy – Serbia : Integrated Governance of productive companies in sectoral clusters (GIFIP)

Morača, S. - Hadžistević, M. - Drstvenšek, I. - Radaković, N.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490

Fig. 6. Ungrouped and grouped parts Finally designing individual technological procedures including utilizing the defined technological procedure for a Group as the starting point.

It is essential that due to similarities of elements in a Group, there are existing technological procedures covering the whole Group which reduces the waste of time in regard to individually defined technologies.

Modification of an application of the Group approach in a Cluster also lies in the fact that the process of designing a Group technology is placed on the Cluster level - which significantly relieves resources of participating companies and decreasing the costs.

In practice, a series of more or less similar systems of classification have been developed.

All developed systems provide gradual

classification in terms of identifying classes, families and groups - types of parts with similar characteristics and specific measurement areas.

Defining operational groups at the clusters level brings certain limitations in the implementation of classification systems. Classification system KSIIS-08* developed for the needs of the industrial systems of geometrically shaped products, basically includes characteristics related to design operational groups in a relatively simple way. The structure of the system is schematically shown in Fig. 7.

Depending on the combination of technical-technological systems of companies, it is later chosen which company will process which group of selected parts including specific operations.

Fig. 7. Structure of the Classification system Application of Group Technology in Complex Cluster Type Organizational Systems

485

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490

Fig. 8. Uses a matrix of part numbers and machine numbers to group families General characteristics of the above mentioned Classification system are the following:  classification label has 14 areas - features (1 to 14),  each feature has 10 fields (0 to 9),  each field has a specific meaning.

Classification System KS-IIS-08*, shown in Fig. 7, represents the modification of the System which has been developed and utilized at the Faculty of Technical Sciences in many projects related to Application of Group Approaches for individual companies. Having in mind that homogenous groups of elements are created in regard to Working Units with extended flexibility – the degree of decomposition of Classification System is being kept on a lower level of details which simplifies the process of classification. It is also important to classify companies, or branches of companies, from the following aspects: type of industry, technicaltechnological potential, the degree of automation and organizational and managerial structures.

In order to reach the optimal choice of companies, in other words, the effective distribution of homogenous groups of objects of work amongst the companies, the following matrix shown in Fig. 8 is used. On its basis a comparison is done, comparing companies’ capabilities and technological requirements of a group of elements. In that way, the problem of participating companies having similar technicaltechnological potentials is being solved.

The result of the above mentioned activiti486

es is demonstrated with decrease of system complexity (Fig. 9), creation of simplified and more effective information flows and creation of the basis for development of effective and efficient organizational and managerial structure of a Cluster. In Fig. 9, the expected result of the Application of Group Technology on the Cluster level is shown. Companies in the Cluster are marked with the characters of Alphabet, and flows of material and information are shown with the lines. Each group of products has its flow, which is defined on the Cluster level which enables easier control and consideration of possible critical points and possibilities for improvement. On the other hand, each innovation implies small changes in the layout of such arranged processing structures of Clusters.

The process of Adjusting organizational and managerial structures of Clusters and member companies is the next phase which shall provide utilization of established processing structures of Clusters.

4 PROGRAMMING AND PLANNING OF PRODUCTION IN CLUSTER In order to achieve balanced utilizations of capacities, companies would have to submit their production plans and engagement of their systems in advance, e.g. by utilizing IT technologies, and on the basis of these plans to make detailed termplans at the cluster level. Any change of termplan is recorded and must be available to all participating enterprises.

Morača, S. - Hadžistević, M. - Drstvenšek, I. - Radaković, N.

StrojniĹĄki vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490

A C

W

B

E

F

D

R

T

J

G

H

L

O

K

N

I

V

P

X

Q

U

M

S

A

A

A

B

B

C

E

G

D

H

K

X

F

O

I

Q

U

D

J

N

L

K

F

G

V

Q

T

U

W

V

M

M

M

S

S

S

Group flows 1

Group flows 2

Group flows h

Fig. 9. Simplified network of flows at the cluster level Many intersections in the system, diversity of procedures mutually connected with connections of different degrees of strength, courses and directions and a lot of feedback connections, hamper the process of managing to the great extent. Directed control procedures, in the case of artificial (man-made) systems, have basically mandatory character which provides designed system operations. However, in the case of natural systems, management procedures based on the homeostasis self-regulating principle, have a natural character and maintain a managed variable on the necessary level in the significantly narrower boundaries of tolerance fields and in significantly longer duration period.

Special environmental requirements, disorders in the work processes, delivery delays, organizational deficiencies and other similar influences condition the need for further settings of operational plans at the time of their performance. Since the above mentioned phenomena are constant in time, the need for settings of operational plans is constant in time too. Only a full harmonization of working elements of the operational plans execution system - working processes â&#x20AC;&#x201C; provides anticipated effects. Here is an illustration of what this concept means: We suppose that firms M, N, U are cooperating in the cluster. Firm M supplies (row materials and components) from firms N

Application of Group Technology in Complex Cluster Type Organizational Systems

487

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490

and U, and firm U supplies from firm N. If we want to apply a group approach, it is considered that every firm has developed a management production system and that at the beginning of making an operating plan for the next period has a correct time schedule for all the processes in a

firm. Plan of processes can be illustrated through matrix (firm M) or through Gantt chart (firms N and U). See Fig. 10. Deviations of the results of given phenomena leads also to deviations of designed effects.

Production plan of Firm M N

Production plan of Firm N

V 4 5 6 7 8 9 10 11 12 13 14 15 16 C P S N P U S C P

17

S N P U S

18 19

C

10r

Q

x

U

N P U S

C P

S N P

9

x 1,3 1

xxx

xxx xx x

xx x 10

14

M

20 21 22 23 24 25 26 27 28 29

P S

x x x 7 7 5

x x x

3,5

x xx xx x

x

7

3,5

xx xx x

xx x

x x

xx x

x x x

1

xx x

S Group flows 2

x xx xx x

x

1.4

Production plan of Firm U

Fig. 10. Integration of production plans The basic problems which condition the need for planning in the most of the production system are reduced to the following elements:  maintenance of delivery deadlines,  control of the level of unfinished production,  minimization of waiting lines,  optimization of the sequence of inputs of orders in the working process,  harmonization of the work load (capacity),  elimination (minimization) of the time in the state of cancellation by providing integrated system support,  maintenance of a balanced relationship between the continuity of flow in the system

488

and the cost of supplies (materials, participants, energy, money).

From the above mentioned, it is clear that data processing and information design about the system status in a specific cross-section, must be done continuously and in real-time in order to have the working process adjusted before entering the next cross-section of the system when needed.

In this sense, there is no use to plan the status for the next day on the basis of the data from the previous week. Information about the status of the cross-section "i" must be the basis for planning the cross-section "i +1". The process must be carried out in real-time – therefore, at the end of the operation that generates the status "i" it is necessary to design the status "i +1".

Morača, S. - Hadžistević, M. - Drstvenšek, I. - Radaković, N.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490

Knowing that with the hierarchy access there is practically no feedback connection between the system programming and system planning, the decisions made by system programming - operating plan (part of the production program stipulating the structure and the quantity that will be produced in the upcoming, accurately specified period of time) is not affecting decisions made in the planning stage, but is limiting them. Therefore, it becomes difficult to carry out the production plan taking into account the precise program for hierarchy systems. It is necessary to make the integration of programming and planning systems for the sake of global optimization of processes in order to have industrial clusters functioning as one entity.

The model of simultaneous planning and programming for more periods was suggested by Birewar & Grossmann [23], where programming decisions are built on the level of planning. It has been shown that planned profit increases significantly when planning and programming decisions are optimized simultaneously. The bad side of this approach is that the model of planning and monitoring is restricted to the specific category of simple problems because it requires an extremely large number of binary variables needed to solve the problems of integrated planning and programming.

The system defined in this way enables high-performance production, and provides optimal use of capacities and great flexibility of the entire system. Such systems enable the production in small series with very low costs.

Since there is a large number of small and medium-sized enterprises, any changes in processing, shaping or any changes of material are solved within a few enterprises either by replacement or purchase of a small number of machines or by including some companies with the required technology in the cluster. By doing so, a very fast reaction to any disorder or any changes is achieved. This means that the development processes are carried out simultaneously because each company is assigned a task to develop a part of a product for which it is specialized. Thus, the development of shorter duration and an increased number of different combinations available for utilization is achieved.

6 REFERENCES [1] [2] [3]

5 CONCLUSIONS Group technology adoption helps small organizations to acquire process competence and better process control.

Investment in measurement and testing equipment leads to long term advantages. They can manufacture high precision products and get price advantage on these value added products as they grow through forward integration [24]. With this approach, a number of structural elements and a variety of relations between them are the basic parameters which define the complexity degree of organizational structure and simultaneously determine the complexity of cluster information flows. Therefore, the complexity degree of organizational structure determined upon those parameters enables a comparison of the designed structure variants using the quality defined as control adequacy. With process expertise they can also develop many new products and cater for the international market [25].

[4] [5]

[6] [7] [8]

Porter, M.E. (1979). How competitive forces shape strategy. Harvard Business Review, vol. 57, no. 2, p. 137-145.

Porter, M.E. (1991). Towards a dynamic theory of strategy. Strategic Management Journal, vol. 12, p. 95-117.

Hamel, G., Prahalad, C.K. (2002).

Competing for the Future. Tata McGrawHill Publishing Company Limited, New Delhi.

Van Kirk, J.E., Noonan, K. (1982). Key factors in strategic planning. Journal of Small Business Management, p. 1-7.

Besant, R. (2001). Nature of competition, firm behaviour, and performance. In: Morris, S., Besant, R., Das, K., Ramachandran, K., Koshy, A. (Eds.). The Growth and Transformation of Small Firms in India, Oxford University Press, New Delhi.

Porter, M. (1990). The Competitive Advantage of Nations, Free Press.

Marshall, A. (1925). Principles of economics, (8th ed.), Macmillan, London, UK. Rosenthal, S.S., Strange, W.C. (2004).

Evidence on the nature and sources of agglomeration economies. In: Henderson.

Application of Group Technology in Complex Cluster Type Organizational Systems

489

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 477-490

[9] [10]

[11] [12]

[13]

[14] [15]

[16]

[17]

[18]

[19]

[20]

[21]

490

Porter, M.E. (1998). Clusters and the new economics of competition. Harvard Business Review, vol. 76, no. 6, p. 77-90.

Malizia, E.E., Feser, E.J. (1999).

Understanding Local Economic Development. Center for Urban Policy Research, New Brunswick, New Jersey.

Burbidge, J.L. (1978). The introduction of group technology. Heineman, London.

S. P. Mitrofanov, (1966) Scientific Principles of Group Technology, English Translation Boston Spa: National Lending Library, London.

Mitrofanov, S.P. (1965). Научние основи технологическој подготовки групового производства. Mašinostroenie, Moscow.

(In Russian) Burbidge, J.L. (1963). Production flow analysis. Journal of the Institution of Production Engineers, vol. 42, p. 742-752.

Maksimovic, R., Lalic, B. (2008) Flexibility and Complexity of Effective Enterprises; Strojniški vestnik - Journal of Mechanical Engineering, vol. 54, no. 11, p. 768-782.

Zelenovic, D.M., Tesic, Z.M. (1988). Period batch control and group technology, International Journal of Production Research, vol. 26, no.4, p. 539-552.

Zelenovic, D., Cosic, I., Maksimovic, R., Radakovic, N. (1998). The IIS - Approach to Design of Effective Industrial Systems Structures.

International Journal of Industrial Systems, vol. 1, no. 1, p. 5-16.

Zelenovic, D., Cosic, I., Sisarica, Z., Sormaz, D., 1986: APOPS - Automated procedure for production systems design. FTN - Industrial Systems Institute, Novi Sad. Wemmerlov, U., Hyer, N.L. (1989). Cellular manufacturing in the U.S. industry: A survey ofusers. International Journal of Production Research, vol. 27, no. 9, p. 1511-1530.

Shankar, R., Vrat, P. (1999). Some design issues in cellular manufacturing using the fuzzy programming approach. International Journal of Production Research, vol. 37, no.

11, p. 2545-2563.

Olorunniwo, F., Udo, G. (2002). The impact of management and employees on cellular manuft. implementation.

International Journal of Production Economics, vol. 76, no. 1, p. 27-38.

[22] [23]

[24] [25]

[26] [27]

[28] [29]

[30]

[31]

[32]

[33]

Porter, M.E. (2003). The economic performance of regions. Regional Studies, vol. 37, no. 6-7, p. 549-578.

Birewar, D.B., Grossman, I.E. (1990).

Simultaneous Synthesis, Sizing, and Scheduling of Multiproduct Batch Plants.

Ind. Eng. Chem. Res., vol. 29, p. 2242-2251.

Murray, J.A. (1984). A concept of entrepreneurial strategy.

Strategic Management Journal, vol. 5, p. 1-13.

Prater, E., Ghosh, S. (2005). Current operational practices of U.S. small and medium-sized enterprises in Europe, Journal of Small Business Management vol. 43 no. 2, p. 155-169.

Henderson, J.V., Thisse, J.-F. (Eds.) (2004).

Handbook of Regional and Urban Economics, vol. 4.

Swann, P., Prevezer, M. (1996). A comparison of the dynamics of industrial clustering in computing and biotechnology.

Research Policy, vol. 25, p. 1139-1157.

Porter, M. (1998). Clusters versus industrial policy. On Competition, HBS Press.

Swamidass, P.M., Newell, W.T. (1987).

Manufacturing strategy, environmental uncertainty and performance: a path analytic model. Management Science, vol. 33, p. 509524.

Boynton, A.C., Victor, B. (1991). Beyond Flexibility: Building and Managing the Dynamically Stable Organization. California Management Review, vol. 34, no. 1, p. 5366.

Broekhuizen, T.L.J., Alsem, K.J. (2002).

Success Factors for Mass Customization: A Conceptual Model. Journal of MarketFocused Management, vol. 5, no. 4, p. 309330.

Hauser, D.P., de Weck, O.L. (2007).

Flexibility in component manufacturing systems: evaluation framework and case study. Journal of Intelligent Manufacturing, vol. 18, no. 3, p. 421-432.

Zelenovic, D., Cosic, I., Maksimovic, R. (1998). Design and Reenginering of Production Systems: Yugoslavian (IISE) Approaches. vol. 16 in: Group Technology and Cellular Manufacturing - State-of-theArt Synthesis of Research and Practice, Kluwer Academic Publishers, Massachusetts, p. 517-537.

Morača, S. - Hadžistević, M. - Drstvenšek, I. - Radaković, N.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 491-496 UDC 621.791.7:004.942

Paper received: 27.05.2009 Paper accepted: 07.07.2010

The Application of a New Formula of Nakaoka Coefficient in HF Inductive Welding Miroslav S. Milićević* High Technical School Beograd, Niš, Srbija The high-frequency welding procedure represents the complex theory which is possible to be used as an approximate calculation of important data such as current, voltage and the power of inductor and the welding object itself as well as the degree of usage. By applying so far known procedures, long and complex calculations are done through the use of the terms as well as many tables and graphic dependences. In this paper, a new analytical dependence is shown by the use of which the value of Nakaoka coefficient is being calculated through specific approximation, and by which important parameters and the range of high-frequency inductive welding of steel pipes are also calculated. By applying the results of this paper, the calculating procedure is shortened and made easier which makes way for optimizing the choice of thermal regime. © 2010 Journal of Mechanical Engineering. All rights reserved.

Keywords: numerical approximation, electromagnetic field, HF welding, equivalence scheme, Nakaoka coefficient, practical application 0 INTRODUCTION The case of high-frequency inductive welding of steel pipes can be reduced to the induction case on the semi-infinite area according [1] to [19]. Thus, the semi-infinite area in Fig. 1. is exposed to the electromagnetic wave influence under the assumption that this is electroconducting board. The wave influence occurs from the dielectric area into the conducting area where displacement currents are being neglected, so that Maxwell equations can be written according [1] to [4]. H z (1)   E y , x E y H z (2)   , x t where H z represents magnetic field, E y electric field, µ magnetic permeability and  specific electric conductivity.

Since to the magnetic field H and electric field E are sine wave function of the time t, we have H  H m e j (t  H )  H m e j H e jt  H m e jt , (3) (4) E  E e j (t  E )  E e j E e jt  E e jt , m

m

m

Fig.1. The influence of electro-magnetic wave on semi-infinite area where H m and E m are complex amplitude value of working fields H m and E m ,  H and  E are relevant phases. Exchanging Eqs. (3) and (4) in (1) and (2), and excluding indexes "y" and "z", after transformations we get d 2 H m (5)  jγH m  0 . dx 2 Solving (5) we have a solution  j  x j  x (6) H  Ae  Be , m

which after getting constants from boundary conditions gives (index "0" value for x = 0) H m  H m0 e



 x 2

e

j

 x 2

.

(7)

1

High Technical School Beograd, Bul. Nemanjića 33/39, 18000 Niš; Serbia; Lavmiro@eunet.yu

491

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 491-496

From the term (7) we conclude that while penetrating the wave becomes suppressed in the conductible area according to the exponential law. Variable Δ  2 / ωμγ is the penetration depth within which the value of surface magnetic field falls down to 1/e value of the field.

There is a term for magnetic field now H m  H m 0 e



x Δ

x

(8)

 ej Δ .

Previously described case of semi-infinite area expands in the [1] to [4] to the real case, that is to say, to welding of the rims of steel band of V loop which is in the zone of inductor and concentration field influence.

Starting from the basic equations of electro-magnetic circuit, an equivalent scheme can be established with concentrated parameters and it is shown in Fig. 2.

In a similar way we get the electric field power along axis x: 

x

x

j    2  (9) H m 0e Δ  e  4 Δ  , Δ which enables us to see that it is being moved in phases and that it is being exponentially suppressed along penetration axis. Penetrating, both fields are more and more late compared to inward field, and electric field is always ahead of magnetic field for π/4.

Current density is: (10) J m    E m , that is after the shift from (9) to (10):

E m 



x

x

j    2  H m 0e   e  4   .  Surface power density is derived Poynting vector: 1 S  Em  H m , 2 where S represents flow energy vector,

J m 





(11) from (12) Em

vector complex amplitude of the electric field and H m conjugated complex amplitude of the magnetic field, which transfer vector to scalar value, after being changed and fixed, gives: 2x

2 2  π  (13) H m0 e  cos  j sin  .

2 4 4  Only active power part performs heating and it is: S 

2x

H2  (14) P  m0 e  .

2 At the entrance into semi-infinite area, x=0 and substitute value of Δ, area power is:  2 P0  0.5 H m0 .

(15) 2 Within the penetration depth ∆, 86% of overall heat is lost.

492

Fig. 2. Changeable scheme with concentrated parameters Fig. 2 is valid for the case of encircling inductor which is most applied in practice, and the signs are: Ri is active resistance of single coil inductor, Rt is active resistance of steel pipe, Xs is inductive resistance of dispersion between inductor and pipes, Xbn is external resistance of dispersion, Rkr is active resistance of the steel band rims, Xkr is inductive resistance of the steel band rims and Zm is impedance conditioned by the resistance of internal pipe opening and internal magnetic concentrator.

External inductive resistance of inductor dispersion is calculated by [1] to [4]: X k X bn  s m , (16) 1 km where km represents Nakaoka coefficient whose values are in the Figure 3. depending on the relationships Di/D and ai/D, where Di represents internal inductor diameter, D is external pipe diameter and ai inductor length.

Nakaoka coefficient km is a function of many variables and its dependability is given in the Fig. 3 within the function Di/D and the range ai/D is taken as a parameter.

If known interpolation formulas are to be used we will get complex analytical terms which are inadequate for use in engineering practice.

In this paper with adequate substitution, before approximation, a simple link km is reached in the function of dependable variables which in the end gives simple analytical dependability.

Milicevic, M.S.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 491-496

The precision of Nakaoka coefficient calculation according to this approximation is illustrated by a practical example where other ranges which characterize the model of highfrequency welding of one solid steel pipe are calculated as well.

n

wk 

x

i

k

, (k = 0, 1, ..., 2m),

(19)

i 1

unknown zl from the relation: n

zl   yi xi , (l = 0, 1, ..., m). l

(20)

i 1

Fig. 3. Nakaoka coefficient values 1 APPROXIMATIVE CALCULATION

BY THE USE OF NUMERIC METHOD In order to calculate Nakaoka coefficient, whose graphic dependences are given in the Fig. 3, numeric method known from [7] as the method of the smallest squares will be used. Having in mind what has been presented in [7] as well as the experience in application and further research of the approximation of the author of this paper, adequate form for its use in programming consists of the following.

If there is an experimental or other cluster (xi, yi) of specific values where (i = 1, 2, …, n), then we can find the polynomial in the form of m

y   ai x i ,

(17)

i 0

which approximates the function given by the cluster of points, where coefficients ai are being determined by the system of m

w

j i a j

 zi , (i = 0, 1, ..., m).

(18)

j 0

The coefficient wk from the system of Eqs. (18) is calculated by the formula:

The precision of approximated function is achieved by an adequate choice of the m degree in the polynomial (17). The way this approximation is laid out in this paper offers the possibility of doing the programming in C language. The system of Eqs. (18) is solved by Gauss’s elimination by means of which we do the calculation of coefficients of ai polynomial (17) which approximates a specific functional dependence which is required.

Nakaoka coefficient, whose Graphic dependences are given in Fig. 3, is a function of 2 variables Di/D and ai/D and it is obvious that those are non-linear functions. Due to this, calculations at HF welding have been done in such a way that the value km has been determined graphically.

To make it simple, we introduce shift: 1 k m  , (21) km and thus by applying cited approximation and C programming, which is not being mentioned in this paper due to its size, linear dependences good at approximation are achieved, with a small bug, coefficient Nakaoka. During the approximation proportion ai/D is used as the parameter and thus how many linear analytical dependences we will get depends on how many approximations we do.

Let us list some of the achieved approximations: ai  D  0.5, k m  0.9712  1.6210  i  1, (22) D  D ai D   0.8, k m  0.9735  0.9987  i  1, (23) D D   ai D   1.1. k m  0.9772  0.7298  i  1.

(24) D D  From the cited relations the change of the coefficient in front of the bracket in the function ai/D can be seen since that function is non-linear, and having in mind other approximations, finding reciprocal values algorithm for determining coefficients is done in a similar way. Based on the approximation bug, on many practical examples,

The Application of a New Formula of Nakaoka Coefficient in HF Inductive Welding

493

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 491-496

we get adequate linear approximation for engineering practice.

Since we introduced shifts in two steps and applied algorithm for approximate calculation of polynomial coefficient twice, by replacing the known and ordering we get: D a k m  1  0.73171 i  1 i (25) D  D and 1 km  .  ai  Di (26)  1 1  0.73171 D D In the Table 1 there are examples of calculation with the help of new analytical dependence of Nakaoka coefficient (26) which were compared to the values from Fig. 3 and the percentage of relative error discrepancies were calculated.

Table 1. Evaluating the approximation accuracy

Di D

ai D

km Eq. (26)

km Fig. 3

ε [%]

1.4

1.2

0.804

0.800

0.50

1.1

0.5

0.8723

0.88

0.87

1.25

1.1

0.8574

0.855

0.28

1.325

0.85

0.78138

0.775

0.80

1.50

1.50

0.8039

0.815

1.36

By analyzing the movement of relative flaw from the Table 1 it can be seen that it does not exceed one percentage which is a very good approximation for engineering practice. That is why the dependence (26) for Nakaoka coefficient with the HF welding will be used further by making calculation process easier.

The inductive resistance of dissipation, between the inductor and steel pipes, from the scheme in the Fig. 2 and (16) is [1] to [7] given in the form of: D 2  D2 , X s  0.25  0 i (27) ai which after the shift in (6) gives D 2  D 2 km .

X bn  0.25  0 i  (28) ai 1  km By substituting the value km according to the new analytical dependence (26) into (28) we get:

494

X bn  0.342  0 Di  D  , (29) through which we come to the new conclusion which was hidden until now, that the exterior induction resistance of inductor dissipation, which is supplied by HF current, does not depend on the length of the inductor. Such a conclusion can be used when optimizing parameters in the process of HF welding which was presented here by the substitution Fig. 2.

In order to check the results of the paper for calculating according to the new dependence and due to the presentation of the special values of the substitution scheme in the Fig. 2, a practical solving example will be given.

Example: Design and calculate the parameters of HF welding of the steel pipe of 76.1 mm in diameter and of 4 mm wall thickness in such a way that production velocity is 60 m/min. The interior diameter of the inductor is 85 mm, generator frequency is 440 kHz, welding temperature is 1500oC, mean distance of the inductor from the joint point of the tape rim 111 mm, 5.36 mm thick conduction pellet in the space of 600 mm. The gap between ferrite and interior rim of the steel tape is b1 = 5 mm and the inductor length is ai = 72 mm. Heating characteristics are adopted in the usual way like in [1] to [6].

Fig. 4. The substitutional scheme with calculated values for the pipe of 76.1 mm in diameter and 4 mm of wall thickness After the application of the procedure from [1] to [19] and this paper we get special values of the equivalent scheme.

In order for this welding to take place, the power Ptr = 163.8 kW which is released on the rims of the steel tape is needed; thus the power flowing through the tape is Itr = 1698 A and voltage of the steel tape noose is Utr = 207.8 V. The coefficient of the power use is ηi = 0.717 and electric coefficient of the useful effect is η = 0.985 and with this the calculation of

Milicevic, M.S.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 491-496

the necessary power for the inductor has been enabled:

Pi 

Ptr  232 kW . i 

(30)

By elementary calculation we get the inductor current I i  2160 A (31) and the voltage on the inductor U i  297 V .

(32) It has been considered that impedance Zm , which is the function of the interior opening of the pipe and the characteristic of ferrite as the concentration field, has an infinite value. The depth of the heat penetration in the steel tape is Δk = 0.824 10-3 m.

Having in mind the approximate losses in the oscilator and other generator appliances as well as done calculation it is proven that it is really possible to make a quality welding of this steel pipe by the chosen velocity of the welding.

2 CONCLUSION The general theory of HF welding has been described in this paper starting with the case of electromagnetic wave influence on half-infinite environment.

A real model has been formed, for a socold V noose formed by the rims of steel tape during the pipe welding and equivalent substitutional scheme has been given. The problem of calculating unknown values is pointed out when using complex analitical expressions and graphic dependences dependent on many parameters.

In order to get analitical dependences, a numerical coefficient approximation has been approached and, so far, it has had only graphic presentation. A numerical approximation method has been chosen and adjusted and it is adequate for solving through the application of the C programming language.

The procedure of numerical approximation is made easier by introducing adequate scheme prior to algorithm application. As a result, the simple formula for calculating Nakaoka coefficient value has been reached as a function of two variables.

By applying it to many practical examples it can be concluded that approximative formula, beside being simple, has good approximation

because relative flaw of line deviation is equal to one procentage. A new conclusion has been drawn that the inductive resistance of inductor dissipation is independent of the inductor length which opens up a new way of optimizing parameters.

The results of this paper are applied to the complete calculation of HF welding of the steel pipe of 76.1 mm in diameter and 4 mm wall thickness which is of great interest for engineering and productional practice and it represents a new way of further research in the field of electromagnetic theory applied in the field of thermia. The application is principal through optimizing parameters with the tendency of achieving greater energy saving and better use coefficient along with improved structure and the qualitiy of welded and thermically treated products.

3. REFERENCES [1] Slukhotskii, A.E. (1968). Application HF current for electrothermia. Mechanical Publ., St. Petersburg. (In Russian) [2] Slukhotskii, A.E., Ryskin, S.E. (1974).

Inductors for induction heating. Mechanical Publ., St. Petersburg. (In Russian) [3] Shamov, N.A., Lunin, V.I., Ivanov, N.V. (1977). High frequency metal welding.

Mechanical Publ., St. Petersburg. (In Russian) [4] Milićević, M., Milićević, V. (2002). Impeder for HF inductive welding of steel tubes. IEE Proceedings, Science, measurement and technology, vol. 149, no. 3., p. 113-116.

[5] Milicevic, M., (2007). Possible defect of the high frequency inductive welding in the steel tubes by applying a ferrite impeder. 39th International October Conference on Mining and Metalurgy, Soko Banja, Serbia & Montenegro, p. 318-333 [6] Mitani, K., Shibuya-Ku, H. (1992). Impeder: How its innovation and design impacts the welding process. The 8th annual world tube congress, Proceedings, Chicago, Illinois, p.

25-33.

[7] Milovanovic, G. (2005). Numerical Analysis.

Faculty of Elecronic Engineering, Nis. (in Serbian).

The Application of a New Formula of Nakaoka Coefficient in HF Inductive Welding

495

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 491-496

[8] Nemkov, V.S., Demidovic, V.B. (1988).

Theory of induction heating, Energy Publ., St. Petersburg (In Rusian).

[9] Wright, J. (1997). Principles of high frequency induction tube welding. Electronic heating equipement, Sumner Inc. [10] Rudnev, I.V. (1997). Induction heat treatment. Steel heat treatment handbook, New York, Basel, Hong Kong.

[11] Ruffini, S.R., Ruffini, T.R., Nemkov, S.V. (1998). Advanced design of induction.

Industrial Heating, Madison Heights, Michigan.

[12] Wade, J. (1990). Effective utilisation of magnetic flux concentrators in induction heating at commercial heat treating plant.

Heat treatements, Cleveland, Ohio.

[13] Electronic Heating Equipment, inc.”, (1998).

1998 Catalog & Applications Guide, Buckley, Washington, p. 1-18.

[14] TDK Impeder core (1999). Technical documentation, TDK Italy.

[15] Todorov, T.S., Ivanov, P.T., Milicevic, M., Madjarov, N., Iliev, D., Aleksiev, D. (2000).

Specialized high frequency power supplies

496

for induction heating applications IHS-2000, Padua, Italy.

[16] Milicevic, M., Milicevic, T. (2005). New solution for improvement of parameters and quality of HF inductive welding of steel tubes. 37th International october conference on mining and metallurgy, Bor, Serbia & Montenegro, p. 466-475.

[17] Milićević, M., Milićević, V., Milićević, T. (2004). MDM impeder for improvement of parameters and quality of HF inductive welding.

36th international october conference on mining and metallurgy, Bor, Serbia & Montenegro, p. 531-537.

[18] Milićević, M., Milićević, V. (2004). Analysis of the transistor converter of power together with energy dosage for the inductive heating and wellding of steel tubes. ETEP, vol. 14, no. 2, p. 111-118.

[19] Milićević, M., Radaković, Z. (2006). Quality improvement of stell pipes produced by seam welding with new magneto-dielectric impeder. Materials transactions, The Japan institute of metals, vol. 47, no. 6, p. 14641468.

Milicevic, M.S.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 497-504 UDC 519.23:658.5:001.895

Paper received: 18.09.2009 Paper accepted: 02.03.2010

Statistical Analysis for Strategic Innovation Decisions in Slovenian Mechanical Industry Peter Fatur* - Borut Likar University of Primorska, Faculty of Management The objective of this study is to identify the main factors influencing the innovation and R&D performance of the machinery and equipment manufacturing industry in the Republic of Slovenia (RS).

The research is based on statistical data from the Statistical Office of RS. Spearman’s coefficient of correlation has been applied to the entire set of input and output variables in calculating the correlation coefficients. Results indicate the existence of two clusters of companies. Both are innovation followers but differ in their capabilities to produce breakthrough innovations and innovation-related turnover. For both of them, no correlation between the innovation outputs and business/financial performance is present.

Based on the empirical findings, we propose some organizational areas where additional managerial effort needs to be invested. Thus, the research also has a practical implication for the enterprises as well as for the national policy makers. © 2010 Journal of Mechanical Engineering. All rights reserved.

Keywords: innovation, R&D, technology, industrial management, productivity and performance management; machinery and equipment manufacturing industry 0 INTRODUCTION The European Union’s (EU) Lisbon goal of becoming the world’s most competitive business environment by the year 2010 has not been met. According to the recent statistical indicators [1] the EU is still losing ground in business exploitation of knowledge and creativity to the United States (US) and Japan. Even though the innovation gap has decreased in the last years (towards US from 41 to 28%, and towards Japan form 42 to 38% in the 2004-2008 period) it remains significant.

The national innovation performances of European countries vary a lot. The European Innovation Scoreboard (EIS) classifies the countries into the following groups [1]: (i) the innovation leaders, including Denmark, Finland, Germany, Israel, Japan, Sweden, Switzerland, the UK and the US. Sweden is the most innovative country, largely due to strong innovation inputs although it is less efficient than some other countries in transforming these into innovation outputs; (ii) the innovation followers include Austria, Belgium, Canada, France, Iceland, Ireland, Luxembourg and the Netherlands; (iii) the moderate innovators group includes Australia, Cyprus, Czech Republic, Estonia, Italy, Norway, Slovenia and Spain; (iv) the catching-up group *

consists of Bulgaria, Croatia, Greece, Hungary, Latvia, Lithuania, Malta, Poland, Portugal, Romania and Slovakia. These country groups appear to have been relatively stable over the last years.

An indicator of the innovation capability is a turnover of new or significantly improved products new to the market as a percentage of total turnover. For the year 2004, it counts 3.5% for medium and 8.5% for large companies for EU27. The relative turnover of new or significantly improved products new to the firm counts 5.1% for medium and 9.3% for large companies.

Pursuant to the national statistical data [2] only 35.1% of Slovenian companies prove to be innovative and 41.2% in the manufacturing sector.

The machinery and equipment manufacturing industry as the subject of our research performs better; however, no more than 47.3% of companies in this industry actively pursue innovation.

What is more, an in-depth analysis noticeably shows that the situation regarding innovation in Slovenian small and medium enterprises (SME) is even worse where the large companies record approximately 50% more innovativeness as the medium-sized ones while the small companies even threefold less than the

Corr. Author's Address: University of Primorska, Faculty of Management, Cankarjeva 5, 6000 Koper, Slovenia, 497 peter.fatur@fm-kp.si

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 497-504

large ones [2] (Here it needs to be taken into consideration that an enterprise – regardless of its size – is classified in the statistical group of innovative enterprises by introducing at least one new product. The latter thus represents a “statistical benefit” for large companies.).

Apparently, Slovenian manufacturing needs an innovation push to outrun the group of innovation followers and catch up with the group of innovation leaders.

Literature tackles different approaches to pursue innovation yet one of the fundamental ones proves to be an analysis of innovation processes based on input, process and output groups of indicators, either individual or composite. Glede na vrstni red manjka citat [3] Individual indicators [4] to [6] measure single influential factors (e.g. the amount of resources invested into the research and development (R&D), the annual number of days dedicated to training of management/employees).

The problem of individual indicators remains their inability to deal with with the complexity of the innovation management field. Consequently, the composite indicators prove to be more appropriate since they regard the inventioninnovation process with due complexity, as an intertwinement of related and correlated factors [7].

The input indicators (also referred to as “investment” indicators) include e.g. expenditure on R&D or employees training; [8] and [9]. The process indicators take into account the organisation or management of innovation processes, the use of appropriate management techniques (market research, problem analysis and idea creation techniques, forecasting techniques, etc.), and innovation environment within a company. The output indicators identify results, e.g. the number of patents and new products, market shares, revenues from the sales of innovations and innovative products etc.; [9] and [11]. Several studies have shown the correlations among the input, process and output variables. Hollenstein shows the correlations between the input (e.g. research input, development input) and the output-oriented indicators (e.g. number of patents, number of innovation projects) and the market-oriented measures (sales share of new products) – thus indicating the innovativeness of a firm [6]. Iansiti shows correlations among input (e.g. technology

498

from suppliers, technology from other groups) and process indicators (e.g. research groups, project management, communication) and technological potential and yield [8].

The results of Parthasarthy's study show that both, the innovation input and the innovation process have implications for innovation frequency, i.e. the number of new products introduced [12]. He realised that R&D intensity, by itself, positively influences the invented technologies; developing them into new products and marketing them frequently requires a corresponding level of functional integration.

Developing a “marketable” product involves a transition form sequential to concurrent product and process development, using apposite product development techniques, e.g. design for assembly and pre-testing of processes by the process simulation tools [13] to [15].

Regardless of the fact that many approaches try to find the key influential factors for an effective management of innovation, an apposite method has thus far not been developed.

The cited methods hold another limitation, namely they were all tested on somewhat small samples of companies and failed to focus on the machinery and equipment manufacturing industry which is the subject of our study.

Thus, the objective of this study is to identify the main influential factors and estimate their effects on the innovation and R&D performance in the machinery and equipment manufacturing industry in the Republic of Slovenia (RS).

1 METHODS Pursuant to the official classification [3], our research encompassed companies headquartered in RS, belonging to the statistical class DK29: Manufacture of machinery and equipment. The sample size was 2500 companies while the subset of the statistical class DK29 comprised a total of 144 companies.

The Statistical Office of the RS (SURS) regularly collects the data on target industry pursuing standardized methodology, [16] to [18].

The statistical survey providing the core data for our research is the most recent Community Innovation Survey (CIS 2006) which in 2007 was carried out across Europe. The Slovenian CIS 2006 survey includes data from the years 2004 to

Fatur, P. – Likar, B.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 497-504

2006 on the enterprises’ product (good or service), organisational and process innovations, innovation activities and expenditures, cooperation in innovation and the effects of innovation. In addition, company’s financial data (balance sheet, profit-and-loss account and some key financial ratios) was collected from the official statistical database on companies (Agency of the RS for Public Legal Records and Related Services), while the third statistical database (Statistical Register of Employment (SRDAP)) provided for data on the educational structure of employees. The employment data refer to the business year 2006 while the financial data comprise the period between 2003 (a year before the CIS survey) and 2007 (the subsequent year).

A group of relevant variables was selected from the statistical databases (Table 1). The two key variables that represent a measurable output from the innovation process have been defined as: RII (“Revenues from innovation index”), i.e. a share of turnover resulting from innovations, and RMI (“Revenues from market innovation index”), i.e. a ratio of turnover from innovations new to the market to turnover from innovations new to the company only. Furthermore, we defined the Lead index (LI) as a contemporary measure of the influence of both RII and RMI. A definition of indices is shown in Table 2.

Spearman’s coefficient of correlation (SCC) was then applied to the entire set of input and output variables in calculating the correlation coefficients. Regardless of the fact that between the two associated variables the Spearman coefficient has less significance than the Pearson’s coefficient, it is suitable for calculating correlations not only among the interval (associated) and ranked (discrete), but among combined variables used in a research.

2 RESULTS We aimed to identify a relationship between the two key output variables from the innovation process, RII and RMI. Surprisingly, the Spearman correlation analysis showed no correlation (SCC=0.01; sig=0.94). Thus, there are a number of companies in the Slovenian machinery and equipment manufacturing industry with both, a high share of turnover from innovations and a high share of turnover from

“radical” innovation in total innovation turnover (high RII and high RMI). These companies are innovation leaders in the industry (Fig. 1).

Fig. 1. The Marketability/Inovativeness matrix On the other side of the matrix (low RII and low RMI), there are companies with little revenues from innovations and the latter are of minor impact (presumably of incremental type, e.g. incremental improvements to their existing products to follow the technology or market trends) – we call them “innovation losers”.

Our research focused on the companies inbetween the two poles. Both groups are market/innovation followers.

The first cluster of companies (high RII and low RMI) makes a notable part of their revenues out of recently introduced products.

However, these products usually act as substitutes to the company’s existing products with no radical improvements incorporated. The result of such a strategy is no influence of new products over companies’ financial performance. We named this type of companies “inertial innovators”.

Companies from the second cluster (“ad hoc innovators”) produce some market inventions, resulting in new products being introduced onto the market before competitors (low RII and high RMI). Such products incorporate a much higher degree of creativity.

However, these companies somehow fail to make substantial revenues out of them. Therefore, the influence of innovation on the companies’ financial performance is again very moderate, if there is any at all.

Statistical Analysis for Strategic Innovation Decisions in Slovenian Mechanical Industry

499

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 497-504

Table 1. Definitions of variables under study Variable name Definition a1 The enterprise is part of an enterprise group Intd_Good The enterprise introduced new or significantly improved goods Intd_Serv The enterprise introduced new or significantly improved services Intd_Proc The enterprise introduced new or significantly improved processes Intd_Org The enterprise introduced new or significantly improved organizational methods a3_1 Markets served: Local / regional within Slovenia a3_2 Markets served: National a3_3 Markets served: Other EU countries, EFTA, or EU candidate countries e3_3 Enterprise received financial support for innovation activities from the EU f31 Co-operation partner: Within your enterprise or enterprise group f32 Co-operation partner: Suppliers of equipment, materials, components or software f33 Co-operation partner: Clients or customers f34 Co-operation partner: Competitors and other firms from the same industry f35 Co-operation partner: Consultants, commercial labs, or private R&D institutes f36 Co-operation partner: Universities or other higher education institutes f37 Co-operation partner: Government or public research institutes f3_slo Partners for cooperation in innovation activities - Slovenia f3_tuj Partners for cooperation in innovation activities - abroad e1_2 The firm performed R&D continuously.

Exp_IntRD_Emp Intramural (in-house) R&D expenditures / Total turnover 2006 Exp_ExtRD_Emp Acquisition of R&D (extramural R&D) / Total turnover 2006 Exp_RD_Emp Intramural + extramural R&D expenditures / Total turnover 2006 Exp_Tot_Emp Total expenditures in innovation / Total turnover 2006 FA1_1 Factor Profit FA2_1 Factor Labor cost prih0607 Total turnover growth 07/06 (%) priht0607 Total turnover growth in foreign markets 07/06 (%) zap_teh_vsi Share of employees with an engineering degree zap_izobr_89 Share of employees with a masters or doctoral degree zap_izobr_6789 Share of employees with at least higher education Table 2. The key output variables of the innovation process RII Revenues from [Turnover from innovations introduced] / [Total turnover] innovation index RMI Revenues from market [Turnover from innovations new to the market] / [Turnover from innovation index innovations new to the company] LI Lead index {[Turnover new to the market] / ([Total turnover] – [Turnover new to the market])} * {([Turnover new to the market] + [Turnover new to the company] / [Total turnover]} What are the specific characteristics of each group?

2.1 The “Inertial Innovators” The “inertial innovators” are the ones with a high correlation between the share of turnover from innovations (RII) and a set of influential

500

(input or process) factors which will be investigated in this section.

The output variable RII correlates with the introduction of new products, either goods (Intd_Good: SCC=0.866; sig=0) or services (Intd_Serv: SCC=0.513; sig=0) and new processes (Intd_Proc: SCC=0.535; sig=0), i.e. “techn(olog)ical” innovations, but much less with

Fatur, P. – Likar, B.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 497-504

the introduction of organizational innovations (Intd_Org: SCC=0.193; sig=0.021). Further, RII correlates with the extent of cooperation in innovation activities with other organizations (f3_slo: SCC=0.774; sig=0), (f3_tuj: SCC=0.733; sig=0). Companies in this group collaborate intensively, in particular with the customers (f33: SCC=0.787; sig=0).

Further, there is a correlation of the above output index with the level of education of the employees (share of employees with at least higher education (zap_izobr_6789: SCC=0.424; sig=0); share of employees with a masters or doctoral degree (zap_izobr_89: SCC=0.300; sig=0)). These companies do not serve the local or regional markets (a3_1: SCC=0.063; sig=0.454) but rather the national (a3_2: SCC=0.348; sig=0) and export (a3_3: SCC=0.290; sig=0) markets, where their revenues are slightly growing (priht0607: SCC=0.249; sig=0.017). A high RII shows no correlation with the financial result (FA1_1: SCC=0.025; sig=0.776).

It seems that this type of innovation followers represents relatively rigid, well established organizations which have traditionally performed the innovation activities but have done them somehow by inertia. We may define their attitude towards innovation as “we always did it this way” – which is not far from routine (in its negative sense) and which results in no breakthrough innovation but rather a continual improvement of existing products and processes to follow the technical/technological developments in the market. The internal organization can (and has to) remain unchanged for long periods. These companies manage to keep their market positions but are condemned to stagnation and vulnerable to new competition with innovative substitutes for their existing products.

This profile of companies intensively cooperates with customers in the field of innovation. The cooperation goes along with the up-to-date concept of open innovation, [19] and [19]. However, it may also imply the company’s position “the customer is a king” which relies on the customer’s demands but very often pushes the company in a defensive position. Namely, customers usually request products already seen at the competition, not products solving their indepth problems, problems they may not even be

aware of. A company following the customer’s requirements without a critical assessment and without a creative insight into deep customer needs may quickly be mislead in the direction of copying existing market solutions and becoming an innovation follower.

2.2 The “Ad Hoc Innovators” The “ad hoc innovators” are the ones with a high correlation between the RMI and some of the influential factors (input or process indices).

This is a less defined group than the group of inertial innovators – the correlations are lower – yet it shows some interesting characteristics.

As shown above, the output indices RMI and RII have no mutual correlation. Furthermore, RMI has none or very little correlation with the indices that correlate high with RII. Data on company size was not available.

However, since there is no correlation in this cluster with the introduction of products (Intd_Good: SCC=0.026; sig=0.857), (Intd_Serv: SCC=0.085; sig=0.551)) and new processes (Intd_Proc: SCC=0.141; sig=0.323) (the three indices being a logical consequence of company size, as discussed in Introduction), it can be assumed that this cluster consists of smaller companies than the cluster of inertial innovators.

These types of companies are often a part of an enterprise company group (a1: SCC=0.397; sig=0.004); other enterprises in the group are their major cooperation partner in innovation (f31: SCC=0.394; sig=0.046).

RMI correlates with the question whether the company performed R&D continuously of occasionally (e1_2: SCC=0.366; sig=0.011).

Obviously, there is a systematic R&D ongoing in companies with a high share of turnover from “radical” innovation in total innovation turnover.

To perform R&D, these companies more often make use of public funding from the EU (e3_3: SCC=0.362; sig=0.009).

Further, there is a correlation with labour costs per employee (FA2_1: SCC=0.336; sig=0.024) and a negative correlation with the share of employees with a technical background of high school level and higher (zap_teh_vsi: SCC=-0.335; sig=0.016) (this may indicate lower salaries of engineering staff in comparison to other profiles (!)).

Statistical Analysis for Strategic Innovation Decisions in Slovenian Mechanical Industry

501

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 497-504

The output index RMI also correlates with the growth of revenues in export (priht0607: SCC=0.314; sig=0.050). However, as in the former case, it shows no correlation with the company financial results (FA1_1: SCC=-0.152; sig=0.319).

At first glance, these companies seem to have a better profile for innovation success. They produce “real” market inventions, resulting in new products being introduced onto the market before competitors. By exploiting knowledge from various sources they manage to incorporate a higher degree of creativity in their products than in the case of inertial innovators. However, these companies somehow fail to make substantial revenues out of their new products. Therefore, the influence of innovation on the companies’ financial performance is again very moderate, if any at all. It looks as they put more effort in knowledge creation than in marketing it effectively.

2.3 The Lead Index There is no correlation between the share of turnover from innovations in total turnover (RII) and the ratio of turnover from innovations new to the market to turnover from innovations new to the company (RMI). Thus, companies with high quality innovations not (necessarily) make a lot of sales out of them (and vice versa). In order to determine the factors that may contemporary influence both output indices RII and RMI we have defined a third output index called the Lead index (LI). Since RMI shows very little correlation with inputs, it is apparent that LI is somehow biased towards RII but is still a good estimate to determine the factors that a company needs to focus upon when trying to improve the innovation system. The findings will be discussed in the subsequent section.

3 DISCUSSION As shown in the Results section, the correlations among groups of influencing indicators and companies’ economic/financial results are not significant, either for the group of inertial innovators or for the ad hoc innovators.

As it is difficult to explain precisely the reason(s), some possible options will be discussed.

502

First, some relevant EU findings need to be investigated. Innovation performance in the EIS is measured as the average performance on both innovation inputs and innovation outputs.

Efficiency analyses among the input and output dimensions show that for most countries there are efficiency gains to be reached. This applies to countries of all levels of performance: many of the innovation leaders (see Introduction) have relatively low innovation efficiency while several of the moderate innovators and catching-up countries have a relatively high efficiency.

Slovenia, besides being ranked in the group of moderate innovators (third out of four groups), combines low efficiency in transforming innovation inputs both in Intellectual property and in Applications (Intellectual property measures the achieved results in terms of successful knowhow; Applications measures the performance in terms of labour and business activities and their value added in innovative sectors) [1].

The same might be the case of the companies from the “ad hoc innovators” cluster.

They produce some market inventions, resulting in new products being introduced onto the market before competitors. Such products incorporate a high degree of creativity. However, these companies are inefficient in exploiting them on the market and thus fail to make substantial revenues out of them. Therefore, the influence of innovation on the companies’ financial performance is very moderate.

Another reason for poor economic performance might be a small share of breakthrough innovations in the companies’ innovation portfolio. Companies are not primarily focused on products which are new to the market but on those which are new to the company only, e.g. a development of the improved products but not completely new ones; an orientation on costs reduction; a reduced time to respond to customer or supplier needs; an improved communication or information sharing etc. Such an approach is a characteristic of the innovation followers, not the leaders. This thesis can be supported by the authors’ several years of experience in the National commission for Innovation rewards at the Chamber of Commerce of RS. The most common patterns/types of best Slovenian innovation projects are improvements of existing products, new products/services connected with relatively unimportant incremental innovations,

Fatur, P. – Likar, B.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 497-504

cost-cutting innovations and other types, which are not connected with break-through ideas – neither via completely new products nor new, highly efficient business models. Such products only lead to little added value. Only a small proportion is the break-through innovations.

Such a situation is prevailing in our first cluster of companies (“inertial innovators”) that invest considerable amounts in innovation and make a notable part of their revenues out of recently introduced products. However, these products usually act as regular substitutes of their existing products with no radical improvements incorporated. The result of such a strategy, as it is clear from the correlation matrix, is no influence of these new products to companies’ financial performance.

Given the complexity of innovation management science, there is evidently no recipe to improve the efficiency of innovation inputs.

However, the Lead Index that contemporarily measures the influence of both the (i) share of turnover resulting from innovations and (ii) share of turnover from innovation new to market in total turnover from innovation indicates some suggestions.

Having creative work undertaken within the enterprise to increase the stock of knowledge and using it to devise new and improved products and processes – in particular in form of the internal R&D process – running continuously and not just occasionally, seems to be one of decisive factors. There is an indication that innovations need to be developed by the company itself and not purchased from outside. However, a well established partnership is required with multiple partners used either for co-creation of innovations or as a source of information (in particular customers, consultants and external research institutions). A share of employees with a masters or doctoral degree has a positive correlation as well. The companies’ marketing activities (the introduction of the innovations to the market) needs to be enforced. Finally, the organizational innovations – in particular innovative business models – that proved to be a weak point in our sample – should receive more attention of the managers.

4 CONCLUSION Despite some encouraging indicators and at times somewhat misleading statistical data, it is obvious that only a moderate portion of the innovative potential of enterprises in the Slovenian mechanical industry is exploited. The incontestable fact remains that the influence of innovation on the companies’ revenues and profit remains too low. A clear strategy of innovation and appropriate further activities are the crucial factors leading to an increase of this influence [21]. The strategy should consistently support the innovation process and strongly focus on most important activities leading to the best innovation performance. Since the innovation introduced to the market is only the last of the links in the invention-innovation chain, a comprehensive and systematic approach is required. In order to asure such an approach, the “innovation of management” [22] in the way that it would be able to manage the innovation process effectively remains a prerequisite.

5 REFERENCES [1] European Innovation Scoreboard 2007 Comparative analysis of innovation performance. Luxembourg: Office for Official Publications of the European Communities. 2008.

[2] Statistical Office of the Republic of Slovenia (2008), Rapid Reports - Research and development, science and technology 200406, , no. 49, 25 p.

[3] Hlavaty, M. (2002). SKD - standard classification as per activities (NACE Rev.1).

Official Gazette of the Republic of Slovenia, 20 p. (in Slovenian).

[4] Coombs, R., Narandren, P., Richards, A. (1996). A literature-based innovation output indicator. Research Policy, vol. 25, no. 3, p.

403-414.

[5] Freel, M.S. (2005). Patterns of innovation and skills in small firms. Technovation, vol.

25, no. 2, p. 123-134.

[6] Hollenstein, H. (1996). A composite indicator of a firm’s innovativeness. An empirical analysis based on survey data for Swiss manufacturing. Research Policy, vol.

25, p. 633-645.

Statistical Analysis for Strategic Innovation Decisions in Slovenian Mechanical Industry

503

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 497-504

[7] Hagedoorn, J., Cloodt, M. (2003). Measuring innovative performance: is there an advantage in using multiple indicators?

Research Policy, vol. 32, no. 8, p. 1365– 1379.

[8] Iansiti, M. (1997). From technological potential to product performance: An empirical analysis. Research Policy, vol. 26, no. 3, p. 345-366.

[9] Carayannis, E., Gonzalez, E., Wetter, J. (2003). The nature and dynamics of discontinuous and disruptive innovations from a learning and knowledge management perspective. International Handbook on Innovation, p. 115-138.

[10] Michalisin, M. (2001). Validity of annual report assertions about innovativeness: an empirical investigation, Journal of Business Research, vol. 53, p. 151-161.

[11] Fatur, P., Likar, B. (2009). Development of performance measurement methodology for idea management. International Journal of Innovation and Learning, vol. 4, p. 422-437.

[12] Parthasarthy, R., Hammond, J. (2002).

Product innovation input and outcome: moderating effects of the innovation process.

Journal of Engineering and Technology Management, vol. 19, no. 1, p. 75-91.

[13] Žargi, U., Kušar, J., Berlec, T., Starbek, M. (2009). A company's readiness for concurrent product and process development. Strojniški vestnik - Journal of mechanical engineering, vol. 55, no. 7/8, p. 427-437.

[14] Anišić Z., Krsmanović, C. (2008). Assembly initiated production as a prerequisite for mass customization and effective manufacturing.

504

Strojniški vestnik - Journal of mechanical engineering, vol. 54, no. 9, p. 607-618.

[15] Perme, T. (2009). Translation of extended Petri net model into ladder diagram and simulation with PLC. Strojniški vestnik Journal of mechanical engineering, vol. 55, no. 10, p. 609-622.

[16] The Community Innovation Survey 2006.

Retrieved on 14. 9. 2009, from http://epp.eurostat.ec.europa.eu/portal/page/ portal/eurostat/home.

[17] Eurostat 2008. (2008). Science, technology and innovation in Europe. European Commission-Statistical Books, p. 83-154.

[18] Fifth Community Innovation Survey: Statistical Office of the Republic of Slovenia.

Retrieved on 14. 9. 2009, from http://www.stat.si/doc/vprasalniki/INOV-PS_2006.pdf.

[19] Chesbrough, H.W. (2003). Open innovation.

The new imperative for creating and profiting from technology. Boston, Harvard Business School Press.

[20] Chesbrough, H., Vanhaverbeke, W., West, J. (2006). Open innovation: researching a new paradigm. Oxford, New York: Oxford University Press Inc. [21] Collins, J.C. (2001). Good to Great. Why some companies make the leap... and others don't. Random House Business Books, London.

[22] Mulej, M., Ženko, Z. (2004). Introduction to systems thinking with application to invention and innovation management.: Management Forum, 2004. Maribor On CD.

Fatur, P. – Likar, B.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 505-510 UDC 685.632:624.014.2:519.6

Paper received: 04.11.2009 Paper accepted: 05.07.2010

Characterization of the Dynamic Behaviour of a Basketball Goal Mounted on a Ceiling Matija Javorski – Primož Čermelj – Miha Boltežar* University of Ljubljana, Faculty of Mechanical Engineering, Slovenia An experimental modal analysis was performed on an existing basketball goal, with the modes, natural frequencies and damping ratios identified. On this basis, a numerical model was created, employing the finite-element method. Due to the correlation between the experimental and the numerical results, the model was accepted as valid. Subsequently, this model was used to analyse the transient response that occurs when a fully loaded goal suddenly becomes unloaded. The resulting oscillation was compared to the valid standard for this type of sports equipment. ©2010 Journal of Mechanical Engineering. All rights reserved.

Keywords: basketball, steel structures, finite element method, experimental modal analysis, oscillations 0 INTRODUCTION The trend of building ever-larger sports halls is a world-wide phenomenon. Inevitably, ceiling heights are also increasing and, supposing that a ceiling basketball goal is used, it must span greater vertical distances. The suitability and quality of a goal, mounted on a ceiling at a height of 13, 14 or 15 m cannot be evaluated merely on the basis of data acquired from a stress analysis, as the flexibility of the goal is not negligible. In the event of a score with a dunk shot, the dynamic properties of the structure come into play and these can significantly influence the subsequent quality of the game. This fact has also been considered by the FIBA organization, which defined the maximum time for which any vibration could remain visible to be 4 seconds [1].

The objective of this research was, firstly, to create a valid numerical model that makes it possible to conduct a valid analysis of the dynamic behaviour of the structure under investigation. For this purpose, an experimental modal analysis (EMA) was conducted on the actual structure, Fig. 1, and the results were then used to create a valid finite-element model.

Secondly, once a valid numerical model was obtained, it was used in a transient numerical analysis. The aim was to investigate whether the structure satisfies the prescribed standards. In case the demands would not have been met or additional requirements emerged, the model could also be used as an evaluation tool in a modification process.

Fig. 1. Ceiling basketball goal assembly; 1) base, 2) upper main arm, 3) lower main arm, 4) upper foldable arm, 5) lower foldable arm, 6) ceiling Since the basketball goal assembly is relatively involved, with welds as well as bolted and pinned joints, and even a nonlinear connection between the foldable arms, building a suitably simple but still valid model is one of the most important steps in such approaches. There has been a great deal of research conducted in the field of joints and their use in structural dynamics, e.g., [2] and [3]. But, for our purpose, the most important are the guidelines on how to incorporate complex joints into the linear

*

Corr. Author's Address: University of Ljubljana, Faculty of Mechanical Engineering, Aškerčeva 6, 1000 Ljubljana, Slovenia, miha.boltezar@fs.uni-lj.si

505

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 505-510

structural dynamics. Such works, mostly emphasizing the validity of the model and valid but simplified models of complex sub-structures, can be found in [4] to [8]. Moreover, as was already established by Ewins and Inman [9] and Dascotte and Swindell [10]: ”The model should be as simple as possible, while still reflecting the most significant properties“. Our approach was also constrained by the methods of linear structural dynamics and thus focused on the overall validity of the model.

custom-made EMA software. More specifically, a generalized least-squares fitting of the MIMO (multiple input, multiple output), frequencydomain algorithm [11] was used to extract the modal parameters

1 MODAL ANALYSIS

with the No output and the Ni input coordinates, and with r being the complex modal vectors and r being the complex modal frequencies. HNoxNi is a matrix of measured FRFs for the structure.

In our case there were 36 input coordinates and one output coordinate. Points 130 and 140 were introduced artificially for the purpose of better visualisation, Fig. 3.

Despite the fact that the structure has many modes in the frequency band from 0 to 100 Hz, we were mostly focused on the frequency band ranging from 0 to 10 Hz. Even with this the modal density was relatively high and a suitable correlation between the experimental and numerical model as well as a good insight into the dynamics of the structure was essential for the subsequent analysis.

1.1 Experimental Modal Analysis In order to acquire the modes, frequencies and damping ratios, the EMA was carried out on the selected basketball goal using the MISO (multiple input, single output) approach with a fixed accelerometer and a roving hammer as the excitation. The measurement setup is displayed schematically in Fig. 2.

rrT r*r*T H No  Ni ( )     i  r* r 1 i  r N

N

 r 1

r

A r AT

i  r



r

A* r A*T

i  r*

(1)

Fig. 2. Measurement setup For the actual modal testing, 14 measurement points (1–14) were selected, mostly on the lower part of the goal, Fig. 3. This was due to the fact that this part of the structure is excited the most. The accelerometer was positioned at point 8 on the lower main arm and oriented in a lateral direction (with respect to the court), while at the other 13 points, the structure was excited in two or three mutually perpendicular directions, depending on the accessibility of the measurement point. Overall, 36 frequencyresponse functions (FRFs) were recorded (Fig. 4) and the modal identification was conducted using

506

Fig. 3. Measurement points on the basketball goal

Javorski, M. – Čermelj, P. – Boltežar, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 505-510

Fig. 4. One of the measured FRFs (solid line) and its reconstruction (dashed line) showing the quality of the EMA 1.2 Numerical Modal Analysis A numerical modal analysis for the basketball goal was carried out using a FEA (finite-element analysis) approach. Since the great majority of the structure is made from long and relatively slender rectangular steel tubes, it was acceptable to model the goal as a system of beams. By default, all the members but the pinned joints, enabling the folding of the goal (Fig. 5), were rigidly coupled at their junctures.

Fig. 6. Connection between the foldable arms

Fig. 5. Folding action of the basketball goal The pinned joints were modelled as a special type of node coupling, with all the relative translations being fixed, while only the rotation around the axis of the joint was released.

However, the most complex connection on the structure is at the location where the adjustable bolt from the upper foldable arm contacts the surface of the lower foldable arm, Fig. 6.

It is evident that only the compressive forces can be transferred across this connection, while the design of the goal is such that its own weight provides these forces, keeping the goal in the operational position. In performed model, the bolt was simplified by creating a binding with a similar geometry to the bolt, rigidly coupled to both folding arms. This was a necessary but reasonable simplification that does not affect the global response of the structure, as will be evident from the results.

Besides the steel tubes, the basketball goal is also composed of the glass backboard and the basket ring, which also influence its dynamics.

Due to the location of these components at the bottom of the structure, accurately modelling their inertial properties was of greater importance than modelling their stiffness. To represent them,

Characterization of the Dynamic Behaviour of a Basketball Goal Mounted on a Ceiling

507

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 505-510

a block with the width and the height of the glass was modelled, while the thickness was selected such that the mass of the block was equal to the sum of the actual masses of the represented objects. For this purpose a linear, 3D solid element was used. The whole structure was rigidly connected to the ceiling.

mode, where a global, torsional motion is well correlated with the numerical approach, Fig 9.

1.3 Results For the purposes of correlation and, finally, to identify the correlated mode pairs that are important for the forthcoming analysis, the results of both approaches were analysed.

Fig. 8. Second mode; experimentally identified motion (left), fEMA = 2.71 Hz, numerically assessed deformation (right), fFEA = 3.18 Hz

Fig. 7. First mode; numerically assessed deformation, fFEA = 2.77 Hz The first mode shape, Fig. 7, was observed only for the numerical model. This is due to the fact that the accelerometer could not detect this local behaviour occurring away from its position.

This local mode is characterized by the bending motion at the folding arms. Due to the simplification in the connection of the arms this numerical solution is only correct up to a certain amplitude of oscillation. As the arms sag under their own weight to the extent of 20 mm, the model correctly represents this oscillation with an amplitude of no more than half of this value. At greater amplitudes, a nonlinear oscillation occurs. fEMA is the natural frequency assessed in the EMA, while fFEA is the natural frequency resulting from the numerical approach.

In contrast to the first, the second mode was indentified with both the experimental and numerical approaches and represents the motion of the goal in the lateral direction, Fig. 8. The same applies to the global motion of the fourth 508

Fig. 9. Fourth mode; experimentally identified motion (left), fEMA = 6.95 Hz, numerically assessed deformation (right), fFEA = 7.20 Hz Table 1 compares the numerical and experimental results, where ξEMA is the damping ratio assessed in the EMA and η is the discrepancy between fEMA and fFEA..

Besides mode 1, modes 3 and 6 were also not identified experimentally. Judging from the numerical results, the reason was again the actual position of the accelerometer and its limited sensitivity, preventing the capture of all the local behaviour of the structure.

Table 1. Comparison of some of the modal results for the numerical and for the experimental approach

Javorski, M. – Čermelj, P. – Boltežar, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 505-510

No. 1 2 3 4 5 6 7 8

fEMA [Hz] ξEMA [/] fFEA [Hz] η [%] / / 2.77 / 2.71 0.08 3.18 17 / / 4.96 / 6.95 0.048 7.20 4 8.24 0.047 8.41 2 / / 8.85 / 8.96 0.038 9.02 1 10.78 0.041 9.76 9

Despite this, the correlation between the model and the actual structure was recognized as being valid. Namely, as will be clearly evident in the next section, our concern was mainly the second numerical mode and its correlation with the experimental one. This mode dominates in the actual, transient response, when the basketball goal is excited laterally, with the initial displacement imitating this mode.

where ε and ν are constants and the following relations hold 2r  r  r  C  r .

T

(3) Due to the orthogonality of the eigenvectors with respect to the mass and stiffness matrices, Eq. (3) can be rewritten as 2r  r    r , 2

(4)

or

r 

1 2r



r 2

,

r  1, 2, ... N .

(5) In Eq. (5) ξr is defined as the modal damping ratio for mode r, while N is the number of modes calculated or experimentally identified.

2 TRANSIENT RESPONSE With a valid numerical model at our disposal, it was then possible to predict a valid transient response of the goal to different, timedependent loadings. It was observed that a crucial dynamic behaviour of the structure takes place when there is a lateral loading applied to the structure of the basketball goal. For our analysis, a load case was conceived, where a static load of 900 N (equal to the static lateral load capacity of the goal [12]) is applied to the ring fixture from the side and then the structure is released to oscillate freely. This kind of loading causes a damped, transient response of the structure, gradually returning the structure into its neutral position. Besides the overall stiffness of the structure, which defines the initial displacements, the damping is one of the most important aspects to account for as accurately as possible.

Experimentally identified damping, as a direct result of the EMA, is thus an important piece of information that influences the validity of the transient response analysis. Moreover, in order for the identified damping to be easily used in the numerical approach, a proportional (or Rayleigh) damping was adopted [13]. In this model it is assumed that the damping matrix of a multipledegree-of-freedom system is proportional to the mass and stiffness matrices. C     K     M  , (2)

Fig. 10. Response of the structure at the place of the ring fixture, where a lateral force of 900 N was applied The response of a continuous system, like the structure of the basketball goal being reviewed, is the sum of all of its modes in general. However, in our case it was observed that in the transient response, due to the actual, lateral force and calculated using the FEA, there is a dominant motion of the second mode shape.

Due to this characteristic, only the identified damping ratio corresponding to the second mode shape was used to derive the coefficients of the proportional damping model to be used in the FEA analysis. The response of the structure to the load case described above is shown in Fig. 10.

One of the purposes of this research was to compare the dynamic attributes of the goal with the regulations for this type of sports equipment.

Because of sufficient damping, the initial

Characterization of the Dynamic Behaviour of a Basketball Goal Mounted on a Ceiling

509

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 505-510

oscillation amplitude of 14 mm substantially drops with every swing, consequently decreasing to a negligible value in a time of about 2.5 s. This is well below the allowed visible vibration duration time of 4 seconds, clearly indicating the suitability of the goal.

[4]

3 CONCLUSION [5] The use of the experimental modal analysis and the finite-element method at the same time proves to be very useful in research on structural dynamics. Experimental results are used to correctly build and, afterwards, validate the numerical model, which can then be used to analyse the various aspects of the behaviour of the structure. In the case of a ceiling basketball goal, the two most common load cases are an impulse from a basketball hitting the backboard or an impulse from a player scoring with a dunk shot. Since both of these impulses are very difficult to assess, a different approach was adopted, using a static force and suddenly eliminating it. With the use of the force, equivalent to the static load capacity of the goal, the maximum static deformation was achieved, ensuring an oscillation of the maximum amplitude was analysed.

The biggest advantage of a numerical model is that it enables us to examine the effect of structural modifications on the behaviour of the structure. Typical modifications are a weight reduction, a material change and thus a manufacturing-costs reduction, or any static or dynamic behaviour improvements to the structure or its components. Having such valid models, the next time this can be used in a prototyping phase, without the necessity to have an existing structure or a prototype.

[6]

[7]

[8]

[9]

[10]

[11]

4 REFERENCES [1] [2]

[3]

510

FIBA regulations: »Official Basketball Rules 2008, Basketball Equipment, As approved by FIBA Central Board«.

Ren, Y., Lim, T.M., Lim, M.K. (1998).

Identification of properties of nonlinear joints using dynamic test data. Journal of Vibration and Acoustics - Transactions of the Asme, 120, p. 324-330.

Oldfield, M., Ouyang, H., Mottershead, J. E. (2003). Bolted joints under dynamic

[12]

[13]

loading. VIII International Conference on Recent Advances in Structural Dynamics, Southampton, UK, 14-16 July, ISVR. Čelič, D., Boltežar, M. (2008).

Identification of the dynamic properties of joints using frequency–response functions.

Journal of Sound and Vibration, 317, p.

158–174. Čelič, D., Boltežar, M. (2009). The influence of the coordinate reduction on the identification of the joint dynamic properties. Mechanical Systems and Signal Processing, 23, p. 1260–1271. Čermelj, P., Boltežar, M. (2006). Modelling localised nonlinearities using the harmonic nonlinear super model. Journal of Sound and Vibration, 298, 1099–1112.

Fotsch, D. W. (2001). Development of Valid FE Models for Structural Dynamic Design. PhD Thesis, Imperial College of Science, Technology and Medicine, University of London.

Liu, W. (2000). Structural Dynamic Analysis and Testing of Coupled Structures.

PhD Thesis, Imperial College of Science, Technology and Medicine, University of London.

Ewins, D.J., Inman, D.J. (2001). Structural Dynamics @ 2000: Current Status and Future Directions. Research Studies Press Ltd., Baldock, Hertfordshire, England.

Dascotte, E., Swindell, R. (2003). Beyond modal animations: modal pre-test planning, structural dynamics modifications & integration with fea. Modal Analysis and Lab Test Simulation Seminar, m+p International, UK. Catbas, F.N., Brown, D.L., Aktan, A.E. (2004). Parameter estimation for multipleinput multiple-output modal analysis of large structures. Journal of Engineering Mechanics-ASCE, vol. 130, p. 921-930.

European standard: »prEN 1270 rev 2004, Playing field equipment - Basketball quipment - Functional and safety requirements, test method«.

Maia, N.M.M., Silva, J.M.M., He, J., Lieven, N.A.J., Lin, R.M., Skingle, G.W., To, W., Urgueira, A.P.V. (1997).

Theoretical and Experimental Modal Analysis. Research Studies Press Ltd., Taunton, Somerset, England.

Javorski, M. – Čermelj, P. – Boltežar, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 511-520 UDC 531.1:621.95

Paper received: 24.06.2009 Paper accepted: 19.05.2010

Analytical Description of Polygonal Holes Boring - General Approach Miodrag Zlokolica* - Maja Čavić - Milan Kostić Faculty of Technical Sciences, University of Novi Sad, Serbia Equilateral polygonal holes can be manufactured on conventional machines such as lathes or drills using special tool. Tool has to rotate about its axis while its center simultaneously follows complex planar trajectory. Tool center motion can be achieved in several ways: by guiding tool with template which shape exactly matches the shape of the hole produced, using cam or planetary gears mechanism etc.

To describe accurately polygonal hole boring process it is necessary to determine tool geometry, tool motion as well as realized hole geometry. In order to prescribe optimal boring regime time-history of tool cutting blade tip speed has to be obtained. Holes are, generally, produced with rounded corners so, having in mind further operations with work piece, it is important to know corner radius.

Analytical approach presented in this paper enables easy and efficient forming of the mathematical model describing geometry and kinematics of polygonal hole boring process. © 2010 Journal of Mechanical Engineering. All rights reserved.

Keywords: polygonal hole, boring, kinematics 0 INTRODUCTION Polygonal contours of different geometry are widely used in many engineering applications – hexagonal hole in screw head is well known example. In case of the screws, deep extrusion process is used as the only one justified considering number of products, required surface quality and price. However, when there is a need for lesser number of products or they can not be produced using casting of forming, machining technologies has to be applied. Internal surfaces are typically produced using broaching, shaping or EDM procedures while for external ones milling is preferable. Generally, broaching and shaping have problems with accuracy and repeatability of dimensions. EDM gives precise dimensions but it is time-consuming process so, frequently, it is discarded due to high price. On the other side, there exist a lot of machining technologies with main motion based on rotation - motion easiest to obtain. They are rather inexpensive and fast, and among them, most interesting for this problem, is boring. In fact, polygonal holes can be manufactured on conventional machines such as lathes or drills using special tool. It is imperative that tool and hole geometry are compatible, that is to say, hole contour must represent an envelope to consecutive tool positions. Tool has to rotate about its axis while its center simultaneously follows complex planar trajectory. Tool center

motion can be achieved in several ways: by guiding tool with template which shape exactly matches the shape of the hole produced (Watts drilling system [1]), using cam mechanism (Formbore system [2[ and [3]) etc. Beside main rotational motion tool has to perform auxiliary linear motion. Watts drill is based on curves of constant width (Roleaux polygons) theory, but it is only applicable on polygonal holes with even number of sides. On the other side when using cams, a different one has to be calculated and applied for each polygonal hole, so it becomes a case study procedure.

In order to prescribe optimal boring regime it is very important to know time-history of tool cutting blade tip speed. Polygonal holes are, generally, produced with rounded corners in order to avoid impact load on the tool, so, having in mind further operations with work piece assembly for instance, corner radius has to be calculated. Using concept of alternative mechanism [4] and centrodes theory for planar motion [5] those two problems were resolved first graphically for square hole [6]. Analytical approach using centrodes generalized to calculate blade tip velocity and hole corner radius for all types of equilateral polygons was proposed in [7] and [8]. Idea was further developed resulting in general analytical approach which enables easy and efficient forming of the complete mathematical model describing geometry and

*

Corr. Author's Address: Faculty of Technical Sciences, Trg Dositeja Obradovića 6, 21000 Novi Sad, Serbia, mzlokolica@uns.ac.rs

511

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 511-520

kinematics of any equilateral polygonal hole boring process which is presented in this paper.

1 BASIC GEOMETRY OF TOOL AND HOLE Cross-section of tool for manufacturing equilateral n-sided polygonal hole is an equilateral n-1-sided polygon. During process, trajectories of two tool cutting blade tips coincide with adjacent hole sides, i.e. they actually produce hole sides.

Other blades cut rounded corners of the hole.

In Fig. 1 tool and hole geometry parameters relevant for analysis are presented, indexes 1 and 2 refer to tool and hole parameters respectively. Point O is hole center while point C denotes tool center.

Each equilateral polygon can be assembled from triangles with two equal sides. Central angles of these triangles are (n is number of sides of produced polygonal hole): 2  1 ( n )  , (1) n 1

 2 ( n) 

2  . n

Peripheral angles of tool hole  2 (n) are: n3 1 (n)   , n 1

1 ( n )

and

(3)

n2  .

(4) n Mutual position of tool and hole is defined with (valid for n>4): 1  1 ( n)   , (5) (n  1)  n (n  2)  (n  3)  .

2  (n  1)  n

(6)

Tool side dimension can be calculated as: a(n)  A  cos(1 (n))  (7)  A  sin(1 (n))  tan( 2 (n)) if n>4 , otherwise a(4)=A. (8) In case when n = 4 tool side dimension can not be calculated accordingly to (7) because tool cutting blade tips will not describe desired hole contour so value defined by (8) is accepted.

512

r1 (n) 

r2 (n) 

a ( n) 2  sin(

1 ( n ) 2

(9)

)

A 2  sin(

 2 ( n) 2

)

(10)

(2)

 2 ( n) 

 2 ( n) 

Radii of circles circumscribed about tool and hole respectively, r1 (n) and r2 (n) , are:

Fig. 1. Tool and hole geometry parameters 2 CHARACTERISTIC POINTS POSITION – FORMING PARAMETRIC EQUATIONS OF MOTION Characteristic points are: tool center C, tool cutting blade tips and instantaneous velocity pole N. Equations of motion are given with respect to tool rotation angle  2.1 Position of Tool Center - Local Parametric Equation Origin of fixed coordinate system xOy is positioned in the polygonal hole center O while y axis coincides with axis of symmetry of the hole corner. In case of n-sided polygonal hole this analysis will define only one part of the tool center trajectory that would be the one around the particular corner (  g (n)     g (n) ). Tool center position parameters are presented in Fig. 2.

Equations of two adjacent hole sides are: y1 ( x, n)  k (n)  x  r2 (n) , (11) y2 ( x, n)  k (n)  x  r2 (n) ,

Zlokolica, M. - Čavić, M. - Kostić, M.

(12)

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 511-520

where k (n) is given as: k (n)   tan(





 2 ( n)

).

(13) 2 2 Parametric equation of the cutting tool side is: y12 ( x,  , n)  tan( )  x  b12 ( , n) , (14)

where parameter  is measured as angle between tool side and x axis.

are ( X 1 ( , n) , Y1 ( , n) ) and ( X 2 ( , n) , Y2 ( , n) ) it can be written that:

Y2 ( , n)  Y1 ( , n) 2    X 2 ( , n)  X 1 ( , n) 2  a (n) 2 .

After transformations, b12 ( , n) is obtained:

(19)

expression

for

b12 ( , n)   r2 (n) 

a(n)  (tan( ) 2  k (n) 2 ) 2

2  k (n)  tan( )  k (n)

2

.

(20)

Tool center C is situated at intersection of y121 ( x, , n) and y122 ( x, , n) - lines which connect respective tool tip and tool center: y121 ( x, , n)  a121 ( , n)  x  b121 ( , n) , (21) y122 ( x, , n)  a122 ( , n)  x  b122 ( , n) ,

(22)

where a121 ( , n) and a122 ( , n) are given as: a121 ( , n)  tan(

Fig. 2. Tool center position parameters Coordinates ( X 1 ( , n) , Y1 ( , n) ) and ( X 2 ( , n) , Y2 ( , n) ) of points 1 and 2 (tool tips), are found as intersection of ( y12 ( x, , n) , y1 ( x , n ) ) and ( y12 ( x, , n) , y2 ( x, n) ) respectively (Fig. 2.).

Relation y1 ( x, n)  y12 ( x, , n) is valid for point 1 so it can be written that: r (n)  b12 ( , n) X 1 ( , n)  2 , (15) tan( )  k (n) Y1 ( , n)  k (n)  X 1 ( , n)  r2 (n) .

(16)

In the same way, following expressions can be obtained for point 2: r (n)  b12 ( , n) X 2 ( , n)  2 , (17) tan( )  k (n) Y2 ( , n)   k (n)  X 2 ( , n)  r2 (n) .

(18)

In those equations b12 ( , n) represents intersection of line y12 ( x,  , n) and y axis.

It can be calculated using the fact that distance between tool tips i.e. intersection points 1 and 2, of lines ( y12 ( x, , n) , y1 ( x , n ) ) and ( y12 ( x, , n) , y2 ( x, n) ) respectively, always equals a(n) . So, if coordinates of point 1 and 2

a121 ( , n)  tan(

 2





 1 ( n) 2

 ) ,

 1 ( n)

 ) .

2 Coefficients b121 ( , n) and b122 ( , n) 2



(23) (24)

are determined using the fact that lines y1 ( x, , n) and y121 ( x, , n) intersects at point 1, while lines y 2 ( x, , n) and y122 ( x, , n) intersects at 2.

b121 ( , n)   Y1 ( , n)  a121 ( , n)  X 1 ( , n) , b122 ( , n)   Y2 ( , n)  a122 ( , n)  X 2 ( , n) .

(25) (26)

Finally, position of tool center C is determined as: b ( , n)  b122 ( , n) X Cl ( , n)  121 (27) a122 ( , n)  a122 ( , n) , YCl ( , n)   a121 ( , n)  X Cl ( , n)  b121 ( , n) .

(28)

It is important to emphasize that Eqs. (27) and (28) are valid in  g (n)     g (n) interval i.e. around axis of symmetry of observed hole corner. Because of that they will be called tool center local equations.

Analytical Description Of Polygonal Holes Boring - General Approach

513

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 511-520

2.2 Position of Fixed Centrode Points – Local Parametric Equations

Fixed centrode presents geometric position of the instantaneous velocity pole (Fig. 3). Since tool cutting blade tips 1 and 2 move along hole sides y1 ( x , n ) and y 2 ( x, n) , respectively, velocities of 1 and 2 are collinear with y1 ( x , n ) and y 2 ( x, n) , so intersection of normals to y1 ( x , n ) and y 2 ( x, n) at points 1 and 2 represents instantaneous velocity pole N. Equations of lines y1n ( x,  , n) and y 2n ( x, , n) are: y1n ( x,  , n)  a1n (n)  x  b1n ( , n) , (29) y 2n ( x,  , n)  a2n (n)  x  b2n ( , n) ,

(30)

where: a1n (n)   a2n (n) 

1 , k ( n)

1 . k ( n)

(31) (32)

Fixed centrode point (instantaneous velocity pole) N is found as intersection of lines y1n ( x, , n) and y 2n ( x, , n) .

Parametric equations defining position of instantaneous velocity pole N are: b ( , n)  b2n ( , n) , X l N ( , n)  1n (35) a 2n (n)  a1n (n) Y l N ( , n)  a1n (n)  X l N ( , n)  b1n ( , n) .

(36)

Again, Eqs. (35) and (36) are valid in  g (n)     g (n) interval i.e. around axis of symmetry of observed hole corner. Because of that they will be called fixed centrode local equations.

2.3 Position of Moving Centrode Points – Local Parametric Equations

Moving centrode is curve rigidly connected to the tool which, while tool moves, rolls without sliding over fixed centrode (Fig. 4.).

If tool center and fixed centrode points coordinates are known, moving centrode can be easily defined – equation defining position of its points is obtained by expressing fixed centrode points coordinates in local, moving coordinate system x1Cy1. It is rigidly attached to tool, axis y1 representing axis of symmetry of the tool side and its origin is situated in the tool center C.

Fig. 3. Instantaneous velocity pole position Coefficients b1n ( , n) and b2n ( , n) are determined knowing that lines y1 ( x , n ) and y1n ( x,  , n) intersects at 1 ( X 1 ( , n) , Y1 ( , n) ), while y 2 ( x, n) and y 2n ( x,  , n) intersects at 2 ( X 2 ( , n) , Y2 ( , n) ) . b1n ( , n)   Y1 ( , n)  a1n ( , n)  X 1 ( , n) ,

b2n ( , n)   Y2 ( , n)  a 2n ( , n)  X 2 ( , n) .

514

(33) (34)

Fig. 4. Fixed and moving centrode From Fig. 4 it can be seen:    rN  rC  rP .

(37)

Using matrix transformation between x1Cy1 and xOy following relation can be written:

Zlokolica, M. - Čavić, M. - Kostić, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 511-520

 X l N ( , n)  l   Y N ( , n)     0   (38) cos( )  sin( ) X lC ( , n)  X lP ( , n)       sin( ) cos( ) Y lC ( , n)    Y lP ( , n)  ,  0   1  0 0     which, after some transformations, gives local equations of moving centrode:  X lP ( , n)  l  Y P ( , n)    0   cos( )   sin( )  0 

 

sin( )  X lC( , n)  cos( )  Y lC( , n)  sin( )  (39) cos( ) X lC( , n)  sin( )  Y lC( , n)  cos( )    0 0 

 X l N( , n)    Y lN ( , n)   1   

2.4 Position of Fixed Centrode Points, Tool Center and Moving Centrode Points – Global Parametric Equations

First, global parametric equations for fixed centrode points will be determined. Locally, fixed centrode is defined with ( X l N ( , n) , Y l N ( , n) ), where angle  is parameter. Manufactured polygonal hole represents centrally symmetrical curve which implies central symmetry of fixed centrode. Furthermore, that suggests transition from Descartes to polar coordinates: r l N ( , n)  

Xl

2

N ( , n)  Y

l2

N ( , n)

,

(40)

 X l N ( , n)   .

(41) l   Y N ( , n)  As problem geometry is same around axis of symmetry of each hole corner it is obvious that fixed centrode has to be periodical curve with period:  NT (n)   2 (n) .

(42)

 l N ( , n)  arctan

Eqs. (40) and (41) are highly nonlinear with respect to parameter, so it is not possible to

obtain analytical expression r l N  r l N ( l N ) . On the other hand, period g of the parameter can be calculated using relationship:

 l N (g, n)   NT (n) .

(43)

In this way, an interval ( g    g ) around one axis of symmetry (one hole corner) in which local equations are valid can be determined. Polar radius of global curve is periodical function of polar angle  l N with period  NT (n)   2 (n) but it is also periodical function of , with period 2g.

In order to calculate global radius rN  rN ( , n) of fixed centrode expansion of (41) to Fourier series will be used on interval  ( n)  ( n)  2   l N ( n)  2 .

2 2 Using property of Fourier series which says that if a function is expanded on interval, than Fourier series not only represents function approximation on that interval but it will also be its periodical continuation, global radius of fixed centrode can be calculated for whole period  N ( , n)  (0  2   ) .

As said before geometry of polygonal hole is symmetrical with respect to y axis (Figs. 2 and 3). Since geometry of the fixed centrode is determined by the geometry of the polygonal hole, centrode itself has to be symmetrical with respect to y axis. This means that local radius (41) has to be even function i.e.: r l N ( l N )  r l N ( l N )

(44)

or r l N ( , n)  r l N ( , n) .

(45)

Because of that, expansion to Fourier series can be simplified and global radius of fixed centrode can be finally expressed as: rN ( , n)  

Af 0 ( n )  i    Af i ( n )  cos(  ) , 2  PNT ( n ) i 1

(46)

where Af i (n) are Fourier coefficients which are determined as: Afi (n)  NT ( n)



2  NT (n)

l

i 

NT ( n)

NT

2



 r N ( , n)  cos(

 )d .

(47)

2

Analytical Description Of Polygonal Holes Boring - General Approach

515

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 511-520

Polar angle function angle is also periodical function with period  NT (n) (parameter period is 2  g ). In order to define global curve of the fixed centrode polar angle has to fill complete interval  N ( , n)  (0  2   ) .

From Fig. 5 it is obvious that global function of the polar angle can be obtained by continuous supervening of polar angle local functions  l N ( , n) until  N ( , n) reaches 2   . So, as in interval (0  2  ) there are n periods ( n   NT (n)  2   ), local function  l N ( , n) has to be added n times in order to form  N ( , n) .

While parameter  changes in interval 0    2  n  g polar angle describes whole fixed centrode i.e.  N ( , n)  (0  2   ) .

Same procedure can be applied in order to form tool center global curve C ( , n) , rC ( , n) .

Some interesting conclusions have been obtained while performing the procedure.

Namely, after forming and analyzing global equation for radius of tool center trajectory rC ( , n) it was concluded that trajectory has nearly circular shape. Eccentricity of trajectory, is introduced as: max(rC ( , n)) . e( n)  (50) min(rC ( , n)) It resumes values significantly close to 1 for all cases of n.

Also, after analysis of polar angle global function C ( , n) (Fig. 6) it has been concluded that it behaves close to function: C ( , n)  (n  1)   .

(51) Differentiation of (51) gives relationship between angular velocities in the following form: (52)  ( , n)  (n  1)   .

C

Fig. 5. Fixed centrode – global curve Index j is introduced, defining an interval in which instantaneous value of parameter  is situated:   j  whole part of   2  g 

 .  

Fig.6. Function C ( , n) with respect to  (case n=5)

(48)

Global function of the polar angle can now be written as:  N ( , n)  (49)   N (  2  j  g, n)  j   N T (n) .

Obtained curve:  N ( , n) , rN ( , n) (eqs.

(46) and (49)) represents fixed centrode global curve (  N ( , n)  (0  2   ) while 0    2  n  g ) (Fig. 5).

516

Fig. 7. Tool motion

Zlokolica, M. - Čavić, M. - Kostić, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 511-520

This means that tool motion (Fig. 7.) can be realized as superposition of two rotations: tool has to rotate around its longitudinal axis with angular velocity  and, at the same time, to rotate around hole centre O with angular velocity C , radius of rotation is equal to average (rC ( , n)) . Such motion can be easily realized using planetary gear mechanism.

Moving centrode global curve  P ( , n) rP ( , n) can be defined following the same procedure. It is also centrally symmetrical and periodical curve, its period depending on its shape. Since tool geometry defines shape of moving centrode period of global polar angle P ( , n) will be: 2    1 (n) .

(53) n 1 Obtained curve gives coordinates in moving coordinate system x1Cy1. Its origin is positioned at the tool center X C ( , n) , YC ( , n) , while axis x1 forms angle  with axis x of fixed coordinate system xOy. Now, moving centrode can be expressed in fixed coordinate system using following matrix transformation:

all tool cutting blades tips m = 1, 2, 3, ..., n-1 (Fig. 8):

 X m ( , n )   Y ( , n )    m   X ( , n )    sin(  ( , n , m ))   C    r1 ( n ) .  YC ( , n )   cos(  ( , n , m )) 

(55)

PT (n) 

 X f P ( , n) cos( )  sin( ) X C ( , n)  f     Y P ( , n)    sin( ) cos( ) YC ( , n)      0 0 0 0      X P ( , n) .   YP ( , n)   1 

where

X

p

( , n )

and

Yp ( , n)

Fig. 8. Position of tool cutting blades tips Tool tips m=1, 2 are situated on hole sides.

Angle  ( , n, m) , is calculated as:

(54)

are

Descartes coordinates of moving centrode in moving coordinate system.

2.5 Position of Tool Cutting Blades Tips

In case of manufacturing n-sided equilateral polygonal hole, tool has to have shape of n-1-sided equilateral polygon, two of its corners being in contact with hole sides.

Parameter m which defines tool cutting blades tip index is introduced. With known global position of tool center X C ( , n) , YC ( , n) , distance r1 (n) between tool center and tool tip as well as angle  ( , n, m) , it is easy to calculate position of

 ( , n, m) 

1 ( n ) 2

  1 ( n )  ( m  2)   .

(56)

Since global coordinates of tool center were used in Eq. (55), obtained X m ( , n, m) and Ym ( , n, m) represent global coordinates of mth tool tip.

After forming trajectory of each tool tip (m=1, 2...n-1) for one period (  0 2  n   g ) they are combined thus forming closed curve which represents contour of real manufactured hole (Fig. 9).

2.6 Radius of Curvature of Tool Cutting Blades Tip Trajectory

In general, if the moving point trajectory is  defined by r ( s ) , (s is the parameter) than its radius of curvature is defined as:  3 r ( s ) (57)  (s)   ,  r ( s )  r( s )

Analytical Description Of Polygonal Holes Boring - General Approach

517

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 511-520

In this case tool tip coordinates are X m ( , n, m) , Ym ( , n, m) , so its position vector in xOy is given as:  rm ( , n, m)    (58)  X m ( , n, m)  i  Ym ( , n, m)  j .

Radius of curvature is then (acc. to Eq. (57)): 3  rm ( , n, m) (59)  ( , n, m)    rm ( , n, m)  rm ( , n, m) or, after differentiation with respect to : (, n, m)  

X m(, n, m)2 Ym(, n, m)2

3

X m(, n, m)Ym(, n, m)  Xm(, n, m)Ym(, n, m)

.

(60)

When tool tip enters hole corner radius of curvature decreases, so it assumes its minimal value  min at positions defined by :

 ρmin  0,2  g,4  g  2  n  g .

(61)

Those positions are determined parameter period 2   g (Fig. 5).

by

  vm ( , n, m)  rm ( , n, m)     X m ( , n, m)  i  Ym ( , n, m)  j .

(62)

Absolute value of velocity is obtained as: vm ( , n, m)  (63)  X m ( , n, m) 2  Ym ( , n, m) 2 .

3 EXAMPLE – PENTAGONAL HOLE BORING Hole profile is equilateral pentagon so tool will have square shape. Necessary geometrical parameters, according to Eqs. (1) to (9) have been calculated:  tool and hole central angles: 1 (5)  90 ,  2 (5)  72 ,  tool and hole peripheral angles: 1 (5)  90  ,  2 (5)  108 ,  angles determining mutual position of tool and hole: 1 (5)  9  ,  2 (5)  27  ,  dimension of tool side: It was adopted that A=1 cm, so a(5)=1.05 cm  radii of circles circumscribed about tool and   hole: r1 5  0.741 cm , r2 5  0.851cm Using procedure derived in Section 2 graphs presenting position of tool center (Fig. 10), fixed centrode (Fig. 11), moving centrode (Fig. 12), real contour of the manufactured polygonal hole (Fig. 9) and tool cutting blade tip velocity (Fig. 13) are obtained.

Fig. 9. Real contour of the manufactured polygonal hole (case n=5) Corner radius  min , minimal corner radius  min , tool tips 1, 2, 3 and 4 position vector and trajectories for one period, as well as complete hole contour for case n=5, are presented in Fig. 9.

2.7 Tool Cutting Blades Tip Velocity

With position of tool tip known it is easy to calculate its velocity vector: 518

Fig. 10. Position of tool center ( C ( ,5) , rC ( ,5) ) Change of tool cutting blade tip velocity with respect to  is presented in Fig. 13.

Zlokolica, M. - Čavić, M. - Kostić, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 511-520

Fig. 11. Fixed centrode (  N ( ,5) , rN ( ,5) )

Fig. 13. Tool cutting blade tip velocity Minimal radius of curvature appears at the hole corners, for pentagonal hole  min  0.17cm .

One period of hole boring process (corresponding to one turn of centrodes) is presented in Fig. 14. Moving centrode rolls over fixed one thus realizing tool motion. During the motion tool center C describes nearly circular curve K.

Fig. 12. Moving centrode ( P ( ,5) , rP ( ,5) )

Fig. 14. Pentagonal hole boring process

Analytical Description Of Polygonal Holes Boring - General Approach

519

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 511-520

In Working model 2D software planetary gear mechanism that realizes desired motion i.e. moves tool so it cuts desired equilateral pentagonal hole has been modeled and tool motion was simulated (Fig. 15).

sides, taking enough members in Eq. (46) ensures satisfactory precision.

Presented approach as well as the method used to solve the problem is general – applicable to all equilateral polygons, not only to 4, 6 an 8sided as methods based on curves of constant width. Even though the holes are cut with radius in corners, the fact that multiple tool tips are in contact with hole sides ensures good dynamics of boring regime (no impact in sharp corners, cutting force is distributed to more cutting blades thus diminishing stress to the tool). Also, a simple practical solution – planetary gear mechanism which can realize desired motion is proposed (no need for complex tool center trajectory realization – cams or robot guidance (EDM)).

5 REFERENCES

Fig. 15. Simulation of tool motion 4 CONCLUSION Using proposed analytical approach it is possible to analyze geometry and kinematics of tool in the process of boring equilateral polygonal hole.

Cross-section of tool for manufacturing equilateral n-sided polygonal hole is an equilateral n-1-sided polygon where cutting blades are positioned in respective tool tips and are always in contact with hole sides. Such geometry implies that hole corner will be cut with respective radius and the difference (distance between tool tip and hole corner) measures less then 5% of hole side A. Value of  min rises with A and, even more, with n (number of hole sides).

For n>10, a radius becomes significant, but having in mind that such a hole, even with precise geometry, slightly differs from a circle, it is not widely used and this fact does not limitate the usefulness of the method. Though existence of the radii is preferable because of better dynamic regime of cutting, in some cases it is necessary to cut hole with precise geometry. This can be done by reshaping the tool and making its center to follow an appropriate trajectory, which, in practice, leads to application of single cutting tip tool with cam mechanism. As for shape of hole

520

[1] Polygon Hole Drill. Retrived on 1.2.2009., fromj http://www.inigerspin.co.uk/polygon.htm.

[2] Koepfer, C. (1995). Boring non round holes, Modern Machine shop. Retrived on 1.02.2009 from http://www.mmsonline.com.

[3] Form drilling or turning device (1996). US patent, Patent Number 5,542,324.

[4] Norton, R.L. (1992). Design of Machinery, McGraw-Hill, Inc., New York, USA. [5] Erdman, A.G., Sandor, G.N. (1984).

Advance Mechanism Design, Analysis and Synthesis, vol. 2. Prentice Hall, Inc., New Jersey, USA. [6] Zlokolica, M. (1975). Kinematics of tools for polygonal drilling. Journal of Technique, Mechanical section, vol. 24, no. 6, p. 15-18.

[7] Zlokolica, M., Čavić, M., Kostić, M. (1999).

Centrodes in the process of kinematic description of the tools for the polygonal holes production. Proceedings of 1st International Conference The Coating in Manufacturing Engineering, p. 663-671, Thessaloniki, Greece, October 14.-15.,.

[8] Zlokolica, M., Sovilj, B., Čavić, M., Kostić, M. (2000). About the blade cutting speed of the tools for boring polygonal holes.

International Proceedings of 10th Conference on Tools, p. 277-281, Miskolc, Hungary, September 6.-8.

Zlokolica, M. - Čavić, M. - Kostić, M.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 521-530 UDC 316.4:001.895

Paper received: 7.10.2009 Paper accepted: 13.1.2010

The Role of the User and the Society in New Product Development Nusa Fain1,* - Niels Moes2 - Jože Duhovnik1 Faculty of Mechanical Engineering, University of Ljubljana, Slovenia 2 Faculty of Industrial Design Engineering, Technological University Delft, the Netherlands 1

Within the knowledge-based economy several institutions are involved in product innovation processes. Literature study has shown that the most researched and cited are the industry-universitygovernment relations, presented in the Triple Helix model of institutional relations within new product development (NPD). Based on a case study of the Academic Virtual Enterprise, we have put the sole input of these institutions in NPD into question. We have tested and supported the claim that the user and the society are equal partners in the product innovation process. We have put forward the Fourfold Helix model that features a new formation of institutional relations where special focus is placed on the involvement of the user and the society in NPD. © 2010 Journal of Mechanical Engineering. All rights reserved.

Keywords: triple helix, user, society, market pull, technology push, fourfold helix, European Global Product Realization course 1 INTRODUCTION Strong competition, the market of customers and increased complexity of products and processes are the characteristics of today's competition. Fast product and process development, combined with timely participation of customers and suppliers together with entering the market at the right time, seem to be the decisive criteria for the market success of a product [1]. In today's competitive environment, every company wants to achieve shorter productdevelopment times, lower costs, higher quality of the product, and, consequently, the satisfaction of its customers. In order to achieve the set goals, the company has to take into account the customers wants and needs during the newproduct development process [2]. Due to these trends and everchanging business environment new product development (NPD) has been changed drastically during the past decades.

As NPD used to be in the domain of the industry, in today’s global economy several other institutions (such as the university and the government) have become new partners of the innovation processes in NPD. One of the reasons for this shift is the emerging significance of knowledge. Nowadays industries do not depend solely on their production capital, but also on their intellectual capital. What matters is not so much the development of technical innovations.

Organizational devices are created to tie these

innovations to social and economic purposes [3].

Since the university is an institution providing intellectual capital, its involvement in NPD is increasing. With these increased relations between the university and industry in NPD, increased importance has also been put on science and technology policies. This means that university and industry relations are also shaped by the government, which is responsible for such policies.

Within knowledge-based economy the interactions between university, industry and government have become more complex and have been acknowledged and researched by several authors. One of the most extensive researches has been done by Leydesdorff and Etzkowitz [4, 5, 6]. They have proposed a model that describes the changing relations among institutions within the context of NPD. The so-called Triple Helix model presumes the involvement and flux of boundaries of several institutions participating in NPD. The relations of the institutions have been represented as a spiral process which emerges from reciprocal university-government-industry relations [4].

We have studied this model for its usability in the context of design education. We have found that it facilitates the setting up of useful relationships with regard to NPD. However, we have also found that it misses an important aspect of NPD, namely that NPD is usually done for artifacts that will be used by human beings and therefore, the processes and

* Corr. Author's Address: Faculty of Mechanical Engineering, University of Ljubljana, Aškerčeva 6, 1000 Ljubljana, Slovenia, nusa@lecad.uni-lj.si

521

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 521-530

interactions within NPD need to be considered by the user. In industrial design engineering this has been manifested in the concept of Human Centered Design (HCD).

In this paper we will first elaborate on the Triple Helix relations as proposed by Leydesdorff and Etzkowitz [4, 5, 6]. We will show that this model does not sufficiently recognize (i) the role of the user and (ii) several important aspects of society. We will then present the research which explores the proposed reorganized model in the field of industrial design engineering education.

1.1 Triple Helix Relations in a KnowledgeBased Economy The Triple Helix as a model of universityindustry-government relations is proposed to be a key component of any future national or international innovation strategy [7]. It postulates that the interaction between university, industry and government is the key to improving the conditions for innovation in a knowledge-based society [8]. It is a spiral model of innovation that elaborates the reciprocal relationships of different institutions that are active in the innovation process. The three institutions are presented as three interacting helices that jointly perform NPD processes (Fig. 1). The spiraling form of the Triple Helix relations represents the interactions taking place among the three institutions in order to improve the local economy through NPD [8].

The industry acts as the centre of production, the government is the source of contractual relations that guarantee stable interactions between the three, and the university is a source of new knowledge and technology, the generative principle of knowledge-based economies [8]. The three institutional spheres (public, private and academic) are now more and more involved in a pattern of spiraling links that emerge in various steps in the process of innovation [9]. Their relations are relatively equal, but interdependent and they overlap as the institutions take the role of the other.

A Triple Helix regime (Fig. 1) typically begins when the university, the industry and the government enter reciprocal relationships, where each institution aims at enhancing the performance of the other [8]. Several important dimensions of institutional cooperation are outlined by the model.

522

Industry

University

Government

Fig. 1. University-Industry-Government relations within the Triple Helix [8] The first dimension of the Triple Helix is the internal transformation in each of the helices/institutions, such as the development of lateral ties among companies through strategic alliances or an assumption of an economic development mission by universities [10]. The second dimension is the influence of one helix upon another. The Slovenian government has, for example, in recent years introduced several schemes for promoting technological development in industry. Industrial research in pre-competitive and near-market areas is subsidized, with special bonuses for co-operations between manufacturing firms, but also between industry and science [11]. The third dimension is the creation of a new overlay of trilateral networks and organizations from the interaction among the three helices, formed for the purpose of developing new ideas and formats for new product development [10]. Centers of Excellence are examples of these networks. Mostly funded by government resources, they are corporate entities, with a view to adequately manage their intellectual property and investing in the development of promising new research avenues.

The entrepreneurial university retains the traditional academic roles of social reproduction and extension of certified knowledge, but places them in a broader context as part of its new role in promoting innovation [8]. Industry, on the other hand, supports university research by extending it beyond technical innovation.

Scientific research is becoming increasingly relevant for socio-economic development.

Fain, N. - Moes, N. - Duhovnik, J.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 521-530

Consequently, governmental policies affecting the innovation process change their strategies in promoting R&D and other entrepreneurial activities to fulfill the “society-centered” forms of governance, which presume the involvement of sectors other than state (such as markets, the society and other non-state actors) in governing the public domain [12].

Although Triple Helix interactions are usually a necessary condition for innovation, they are often not a sufficient condition [13]. As proposed above, the innovation processes are becoming more complex and the involvement of the user is both, necessary and natural in the innovation processes within the knowledge-based economy. At the same time all the participating institutions within the innovation process are part of the society that shapes them and reshapes their relationships.

Society

Industry

University USER

Government

Fig. 2. The Fourfold Helix relations within NPD process We propose a new framework (Fig. 2) of university – industry - government relations, where the user is put into the center of innovative relations among the studied institutions and where the society as a whole is the institution binding all of the actors together.

1.2 The Role of the User and Society in NPD In its broadest sense, innovation is about creating a climate or culture which promotes implementation of productive change in order to improve the wealth creating capacity of society

[14]. In this sense, it is an idea, practice, or object that is perceived as new by an individual or some other unit of adoption [15]. If we follow this definition and presume that innovation is about social and cultural change, then it follows that change, and the circumstances leading to the adoption and ultimate success of those changes will take place within unique systems and cultures – the society.

As stated in [16], the adoption of innovation is a decision of the user to make full use of a new idea as the best course of action.

This, consequently means that the society, and more specifically, the users of products, will have the highest effect on the success of innovations.

Their adoption rate – the speed with which an innovation is adopted by the members of a social system [16] – will dictate (1) the diffusion of an innovation into the society and (2) its success rate.

If, on the other hand, innovation is limited only to the diffusion into society, whereby the adoption of innovation is disregarded, only the Triple Helix relations are relevant. Diffusion is the process through which an innovation is communicated through certain channels (i.e. marketing, word of mouth, etc.) over time, among the members of the social system [17]. The three institutions of the Triple Helix are the communicators in the diffusion process. They push the innovation onto the market, whereas the acceptability and the speed of adoption of the innovation are not in their domain. The market point of view, where the user is the key player, is disregarded by the Triple Helix model. The importance of the involvement of the user and the society in the NPD is presented in Fig. 3.

Empirical research in a number of fields has shown that users are frequently the first to develop and use the prototype versions of what later become commercially significant new products and processes [18]. Although most of the users that innovate are “lead users”, meaning they are at the leading edge of market trends, many of the novel products they develop for their own use will appeal to other users and could provide the basis for the products manufacturers might wish to commercialize [19]. As stated in [20], users (1) can be the source of incremental technical changes, (2) can develop unconventional design solutions or (3) can find and test new applications of a product. Since

The Role of the User and the Society in New Product Development

523

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 521-530

products are always culturally evaluated, users in the widest sense always play a role in innovation.

Their role is essential in the early development and diffusion phases of an innovation [21]. In many cases technical improvements are realized during the diffusion phase (the phase of product commercialization and adoption) by user feedback or by re-invention by the user.

User practices and the wider socioeconomic environment are strongly related.

Society may directly influence user-relevant characteristics of any technology [21].

According to [18] user innovation occurs when (1) a local community has unique needs and (2) when it is cheaper to invent anew than it is to search for and acquire a needed innovation that may exist elsewhere. To be able to use new technology successfully, many changes in user practices and the socio-economic environment may be required [13]. The social, the economic and the cultural aspects of the wider context are crucial for the innovation to be successful.

Triple Helix Market pull NPD

User Diffusion Adoption Technology push

S o c i e t y FOURFOLD HELIX

Fig. 3. The importance of the user in NPD Therefore, the user is, on one hand the source of innovation and on the other, the evaluator of new products. Being a part of the wider society, neglecting this perspective might lead to NPD failure. According to [22], industry can only be successful if it is embedded in a healthy society. When a society is considered to

524

be healthy, the following conditions are fulfilled: (1) education, health care, and equal opportunity are essential to a productive workforce; (2) safe products and working conditions attract customers; (3) efficient utilization of land, water, energy and other natural resources makes business more effective; (4) good government, the rule of law, and property rights are essential for efficiency and innovation; (5) strong regulatory standards protect both, customers and competitive companies from exploitation; and (6) a healthy society creates an expanding demand for business, as more human needs are met and aspirations grow [22]. A healthy society therefore, needs universities, industries, governments and users to properly function.

There are two main aspects that we address in this research. First, we test the Triple Helix model on a practical case of NPD to research the institutional relations that are formed during that process. Secondly, we test the assumption that the user acts as an equal institution in these relations within NPD. In this way, we intend to investigate the proposal of extending the Triple Helix model into a Fourfold Helix of institutional relations in NPD. It is imperative that the institutions involved in innovation effectively capture unmet needs, continuously validate technology assumptions with customers and combine the technical and marketing functions in order to achieve NPD success.

The research questions are as follows: 1. Does the involvement of the user and the society reshape institutional relations in product innovation processes, proposed by the Triple Helix model? If so, how are these relations reshaped by the involvement of the user and the society?

2. Is the user an equal partner in the reshaped relations of NPD? 3. Should the user and the society be added to the Triple Helix model, thus forming a Fourfold Helix?

According to the constructs identified in the literature review presented, we will analyze how the institutional boundaries are being reshaped in product innovation processes on a case of an Academic Virtual Enterprise (AVE), embodied in the European Global Product Realization (EGPR) course. We will study all four institutions that are proposed to be a part of the innovation process within the selected course

Fain, N. - Moes, N. - Duhovnik, J.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 521-530

and their relations. Our objective is to present the role of the user and the society within these institutional relations. The theoretical focus is placed on the Triple Helix model and its enhancement with the society perspective, whereby the unit of analysis is the EGPR course implemented into design education of several European Universities.

There were several criteria taken into account when choosing a design course instead of a real-life NPD project for the case study. First of all, there was a need for a case study where the Triple Helix relations were present and the NPD process involved a common consumer product.

Although the involvement of the university and the government in NPD is increasing and becoming more relevant, the majority of NPD projects is still strictly in the domain of the industry, whereby most of the innovative processes are done within business to business environment. The EGPR course presumes and supports the involvement of all the relevant stakeholders and its final result is a product intended for the consumer market. Second, the NPD processes within the EGPR course are conducted according to real-life design practice in order to enable the development of good designers. Finally, the rapid developments in design practice call for proactive educational responses. Design education should enable students to aquire the necessary competences that will allow them to face challenges yielded by new trends in current real-world design problems when they become professional designers.

Experience shows that many graduates cannot cope with such NPD practice. According to this presumption, the EGPR course was formed as an application of AVE, with the notion to put the students into the real-life NPD environment. As such, it is presumed to be suitable to investigate the institutional relations arising from the NPD practice.

2 RESEARCH METHODOLOGY To analyze the value of the Fourfold Helix framework, case study methodology was chosen.

A case study maps real-world data onto an abstract, general framework, and expresses it in terms of a detailed substantive argument, which is in effect a theory [23]. The case study is suitable to answer questions like “how” and “why”,

whenever the empirical analysis focuses on reallife context [24]. Since the present work aims at identifying how the involvement of the society and the user in innovation processes reshapes the institutional relationships within the Triple Helix model, it will be necessary to conduct an explanatory study.

To increase the validity of the conclusions, the “quasi judicial” (QJ) method of case study will be used. It relies on assessing and weighing the evidence provided on a case-study for its causal arguments. In contrast to other case methods, it uses a systematic inductive procedure to construct arguments about a specific case and tries to establish the causal connections referred to in the substantive arguments used to describe and analyze the case [23].

Therefore, Q, C

D

W

R

B LEGEND: C (claim): conclusion D (data): relevant empirical data Q (modal qualifier): subjectively determined probability that the claim is actually valid W (inference warrant): presumptions, rules and theory that support the claim (C) on the basis of data (D) B (background data): contextual information from background that supports the arguments (W) R (rejection): conditions under which the arguments would fail

Fig. 4: Representation of the QJ method [23] and [25]discussed In [25] the author helps codify the “quasijudicial” method by setting forth eight formal steps for applying this method to clinical or social science research (see Fig. 4). These are the following: 1) the initial problems and issues of the case must be clearly stated, 2) background information should be collected to provide a context in which to understand the problems and issues of the case, 3) existing explanations of the case must be evaluated to determine whether they fit the evidence and to discern what they lack, 4) a new explanation should be set forth fixing the problems identified in previous explanations, 5)

The Role of the User and the Society in New Product Development

525

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 521-530

the sources of evidence and the evidence itself used in the new explanation must be evaluated or “cross-examined,” 6) the internal coherence and logic of the new explanation including itscompatibility with the evidence should be critically examined, 7) the conclusions of the new explanation regarding the case must be presented, and 8) the implications of the new explanation for comparable cases must be the elements presented in Fig. 4 the questions are as follows [23], [25].

These steps can be achieved by posing several questions, which are used to construct or dissect an argument. With regard to  C: What are we trying to prove?  D: What evidence do we have to go on?  Q: How likely is it that our conclusions are correct?  W: What entitles us to draw these conclusions from that evidence?  B: What is the justification for our line of reasoning?  R: What assumptions are we making?

The pattern of argument revealed by the QJ method may permit findings to be generalized to a class of similar cases and will be used in this paper on the case of the EGPR course, conducted within an AVE. 3 COMBINING THEORY AND PRACTICE The concept of an academic virtual enterprise (AVE) was invented to establish a stimulating learning and working environment for students [27]. It is a project-oriented educational agreement, which is based on volatile alliance of industrial and academic partners for mutual advantages. As a result of a multi-year educational design, an international AVE was set up as a learning environment for an EGPR design course. Together with university educators and company experts, the students of several universities from different countries form the labor capacity of the AVE. In 2007 five European universities (University of Zagreb, Ecole Polytechnique Federale de Lausanne, University of Ljubljana, City University London and Delft University of Technology) and an industrial partner participated in an AVE. The industrial partner provides a problem to be solved by the international teams of students. The AVE connects theoretical knowledge and practice (by combining academic

526

and practical knowledge) to solve a real-life NPD problem. Its main characteristic is the formation of virtual teams of students that only know each other through the video-conferencing meetings.

All the communication and work in such an enterprise is done with the help of IT technologies, as the participants are located in different parts of the world.

The goal of EGPR courses is to enable students to develop abilties that are needed to solve complex real-life NPD problems, to generate product ideas and forward them to the status of a working product prototype and to manage their knowledge inquiry and skill development for their future work as professional designers [27]. Through the EGPR course the students work in multicultural, multinational and multidisciplinary teams with the objective to solve a global product development problem using the knowledge acquired during the EGPR course, the knowledge learned during other courses at their universities and information provided by the industrial partner [28].

EGPR is a one-semester course for Master of Science level students. It comprises several steps, such as market analysis, financial issues, product specifications, vision formation, concept generation, concept solution, materialization, prototyping and testing [28]. Teams are formed in such a manner that each team consists of several students from each of the participating universities. Therefore, students's profiles within a team are very different, which has the advantage of providing complementary knowledge and expertise that are needed for the development of a global product. On the other hand, it poses a problem of handling the discrepancies not only in skills and expertise but also in view points about the same subjects [28].

One of the advantages of this course is that the students can engage in more risky activities than the present industries, because of their supporting learning and developmental objectives. In doing so, they put the interactive Triple Helix in action. However, the course also actively considers the user and the wider society in the process of innovation, and in doing so it also implements the proposed fourth helix. As we argued earlier, the relations of the three helices are not enough for a successful innovation process. The user is the center of innovation, as he/she is present in every step of the NPD process

Fain, N. - Moes, N. - Duhovnik, J.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 521-530

as an innovator or evaluator of the innovation and therefore, affects all three helices in innovation.

The user is also part of the wider society, and in constant interaction and relation with the three institutions (government, university, industry).

Together they form a society that can only be healthy and progressive when all the spheres are in creative and constructive relations. The students of the EGPR course take these relations into account and put them in to use to produce innovations that are creative, socially and environmentally acceptable and provide a competitive advantage for the company (Fig. 5).

The AVE has a specific organizational framework where cooperation and the flux of boundaries between different institutional spheres are enabled to provide the best possible innovation output.

The university provides theoretical and practical knowledge on innovation strategies, policies and the competitive environment, enabling students to design a product that is suitable for the involved company.

The student teams are introduced to various facets of global product realization through selected lectures by experts from both, academic and industry sectors. Lectures balance between practical and theoretical issues in order to provide the students with efficient tools to deal with global product development projects in a structured way [28].

From the educational point of view, design is mainly characterized by the need to combine theoretical knowledge and practical skills [26]. A strategy of paralleling theory and practice

D1: NPD problem is defined by a company D2: NPD is performed by students of several European universities D3: university curriculum is funded and determined by governments D4: product is developed for a specific user group within a market

(students should learn about design and how to design), as proposed by the Triple Helix relations, should therefore, be adopted in teaching design.

The EGPR course does that with (1) parallelling academic lectures and strong theoretical knowledge with product development in EGPR project, and (2) involving the application of intensive practical skills.

The EGPR course is also regarded as an opportunity for a closer cooperation between the university and industry [29]. The relation is seen primarily in providing students with a real life problem that they can solve with designing an appropriate product. Consequently, the solution provides an opportunity for the participating industry, as all the activities necessary in NPD (from market analysis to prototyping) are executed by the participating students. In a constant search for new market opportunities and developmental potentials companies certainly support the EGPR course [29]. Both, short term benefits and long term advantages for the future are expected. The industrial partners present a real-life problem for the products to be developed and provide the information and data about the existing models in the comparable families of products. Their practice of product development allows the students to deepen their understanding of problem analysis, product development processes, to improve their professional skills and also to gain experience in multi cultural, multi national and multi disciplinary cooperation. The students are thebridge between the academic knowledge and industrial application [28].

Therefore, Q1 presumably, C1: there are fourfold relations within NPD in EGPR C2: the user plays a unique role within the NPD W1: cooperation of all process partners with the user W2: HCD W3: global trends in design

B1: literature/papers/publications B2: experience B3: surveys/research

R1: the design education project is a representation of NPD practice R2: equal involvement of all partners as in NPD practice

Fig. 5. The Fourfold Helix relations within EGPR

The Role of the User and the Society in New Product Development

527

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 521-530

In regard to the EGPR course, the involvement of the government is subtle and indirect, but it is present. The government is responsible for providing the rules of the game and also for making new venture capital available to help start new enterprises [8]. As EGPR is a part of a broader curriculum of the participating universities that is supported by governments of all participating countries, the involvement of the governmental institution is not questionable.

Fig. 6. An example of the final prototype in EGPR 2007 The role of the society within the EGPR course is, however, more obvious. The fourth helix is an important part of the course of EGPR from the very beginning. The user perspective is introduced from the beginning of student work with market research being done on regional, national and global scale. The design course is in this way not only technology-oriented, but also market-oriented. In 2007, for example, the project task was to develop a technologically and technically advanced Point of Purchase display, which would be used for displaying male grooming products produced by a brand at the premium range (Fig. 6). The display should have emphasized the brands’ products and have them presented in such a way that the client understood and felt the brand advertised. In order to be ahead of their competition, the company wanted to offer 528

its customer a display that managed to draw attention of male buyers by incorporating cutting edge technology and creating an interactive male grooming display. This should have helped the company’s customers to increase the sales of their products.

Students' work was structured into several consecutive phases, whereby the user perspective was mostly investigated in the first phases – the so called fuzzy front end of the NPD process.

During this phase the main concern of the teams was to get an insight into what the market needs regarding the specified product were, what the competition was like and what the trends within the given market were.

With such an analysis, the requirements, important for the next stages of product development, were found. As the assignment process had already been specified by the company involved, further insight into the market needs and company competition was done by a survey including a questionnaire and a focus group. Through the analysis of market trends and with the guidelines discovered from the survey, a list of requirements was put out as an introduction into the idea generation phase. The final result within the fuzzy front end phase were several ideas for the design of male grooming displays.

They were further evaluated and developed in the next phases of the project. The main mission of the first stages of student work was therefore, getting to know the market and specifying the requirements that the male grooming display should satisfy to be successful within its target group.

The students did research in all five participating countries, so that the results could be applied on a more global general scale. They did extensive research to explore the market needs and the solutions already available on the market.

With the user perspective analyzed they were able to see the user expectations as well as which other factors had to be involved in NPD to obtain a result that was acceptable for the end user. In other words, further socio-economic and sociopolitical aspects of developing new products were well investigated during the course. For example, by testing the acceptability of proposed concepts by various standards and by taking into account the environmental aspects of the society, several indicators of how to keep a healthy society were taken into account.

Fain, N. - Moes, N. - Duhovnik, J.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 521-530

4 DISCUSSIONS AND CONCLUSION In our study we have tested the validity of the Triple Helix model on an Academic Virtual Enterprise. With the QJ method we have found evidence to support our claim, that only university-industry-government relations within NPD processes are not enough for a successful product innovation process. Our proposed Fourfold Helix model that also involves the user and the role of the society in the NPD processes has been supported by empirical evidence. It has shown that technology-push and market-pull strategies need to be a part of NPD processes in order to make them successful. However, the presented case was a study of NPD within an educational area. Moreover, it is a single case study; therefore further research is needed to provide further support for our model.

The main differences between educational and real-life NPD practice that need to be taken into account in regard to our study are mostly related to the NPD processes. Professors at universities tend to teach the ideal case NPD process, meaning that the students go through all the phases of NPD process and consider all the institutions and actors involved in such processes.

In practice, however, the environment of the firm and the NPD project needs to be taken into account and several risks need to be considered as well. The NPD process in real environments can go faster and must be cost efficient in order to bring success. Students can make mistakes in their processes that are reflected only in their marks. Furthermore, students can also take more risks and try satisfying user needs that are more complex and less feasible than in real NPD environments.

However, our study has efficiently proven that there are several more institutions than just the Triple Helix ones that significantly contribute to product innovation processes and consequently, provide successful NPD. Therefore, although in need of further confirmation, our Fourfold Helix model can be used as guidance in evolving institutional relations in NPD. 5 REFERENCES [1] Žargi U., Kušar J., Berlec T. and Starbek M. A Company's readyness for concurrent

product and process development. Strojniški vestnik-journal of Mechanical Engineering, 2009, Vol. 55, No. 7-8, p. 427-437.

[2] Kušar J., Duhovnik J., Tomaževič R. and Starbek M. Finding and Evaluating Customers' Needs in the ProductDevelopment Process. Strojniški vestnikjournal of Mechanical Engineering, 2007, Vol. 53, No. 2, p. 78-104.

[3] Yapp, C. (2000). The Knowledge Society: the challenge of transition. Business Information Review, vol. 17, no.2, p. 59-65.

[4] Leydesdorff, L., Etzkowitz, H. (1996).

Emergence of Triple Helix as a model for innovation studies. Science and Public Policy, vol. 25, p. 195-203 [5] Etzkowitz, H., Leydesdorff, L. (1998). The endless transition: A Triple Helix of university-industry-government relations.

Minerva, , vol. 36, no. 3, p. 203-208.

[6] Leydesdorff, L., Etzkowitz, H. (2001). The transformation of university – industry government relations. Electronic Journal of Sociology. Retrieved on 22.11.2007, from: http://www.sociology.org/content/vol005.00 4/th.html [7] Etzkowitz, H., Leydesdorff, L. (2001).

Universities and the global knowledge economy: A Triple Helix of University – industry - government relations. Continuum, London,. ISBN 0-8264-5673-1.

[8] Etzkowitz, H. (2003). Innovation in innovation: The Triple Helix of university – industry - government relations. Social Science Information, vol. 42, no. 3, p. 293337.

[9] Marques, J. P. C., Caraca, J. M. G, Diz, H. (2006). How can university – industry government interactions change the innovation scenario in Portugal? – The case of the University of Coimbra. Technovation, vol. 26, no. 4, p. 534-542.

[10] Etzkowitz, H. (2002). The triple helix of university – industry - government implications for policy and evaluation. working paper, Retrieved on 5.11.2007, from: http://www.sister.nu/pdf/ wp_11.pdf [11] Koschatzky, K., Bross, U., Stanovnik, P. (2001). Development and innovation potential in the Slovene manufacturing industry. Technovation, vol. 21, no. 5, p.

311-324.

The Role of the User and the Society in New Product Development

529

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 521-530

[12] Mok, K. H. (2005).

Fostering entrepreneurship: Changing role of government and education governance in Hong Kong. Research Policy, vol. 34, no. 4, p. 537-554.

[13] Bunders, J. F. G., Broerse, J. E. W., Zweekhorst, M. B. M. (1999). The triple helix enriched with the user perspective: A view from Bangladesh. Journal of Technology Transfer, vol. 24, p. 235-246.

[14] Duggan, R. (1996). Promoting innovation in industry, government and higher education.

Long range planning, vol. 29, no. 4, p. 503513.

[15] Rogers, E. M. (2002). Diffusion of preventive innovations. Addictive behaviors, vol. 27, no. 6, p. 989-993.

[16] Rogers, E. M., Shoemaker, F. F. (1972).

Communication of innovation: a crosscultural approach. 2nd ed. Free Press, New York.

[17] Rogers, E. M. (1995). Diffusion of innovations. 4th ed. Free Press, New York.

[18] Morrison, P. D., Roberts, J. H., von Hippel, E. (2000). Determinants of user innovation and innovation sharing in a local market.

Management Science, vol. 46, no. 12, p.

1513-1527.

[19] von Hippel, E. (2005). Democratizing innovation. The MIT Press, Cambridge.

[20] Ornetzeder, M., Rohracher, H. (2006). Userled innovations and participation processes: lessons from sustainable energy technologies. Energy Policy, vol. 34, no. 2, p. 138-150.

[21] Truffer, B. (2003). User-led innovation processes: the development of professional

530

[22]

[23]

[24] [25] [26]

[27]

[28]

[29]

car sharing by environmentally concerned citizens. Innovation, vol. 16, no. 2, p. 139154.

Porter, M. E., Kramer, M. R. (2006).

Strategy and society: The link between competitive advantage and corporate social responsibility. Harvard Business Review, vol. 84, no. 12, p. 78-92.

Bromley, D. B. (2000). Comparing corporate reputations: leogne tables, quotients, benchmarks or case studies?.

Corporate Reputation Review, vol. 5, no. 1, p. 35-50.

Yin, K. I. (2003). Case study research: design and methods, 3rd ed. Sage Publications, Inc. Thousand Oaks.

Bromley, D. B. (1986). The case study method in psychology and related disciplines, John Wiley & Sons, Chichester.

Kline, M., Berginc, D. (2004). Transfer of image of a turistical brand of state to its other brands. Theory and practice, vol. 41, no. 5-6, p. 962-978 (in Slovenian).

Horvath, I. (2006). Design competence development in an academic virtual enterprise. Proceedings of IDETC/CIE 2006, September 10-13, Philadelphia.

Bufardi, A., Xirouchakis, P., Duhovnik, J., Horvath, I. (2005). Collaborative design aspects in the European global product realization. International Journal of Engineering Education, vol. 21, no. 5, p.

950-963. Žavbi, R., Tavčar, J. (2005). Preparing undergraduate students for work in virtual product development teams. Computers & Education, vol. 44, no.4, p. 357-376.

Fain, N. - Moes, N. - Duhovnik, J.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 531-539 UDC 628.8:697.353:069

Paper received: 01.12.2009 Paper accepted: 02.03.2010

Alternative to the Conventional Heating and Cooling Systems in Public Buildings Mitja Košir* - Aleš Krainer - Mateja Dovjak - Rudolf Perdan - Živa Kristl University of Ljubljana, Faculty of Civil and Geodetic Engineering, Slovenia The paper presents an alternative system for heating and cooling in public buildings. The system was designed for the retrofitted building of the Slovene Ethnographic Museum (SEM) where it was also extensively tested. The installed system includes radiant wall mounted panels for heating and cooling, localized automated tangential fans for cooling and ventilation and a centralized building management system for the regulation and supervision of the performance. The efficiency of the system was thoroughly investigated through a series of experiments conducted prior to the renovation of the building as well as after the museum was put into service. The application of the described system resulted in substantial reduction of energy consumption, better internal thermal conditions and lower investment costs for the Heating, Ventilation and Ait Conditioning (HVAC) system of the entire building. © 2010 Journal of Mechanical Engineering. All rights reserved.

Keywords: heating, cooling, ventilation, low temperature system, radiant panels 0 INTRODUCTION The paper presents a system for indoor temperature regulation with the use of lowtemperature radiant heating/cooling panels and automated natural ventilation.

Extensive experimentation with the low-temperature wall mounted heating/cooling panels was conducted before and during the retrofitting of the Slovene Ethnographic Museum (SEM) in Ljubljana, Slovenia [1]. The in-situ experimental results, simulations and later measurements of the building performance in real time conditions proved high efficiency of the system. The wall mounted low-temperature radiant heating and cooling panels present an alternative to the air heating and air cooling systems originally proposed for the museum building. Because the majority of heat transport in the heated or cooled spaces equipped with low-temperature systems is conducted by radiation and not by convection, a smoother temperature profile preferred by the majority of users is achieved [2]. Lowtemperature heating systems that operate close to the environmental temperatures are in addition to low energy also low exergy systems [3], although the use of high exergy fuels (e.g. electricity or fossil fuels) where low exergy work is needed somewhat reduces this effect. The previously proposed mechanical centralized ventilation was replaced by a localized automated ventilation

system utilizing small tangential fans integrated into the window sills. The system enabled the necessary physiological ventilation during museum opening hours as well as cooling via the night ventilation.

Fig. 1. North-east external view of the retrofitted museum building The geometry of the museum building is presented in Fig. 2. Exhibition spaces equipped with panels are located in the east wing of the ground floor and on the 1st, 2nd and 3rd floors of the building; the total floor area is 2575 m2, while the floor area of the entire building is 5214 m2.

The existing exterior walls were composed of external rendering applied to a 50 cm thick brick wall with internal rendering removed. Floors

*

Corr. Author's Address: University of Ljubljana, Faculty of Civil and Geodetic Engineering, Chair for Buildings and Constructional Complexes, Jamova cesta 2, Ljubljana, Slovenia, MKosir@fgg.uni-lj.si

531

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 531-539

were mostly brick vaulted, gravel filled and finished with wood decking.

1 CONCEPT OF HEATING/COOLING AND VENTILATION SYSTEM The main objectives of the SEM – Museums project were to assure optimal conditions for exhibitions and for the storage of museum’s exhibits in accordance with international standards, to assure appropriate environment for visitors from the visual and from the thermal point of view and rational use of energy without reducing the quality of functional use. When studying various options for achieving these goals, the decision was made that the problems had to be dealt with holistically (and not every aspect of internal conditions solved with a separate system) and that the rational use of energy for heating and cooling was to be achieved by reduction of operating costs controlled by a building management system (BMS).

There were seven main spheres of activities resulting in the framework of the following interventions: thermal energy with heating and cooling part, ventilation, daylight (not presented in this paper), control and management, constructional complexes, simulations, testing and measurements.

1.1 Heating and Cooling System The design and performance of the wall mounted low-temperature heating/cooling system progressed through four distinct experimental phases. The conducted measurements spanned over four years, from those executed prior to the building renovation to those carried out during the first year of building operation [4, 5]. The first phase of measurements (conducted during the summer and autumn months of 2000) encompassed the recording of values for the visual and thermal environment, while the building was un-refurbished and in “free run”.

The measurements showed that during summer season internal temperatures were maintained mostly within the 18°C – 25°C zone. Thermal mass of the building compensated for night lows (T 10K) and day highs (T 4K). During this time the possibility of installing a lowtemperature heating and cooling system was discussed. The decision was made to investigate the effectiveness of such a system by conducting experiments in real environment of the building.

The experience based on preliminary measurements and simulations also put in question the necessity of an air conditioning system proposed in the original design.

During the second phase of experiments (conducted from the end of autumn 2000 till

Fig. 2. Floor plan of the renovated SEM building with annex marked in dark grey; the position of test rooms used in the preliminary experiments is also marked

532

Košir, M. – Krainer, A. – Dovjak, M. – Perdan, R. – Kristl, Ž.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 531-539

March 2001) the response of test rooms to various heating modes was tested. Two experimental test rooms were completely renovated in the SW corner in the 1st and 2nd floors (the position of the rooms relative to the retrofitted state of the building is shown in Fig. 2). Both rooms were additionally thermally insulated and outfitted with new double glazed windows with low-e coatings as well as equipped with electrically powered heating units (conventional electrical radiators), as shown in Fig. 3. Measurements and computer simulations showed that annually the system would use 10 kWh/m2 of heating energy less than the proposed air system [1, 4, 5], if an alternative radiant system was to be used. In time for the beginning of the third phase of experiments (conducted from the end of March 2001 till July 2002) the prototypes of wall mounted heating/cooling panels were constructed and installed in the renovated test room on the 1st floor (dubbed the model room), while the second room on the 2nd floor (dubbed the reference room) remained the same as in the second phase (Fig. 4). Thorough measurements of wall panels were executed under real time environmental conditions. Example of results acquired during winter testing is shown in Fig. 5. For the model room the average energy consumption for heating was 12% lower than in the reference room.

Difference in energy consumption between the two test rooms is the result of different temperatures of heating media and modes of heat transport. The cooling energy consumption was measured during two summer seasons, from August 2001 till June 2002. Different set point temperature series were tested: 24 h/day continuous cooling and two modes of intermittent cooling, from 08-20 h with 22.5oC and 25oC set point temperatures, respectively. Collected results were derived into seasonal specific cooling energy consumption between 10 kWh/m2 (set point 25°C) and 15 kWh/m2 (set point 22.5°C) for intermittent cooling and 25 kWh/m2 (set point 25°C) to 30 kWh/m2, (set point 22.5°C) for continuous cooling. Such low set point temperature was defined in order to test the cooling performance of the system and any possible occurrence of surface condensation. The acquired data showed important differences in energy consumption between the reference room and the model room for heating and small specific energy consumption for cooling.

Fig. 3. Reference room on the 2nd floor of the SEM building heated with conventional radiator system

Fig. 4. Model room on the 1st floor equipped with the prototype heating/cooling wall mounted panels In Fig. 5 daily temperature profiles for the model room during winter period (December) in intermittent heating mode are presented. The outside air temperatures were very low, between -15oC and -3oC which is low even for Ljubljana.

In the diagram also the temperature profile for the reference room exhibiting a “wave” pattern is presented. In the model room the temperature profile is smooth and follows very well the prescribed set-point temperature profile. The experiments showed that the panels reacted well in winter, summer and mid-season conditions and consistently maintained the indoor temperatures close to the set-point temperatures without any difficulties.

1.2 Ventilation System For the ventilation of exhibition rooms new automated and localized natural ventilation system was designed. Small tangential fans were integrated into the window sills (Fig. 6) of new

Alternative System for Heating and Cooling in Public Buildings

533

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 531-539

Fig. 5. Winter time experiment conducted in the model room during the 25th of December; shown are: external air temperature (Toutside), internal air temperature in the model room (TIAverage), vertical surface temperature distribution at the bottom of the panels (measuring nodes 7/1, 7/3, 7/5, 7/7, 7/9), core temperature of the panel (5/3), inlet (3/1) and outlet (3/2) water temperatures; as a reference the internal air temperature in the reference room (TIIAverage) and the set-point temperature (Tset-point) are shown windows and were controlled with the BMS. The integration of the ventilation system with the window was necessary due to the minimal impact on the building exterior prescribed by the strict conservation standards.

The window integrated ventilation system is used for the necessary physiological ventilation during opening hours and for cooling purposes (night cross ventilation), when the conditions are favourable. The cooling by ventilation system is harmonised with the functioning of the wall cooling system. Generally the wall cooling

534

system is activated if the outside conditions do not enable cooling of the building with ventilation. The ventilation system also uses the microclimatic conditions surrounding the building in a way that optimizes the use of fresh air on internal comfort conditions. This means that during heating season the air is supplied from the south façade and expelled on the north side of the building. The situation is reversed during cooling season, as the supplied air is taken from the north side and expelled on the south side of the museum (Fig. 7.).

Košir, M. – Krainer, A. – Dovjak, M. – Perdan, R. – Kristl, Ž.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 531-539

2 FUNCTIONING OF SYSTEM IN THE BUILDING 2.1 Heating & Cooling In the starting phase of the project no requirements for thermal insulation of buildings under historical monument preservation protection were foreseen. Firstly, the decision was made to place a 10 cm thermal insulation with the corresponding vapour barrier and plaster boards on the inner side of the existing outer brick wall.

On one hand, this reduced the vast thermal mass of the building and on the other hand it reduced the U value of the outer wall from 1.16 W/m2K to 0.30 W/m2K. With this intervention several benefits were gained. First, quick thermal response of the building was achieved, which enables effective intermittent heating and cooling.

At the same time the outer side of the protected facade was not touched and it could retain its original structure an appearance. Second, lowtemperature wall mounted heating/cooling vertical system was used. Third, non-manageable thermal mass of the original wall was excluded from the wall’s thermal conduction transport system, but at the same time it was replaced by a designed thermal mass in the reinforced concrete wall panels separated from the other parts of the outer wall structure. Forth, thermal comfort was improved due to the surface temperature to air temperature relation and lateral radiation effect.

Fifth, consideration of the new design of combined heating/cooling wall panel system resulted in the decision to omit the designed air conditioning system. This sets free 158 m2 of space for depository area, and the investment was reduced by about € 100.000 – 150.000.

2.2 Ventilation Each floor is divided into two zones (east and west zones). A set of tangential fans is integrated into the sills of windows on the north and south side of the building. In zone 1 (ground floor, SE part of the building) there is a CO2 sensor, triggering physiological ventilation during working hours when critical levels (700 ppm) are reached. The functioning of the ventilation in other parts of the museum is operated according to time dependent ventilation protocols that were derived from the measurements of CO2 concentration in the SE part of the ground floor.

Fig. 6. Scheme of the tangential fan integrated into the window sill

Fig. 7. Functioning of the localized ventilation system according to the seasonal microclimatic conditions (W – winter operation, S – summer operation, continuous line – supplied air, dotted line – expelled air) The fans of the ventilation system are also used for cross ventilation and cooling of exhibition rooms in the case of convenient outside temperature and relative humidity conditions. Cooling with ventilation is enabled when the external air temperature is 1 K lower than the internal set-point air temperature. If the cooling by ventilation is not sufficient (the internal air temperature is 1 K higher than the internal set-point air temperature), the system switches to the wall mounted cooling panels. In this case the fans are activated according to the physiological ventilation protocols.

Alternative System for Heating and Cooling in Public Buildings

535

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 531-539

2.3 Control and Management Computer simulations of the building’s energy consumption in the pre-retrofitted state showed the annual value of 156 kWh/m2 of energy consumption for heating and cooling. This number was used as reference for the evaluation of energy performance of the building after the proposed interventions had been carried out. It was evident that the energy efficiency of the building could be improved with the application of the proposed interventions to the building envelope (installation of thermal insulation) and by using low-temperature heating and cooling system. If these interventions were considered in the TRNSYS [6] simulations, the reduction of heating and cooling energy consumption of 46.5% could be achieved in comparison to the reference state (Table 1). Simulation results predicted an annual energy consumption of 73 kWh/m2 for heating and 10.5 kWh/m2 for cooling, totalling at combined consumption 83.5 kWh/m2 annually. The simulated energy consumption for the ventilation was predicted at 2.37 kWh/m2 annually.

For the purposes of BMS the exhibition area with the total floor surface of 2884 m2 is divided into seven zones: the East wing of ground floor and in the 1st, 2nd and 3rd floors (East and West zone in each floor), which are separately controlled by BMS. The scheme of the central control system and an example of opening BMS screen for heating-cooling panel system are presented in Fig. 8 and Fig. 9, respectively.

The heating system of the building is connected to the city district heating system. It is divided into 7 control zones (East part and West part of the building in each floor).

Temperature/time/season sensitive BMS system is used and enables establishing of different setpoint temperature profiles during opening and non-opening hours. During heating season the inlet water temperature was 35°C with occasional peaks reaching 40°C at times of extreme loads.

For the cooling the wall mounted panels are connected to a common cooling plant (McQuay AGF-XN 070.2, cooling power: 218 kW, electric 88 kW, 2 compressors, 4 steps of 25, 50, 75, 100%). The temperature of cooled inlet water was typically kept around 15°C with occasional lows of 11°C. The same division of spaces as for heating is used for the control of cooling.

Temperature/time/season sensitive control wall

536

mounted system is supplemented and harmonized with the ventilation cooling system. The activation of cooling panels is enabled when the external conditions do not allow cooling of the building via ventilation. Both systems are linked and harmonized. The possibility of manual override is foreseen for all zones. In addition to cooling purposes the localized automated ventilation system is also used for physiological ventilation of the museum. For physiological requirements daily/weekly regime of performance is executed. During visiting hours the air exchange level is set to 0.5 h-1. This means that all fans in zones 2 to 7 (1st, 2nd and 3rd floor, East and West wings) are switched on every 15 minutes for the duration of 15 minutes. When the museum is closed, the ventilation is switched off.

The East wing on the ground floor (zone 1) is ventilated according to the levels of CO2 concentration.

Fig.8. Scheme of the BMS.

Fig.9. BMS screen with controls of the heatingcooling panels for the East wing of the ground floor. (angleška besedila v sliki)

Košir, M. – Krainer, A. – Dovjak, M. – Perdan, R. – Kristl, Ž.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 531-539

Table 1. Results of simulations and measurements conducted during 2004 for the annual energy consumption of the exhibition part of the SEM; the pre-retrofitted reference state (simulated) energy consumption for heating and cooling was 156 kWh/m2 annually MEASURED RESULTS (2004)

Qc (cooling)

Qv (ventilation)

Qh + Qc (combined)

Qh (heating)

Qc (cooling)

Qv (ventilation)

Qh + Qc (combined)

0.0 0.0 0.0 0.0 0.0 2.1 4.2 4.2 0.0 0.0 0.0 0.0 10.5

0.18 0.18 0.18 0.18 0.18 0.25 0.25 0.25 0.18 0.18 0.18 0.18 2.37

17.0 12.0 9.0 3.3 1.0 2.1 4.2 4.2 0.0 6.0 9.0 16.0 83.5

21.0 12.7 5.6 1.3 1.0 0.0 0.0 0.0 0.0 1.6 3.8 4.4 50.4

0.0 0.0 0.0 0.0 0.0 0.0 5.6 5.6 0.0 0.0 0.0 0.0 11.2

0.06 0.06 0.06 0.06 0.06 0.06 0.19 0.19 0.06 0.06 0.06 0.06 0.98

21.0 12.7 5.6 1.3 1.0 0.0 5.6 5.6 0.0 1.6 3.8 4.4 61.6

[kWh/m2]

17.0 12.0 9.0 3.3 1.0 0.0 0.0 0.0 0.0 6.0 9.0 16.0 73.0

Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Annual

[%]

Qh (heating)

SIMULATED RESULTS

Annual reduction compared to the reference state

The BMS permanently controlled and collected the following quantities: • Microclimate: ambient air temperature, ambient air humidity.

• Energy Systems: heating consumption (district heating, each zone separately and total consumption), cooling consumption (electricity), lighting consumption (electricity), total electricity consumption.

• Indoor Comfort: indoor air temperature, indoor air humidity, lighting levels.

Measurements of the whole building performance that were performed during the whole year 2004 were collected using installed BMS (Johnson Controls Metasys with FX15 Controllers) and the following sensors/meters: • Air temperature: JC Series A99 sensors.

• Air humidity: JC Series HT-9000 sensors.

• Heating/cooling energy: Allmess, type CF Echo.

• Electricity: Iskra instrumenti, d.d., type WS1202.

• CO2: Siemens, type QPA63.1.

Protocols for on-line monitoring of the control system during the operation of the building were prepared for different day-night, summer-winter regimes, for different sources: heating-cooling panels for heating and cooling

-46.5

-60.5

function, physiological ventilation, for the combination of cooling and relative humidity with corresponding descriptions of interventions in tabular form. The following information is available and stored by the BMS during the operation of the building: • Review of conditions on PC: outside air temperature and relative humidity, temperature and relative humidity in zones, temperature of heating medium by zones, energy use (heating, cooling), daily/seasonally by zones, electrical energy use daily/seasonally by zones, condition of panels and ventilators, both for the part of the building treated in the framework of the MUSEUMS project and for the other part of the SEM exhibition building.

• Possibilities of data storage on PC: outside temperature and relative humidity (hourly average), temperature and relative humidity by zones (hourly average), temperature of heating medium by zones (hourly average), energy use (heating, cooling), full hourly data, electrical energy use, full hourly data, condition of panels and ventilators.

There were two different energy use patterns during the measurements conducted in 2004, the first in the beginning and the second at the end of the year. In the first part energy

Alternative System for Heating and Cooling in Public Buildings

537

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 531-539

consumption was higher by 8.6% in the part of the museum with installed wall panels than in the rest of the building. During this period the controller maintained constant heating media temperature for 24 hours per day for the whole building. In the second part of the heating season the wall panels functionality was optimized by the BMS which resulted in 5 times smaller consumption in December 2004 compared to January 2004 (Fig. 10).

3 CONCLUSION Low-temperature radiative heating and cooling systems represent an efficient solution for creating good thermal environment. Lowtemperature systems enable the transport of heat through radiation and this eliminates the problems of user discomfort due to annoying air movement [2]. In addition to better thermal comfort of users such systems also exhibit improved energy efficiency due to utilization of lower temperatures of heating and/or cooling medium, which results in direct energy savings due to better boiler efficiency and lower thermal losses of the entire system [7]. Nonetheless, systems that utilize low temperatures of heating and cooling medium have special configuration compared to conventional systems. The most obvious difference is that large surfaces have to be heated or cooled for efficient functioning.

In the case of the SEM optimal relation between air temperature and surface temperature in the museum building is achieved with the use of heating/cooling wall mounted panels. They represent the main intervention in the framework of the construction. The system is connected to the district heating system for heating purposes in winter and to cooling plant for cooling purposes in summer. Window integrated BMS controlled ventilation system (small tangential fans) is used for the necessary ventilation during opening hours and for cooling purposes (night ventilation), harmonised with the wall cooling system.

Due to “high risk” nature of the proposed innovative heating/cooling system a series of experiments were conducted in various phases of its development. On the basis of these findings the decision for using the proposed system in the exhibition area of the building was adopted. After the execution of the planned interventions and the installation of the heating/cooling and ventilation system, the building was put into operation. The performance of the building was closely monitored during the whole first year (2004) of the museums operation. The actual measured energy consumption of the heating system was even lower than had been indicated by the computer simulations, as the exhibition area that encompasses approximately one half of the museum floor spaces consumed only 14% of the total heating energy consumption of the whole building.

Fig. 10. Heating energy consumption of the low-temperature wall panels installed in the exhibition space compared to the overall consumption of the whole SEM building

538

Košir, M. – Krainer, A. – Dovjak, M. – Perdan, R. – Kristl, Ž.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 531-539

The most important, difficult and problematic part of the project was design and tuning of harmonised control of temperatures, relative humidity, CO2 and cooling oriented ventilation with the application of central control system designed specially for this project. In the end the gross energy demand for heating and cooling is reduced by 60.5% (Table 1), from 156 kWh/m2 annually (simulated pre-renovation state with presumed continuous heating) to 61.6 kWh/m2 annually (measured - combined energy consumption for heating and cooling). The average measured consumption in similar buildings is usually more than 140 kWh/m2 annually (based on simulated cases). The selection of a localized automated ventilation system integrated under the window resulted in a negligible quantity of only 0.98 kWh/m2 of energy used for ventilation purposes per year. The quantity of the foreseen blown-in air was reduced from 36500 m3/h (predicted in the original project) to 10000 m3/h (implemented system).

The power of heating station was reduced by 110 kW and for cooling by 62 kW. The result of the project is also the reduction of the investment budget in the field of HVAC from € 530,319 to € 434,075 [1, 4, 5]. As a result of the installation of the wall mounted heating/cooling system and window integrated ventilation system 158 m2 of floor space were liberated for critical deposit area. The introduction of heating-cooling wall panels resulted besides in considerable energy reduction also in better indoor comfort because of large vertical heating and cooling areas with optimal surface temperatures.

The obtained results pointed to the importance of proper system regulation and automatic control [8], as realised energy savings would not be possible without sufficient control provided by the BMS. In the end the use of wallmounted heating/cooling system in a renovated building also showed that successful use of lowtemperature systems can be achieved in retrofitting projects if they are well coordinated throughout the project activities.

4 ACKNOWLEDGEMENT This project has been supported by the European Commission 5th Framework Programme MUSEUMS Energy Efficiency and Sustainability in Retrofitted and New Museum Buildings, NNE5/1999/20, Ministry of Education, Science and Sport, Republic of Slovenia, Research Programme Renewable Energy Sources, P0-504-0792-02 and Chair for Buildings and Constructional Complexes, Faculty of Civil and Geodetic Engineering, University of Ljubljana, Slovenia.

5 REFERENCES [1] Krainer, A., Rudi, P. (2005). Slovene Etnographic Museum. Slovene Etnographic Museum, Ljubljana.

[2] Imanari, T., Omori, T., Bogaki K. (1999).

Thermal comfort and energy consumption of the radiant ceiling panel system: Comparison with the conventional all-air system. Energy and Buildings, vol. 30, no. 2, p. 167-175.

[3] Asada, H. (2004). Exergy analysis of low temperature radiant heating system. Building Service Engineering Research and Technology, vol. 25, no. 3, p. 197-209.

[4] Krainer, A., Perdan, R., Krainer, G. (2007).

Retrofiting of the Slovene Ethnographic Museum. Bauphysik, vol. 29, no. 5, p. 350365.

[5] Krainer, A., Perdan, R., Krainer, G. (2006).

Slovene Ethnographic Museum: SEM, a case study. International Journal of Sustainable Energy, vol. 25, no. 3, p. 131-151.

[6] TRNSYS. A transient System Simulation Program, version 15.0. Solar Energy Laboratory, The Centre Scientifique et Technique du Batiment, Transsolar Energietechnik GmBH and Thermal Energy Systems Specialists, 2002.

[7] Paić, Z. (2002). Sustavi površinskog grijana i hlađenja. Energetika marketing, Zagreb.

[8] Košir, M. (2003). Integrated regulating system of internal environment on the basis of fuzzy logic use: doctoral thesis, Ljubljana: University of Ljubljana, Faculty of Civil and Geodetic Engineering.

Alternative System for Heating and Cooling in Public Buildings

539

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 541-542 Instructions for Authors

Instructions for Authors For full instructions see the Authors Guideline section on the journal's website: http://en.sv-jme.eu/ Send to: University of Ljubljana Faculty of Mechanical Engineering SV-JME Aškerčeva 6, 1000 Ljubljana, Slovenia Phone: 00386 1 4771 137 Fax: 00386 1 2518 567 E-mail: info@sv-jme.eu strojniski.vestnik@fs.uni-lj.si All manuscripts must be in English. Pages should be numbered sequentially. The maximum length of contributions is 10 pages. Longer contributions will only be accepted if authors provide justification in a cover letter. Short manuscripts should be less than 4 pages.

Prior or concurrent submissions to other publications should be included in the cover letter.

We also ask you to specify the typology of the manuscript – you can define your paper as an original, review or short paper. The required contact information (e-mail and mailing address) and a suggestion of at least two potential reviewers should be provided in the cover letter. Reasons for a certain reviewer to be excluded from the review process can also be provided in the cover letter.

Every manuscript submitted to the SV-JME undergoes the course of the peer-review process.

THE FORMAT OF THE MANUSCRIPT The manuscript should be written in the following format: - A Title, which adequately describes the content of the manuscript.

- An Abstract should not exceed 250 words. The Abstract should state the principal objectives and the scope of the investigation, as well as the methodology employed. It should summarize the results and state the principal conclusions.

- 6 significant key words should follow the abstract to aid indexing.

- An Introduction, which should provide a review of recent literature and sufficient background information to allow the results of the article to be understood and evaluated.

- A Theory or experimental methods used.

- An Experimental section, which should provide details of the experimental set-up and the methods used for obtaining the results.

- A Results section, which should clearly and concisely present the data using figures and tables where appropriate.

- A Discussion section, which should describe the relationships and generalizations shown by the results and discuss the significance of the results making comparisons with previously published work. (It may be appropriate to combine the Results and Discussion sections into a single section to improve the clarity).

- Conclusions, which should present one or more conclusions that have been drawn from the results and subsequent discussion and do not duplicate the Abstract.

- References, which must be cited consecutively in the text using square brackets [1] and collected together in a reference list at the end of the manuscript.

Units - standard SI symbols and abbreviations should be used. Symbols for physical quantities in the text should be written in italics (e.g. v, T, n, etc.). Symbols for units that consist of letters should be in plain text (e.g. ms-1, K, min, mm, etc.)

Abbreviations should be spelt out in full on first appearance, e.g., variable time geometry (VTG).

Meaning of symbols and units belonging to symbols should be explained in each case or quoted in a special table at the end of the manuscript before References.

Figures must be cited in a consecutive numerical order in the text and referred to in both the text and the caption as Fig. 1, Fig. 2, etc. Figures should be prepared without borders and on white grounding and should be sent separately in their original formats.

Pictures may be saved in resolution good enough for printing in any common format, e.g. BMP, GIF or JPG. However, graphs and line drawings should be prepared as vector images, e.g. CDR, AI. When labeling axes, physical quantities, e.g. t, v, m, etc. should be used whenever possible to minimize the need to label the axes in two languages. Multi-curve graphs should have individual curves marked with a symbol. The meaning of the symbol should be explained in the figure caption.

Tables should carry separate titles and must be numbered in consecutive numerical order in the text and referred to in both the text and the caption as Table 1, Table 2, etc. In addition to the physical quantity, e.g. t (in italics), units (normal text),

541

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, 541-542

should be added in square brackets. The tables should each have a heading. Tables should not duplicate data found elsewhere in the manuscript.

Acknowledgement of collaboration or preparation assistance may be included before References.

Please note the source of funding for the research.

REFERENCES A reference list must be included using the following information as a guide. Only cited text references are included. Each reference is referred to in the text by a number enclosed in a square bracket (i.e., [3] or [2] to [6] for more references). No reference to the author is necessary.

References must be numbered and ordered according to where they are first mentioned in the paper, not alphabetically. All references must be complete and accurate. Examples follow.

Journal Papers: Surname 1, Initials, Surname 2, Initials (year). Title.

Journal, volume, number, pages.

[1] Zadnik, Ž., Karakašič, M., Kljajin, M., Duhovnik, J. (2009). Function and Functionality in the Conceptual Design Process. Strojniški vestnik Journal of Mechanical Engineering, vol. 55, no.

7-8, p. 455-471.

Journal titles should not be abbreviated. Note that journal title is set in italics.

Books: Surname 1, Initials, Surname 2, Initials (year). Title.

Publisher, place of publication.

[2] Groover, M.P. (2007). Fundamentals of Modern Manufacturing. John Wiley & Sons, Hoboken.

Note that the title of the book is italicized.

Chapters in Books: Surname 1, Initials, Surname 2, Initials (year).

Chapter title. Editor(s) of book, book title. Publisher, place of publication, pages.

[3] Carbone, G., Ceccarelli, M. (2005). Legged robotic systems. Kordić, V., Lazinica, A., Merdan, M. (Editors), Cutting Edge Robotics.

Pro literatur Verlag, Mammendorf, p. 553-576.

Proceedings Papers: Surname 1, Initials, Surname 2, Initials (year). Paper title. Proceedings title, pages.

[4] Štefanić, N., Martinčević-Mikić, S., Tošanović, N. (2009). Applied Lean System in Process

542

Industry.

MOTSP 2009 Conference Proceedings, p. 422-427.

Standards: Standard-Code (year). Title. Organisation. Place.

[5] ISO/DIS 16000-6.2:2002. Indoor Air – Part 6: Determination of Volatile Organic Compounds in Indoor and Chamber Air by Active Sampling on TENAX TA Sorbent, Thermal Desorption and Gas Chromatography using MSD/FID. International Organization for Standardization. Geneva.

WWW pages: Surname, Initials or Company name. Title, from http://address, date of access.

[6] Rockwell Automation.

Arena, from http://www.arenasimulation.com, accessed on 2009-09-07.

COPYRIGHT Authors submitting a manuscript do so on the understanding that the work has not been published before, is not being considered for publication elsewhere and has been read and approved by all authors. The submission of the manuscript by the authors means that the authors automatically agree to transfer copyright to SV-JME and when the manuscript is accepted for publication.

All accepted manuscripts must be accompanied by a Copyright Transfer Agreement, which should be sent to the editor. The work should be original by the authors and not be published elsewhere in any language without the written consent of the publisher.

The proof will be sent to the author showing the final layout of the article. Proof correction must be minimal and fast. Thus it is essential that manuscripts are accurate when submitted.

Authors can track the status of their accepted articles on http://en.sv-jme.eu/.

PUBLICATION FEE For all articles authors will be asked to pay a publication fee prior to the article appearing in the journal. However, this fee only needs to be paid after the article has been accepted for publishing.

The fee is 180.00 EUR (for articles with maximum of 6 pages), 220.00 EUR (for articles with maximum of 10 pages), 20.00 EUR for each addition page. Additional costs for a color page is 90.00 EUR.

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8 Vsebina

Vsebina Strojniški vestnik - Journal of Mechanical Engineering letnik 56, (2010), številka 7-8 Ljubljana, avgust 2010 ISSN 0039-2480 Izhaja mesečno

Povzetki člankov Gregor Škorc, Simon Zapušek, Jure Čas, Riko Šafarič: Virtualni uporabniški vmesnik za oddaljeno vodenje nano-robotske celice s pomočjo haptične naprave Zoran Stefanović, Ivan Kostić: Analiza končnih priletov jadralnega letala s spreminjanjem hitrosti po kosinusnem zakonu Halil Demir, Abdulkadir Gullu, Ibrahim Ciftci, Ulvi Seker: Raziskava vplivov velikosti zrna in parametrov brušenja na površinsko hrapavost in sile pri brušenju Dragan Antić, Marko Milojković, Zoran Jovanović, Saša Nikolić: Optimalna zasnova mehkega krmiljenja enosmernega servopogona v drsnem režimu Gašper Benedik, Brane Širok, Janez Rihtaršič, Marko Hočevar: Tokovne karakteristike brezlopatičnega rotorja iz odprto celičnega poroznega materiala Slobodan Morača, Miodrag Hadžistević, Igor Drstvenšek, Nikola Radaković: Uporaba skupinske tehnologije v sestavljenih sistemih grozdov Miroslav S. Milićević: Uporaba nove formule za izračun Nakaokinega koeficienta pri visokofrekvenčnem indukcijskem varjenju Peter Fatur, Borut Likar: Statistična analiza kot podlaga za strateško odločanje glede inoviranja v slovenski strojni industriji Matija Javorski, Primož Čermelj, Miha Boltežar: Karakterizacija dinamičnega obnašanja stropnega košarkaškega koša Miodrag Zlokolica, Maja Čavić, Milan Kostić: Analitični opis vrtanja poligonalnih izvrtin – splošni pristop Nusa Fain, Niels Moes, Jože Duhovnik: Vloga uporabnika in družbe v procesu razvoja novega izdelka Mitja Košir, Aleš Krainer, Mateja Dovjak, Rudolf Perdan, Živa Kristl: Alternativni sistem ogrevanja in hlajenja javnih stavb Osebne vesti Doktorati, magisteriji, specializacija in diplome Navodila avtorjem

SI 95 SI 96 SI 97 SI 98 SI 99 SI 100 SI 101 SI 102 SI 103 SI 104 SI 105 SI 106

SI 107 SI ??

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, SI 95 UDC 621.398:004.8

Prejeto: 23.04.2009 Sprejeto: 16.04.2010

Virtualni uporabniški vmesnik za oddaljeno vodenje nano-robotske celice s pomočjo haptične naprave Gregor Škorc1,* _ Simon Zapušek2 - Jure Čas3 - Riko Šafarič2 RESISTEC UPR d.o.o. & Co. k.d., Kostanjevica na Krki, Slovenija 2 Univerza v Mariboru, Fakulteta za Elektrotehniko, Računalništvo in Informatiko, Maribor, Slovenija 3 EM. TRONIC d.o.o., Maribor, Slovenija 1

V prispevku je predstavljen razvoj virtualnega uporabniškega vmesnika za oddaljeno vodenje proizvodne, nano-robotske celice. Uporabniški vmesnik je sestavljen iz dveh različnih programskih aplikacij, zgrajenih na dveh različnih programskih platformah. Prva, gostiteljska aplikacija je zasnovana na programskem paketu LabView 8.5 in teče na realno časovnem krmilniku. Uporabljena je kot komunikacijski vmesnik med nano-robotsko celico in oddaljenim uporabniškim vmesnikom. Druga, oddaljena aplikacija je bila zgrajena znotraj programskega paketa Microsoft Visual Studio 6.0, z uporabo programskega jezika C++. Uporabljena je za oddaljeno, virtualno vodenje nano-robotske celice. Glede na proizvodne zahteve lahko uporabnik izbira med dvema različnima načinoma vnosa trajektorije gibanja nano-robotske celice. Prvi, klasični vnos je direktni vnos trajektorije preko uporabniškega vmesnika. Na tak način je mogoče voditi vsako os posamično ali pa vse osi hkrati. Drugi režim vnosa omogoča vnos trajektorije s pomočjo haptične naprave. Pri vodenju v tem režimu, ima uporabnik ves čas na voljo povratno informacijo v obliki povratne sile, kar naredi oddaljeno vodenje še toliko bolj realistično. Oba režima vnosa sta podprta z animiranim, virtualnim, VRML modelom ciljne aplikacije. VRML model je uporabljen za izvajanje simulacij v načinu brez povezave ali pa za opazovanje dogajanja na ciljni aplikaciji v načinu z vzpostavljeno povezavo. Kot osnovni komunikacijski protokol med oddaljeno in ciljno aplikacijo je uporabljen UDP protokol. © 2010 Strojniški Vestnik. Vse pravice pridržane.

Ključne besede: oddaljeno virtualno vodenje, nano-pozicioniranje, VRML, LabVIEW Real Time, sestavljanje MEMS-ov

Slika 1. Sistemske komponente

*

Naslov odgovorneg avtorja: Resistec UPR d.o.o. & Co. k.d., Krška cesta 8, 8311 Kostanjevica na Krki, Slovenija, gregor.skorc@resistec.si

SI 95

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, SI 96 UDC 629.734.33:351.814.343

Prejeto: 10.12.2007 Sprejeto: 02.07.2010

Analiza končnih priletov jadralnega letala s spreminjanjem hitrosti po kosinusnem zakonu Zoran Stefanović - Ivan Kostić* Univerza v Beogradu, Fakulteta za strojništvo, Srbija Sodobna jadralna letala so po zaslugi visokih razmerij med vzgonom in uporom energetsko najučinkovitejše letalne naprave. Po drugi strani pa lahko prav ta njihova sposobnost postane resna pomanjkljivost med pristankom, če pride med letom do odpovedi naprav za aerodinamično zaviranje. Če letalo ne uspe hitro oddati odvečne energije, ko je blizu tal, lahko preleti teren za pristanek in konča pred ovirami s preveč energije za pristanek in s premalo energije, da bi ovire preletelo.

Razen bočnega drsenja pri končnem priletu, kjer letalo izgublja energijo zaradi povečanega bočnega upora, je v številnih člankih predlagana tudi druga rešitev tega problema. Numerične analize so pokazale, da je pristajalno razdaljo v takih primerih možno skrajšati s precej kompleksnimi oscilirajočimi potmi leta v vertikalni ravnini. Čeprav takšne poti omogočajo pomembno skrajšanje pristajalne razdalje, so za njihovo praktično izvedbo potrebne izjemne pilotske veščine. V tem članku so namesto tega analizirani veliko enostavnejši profili prileta na osnovi dveh vrst kosinusnega spreminjanja hitrosti s konstantno periodo in amplitudo, ki ga lahko izvajajo tudi piloti s povprečnimi letalnimi izkušnjami. Po določitvi hitrega konvergenčnega algoritma so predstavljene numerične rešitve za več tipičnih primerov dveh splošnih vrst spreminjanja hitrosti. Uporabljena so bila enaka začetna in končna referenčna stanja energije. Čeprav so skrajšanja razdalje manjša kot tista, ki jih dobimo s tehnikami za določanje minimalne pristajalne razdalje, so predstavljene tehnike uporabne za to kategorijo problemov zaradi enostavnosti operativne izvedbe in nekaterih drugih specifičnih prednosti. ©2010 Journal of Mechanical Engineering. All rights reserved.

Ključne besede: jadralno letalo, končni prilet, nedelujoče zračne zavore, kosinusno spreminjanje hitrosti

Sl. 4: Običajni pristanek (A) vključuje: (1) končni prilet, kjer je  ≈ konst., (2) fazo ravnanja, (3) fazo upočasnitve in (4) prizemljitev (v tem članku ni obravnavana); v članku je analiziran kosinusni prilet (B), kjer je faza (2) integralni del faze (1), pri čemer  ≠ konst.

SI 96

* Naslov odgovornega avtorja: University of Belgrade, Faculty of Mechanical Engineering, Aeronautical Department, Kraljice Marije 16, 11120 Belgrade 35, Serbia, ikostic@mas.bg.ac.rs

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, SI 97 UDK 621.795:621.922.025

Prejeto: 24.08.2009 Sprejeto: 24.05.2010

Raziskava vplivov velikosti zrna in parametrov brušenja na površinsko hrapavost in sile pri brušenju 1

Halil Demir1,* - Abdulkadir Gullu2 – Ibrahim Ciftci1 – Ulvi Seker2 Univerza Karabuk, Tehniška fakulteta, 78050, Baliklar Kayasi, Karabuk, Turčija 2 Univerza Gazi, Tehniška fakulteta, 06500, Besevler, Ankara, Turčija

Opravljena je bila raziskava vpliva velikosti zrn na površinsko hrapavost obdelovanca in sile pri površinskem brušenju jekla AISI 1050. Za merjenje in beleženje sil pri brušenju je bil uporabljen dinamometer, ki je bil zasnovan in izdelan že pred tem. Opravljeni so bili preizkusi brušenja z različnimi brusilnimi koluti z različnimi zrni. Opravljene so bile tudi meritve hrapavosti brušene površine. Rezultati kažejo, da ima velikost zrn signifikanten vpliv na sile pri brušenju in vrednosti površinske hrapavosti.

Povečanje velikosti zrn in globine reza prinaša povečanje sil pri brušenju in vrednosti površinske hrapavosti. Pri globinah reza 0,01 mm in 0,02 mm pri različnih velikostih zrn ni signifikantnih razlik v silah pri brušenju, dodatno povečanje globine reza pa prinaša tudi do 50-odstotno spremembo sil pri brušenju. ©2010 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: površinsko brušenje, sile pri brušenju, površinska hrapavost, velikost zrn brusilnega koluta

a)

0.5 mm

0.5 mm b)

Sl. 6: Na sliki brusilnega koluta z zrni velikosti 80 je pri globini reza 0,04 mm vidna izpolnitev poroznosti z materialom obdelovanca

*

Naslov odgovornega avtorja: Karabuk University Tech. Edu. Faculty, 78050 Baliklar Kayasi, Karabuk, Turkey, hdemir@karabuk.edu.tr

SI 97

Strojniški vestnik - Journal of Mechanical Engineering 56 (2010)7-8, SI 98 UDK 621.3:681.5:004.94

Prejeto: 16.01.2009 Sprejeto: 07.07.2010

Optimalna zasnova mehkega krmiljenja enosmernega servopogona v drsnem režimu Dragan Antić – Marko Milojković* – Zoran Jovanović – Saša Nikolić 1 Univerza v Nišu, Fakulteta za elektroniko, Srbija V članku je predstavljen nov krmilnik za optimalno krmiljenje hitrosti enosmernega motorja.

Krmilnik uporablja različico mehkega drsnega režima, optimizirano z genetskim algoritmom. Predlagani krmilnik ima več prednosti, med njimi zadovoljivo zmogljivost krmiljenja pri različnih obratovalnih pogojih, hitrejši odziv od običajnih krmilnikov in omejen pojav drhtenja. Simulacije, rezultati eksperimentov in primerjalne analize potrjujejo učinkovitost, odlično zmogljivost in robustnost takega krmilnika za primer enosmernega servomotorja. ©2010 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: krmiljenje v drsnem režimu, krmiljenje z mehko logiko, genetski algoritem, enosmerni motor

Sl. 6: Električni model enosmernega motorja

*

SI 98

Naslov odgovornega avtorja: Univerza v Nišu, Fakulteta za elektroniko, A. Medvedeva 14, 18000 Niš, Srbija, marko.milojkovic@elfak.ni.ac.rs

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, SI 99 UDK 532.57:621.63

Paper received: 28.08.2009 Paper accepted: 29.04.2010

Tokovne karakteristike brezlopatičnega rotorja iz odprto celičnega poroznega materiala 1

Gašper Benedik1 - Brane Širok2 - Janez Rihtaršič3 - -Marko Hočevar2 Domel d.d., Železniki in Univerza v Ljubljani, Fakulteta za strojništvo, Slovenija 2 Univerza v Ljubljani, Fakulteta za strojništvo, Slovenija 3 Domel d.d., Železniki, Slovenija

Prispevek opisuje brezlopatični rotor turbostroja, pri katerem se energija iz rotorja na fluid prenaša preko strukture poroznega materiala. Opisana je konstrukcijska izvedba turbokolesa in podane so teoretične osnove za popis toka fluida skozi rotirajoč porozen medij. Izmerjeni so tlačni padci v odvisnosti od volumskega pretoka, ki so prisotni pri volumskem pretoku zraka skozi rotorje, narejene iz različnih materialov in z različnimi konstrukcijskimi parametri. Meritve lokalnih hitrosti zračnega toka v bližini oboda rotorja so bile izvedene na stacionarnem rotorju z enokomponentnim anemometrom na vročo žičko. Analizirano je hitrostno polje v odvisnosti od lokalne strukture materiala in volumskega pretoka zraka. Predstavljene so meritve integralnih karakteristik rotorjev iz različnih poroznih materialov ter z različnimi konstrukcijskimi parametri. Prikazane so lokalne meritve radialnih in tangencialnih hitrosti v bližini oboda rotirajočega rotorja. ©2010 Strojniški vestnik.Vse pravice pridržane.

Ključne besede: brezlopatični rotor, turbostroj, porozen odprtocelični material, hitrostno polje, anemometer na vročo žičko

Slika 2 a). Brezlopatični rotor brez in b) brezlpatični rotor z radialnim inducerjem.

*

Naslov odgovornega avtorja: Domel d.d., Otoki 21, SI-4228, Železniki, Slovenija, gasper.benedik@domel.si SI 99

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, SI 100 UDK 005.72: 005.74

Prejeto: 17.03.2009 Sprejeto: 25.01.2010

Uporaba skupinske tehnologije v sestavljenih sistemih grozdov Slobodan Morača1,* - Miodrag Hadžistević1 - Igor Drstvenšek2 - Nikola Radaković 1 1 Univerza v Novem Sadu, Fakulteta za tehniske vede, Novi Sad, Srbija 2 Univerza v Mariboru, Fakulteta za strojnistvo, Maribor, Slovenija Cilj te raziskave je prispevek k prizadevanjem za razvoj postopkov strukturiranega načrtovanja kompleksnih grozdnih organizacijskih sistemov. Industrijski grozdi pomagajo podjetjem k boljšemu položaju na trgu, večji učinkovitosti, produktivnosti in boljši kakovosti izdelkov. Organizacija proizvodnje je zelo zapleten proces že v samem podjetju, če pa nalogo prestavimo na nivo grozda, dobimo še bolj zapleten in težko rešljiv problem. V ta namen v prispevku analiziramo pogoje in možnosti, ki bi omogočale takšnim strukturam prilagajanje spremembam v okolju – s prilagodljivostjo in ustreznim vodenjem proizvodnih in organizacijskih struktur – z zmanjšanjem stopnje kompleksnosti.

Trenutno ni na voljo razvitih preprostih modelov, ki bi omogočali povečanje učinkovitosti procesov v kompleksnih organizacijski sistemih, kot so industrijski grozdi. Ena izmed možnosti za zmanjšanje kompleksnosti pretokov in povečanje učinkovitosti procesov v industrijskem grozdu je uporaba skupinskega pristopa. © 2010 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: Industrijski grozdi, skupinska tehnologija, načrtovanje, delovne celice, kompleksnost, prilagodljivost

Izvršilni odbor

Revizijski odbor

Pisarna grozda Podjetje B

Podjetje A

Podjetje E Podjetje C

Podjetje H

Podjetje F

Podjetje D

Podjetje G

Plastični izdelki PROIZVODNI

Izdajatelj orodij

Vstop materiala

Rotacijski izdelki Ozobljeni izdelki

Preddelavec Načrtovalec

Prizmatični izdelki

Pločevinasti izdelki

Podjetje – član grozda Izhod izdelkov Nadzor kakvosti

Operativno vzdrževanje

PROGRAM

Slika 5. Proizvodnja v grozdu, v skladu s skupinskim pristopom

SI 100

*

Naslov odgovornega avtorja: Univerza v Novem Sadu, Fakulteta za tehniske vede, Trg Dositeja Obradovica 6, Novi Sad, Srbija, moraca@uns.ac.rs

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, SI 101 UDK 621.791.7:004.942

Prejeto: 27.05.2009 Sprejeto: 07.07.2010

Uporaba nove formule za izračun Nakaokinega koeficienta pri visokofrekvenčnem indukcijskem varjenju 1

Miroslav S. Milićević1 Visoka tehniška šola, Srbija

Postopek visokofrekvenčnega varjenja je opisan s kompleksno teorijo, ki jo je možno uporabiti za približni izračun pomembnih podatkov kot so tok, napetost ter moč induktorja in varjenca, kakor tudi stopnje izkoristka. Pri dosedanjih postopkih se izvajajo dolgi in zahtevni preračuni, ki vključujejo enačbe ter mnoge preglednice in grafične odvisnosti. V tem članku je predstavljena nova analitična metoda, ki omogoča izračun vrednosti Nakaokinega koeficienta s posebno aproksimacijo. Metoda omogoča tudi izračun pomembnih parametrov pri visokofrekvenčnem indukcijskem varjenju jeklenih cevi. Rezultati tega članka omogočajo skrajšanje in poenostavitev postopka izračuna, s tem pa optimizacijo izbire toplotnega režima. © 2010 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: numerična aproksimacija, elektromagnetno polje, visokofrekvenčno polje, ekvivalenčna shema, Nakaokin koeficient, praktična uporaba

Sl. 3. Vrednosti Nakaokinega koeficienta

*

Naslov odgovornega avtorja: Visoka tehnična šola Beograd, Bul. Nemanjića 33/39, 18000 Niš; Serbia; Lavmiro@eunet.yu

SI 101

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, SI 102 UDK 519.23:658.5:001.895

Prejeto: 18.09.2009 Sprejeto: 02.03.2010

Statistična analiza kot podlaga za strateško odločanje glede inoviranja v slovenski strojni industriji Peter Fatur* - Borut Likar Univerza na Primorskem, Fakulteta za management, Slovenija Cilj raziskave je opredeliti ključne dejavnike, ki vplivajo na inovacijske in R&R zmogljivosti industrijske panoge Proizvodnja strojev in naprav v Sloveniji. Raziskava temelji na podatkih Statističnega urada republike Slovenije. Za izbrano množico statističnih spremenljivk smo izračunali Spearmanove koeficiente korelacije. Rezultati nakazujejo obstoj dveh skupin podjetij. Podjetja iz obeh skupin so inovacijski sledilci, vendar se med seboj razlikujejo v svojih zmožnostih ustvarjanja radikalnih inovacij in inovacijskih prihodkov. Pri nobeni od skupin pa ni značilne povezave med inovacijskimi rezultati in poslovno/finančno uspešnostjo podjetja.

Raziskava ima uporabno vrednost tako za podjetja kot za tvorce nacionalnih inovacijskih politik, saj na podlagi empiričnih ugotovitev predlaga nekatera organizacijska področja, ki bi se jim moral management posvečati bolj poglobljeno kot doslej. © 2010 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: inoviranje, R&R, tehnologije, primerjalna presoja, proizvodnja strojev in naprav, industrijski management, management produktivnosti

inercijski inovatorji (novo za podjetje)

T R Ž N O S T

inovacijsko vodilni

inovacijski izgubarji

ad-hoc inovatorji (novo za trg)

INVENTIVNOST Sl. 1. Matrika tržnost/inventivnost

SI 102

*

Naslov odgovornega avtorja: Univerza na Primorskem, Fakulteta za management, Cankarjeva 5, 6000 Koper, Slovenija, peter.fatur@fm-kp.si

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, SI 103 UDK 685.632:624.014.2:519.6

Prejeto: 04.11.2009 Sprejeto: 05.07.2010

Karakterizacija dinamičnega obnašanja stropnega košarkaškega koša 1

Matija Javorski1 – Primož Čermelj2 – Miha Boltežar1,* Univerza v Ljubljani, Fakulteta za strojništvo, Slovenija 2 Iskra Mehanizmi d.d., Lipnica, Slovenija

Izvedena je bila eksperimentalna modalna analiza na obstoječem stropnem košarkaškem košu. Pri tem so bili identificirane lastne frekvence, lastne oblike ter razmerniki dušenja. Dodatno je bil zgrajen numerični model koša v okviru metode končnih elementov. Na osnovi primerjave med eksperimentalnimi ter numeričnimi rezultati je bila potrjena veljavnost slednjega. V nadaljevanju je bil numerični model uporabljen za določitev prehodnega odziva koša v primeru trenutne statične razbremenitve. Rezultirajoča nihanja so bila primerjana z napotki v veljavnih standardih na področju te športne opreme. ©2010 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: košarkaški koš, jeklene konstrukcije, metoda končnih elementov, eksperimentalna modalna analiza, nihanja

Sl. 1. Četrta nihajna oblika; eksperimentalno določena (levo), fEMA = 6.95 Hz, numerično določena (desno), fFEA =7.20 Hz

* Naslov odgovornega avtorja: Univerza v Ljubljani, Fakulteta za strojništvo, Aškerčeva 6, 1000 Ljubljana, Slovenija, miha.boltezar@fs.uni-lj.si

SI 103

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, SI 104 UDC 531.1:621.95

Prejeto: 24.09.2009 Sprejeto: 15.05.2010

Analitični opis vrtanja poligonalnih izvrtin – splošni pristop Miodrag Zlokolica* - Maja Čavić - Milan Kostić Tehniška fakulteta, Univerza v Novem Sadu, Srbija Enakostranične poligonalne izvrtine je možno izdelovati s posebnimi orodji na običajnih strojih, kot so stružnice in vrtalni stroji. Orodje se mora vrteti okrog svoje osi, medtem ko se njegovo središče giblje po kompleksni ravninski krivulji. Gibanje središča orodja je možno doseči na več načinov: z vodenjem orodja po šabloni, katere oblika se natančno ujema z obliko izvrtine, z odmikači, planetarnimi zobniškimi mehanizmi itd.

Za natančen opis procesa vrtanja poligonalne izvrtine je treba določiti geometrijo orodja, gibanje orodja in geometrijo izdelane izvrtine. Za predpis optimalnega režima vrtanja je treba pridobiti zgodovino hitrosti konice rezila orodja. Izdelane izvrtine imajo v splošnem zaokrožene kote, zato je za nadaljnje operacije obdelave nujno poznavanje polmerov kotov.

Analitični pristop, ki je predstavljen v tem članku, omogoča enostavno in učinkovito izdelavo matematičnega modela, ki opisuje geometrijo in kinematiko procesa vrtanja poligonalnih izvrtin. ©2010 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: poligonalna izvrtina, vrtanje, kinematika

Sl. 8. Položaj konic rezil orodja

*

Naslov odgovornega avtorja: Tehniška fakulteta, Univerza v Novem Sadu, Trg Dositeja Obradovića 6, 21000 Novi Sad, Srbija, mzlokolica@uns.ac.rs

SI 104

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, SI 105 UDK 316.4:001.895

Prejeto: 7.10.2009 Sprejeto: 13.1.2010

Vloga uporabnika in družbe v procesu razvoja novega izdelka Nusa Fain1,* - Niels Moes2 - Jože Duhovnik1 Fakulteta za strojništvo, Univerza v Ljubljani, Slovenija 2 Fakulteta za industrijski dizajn, Tehnična univerza Delft, Nizozemska 1

Znotraj gospodarstva podprtega z znanjem se v procesu razvoja novih izdelkov povezujejo številne institucije. Študij literature je pokazal, da so najbolj raziskani odnosi znotraj tovrstnih procesov odnosi med industrijo, univerzo in državo, ki jih prikazuje model trojne vijačnice. Na podlagi študije primera virtualnega akademskega podjetja odpiramo vprašanje, ali za razvoj novega izdelka zadostujejo samo te tri institucije. Testiramo in podpremo trditev, da sta uporabnik in družba enakovredna partnerja znotraj inovacijskih procesov. Predstavimo nov model četverne vijačnice, kjer je poseben poudarek na uporabniku in družbi v procesih razvoja novih izdelkov. © 2010 Strojn iški vestnik. Vse pravice pridržane.

Ključne besede: trojna vijačnica; uporabnik; družba;tržni poteg; tehnološki pritisk; četverna vijačnica; študijski predmet EGPR

Trojna vijačnica

Tržni poteg Razvoj izdelka

uporabnik

razpršit sprejetje Tehnološki potisk

Družba ČETVERNA VIJAČNICA Sl. 3. Pomen uporabnika v procesu razvoja novega izdelka

*

Naslov odgovornega avtorja: Fakulteta za strojništvo, Univerza v Ljubljani, Aškerčeva 6, 1000 Ljubljana, Slovenija, nusa@lecad.uni-lj.si

SI 105

Strojniški vestnik - Journal of Mechanical Engineering 56(2010)7-8, SI 106 UDK 628.8:697.353:069

Prejeto: 01.12.2009 Sprejeto: 02.03.2010

Alternativni sistem ogrevanja in hlajenja javnih stavb 1

Mitja Košir1,* - Aleš Krainer1 - Mateja Dovjak1 - Rudolf Perdan1 - Živa Kristl1 Univerza v Ljubljani, Fakulteta za gradbeništvo in geodezijo, Katedra za Stavbe in Konstrukcijske Elemente, Slovenija

V pričujočem članku je predstavljen alternativni sistem ogrevanja in hlajenja javnih stavb.

Predstavljen sistem je bil načrtovan in ekstenzivno preizkušen v prenovljeni stavbi Slovenskega etnografskega muzeja (SEM). Vgrajen sistem sestavljajo radiacijski ogrevalno-hladilni stenski paneli, tangencialni ventilatorji za hlajenje in lokalizirano prezračevanje ter centralni nadzorni sistem za upravljanje in nadzorovanje delovanja. Učinkovitost delovanja predlaganega ter kasneje vgrajenega sistema je bila temeljito preverjena skozi serijo eksperimentov izvedenih na stavbi pred in po izvedeni prenovi. Uporaba opisanega sistema v prenovljenih prostorih muzeja je posledično doprinesla k izrazitem zmanjšanju porabljene energije, boljšemu notranjemu toplotnemu okolju ter nižjim investicijskim stroškom za HAVC sistem celotne stavbe. © 2010 Strojniški vestni. Vse pravice pridržane.

Ključne besede: ogrevanje, hlajenje, prezračevanje, nizko temperaturni sistem, radiacijski paneli

Sl. 10. Poraba energije za ogrevanje nizko temperaturnih stenskih panelov vgrajenih v razstavnem delu muzeja v primerjavi s porabo celotne stavbe SEM. V prvih štirih mesecih leta so paneli delovali brez avtomatske regulacije, v zadnjih treh mesecih pa so bili vodeni s strani centralnega nadzornega sistema

SI 106

* Corr. Author's Address: Univerza v Ljubljani, Fakulteta za gradbeništvo in geodezijo, Katedra za stavbe in konstrukcijske elemente, Jamova cesta 2, Ljubljana, Slovenija, MKosir@fgg.uni-lj.si