Rapid Application Development Tool Tesla for Fast Prototyping of Electrical
Machines 12 Peter van Duijsen, Davor Gospodarić
1 Simulation Research
P.O.Box 397, NL-2400 AJ, Alphen aan den Rijn, The Netherlands
Tel: +31 172 492353, Fax: +31 172 492477
www.caspoc.com, e-mail: firstname.lastname@example.org
Karl-Benz-Str. 19 D-70794 FILDERSTADT
Tel +49 711 700 1400
Abstract Induction, DC, Permanent Magnet, Universal, Designers of electrical machines and drives Synchronous and BLDC Machines. Using the need a Rapid Application Development (RAD) tool the designer can enter geometric data, tool during the design phase of an electrical material data and mechanical data, and obtain machine in order to perform fast prototyping. the performance of the machine such as Using the tool various prototypes of machines speed-power, speed-efficiency, and speed-can be designed. torque characteristics.
Also simulation models are generated for I Introduction testing the prototype in a system simulator, In this paper a Rapid Application Development where it is connected to the power electronics, tool is described for fast prototyping of driving mechanics and the control of the drive.
Figure 1: Coupling between the design tool and simulators
Coupling to FEM/BEM tools is provided for a ; D.C motor with compound
more detailed design of the machine where a excitation
magnetic, thermal and mechanical strength TESLA-P
analysis can be performed. Optimization can Calculates permanent magnet D.C be performed on the design of the machine or motors. TESLA-P considers the on the overall design of the drive system, numerical possibilities of the iterative where the machine is one of the components. treatment of non-linearity and the The tool is based on the idea of a construction effect of magnetic field calculation. kit where basic designs of typical parts, such The program takes into account the as rotor, fan, windings, magnetic material, effects of demagnetization and bearings, rotor bars, etc, can be constructed temperature distribution. that are manufacturer specific. Using such TESLA-S
libraries of components, the user can prototype Calculates the operational various variants of his design based on characteristics of synchronous motors. standard parts already designed. TESLA-U
The calculation of the characteristics is done Calculates universal series motors. using analytical methods and dedicated TESLA-E
numerical routines inside the Rapid Application Calculates brushless D.C. motors. Development tool and therefore the response TESLA-E solves both 2 phase and 3 to the user to a change in parameter is phase brushless D.C. motors with both interactive. In the tool the complete magnetic slotless and slotted stator windings. circuit is analyzed, resulting in the
characteristics of the machine. Here the non-III. Requirements of the design tool linear material characteristics are used for the
applied magnetic materials. Wire parameters A design tool for electrical machine would are based on standards such as DIN, SEV, need the following features:
JASO, etc. For the geometric data the ; Direct interface to external field calculation designer can mix SI and US metrics such as software, such as Ansys, DynaCom, "inch" and "mm" in one design and couplings Amperes and Magneto and simulation exists to popular CAD packages. tools like Spice, Simulink and CASPOC Compared to lengthy Finite Element ; Library of pre-calculated flux curves simulations the user is able to prototype a new ; Motor component library for easily type of electrical machine in a very fast way. designing stators, rotors and commutators ; Non-linear analysis capabilities II Basic Modules ; Calculation of mechanical and electrical losses due to air friction, bearing loss, The basic modules in the Tesla suite of design blower/fan loss, copper loss, hysteresis tools are broken down into six specific loss and eddy current loss modules for various types of motors. ; Outputs such as power, currents, efficiency, lamination data and TESLA-A dimensioning, speed/torque curve, etc Calculates the operational ; Graphical outputs such as lamination characteristics of 3-phase induction geometry, magnetization curves, current, motors. The software utilizes classical efficiency, speed, torque, power and power numerical theory for solving induction losses machines. ; Archiving capabilities of all motors and TESLA-G motor components allowing for search Offers the possibility of practice-capabilities on past motor designs. oriented calculations of D.C commutator motors. TESLA-G is used IV Coupling to model: ; Shunt connected commutator The design tool should have the possibility to motors (self-excited) couple directly with CAD and Simulation ; Shunt connected commutator packages. motors (separately excited) In the first place the prototype is build from a ; Series motors basic geometry. From these data a 2D or 3D
Figure 2: Generator simulation with stator design templateCAD model can be exported for further the power converter, mechanical load and mechanical and electromagnetic design. control circuitry or embedded control. In many cases a rotational 2D model would be Figure 2 shows the simulation of a satisfactory. Also material data for the synchronous generator with rectifier and buck properties of the non-linear material has to be converter. The design of the generator is supplied to the CAD package. For 3D carried out in the RAD tool Tesla. The modeling in CAD packages the export of the parameters of the behavioral model of the lengths and an eventual rotor skewing has to generator are used for the model in Caspoc. be done. The templates for the machine are selected In the second place the prototype has to be from the project manager in the design tool. tested in a complete drive simulation. For each component a template exists that Therefore the electromagnetic model has to be guides the designer to enter specific values for exported into a behavioral model required for the design of that particular component. simulation. Not only should the model and the More components can be defined and stored parameters contain the electrical parameters, in a project. For example, 3 different types of but also the mechanical parameters are stators can be defined and each of them can required for the simulation, such as the rotor be used in the total design.
inertia and friction of the bearings. The parameters for the prototype are Since the prototyping takes place in the design automatically transferred to the simulation tool and because of its connections to the tools, so changes in the design are system simulation, optimization can be immediately effective in the simulation. This performed based on the prototype of the allows fast prototyping of not only the machine machine inside the complete drive, including but also performance of the machine in the
total drive system.
Figure 3: Design Process
V Machine EM design Manufacturers provide libraries with data for
their materials, such as, for example, iron and The electrical machine is mainly designed for magnetic materials. From this data losses such the electromagnetic properties. In figure 3 the as hysteresis and eddy current losses are process from geometric data to design info is calculated.
shown. The data for the windings allows copper losses The input for the design is not only geometrical to be calculated.
data, but also material data has to be supplied In the design tool the behavior of the machine Important are the non-linear materials, see is calculated directly using analytical figure 4. expressions. This allows the fast determination
of specific design considerations, such as
power and speed curves. Also, for example,
the efficiency as function of the speed of the
rotor can be displayed graphically. From these
data the machine can be optimized.
Figure 5 shows the analytical results from
Tesla. Shown are the efficiency (*10%), power
(*1000 w), current (*100 A) and speed (*1000
rpm) as function of the torque (Nm)
The parameters for the behavioral model in
Caspoc are exported in a model file. For
example, parameters for a Switched
Reluctance Machine are exported as:
.model srm68_1 user Figure 4: Non-linear material properties br=0.436 bs=0.349 lmax=110m lmin=10m
nr=6 ns=8 r=50 jr=10m
Figure 5: Graphical results from Tesla
For a more detailed analysis of the electrical provided in addition to a rotational symmetrical machine the prototyped design can be 2D model. Figure 7 shows the model of a exported into a FEM/BEM package. Since all Switched Reluctance Machine in Ansys. geometric data is entered in the design tool, Also mechanical components are prototyped in the model is exported automatically to the the design tool. In figure 6 the template for a FEM/BEM package Only data such as the fan is shown.
length of the rotor and rotor skewing has to be
Figure 6: Prototyping of mechanical components
Figure 7: Model of a Switched Reluctance Machine in a FEM package
 Killat U., Otto J., van Duijsen P.J., Summary Parameter extraction in FEM models for A new Rapid Application Development tool is dynamic system simulations. PCIM, presented for fast prototyping of electrical Conference proceedings IM, 2002, machines. Models from the prototype can be Nürnberg, Germany 2002
used in CAD-, Finite Element and Simulation  Matveev A., van Duijsen P.J., Novel software. The electromechanical Caspoc-based software for multilevel characteristics are obtained using the RAD tool simulation of switched reluctance drives. and enable the designer to optimize the design Proceedings PCIM 2003
in a short time and very cost efficient.  ANSYS Theory Manual, Release 5.7,
Swanson Analysis Systems, Inc., 2001 Literature  Simulation Research / Trimerics, Tesla
users manual 2003,
 Gospodaric D., Schmenkel A., Simulation  IES Amperes / Magneto, Users manual
of Power Trains for Hybrids and Electrical 2003
Cars, Proceedings PEMC Cavtat 2002  Simulation Research, Caspoc Users  Duijsen. P.J. van, Simulation and manual, 2003, www.caspoc.com
animation of power electronics and
electrical drives, PCIM Europe, December
2001, p 26-28. www.caspoc.com
 Otto J., Killat U. van Duijsen P.J., Energy
Based Model Synthesis for Electrical
Actuators and Sensors, PCIM, Conference
proceedings IM, 2002, Nürnberg, Germany