SPEEDLink quickly and easily imports any model created with the University of Glasgow's SPEED motor design program into MagNet. The geometric, material, coils, and boundary condition data in the SPEED file is used to automatically build and solve a MagNet model of the same motor design.
What SPEEDLink Assistant extracts from the SPEED model
- 2D geometric, material, and coil data
- Stack length data that allows for the construction of a true 3D model
- Stator, rotor, and air gap data as a multiplicity of identical shapes with common material properties that are repeatedly copied to form the complete model
What SPEEDLink Assistant does with the data imported from SPEED
- Uses Boolean union operations to coalesce SPEED sub-components into single stator iron and rotor iron components. This grouping of the components into two bodies (i.e. rotor and stator) allows easy set-up for:
- Dynamic simulation of motion effects, usually used for simulating startup or supply-switching or fault conditions
- Parametric study in MagNet and optimization in OptiNet, Infolytica's Design Optimization software
- Automatically applies the corresponding periodic boundary conditions to part models, such as one or two poles of a multi-pole machine, when it is required for a MagNet simulation
- Adds a limited number of additional regions to represent the air gap and the other major parts of the motor
- Allows users to set up the analysis for one or more rotor positions (e.g. from 0-30 degrees, analyze for every one degree)
- The stack length data can be redefined
The materials used in the SPEED model are imported as-is, and may be copied to permanent new material names, or replaced by corresponding materials already in the library.
Automatic mesh controls are included to ensure that the air gap and curved slot shapes are accurately represented and that the results are also accurate. These mesh controls may be readily overridden by the user, if preferred.
An accompanying log file is produced, ensuring that all the information covering the conversion is available for review later.