| Thermal simulation gives a lift to elevator design |
| Written by Mark Fletcher | |||||||
| Friday, 03 April 2009 | |||||||
| MagneMotion recently applied its unique linear synchronous motor (LSM) technology to develop an elevator with a vertical lift capacity of many tons. The design utilized end-to-end serially aligned LSM rails at each corner of the elevator shaft forming columns extending the full height of the elevator shaft. MagneMotion contracted with AnJen Solutions, a mechanical engineering consulting company specializing in thermal simulation, to evaluate and optimize the LSM rail baseline heatsink. “FloTHERM software from the Mentor Graphics Mechanical Analysis Division (formerly Flomerics) provided a detailed understanding of the conductive heat transfer between the heatsink and the LSM rail support structure and convective heat transfer to the surrounding environment,” said Michael Rigby of AnJen Solutions. “The simulation demonstrated that reducing the number of fins and changing the fin spacing and thickness would reduce the weight of the heatsink by 1/3 while providing the same thermal performance as the initial design.” LSM thermal considerationsAn LSM elevator is made up of encapsulated arrays of copper coils (LSM rail) mounted in each corner of the elevator shaft and stationary magnets (stators) mounted to the elevator platform. Electrical current passing through the copper coils creates a magnetic field which propels the elevator platform. A control system synchronizes the traveling magnetic field and the elevator platform. In conventional horizontal material transport systems, where the payload weight is relatively low and environment benign, MagneMotion engineers use two-dimensional finite element analysis (FEA) to predict coil and encapsulation temperatures throughout the rail. This approach requires an approximation of the convective heat transfer from the rail elements to the surrounding air. The elevator application introduces significantly greater weight, vertical rail orientation, and potentially severe surrounding environment. These new design constraints required a more accurate characterization of the surface-to-air boundary condition. CFD completely simulates LSM thermal managementMagneMotion contracted with AnJen Solutions to perform CFD on the vertical lift LSM. The advantage of CFD is its ability to model the airflow around the LSM which makes it possible to predict convection accurately. FloTHERM software is specially designed for the challenges of modeling thermal management of electronic and electrical systems. “FloTHERM offers a wide range of features such as an automatic optimizer and compact models that make it possible to improve cooling performance and reduce engineering time,” Rigby said. “These and other capabilities of the software made it possible to optimize the design of the heatsink which was important because the overall weight of the LSM was a critical concern to MagneMotion’s customer.” MagneMotion provided AnJen with the LSM geometry in a STEP file. The model consists of seven individual aluminum heatsinks conductively coupled with aluminum blocks of the same thickness at the heatsink bases. Rigby meshed the structure using cuboid elements. He modeled the horizontal and vertical members as aluminum blocks. The horizontal members have square holes with the same flow area as the round ventilation holes in the baseline mechanical database The total power entering the heat sink is 600W uniformly distributed across the heatsink back surface. The distance from the heatsink back surface (the interface surface with LSM rail) to the wall of the elevator shaft is 130.25mm. The flow cross section is 174.92 x 431.6mm. Natural convection induced by the 600W heat load was the only flow condition considered. FloTHERM solved the complete thermal problem including conduction from the motor through the mechanical structure and the heatsink and convection from the mechanical structure and heatsink to the air. FloTHERM solved the Navier-Stokes equations to determine the airflow caused by the heat loading. The initial design used a heatsink with a fin height of 94.65mm, a fin thickness of 11.47mm and a base thickness of 10.2mm. This geometry provided a weight of 68lb. The temperature at the interface between the motor and the structure was 114.7°F. The temperature was safely below the maximum interface temperature of 150°F. Rigby’s next task was to see if he could reduce the weight of the heatsink without having a major effect on the interface temperature. Iterating to an optimized designRigby evaluated 11 different design scenarios by varying the heatsink fin count, spacing and thickness in the model. A fin count of 15 and fin thickness of 3mm provided the lowest weight while still meeting the interface temperature requirements. The weight of 39lb for the optimal heatsink configuration provides a weight savings of more than 1/3 compared to the baseline configuration. Rigby also ran a number of additional cases to investigate other potential design alternatives. The first run showed that decreasing the fin height resulted in an increase in the interface temperature. The second run showed that use of a bonded fin with a fin count of 30 and thickness of 1mm did not provide any substantial improvement in interface temperature. The third run showed the effect of bolting the heatsink directly to the wall of the elevator increased the interface temperature. The fourth run looked at the effect of using one large vent hole rather than the five smaller holes used in the baseline case. This run showed only a slight improvement but was more expensive because it introduced a more costly machining process. The fifth run investigated the option of eliminating the vent holes which turned out to have a negative impact on the interface temperature. “FloTHERM provided a detailed understanding of the convective heat transfer from the LSM heatsinks and structural supports and of the conductive heat transfer within the structure,” Rigby concludes. “The CFD simulation saved engineering time and reduced time to market by enabling us to verify the thermal design of the entire LSM without having to build a prototype. Simulation also enabled a substantial reduction in weight and hence material cost of the heatsink.”
|
|||||||
| < Previous | Next > |
|---|




