THE SIMILARITY LAW FOR MEMS GYRO WITH TEM- PERATURE CHANGES Bohua Sun Centre for Mechanics and Technology, Cape Peninsula University of Technology, Cape Town, South Africa e-mail: bohua.sun@gmail.com The performance of MEMS gyro is affected by several factors, one of them is the change of temperature. To design a higher performance MEMS gyro, the temperature compensation has to be taken into account. Due to its complicated nature, the temperature data has to been collected via experiments. The dimensional analysis has been used to find he similarity law of temperature effects on resonant frequency, Q factor and voltage output. Since we have giving no any restriction on the MEMS gyro structure, it seams that all our similarity laws might be valid to any other mechanical or electro-mechanical system. 1. Introduction Because of rapid development of Internet of Things, more and more motion sensor such as MEMS gyros are needed desperately. However, the performance of the MEMS gyro is affected badly by the change of temperature, in general the MEMS gyro can not be used for high-end field if no temperature compensation. In practice, the temperature compensation of each MEMS gyro has to be tested one-by-one, this process is very costly and time consumable. There are numbers of publica- tions investigated the temperature problem of the MEMS gyro and have made some achievement for particular case[1,2,3,4]. However, due to the complicated nature of the problem, there is no any result on the general consideration on the problem. In this paper, we report our investigation on the problem by using dimensional analysis. Some similarity laws have been formulated. Those general results must be benefit to the future development high performance MEMS gyro. 2. Dimensional analysis and Buckingham Pi theorem[5,6] In many cases in real-life engineering, the equations are either not known or too difficult to solve; oftentimes experimentation is the only method of obtaining reliable information. In most experiments, to save time and money, tests are performed on a geometrically scaled model, rather than on the full-scale prototype. In such cases, care must be taken to properly scale the results. All mathematical equations must be dimensionally homogeneous; this fundamental principle can be applied to equations in order to nondimensionalize them and to identify dimensionless groups, also called non-dimensional parameters. A powerful tool to reduce the number of necessary inde- pendent parameters in a problem is called dimensional analysis. Dimensional analysis is useful in all disciplines, especially when it is necessary to design and conduct experiments.The beauty of dimen- sional analysis is that the only other thing we need to know is the primary dimensions of each of these ICSV21, Beijing, China, July 13-17, 2014 1