Authors: Zhao Yun Zhang, Wei Su, Zhi Gui Shi, Bin Tang, Zhuang Xiong, Yin Hui Chen, Bo Peng
Abstract: This paper presents the fabricationof SOI micro-accelerometer by using the sacrificial process. The structure ofthe SOI micro-accelerometer is designed and analyzed by the finite element modeling.As for the fabrication issue, the problem of electrode metal layer to standagainst HF etching is first studied. Second, to prevent the over-etching of theBOX layer during structure releasing process, the etching rate of the BOX layeris carefully investigated and an optimal etching duration is obtained. Third,the adhesion phenomenon between comb fingers during releasing process isstudied and optimized finger geometry is proposed to solve such problem.Devices based on the sacrificial process is carried out successfully, themeasurement results show that the sensitivity of the accelerometer is about 35mV/g, with a maximal measurement error of 12mg, and a maximal nonlinear error of0.41% within 50g.
616
Authors: Jian Yang, Liang Liu, Qiu Ye Lv, Xiao Wei Liu, Liang Yin
Abstract: In this paper a high-order sigma-delta modulator applied in micro-accelerometer is designed. The modulator chooses the distributed feedback structure. And the signal bandwidth is 500Hz, the oversampling ratio is 250 and sampling frequency is 250KHz. By the MATLAB Simulink simulation, when the input signal is 1g, and the signal frequency is 250Hz, the simulation result is that the noise level is-160dBV at the signal frequency in the ideal situation. And when considering the non-ideal factors, the simulation result shows that the noise level at the input accelerated signal is 20dBV higher than the ideal. The overall circuit was implemented under 0.5 um CMOS process and simulated in Cadence Spectre. The final simulation results show that the signal to noise ratio (SNR) is 97.1dB.
1266
Authors: Zhao Yun Zhang, Zhi Gui Shi, Xiao Ping He, Lian Ming Du, Hui Zhang, Bo Peng
Abstract: In order to develop the SOI micro-accelerometer front release process, this paper discusses in details the key technologies of the process. There are three problems need to be resolved: the corrosion of the electrode, the corrosion rate of the buried silicon dioxide layer as well as the anti-adhesion of the micro-structure. The corrosion characteristic of the electrode is studied, and a metal electrode of high ability of anti-HF acid corrosion is designed, after release of the micro-structure, the electrode does not fall off. The corrosion property of the buried silicon dioxide is studied, and the corrosion rate is exactly known for 2um and 5um thick buried silicon dioxide layer. Based on this, the buried silicon dioxide layer etching time can be controlled, preventing over-etching of the oxide layer. The adhesion of comb fingers and the mass with the substrate is settled at last. By electron microscopy tests, found that the process can get a good micro-structure surface, with smaller footing effects. By test, the sensitivity of the accelerometer is about 144.5mv/g.
192
Authors: Yang Gao, Ran Qin, Chen Qiu Guan, Mei Li
Abstract: In order to achieve thedynamic flight environment simulation assessment for the micro-accelerometer usedin the inertial navigation or guidance systems, it is necessary to reform the precisionstatic centrifuge’s mechanical and electronic architecture, accompanied withthe development of the novel data acquisition system. System design, hardwareand software implementation of the DAQ system for the micro-accelerometer’s dynamiccentrifuge test is reported, which is a subsystem of the dynamic precisioncentrifuge, Involving its system architecture, hardware and softwareimplementation. The DAQ system is developed based on the virtual instrument architecture,which has the ability to handle 4 analog or digital output micro-accelerometerssimultaneously. Error analysis shows that when the micro-accelerometer’s outputvoltage signal range is +2.5 V ~-2.5V, the comprehensive measurement accuracyis less than 2.5mV; when the output voltage signal range is +12.5 V ~-12.5V, thecomprehensive measurement accuracy is less than 12.5mV. When the sampling timeinterval is 20ms, synchronization error between each sampling point and thecentrifuge controller is less than 0.1ms.
214
Authors: Li Shen, Shi Qiao Gao, Yan Wei Guan
Abstract: Casimir had measured the Casimir force through experimental methods, and fitted the empirical formula between the plates. On its basis we analyse that when the gap of two separating surfaces achieves submicron in micro devices, the effect of the Casimir force to the micro-devices is great and this Casimir effect can not be ignored. The results show that: Considering the Casimir force or not, the extreme value of acceleration which can be detected by micro-accelerometer is different, and we have got the specific numerical fitting figure by Matlab .
247
Authors: Ping Li, Shi Qiao Gao, Lei Jin, Hai Peng Liu, Yun Bo Shi
Abstract: Package is one of the key technologies for micro-accelerometer. This paper researched the influence of package on the sensitivity of self-designed piezoresistive accelerometer. Based on the designed package method, the effect of thickness and elastic modulus of die adhesive, package shell materials and signal connection wires on the sensitivity of the sensor were studied by theory analysis, simulation and experimental test. According to results, the sensitivity of micro-accelerometer would be increased with the increasing thickness of die adhesive and decreased with increasing elastic modulus and the quantity of the die adhesive; besides, compared with accelerometer packaged by ceramic shell, the sensitivity of accelerometer packaged with the stainless steel shell was much bigger; meanwhile, the output sensitivity of the sensor varied with length of signal connection wire.
1367
Authors: Yu Bin Jia, Qing Wen Huang
Abstract: The basic analysis methodology of micromachined accelerometer reliability are developed. The reliability of micro-accelerometer system are basically analyzed mathematically based on probability theory, indicating that the fracture strength distribution of accelerometer structure keep to Weibill distribution, and the stress distribution keep to normal distribution. Fracture reliability degree for micromachined accelerometer is defined as the probability that strength is greater than stress. From this, we can predict and evaluate various micro-accelerometer fracture reliability under shock load. The methodology of reliability is also commonly suitable for other MEMS devices.
790
Authors: Xi Hong Ma, Chang Long Li
Abstract: In order to quickly determine the weak links of micro-accelerometer in vibration environment, the test profile of step stress and sustained stress were designed in the vibration environment on micro-accelerometer, the relevant reliability enhancement test was actualized for micro-accelerometer of 10g, the failure time was given in the sustained stress. The failure mode and failure mechanism were analyzed by detailed test results, the ANSYS software was used to simulate the move route and stress distribution of cantilever beam in the conditions of being adhesion strength and without it. According to the step stress test and constant stress test of RET, the various stress limit, the relevant failure mode and the failure mechanism were given in the vibration environment.
1881
Authors: Bing Jun Lv, Peng Fei Wang, Dong Bo Wang, Jun'an Liu, Xiao Wei Liu
Abstract: In this paper a high-performance closed-loop fourth-order sigma-delta (ΣΔ) micro-accelerometer is presented. After a introduction of sigma-delta accelerometer, system-level analysis and design of a fourth-order sigma-delta micro-accelerometer is given. The simulation result shows that an accelerometer with 107dB signal to noise ratio (SNR) and 17.5 bits effective number of bits (ENOB) is achieved. Through the root locus analysis, it is got that accelerometer is stable when quantization gain is bigger than 0.262. The accelerometer gets a good linearity and it becomes overload when input signal level is greater than -5dBFS.
134
Authors: Yan Liu, Yu Long Zhao, Lu Sun
Abstract: Dynamic and static performances are the most important parameters for accelerometers. The natural frequency decides the sensor’s working frequency band, and the accompanying stress represents the measurement sensitivity. In this paper, a novel sensing structure, cantilever-membrane structure, for piezoresistive accelerometers is studied, in order to detect the structural dimension’s effect on the sensor. With the help of FEM (Finite element method) software, the first order natural frequency of the cantilever-membrane based accelerometer is investigated with the different combinations of membrane’s dimensions. The accompanying stress of the sensing structure is also simulated in this paper. The results show that the membrane’s dimensions affect the frequency and stress more tempestuously when the membrane is short, but the tendency become gentle when the width of the membrane increases.
87