Description

  The design and system level simulations of a reliable dual axis capacitive MEMS accelerometer is presented. The reliability of accelerometer is analyzed with respect to temperature and residual stresses. The temperature reliability of proposed MEMS accelerometer allows it to work in temperature range of -40 ℃ to 100 ℃. The residual stresses are characterized by FEM simulations to observe their effect on frequency of system. The simulation results show that the effect of residual stresses on the performance characteristics of MEMS accelerometer is negligible and can be ignored. The device is integrated with readout circuit to observe the variation in capacitance for various input accelerations. The proposed MEMS accelerometer can detect an input acceleration in the range of ± 50 g and has a capacitive sensitivity of 38 fF/g..

Structural Design

  The proposed dual axis MEMS accelerometer is designed by using the design constraints of commercially available MEMS Integrated Design for Inertial Sensors (MIDIS) microfabrication process. Fig. 1. shows the schematic of MEMS accelerometer. The proof mass is supported by 4 sets of L shape mechanical suspension beams. The internal anchors are designed in symmetry around the center. The sensing combs are attached at the outer periphery of the proof mass and are arranged in differential scheme with a smaller gap of 3.5 µm and large gap of 10.5 µm.

FEM Analysis

Modal Analysis

   Modal analysis is performed to observe the natural behavior of the device, it only provides us the information regarding the position of resonance frequencies and the mode shapes of the device at those frequencies. Fig. 4. Shows the first two modes of proposed MEMS accelerometer.

Harmonic Analysis

  To observe the behavior of the system under an external load, a harmonic analysis is performed. An external load in the form of acceleration is applied on the system at a range of frequencies and the amplitude of the response is observed. The result of harmonic analysis performed in CoventorWare MEMS+ module is shown in Fig. 5. Different input accelerations are applied with values ranging from 10 g to 50 g. The results show that at resonance the amplitude is 23.1 µm which is much higher than the gap between combs, so the system is operated in the linear region between 0 and 400 Hz, which gives us an operational bandwidth of DC-400 Hz. Here the proof mass displacement remains constant and the amplitude is small, at 50 g acceleration the proof mass displaces by 1.01 µm which is below the pull-in threshold, which is discussed in detail in section C.

Sagging Analysis

   MEMS devices like everything else, experience a gravitational pull and it is important to analyze the effect of gravity while designing the device. An acceleration of 1 g applied in the z direction can cause the mass to sag and can be the reason for misalignment of combs. It can also deflect the low stiffness mechanical springs, so it is important to estimate the effect of gravity. A sagging analysis is performed in CoventorWare MEMS+ module to characterize this behavior as shown in Fig. 6., a constant acceleration of 1 g is applied in the z direction and the movement of proof mass is observed. The proof mass displaces in the negative Z direction by 17.6 nm, this misaligns the sensing parallel plates by a minute amount but compared to the thickness of the parallel plates this value is too small to have any effect.

Thermal Analysis

   The performance of system under thermal conditions is observed by FEM based thermal analysis. The thermal expansion of the solid material causes the decrease in the small gap d1 which limits the range of the MEMS accelerometer. The differential arrangement of the sensing combs causes the pull-in acceleration to decrease for both high and low temperatures because the gap between the combs decreases in both cases.The net effect of temperature on the performance of device is cancelled-out because in case of thermal expansion or contraction the capacitance change is not affected due to the symmetry of combs around the proof mass. Fig. 7 shows the total thermal expansion and contraction of the proof mass for -40  and 100 . Fig. 8. shows the effect of temperature on the range of accelerometer due to change in gap ݀ଵ between parallel plates with change in temperature.