Transcript Design and Simulation of a Novel MEMS Dual Axis - Asee

Design and Simulation of a Novel MEMS
Dual Axis Accelerometer
Zijun He, Advisor: Prof. Xingguo Xiong
Department of Electrical and Computer Engineering, University of Bridgeport, Bridgeport, CT 06604
Abstract
The intrinsic frequency of the device along Y-axis is
In this project, the design and simulation of a novel bulk-micromachined capacitive
MEMS dual axis accelerometer based on Silicon-on-Glass (SoG) structure is
proposed. The device has 32 differential capacitance groups on top and bottom of an
H-shaped central mass to sense acceleration along X direction. It also has 32
differential capacitance groups on the left and right side of the central mass to sense
acceleration along Y direction. The H-shape movable mass is supported via four
folded beams which can bend along X direction, and in turn the folded beams are
connected to substrate anchors via two straight beams which can bend along Y
direction. This unique structure allows the device to sense acceleration along both X
and Y directions. Theoretical model is used to analyze the working principle of the
accelerometer. Based on analysis, a set of optimized design parameters are
suggested. ANSYS simulation is used to verify the function of the device. The
proposed accelerometer can be used for 2-axis inertial navigation applications.
Introduction
MEMS (Microelectromechanical Systems) refer to devices and systems integrated
with electrical and mechanical components in the scale of microns (1μm=10-6m). Due
to their small size, low weight, low cost and low energy consumption, MEMS devices
have achieved great commercial success in recent decades. MEMS accelerometers
have been widely used in automobile airbag deployment systems, inertial navigations,
etc. In this poster, the design and simulation of a novel bulk-micromachined capacitive
MEMS dual axis accelerometer based on Silicon-on-Glass (SoG) structure is
proposed. The H-shape movable mass, which has many differential capacitance
groups surrounding it, is supported by a set of vertical folded beams and horizontal
straight beams. The vertical and horizontal beams can bend along X direction and Y
direction separately. Therefore, the device can detect acceleration along these two
directions. The proposed accelerometer can be used for 2-axis inertial navigation
applications.
Accelerometer Design
fy 
1
K y _ tot
2
Mh
Based on above analysis, we plot the relationship between X-axis (Y-axis)
displacement sensitivity and the width of folded beams (straight beams) respectively,
as shown in Fig. 3 and Fig. 4. As shown in the figures, the X-axis (Y-axis) sensitivity is
very sensitive to the width of folded beams (straight beams). If the beam width is
reduced, the sensitivity increases very rapidly. The curves can guide us in the device
design optimization.
Design Optimization and Simulation
Based on above analysis, we derived a set of optimized design parameters of the
MEMS accelerometer, as shown in Table 1. The gap between two movable fingers is
20μm, the thickness of the device is 80μm.
Table 1. The optimized design parameters of the MEMS accelerometer
Name
Amount
Length(μm)
Width(μm)
Central mass
1
100
500
Side mass
4
200
50
Movable Fingers (horizontal)
32
160
4
Movable fingers (vertical)
32
160
4
Folded beams
4
500
4
Straight beams
2
300
4
Connection mass
2
40
200
Connector1
4
15
20
Connector2
2
40
20
ANSYS simulation is used to verify the function of the device design. We performed
ANSYS sensitivity analysis to simulate the displacement of the movable mass in
response to 1g input acceleration along X and Y direction, and the results are shown
in Figure 5 and 6 respectively. The results are in good agreement with hand
calculation results.
The novel bulk-micromachined capacitive MEMS dual axis accelerometer structure is
shown in Fig. 1.
(a). Without acceleration (a=0)
(b). With acceleration (a≠0)
Figure 1. Structure design of the novel
Figure 2. Differential capacitance
MEMS dual axis accelerometer
sensing in MEMS accelerometer
As shown in Figure 1, H-shape central mass is connected to four folded beams,
which are in turn connected to two straight beams with one end anchored to
substrate. There are 32 horizontal and 32 vertical movable fingers extruding from
side and central mass respectively. Vertical movable fingers constitute differential
capacitance with left/right fixed fingers, and horizontal movable fingers constitute
differential capacitance with top/bottom fixed fingers. If there is horizontal
acceleration input, folded beams will deflect due to inertial force. By sensing
horizontal differential capacitance change (see Fig. 2), the input acceleration along
X direction can be measured. By sensing. Similarly, vertical acceleration input will
cause straight beams to deflect. By sensing vertical differential capacitance change,
the input acceleration along Y direction can be measured. This is the working
principle of the dual-axis MEMS accelerometer.
Assume the width, length and thickness of the folded beams are Wb1, Lb1 and tb1
separately, and the width, length and thickness of straight beams are Wb2, Lb2 and tb2
separately. The mass of H-shape mass is Mh. Young’s modulus of Si material is E.
The total spring constant of four folded beams is K x _ tot 
4
2
K b1 
L b1
1
24 EI b 2
3
b2
The intrinsic frequency of the device along X-axis is
0.45
4.5
0.4
4
Displacement Sensitivity (um/g)
Displacement Sensitivity (um/g)
5
0.35
0.3
0.25
0.2
Sd=0.138um per g
1
12
F x _ inertial
S dx 
3
Lb 2
3
W b 2tb 2
M hg

K x _ tot
F y _ inertial
1
K x _ tot
2
Mh
K x _ tot
M hg

K y _ tot
K y _ tot
(per g)
(per g)
Figure 9. The fabrication sequence of the accelerometer
3.5
Conclusions and Future Work
3
2.5
Sd=1.96um per g
2
1.5
0.5
0.05
3
3.5
4
Width of folded Beam (um)
4.5
Device Fabrication
2 EW b 2 t b 2
1
0.1
0
2.5
3

Figure 8. Stress distribution when
input Y-axis acceleration ay=5g
Relationship between the width of straight beam and Sensitivity
Relationship between the width of folded beam and Sensitivity
0.5
0.15
fx 
3
t b 1W b 1
L
where Ib2 is moment of inertia of one section of straight beam, I b 2 
The Y-axis displacement sensitivity of the accelerometer is
12
Figure 7. Stress distribution when input
X-axis acceleration ax=5g
The fabrication sequence of the bulk-micromachined comb accelerometer is shown in
Figure 9. The device is based on Silicon-on-Glass (SoG) compound structure. Siliconglass anodic bonding and DRIE etching are used in the fabrication. The fabrication
flow is shown in Figure 9.
3
L
The total spring constant of two straight beams is K y _ tot  2 K b 2 
S dy 
2 Et b 1W b 1

3
b1
where Ib1 is moment of inertia of one section of the folded beam, I b 1 
The X-axis displacement sensitivity of the accelerometer is
3
24 EI b 1
Figure 6. Displacement sensitivity
Figure 5. Displacement sensitivity analysis
analysis when Y-axis acceleration ay=1g
when X-axis acceleration ax=1g
We also performed stress analysis on the accelerometer to show its stress distribution
in response to 5g acceleration along X and Y directions, as shown in Fig. 7 and 8
respectively. We can see the maximum stress occurs at the end of beams. Thus it may
be helpful to widen those regions in the design for improved device reliability.
5
Figure 3. X-axis displacement sensitivity
versus the width of folded beams
0
1.5
2
2.5
3
3.5
4
Width of Straight Beam (um)
4.5
5
Figure 4. Y-axis displacement sensitivity
versus the width of straight beams
In this project, the design and simulation of a novel bulk-micromachined capacitive
MEMS dual axis accelerometer is proposed. Due to the design of beams, the H-shape
mass can move along X and Y direction separately under corresponding inertial forces.
Hence, the input acceleration along X and Y directions can be measured by differential
capacitance sensing. Based on the analysis, an optimized design is suggested.
ANSYS simulation is used to verify the device performance. The fabrication flow of the
device is also proposed. In the future, we will look into the signal coupling between X
and Y axis of the device, and minimize its influence by improving the device design.