Inertial sensors are commonly used in many different applications including vehicle rollover sensors, aircraft sensors, and others. The sensors should be capable being used in many different environments, and be relatively ruggedized. In addition, it is important that the sensors produce output signals which are accurate. Various kinds of environmental noise, of various forms, may effect the accuracy of such sensors.
The present application describes the formation of an inertial sensor on a silicon substrate, and in an embodiment is formed using Micro Electro Mechanical Systems or MEMS. An aspect disclosed herein describes mitigating the vibration susceptibility of the sensor. Another aspect describes three-dimensional sensors, and ways of isolating the different orthogonal axes of information.
a and 1b show respective views of an angular rate sensor of an embodiment;
a shows a piezo resistive element on the sensing element;
a and 4b show an anti-phase vibration mitigation embodiment;
a-9c illustrate the x and z axis acceleration sensing of the accelerometer;
a-10d illustrate how the Wheatstone bridges can sense the acceleration parameters; and
a and 11b illustrate shock survival techniques for the sensor.
The general structure and techniques, and more specific embodiments which can be used to effect different ways of carrying out the more general goals are described herein.
An embodiment is shown in
An anchor part 100 supports the entire silicon beam-mass structure, and acts as the support for the system. The driving element 110 may include a ZnO piezoelectric film deposited on a Si3N4 beam. In operation, actuating the driving element 110 causes the structure to vibrate in the vertical plane.
A decoupling island 120 may decouple the driving part 110 from the sensing parts, which includes a vertical beam 130, sensing elements 150, 155, and proof mass 140. The beam 130 connects to a connection surface of the proof mass which is along a surface of the proof mass that faces to the decoupling island. Other surfaces of the mass, which are closest to the edges of the mass, abut against the sensing elements.
Vertical supporting beam 130 is connected to the decoupling island 120 to support the proof mass 140, while allowing the proof mass to move in specified ways. The supporting beams such as 130 effectively forms springs. The supporting beams are columnar in shape, with a rectangular, non-square cross section that defines a thicker thickness, which is in the z direction in
The strain on the sensing elements 150, 155, which are located around the vertical beam 130, indicates the amount of movement of the proof mass. The sensing elements may produce respective output signals which are detected by and analyzed by electronic circuitry shown as 160.
A first embodiment uses piezoresistive sensing to detect the movement.
In order to maintain the standards for a rollover application, the system is much more lenient. The area of rollover specification is shown as 205. In the embodiment, the sensing mode resonant frequency is between 352 and 595 hertz, in order to maintain it within the rollover sensitivity. This only requires a difference between sensing mode and driving mode as being less than 25% or less.
Table 1 illustrates a set of structural design parameters that satisfy the specifications for both stability control and rollover applications
The total size of the sensor chip is 6000 um×3000 um by 530 um. The driving mode resonant frequency is 477.5 Hz. The sensing mode resonant frequency is 502.5. Sensitivity of the angular rate sensor can be up to 0.00635*(V. supply)/degrees/sec.
Table 2 illustrates the structural design parameters which allow the piezoelectric sensing to satisfy the specifications for stability control and rollover applications. Again, the total size of the sensor chip is 6000 μm by 3000 μm by 530 μm. The driving mode resonant frequency is 477.5 Hz and the sensing mode resonant frequency is 498 Hz. The sensitivity of the angular rate sensor can be as high as 0.00608*(V.supply/degrees/seconds).
This system can satisfy the specifications for automotive stability and rollover controls. Moreover, the fundamental resonant frequency of the sensor can be around 500 Hz in order to satisfy the specification that the frequency response should be between 10 and 50 Hz.
The sensing signal is from two different elements 150 and 155. Effectively that sensing signal is a differential mode signal where one beam receives tensile stress and the other beam receives compressive stress. Because of this, the system is not sensitive to vertical vibration and acceleration. These vibrations produce a common mode signal on the two sensing beans which cancel each other out.
Another aspect describes embodiments to mitigate the vibration susceptibility of this sensor.
A first embodiment takes advantage of the resonance and drives the resonant frequency of the driving mode of the sensor within a specified range that is outside the range of expected vibration and hence provides some vibration independence. For example, the driving mode resonant frequency may be set to around 3000 Hz, taking it about 1000 Hz away from the vibration environment frequency range of 20-2000 Hz.
This embodiment may use a PZT film instead of zinc oxide as the piezoelectric driver, to provide a higher d31. More specifically, by selecting the sensing mode resonant frequency to be within the range of 2300 to 4000 Hz, this also maintains the frequency outside the specification for automobile stability control.
In this embodiment, the difference between the sensing mode and the driving mode resonant frequency is less than 23%. If only the vertical beam is changed, while the other structural parameters are kept the same this can use a vertical beam having a size between 96.6 and 140.4 μm.
Table 3 satisfies the specification for both stability control and rollover applications. The total size of the sensor chip is around 6000 μm×3000 μm×530 μm. The targeted driving mode and sensing mode resonant frequencies are around 3000 and 3500 Hz, respectively. The sensitivity of the angular rate sensor can be up to 0.0102*Vsupply/degrees/seconds.
A second embodiment of vibration independence is explained with reference to
In-phase oscillations may cause vibrations along the z-axis to be transmitted via the anchor block and from there, into the sensor's supporting structure. This type and magnitude of vibration could be transmitted into the electronics PCB where it can cause electronic component failures and other undesirable effects. Transmitted vibrations are, therefore undesirable. An embodiment changes the driven oscillation mode to one where each mass is induced to vibrate with a 180 degrees phase shift with respect to each other. That is, both masses will vibrate at the same frequency. However, when one mass is at its maximum deflected position in the +Z direction, the other mass is at its maximum deflected position in the −Z direction.
Ad1=Ad+B (1)
Ad2=−Ad+B (1)
From equation 1, it follows that:
Equation 3 shows that the vibration amplitude produced by the environmental vibration is canceled out, thereby providing vibration independence.
Another embodiment shown in
In this embodiment, the driving element 500 uses silicon thin beams 520, 522, in place of the silicon nitride parts used in other embodiments. As in these other embodiments, this embodiment uses resistive sensors 510, 512 on the sensing element beams 514, 516, respectively. The additional pair of piezoresistors 520, 522 are used as R3 and R4, on the silicon flat beam.
The resistors R1/R3 forms a first half Wheatstone bridge, and the resistors R2/R4 form the other half Wheatstone bridge. Each produces an output voltage. In the embodiment, each of the four resistors has a resistance value of R when unstrained. The difference between the base value R, and the strained value, produces the output voltages as follows.
The driving amplitude:
X=Ad sin(ωt) (4)
The sensing amplitude:
The outputs from the two half Wheatstone bridges:
The phases of Vout1 and Vout2 when Vout1 and Vout2 are equal to 0:
The phase difference:
The phase difference is proportional to the angular rate Ω and is not related to the vibration amplitude. S1, S2 and G are determined by the structural parameters, driving frequency, resonant frequency of sensing mode, and quality factor. Thus, the phase-difference detecting scheme can reduce the sensitivity to environmental vibrations.
Another embodiment describes a three axis accelerometer. The three axis accelerator uses two masses to sense acceleration along three orthogonal axes. The embodiment includes two symmetric parts, each of which includes a proof mass, shown as proof mass 1 (600) and a proof mass 2 (610). Each part also includes one vertical beam (602 and 612), one decoupling island (604, 614), and two sensing beams (620,622 and 624,626). The beams can sense as described in previous embodiments. In this embodiment, piezoelectric films can be deposited on the sensing beams to allow piezoelectric sensing. The sensing beams 620, 622, 624, 626 sense the movement in the “y” direction, as described herein.
Coupling beams 640, 642 connect the decoupling island 604 to anchor 650, which holds the first and second parts together. Analogously, the coupling beams 644, 646 connect the other decoupling island 614 to the anchor 650. The anchor 650 connects to two groups of sensors—a first group sensitive to the x-axis acceleration, and a second group sensitive to the z axis acceleration.
The supporting beams 640, 642, 644, 646 are separated from the sensing elements 620, 622, 624, 626. This may optimize the sensing elements and the. supporting beam separately in order to satisfy bandwidth requirements and maximize the sensitivity. Cross axis sensitivity can also be minimized. Tables 4 and 5 show the parameters of an embodiment.
Therefore, in summary, the Three-axis Accelerometer with Piezoresistive Sensing can have the following characteristics:
(1) input voltage: 5V
(2) un-amplified sensitivity:
(3) Minimum detectable signal:
(4) cross-axis sensitivity: <2.2%
(5) Bandwidth: DC-70 Hz
(6) Shock survival: need to include shock protection structure appropriate to the package.
The support beams contribute mainly to the resonant frequency of the accelerometer, while the sensing beams affects its sensitivity. This provides the flexibility to optimize the structure parameters of the support beam and sensing beam to maximize the figure of merit (i.e., sensitivity*ω2).
For example, the vertical beam 602, 612 is the support beam for Y-axis acceleration sensing, while the sensing beams, also called “tiny beams” because of their thin construction in certain embodiments, are on both sides of the vertical beam 620, 622, 624, 626.
The thickness and other structural parameters of the vertical beam 602, 612 may be determined by the resonant frequency and bandwidth requirement for a specific application. The thickness of the vertical beam, and also the length and width of the vertical beam, as well as the two sensing beams to some minor degree, may effect the resonant frequency.
In the structure-parameter design summarized in Table 4, the width of the vertical beam is chosen to be 28 μm. This vertical beam thickness can be achieved with Deep Reactive Ion Etching (DRIE). The sensitivity is determined by the thickness, length, width and thickness of the tiny sensing beam with trade-off among those parts.
Table 5 shows selecting the thickness of the tiny beam to be 2 micron, which can be achieved by using silicon wafers with 2 μm thick N-type epitaxial layer on a P-type substrate. In an alternative embodiment, these may use silicon-on-insulator wafers. Electrochemical etching can be used to etch the P substrate, so that the etching stops exactly at the P-N junction of the epi wafer, leaving a 2 μm thick N-type diaphragm or beams with very good uniformity and repeatability.
The structure 612 has a rectangular shape to constrain the movement of the mass 112 to the y axis. This forces substantially pure compression on-the sensor 626; and substantially pure tension on the other sensor 624.
a-9c show a force distribution of the
a-10d illustrate how these resistances, formed into a Wheatstone Bridge configuration, can be connected to sense the different kinds of acceleration.
In order to optimize parameters of the device, a finite element analysis of the device may be carried out for the twin mass structure. When 1 G of acceleration is applied, stresses as high as 390 MPa can be achieved. Sensitivity can be up to 257 mV per G per V, and resonant frequency can be as high as 333 Hz. One G of Y axis acceleration produces minimal stress on the x and z axis. Therefore, there is minimal cross beam sensitivity.
The stress on the sensing beams can be up to 180 MPa. Device function and structure may be improved by optimizing the structural parameters such that the asymmetric stress distribution on the beam is minimized.
Even higher stresses may cause damage if the stress is high enough to exceed the structural limits of the system
a-7f illustrate the device formation, with each figure illustrating both a side view and cross-sectional view of each step.
a illustrates a first operation. A P type silicon substrate 700 with an epitaxial layer 701 is deposited with a layer of silicon nitride. Low dose boron is implanted in at least one hole 702 in silicon nitride layer 705 to form piezo resistors. The epitaxial layer can be 2-3 μm thick, and can be an n type epitaxial layer on a p type substrate. An electrochemical self etching stop can be used to form the tiny sensing beans in this way.
After the initial low dose boron ion implantation, a higher dose boron ion implantation is used to form the ohmic contacts 710 as shown in
The low dose boron parts are used as etch stops in
The sensors described herein may produce their outputs to be displayed on any kind of display, e.g., an analog dial or a digital display. The outputs may also be processed by one or computers which may be any kind of computer, either general purpose, or some specific purpose computer such as a workstation. The computer may be a Pentium class computer, running Windows XP or Linux, or may be a McIntosh computer. The programs may be written in C, or Java, or any other programming language. The programs may be resident on a storage medium, e.g., magnetic or optical, e.g. the computer hard drive, a removable disk or other removable medium. The programs and the data may also be run over a network.
Although only a few embodiments have been disclosed in detail above, other embodiments are possible and are intended to be encompassed within this specification. The specification describes specific examples to accomplish a more general goal that may be accomplished in other way. This disclosure is intended to be exemplary, and the claims are intended to cover any modification or alternative which might be predictable to a person having ordinary skill in the art.
For example, the above has described certain parameters with reference to x, y, z orthogonal axes. It should be understood that the parameters can be sensed in different ways. In addition, the above has described different kinds of driving and sensing mechanisms, for example it has described piezo based driving mechanisms, and piezo resistive (as well as piezoelectric) based sensing mechanisms. It should be understood that other driving mechanisms, including magnetic, and other driving mechanisms can be used. Moreover, the sensor can be any kind of sensor.
The above has also described an embodiment formed using MEMS, but it should be understood that other formation techniques can be used.
Also, only those claims which use the words “means for” are intended to be interpreted under 35 USC 112, sixth paragraph. Moreover, no limitations from the specification are intended to be read into any claims, unless those limitations are expressly included in the claims.
This application claims priority from provisional application Ser. Nos. 60/614,858 and 60/614,909, both filed Sep. 30, 2004, the contents of which are herewith incorporated by reference.
Number | Date | Country | |
---|---|---|---|
60614858 | Sep 2004 | US | |
60614909 | Sep 2004 | US |