U.S. patent application number 11/163388 was filed with the patent office on 2007-04-19 for mems sensor package leak test.
This patent application is currently assigned to HONEYWELL INTERNATIONAL INC.. Invention is credited to Mark J. Jarrett.
Application Number | 20070084270 11/163388 |
Document ID | / |
Family ID | 37946928 |
Filed Date | 2007-04-19 |
United States Patent
Application |
20070084270 |
Kind Code |
A1 |
Jarrett; Mark J. |
April 19, 2007 |
MEMS SENSOR PACKAGE LEAK TEST
Abstract
Methods and apparatus are provided for detecting leaks in a MEMS
sensor package, and in particular, a MEMS sensor package that
includes an oscillating structure or element that has a Quality (Q)
value. The method and apparatus may include measuring the Q value
of the MEMS sensor at a first time, applying a pressure to the
outside of the MEMS sensor package, and measuring the Q value of
the MEMS sensor at a second time after pressure has been applied
for a period of time. A change in the measured Q values between the
first time and the second time may be determined, which may then be
correlated to a leak rate for the particular MEMS sensor package.
In some cases, a leak rate of 2.times.10.sup.-13 He atm.cc/s or
less may be detected.
Inventors: |
Jarrett; Mark J.; (Lake
Elmo, MN) |
Correspondence
Address: |
HONEYWELL INTERNATIONAL INC.
101 COLUMBIA ROAD
P O BOX 2245
MORRISTOWN
NJ
07962-2245
US
|
Assignee: |
HONEYWELL INTERNATIONAL
INC.
101 Columbia Road
Morristown
NJ
|
Family ID: |
37946928 |
Appl. No.: |
11/163388 |
Filed: |
October 17, 2005 |
Current U.S.
Class: |
73/49.2 |
Current CPC
Class: |
G01M 3/3281
20130101 |
Class at
Publication: |
073/049.2 |
International
Class: |
G01M 3/04 20060101
G01M003/04 |
Claims
1. A method for detecting a leak in a MEMS sensor package, wherein
the MEMS sensor package includes a sensor cavity that encloses a
MEMS sensor, wherein the MEMS sensor includes a mechanical
oscillating element with a Quality (Q) value, the method
comprising: measuring a first Q value of the MEMS sensor at a first
time; applying a pressure differential between the sensor cavity
and the exterior of the MEMS sensor package; and measuring a second
Q value of the MEMS sensor at a second time, wherein the second
time is later than the first time.
2. The method of claim 1 wherein the first Q value is measured
before the pressure differential is applied.
3. The method of claim 1 wherein the first Q value is measured
after the pressure differential is applied.
4. The method of claim 1 wherein the second Q value is measured
while the pressure differential is applied.
5. The method of claim 1 further comprising the step of removing
the pressure differential between the sensor cavity and the
exterior of the MEMS sensor package, and wherein the second Q value
is measured after the pressure differential is removed.
6. The method of claim 1 wherein the differential pressure applying
step includes the steps of: inserting the MEMS sensor into a
chamber; and pressurizing the chamber with a gas for a period of
time.
7. The method of claim 1 wherein the second Q value is measured
after the pressure differential between the sensor cavity and the
exterior of the MEMS sensor package has been applied for a period
of time.
8. The method of claim 7 wherein the period of time is less than
100 hours.
9. The method of claim 7 wherein the period of time is less than 50
hours.
10. The method of claim 7 wherein the period of time is less than
20 hours.
11. The method of claim 1 wherein the pressure differential applied
by the applying step is less than 200 psig.
12. The method of claim 1 wherein the pressure differential applied
by the applying step is less than 100 psig.
13. The method of claim 1 wherein the pressure differential applied
by the applying step is less than 50 psig.
14. The method of claim 1 wherein the pressure differential applied
by the applying step is less than 20 psig.
15. The method of claim 1 further comprising: determining a delta Q
value, which is the difference between the first Q value and the
second Q value; and comparing the delta Q value to a predetermined
delta Q value or a predetermined range of delta Q values.
16. The method of claim 15 further comprising: determining that the
MEMS sensor package does not meet a vacuum integrity requirement
when the delta Q value is greater than the predetermined delta Q
value or the predetermined range of delta Q values.
17. The method of claim 1 wherein the leak detected is less than
5.times.10-12 He atm.cc/s.
18. A method for detecting a leak in a MEMS sensor package, wherein
the MEMS sensor package includes a sensor cavity that encloses a
MEMS sensor, wherein the MEMS sensor includes a mechanical
oscillating element with a Quality (Q) value, the method
comprising: waiting for a period of time; measuring the Q value of
the MEMS sensor; and using the measured Q value to determine if the
MEMS sensor package has an unacceptable leak rate.
19. The method of claim 18 wherein the period of time is greater
than 10 hours.
20. The method of claim 18 wherein the period of time is greater
than 15 hours.
21. The method of claim 18 wherein the period of time is greater
than 20 hours.
22. The method of claim 18 further comprising: measuring the Q
value of the MEMS sensor at two or more different times; and using
two or more of the measured Q values to determine if the MEMS
sensor package has an unacceptable leak rate.
23.-26. (canceled)
27. The method of claim 18 further comprising the step of putting
the MEMS sensor package into a pressure chamber, and pressurizing
the pressure chamber, prior to the waiting step.
28. The method of claim 27 wherein the pressurized chamber is
pressurized with an inert gas.
29. The method of claim 27 wherein the pressurized chamber is
pressurized to greater than 100 psig.
30. The method of claim 27 wherein the pressurized chamber is
pressurized to greater than 100 psig for more than 10 hours.
Description
FIELD
[0001] The present invention relates generally to
micro-electro-mechanical systems (MEMS) sensors, and more
particularly, to methods and systems for detecting leaks in
packages that house a MEMS sensor that includes a mechanical
oscillator element.
BACKGROUND
[0002] Many MEMS sensors include a mechanical oscillator element.
For example, MEMS gyroscope and/or accelerometer type sensors often
include one or more proof masses, tuning forks or other oscillating
structures that are electrostatically driven at a resonance
frequency. Movements of the sensor housing, such as rotational
movement, lateral movement, acceleration, or other movement can
then be detected by sensing certain behavior in the oscillating
structure. For example, the oscillating structure may move in a
direction that is perpendicular to the oscillating direction due to
externally applied forces, such as coriolis forces, acceleration
forces, or other forces, depending on the application.
[0003] The operational performance characteristics of some MEMS
sensors, such as MEMS gyroscope or MEMS accelerometer type sensors,
are often related to the resonator Quality value (Q) of the sensor.
For example, the start-up time of the mechanical oscillator
element, the ring-down time, the sensitivity of the sensor, as well
as other performance characteristics are often affected by the Q
value of the sensor. The Q value of the sensor is dependent on a
number of factors, including the overall sensor design.
[0004] Known dampening mechanisms within the sensor can affect the
Q value of the sensor. One known dampening mechanism is dependent
on the energy lost due to collisions of the mechanical oscillator
element with gas molecules within the sensor cavity of the sensor
package. To reduce this dampening mechanism, and to obtain higher Q
values, such sensors are often packaged in a sensor cavity that is
under low pressure. Such sensor packages are often referred to as
vacuum packages, even though an absolute vacuum may not be
used.
[0005] The packages for many MEMS sensors often do not have perfect
seals, which results in gas leakage into or out of the sensor
cavity. Over time, these leaks can change the internal package
pressure, and thus may affect the Q value of the sensor. In some
cases, a relatively small leak can cause a relatively large change
in pressure in the sensor cavity, particularly over long periods of
time. For some applications, this can cause the sensor to cease to
operate in accordance with required design parameters after a
certain period of time.
[0006] Recently, there has been an increased demand for MEMS
sensors that have an extended useful life, such as 15 to 20 years.
For these and other applications, a MEMS sensor must have a small
enough leak rate so that the pressure in the sensor cavity does not
exceed some pressure limit over the expected lifetime of the
sensor. Currently, conventional methods for testing leak rates of
sensor packages are in the 5.times.10.sup.-12 He atm.cc/s range,
which is often not sensitive enough to test sensor packages with
expected lifetimes of 15 to 20 years. Therefore, there is need for
improved methods and systems for detecting leaks in packages that
house MEMS sensors, and in particular, MEMS sensors that have a
mechanical oscillating element.
SUMMARY
[0007] The following summary of the invention is provided to
facilitate an understanding of some of the innovative features
unique to the present invention and is not intended to be a full
description. A full appreciation of the invention can be gained by
taking the entire specification, claims, drawings, and abstract as
a whole.
[0008] The present invention relates generally to MEMS sensors, and
more particularly, to methods and systems for detecting leaks in
packages that house a MEMS sensor that includes a mechanical
oscillator element. In one illustrative embodiment, a method for
detecting a leak in a MEMS sensor package includes measuring a
packaged MEMS sensor parameter, such as the Q value, at a first
time, inserting the packaged MEMS sensor into a pressure chamber,
pressurizing the chamber (e.g. with a positive or negative
pressure) for a period of time, and then measuring the packaged
MEMS sensor parameter at a second time.
[0009] The method may further include determining a change in the
measured sensor parameter from the first time to the second time,
and comparing the change in the sensor parameter to a predetermined
value or range of values. A greater change in the sensor parameter
may indicate that the MEMS package does not meet vacuum integrity
requirements. In some cases, the sensor parameter may be dependent
on the pressure in the MEMS sensor package, such as the Q value of
the packaged MEMS sensor.
[0010] In some cases, the chamber may be pressurized before the
sensor parameter is measured at the first time. Also, the chamber
may be depressurized before the sensor parameter is measured at the
second time, or the pressure in the chamber may be maintained while
the sensor parameter is measured at the second time. The pressure
leak detected in the illustrative method may be less than, for
example, 5.times.10.sup.-12 He atm.cc/s.
[0011] In another illustrative embodiment, an apparatus for testing
the leak rate of a MEMS sensor package may include a MEMS sensor
situated in a MEMS sensor package, a chamber defined by chamber
walls that are sized to house the MEMS sensor package, a pump for
pressurizing the chamber (e.g. with a positive or negative
pressure). The apparatus may further include a controller for
measuring the sensor parameter of the MEMS sensor at a first time
and at a second time, where the measurements are spaced over a
period of time. The controller may detect a change in the sensor
parameter from the first time to the second time, and such change
may indicate if the MEMS sensor package meets certain vacuum
integrity requirements. In some cases, the sensor parameter may be
the Q value of the MEMS sensor.
BRIEF DESCRIPTION
[0012] FIG. 1 is a schematic diagram of an illustrative embodiment
of an apparatus for testing the leak rate of a MEMS sensor;
[0013] FIG. 2 is a flow diagram of an illustrative method for
detecting a leak in a MEMS sensor package;
[0014] FIG. 3 is a graph showing illustrative pressure rises in a
MEMS sensor package over time at various leak rates;
[0015] FIG. 4 is a graph showing predicted leak detector signals
for a MEMS sensor package over time at various leak rates during a
helium bomb test;
[0016] FIG. 5 is a graph showing illustrative pressure rises inside
a MEMS sensor package for various leak rates when exposed to a
pressurized container or chamber (e.g. during a helium bomb
test);
[0017] FIG. 6 is a graph showing the change in motor Q value and
sensor Q value as a function of the package pressure for an actual
MEMS gyro sensor;
[0018] FIG. 7 is a graph showing the change in pressure in an
illustrative MEMS sensor package over a period of time.
DETAILED DESCRIPTION
[0019] The following description should be read with reference to
the drawings wherein like reference numerals indicate like elements
throughout the several views. The detailed description and drawings
show several embodiments, which are meant to be illustrative of the
claimed invention.
[0020] In some applications, it is desirable to provide a MEMS
sensor with an expected useful life in the range of 10 to 20 years,
or greater. MEMS sensors for the automotive industry may be one
such application. To have a relatively long life, a MEMS sensor
package must typically maintain a pressure limit in a sensor cavity
over the expected useful lifetime of the MEMS sensor. For example,
and in one application, the pressure in the sensor cavity must be
less than 47 mTorr over the expected lifetime of the MEMs sensor.
Other applications may have higher or lower pressure limits for the
MEMS sensor package, as desired. In one case, the MEMS sensor
package may have a sensor cavity with a volume of 0.1420 cubic
centimeters (cc). The illustrative MEMS sensor may occupy some of
the volume of the MEMS sensor package, thus, the free volume of the
MEMS sensor package may be less than the total volume. In one case,
the remaining free volume of the MEMS sensor package may be 0.1305
cc. Thus, with this free volume, and in some cases, for a MEMS
sensor to have a relatively long life of 15 to 20 years, the leak
rate of the MEMS sensor package may need to be 2.times.10.sup.-13
He atm.cc/s or less.
[0021] To help extend the lifetime of the MEMS sensor, the MEMS
sensor package may also include a getter, however, this is not
required. The getter may absorb residual non-inert gas in the MEMS
sensor package and/or non-inert gases that leak into the MEMS
sensor package over time. This may help extend the life of the MEMS
sensor by reducing the pressure change inside the sensor cavity of
the MEMS sensor package. In some cases, the getter may be activated
by heat.
[0022] FIG. 1 is a schematic diagram of an illustrative embodiment
of an apparatus 10 for testing the leak rate of a MEMS sensor
package 14. The illustrative apparatus 10 includes a MEMS sensor 12
situated in a cavity 17 of the MEMS sensor package 14. The MEMS
sensor package 14 is shown positioned in a pressure chamber 16,
which is defined by chamber walls 19. The chamber 16 may be
pressurized with a gas to a desired pressure using, for example, a
pump 18. The pump 18 may be a mechanical pump, a gas cylinder, an
evacuated container or any other suitable device or pressure source
that can change the pressure in the pressure chamber 16. In some
cases, the pressure may be a pressure above atmospheric pressure,
while in other cases, the pressure may be a pressure below
atmospheric pressure.
[0023] The MEMS sensor 12 may be packaged and sealed in the sensor
cavity 17 of the MEMS sensor package 14, sometimes under a negative
pressure. A getter 15 may also be provided in the sensor cavity 17,
but this is not required. In some cases, the MEMS sensor package 14
and/or seal may include some small leaks. To detect the magnitude
of these leaks, if present, the MEMS sensor package 14 may be
situated in the chamber 16. The chamber 16 may be defined by
chamber walls 19, which may define a chamber space that is at least
sufficiently large to hold the MEMS sensor package 14. The chamber
16 may then be pressurized via pump 18, either positively or
negatively, depending on the application. In some cases, the
chamber 16 is pressurized with a pressure of gas.
[0024] The chamber 16 may have an inlet port 22 and/or an outlet
port 24 to facilitate the pressurization, and depressurization if
desired, of the chamber 16. In some cases, there may be a valve 24
situated in fluid communication with the inlet port 22 and/or a
valve 26 situated in fluid communication with the outlet port 24 to
help seal the chamber 16 during testing of the MEMS sensor package
14. Additionally, in some cases, the chamber 16 may have a cover
member or door (not shown). The cover member or door may allow the
inserting and removing of the MEMS sensor package 14 from the
chamber 16. More generally, it is contemplated that any suitable
chamber 16 that is capable of pressurizing the space around a MEMS
sensor package 14 may be used, as desired. In many cases, the
sensor cavity 17 of the MEMS sensor package 14 is back-filled to
some degree with an inert gas such as helium and/or argon. An inert
gas is often used because the getter 15, when provided, may absorb
non-inert gas. In some cases, the chamber 16 is positively
pressurized by pumping a gas, such as an inert gas, into the
chamber 16. More generally, however, it is contemplated that any
suitable gas may be used to positively pressurize the chamber 16,
as desired.
[0025] The MEMS sensor 12 may be any type of MEMS sensor. In some
cases, the MEMS sensor 12 includes a mechanical oscillator element.
For example, the MEMS sensor 12 may be a gyroscope and/or
accelerometer type sensor, and may include one or more proof
masses, tuning forks or other oscillating structures that are
electrostatically driven at a resonance frequency. In some cases,
the MEMS sensor 12 may have a quality value (Q), which is dependant
on the pressure in the cavity 17 of the MEMS sensor package 14.
[0026] Known dampening mechanisms within the MEMS sensor 12 and/or
MEMS sensor package 14 can affect the Q value of the sensor. One
known dampening mechanism is dependent on the energy lost due to
collisions of the mechanical oscillator element of the MEMS sensor
12 with gas molecules within the sensor cavity 17 of the MEMS
sensor package 14. To reduce this dampening mechanism, and to
obtain higher Q values, such MEMS sensors 12 are often packaged in
a sensor cavity 17 that is under low pressure. Such MEMS sensor
packages 14 are often referred to as vacuum packages, even though
an absolute vacuum may not be used.
[0027] As noted above, the MEMS sensor package 14 may not have a
perfect seal, which may result in gas leakage into or out of the
sensor cavity 17. Over time, these leaks can change the internal
package pressure in the sensor cavity 17 over time, and thus may
affect the Q value of the MEMS sensor 12. In some cases, a
relatively small leak can cause a relatively large change in
pressure in the sensor cavity 17, particularly over long periods of
time. For some applications, this can cause the MEMS sensor 12 to
cease to operate in accordance with required design parameters
after a certain period of time.
[0028] The Q value may be related to many performance
characteristics of the MEMS sensor 12, such as, for example, the
sensitivity of the MEMS sensor 12, the ring-down time of the
mechanical oscillator element, the start-up time of the mechanical
oscillator element, as well as others. For certain applications, it
may be desirable to have a desired Q value that provides a specific
performance characteristic for the MEMS sensor 12, often depending
on the application. Thus, a desired pressure or pressure range in
the sensor cavity 17 over the expected lifetime of the MEMS sensor
12 may be desirable.
[0029] A controller 23 may be provided, and may be electrically
coupled to the MEMS sensor package 14. The controller may be
capable of controlling the MEMS sensor 12, at least sufficiently to
determine a sensor parameter such as the "Q" value of the MEMS
sensor 12. In some cases, the controller 23 may measure the Q value
of the MEMS sensor at a first time. After pressure has been applied
for a period of time to the exterior of the MEMS sensor package via
the pressurized chamber 16, the controller 23 may again measure the
Q value of the MEMS sensor at a second time. A change in the
measured Q values between the first time and the second time may
indicate a leak, and the magnitude of the change in Q value may be
correlated to a leak rate for the particular MEMS sensor package.
In some cases, a leak rate of 2.times.10.sup.-13 He atm.cc/s or
less may be detected using this technique.
[0030] FIG. 2 is a flow diagram of an illustrative method for
detecting a leak in a MEMS sensor package. In some applications,
such as, for example, the automotive, space, aeronautic and other
industries, there may be a desire for MEMS sensors to have an
expected lifetime of between 15 and 20 years or more. To have this
relatively long life, the MEMS sensor package 14 must typically
prevent the pressure in the sensor cavity 17 from falling outside
some pressure limit or range. For example, for some MEMS gyroscope
sensors, the pressure inside the sensor cavity 17 should not exceed
some pressure limit, such as, for example, 47 mTorr. It has been
found that the illustrative method for detecting MEMS sensor
package leaks may be able to detect relatively small leaks, such
as, for example, 2.times.10.sup.-13 He atm.cc/s, or less. In some
cases, these relatively small leaks may, over time, increase the
pressure in the sensor cavity 17 of the MEMS sensor package 14 to a
pressure that is greater than the designated pressure limit (e.g.
47 mTorr or higher). Being able to detect these relatively small
leaks, such as, leaks of 2.times.10.sup.-13 He atm.cc/s, may help
determine in advance the expected lifetime of particular MEMS
sensors.
[0031] Being able to determine in advance the expected lifetime of
particular MEMS sensors may allow the MEMS sensors to be binned or
otherwise sorted. In some cases, MEMS sensors with shorter expected
lifetimes may be used or sold in applications that require shorter
lifetimes, while MEMS sensors with longer expected lifetimes may be
used or sold in applications that require longer lifetimes. In some
cases, this may increase the effective yield of the MEMS sensors
that are produced.
[0032] To detect leaks in the MEMS sensor package 14, a sensor
parameter of the MEMS sensor 12 that is housed by the MEMS sensor
package 14 may be used. One illustrative sensor parameter may be
the Q value of the MEMS sensor 12, as discussed above. As shown at
step 30 of FIG. 2, the sensor parameter (e.g. Q value) may be
measured at a first time. When the Q value is used, it is
contemplated that the Q value may be measured using any suitable
technique, including measuring the ring-down time, the start-up
time, or any other suitable parameter of the MEMS sensor 12, as
desired. Next, and as shown at step 32, the MEMS sensor package 14
may be placed in a chamber 16, and the chamber 16 may be
pressurized for a period of time. When the pressure inside the
sensor cavity 17 of the MEMS sensor package 14 is low, the chamber
16 may be pressurized in a positive direction relative to
atmosphere, thereby creating a greater pressure gradient across the
MEMS sensor package 14. This will tend to temporarily increase the
leak rate, if any, into the sensor cavity 17. In some cases, the
gas used to pressurize cavity 16 around the MEMS sensor package 14
may be a non-inert gas, as discussed previously.
[0033] In some cases, the period of time that the MEMS sensor
package 14 is pressurized is in the range of 2 to 120 hours, but
other times may also be used, depending on the circumstances. It is
contemplated the MEMS sensor parameter (e.g. Q value) may be
measured at the first time before the chamber 16 is pressurized,
after the chamber 16 is pressurized, during the pressurization, or
at any other time, as desired.
[0034] After a time period has elapsed following the first
measurement time, and after the MEMS sensor package 14 has been
pressurized in the chamber 16 for a period of time, the MEMS sensor
parameter (e.g. Q value) may again be measured at a second
measurement time, as shown at step 34. The MEMS sensor parameter
(e.g. Q value) may be measured using the same method as at the
first measurement time, if desired. In some cases, the cavity 16
may be depressurized prior to the measuring the sensor parameter at
the second measurement time. In other cases, the cavity 16 may be
maintained at an elevated pressurize level when measuring the
sensor parameter at the second measurement time.
[0035] Next, and as shown at step 36, a change in the MEMS sensor
parameter from the measurement at the first time to the measurement
at the second time may be determined. The change in the MEMS sensor
parameter may relate or correspond to a change in the pressure in
the sensor cavity 17 of the MEMS sensor package 14. In some cases,
the detected change in the sensor parameter may be compared to an
expected value for acceptable leak rates, as shown in step 38. A
change in the MEMS sensor package pressure that is greater than
that expected by an acceptable leak rate may indicate that the MEMS
sensor package 14 does not meet the vacuum integrity requirement
for the desired lifetime of the MEMS sensor.
[0036] In some cases, prior to measuring the sensor parameter at
the first measurement time, the MEMS sensor package 14 may be set
aside for a period of time. Once the period of time has elapsed,
the MEMS sensor parameter may be measured at the first measurement
time, and the measured sensor parameter (e.g. Q value) may be used
to detect relatively gross leak rates in the MEMS sensor package
14. In some cases, a Q value that is relatively low may be used to
detect gross leak rates in the MEMS sensor package. The MEMS sensor
package 14 may be set aside, or in some cases "quarantined", in a
pressurized chamber, or at atmosphere, depending on the
circumstances.
[0037] FIG. 3 is a graph showing illustrative pressure rises in a
MEMS sensor package 14 over time at various leak rates. The
illustrative graph shows the MEMS internal inert gas pressure in
the sensor cavity 17 for various leak rates. The initial residual
pressure in the sensor cavity after fabrication was assumed to be 1
mTorr. There are seven leak rates shown on the graph, ranging from
1.times.10.sup.-8 He atm.cc/s to 1.times.10.sup.-15 He atm.cc/s.
Line 302 shows a leak rate of 1.times.10.sup.-15 He atm.cc/s. Line
304 shows a leak rate of 1.times.10.sup.-13 He atm.cc/s. Line 306
shows a leak rate of 1.times.10.sup.-12 He atm.cc/s. Line 308 shows
a leak rate of 1.times.10.sup.-11 He atm.cc/s. Line 310 shows a
leak rate of 1.times.10.sup.-10 He atm.cc/s. Line 312 shows a leak
rate of 1.times.10.sup.-9 He atm.cc/s. Line 314 shows a leak rate
of 1.times.10.sup.-8 He atm.cc/s.
[0038] As illustrated, initially, the illustrative leak rates may
cause a relatively large percentage change in the MEMS sensor
package pressure. Over time, such as from 10 to 20 years, the
pressure change or pressure curves flatten out, and the percentage
change in pressure in the MEMS sensor package is relatively
smaller. Line 340 shows a pressure limit of 47 mTorr, which is one
illustrative maximum pressure limit for the cavity 17 of a MEMS
sensor package 14 in order for the MEMS sensor 12 to operate
properly. Thus, and using this pressure limit, the expected
lifetime of the MEMS sensor 12 is indicated by when the leak rate
line crosses the pressure limit line shown at 340. As can be seen,
leak rate lines 302 and 304 are always less than pressure limit
line 340, and thus are acceptable leak rates and the MEMS sensor
package 14.
[0039] From the illustrative graph, the highest leak rate that will
result in an expected sensor lifetime of 20 years is about
2.times.10.sup.-13 He atm.cc/s. However, due to the uncertainty in
the residual pressure of the MEMS sensor package after fabrication,
which may be anywhere between 1 mTorr up to 10 mTorr, and because
the maximum pressure limit (maximum pressure allowable) may range
from anywhere from 20 mTorr to 47 mTorr, the leak limit may be
anywhere between 2.times.10.sup.-13 He atm.cc/s to
2.times.10.sup.-14 He atm.cc/s in the illustrative case.
[0040] Commercial helium leak detectors are not capable of finding
these leaks, even when using a helium bomb test method. In a
typical helium bomb test, the sensor package is placed in a
container that is pressurized with helium at about two atmospheres
or higher. After a set time (usually 4 to 12 hours), the sensor
package is removed from the bomb and placed in a leak detector
canister. The total helium in the canister is then measured and
compared with a calculated signal for the leak specification limit.
Because the leak rate is a throughput measurement and dependent on
pressure difference, the leak detector signal for a given leak rate
may be given by the equation: R = LP E P O .times. ( 1 - e - Lt 1 /
VP O ) .times. e - Lt 2 / VP O ##EQU1##
[0041] where:
[0042] R is the leak signal in He atm.cc/s
[0043] L is the actual helium leak rate in He atm.cc/s
[0044] P.sub.E is the helium exposure pressure in atmospheres
[0045] P.sub.O is the atmospheric pressure in atmospheres
[0046] V is the package volume in cc
[0047] t.sub.1 is the helium exposure time in seconds
[0048] t.sub.2 is the dwell time after release in pressure in
seconds.
[0049] A graph of this equation for various leak rates L and
exposure times t.sub.1 is shown in FIG. 4. Line 402 shows a leak
detector signal that corresponds to a leak rate of
1.times.10.sup.-10 He atm.cc/s. Line 404 shows a leak detector
signal that corresponds to a leak rate of 1.times.10.sup.-9 He
atm.cc/s. Line 406 shows a leak detector signal that corresponds to
a leak rate of 1.times.10.sup.-8 He atm.cc/s. Line 408 shows a leak
detector signal that corresponds to a leak rate of
1.times.10.sup.-7 He atm.cc/s. Line 410 shows a leak detector
signal that corresponds to a leak rate of 1.times.10.sup.-6 He
atm.cc/s. Finally, line 412 shows a leak detector signal that
corresponds to a leak rate of 1.times.10.sup.-5 He atm.cc/s.
[0050] In a controlled lab environment, the smallest leak signal
that could be observed using a canister test is about
1.times.10.sup.-9 He atm.cc/s, shown at line 440. However, the
background noise would most likely be higher in a production
environment, so a reasonable leak detector signal that could be
observed in production may be about 1.times.10.sup.-8 He atm.cc/s,
shown at line 442. In either case, a leak of 1.times.10.sup.-8 He
atm.cc/s cannot be detected in a reasonable amount of time. This
can be explained by considering the amount of helium entering a
sensor package during helium bombing. Because partial pressure
difference is what drives the movement of helium through the leak,
more helium can get into the package cavity over a short time (e.g.
200 psi difference) than can get out (e.g. less than 1 mTorr
difference). Thus, to detect a leak of 2.times.10.sup.-13 He
atm.cc/s, a leak detector capable of detecting 1.times.10.sup.-18
He atm.cc/s may be needed and currently, the best commercial helium
detectors are rated only for about 5.times.10.sup.-12 He atm.cc/s.
This illustrates the short comings of the conventional helium bomb
test.
[0051] FIG. 5 is a graph showing illustrative pressure rises inside
a MEMS sensor package for various leak rates when exposed to a
pressurized container or chamber (e.g. during a helium bomb test).
In the illustrative graph, a MEMS sensor package 14 is exposed to
200 psig of helium for a period of time. As depicted in the graph,
there are multiple leak rates ranging from 1.times.10.sup.-13 He
atm.cc/s to 1.times.10.sup.-5 He atm.cc/s. Line 502 corresponds to
a leak rate of 1.times.10.sup.-13 He atm.cc/s. Line 504 corresponds
to a leak rate of 1.times.10.sup.-12 He atm.cc/s. Line 506
corresponds to a leak rate of 1.times.10.sup.-11 He atm.cc/s. Line
508 corresponds to a leak rate of 1.times.10.sup.-10 He atm.cc/s.
Line 510 corresponds to a leak rate of 1.times.10.sup.-9 He
atm.cc/s. Line 512 corresponds to a leak rate of 1.times.10.sup.-8
He atm.cc/s. Line 514 corresponds to a leak rate of
1.times.10.sup.-7 He atm.cc/s. Line 516 corresponds to a leak rate
of 1.times.10.sup.-6 He atm.cc/s. Line 518 corresponds to a leak
rate of 1.times.10.sup.-5 He atm.cc/s. The pressure limit of the
MEMS sensor package 14 in the illustrative graph is shown at 47
mTorr, as shown by line 540.
[0052] The Q value for a MEMS sensor having an oscillating
structure or element is dependent on the pressure in the cavity
that houses the MEMS sensor. FIG. 6 is a graph showing the change
in motor Q value and sensor Q value as a function of the package
pressure for an actual MEMS gyro sensor. The sensor Q value
represents the sensitivity of the sensor, while the motor Q value
represents the Q value of the oscillating structure as it is
driven. The graph shows the change in motor Q at line 604 and the
change in sensor Q at line 602.
[0053] As can be seen, as the pressure in the cavity 17 of the MEMS
sensor package 14 increases, the change in the respective Q values
also increase. Furthermore, the illustrative pressure limit of 47
mTorr is shown at line 640. In the illustrative case, the total
change in Q value if the pressure limit is reached is in the range
of 35,000 to 45,000.
[0054] The repeatability of the Q value measurements is about 0.2%,
so a pressure change near 1 mTorr is needed for a substantial
measurable Q shift. A 1 mTorr change in pressure in the sensor
cavity 17 of a MEMS sensor package 14 may require the following
bomb conditions: TABLE-US-00001 Bomb Pressure Time to 1 mTorr with
Package 2E-13 Leak 2 atm (14.7 psig) 116 hours 3 atm (29.4 psig) 78
hours 4 atm (44.1 psig) 58 hours 5 atm (58.8 psig) 47 hours 6 atm
(73.5 psig) 39 hours 14.6 atm (200 psig) 16 hours
[0055] As can be seen, a substantial shift in Q value may be
detected after bombing for only 16 hours at 200 psig if the MEMS
sensor package had a 2E-13 He atm.cc/s leak rate.
[0056] In some cases, and prior to measuring a sensor parameter
(e.g. Q value) at a first measurement time, the MEMS sensor package
14 may be set aside for a period of time. Once the period of time
has elapsed, the MEMS sensor parameter may be measured at the first
measurement time, and the measured sensor parameter (e.g. Q value)
may be used to detect a relatively gross leak rate in the MEMS
sensor package 14. In some cases, the MEMS sensor package 14 may be
set aside, or in some cases "quarantined", in a pressurized
chamber, or at atmosphere, depending on the circumstances.
[0057] FIG. 7 is a graph showing the change in pressure in an
illustrative MEMS sensor package over a period of time. As
discussed previously, gross leaks may be detected by setting aside
the MEMS sensor package 14 for a period of time prior to measuring
the Q value at the first time. In the illustrative graph, the MEMS
sensor package 14 is placed in a chamber 16 and pressurized with
200 psig of helium. The change in pressure is shown at line 802,
and a pressure limit of 47 mTorr is shown at line 804. If the
illustrative MEMS sensor package has a getter, and the getter is
activated so there is only inert gas in the sensor cavity 17, the
pressure in the sensor cavity 17 may exceed 47 mTorr in 18 hours if
it has a 4.2.times.10.sup.-8 He atm.cc/s leak rate. If the MEMS
sensor package does not have a getter, or the getter is not active,
then the pressure in the sensor cavity 17 may exceed 47 mTorr in 18
hours if it has a 3.4.times.10.sup.-10 He atm.cc/s leak rate.
[0058] It is contemplated that the MEMS sensor package 14 may be
set aside for a period of time during the
manufacturing/assembly/testing process before the Q value is
measured at the first or subsequent time. That is, the MEMS sensor
package 14 may be provided on a shelf, inserted into a pressure
chamber, or otherwise stored for a period of time during the
manufacturing/assembly/testing process, before the Q value is
measured at the first or subsequent time. Alternatively, or in
addition, the MEMS sensor package 14 may be shipped and used for a
period of time, before measuring the Q value at the first or
subsequent time. In this latter case, a leak test may be performed
in the field, which may help check the seal integrity of fielded
devices, which may provide some insight during failure analysis
testing. A reason to wait for a period of time before measuring the
Q value at the first or subsequent time may be to allow an
increased pressure change inside the MEMS sensor package 14, which
may be easier to detect. Smaller leak rates than that shown in FIG.
7 may be detected in older sensors.
[0059] Having thus described the preferred embodiments of the
present invention, those of skill in the art will readily
appreciate that yet other embodiments may be made and used within
the scope of the claims hereto attached. Numerous advantages of the
invention covered by this document have been set forth in the
foregoing description. It will be understood, however, that this
disclosure is, in many respect, only illustrative. Changes may be
made in details, particularly in matters of shape, size, and
arrangement of parts without exceeding the scope of the invention.
The invention's scope is, of course, defined in the language in
which the appended claims are expressed.
* * * * *