Cross-Reference to Related Applications
This application is claiming priority from
U.S. Provisional Patent Application No. 60/783,448, filed on March 20, 2006
, entitled COMPENSATING ACCELEROMETER WITH OPTICAL ANGLE SENSING, which
is incorporated by reference herein in its entirety.
Background of the Invention
Field of the Invention
The present invention relates to measurement equipment,
and more particularly to compensating pendulum-type linear accelerometers with flexible
suspension of the sensing element.
Description of the related art
An example of a conventional accelerometer is illustrated
Japanese Patent No. 188924/1981
. This accelerometer includes an inertial mass (in this case, a pendulum)
that is placed in a sealed housing. Inside the sealed housing there is typically
a vacuum or an inert gas, such as helium. The inertial mass is generally in the
shape of a movable rod, whose lower portion is attached, using flexible suspension,
to a plate. The plate is then affixed, using screws, to a base surface of one of
the frames. The rod can rotate, using the flexible suspension, around the plate
in the direction of the measurement axis of the accelerometer. The rod of the pendulum
has two cylindrically shaped coils attached to it.
The axial magnetic systems of the accelerometer include
magnetic conductors, in this case first and second frames, which are made of magnetically
soft material. The magnetic axial system also include permanent magnets and field
concentrators, such that the movable coils are located in the gaps formed between
the surfaces of the magnetic conductors and the field concentrators. The above elements
together with the movable coils comprise the momentum sensor of the accelerometer.
Both frames are rigidly connected to each other, and each
of the frames includes stoppers for limiting the range of motion of the pendulum.
This range of motion can be regulated by moving the stoppers in the threaded openings
in the frames.
Inside the housing there is an angle sensor that measures
the movement of the pendulum. The angle sensor consists of a single light source
and two light detectors. The pendulum is located between the frames, such that the
light source is located on one side of the pendulum, and the light detectors are
located on the other side of the pendulum. Thus, the pendulum acts as a shading
element that blocks the light from the light source to the light detectors. When
there is no measured acceleration along the measurement axis, the pendulum is in
a neutral position, and some of the light from the light source is shielded by the
pendulum, while the remainder of the light is evenly distributed between the two
light detectors. Thus, the current produced by the light detectors (if, for example,
photodiodes are used) is equal. Note that the surfaces of the photodiodes (light
detectors) need not be entirely illuminated, but so long as the illumination of
each detector is the same, the output currents are the same, indicating that the
measured acceleration is zero.
When the sensor experiences acceleration along the measurement
axis, the pendulum is displaced from its neutral position due to inertia force.
As a result, the light distribution between the two detectors changes, and therefore
the relationship between the current from the two detectors also changes. The difference
between the two currents is related to the acceleration, and using conventional
electronics, can be converted to an acceleration value. The current can also be
used in a feedback circuit; it passes through the movable coils interacts with the
magnetic field of the permanent magnets and returns the pendulum to its neutral
position. The magnitude and polarity of the feedback current therefore permit a
measurement of the acceleration.
The conventional accelerometer described above has a number
of disadvantages. One of the disadvantages is that due to the relatively large dimensions
of the rod of the pendulum, the light detectors need to be placed relatively far
apart, which tends to increase noise and reduces the sensitivity of the measurement.
Another disadvantage is that the neutral position of the pendulum typically does
not precisely correspond to a zero output signal of the accelerometer. Yet another
disadvantage is that mechanical tuning of the accelerometer tends to be difficult,
particularly with regards to the zero bias signal.
Another example of a conventional accelerometer is described
U.S. Patent No. 4,649,748
. The accelerometer described in this patent shares some commonalities
with the earlier-described accelerometer. However, the accelerometer in
U.S. Patent No. 4,649,748
has the following differences compared to the device of
Japanese Patent No. 188924/1981
: first, the first frame - the magnetic conductor of the magnetic system
- is used as a mounting base of the accelerometer, which permits simplifying the
manufacturing of the accelerometer, and also reduces its mass.
Additionally, the free end of the pendulum rod, which is
located between the light source and the light detectors, is formed as a thin plate
which permits the light source and the light detectors to be closer to each other.
This increases the sensitivity of the angle sensor, and reduces the noise in its
output. One of the options is for the thin plate to also have the shape of a rod,
another option is to have the plate shaped as a thin rod with a slit along the axis
of the rod, such that part of the light passes through the slit.
The various elements of the overall construction where
the light source and light detectors are housed are pressed against the frames using
a flat spring, which increases the precision of the mechanical tuning of the zero
The device described in
U.S. Patent No. 4,649,748
has a number of disadvantages. For example, one of the disadvantages is
due to the use of two magnetic systems. In case of need to increase range of measure
acceleration this leads to a relatively values for currents that might flow through
the coils, which in turn leads to a large amount of heat being dissipated, which
in turn leads to an error source in the measurement, relating to the waste heat.
Additionally, the use of two different magnetic systems increases the manufacturing
cost and complexity of the device. These disadvantages are also present in the device
Japanese Patent No. 188924/1981
, discussed above.
Furthermore, the mechanical tuning of the zero bias signal
is relatively coarse. Furthermore, a relatively small range of displacement of the
light source and detectors relative to the housing makes it difficult to tune the
angle sensor when there is a relatively large angle between the axis of the pendulum
rod in the neutral position, and the base mounting surface of the accelerometer.
Summary of the Invention
The present invention relates to compensating accelerometers
that substantially obviates one or more disadvantages of the conventional accelerometers.
More particularly, in an exemplary embodiment of the present
invention, a compensating accelerometer includes a housing, and a sensing element
mounted in the housing. The sensing element includes a pendulum flexibly suspended
on a base. The pendulum includes a coil mounted on a movable thin plate, a curtain
having a slit, and a load mass affixed to the curtain. An angle sensor within the
housing includes the curtain, a fork, a light source and a differential light detector.
A momentum sensor within the housing includes a permanent magnet, an inner magnetic
conductor, an outer magnetic conductor and the coil. The permanent magnet is magnetized
in a direction of an axis of the pendulum. A stopper is used to adjust position
of the fork. A spring is on opposite side of the fork, for taking up slack in the
fork. A fixator has an eccentric, for adjusting a gap between the curtain and the
Additional features and advantages of the invention will
be set forth in the description that follows, and in part will be apparent from
the description, or may be learned by practice of the invention. The advantages
of the invention will be realized and attained by the structure particularly pointed
out in the written description and claims hereof as well as the appended drawings.
It is to be understood that both the foregoing general
description and the following detailed description are exemplary and explanatory
and are intended to provide further explanation of the invention as claimed.
Brief Description of the Figures
The accompanying drawings, which are included to provide
a further understanding of the invention and are incorporated in and constitute
a part of this specification, illustrate embodiments of the invention and together
with the description serve to explain the principles of the invention. In the drawings:
FIG. 1 illustrates a cross-sectional view of the accelerometer of one embodiment
of the invention, with the view shown along the sensitivity axis SA and pendulum
FIG. 2 shows the view in the direction of arrow A in FIG. 1,
with element 21 removed.
FIG. 3 shows the view in the direction of arrow B in FIG. 1,
with element 21 removed.
FIGs. 4A and 4B illustrate the sensing element.
FIG. 5 illustrates a functional electrical circuit diagram of the accelerometer.
FIG. 6A and 6B illustrates the principle of operation of the sensing
element of the exemplary accelerometer of FIG. 1.
FIG. 7 illustrates the differences between the described accelerometer and
Detailed Description of the Preferred Embodiments
Reference will now be made in detail to embodiments of
the present invention, examples of which are illustrated in the accompanying drawings.
FIGs. 1-3, 4A and 4B illustrate an exemplary embodiment of the accelerometer
of the present invention. FIG. 1 illustrates a cross-sectional view of the
accelerometer of one embodiment of the invention, with the view shown along the
sensitivity axis SA and pendulum axis PA. FIG. 2 shows the view in
the direction of arrow A in FIG. 1, with element 21 removed.
FIG. 3 shows the view in the direction of arrow B in FIG. 1,
with element 21 removed. FIGs. 4A and 4B illustrate the sensing
The accelerometer includes a housing 1, with a sensing
element located in its upper portion. The sensing element includes an inertial mass,
formed by a movable plate 3, a coil 8 mounted on the plate
3, a curtain 14 that has a slit 26, with the slit oriented
along the axis PA of the pendulum and a load mass 13, which is rigidly
connected to the curtain 14. (As an option, the load mass 13 and the
curtain 14 can be manufactured as a single element.) The sensing element
also includes a fixed base 4 and a flexible suspension 5, used to
attach the plate 3 to the base 4. Attachments 2 are used to
connect the coil 8 to the plate 3. The base 4 is fixedly mounted
to the upper portion of the housing 1, for example, using a screw
6 and a spring washer 7.
The housing 1 includes a differential optical angle
sensor and a momentum sensor. The momentum sensor includes a magnetic system and
the coil 8. The magnetic system is located in the upper cylindrical part
of the housing 1, and includes an outer magnetic conductor 10, an
inner magnetic conductor 11, and a ring-shaped magnet 12, which has
a magnetization direction along the axis PA of the sensing element. The ring-shaped
magnet 12 is typically a permanent magnet, although an electromagnet can
also be used. The suspension axis HA of the sensing element is located in
the neutral plane of the permanent magnet 12. The coil 8 is located
in the gap between the inner surface of the outer magnetic conductor 10 and
the outer surface of the permanent magnet 12.
The optical angle sensor includes a fork 23. A light
source 24 (e.g., an LED) and a two-element photodetector 25 are mounted
in the fork 23. The photodetector can be, for example, a photodiode. The
curtain 14, with the slit 26, is mounted on the plate 3 of
the sensing element. The lower portion of the housing 1 includes a threaded
opening 20, in which a threaded stopper 19 can be placed. Also, the
housing 1 includes an opening 27 with a thread 28 in its lower
part. A fixator 9 that includes an eccentric 29 can be located in
the opening 27. The fork 23, with the light source 24 and the
photodetector 25 mounted on it, is arranged in the housing 1 in a
manner that permits it to move up and down along guides 18, and to ultimately
be affixed relatively to the housing 1 by using the screws 22.
The lower surface of the fork 23 contacts against
the stopper 19. The upper surface of the fork 23 contacts against
two springs 16, which abut against a flange 15. The flange
15 is fixed to the housing using screws 17. When the stopper
19 moves in the opening 20, the fork 23 moves along the axis
SA along the guides 18. The springs 16 are used to take up
any slack in the movement of the fork 23.
A casing 21 can be attached to the housing
1, to protect the accelerometer from external influences. The housing
1 also includes a base mounting surface C, which is perpendicular
to the measurement axis SA. The functional electrical schematic of the accelerometer
is show in FIG. 5. The accelerometer includes an optical angle sensor
41, a correcting element 42, an amplifier 43, coils
8, a resistor 45 and a filter 46, as shown in FIG. 5.
UOP is the output voltage of the accelerometer. The angle sensor
41 includes a light emitting diode 24, a two-element photodiode
25, a curtain 14 with a slit 26, preamplifiers 47 and
48, summers 49 and 50, a current regulator 51, and a
reference voltage source 44 that outputs a reference voltage UREF,
all connected as shown in FIG. 5.
The accelerometer works as follows: when no acceleration
a is applied along the axis SA, the inertial mass of the sensing element
is in a neutral position. Therefore, the curtain 14 is also in a neutral
position relative to the LED 24 and the dual element photodiode
25. In the neutral position, when the zero bias signal is being tuned mechanically,
the illuminated areas of the two-element photodiode 25 are equal, and each
is equal to half of the active area.
Each of the elements of the photodiode 25 generates
an electric current, whose value is proportional to the area receiving light from
the LED 24, as well as to the intensity of the light. In the neutral position,
the areas receiving light and the intensities are all equal, therefore, the output
voltages at the input of the preamplifiers 47 and 48 also equal, and
proportional to the photo-induced currents. In the summer 49, these voltages
are subtracted, which results in a zero output voltage, which also is the zero output
of the angle sensor. When the acceleration a is applied along the axis
SA, the inertial mass, due to the moment of inertia force M
i is deflected, due to the flexible suspension 5, relative to
the housing 1 of the accelerometer. If the acceleration a is positive, the
inertial mass is displaced towards the surface C, and if it is negative,
away from the surface C.
The curtain 14 also moves, together with the inertial
mass, relative to the LED 24 and the photodiode 25. The movement of
the curtain 14 results in one of the elements of the photodiodes having being
exposed to more light, and the other one being exposed to less light. Therefore,
the voltages at the outputs of the preamplifiers 47 and 48 also change,
the output voltage at the output of the summer 49 becomes non-zero, and its
value is proportional to the deflection of the inertial mass, while its sign corresponds
to the direction of the deflection. Afterwards, the voltage passes through the correction
element 42 and the amplifier 43, whose output, in the form of a negative
feedback signal, is supplied to coil 8. The current IOP through
the coil 8 interacts with the magnetic field of the permanent magnet
12, and creates a feedback moment MF around the suspension axis
HA. This balances out the moment of inertia force Mi of the measured
acceleration, and returns the inertial mass to the neutral position. The current
IOP, which passes through the resistor 45, creates a voltage UOP
= IOP R, where R is the resistance of the resistor 45.
This voltage, after passing through the filter 46, is the output signal of
The eccentric 29 of the fixator 9 is located
in the slit 30 of the base 4, and is intended for regulating the gap
31 between the curtain 14 and the photodiode 25, as shown in
FIGs. 2 and 3. The size of the gap 31 therefore is related to the
areas of the photodiode 25 which receive light from the LED 24. At
a nominal value of the gap 31, in the neutral position of the inertial mass,
the area irradiated by the light from the light source 24 should be equal
to one half of the total area. Rotating the fixator 9, for example, using
a screwdriver, leads to a rotation of the axis PA relative to the measurement
axis SA. At the same time, there is a change in the gap 31 between
the curtain 14 and the photodiode 25. Therefore, the gap
31 can be tuned until its value reaches the sought nominal value.
Mechanical tuning of the zero bias signal of the accelerometer
is performed by moving the fork 23 along the guides 18, which moves
the fork 23
along the axis SA. To mechanically tune the zero bias signal the accelerometer
is arranged in such a manner that the inertial mass of the sensing element is located
in the neutral position. The movement of the fork 23 is achieved by moving
the stopper 19, which contacts the fork 23, along the thread
20. Together with a fork 23, the LED 24 and the photodiode
25 also move relative to the slit 26 of the curtain 14, which
is positioned in the neutral position of the inertial mass of the accelerometer.
Due to this movement, the polarity of the current IOP also changes, which
means that the sign of the output at the summer 49, representing the output
signal of the accelerometer, also changes.
The position of the fork 23 relative to the guides
18, when the output voltage of the summer 49 is zero, is the position
when the mechanical tuning of the zero bias signal is performed. In this position,
the fork 23 is fixed in place along the guides 18 by using the screws
When the accelerometer is functioning, its temperature
can change in a relatively broad range, which can lead to a change in the parameters
of its various electronic and mechanical components, as shown in FIG. 5.
To stabilize the transfer coefficient of the angle sensor when the temperature changes,
an additional feedback circuit is used. The feedback circuit includes a summer
50, an LED current regulator 51, and a reference source
44. Using the summer 50, the sum of the voltages at the outputs of
the preamplifiers 47 and 48 is compared with the reference voltage
UREF from the reference voltage source 44.
When the transfer coefficient of the angle sensor is equal
to a nominal value (i.e., the value calculated based on sensor characteristics and
geometry), the voltage at the output of the summer 50 is equal to zero, and
the regulator 51 maintains the current at the LED 24 unchanged. For
example, when the transfer coefficient of the angle sensor 41 increases,
the voltage at the outputs of the preamplifiers 47 and 48 also increases,
the voltage at the output of the summer 50 becomes negative, which leads
to a reduction in the current through the LED 24,
and therefore a reduction in the light detected by the two elements of the photodiode
25. Therefore, the transfer coefficient of the angle sensor 41 reduces
to its nominal value. Thus, the transfer coefficient of the angle sensor is stabilized,
and the temperature-caused variation in the output of the accelerometer is reduced.
FIGS. 6 and 7 explain the differences between the accelerometer described
herein and conventional accelerometers. For the accelerometer described herein,
with the variables: an inertial mass m&Sgr;, a momentum Mi
due to the inertia force Fi, feedback momentum MF and a feedback
current IOP. Here, the index "I" refers to the accelerometer of the present
invention, and the index "II" refers to the conventional accelerometer.
where mB, LB - mass and center of
mass coordinate of the left portion ("B") (relative to the axis HA) of the
sensing element (i.e., half of the plate 3, and half of the coil
8); mA, LA - mass and center of mass coordinate of
the right portion (circled "A") (relative to the axis HA) of the sensing
element, whose mass consists of the mass mB and the mass of the load
13, together with the curtain 14; FiB, FiA -
inertia forces of the acceleration of the portions "A" and "B" of the inertial mass;
Fim - inertia force of the load mass 13 and the curtain
14; Lm - center mass coordinate of the load mass 13 with
the curtain 14; B is the induction in the working gap of the magnetic system
of the momentum sensor; rK - radial dimension of the coil 8, Lw
is the length of the wire of the coil 8.
The product mLm is called the pendulum value
of the accelerometer. When current IOP 1 flows through the coil
8, the amount of power PI is the waste heat generated in the coil
8, defined by the expression
where RK - resistance of the coil 8; &rgr;, Lw,
S - wire resistivity, wire length and wire cross-section of the coil
For the conventional accelerometers described above, the
parameters below are designated by the index "II," and are defined by the geometry
and construction of those accelerometers as follows:
1 II - mass of the rod of the inertial mass; m
2 II - mass of both coils mounted on the rod of the inertial mass. In
this design of the sensing element, in order to increase the maximum acceleration
that can be measured, it is desirable to have the mass of the rod of the proof mass
1 II to be substantially less than the mass m
2 II of the coils, therefore, in equation 6, it is assumed that
the mass of the entire inertial mass is substantially defined by the mass of the
coils; &ggr;, Lw, S - wire density, total wire length and wire cross
section of the coils, respectively; Lm - distance to the inertial mass
center; LF - distance to the point of application of the restoring force,
created by the momentum sensor (which coincides with the axis of symmetry of the
coils); B - magnetic induction in the working gap of the magnetic systems of the
The product m
is called the pendulum value of the conventional accelerometers. Similar to
the equations described above, the feedback current IOP II also generates
waste heat, designated by Pll, which is given by
is the total resistance of the coils of the momentum sensor, connected in series;
&rgr;, LW, S are resistivity, length and cross section of the wire
of the coils, respectively.
Based on equations 1-11, we can compare the conventional
accelerometer and the accelerometer described herein. For purposes of comparison,
it can be taken that all the accelerometers have the same maximum acceleration amax
that they can measure, the same pendulum value
which would be the case if they had the same distance from the axis HA to
their center masses Lm and the same wire length LW of the
wires in the coil 8 and in two coils in the conventional accelerometer. Then,
when the following condition is true
the values of the maximum feedback current occurs at the maximum acceleration IOP
(amax), which flows in the coils and will be equal, as well as the amount
of waste heat dissipated due to the current. Therefore, the maximum temperature
overheating of magnetic systems is due to the feedback current IOP approximately
equal and leads to approximately equal multiplicative error of accelerometer (depending
on the measured acceleration a).
Consider the situation where it is necessary to increase
the maximum acceleration amax that the accelerometer needs to measure,
by a factor of n, without increasing the pendulum value of the accelerometer (since
reducing the pendulum value leads to a decrease in sensitivity of the accelerometer
and a greater error in the measurement).
In the conventional accelerometer, based on equation
10, the current IOP II(amax), flowing through the coils,
will increase by a factor of n as well, which means that, based on equation
11, the dissipated waste heat PII(amax) in the momentum
sensor will increase by a factor of n2. Therefore, the temperature of
the permanent magnets will also increase as the square of the current increase.
However, for the conventional accelerometer, the current IOP II(amax)
is limited by the cross-section S of the wires of the coils of the momentum sensor.
In the proposed accelerometer, when amax increases
by a factor of n, the length of wire in the coil 8 can increase by a factor
of n, without changing the pendulum value of the accelerometer, since the coil
8 is mounted symmetrically on the plate 3 relative to the axis
HA. Increasing the length of the wire in the coil 8 does not lead
to a change in pendulum value - the pendulum value is determined only the load mass
13 and the curtain 14.
When the length of the wire in coil 8 increases
by a factor of n, based on equations 3 and 4, the maximum compensating
moment MF I will also increase by n, balancing out the maximum inertial
moment Mi I(amax) without an increase in the feedback current
IOP I(amax) that flows in the coil 8. This means that,
based on equation 5, the maximum waste heat dissipation PI(amax)
increases only by a factor of n, not n2. Therefore, the overheating of
the permanent magnet will also increase only by a factor of n, not n2,
and the errors in the measurement have a linear dependence on amax, rather
than a non-linear dependence.
For the proposed accelerometer, there is no theoretical
limit on amax, since, based on equation 4, even when amax
increases very significantly, the length of the wire in the coil 8 can be
increased very significantly as well, so that IOP I (amax)
will not be greater than the maximum allowed based on the cross-section S. Thus,
the proposed accelerometer has a significant advantage in applications that require
a large amax.
Having thus described an embodiment of the invention, it
should be apparent to those skilled in the art that certain advantages of the described
method and apparatus have been achieved. It should also be appreciated that various
modifications, adaptations, and alternative embodiments thereof may be made within
the scope and spirit of the present invention. The invention is further defined
by the following claims.