ADXL276QC38 [ADI]
IC SPECIALTY ANALOG CIRCUIT, CDSO14, CERAMIC, SMT-14, Analog IC:Other;型号: | ADXL276QC38 |
厂家: | ADI |
描述: | IC SPECIALTY ANALOG CIRCUIT, CDSO14, CERAMIC, SMT-14, Analog IC:Other CD |
文件: | 总12页 (文件大小:181K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
Single and Dual Axis
Automotive iMEMS® Accelerometers
a
ADXL76, ADXL269, ADXL276
FEATURES
FUNCTIONAL BLOCK DIAGRAMS
Complete Acceleration Measurement System on a
Single Monolithic IC
2-Pole Filter On-Chip
V
2
S
ADXL76
+V
S
No External Components Required for Direct Interface
to ADC or Controller
Complete Mechanical and Electrical Self-Test on Digital
Command
Sensitivity and Offset Are Ratiometric to Supply
5 V Supply Voltage with Operation Down to 4 V
Sensitive Axes in the Plane of the Chip
Linearity (0.2% of Full-Scale)
OSCILLATOR
AMP
5k⍀
V
FILTER
OUT
SENSOR
DEMODULATOR
25k⍀
SELF-TEST
ZERO-g ADJUST
COMMON
DC Response
Low Noise (1.0 mg/√Hz)
+X ZERO-g ADJUST
25k⍀
Low-Power Consumption (1.8 mA per Axis)
؎35 g, ؎50 g Ranges Available (for ؎125 g or ؎50 g
Contact Factory)
+V
S
ADXL269
SENSOR
AMP
DEMODULATOR
FILTER
V
V
(+X)
(–X)
–
+
Surface Mount Package Available
OUT
5k⍀
5k⍀
V
2
S
OSCILLATOR
AMP
GENERAL DESCRIPTION
FILTER
OUT
SENSOR
The ADXL76, ADXL276, and ADXL269 family of products are
the third generation surface micromachined iMEMS accelero-
meters from Analog Devices with enhanced performance and lower
cost. Designed for use in front and side impact airbag applica-
tions, these products also provide complete cost-effective solutions
useful for a wide variety of other applications.
DEMODULATOR
25k⍀
+
–
SELF-TEST
–X ZERO-g ADJUST
COMMON
Y ZERO-g ADJUST
25k⍀
The ADXL276 is a monolithic two-axis (XY) version of the
ADXL76 with the sensor axes orthogonal (90°) to each other
and in the plane of the chip. It can be used for sensing crashes in
the front or side of the vehicle and can be used to determine the
angle of impact. The ADXL269 is a monolithic two-axis (+X/–X)
version of the ADXL76 with the sensor axes arranged in anti-
parallel orientation (180°) to provide redundancy and elimination
of the need for an external mechanical safing/arming sensor.
+V
ADXL276
S
AMP
DEMODULATOR
FILTER
V
V
OUT (Y)
5k⍀
5k⍀
V
2
S
OSCILLATOR
AMP
FILTER
OUT (X)
The ADXL76, ADXL269, and ADXL276 are temperature stable
and accurate over the automotive temperature range, with a self-
test feature that fully exercises all the mechanical and electrical
elements of the sensor with a digital signal applied to a single pin.
SENSOR
DEMODULATOR
25k⍀
SELF-TEST
X ZERO-g ADJUST
COMMON
i
MEMS is a registered trademark of Analog Devices, Inc.
REV. 0
Information furnished by Analog Devices is believed to be accurate and
reliable. However, no responsibility is assumed by Analog Devices for its
use, nor for any infringements of patents or other rights of third parties
which may result from its use. No license is granted by implication or
otherwise under any patent or patent rights of Analog Devices.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 781/329-4700
Fax: 781/326-8703
World Wide Web Site: http://www.analog.com
© Analog Devices, Inc., 2000
ADXL76, ADXL269, ADXL276–SPECIFICATIONS
(@ TA = –40؇C to +85؇C, VS = 5 V ؎ 5%, Acceleration = 0 g; unless otherwise noted)
ADXLxxQx
ADXLxxQx
ADXLxxQx
551
181
381
Parameter
Conditions
Min
Typ
Max Min
Typ
Max Min
Typ
Max Unit
SENSOR
Guaranteed Full-Scale Range
ADXL76 Only, Offset Adjusted2
ADXL76 Only, Offset Adjusted
115
105
g
g
50
45
35
30
g
g
Nonlinearity
0.2
1
%
Degrees
Package Alignment Error3
Sensor-to-Sensor Alignment Error ADXL276, ADXL269
0.1
2
Degrees
%
Transverse Sensitivity4
SENSITIVITY
Sensitivity
18
38
55
11
mV/g
%
%
Sensitivity Tolerance5, 6
Temperature Drift
Ratiometric
From 25°C to TMIN to TMAX
–8
+8
0.5
ZERO-g OFFSET LEVEL
Output Zero-g Voltage6, 7
Temperature Drift
Offset from VS/2
From 25°C to TMIN to TMAX
–198
+198 –418
+418 –605
+605 mV
3.6
7.6
mV
ZERO-g ADJUSTMENT
Voltage Gain
Input Resistance
∆VOUT/∆VADJ PIN
0.45
20
0.5
30
0.1
0.55
40
V/V
kΩ
%
Input Resistor Network Ratio %:1
NOISE PERFORMANCE
Noise Density
Clock Noise
10 Hz to Nominal Bandwidth
See Figure 18
4
12
1
5
3
1
3
mg/√Hz
mV p-p
FREQUENCY RESPONSE
–3 dB Frequency8
Bandwidth Options: See Ordering Guide
–10
+10
%
Bandwidth Temperature Drift
Sensor Resonant Frequency
20
24
Hz
kHz
Q = 5
SELF-TEST9
@ 5 V
Output Change
Logic “1” Voltage
Logic “0” Voltage
Input Resistance
135
180
270
285
3.5
380
50
570
1.0
413
550
825
mV
V
V
To GND
30
kΩ
OUTPUT AMPLIFIER
Output Voltage Swing
IOUT = +100 µA
IOUT = –100 µA
VS–0.25
1000
V
V
pF
0.25
Capacitive Load Drive
POWER SUPPLY (VS)
Operating Voltage Range
Functional Voltage Range
Quiescent Supply Current
4.75
4.0
5.25
6.0
3
V
V
mA
mA
ADXL76, ADXL276,
ADXL269
18
3.5
5
TEMPERATURE RANGE10
NOTES
–40
+85
°C
1For example: ADXLxxQx38 describes devices with a nominal sensitivity of 38 mV/g. See Ordering Guide for full part number.
2Trimmed at the factory to user specifications. May require 0-g adjust circuitry. Reference section on Zero-g Adjustment and Dynamic Range.
3Alignment error is specified as the angle between the true axis of sensitivity and the edge of the package.
4Transverse sensitivity is measured with an applied acceleration that is 90 degrees from the indicated axis of sensitivity.
5Ratiometric: VOUT (accel, VS) = [VS/2 (a VS/5 V)] +[(accel)(b VS + c VS2)(1 0.08)] where a = offset range in volts. For a 38 mV/g sensor: b = 5.725 × 10–3 1/g, c = 0.375 ×
10–3 1/g/V. (For a 55 mV/g sensor: b = 8.284 × 10–3 1/g, c = 0.542 × 10–3 1/g/V. See Figures 5 and 13. Spec includes temperature drift, life drift, and nonlinearity. Test conditions:
100 Hz, 50 g for the 38 mV/g and 18 mV/g sensor; 100 Hz 35 g for the 55 mV/g sensor.
6Error included in full-scale range specification.
7Proportional to VS/2. See Figures 6 and 19.
8Includes Temperature Drift.
9ST pin from Logic “0” to “1.” For the ADXL76: VOUT change = (VOUT change @ 5 V) × (VS/5 V). For the ADXL269 and the ADXL276: VOUT change = (VOUT change @
5 V) × (VS/5 V)3.
10A higher temperature range available.
Specifications subject to change without notice.
–2–
REV. 0
ADXL76, ADXL269, ADXL276
ABSOLUTE MAXIMUM RATINGS*
Acceleration (Any Axis, Unpowered for 0.5 ms) . . . . . 2000 g
Supply Voltage VS . . . . . . . . . . . . . . . . . . . . –0.3 V to +7.0 V
Output Short Circuit Duration (VOUT, to Ground) . . . .Indefinite
Storage Temperature . . . . . . . . . . . . . . . . . . –65°C to +150°C
Lead Temperature Range (Soldering, 10 sec) . . . . . . . 245°C
Drop Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 m
*Stresses above those listed under Absolute Maximum Ratings may cause perma-
nent damage to the device. This is a stress rating only; the functional operation of
the device at these or any other conditions above those indicated in the operational
sections of this specification is not implied. Exposure to absolute maximum rating
conditions for extended periods may affect device reliability.
CAUTION
ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily
accumulate on the human body and test equipment and can discharge without detection. Although
the ADXL76, ADXL269, ADXL276 features proprietary ESD protection circuitry, permanent
damage may occur on devices subjected to high-energy electrostatic discharges. Therefore, proper
ESD precautions are recommended to avoid performance degradation or loss of functionality.
WARNING!
ESD SENSITIVE DEVICE
ORDERING GUIDE
Sensitivity
[mV/g]
Bandwidth
(–3 dB) [Hz]
Package
Option
Standard Devices
Part Number
Package Description*
ADXL76Q38
AD22215
AD22227
AD22229
AD22224
AD22226
AD22235
AD22239
AD22237
38
38
38
55
55
38
38
55
400
400
1000
400
400
400
400X; 1000Y
400
Ceramic Dual-In-Line
Surface Mount Ceramic
Surface Mount Ceramic
Ceramic Dual-In-Line
Surface Mount Ceramic
Surface Mount Ceramic
Surface Mount Ceramic
Surface Mount Ceramic
Q-8
ADXL76QC38
ADXL76QC38-1
ADXL76Q55
ADXL76QC55
ADXL276QC38
ADXL276QC38
ADXL276QC55
QC-14
QC-14
Q-8
QC-14
QC-14
QC-14
QC-14
Sensitivity
[mV/g]
Bandwidth
(–3 dB) [Hz]
Package
Option
Standard Devices
Part Number
Package Description*
ADXL76QC18
ADXL76Q38-1
ADXL269QC38
ADXL269QC55
ADXL276QC38-1
ADXL276QC
18
38
38
55
38
38X; 70Y
55
400
1000
400
400
1000
400
1000
Surface Mount Ceramic
Ceramic Dual-In-Line
Surface Mount Ceramic
Surface Mount Ceramic
Surface Mount Ceramic
Surface Mount Ceramic
Surface Mount Ceramic
QC-14
Q-8
QC-14
QC-14
QC-14
QC-14
QC-14
AD22228
AD22269
AD22236
AD22204
AD22238
ADXL276QC55-1
*Contact factory for availability.
NOTE: Surface mount packages are shipped taped and reeled with a full reel quantity of 750 pieces. Dual-in-line packages are
shipped in tubes (45 pieces per tube), 1080 pieces per box. Samples for preproduction development can be shipped in less than
full box or full reel quantities.
–3–
REV. 0
ADXL76, ADXL269, ADXL276
PIN CONFIGURATIONS
PIN FUNCTION DESCRIPTIONS
8-Lead Cerdip
1
14
13
12
V
V
NC
S
S
1
2
3
4
8
7
6
5
V
S
NC
NC
Pin
No.
2
3
4
5
6
7
/NC
NC
NC
ADXL76
V
OUT
ADXL76
Description
NC
TOP VIEW
ADXL76
TOP VIEW
(Not to Scale)
SELF-TEST
TP
(Not to Scale)
11 NC
NC
1
2
3
4
5
6
7
8
NC
NC
TP
COM
Zero-g Adj
Self-Test
VOUT
VS
Do Not Connect
Do Not Connect
Do Not Connect
Common
ZERO-g ADJ
COM
10
9
V
OUT
TP
NC = NO CONNECT
SELF-TEST
NC
ZERO-g ADJ
8
COM
Zero-g Offset Input (Optional)
Self-Test Input
Voltage Output
Power Supply
NC = NO CONNECT
1
14
13
12
V
V
NC
S
2
3
4
5
6
7
ZERO-g ADJ –X
S
NC
V
–X
TP
NC
NC
OUT
14-Lead Cerpak
Description
ADXL269
TOP VIEW
(Not to Scale)
11 NC
Pin
No.
10
9
V
+X
OUT
ADXL76
SELF -TEST
ZERO-g ADJ +X
8
COM
1
NC
Do Not Connect
2
3
NC
NC
Do Not Connect
Do Not Connect
NC = NO CONNECT
4
NC
Do Not Connect
5
6
7
8
TP
NC
Do Not Connect
Do Not Connect
Common Ground
Optional Offset Adjust Input for X Axis
Self-Test Input Pin
Voltage Output of X Axis
Do Not Connect
Do Not Connect
Do Not Connect/Power Supply Pin
1
2
3
4
5
6
7
14
V
S
NC
ZERO-g ADJ Y
13
12
V
S
COM
Zero-g Adj
Self-Test
VOUT
NC
NC
NC/VS*
NC
V
Y
OUT
ADXL276
TOP VIEW
(Not to Scale)
11 NC
NC
TP
9
10
9
V
X
OUT
10
11
12
13
14
SELF-TEST
NC
ZERO-g ADJ X
8
COM
NC = NO CONNECT
VS
Power Supply Pin
*For new designs only: Pin 13 should be connected to Power Supply.
ADXL76QC18: Pin 13 must be connected to VS.
Pin
No. ADXL276
ADXL269
Description
1
2
NC
NC
Do Not Connect
Zero-g Adj Y Zero-g Adj (–X) Optional Offset Adjust
Input for Y or –X Axis
3
V
OUT Y
VOUT (–X)
Voltage Output of Y or
–X Axis
4
5
6
7
8
NC
TP
NC
COM
TP
NC
NC
COM
Do Not Connect
Do Not Connect
Do Not Connect
Common Ground
Zero-g Adj X Zero-g Adj (+X) Optional Offset Adjust
Input for X or +X Axis
9
Self-Test
Self-Test
VOUT (+X)
NC
NC
VS
Self-Test Input Pin
Voltage Output of X Axis
Do Not Connect
10
11
12
13
VOUT
NC
NC
VS
X
Do Not Connect
Power Supply Pin,
Connect Also to Pin 14
Power Supply Pin,
Connect Also to Pin 13
14
VS
VS
–4–
REV. 0
ADXL76, ADXL269, ADXL276
THEORY OF OPERATION
PLATE
The ADXL76, ADXL269, and ADXL276 are fabricated using a
proprietary iMEMS surface micromachining process that has been in
high-volume production since 1993. The fabrication technique
uses standard integrated circuit manufacturing methods enabling
all the signal processing circuitry to be combined on the same chip
with the sensor. The higher levels of integration and economies of
volume production that exist for other ICs exist for ADI’s surface
micromachined process as well. This can be seen by comparing
prior generation ADXL50 and ADXL75 to this third generation
ADXL76 product family.
CAPACITANCES
MOVEABLE BEAM
(SINGLE AXIS)
+
+
FIXED
PLATE
PLATE
CAPACITANCES
FIXED PLATE
MOVEABLE PLATE
The ADXL76 family requires no external components, other than
a standard power supply bypass capacitor. A 2-pole switched-
capacitor Bessel filter is included on the chip.
ANCHOR
Figure 1. Simplified View of Sensor Under Acceleration
The fixed plates are driven by complementary 100 kHz clocks.
When no acceleration is applied the signal output is zero. When
the beam is deflected due to an acceleration the resulting signal
is proportional to the acceleration (Figure 2). After amplification
the signal is synchronously demodulated. This removes the
oscillator frequency from the signal and provides a dc output
signal. In the final stage of the accelerometer IC the dc signal is
further amplified and filtered (Figure 3).
The surface micromachined sensor element is made by deposit-
ing polysilicon on a sacrificial oxide layer that is then etched away
leaving the suspended sensor element. Figure 1 is a simplified
view of the ADXL76 sensor structures. In the ADXL276 there
are two sensing elements orthogonal to each other and in the
ADXL269 there two sensing elements of opposite polarity.
Each sensor has 42 unit sensing cells to sense acceleration. This
differential capacitor sensor is composed of fixed plates and
movable plates attached to the beam that moves in response to
acceleration. Movement of the beam changes the differential
capacitance which is measured with the on-chip circuitry.
V
IN+
+
V
OUT
+
AMP
V
IN–
Figure 2. Drive signals to the differential capacitors are 180 degrees out of phase. When the differential capacitors are
unequal, there is a signal out of the sensor which is proportional to the force on the beam. This modulated output from
the beam is then demodulated to create a dc output (not shown).
–5–
REV. 0
ADXL76, ADXL269, ADXL276
ACCELERATION RESPONSE
SELF-TEST RESPONSE
+
MOVEABLE
+X AXIS SENSOR
+
V
+X
V
+X
OUT
OUT
UNIT SENSING
CELL
UNIT SENSING
CELL
SELF-TEST
INPUT
+
MOVEABLE
–X AXIS SENSOR
+
V
–X
V
–X
OUT
OUT
Figure 3. Simplified Top-Down View of +X and –X Sensors Under Acceleration in the ADXL269
Self-Test Feature
Sensors with different bandwidths will have different output
signal rise times during self-test actuation. A 400 Hz sensor can
be differentiated from a 1000 Hz sensor by monitoring the
output. This simple method could be used to verify the chosen
bandwidth option during an electrical in-circuit test.
Each sensor element has 12 unit forcing cells for electrostatically
moving the beam during a self-test (see Figure 1). Self-test is
activated by the user with a dc logic high signal on the self-test
input pin. During a logic high, an electrostatic force acts on the
beam, equivalent to approximately 10 gs of acceleration input,
and thus creates a proportional voltage change to appear on the
output pin. When activated, the self-test feature exercises the
entire mechanical structure and electrical circuitry.
ADXL269 Features and Benefits
The ADXL269 is specially designed for redundant sensor opera-
tion to protect against general system level faults. In the ADXL269
the sensing fingers are opposite in polarity for the two sensing
elements. The output of each sensor, because of the orientation
of the two sensors, provides the same voltage with different
polarities resulting in symmetrical output signals (Figure 3). All
the circuitry is comparable for each sensing channel (see the sensor
diagram on the first page). The ADXL269 self-test has its own
unique feature. When self-test is activated via the self-test pin the
beams move in opposing directions. As a result of this movement,
the sensor outputs will have identical polarity (i.e., both chan-
nels give a positive output, Figure 3), an improbable situation in
a real world acceleration event. All the circuitry needed to drive
the sensor and convert the capacitance change to voltage is incor-
porated on the chip requiring no external components, except
for standard power supply decoupling.
As previously mentioned, the capacitive sensor is internally
controlled with a series of clocked events. These events are not
synchronized with the self-test input. When the self-test is applied
during certain internal clock states, this will cause an initial tran-
sient at the output of less than 10% of the self-test output voltage.
The shape of the initial transient will not affect the overall set-
tling time or self-test response amplitude (see Figure 4).
SELF-TEST OUTPUT
(0.2V/DIV)
More Notes On the XL76, XL269, and XL276
Both sensitivity and the zero-g value are essentially ratiometric
to the supply voltage, so that a ratiometric ADC and the acceler-
ometer scale factors track each other if the supply voltage changes.
SELF-TEST INPUT
(2V/DIV)
The output voltage (VOUT) is a function of both the acceleration
input and the power supply voltage (VS) as follows:
V
OUT (accel, VS) = VS/2– [(accel)(b VS + c VS2 )]
TIME (2ms/DIV)
The coefficients b and c are dependent on the sensitivity of the
accelerometer used. (See the Specifications page, Note 4 for the
coefficient values and offset range values.) Typically, the differ-
ences between a true ratiometric response and the actual output
are small (see Figures 5 and 6).
Figure 4. Typical Self-Test Response
The self-test function is designed to be a dc test. Applying a
repetitive signal to the self-test pin is not recommended.
–6–
REV. 0
ADXL76, ADXL269, ADXL276
5.0
5.0
6
0
4.0
3.0
–6
–12
–18
–24
–30
–36
–42
–48
–54
2.0
1.0
PACKAGE
0.0
RESONANCE
–1.0
–2.0
–3.0
–4.0
–5.0
BEAM
RESONANCE
4.0
4.5
5.0
5.5
6.0
100
1k
10k
10k
POWER SUPPLY VOLTAGE
FREQUENCY – Hz
Figure 5. Typical Sensitivity Error from Ideal Ratiometric
Response for a Number of Units
Figure 7. Typical Output Response of an ADXL76Qxx vs.
Frequency on a PC Board
6
0
0
2.5
2.0
1.5
–6
–12
–18
–24
–30
–36
–42
–48
–54
CERPAK
PACKAGE
RESONANCE
1.0
0.5
0.0
BEAM
RESONANCE
–0.5
–1.0
–1.5
–2.0
4.0
4.5
5.0
5.5
6.0
100
1k
10k
10k
POWER SUPPLY VOLTAGE
FREQUENCY – Hz
Figure 6. Offset Error of Zero-g Level from Ideal Ratiometric
Response as a Percent of Full-Scale for a Number of Units
Figure 8. Typical Output Response of an ADXL76QCxx vs.
Frequency on a PC Board
The final output stage of the accelerometer is a 2-pole Bessel
switched-capacitor filter. Bessel filters, also called linear phase
filters, have a step response with minimal overshoot and a maxi-
mally flat group delay. The –3 dB frequency of the poles is preset
at the factory to either 400 Hz or 1 kHz. (See Figures 7, 8, and
9.) These filters are also completely self-contained and buffered,
requiring no external components.
500g INPUT
OUTPUT RESPONSE
0
0.2
0.4
0.6
0.8
10
12
14
16
18
TIME – ms
Figure 9. Typical 500 g Recovery at the Output
–7–
REV. 0
ADXL76, ADXL269, ADXL276
Power-On Settling
VADJ = 280 mV, set the three-state I/O pin to a logic high (5 V)
The rising edges during power-on settling will have different
shapes depending on the charge state of internal capacitors in
the filter. Therefore, both the power supply voltage and the
charge state of the filter/sensor will affect the shape of the out-
put rising edge. Again, the final settling time is independent of
the initial edge shape (see Figure 10).
and use a resistor (RADJ) value of 104 kΩ. To adjust down by
280 mV set the logic level low. To calculate the value of RADJ
decide on the desired VADJ and apply the following equation:
,
3 × R5 kΩ × VS
RADJ
=
– R5 kΩ – R25 kΩ
2 × VADJ
This can be further simplified to:
5.5
V
S
5.0
4.5
4.0
3.5
3.0
2.5
7.5 × VS
VADJ
0.5V
RADJ
=
– 30 kΩ
V
–50g
OUT
V
–0g
OUT
Note: R5 kΩ and R25 kΩ can vary—see Specification.
VS is the supply voltage and R5 kΩ and R25 kΩ are the sensor’s
internal resistors (nominaly 5 kΩ and 25 kΩ).
2.0
1.5
1.0
0.5
0
V
+50g
OUT
Table II. Adjustment Ranges Using One Bit
Zero-g Error
Reading (mV)
I/O
Setting
Nominal Adjust
Amount (mV)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
TIME – ms
–140 to –420
–140 to +140
140 to 420
H
TS
L
280
0
–280
Figure 10. Typical power-on settling with full-scale Input.
Time constant of internal filter dominates the response
when a signal is present.
Figure 11 shows the zero-g adjust pin connected to one and two
three-state I/O pins on the microcontroller. The resistor values
shown do not include the on-resistance of the driving gates.
The capacitor helps to filter noise on the power supply line and
noise coming through the gate.
Zero-g Adjustment and Dynamic Range
In some cases the user may want to fine-adjust the zero-g output
level (sometimes referred to as offset level) to obtain maximum
dynamic range or achieve an asymmetrical output signal. The
user may adjust the zero-g voltage level of the accelerometer by
supplying a voltage to the zero-g adjustment pin. Any voltage
difference between the zero-g adjustment pin and VS/2 is reduced
by a factor of six by the internal resistor divider. This value is
then scaled up by a factor of three in the output stage for a total
gain of 0.5 for the zero-g adjustment. (Note: The ratio of the
resistors in the divider is consistent from part-to-part; however,
the absolute values can have a tolerance as listed in the speci-
fications.) The zero-g adjustment voltage can be setup by a
variety of methods including a PWM signal or a simple three-
state digital signal.
To adjust the output, the microcontroller reads the amplifier
output and sets the I/O pins to the proper level, as described in
Table II. A voltage is applied by the microcontroller depending
on the states of the I/O pin, thus creating three different adjust
levels. Table III describes the adjustment ranges when two bit
adjustment is used. Two bits of adjustment results in nine different
adjust levels. The zero-g error reading described in Tables II and
III refers to the ADXL output error from VS/2 before adjustment.
V
2
V
S
2
S
Table I. Achievable Zero-g Output Error
5k⍀
5k⍀
V
V
OUT
OUT
No Compensation
Three-State Compensation
1 Bit I/O
2 Bit I/O
PWM Compensation
DAC
11 g
ACCELERATION
SIGNAL
ACCELERATION
SIGNAL
5 g1
25k⍀
25k⍀
2.2 g1
0 g1
0 g2
ZERO-g ADJUST
ZERO-g ADJUST
104k⍀ 0.01F
104k⍀ 95k⍀ 0.01F
NOTES
1Based on an external resistor tolerance of 5%, VS = 5 V, 38 mV/g
version.
THREE-STATE
I/O-1
ADC
INPUT
THREE- THREE-
STATE STATE
ADC
INPUT
2Achievable to zero-g output error within the resolution of the PWM
or digital-to-analog converter (DAC).
I/O-1
I/O-2
CONTROLLER
CONTROLLER
Three-State Adjustment
Figure 11. Optional Zero-g Adjust using three-state signals.
On the left is a one-bit configuration and on the right is a
two-bit configuration. A filter capacitor, shown here, may
be required depending on the amount of noise on the
microcontroller I/O pins.
An effective and easy method that provides increased dynamic
range and coarse offset adjustment is to apply a voltage to the
zero-g adjust pin from a microcontroller’s three-state I/O pin
(Figure 11). Using one three-state I/O pin creates three different
levels of adjustment. For example, to adjust the zero-g level up by
–8–
REV. 0
ADXL76, ADXL269, ADXL276
Table III. Adjustment Ranges Using Two Bits
from a microcontroller’s counter/timer port can be used. PWM
signals have an average dc level that can be used to apply a
precise dc voltage to the zero-g offset adjust pin.
Zero-g Error
Remaining Nominal
Offset Adjust Amount
Setting Setting (mV)
Reading
(mV)
I/O-1
I/O-2
When zero-g adjustment is done by a PWM signal, as shown in
Figure 12, the resolution of zero-g adjustment is determined by
the number of bits in the counter. For example, a 60 kΩ resistor
with an 8-bit counter provides an adjust resolution of 3.26 mV and
(mV)
–420 to –358
–357 to –246
–246 to –152
–152 to –56
–56 to +56
56 to 152
152 to 246
246 to 357
357 to 420
H
H
TS
H
TS
L
TS
L
L
H
TS
H
L
TS
H
L
TS
L
–5 to +57
415
–57 to +54 300
–55 to +39 191
–39 to +57 113
a
418 mV adjustment range.
To understand how the PWM adjustment works, consider the
case at the extreme of the adjustment range when the output from
the microcontroller is always low or always high. The adjustment is
then very similar to the three-state method. When the pulses
are high for the same amount of time as they are low, the average
dc level is VS/2 and there is no adjustment. Everywhere between
these levels the dc voltage applied is proportional to the difference
between the time high and the time low. The adjustment resolu-
tion is limited by the number of bits in the PWM counter.
–56 to +56
0
–57 to +39 –113
–39 to +55 –191
–54 to +57 –300
–57 to +5
–415
Variations in the values for the zero-g adjust pin input imped-
ance, the zero-g adjust gain, resistance of the driving gate, and
external resistor will affect the adjustment value. The nominal
adjust column in Tables II and III refers to the adjustment value
when all of the above parameters are at their nominal values.
Therefore, determining the microcontroller output settings that
produce the lowest zero-g error may require an iterative process.
Noise energy at the internal clock frequency on the zero-g adjust
pin will cause output errors (Figure 13). Therefore, high-frequency
switching noise must be sufficiently filtered with a resistor and
capacitor as shown in Figure 12.
Using the Sensor Without Zero-g Output Adjustment
To determine the dynamic range without adjustment it is impor-
tant to determine the maximum output voltage swing; this is the
upper voltage limit (VS–0.25 V) minus the lower voltage limit
(0.25 V). If a 5 V power supply is used, the maximum output
swing would be 2.5 V 2.25 V. Therefore, without adjustment,
the maximum offset and sensitivity tolerance leads to a maximum
voltage output of:
V
2
S
ADXL76/276 OR
ADXL269
5k⍀
V
OUT
ACCELERATION
SIGNAL
GAIN = 3
25k⍀
(VS/2 – 0.25 V) – Zero-g Output (max) × Sensitivity (max)
= 2.25 V – 418 mV × 1.08 = 1.799 V
ZERO-g ADJUST
60k⍀
this corresponds to a measurable acceleration maximum of:
33nF
1.799 V
= 43.8 g
OUTPUT
PORT
ADC
INPUT
38 mV /g × 1.08
CONTROLLER
In this example with maximum tolerances, the positive swing is
limited to 43.8 g. The offset was assumed to be positive in this
case so the measurable acceleration in the negative direction
would be larger:
Figure 12. Optional Zero-g Adjust Using PWM
30
(VS/2 – 0.25 V) + Offset (max) × Sensitivity (max)
= 2.25 V + 418 mV × 1.08 = 2.701 V
25
20
15
10
5
this corresponds to a measurable acceleration maximum of:
2.701V
= –65.8 g
38 mV /g × 1.08
So in this particular example of a 38 mV/g accelerometer using
the maximum offset and sensitivity specification, the measurable
acceleration range without offset adjustment would be worst-case
+43.8 g/–65.8 g for an individual part. Since the sensitivity
tolerance can be up to +8% or –8% the worst dynamic range for
a population of accelerometers would have to be 43.8 g. Test
procedures ensure that the maximum offset does not coincide with
the maximum sensitivity in a single part, which allows a minimum
specification that is slightly higher than calculated above.
0
100
1000
10000
Figure 13. Supply voltage line interference at odd har-
monic frequencies to the sensor oscillator frequency may
also create output errors. This figure shows a typical noise
amplitude appearing on the output signal if a 1 mV rms
signal with these odd harmonics is present on the supply
voltage. This noise is a baseband error and can be at any
frequency in the baseband or at dc.
PWM Adjustment
In order to maximize the dynamic range it is important to minimize
the zero-g offset error. A pulsewidth modulated signal (PWM)
–9–
REV. 0
ADXL76, ADXL269, ADXL276
The corner frequency (–3 dB point) of the filter in Figure 12 is
80 Hz and therefore sufficient in applications with 8-bit PWM
and PWM frequencies of 12 kHz or more. At a corner frequency
of 80 Hz the ripple on the ADXL output is below one LSB (@
8-bit resolution, referred to VOUT) for PWM frequencies of 12 kHz
or higher.
Table V. Optional Bandwidth Adjust Component Values
Bandwidth
C1 Value
R1 Value
100 Hz
30 Hz
10 Hz
3 Hz
0.0015 µF
0.0047 µF
0.015 µF
0.047 µF
1 MΩ
1 MΩ
1 MΩ
1 MΩ
The microcontroller output port frequency should be chosen to
avoid harmonics with the accelerometer’s demodulator frequency.
For maximum filtering, the effective RC time constant must be
as large as possible. Using a larger capacitor will allow the reduc-
tion of the PWM frequency
Any zero-g deviation from VS/2 will be amplified along with the
signal. It is important to adjust the zero-g level at VOUT to VS/2
to ensure the largest possible output voltage swing. The adjust-
ment can be done at the zero-g adjust pin of the accelerometer,
as previously described, or it can be done at the input of the
external summing amplifier as shown in Figure 14.
The PWM adjustment method can provide continuous offset
compensation. This method is well suited to eliminate tempera-
ture drifts since the temperature change in many real-world
systems is only a few degrees per minute.
The voltage of the zero-g adjust pin is at VS/2 and can be used
as a reference for the bias level of the external amplifier. If this is
done then the zero-g adjust pin can no longer be used for adjust-
ment of the zero-g level. A 100 nF capacitor at the positive input of
the amplifier connected to ground helps against high frequency
noise pickup.
Asymmetrical Ranges
Zero-g offset level shifts can be used to create asymmetric mea-
surement ranges. This can be achieved by connecting the zero-g
adjust pin to VS or to GND via an appropriate resistor. The
asymmetry should not exceed 25%. Contact the factory for
ranges beyond this amount. Note that the sensors are tested at
50 g or 35 g, according to the sensor´s nominal sensitivity.
R1
R2
GV
=
OUT
Low-g Applications
Connection Diagram and Layout Guidelines
To use the ADXL76/ADXL276/ADXL269 in low-g applications,
it is possible to add an output amplifier with filtering, as shown
in Figure 14. Gain is set by R1/R2, and a single-pole low-pass
filter is formed by C1 and R1.
The basic connection diagram to a microcontroller is shown in
Figure 15. The microcontroller ADC reference, VRH, and the
accelerometer supply voltage should be connected at exactly the
same voltage. Since the accelerometer is ratiometric, the use
of a ratiometric ADC will reduce system errors due to supply
V
2
V
S
S
C1
ADXL76/276
OR ADXL269
voltage changes.
5V
GAIN = 3
100k⍀
5k⍀
5V
R1
V
OUT
ACCELERATION
SIGNAL
R2
BUFFER
FILTER
V
V
V
SS
S
RH
OPTIONAL
EXTERNAL
OP AMP
25k⍀
V
O
PE
68HC11
90k⍀
C
C
ADXL76/
ADXL276
ZERO-g ADJUST
V
RL
V
COM
DD
Figure 14. Optional Output Amplifier and Filter
To illustrate this Table IV lists different sensitivities based on a
55 mV/g sensor and different R2 values.
Figure 15. Basic Connection to a Microcontroller with ADC
Table IV. Optional Gain Adjust Resistor Values
for 55 mV/g-Versions
Also, the ground for the accelerometer should be directly con-
nected to VRL of the microcontroller. Note that the analog and
digital grounds are separated. It is important to keep analog and
digital grounds separate to avoid ground loops. To minimize
pickup and noise in the design of the PC Board, standard ana-
log design techniques are recommended (e.g., ground planes,
avoiding long routes to external components, avoiding placement
near “noise generators” like crystals, switching transistors/logic
ICs, and microcontrollers). Consider using the microcontrol-
ler manufacturer’s recommended signal isolation components.
Desired
Sensitivity
Full-Scale
Range
Gain
R1 Value R2 Value
100 mV/g
200 mV/g
400 mV/g
1.79
3.64
7.30
20 g
10 g
5 g
1 MΩ
1 MΩ
1 MΩ
560 kΩ
275 kΩ
137 kΩ
The bandwidth of the circuitry is determined by C1 and R1.
(–3 dB) = 1 / (2 × 1/2 × R1 × C1)
F
The ADXL76, ADXL269, and ADXL276 use synchronous
demodulation architecture, commonly used in accelerometers
and other sensors. Noise on the supply that is synchronous with
or near the clock frequency (approximately 100 kHz) or its
odd harmonics will exhibit baseband errors at the output.
These error signals are the beat frequency signals between the
clock and the supply noise (see Figures 13 and 16).
–10–
REV. 0
ADXL76, ADXL269, ADXL276
20
Such noise can be generated by digital oscillations (power sup-
ply, microcontrollers, crystal, etc.) elsewhere in the system and
must be attenuated by proper bypassing as shown in Figure 15.
By inserting a small resistance in series with the accelerometer
and ADC reference supply, a filter is created with the bypass
capacitor. For example if R = 3 Ω and C = 2.2 µF, a 24 kHz
filter is created which is usually adequate to attenuate noise
on the supply from other digital circuits with proper ground
and supply layout. Other values for the resistors and capacitors
are possible. In this example, however, the voltage drop over the
resistor is smaller than half of an LSB for an 8-bit 5 V input ADC.
Thus the positive reference of the ADC VRH is very close to 5 V
and the span of the other ADC channels of the microcontroller
are not significantly limited by the voltage drop over the resistor.
15
10
5
0
–5
–10
–15
–20
NOISE FROM INTERNAL CLOCK
0
2
4
6
8
10
12
14
16
18
20
TIME – s
30
25
20
Figure 18. Typical Output Noise Voltage for ADXL76Q38
with Spikes Generated by Internal Clock. Measurement
Bandwidth > 100 MHz.
30
15
10
20
10
0
5
0
–10
–20
–30
95
100
105
110
FREQUENCY – kHz
Figure 16. If high frequencies are present on the voltage
supply line, output errors may result. This figure shows
a typical noise amplitude appearing on the output signal
if a 1 mV rms signal similar in frequency to the sensor’s
internal oscillator frequency is present on the supply volt-
age. In this case, the internal oscillator frequency was
approximately 102.5 kHz. Using appropriate supply volt-
age filtering methods, this type of interference can be
avoided.
–30 –20 –10
0
10 20 30 40 50 60 70 80
TEMPERATURE
Figure 19. Zero-g Drift from –40°C to +25°C and from
+25°C to +85°C for a Number of Units for ADXL76Q38
2.4
2.2
+105؇C
2.0
1.8
–40؇C
+25؇C
1.6
1.4
1.2
4
4.5
5
5.5
6
SUPPLY VOLTAGE – V
Figure 17. Typical Accelerometer Supply Current vs.
Supply Voltage
–11–
REV. 0
ADXL76, ADXL269, ADXL276
OUTLINE DIMENSIONS
Dimensions shown in inches and (mm).
ORIENTATION OF SENSITIVE AXES
Ceramic Dual-In-Line
(QC-8)
14
13
12
11
10
9
14
13
12
11
10
9
1
2
3
4
5
6
7
1
2
3
4
5
6
7
+X
+X
90؇
8
5
0.310 (7.874)
0.220 (5.588)
+Y
PIN 1
1
4
ADXL276
ADXL76
8
8
0.100 (2.540) BSC
0.345 (8.763)
0.290 (7.366)
0.485 (12.319)
MAX
0.015
(0.381)
MIN
ADXL76 SENSITIVE AXIS
ADXL276 SENSITIVE AXIS
POSITIVE A => NEGATIVEV
POSITIVE A => NEGATIVEV
0.230
(5.842)
MAX
OUT
OUT
(TOPVIEW)
(TOPVIEW)
0.230
(5.842)
MAX
0.200 (5.080)
0.115 (2.921)
0.015 (0.381)
0.008 (0.203)
14
13
12
11
10
9
1
2
3
4
5
6
7
SEATING
15°
0°
0.023 (0.584) 0.070 (1.778)
0.014 (0.356) 0.038 (0.965)
PLANE
–X
+X
POSITIVE
ACCELERATION
POSITIVE
ACCELERATION
Surface Mount Ceramic
(QC-14)
8
ADXL269
0.485 (12.319)
MAX
ADXL269 SENSITIVE AXIS
POSITIVE A => NEGATIVEV
(+X-AXIS)
OUT
OUT
14
8
=> POSITIVEV
(TOPVIEW)
(–X-AXIS)
0.310 (7.874)
0.220 (5.588)
0.419 (10.643)
0.394 (10.008)
7
1
Figure 20. Sensitive Axes
0.345 (8.763)
0.290 (7.366)
PIN 1
0.300 (7.62)
0.195 (4.953)
0.115 (2.921)
0.215 (5.461)
0.119 (3.023)
8؇
0؇
0.020 (0.508)
0.004 (0.102)
SEATING
PLANE
0.050 0.020 (0.508)
0.013 (0.318)
0.009 (0.229)
0.050 (1.270)
0.016 (0.406)
(1.27)
BSC
0.013 (0.330)
–12–
REV. 0
相关型号:
©2020 ICPDF网 联系我们和版权申明