a 61 g to 65 g Single Chip Accelerometer

with Signal Conditioning

ADXL05*
FEATURES floor is 500 µg/√Hz, (12× less than the ADXL50), allowing sig-
5 milli-g Resolution nals below 5 milli-g to be resolved. The ADXL05 is a force bal-
Noise Level 123 Less than the ADXL50 anced capacitive accelerometer with the capability to measure
User Selectable Full Scale from 61 g to 65 g both ac accelerations (typical of vibration) or dc accelerations
Output Scale Selectable from 200 mV/g to 1 V/ g (such as inertial force or gravity). Three external capacitors and
Complete Acceleration Measurement System on a a +5 volt regulated power supply are all that is required to
Single Chip IC measure accelerations up to ± 5 g. Three resistors are used to
Self Test on Digital Command configure the output buffer amplifier to set scale factors from
+5 V Single Supply Operation 200 mV/g to 1 V/g. External capacitors may be added to the
1000 g Shock Survival resistor network to provide 1 or 2 poles of filtering. No addi-
tional active components are required to interface directly to
APPLICATIONS
most analog to digital converters (ADCs).
Low Cost Sensor for Vibration Measurement
Tilt Sensing with Faster Response than Electrolytic or The device features a TTL compatible self-test function that
Mercury Sensors can electrostatically deflect the sensor beam at any time to verify
More Sensitive Alarms and Motion Detectors that the sensor and its electronics are functioning correctly.
Affordable Inertial Sensing of Velocity and Position The ADXL05 is available in a hermetic 10-pin TO-100 metal
can, specified over the 0°C to +70°C commercial, and –40°C to
GENERAL DESCRIPTION +85°C industrial temperature ranges. Contact factory for avail-
The ADXL05 is a complete acceleration measurement system ability of automotive grade devices.
on a single monolithic IC. The ADXL05 will measure accelera-
tions with full-scale ranges of ± 5 g to ± 1 g or less. Typical noise

FUNCTIONAL BLOCK DIAGRAM

ADXL05
+3.4V
VREF
REFERENCE 6
OUTPUT
OSCILLATOR
DECOUPLING
CAPACITOR +1.8V
4 OSCILLATOR SENSOR DEMODULATOR
C2
PREAMP

SELF-TEST
7
(ST) BUFFER
AMP

5 1 2 3 8 10 9
COM C3 C1 C1 VPR VIN–
R1 R3
VOUT
DEMODULATOR R2
CAPACITOR
+5V

*Patents pending.

REV. B
Information furnished by Analog Devices is believed to be accurate and © Analog Devices, Inc., 1996
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 One Technology Way, P.O. Box 9106, Norwood. MA 02062-9106, U.S.A.
otherwise under any patent or patent rights of Analog Devices. Tel: 617/329-4700 Fax: 617/326-8703

This datasheet has been downloaded from http://www.digchip.com at this page

 (TA = TMIN to TMAX, TA = +258C for J Grade Only, VS = +5 V, @ Acceleration = 0 g,
ADXL05–SPECIFICATIONS unless otherwise noted)
ADXL05J/A
Parameter Conditions Min Typ Max Units
SENSOR INPUT
Measurement Range Guaranteed Full Scale –5 +5 g
Nonlinearity Best Fit Straight Line, 5 g FS 0.2 % of FS
Alignment Error1 ±1 Degrees
Transverse Sensitivity2 ±2 %
SENSITIVITY
Initial Sensitivity at VPR +25°C 175 200 225 mV/g
Initial Sensitivity at VOUT +25°C, R3/R1 = 5 0.875 1.000 1.125 V/g
Temperature Drift3 ± 0.5 % of Reading
ZERO g BIAS LEVEL at VPR
Initial Offset 1.50 1.80 2.10 V
vs. Temperature3 ± 25/40 mV
vs. Supply VS = 4.75 V to 5.25 V 10 32 mV/V
NOISE PERFORMANCE at VPR
Voltage Noise Density BW = 4 Hz to 1 kHz 500 1000 µg/√Hz
Noise in 100 Hz Bandwidth 5 mg rms
Noise in 10 Hz Bandwidth 1.6 mg rms
FREQUENCY RESPONSE
3 dB Bandwidth4 C1 = 0.022 µF (See Figure 9) 1000 1600 Hz
3 dB Bandwidth4 C1 = 0.010 µF 4 kHz
Sensor Resonant Frequency 12 kHz
SELF TEST INPUT
Output Change at VPR5 ST Pin from Logic “0” to “1” –0.85 –1.00 –1.15 V
Logic “1” Voltage 2.0 V
Logic “0” Voltage 0.8 V
Input Resistance To Common 50 kΩ
+3.4 V REFERENCE
Output Voltage 3.350 3.400 3.450 V
Output Temperature Drift3 ±5 mV
Power Supply Rejection DC, VS = +4.75 V to +5.25 V 1 10 mV/V
Output Current Sourcing 500 µA
PREAMPLIFIER OUTPUT
Voltage Swing 0.25 VS – 1.4 V
Current Output Source or Sink 30 80 µA
Capacitive Load Drive 100 pF
BUFFER AMPLIFIER
Input Offset Voltage6 Delta from Nominal 1.800 V ± 10 ± 25 mV
Input Bias Current 5 20 nA
Open-Loop Gain DC 80 dB
Unity Gain Bandwidth 200 kHz
Output Voltage Swing IOUT = ± 100 µA 0.25 VS – 0.25 V
Capacitive Load Drive 1000 pF
Power Supply Rejection DC, VS = +4.75 V to +5.25 V 1 10 mV/V
POWER SUPPLY
Operating Voltage Range 4.75 5.25 V
Quiescent Supply Current 8.0 10.0 mA
TEMPERATURE RANGE
Operating Range J 0 +70 °C
Specified Performance A –40 +85 °C
Automotive Grade* –40 +125 °C
NOTES
1
Alignment error is specified as the angle between the true and indicated axis of sensitivity, (see Figure 2).
2
Transverse sensitivity is measured with an applied acceleration that is 90° from the indicated axis of sensitivity. Transverse sensitivity is specified as the percent of
transverse acceleration that appears at the V PR output. This is the algebraic sum of the alignment and the inherent sensor sensitivity errors, (see Figure 2).
3
Specification refers to the maximum change in parameter from its initial at +25°C to its worst case value at T MIN to TMAX.
4
Frequency at which response is 3 dB down from dc response assuming an exact C1 value is used. Maximum recommended BW is 6 kHz using a 0.010 µF capacitor, refer to
Figure 9.
5
Applying logic high to the self-test input has the effect of applying an acceleration of –5 g to the ADXL05.
6
Input offset voltage is defined as the output voltage differential from 1.800 V when the amplifier is connected as a follower. The voltage at this pin has a temperature drift
proportional to that of the 3.4 V reference.
*Contact factory for availability of automotive grade devices.
All min and max specifications are guaranteed. Typical specifications are not tested or guaranteed.
Specifications subject to change without notice.

–2– REV. B
 System Performance Specifications–ADXL05
C2
4 1 +5V
0.022µF
ADXL05
1.8V C3
C1 PRE-AMP BUFFER 0.1µF
2
AMP
0.022µF
3 9 VOUT
C1
5
COM
6 8 10
NOMINAL VALUES:
+3.4V VPR C4 R1 VIN– R3
R1 = 49.9kΩ
REF
R3 = 249kΩ
R2
R2 = 640kΩ

AC COUPLED CONNECTION (61.5 g Full Scale)

(@ VOUT Terminal (Pin 9), unless otherwise noted. 0 g Bias Level = +2.5 V, C1 = 0.022 mF, R2 = 2.57 R3

ADXL05J/A

Parameter Conditions Min Typ Max Units

Buffer Gain G = R3/R1* 5
FULL-SCALE RANGE –1.5 +1.5 g
SENSITIVITY @ +25°C 875 1000 1,125 mV/g
Temperature Drift TMIN to TMAX ± 0.5 % of Reading
ZERO g BIAS LEVEL @ +25°C 2.5 V
Temperature Drift +25°C to TMIN or TMAX 2/5 mV
FREQUENCY RESPONSE C4 = 3.3 µF, R1 = 49.9 kΩ 1 1000 Hz
*Note: Resistor tolerance will affect system accuracy. Use of ± 1% (or better) metal film resistors is recommended.

C2
4 1 +5V
0.022µF
ADXL05
1.8V C3
C1 PRE-AMP BUFFER 0.1µF
2
AMP
0.022µF
3 9 VOUT
C1
5
COM
6 8 10
NOMINAL VALUES:
+3.4V VPR R1 VIN– R3 R1 = 49.9kΩ
REF
R3 = 100kΩ (G=2)
R2
R2 = 255kΩ (G=2)

DC COUPLED CONNECTION (62 g Full Scale)

(@ VOUT Terminal (Pin 9), unless otherwise noted. 0 g Bias Level = +2.5 V, C1 = 0.022 mF, R2 = 2.57 R3)

ADXL05J/A
Parameter Conditions Min Typ Max Units
Buffer Gain G = R3/R1* 2
FULL-SCALE RANGE –2 +2 g
SENSITIVITY @ +25°C 350 400 450 mV/g
Temperature Drift TMIN to TMAX ± 0.5 % of Reading
ZERO g BIAS LEVEL @ +25°C 1.75 2.5 3.2 V
Temperature Drift +25°C to TMIN or TMAX ± 50/80 mV
FREQUENCY RESPONSE dc 1000 Hz
*Note: Resistor tolerance will affect system accuracy. Use of ± 1% (or better) metal film resistors is recommended.

REV. B –3–
ADXL05
ABSOLUTE MAXIMUM RATINGS* Package Characteristics
Acceleration (Any Axis, Unpowered for 0.5 ms) . . . . . . 1000 g Package uJA uJC Device Weight
Acceleration (Any Axis, Powered for 0.5 ms) . . . . . . . . . . 500 g
+VS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . –0.3 V to +7.0 V 10-Pin TO-100 130°C/W 30°C/W 5 Grams
Output Short Circuit Duration
(VPR, VOUT, VREF Terminals to Common) . . . . . . . Indefinite
Operating Temperature . . . . . . . . . . . . . . . . . –55°C to +125°C ORDERING GUIDE
Storage Temperature . . . . . . . . . . . . . . . . . . . –65°C to +150°C Model Temperature Range
*Stresses above those listed under “Absolute Maximum Ratings” may cause
permanent damage to the device. This is a stress rating only; the functional ADXL05JH 0°C to +70°C
operation of the device at these or any other conditions above those indicated in the ADXL05AH –40°C to +85°C
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. WARNING!
Although the ADXL05 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. ESD SENSITIVE DEVICE

Drops onto hard surfaces can cause shocks of greater than 1000 g
and exceed the absolute maximum rating of the device. Care
should be exercised in handling to avoid damage.

PIN DESCRIPTION CONNECTION DIAGRAM

10-Header (TO-100)
+5 V The power supply input pin.
C2 Connection for an external bypass capacitor (nominally 0.022 µF) TOP VIEW
used to prevent oscillator switching noise from interfering with COM
other ADXL05 circuitry. Please see the section on component VREF C2
5
selection. 66 4
C1 Connections for the demodulator capacitor, nominally 0.022 µF.
SENSITIVITY

ST C1
AXIS OF

7 3
See the section on component selection for application information.
COM The power supply common (or “ground”) connection. 88 2
VPR NOTES:
C1
VREF Output of the internal 3.4 V voltage reference. 9 1
AXIS OF SENSITIVITY IS ALONG A LINE
BETWEEN PIN 5 AND THE TAB.
10
ST The digital self-test input. It is both CMOS and TTL compatible. VOUT +5V
THE CASE OF THE METAL CAN
PACKAGE IS CONNECTED TO PIN 5
VPR The ADXL05 preamplifier output providing an output voltage of (COMMON).
VIN– ARROW INDICATES DIRECTION OF
200 mV per g of acceleration. POSITIVE ACCELERATION ALONG AXIS
OF SENSITIVITY.
VOUT Output of the buffer amplifier.
VIN– The inverting input of the uncommitted buffer amplifier.

–4– REV. B
 ADXL05
GLOSSARY OF TERMS Polarity of the Acceleration Output
Acceleration: Change in velocity per unit time. The polarity of the ADXL05 output is shown in the Figure 1.
Acceleration Vector: Vector describing the net acceleration When oriented to the earth’s gravity (and held in place), the
acting upon the ADXL05 (AXYZ). ADXL05 will experience an acceleration of +1 g. This corre-
sponds to a change of approximately +200 mV at the VPR out-
g: A unit of acceleration equal to the average force of gravity put pin. Note that the polarity will be reversed to a negative
occurring at the earth’s surface. A g is approximately equal to going signal at the buffer amplifier output VOUT, due to its
32.17 feet/s2, or 9.807 meters/s2. inverting configuration.
Nonlinearity: The maximum deviation of the ADXL05 output TAB
voltage from a best fit straight line fitted to a plot of acceleration
vs. output voltage, calculated as a % of the full-scale output +1g +
–
voltage (@ 5 g). INDICATED POLARITY IS THAT
PIN 5 OCCURRING AT V PR .
Resonant Frequency: The natural frequency of vibration of
the ADXL05 sensor’s central plate (or “beam”). At its resonant
frequency of 12 kHz, the ADXL05’s moving center plate has a
peak in its frequency response with a Q of 3 or 4.
Sensitivity: The output voltage change per g unit of accelera- Figure 1. Output Polarity at VPR
tion applied, specified at the VPR pin in mV/g. Acceleration Vectors in Three Dimensions
Sensitive Axis (X): The most sensitive axis of the accelerom- The ADXL05 is a sensor designed to measure accelerations that
eter sensor. Defined by a line drawn between the package tab result from an applied force. The ADXL05 responds to the
and Pin 5 in the plane of the pin circle. See Figures 2a and 2b. component of acceleration on its sensitive X axis. Figures 2a
and 2b show the relationship between the sensitive “X” axis and
Sensor Alignment Error: Misalignment between the
the transverse “Z” and “Y” axes as they relate to the TO-100
ADXL05’s on-chip sensor and the package axis, defined by
Pin 5 and the package tab.
Z
Total Alignment Error: Net misalignment of the ADXL05’s TRANSVERSE Z AXIS
SIDE VIEW
on-chip sensor and the measurement axis of the application.
This error includes errors due to sensor die alignment to the
package, and any misalignment due to installation of the sensor
package in a circuit board or module.
X X
Transverse Acceleration: Any acceleration applied 90° to the PIN 5 SENSITIVE (X) AXIS
TAB
axis of sensitivity.
Transverse Sensitivity Error: The percent of a transverse ac-
celeration that appears at the VPR output. For example, if the
Z
transverse sensitivity is 1%, then a +10 g transverse acceleration
will cause a 0.1 g signal to appear at VPR (1% of 10 g). Trans- Figure 2a. Sensitive X and Transverse Z Axis
verse sensitivity can result from a sensitivity of the sensor to
transverse forces or from misalignment of the internal sensor to Y
its package. TRANSVERSE Y AXIS
TOP VIEW
Transverse Y Axis: The axis perpendicular (90°) to the pack-
age axis of sensitivity in the plane of the package pin circle. See
Figure 2.
X X
Transverse Z Axis: The axis perpendicular (90°) to both the PIN 5
TAB SENSITIVE (X) AXIS
package axis of sensitivity and the plane of the package pin
circle. See Figure 2.
Y

Figure 2b. Sensitive X and Transverse Y Axis

REV. B –5–
ADXL05
package. Figure 2c describes a three dimensional acceleration Table I. Ideal Output Signals for Off Axis Applied
vector (AXYZ) which might act on the sensor, where AX is the Accelerations Disregarding Device Alignment and
component of interest. To determine AX, first, the component Transverse Sensitivity Errors
of acceleration in the XY plane (AXY) is found using the cosine
law: % of Signal Appearing Output in gs for a 5 g
θX at Output Applied Acceleration
AXY = AXYZ (cosθXY) then
AX = AXY (cosθX) 0 100% 5.000 (On Axis)
1° 99.98% 4.999
Therefore: Nominal VPR = 200 mV/g (AXYZ) (cosθXY) cosθX 2° 99.94% 4.997
3° 99.86% 4.993
–Z AXIS 5° 99.62% 4.981
10° 98.48% 4.924
Axyz
30° 86.60% 4.330
45° 70.71% 3.536
60° 50.00% 2.500
80° 17.36% 0.868
85° 8.72% 0.436
θxy 87° 5.25% 0.263
Ax
X AXIS 88° 3.49% 0.175
θx 89° 1.7% 0.085
90° 0% 0.000 (Transverse Axis)
Axy
Mounting Fixture Resonances
Y AXIS
A common source of error in acceleration sensing is resonance
of the mounting fixture. For example, the circuit board that the
Figure 2c. A Vector Analysis of an Acceleration Acting
ADXL05 mounts to may have resonant frequencies in the same
Upon the ADXL05 in Three Dimensions
range as the signals of interest. This could cause the signals
Note that an ideal sensor will react to forces along or at angles measured to be larger than they really are. A common solution
to its sensitive axis but will reject signals from its various trans- to this problem is to dampen these resonances by mounting the
verse axes, i.e., those exactly 90° from the sensitive “X” axis. ADXL05 near a mounting post or by adding extra screws to
But even an ideal sensor will produce output signals if the trans- hold the board more securely in place.
verse signals are not exactly 90° to the sensitive axis. An accel-
When testing the accelerometer in your end application, it is
eration that is acting on the sensor from a direction different
recommended that you test the application at a variety of fre-
from the sensitive axis will show up at the ADXL05 output at a
quencies in order to ensure that no major resonance problems
reduced amplitude.
exist (refer to Analog Devices Application Note AN-379).
Table I shows the percentage signals resulting from various θX
angles. Note that small errors in alignment have a negligible
effect on the output signal. A 1° error will only cause a 0.02%
error in the signal. Note, however, that a signal coming 1° off of
the transverse axis (i.e., 89° off the sensitive axis) will still con- 100
tribute 1.7% of its signal to the output. Thus large transverse 90

signals could cause output signals as large as the signals of

interest. Table I may also be used to approximate the effect of
the ADXL05’s internal errors due to misalignment of the die to
the package. For example: a 1 degree sensor alignment error will
allow 1.7% of a transverse signal to appear at the output.
10
0%

0.5ms

Figure 3. 500 g Shock Overload Recovery. Top Trace, PCB

Reference Accelerometer Output: 500 g/Vertical Division.
Bottom Trace, ADXL05 Output at VPR

–6– REV. B
 ADXL05
Typical Characteristics (@ +258C, C1 = C2 = 0.022 mF, V = +5 V unless otherwise noted) S

6 0.2

NONLINEARITY IN % OF FULL SCALE

3
NORMALIZED SENSITIVITY – dB

0 0.1

–3

–6 0

–9

–12 –0.1

–15
–0.2
–18

–21
1 10 100 1k 10k 0 1 2 3 4 5
FREQUENCY – Hz g LEVEL APPLIED

Figure 4. Normalized Sensitivity vs. Frequency Figure 7. % Nonlinearity vs. g Level Applied

40
2

30
1
–3dB BANDWIDTH CHANGE – %

CHANGE IN NOISE – %

20
0

10
–1

0
–2

–10
–3

–4 –20
–50 0 50 100 150 –50 0 50 100 150
TEMPERATURE – °C
TEMPERATURE – °C

Figure 5. –3 dB Bandwidth vs. Temperature Figure 8. % Change in Noise from +25°C vs. Temperature

10k
0.30

0.20

0.10
SENSITIVITY CHANGE – %

–3dB BANDWIDTH – Hz

1k
0.00

–0.10

–0.20
100

–0.30

–0.40

10
–0.50 0.01 0.1 1
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5
DEMODULATOR CAPACITANCE – µF
SUPPLY VOLTAGE – Volts

Figure 6. Sensitivity Change at VPR vs. Supply Voltage Figure 9. –3 dB Bandwidth vs. Demodulator
Capacitance

REV. B –7–
ADXL05
Typical Characteristics (@ +258C, C1 = C2 = 0.022 mF, V = +5 V unless otherwise noted) S

80 80

TA = +25°C
60 VS = +5V + (0.5Vp-p)
CHANGE IN 0g BIAS LEVEL – mV

E
A
40

+3.4V VREF PSRR – dB

60
B
20
C
B
0 D

C 40
–20
A
D
E
–40

–60 20
–40 0 40 80 120 1 10 100 1k 10k 100k
TEMPERATURE – °C FREQUENCY – Hz

Figure 10. Change in 0 g Bias Level vs. Temperature Figure 13. +3.4 V VREF PSRR vs. Frequency
(Characteristic Curves from Five Typical Units)

40 100

TA = +25°C, ACL = 2
30
80
OUTPUT IMPEDANCE – Ω

20
VPR 0 g PSRR – dB

60

10

40
0 NOTE: AT THIS FREQUENCY, THE SIGNAL ON THE
POWER SUPPLY IS IN SYNCHRONISM WITH THE
ACCELEROMETER'S INTERNAL CLOCK OSCILLATOR
(SEE EMI/RFI SECTION) 20
–10

–20 0
10 100 1k 10k 100k 1M 10 100 1k 10k
FREQUENCY – Hz FREQUENCY – Hz

Figure 11. 0 g PSRR vs. Frequency Figure 14. Buffer Amplifier Output Impedance vs.
Frequency

0.05 30

25 TA = +25°C
0 G =10
20
% VREF DRIFT FROM +25°C

15
–0.05
10
GAIN – dB

G=2
–0.1 5

0
–0.15
–5

–10
–0.2
–15

–0.25 –20
–50 0 50 100 150 10 100 1k 10k 100k 1M
TEMPERATURE – °C FREQUENCY – Hz

Figure 12. % VREF Drift vs. Temperature Figure 15. Buffer Amplifier Closed-Loop Gain vs.
Frequency

–8– REV. B
 ADXL05
THEORY OF OPERATION Figure 17 shows the sensor responding to an applied accelera-
The ADXL05 is a complete acceleration measurement system tion. When this occurs, the common central plate or “beam”
on a single monolithic IC. It contains a polysilicon surface- moves closer to one of the fixed plates while moving further
micro machined sensor and signal conditioning circuitry which from the other. This creates a mismatch in the two capacitances,
implements a force-balance control loop. The ADXL05 is ca- resulting in an output signal at the central plate. The output
pable of measuring both positive and negative acceleration to a amplitude of the signal varies directly with the amount of accel-
maximum level of ± 5 g. eration experienced by the sensor.
Figure 16 is a simplified view of the ADXL05’s acceleration
sensor at rest. The actual structure of the sensor consists of 46 TOP VIEW
unit cells and a common beam. The differential capacitor sensor
APPLIED
consists of independent fixed plates and central plates attached ACCELERATION BEAM
to the main beam that moves in response to an applied accelera- CENTER PLATE
tion. The two capacitors are series connected, forming a
capacitive divider with a common movable central plate. The
FIXED
sensor’s fixed capacitor plates are driven differentially by a OUTER
PLATES
1 MHz square wave: the two square wave amplitudes are equal
but are 180° out of phase from one another. When at rest, the UNIT CELL
CS1 CS1 < CS2
values of the two capacitors are the same, and therefore, the
DENOTES ANCHOR
voltage output at their electrical center (i.e., at the center plate) CS2

is zero.
Figure 17. The ADXL05 Sensor Momentarily Responding
to an Externally Applied Acceleration
BEAM
Figure 18 shows a block diagram of the ADXL05. The voltage
CENTER PLATE output from the central plate of the sensor is buffered and then
applied to a synchronous demodulator which is clocked, in
FIXED
phase, with the same oscillator that drives the fixed plates
OUTER
PLATES
of the sensor. If the applied voltage is in sync and in phase
with the clock, a positive output will result. If the applied volt-
UNIT CELL
CS1 CS1 = CS2
age is in sync but 180° out of phase with the clock, then the
CS2
DENOTES ANCHOR
demodulator’s output will be negative. All other signals will be
rejected. An external capacitor, C1, sets the bandwidth of the
demodulator.
Figure 16. A Simplified Diagram of the ADXL05
Sensor at Rest

DENOTES EXTERNAL +3.4V +3.4V +5V

PIN CONNECTION
33kΩ 33kΩ
75Ω PREAMP
C2
C1 EXTERNAL C1 VPR
DEMODULATION
EXTERNAL CAPACITOR
OSCILLATOR +1.8V
0° CS1
DECOUPLING +1.8V
CAPACITOR 1MHz
OSCILLATOR BEAM
CS2 SYNCHRONOUS
180° DEMODULATOR
LOOP GAIN = 10 INTERNAL
SYNC FEEDBACK
+5V +5V LOOP
3MΩ
RST
+5V COMMON
+0.2V
INTERNAL
VREF BUFFER
REFERENCE +3.4V
+3.4V AMPLIFIER
+1.8V 50kΩ
VOUT
VIN–
+3.4V +1.8V +0.2V SELF–TEST
COM (ST)

Figure 18. Functional Block Diagram

REV. B –9–
ADXL05
The output of the synchronous demodulator drives the amount equal to –5 g (the negative full-scale output of the de-
preamp—an instrumentation amplifier buffer which is refer- vice). Note that the ± 15% tolerance of the self-test circuit is not
enced to +1.8 volts. The output of the preamp, VPR, is fed back proportional to the sensitivity error, see Self-Test section.
to the outer plate of the sensor through a 3 MΩ isolation resis-
tor. The VPR voltage electrostatically resets the sensor back to its BASIC CONNECTIONS FOR THE ADXL05
0 g position and is a direct measure of the applied acceleration. Figure 19 shows the basic connections needed for the ADXL05
to measure accelerations in the ± 5 g range with an output scale
The output of the ADXL05’s preamplifier is 1.8 V ± 200 mV/g,
factor 400 mV/g, a 2.5 V 0 g level, a ± 2.0 V full-scale swing
with an output range of ± 1 V for a ± 5 g input. An uncommit-
around 0 g, and a 3 dB bandwidth of approximately 1.6 kHz.
ted buffer amplifier provides the capability to adjust the scale
factor and 0 g offset level over a wide range. An internal refer- Using the circuit of Figure 19, the overall transfer function is:

( ) ( )
ence supplies the necessary regulated voltages for powering the
chip and +3.4 volts for external use. VOUT = R3 (1.8 V – V PR ) + R3 (1.8) + 1.8 V
R1 R2
A self-test is initiated by applying a TTL “high” level voltage
(> +2.0 V dc) to the ADXL05’s self-test pin, which causes the
chip to apply a deflection voltage to the beam which moves it an

ADXL05
+3.4V
VREF
REFERENCE 6
OUTPUT
OSCILLATOR
DECOUPLING
CAPACITOR +1.8V
4 OSCILLATOR SENSOR DEMODULATOR
C2
0.022µF
PREAMP

SELF-TEST
7
(ST) BUFFER
AMP

5 1 2 3 8 10 9
COM C3 C1 C1 VPR R1 VIN– R3
0.1µF 49.9kΩ 100kΩ
VOUT
0.022µF

+5V DEMODULATOR
R2
CAPACITOR
274kΩ

Figure 19. ADXL05 Application Providing an Output Sensitivity of 400 mV/g,

a +2.5 V 0 g Level and a Bandwidth of 1 kHz

–10– REV. B
 ADXL05
USING THE INTERNAL BUFFER AMPLIFIER TO VARY
THE ACCELEROMETER’S OUTPUT SCALE FACTOR C2
4 1 +5V
AND 0 g BIAS LEVEL 0.022µF ADXL05
1.8V 0.1µF
The ADXL05 accelerometer has an onboard buffer amplifier C1
2 PRE-AMP BUFFER
AMP
that allows the user to change the output scale factor and 0 g 0.022µF
3 9 VOUT
bias level. C1

The output scale factor of an accelerometer is simply how many COM

5

volts output are provided per g of applied acceleration. This 6 8 10

VPR VIN–
should not be confused with its resolution. The resolution of the +3.4V
REF
R3

device is the lowest g level the accelerometer is capable of mea- R1a R1b
suring. Resolution is principally determined by the device noise OUTPUT SCALE FACTOR =
R3
x VPR OUTPUT
(R1a + R1b)
and the measurement bandwidth. VPR OUTPUT: 200mV/g

The 0 g bias level is simply the dc output level of the accelerom-

eter when it is not in motion or being acted upon by the Earth’s Figure 21. External Scale Factor Trimming
gravity. Setting the Accelerometer’s 0 g Bias Level, AC Coupled
Setting the Accelerometer’s Scale Factor Response
Figure 20 shows the basic connections for using the onboard If a dc (gravity) response is not required—for example in motion
buffer amplifier to increase the output scale factor. The nominal sensing or vibration measurement applications—ac coupling can
output level in volts from VPR (the preamplifier output) is equal be used between the preamplifier output and the buffer input as
to the g forces applied to the sensor (along its sensitive axis) shown in Figure 22. The use of ac coupling between VPR and
times 200 mV/g. The use of the buffer is always recommended, the buffer input virtually eliminates any 0 g drift and allows the
even if the preset scale factor is adequate, as the buffer helps maximum buffer gain without clipping.
prevent any following circuitry from loading down the VPR Resistor R1 and capacitor C4 together form a high pass filter
output. whose corner frequency is 1/(2 π R1 C4). This means that this
simple filter will reduce the signal from VPR by 3 dB at the
C2
corner frequency, and it will continue to reduce it at a rate of
4 1 +5V 6 dB/octave (20 dB per decade) for signals below the corner
0.022µF ADXL05
C1
1.8V C3
0.1µF
frequency.
PRE-AMP BUFFER
2
AMP
0.022µF Note that capacitor C4 should be a nonpolarized, low leakage
C1
3 9 VOUT type. If a polarized capacitor is used, tantalum types are pre-
5
ferred, rather than electrolytic. With polarized capacitors, VPR
COM
8
should be measured on each device and the positive terminal of
6 10
+3.4V VPR R1 VIN– R3
the capacitor connected toward either VPR or VIN—whichever is
REF more positive
R3
OUTPUT SCALE FACTOR =
R1
x VPR OUTPUT The 0 g offset level of the ADXL05 accelerometer is preset at
VPR OUTPUT: 200mV/g
+1.8 V. This can easily be changed to a more convenient level,
such as +2.5 V which, being at the middle of the supply voltage,
Figure 20. Basic Buffer Connections provides the greatest output voltage swing.
In Figure 20, the output scale factor at Pin 9 (VOUT) is the out- When using the ac coupled circuit of Figure 22, only a single re-
put at VPR times the gain of the buffer, which is simply the value sistor, R2, is required to swing the buffer output to +2.5 V.
of resistor R3 divided by R1. Choose a convenient scale factor, Since the “+” input of the buffer is referenced at +1.8 V, its
keeping in mind that the buffer gain not only amplifies the sig- summing junction, Pin 10, is also held constant at +1.8 V.
nal but any noise or drift as well. Too much gain can also cause Therefore, to swing the buffer’s output to the desired +2.5 V
the buffer to saturate and clip the output wave form. 0 g bias level, its output must move up +0.7 V (2.5 V – 1.8 V =
The circuit of Figure 20 is entirely adequate for many applica- 0.7 V). Therefore, the current needed to flow through R3 to
tions, but its accuracy is dependent on the pretrimmed accuracy cause this change, IR3, is equal to:
of the accelerometer and this will vary by product type and
0.7 Volts
grade. For the highest possible accuracy, an external trim is rec- IR3 =
R3 in Ohms
ommended. As shown by Figure 21, this consists of a potenti-
ometer, R1a, in series with a fixed resistor, R1b.

REV. B –11–
ADXL05
With a dc coupled connection, any difference between a non-
C2 ideal +1.8 V 0 g level at VPR and the fixed +1.8 V level at the
4 1 +5V
0.022µF ADXL05 buffer’s summing junction will be amplified by the gain of the
1.8V C3
C1
2 PRE-AMP BUFFER 0.1µF buffer. If the 0 g level only needs to be approximate and the
AMP
0.022µF buffer is operated a low gain, a single fixed resistor, R2, can still
3 9 VOUT
C1 be used. But to obtain the exact 0 g output desired or to allow
5
SCALE FACTOR =
R3 the maximum output voltage swing from the buffer, the 0 g
COM
6 8 10
R1 offset will need to be externally trimmed using the circuit of Fig-
1
+3.4V VPR C4 R1 VIN– R3 C4 =
2πR1 FL
ure 23. Normally, a value of 100 kΩ is typical for R2.
REF
FOR A +2.5V 0g LEVEL,
IN AN AC COUPLED
The buffer’s maximum output swing should be limited to
R2
CONFIGURATION, ± 2 volts, which provides a safety margin of ± 0.25 volts before
R2 = 2.57 R3
clipping. With a +2.5 volt 0 g level, the maximum gain the
RECOMMENDED COMPONENT VALUES buffer should be set to (R3/R1) equals:
SCALE DESIRED R2 VALUE
FULL
SCALE
FACTOR LOW
FREQUENCY
R1
IN
CLOSEST
C4
R3
IN
IN kΩ FOR 2Volts
IN +2.5V 0g
RANGE mV/g LIMIT, FL kΩ VALUE kΩ LEVEL 200 mV/g Times the Max Applied Acceleration in gs
±2 g 1000 30Hz 49.9 0.10µF 249 640
Note that the value of R1 should be kept as large as possible,
±5 g 400 30Hz 127 0.039µF 249 640
20 kΩ or greater, to avoid loading down the VPR output.
±2 g 1000 3Hz 49.9 1.0µF 249 640
The device scale factor and 0 g offset levels can be calibrated us-
±5 g 400 1Hz 127 1.5µF 249 640
ing the Earth’s gravity as explained in the section “calibrating
±5 g 400 0.1Hz 127 15µF 249 640
the ADXL05.”

Figure 22. Typical Component Values for AC Coupled C2

Circuit 4 ADXL05 1 +5V
0.022µF
1.8V 0.1µF
In order to force this current through R3, the same current C1
2 PRE-AMP BUFFER
AMP
needs to flow from Pin 10 to ground through resistor R2. Since 0.022µF
3
Pin 10 is always held at +1.8 V, R2 is equal to: C1
9 VOUT

5
1.8 Volts
R2 = COM
8
IR3 6 10
+3.4V VPR R1 VIN– R3
REF
Therefore, for an ac coupled connection and a +2.5 V 0 g VX
output: 50kΩ R3
0g R2 S.F. =
R1
1.8Volts × R3 LEVEL 100kΩ
R1 ≥ 20kΩ
R2 = = 2.57 × R3 TRIM
0.7Volts
RECOMMENDED COMPONENT VALUES FOR
If ac coupling is used, the self-test feature must be monitored at
VARIOUS OUTPUT SCALE FACTORS
VPR, rather than at the buffer output (since the self test output is
FULL mV R1 R3
a dc voltage). SCALE per g kΩ kΩ

±1 g 2000 30.1 301

Setting the Accelerometer’s 0 g Bias Level, DC Coupled
Response ±2 g 1000 40.2 200

When a true dc (gravity) response is needed, the output from ±4 g 500 40.2 100
the preamplifier, VPR, must be dc coupled to the buffer input. ±5 g 400 49.2 100
For high gain applications, a 0 g offset trim will also be needed.
The external offset trim permits the user to set the 0 g offset
voltage to exactly +2.5 volts, since this is at the center of the +5 Figure 23. Typical Component Values for Circuit with
volt power supply it will allow the maximum output swing from External 0 g Trimming
the buffer without clipping.

–12– REV. B
 ADXL05
DEVICE BANDWIDTH VS. MEASUREMENT The equivalent rms noise of the bandpass filter is equal to
RESOLUTION 500 µg / Hz × (1.5 FH ) – (FL /1.5) .
Although an accelerometer is usually specified according to its
full scale (clipping) g level, the limiting resolution of the device, For example, the typical rms noise of the ADXL05 using 1 pole
i.e., its minimum discernible input level, is extremely important ac coupling with a bandwidth of 10 Hz and 1 pole low-pass
when measuring low g accelerations. filter of 100 Hz is:
The limiting resolution is predominantly set by the measure- Noise (rms ) = 500 µg / Hz × 1.5(100) – (10 /1.5)
ment noise “floor” which includes the ambient background = 5,987 µg rms or ≈ 5.9 mg rms
noise and the noise of the ADXL05 itself. The level of the noise
floor varies directly with the bandwidth of the measurement. As Because the ADXL05’s noise is for all practical purposes
the measurement bandwidth is reduced, the noise floor drops, Gaussian in amplitude distribution, the highest noise amplitudes
improving the signal-to-noise ratio of the measurement and in- have the smallest (yet nonzero) probability. Peak-to-peak noise
creasing its resolution. is therefore difficult to measure and can only be estimated due
to its statistical nature. Table II is useful for estimating the
The bandwidth of the accelerometer can be easily reduced by
probabilities of exceeding various peak values, given the rms
adding low-pass or bandpass filtering. Figure 24 shows the typi-
value.
cal noise vs. bandwidth characteristic of the ADXL05.
Table II.
100mg 660mg

Nominal Peak-to- % of Time that Noise Will Exceed

NOISE LEVEL – Peak to Peak Peak Value Nominal Peak-to-Peak Value
2.0 × rms 32%
NOISE LEVEL – rms

4.0 × rms 4.6%

6.0 × rms 0.27%
10mg 66mg
6.6 × rms 0.1%
8.0 × rms 0.006%

RMS and peak-to-peak noise (for 0.1% uncertainty) for various

bandwidths is estimated in Figure 24. As shown by the figure,
device noise drops dramatically as the operating bandwidth is
1mg 6.6mg
10 100 1k
reduced. For example, when operated in a 1 kHz bandwidth,
3dB BANDWIDTH – Hz the ADXL05 typically has an rms noise level of 19 mg. With
± 5 g applied accelerations, this 19 mg resolution limit is nor-
Figure 24. Noise Level vs. 3 dB Bandwidth mally quite satisfactory; but for smaller acceleration levels the
The output noise of the ADXL05 scales with the square root of noise is now a much greater percentage of the signal. As shown
the measurement bandwidth. With a single pole roll-off, the by the figure, when the device bandwidth is rolled off to 100 Hz,
equivalent rms noise bandwidth is π divided by 2 or approxi- the noise level is reduced to approximately 6 mg, and at 10 Hz it
mately 1.5 times the 3 dB bandwidth. For example, the typical is down to less than 2 mg.
rms noise of the ADXL05J using a 100 Hz one pole post filter is: Alternatively, the signal-to-noise ratio may be improved consid-
erably by using a microprocessor to perform multiple measure-
Noise (rms ) = 500 µg/ Hz × 100(1.5) = 6,124 µg or 6.1 mg rms ments and then compute the average signal level. When using
For the bandpass filter of Figure 27 where both ac coupling and this technique, with 100 measurements, the signal-to-noise ratio
low pass filtering are used, the low frequency roll-off, FL, is de- will be increased by a factor of 10 (20 dB).
termined by C4 and R1 and the high frequency roll-off, FH, is
determined by the 1-pole post filter R3, C5.

REV. B –13–
ADXL05
Low-Pass Filtering
The bandwidth of the accelerometer can easily be reduced by
ADXL05
using post filtering. Figure 25 shows how the buffer amplifier 1.8V OPTIONAL CAPACITOR
PRE-AMP BUFFER
can be connected to provide 1-pole post filtering, 0 g offset trim- AMP
FOR 3-POLE FILTERING

ming, and output scaling. The table provides practical compo- 9

VOUT

nent values for various full-scale g levels and approximate circuit R5

bandwidths. For bandwidths other than those listed, use the 6 8 10
formula: VREF VPR VIN– C4 +5V

1 0.01µF
C4 = R5
(2 π R3)Desired 3dB Bandwidth in Hz 42.2kΩ R4a R4b
2 7
OUTPUT
R1 R3
or simply scale the value of capacitor C4 accordingly, i.e., for an 82.5kΩ
C3
82.5kΩ SCALE
AD820 6

application with a 50 Hz bandwidth, the value of C4 will need 2-POLE FILTER

FACTOR 3 4 OFFSET AND
TRIM
to be twice as large as its 100 Hz value. If further noise reduc- +3.4V SCALING
AMPLIFIER
tion is needed while maintaining the maximum possible band- R6
40.2kΩ
width, then a 2- or 3-pole post filter is recommended. These
provide a much steeper roll-off of noise above the pole fre- 20kΩ
0g
quency. Figure 26 shows a circuit that uses the buffer amplifier R7 LEVEL
71.5kΩ TRIM
to provide 2-pole post filtering. Component values for the 2-
pole filter were selected to operate the buffer at unity gain. Ca- 2-POLE FILTER OFFSET AND SCALING
COMPONENT VALUES AMPLIFIER COMPONENT VALUES
pacitors C3 and C4 were chosen to provide 3 dB bandwidths of
3dB FULL mV R4a R4b R5
10 Hz, 30 Hz, 100 Hz, and 300 Hz. BW (Hz) C3µF C4µF SCALE per g GAIN kΩ kΩ kΩ

In this configuration, the nominal buffer amplifier output will be 300 0.027 0.0033 ±1 g 2000 10.00 10 24.9 301
+1.8 V ± the 200 mV/g scale factor of the accelerometer. An 100 0.082 0.01 ±2 g 1000 4.98 10 35.7 200
AD820 external op amp allows noninteractive adjustment of 0 g
30 0.27 0.033 ±4 g 500 2.50 10 35.7 100
offset and scale factor. The external op amp offsets and scales
±5 g
the output to provide a +2.5 V ± 2 V output over a wide range 10 0.82 0.1 400 2.00 10 45.3 100

of full-scale g levels.
Figure 26. Two-Pole Filtering Circuit with Gain and 0 g
Offset Adjustment
C2
4 1 +5V
0.022µF ADXL05
C1
1.8V 0.1µF Bandpass Filtering
PRE-AMP BUFFER
2
AMP Figure 27 shows how the combination of ac coupling and low-
0.022µF
3 9 VOUT pass filtering together form a bandpass filter that provides an
C1
even greater improvement in noise reduction.
5
COM
6 8 10
+3.4V VPR R1a VIN–
REF R1b R3 ADXL05 1.8V
BUFFER
OPTIONAL SCALE PRE-AMP AMP
FACTOR TRIM* VPR VOUT
0g C4
9
LEVEL 50kΩ
TRIM R2 *TO OMIT THE OPTIONAL SCALE FACTOR
TRIM, REPLACE R1a AND R1b WITH A FIXED
8 10
VALUE 1% METAL FILM RESISTOR.
VPR VI N –
SEE VALUES SPECIFIED IN TABLES BELOW. C4 R1 R3
NOTE: FOR NONINTERACTIVE TRIMS, COMPONENT
SET SCALE FACTOR FIRST, THEN OFFSET. VALUES ARE
R2
COMPONENT VALUES FOR VARIOUS APPROXIMATE.
FULL-SCALE RANGES AND BANDWIDTHS C5

FULL mV 3dB R1a R1b R3 R2 C4

SCALE per g BW (Hz) kΩ kΩ kΩ kΩ µF DESIRED DESIRED
SCALE VALUE
FACTOR LOW R1 CLOSEST HIGH R3 CLOSEST OF R2
±1 g 2000 10 10 24.9 301 100 0.056 FREQUENCY VALUE FREQUENCY
IN C4 IN C5 FOR +2.5V
mV/g LIMIT, FL IN kΩ VALUE LIMIT, FH kΩ VALUE 0g LEVEL
±2 g 1000 100 10 35.7 200 100 0.0082

±4 g 500 200 10 35.7 100 100 0.0082 1000 30 49.9 0.10µF 300 249 0.002µF 640kΩ
200 30 249 0.022µF 300 249 0.002µF 640kΩ
±5 g 400 300 10 45.3 100 100 0.0056
1000 3 49.9 1.0µF 100 249 0.0068µF 640kΩ
1
3dB BW = 200 1 249 0.68µF 100 249 0.0068µF 640kΩ
2π R3 C4
200 0.1 249 6.8µF 10 249 0.068µF 640kΩ

Figure 25. Using the Buffer Amplifier to Provide 1-Pole Figure 27 AC Coupling and Low-Pass Filtering Used
Post Filtering Plus Scale Factor and 0 g Level Trimming Together to Provide a Bandpass Function

–14– REV. B
 ADXL05
Additional Noise Reduction Techniques
Shielded wire should be used for connecting the accelerometer to
any circuitry that is more than a few inches away—to avoid 60 Hz
pickup from ac line voltage. Ground the cable’s shield at only one
end and connect a separate common lead between the circuits; θ θ
this will help to prevent ground loops. Also, if the accelerometer
is inside a metal enclosure, this should be grounded as well. 1g 1g
Methods for Reducing 0 g Offset Drift
When using any accelerometer with a dc (gravity sensing) re-
sponse, the 0 g offset level will exhibit some temperature drift. Figure 28. Two Possible Orientations for Tilt Measurement
For very high accuracy applications, one very straightforward
approach is to use a low cost crystal oven to maintain the accel- Conversely, for a given acceleration signal and assuming no
erometer at a constant temperature. These ovens are available in other changes in the axis or interfering signals, the tilt angle is
a variety of different temperatures. After the circuit has been built proportional to the voltage output as shown in Figure 29. The
and is operating correctly, the crystal oven can be mounted over angle, θ can be calculated using:
the accelerometer and powered off the same +5 V power supply.
 V – zero g output(V ) 
θ = arcsin 1g ×  OUT 
The ovens may be purchased from Isotemp Research, Inc., P.O.   Scale Factor (V / g ) 
Box 3389, Charlottesville, VA 22903, phone 804-295-3101. For
more details on crystal oven compensation, refer to application 500
note AN-385.
400
Other methods for 0 g drift compensation include using a low
300
cost temperature sensor such as the AD22100 to supply a mi-
croprocessor with the device temperature. If the drift curve of 200

VOUT @ 500mV/g
the accelerometer is stored in the µP, then a software program 100
can be used to subtract out the drift. Alternatively, a simple 1st 0
order (straight line) correction circuit can be used to subtract
–100
out the linear portion of the accelerometer’s drift by using a
temperature sensor and op amp to supply a small compensation –200

current. For more details on software and hardware drift com- –300
pensation, refer to application note AN-380. –400

–500
ACCELEROMETER APPLICATIONS –90 –70 –50 –30 –10 10 30 50 70 90
ANGLE OF TILT
Popular applications for low g accelerometers tend to fall into
three categories: measurement of tilt and orientation, inertial
measurement of acceleration, velocity and distance, and vibra- Figure 29. VOUT vs. Tilt Angle
tion or shock measurement. The use of an accelerometer in tilt applications has several ad-
The ADXL05 is a “dc” accelerometer, meaning that it is ca- vantages over the use of a traditional tilt sensor. A traditional tilt
pable of measuring static accelerations such as the Earth’s grav- sensor consists of glass vial filled with a conductive liquid, typi-
ity. The ADXL05 differs from other acceleration measurement cally a mercury or electrolytic solution. Besides being larger
technologies such as piezoelectric and piezofilm sensors which than an XL05, it requires additional signal conditioning cir-
can only respond to ac signals greater than approximately 1 Hz. cuitry. The settling time and frequency response is limited by
This dc capability is required for tilt and inertial measurement. the amount of time required for the liquid to stop sloshing
For ac shock or vibration the ADXL05 can measure frequencies around in the vial. In high vibration environments, or where
of up to 4 kHz and has the added benefit of measuring all the high lateral accelerations may be present, it may not be possible
way down to dc. to resolve the tilt signal above the “slosh” noise. The acceler-
ometer has faster frequency (up to 50 ×) response and set-
Using the ADXL05 in Tilt Applications tling time. Interfering vibrations may be filtered out if
The ADXL05’s precision dc characteristics make it suitable for necessary, an impossibility with a liquid tilt sensor, since one
tilt measurement. It can directly measure the Earth’s gravity and cannot filter the liquid. Finally, in the presence of lateral accel-
use this constant force as a position reference to determine incli- erations, an accelerometer provides more useful information,
nation. As shown in Figure 28, the accelerometer should be i.e., an acceleration signal, which if cleverly signal processed,
mounted so that its sensitive axis is perpendicular to the force of can provide both a tilt and an acceleration output. A single ac-
gravity, i.e., parallel to the Earth’s surface. In this manner, it celerometer can be used to measure tilt over a 180° range; two
will be most sensitive to changes in orientation (when it is orien- accelerometers gives a complete 360° of measurement.
tated 90° to the force of gravity). Its output can be then de-
scribed by the sine function; a tilt occurring at an angle θ will An important characteristic for an accelerometer used in a tilt
cause a voltage output equal to: application is its 0 g offset stability over temperature. The
ADXL05 typically exhibits offsets that deviate no more than
VOUT =  Accelerometer Scale Factor  g  × sin ( θ ) ×1g  + zero g output(V )
V 0.1 g over the 0°C to +70°C temperature range, corresponding
   

REV. B –15–
ADXL05
to a 5° tilt error over the entire temperature range. Straight- control of system and mounting resonances are critical to proper
forward calibration schemes discussed in this data sheet may be measurement. Refer to the application note AN-379, available
used to reduce or compensate for temperature drift to improve from Analog Devices.
the absolute accuracy of the measurement.
CALIBRATING THE ADXL05
Using the ADXL05 in Inertial Measurement Applications
If a calibrated shaker is not available, both the 0 g level and scale
Inertial measurement refers to the practice of measuring accel-
factor of the ADXL05 may be easily set to fair accuracy by using
eration for the purpose of determining the velocity of an object
a self-calibration technique based on the 1 g (average) accelera-
and its change in position, or distance traveled. This technique
tion of the earth’s gravity. Figure 30 shows how gravity and
has previously required expensive inertial guidance systems of
package orientation affect the ADXL05’s output. Note that the
the type used in commercial aircraft and military systems. The
output polarity is that which appears at VPR; the output at VOUT
availability of a low cost precision dc accelerometer such as the
will have the opposite sign. With its axis of sensitivity in the
ADXL05 enables the use of inertial measurement for more cost
vertical plane, the ADXL05 should register a 1 g acceleration,
sensitive industrial and commercial applications.
either positive or negative, depending on orientation. With the
Inertial measurement makes use of the fact that the integral of axis of sensitivity in the horizontal plane, no acceleration (the
acceleration is velocity and the integral of velocity is distance. 0 g bias level) should be indicated.
By making careful measurements of acceleration and math-
ematically integrating the signals, one can determine both veloc-
ity and the distance traveled. The technique is useful for
applications where a traditional speed and distance measure-
0g 0g –1 g +1 g
ment is impractical, or where a non-contact, relative position (a) (b) (c) (d)
measurement must be made.
A practical inertial measurement system uses multiple acceler-
ometers to measure acceleration in three axes, and gyroscopes to INDICATED POLARITY IS THAT
measure rotation in three axes, the requirement for a 6 degree of OCCURRING AT V PR .

freedom system. For simpler systems where one or more of the

axes can be constrained, it is possible to build a system with
fewer accelerometers and gyros.
The measurement system must take the acceleration sensor and
calibrate out all static errors including any initial inaccuracy or Figure 30. Using the Earth’s Gravity to Self-Calibrate the
temperature drift. A mathematical model is used to describe the ADXL05
performance of the sensor in order to calibrate it. If these errors To self-calibrate the ADXL05, place the accelerometer on its
are not removed, then the process of double integration will side with its axis of sensitivity oriented as shown in “a.” The 0 g
quickly cause any small error to dominate the result. Most prac- offset potentiometer, Rt, is then roughly adjusted for midscale:
tical systems use microprocessors for error correction and a tem- +2.5 V at the buffer output (see Figure 25).
perature sensor for temperature drift compensation. Another
approach is to maintain all of the sensors at a controlled tem- Next, the package axis should be oriented as in “c” (pointing
perature. The microprocessors have the additional advantage of down) and the output reading noted. The package axis should
providing a low cost method of performing the single and then be rotated 180° to position “d” and the scale factor poten-
double integration of the acceleration signal. tiometer, R1a, adjusted so that the output voltage indicates a
change of 2 gs in acceleration. For example, if the circuit scale
The stability and repeatability of the accelerometer is the most factor at the buffer output is 400 mV per g, then the scale factor
important specification in an inertial system. The ADXL05 is trim should be adjusted so that an output change of 800 mV is
“well behaved” that is, its response and temperature characteris- indicated.
tics are easy to model and correct, and once modeled they are
very repeatable. For example, temperature performance can be Adjusting the circuit’s scale factor will have some effect on its
adequately modeled using first order, (straight line) approxima- 0 g level so this should be readjusted, as before, but this time
tions for most applications, and other errors such as on-axis and checked in both positions “a” and “b.” If there is a difference in
pendulous rectification are minimal. This greatly simplifies the the 0 g reading, a compromise setting should be selected so that
math required to correct the sensor. the reading in each direction is equidistant from +2.5 V. Scale
factor and 0 g offset adjustments should be repeated until both
Vibration and Shock Measurement Applications are correct.
The ADXL05 can measure shocks and vibrations from dc to
4 kHz. Typical signal processing for vibration signals includes REDUCING POWER CONSUMPTION
fast Fourier transforms, and single and double integration for The use of a simple power cycling circuit provides a dramatic
velocity and displacement. It is possible to build a single inte- reduction in the ADXL05’s average current consumption. In
grator stage using the ADXL05’s output buffer amplifier in low bandwidth applications such as shipping recorders, this
order to provide a velocity output. simple, low cost circuit can provide substantial power reduction.
The sensitivity of the accelerometer will typically vary only If a microprocessor is available, only the circuit of Figure 31 is
± 0.5% over the full industrial temperature range, making it one needed. The microprocessor supplies a TTL clock pulse to gate
of the most stable vibration measurement devices available. In buffer transistor Q1 which inverts the output pulse from the µP
vibration measurement applications, mechanical mounting and

–16– REV. B
 ADXL05
so that the duty cycle is correct when the pulse is re-inverted
again by transistor, Q2, which cycles the accelerometer’s supply 2V 200µs

voltage on and off. 100

90

+5V

0.1µF
100kΩ
BUFFER
10kΩ 10
Q1 2N3906 0%
FROM µP 10kΩ
OR 2N2222 Q2 500mV
1
FIGURE 1b
ADXL05
VPR VIN– VOUT Figure 33. Top Trace: Voltage at Pin 1; Bottom Trace:
Buffer Output With R1 = R3 = 100 kΩ, CF = 0.01 µ F
5 8 R1 10 R3 9

VOUT
COMPONENT SELECTION
CF LOAD DRIVE CAPABILITIES OF THE V PR AND BUFFER
OUTPUTS
Figure 31. Basic Power Cycling Circuit The VPR and the buffer amplifier outputs are both capable of
Figures 32 and 33 show typical waveforms of the accelerometer driving a load to voltage levels approaching that of the supply
being operated with a 10% duty cycle: 1 ms on, 9 ms off. This rail. However, both outputs are limited in how much current
reduces the average current consumption of the accelerometer they can supply, affecting component selection.
from 8 mA to 800 µA, providing a power reduction of 90%. The VPR Output
µP should sample acceleration during the interval between the The VPR pin has the ability to source current up to 500 µA but
time the 0 g level has stabilized (≈ 400 µs using a 0.022 µF only has a sinking capability of 30 µA which limits its ability to
demod cap) and the end of the pulse duration. The measure- drive loads. It is recommended that the buffer amplifier be used
ment bandwidth of a power-cycled circuit will be set by the in most applications, to avoid loading down VPR. In standard
clock pulse rate and duty cycle. In this example, 1 sample can ± 5 g applications, the resistor R1 from VPR to VIN– is recom-
be taken every 10 ms which is 100 samples per second or 100 mended to have a value greater than 20 kΩ to reduce loading
Hz. As defined by the “Nyquist criteria,” the best case measure- effects.
ment bandwidth is FS/2 or half the clock frequency. Therefore
50 Hz signals can be processed if adequate filtering is provided. Capacitive loading of the VPR pin should be minimized. A load
capacitance between the VPR pin and common will introduce an
offset of approximately 1 mV for every 10 pF of load. The VPR
pin may be used to directly drive an A/D input or other source
2V 200µs as long as these sensitivities are taken into account. It is always
100
preferable to drive A/D converters or other sources using the
90
buffer amplifier (or an external op amp) instead of the VPR pin.
Buffer Amplifier Output
The buffer output can drive a load to within 0.25 V of either
power supply rail and is capable of driving 1000 pF capacitive
10
loads. Note that a capacitance connected across the buffer
0% feedback resistor for low-pass filtering does not appear as a
1V capacitive load to the buffer. The buffer amplifier is limited to
sourcing or sinking a maximum of 100 µA. Component values
for the resistor network should be selected to ensure that the
Figure 32. Top Trace: Voltage at Pin 1; buffer amplifier can drive the filter under worst case transient
Bottom Trace: Output at VPR conditions.
Higher measurement bandwidths can be achieved by reducing Self-Test Function
the size of the demodulation capacitor below 0.022 µF and in- The digital self-test input is compatible with both CMOS and
creasing the pulse frequency. A 0.01 µF capacitor was con- TTL signals. A Logic “l” applied to the self-test (ST) input will
nected across the feedback resistor of the ADXL05 buffer to cause an electrostatic force to be applied to the sensor which
improve its transient characteristics. The optimum value for this will cause it to deflect to the approximate negative full-scale
capacitor will change with buffer gain and the cycling pulse rate. output of the device. Accordingly, a correctly functioning accel-
For more details, refer to application note AN-378. erometer will respond by initiating an approximate –1 volt

REV. B –17–
ADXL05
output change at VPR. If the ADXL05 is experiencing an low-pass filtering generally results in smaller capacitance values
acceleration when the self-test is initiated, the VPR output will and better overall performance. It is also a convenient and more
equal the algebraic sum of the two inputs. The output will stay precise way to set the system bandwidth. Post filtering allows
at the self-test level as long as the ST input remains high and bandwidth to be controlled accurately by component selection
will return to the 0 g level when the ST voltage is removed. and avoids the ± 40% demodulation tolerance. Note that signal
A self-test output that varies more than ± 15% from the nominal noise is proportional to the square root of the bandwidth of the
–1.0 V change indicates a defective beam or a circuit problem ADXL05 and may be a consideration in component selection—
such as an open or shorted pin or component. see section on noise.

Operating the ADXL05’s buffer amplifier at Gains > 2, to pro- Care should be taken to reduce or eliminate any leakage paths
vide full-scale outputs of less than ± 5 g, may cause the self-test from the demodulator capacitor pins to common or to the +5 V
output to overdrive the buffer into saturation. The self-test may pin. Even a small imbalance in the leakage paths from these pins
still be used in the case, but the change in the output must then will result in offset shifts in the zero-g bias level. As an example,
be monitored at the VPR pin instead of the buffer output. an unbalanced parasitic resistance of 30 MΩ from either de-
modulator pin to ground will result in an offset shift at VPR of
Note that the value of the self-test delta is not an exact indica- approximately 50 mV. Conformal coating of PC boards with a
tion of the sensitivity (mV/g) of the ADXL05 and, therefore, high impedance material is recommended to avoid leakage prob-
may not be used to calibrate the device for sensitivity error. lems due to aging or moisture.
In critical applications, it may be desirable to monitor shifts in
the zero-g bias voltage from its initial value. A shift in the 0 g MINIMIZING EMI/RFI
bias level may indicate that the 0 g level has shifted which may The architecture of the ADXL05 and its use of synchronous de-
warrant an alarm. modulation make the device immune to most electromagnetic
(EMI) and radio frequency (RFI) interference. The use of syn-
Power Supply Decoupling
chronous demodulation allows the circuit to reject all signals ex-
The ADXL05 power supply should be decoupled with a 0.1 µF
cept those at the frequency of the oscillator driving the sensor
ceramic capacitor from +5 V pin of the ADXL05 to common
element. However, the ADXL05 does have a sensitivity to RFI
using very short component leads. For other decoupling consid-
that is within ± 5 kHz of the internal oscillator’s nominal fre-
erations, see EMI/RFI section.
quency of 1 MHz and also to any odd harmonics of this fre-
Oscillator Decoupling Capacitor, C2 quency. The internal oscillator frequency will exhibit part to
An oscillator decoupling capacitor, C2, is used to remove part variation in the range of 0.5 MHz to 1.4 MHz.
1 MHz switching transients in the sensor excitation signal, and
In general the effect is difficult to notice as the interference must
is required for proper operation of the ADXL05. A ceramic
match the internal oscillator within ± 5 kHz and must be large
capacitor with a minimum value of 0.022 µF is recommended
in amplitude. For example: a 1 MHz interference signal of
from the oscillator decoupling capacitor pin to common. Small
20 mV p-p applied to the +5 V power supply pin will produce a
amounts of capacitor leakage due to a dc resistance greater than
200 mV p-p signal at the VPR pin if the internal oscillator and
1 MΩ will not affect operation (i.e., a high quality capacitor is
interference signals are matched exactly or at odd harmonics. If
not needed here). As with the power supply bypass capacitor,
the same 20 mV interference is applied but 5 kHz above or be-
very short component leads are recommended. Although
low the internal oscillator’s frequency, the signal level at VPR will
0.022 µF is a good typical value, it may be increased for reasons
only be 20 mV p-p in amplitude.
of convenience, but doing this will not improve the noise perfor-
mance of the ADXL05. Power supply decoupling, short component leads (especially for
capacitors C1 and C2), physically small (surface mount, etc.)
Demodulator Capacitor, C1
components and attention to good grounding practices all help
The demodulator capacitor is connected across Pins 2 and 3 to
to prevent RFI and EMI problems. Good grounding practices
set the bandwidth of the force balance control loop. This capaci-
include having separate analog and digital grounds (as well as
tor may be used to approximately set the bandwidth of the ac-
separate power supplies or very good decoupling) on the printed
celerometer. A capacitor is always required for proper operation.
circuit boards. A single ground line shared by both the digital
The frequency response of the ADXL05 exhibits a single pole and analog circuitry can lead to digital pulses (and clock signals)
roll-off response, see Figure 4. interfering with the sensor’s onboard oscillator. In extreme
A nominal value of 0.022 µF is recommended for C1. In gen- cases, a low cost radio frequency choke (≈10 µH) may be
eral, the design bandwidth should be set 40% higher than the needed in series with the accelerometer’s power supply pin.
minimum desired system bandwidth due to the ± 40% tolerance, This, together with the recommended 0.1 µF power supply by-
to preserve stability C1 should be kept > 0.01 µF. pass capacitor, will form an effective RF filter. The use of an RF
choke is preferred over a resistor since any series resistance in
The demodulation capacitor should be a low leakage, low drift
the power supply will “unregulate” the device from the supply,
ceramic type with an NPO (best) or X7R (good) dielectric.
degrade its power supply rejection and reduce its supply voltage.
In general, it’s best to use the recommended 0.022 µF capacitor
across the demodulator pins and perform any additional low-
pass filtering using the buffer amplifier. The use of the buffer for

–18– REV. B
 ADXL05
OUTLINE DIMENSIONS
Dimensions shown in inches and (mm).
REFERENCE PLANE
0.750 (19.05)
0.500 (12.70) 0.160 (4.06)
0.185 (4.70)
0.165 (4.19) 0.110 (2.79)

6
7
5
0.335 (8.51) 8 0.045 (1.14)
0.305 (7.75) 0.115
4 0.027 (0.69)
(2.92)
0.370 (9.40) BSC 9
3
0.335 (8.51)
10
2
1

0.230 (5.84)
0.040 (1.02) MAX
BSC 0.034 (0.86)
0.045 (1.14) 0.027 (0.69)
0.010 (0.25)
SEATING PLANE

REV. B –19–
–20–
PRINTED IN U.S.A. C2028b–5–3/96