Datasheet ADXL345 (Analog Devices) - 34

Data Sheet ADXL345
NOISE PERFORMANCE 10k X-AXIS
The specification of noise shown in Table 1 corresponds to
Y-AXIS
the typical noise performance of the ADXL345 in normal power
Z-AXIS )
operation with an output data rate of 100 Hz (LOW_POWER bit
g (µ 1k
(D4) = 0, rate bits (D3:D0) = 0xA in the BW_RATE register,
N IO T
Address 0x2C). For normal power operation at data rates below
IA V
100 Hz, the noise of the ADXL345 is equivalent to the noise at 100
E D N
Hz ODR in LSBs. For data rates greater than 100 Hz, the noise
A L 100 L
increases roughly by a factor of √2 per doubling of the data rate.
A
For example, at 400 Hz ODR, the noise on the x- and y-axes is typically less than 1.5 LSB rms, and the noise on the z-axis is typically less than 2.2 LSB rms.
10
1
0.01 0.1 1 10 100 1k 10k
5 -2 5 For low power operation (LOW_POWER bit (D4) = 1 in the 2
AVERAGING PERIOD, (s)
9 7 0 BW_RATE register, Address 0x2C), the noise of the ADXL345 is Figure 52. Root Allan Deviation constant for all valid data rates shown in Table 8. This value is
130
typically less than 1.8 LSB rms for the x- and y-axes and typically
)
less than 2.6LSB rms for the z-axis.
(% E 120 IS O
The trend of noise performance for both normal power and low
N D X-AXIS
power modes of operation of the ADXL345 is shown in Figure 51.
E 110 Y-AXIS IZ L Z-AXIS A
Figure 52 shows the typical Allan deviation for the ADXL345.
M R 100 O
The 1/f corner of the device, as shown in this figure, is very low,
N F
allowing absolute resolution of approximately 100 µg (assuming
O E 90
that there is sufficient integration time). Figure 52 also shows
G A T
that the noise density is 290 µg/√Hz for the x-axis and y-axis
EN C 80
and 430 µg/√Hz for the z-axis.
PER
Figure 53 shows the typical noise performance trend of the
70
2
2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6
5 -2 5 ADXL345 over supply voltage. The performance is normalized 2 9
SUPPLY VOLTAGE, V
7
S (V)
0 to the tested and specified supply voltage, VS = 2.5 V. In general, Figure 53. Normalized Noise vs. Supply Voltage, VS noise decreases as supply voltage is increased. It should be noted, as shown in Figure 51, that the noise on the z-axis is typically higher
OPERATION AT VOLTAGES OTHER THAN 2.5 V
than on the x-axis and y-axis; therefore, while they change roughly The ADXL345 is tested and specified at a supply voltage of the same in percentage over supply voltage, the magnitude of change VS = 2.5 V; however, it can be powered with VS as high as 3.6 V on the z-axis is greater than the magnitude of change on the or as low as 2.0 V. Some performance parameters change as the x-axis and y-axis. supply voltage changes: offset, sensitivity, noise, self-test, and
5.0
supply current.
4.5 X-AXIS, LOW POWER
Due to slight changes in the electrostatic forces as supply voltage
Y-AXIS, LOW POWER Z-AXIS, LOW POWER 4.0
is varied, the offset and sensitivity change slightly. When operating
) X-AXIS, NORMAL POWER s Y-AXIS, NORMAL POWER
at a supply voltage of V
3.5
S = 3.3 V, the x- and y-axis offset is typically
rm Z-AXIS, NORMAL POWER B S
25 mg higher than at Vs = 2.5 V operation. The z-axis is typically
3.0 (L E
20 mg lower when operating at a supply voltage of 3.3 V than when
IS 2.5 O
operating at VS = 2.5 V. Sensitivity on the x- and y-axes typically
N T 2.0 U
shifts from a nominal 256 LSB/g (full-resolution or ±2 g, 10-bit
P T 1.5 U
operation) at VS = 2.5 V operation to 265 LSB/g when operating
O 1.0
with a supply voltage of 3.3 V. The z-axis sensitivity is unaffected by a change in supply voltage and is the same at V
0.5
S = 3.3 V operation as it is at VS = 2.5 V operation. Simple linear interpolation can be
0
0 5
3.13 6.25 12.50 25 50 100 200 400 800 1600 3200
-2 used to determine typical shifts in offset and sensitivity at other 5 2 9
OUTPUT DATA RATE (Hz)
7 0 supply voltages. Figure 51. Noise vs. Output Data Rate for Normal and Low Power Modes, Full-Resolution (256 LSB/g) Rev. E | Page 33 of 40