Characteristics and accuracy of the hottest accele

  • Detail

The adxl150 acceleration sensor is a complete acceleration measurement system integrated on a monolithic integrated circuit. It is very suitable for harsh industrial and automotive environments. This paper will introduce the characteristics and application methods of the sensor, and analyze the main factors affecting its measurement accuracy

the monolithic integrated acceleration sensor ADXL50 is a product of American ad company. Its main characteristics are small size, good directivity, high precision, and small time drift and temperature drift; Working under +5 V single power supply, it is easy to use and requires less peripheral components; It has high reliability. It can withstand the acceleration impact of 500 g when powered on and 2000 g when not powered on without damage. Therefore, it is especially suitable for working in harsh industrial environment and automobile test system

1. The main technical indicators of ADXL50

measurement sensitivity: 19 mv/g

full scale measurement range: ± 50 g

frequency response range: DC1 kHz

self inspection can be carried out under the control of switch signal

built in buffer amplifier can be used for output sensitivity and zero acceleration output level adjustment

2. Working principle of ADXL50

Figure 1 schematic block diagram and pin diagram of acceleration sensor

adxl50 is a complete acceleration measurement system. It integrates a sensor component and the circuits related to the completion of acceleration measurement, including oscillator, demodulator, preamplifier, buffer amplifier, reference power supply and temperature compensation circuit, into a silicon chip. The chip is packaged in a 10 pin to100 shell as shown in Figure 1 (b). These 10 pins are defined as follows:

① is a 5 V power supply

② and ③ are external demodulator capacitors

④ is the decoupling capacitance of the oscillator

⑤ is the public end

⑥ refers to 3.4 V reference power output

⑦ refers to self-test digital signal input

⑧ refers to preamplifier output

⑨ is buffer amplifier output

⑩ is the inverting input of buffer amplifier

Figure 1 (a) is the principle block diagram of the acceleration sensor. The sensor component in the figure is the key component of the whole device, and its principle is shown in Figure 2. Only one capacitance unit of the sensor component is shown in Figure 2, which is composed of one center plate and two mutually independent fixing plates, and the two fixing plates are equidistant arranged on both sides of the center plate. The actual sensor components include a group of 42 center pieces, which are fixed on a center arm, and the center arm can move along the horizontal direction in the figure with 42 center pieces

each center piece and two fixed pieces on both sides of each center piece form a pair of capacitors CS1 and CS2. CS1 and CS2 are connected in series on the circuit to form a capacitive voltage divider. The midpoint of the voltage divider is the movable center piece. The oscillator in the acceleration sensor (see Figure 1 (a)) generates a pair of 1 MHz square wave signals with equal amplitude and opposite phase when powered on. The pair of square wave signals are added to different fixed plates on the side of the center plate. When there is no acceleration, the center plate is in the center of the two fixed plates. At this time, cs1=cs2. Therefore, the voltage on the center plate at the midpoint of the capacitive voltage divider is 0. When the sensor accelerates, due to the inertia, the center arm with the center piece moves in the direction opposite to the acceleration relative to the fixed piece. At this time, CS1 and CS2 are not equal (see Figure 2), which generates a voltage signal on the center piece. This signal is processed by the demodulator and amplified by the preamplifier. At the output end of the preamplifier, a voltage signal Vpr proportional to the acceleration is obtained and connected to the eighth pin of the ADXL50

Figure 2 sensor components

3. Basic measuring circuit of ADXL50

the basic measuring circuit composed of ADXL50 is shown in Figure 1 (a). Three external capacitors and three external resistors are used

c1 is the demodulator capacitance to determine the bandwidth of the measurement system, C2 is the oscillator decoupling capacitance, and C3 is the power decoupling capacitance. The capacity of C1 and C2 is generally 0.022 μ F. C3 is generally 0.1 μ F is OK

adxl50 before delivery, the manufacturer has adjusted the measurement sensitivity of its preamplifier to 19 mv/g, and its output voltage Vpr at 0 g is 1.8 v. Therefore, if the acceleration is ± 50 g, vpr=1.8 ± 0.95 V; if the acceleration is ± 20 g, vpr=1.8 ± 0.38 v. Because the signal on Vpr is not standardized, it is not suitable for direct reading for display or digital processing. In addition, the maximum injection current of ADXL50 preamplifier can only be 25 μ A. The driving capacity is very small. Therefore, the signals on the Vpr need to be further processed to meet the needs. Generally, several resistors are connected externally to form an amplification link with the buffer amplifier to adjust the 0 g potential of the Vpr of the sensor and improve the sensitivity of the output signal. The variation range of the buffered and amplified acceleration signal should preferably be between 0.5 ~ 4.5 V, so that there is a certain margin from the high and low ends of the power supply. If the change amplitude of the acceleration in the positive and negative directions is similar, the 0 g potential can be adjusted around the huge market demand of 2.5 v. In this way, the positive and negative acceleration signals have a variation range of ± 2 V on the sensor output port Vout. If the range of acceleration to be measured is ± 50 g, then vpr=1.8 ± 0.95 v. to make vout=2.5 ± 2.0 V, it is necessary to amplify the AC component in the signal. It can be seen from Figure 2 that the amplification factor of the buffer amplifier is -r3/r1. Therefore, taking r1=50 K Ω, r3=105 K Ω can be determined

r3/r1=vout change/vpr change =2.0/0.95=2.10

in Figure 1 (a), one end of R2 is grounded. In order to make vout=2.5 V at 0 g, R2 should meet the following conditions:

r2= (1.8 V × R3)/(vout-1.8 V) =270 K Ω (1)

Figure 3 shows a more common circuit. The 3.4 V reference voltage provided by ADXL50 chip is used, and a potentiometer RT is added. At this time, Vout is determined by the following formula:

vout=r3/r1 (1.8 v-vpr) +r3/r2 (1.8 v-vx) +1.8 V (2)

it can be seen that 0 g voltage of Vout can be adjusted in a larger range, generally r2=100 K Ω

Figure 3 common circuit of ADXL50

now an example is given to illustrate the selection of circuit parameters. The maximum positive acceleration of a mechanical device can reach 10 g, and the maximum negative acceleration can reach -2g. The mechanical characteristics can be measured with ADXL50, and the circuit parameters can be selected as follows. The speed variation range of this device is 10 g- (-2g) =12g, especially when the equipment is abnormal occasionally. If the variation range of output voltage Vout is 4.5 v-0.5 v=4.0 V, the sensitivity of the system is 4.0 v/12 g=0.333 v/g. It is known that the sensitivity of Vpr pin is 0.019 v/g, so the amplification factor of buffer amplifier should be 0.333 v/0.019 v=17.53. Thus, r3/r1=17.53 is obtained. When selecting R1, note that the injection current of Vpr port shall not be greater than 25 μ A。 Select r1=25 K Ω to get r3=25k Ω × 17.53=438kΩ。 The 0g potential of the system can be set as 0.5 v+0.667 v=1.167 V, and vx=1.95 V can be obtained from formula (2). Select r2=100 K Ω and adjust RT to make vx=1.95 v

during circuit design, the following points should be noted:

(1) when selecting R1, the injection current of the preamplifier should be less than 25 μ A。

(2) when the acceleration is small, such as less than ± 5 g, the signal-to-noise ratio of the measured signal will be much lower than the value of ± 50 G. When the bandwidth requirement is not high, the demodulator capacitance C1 can be appropriately increased, the response speed of some sensors can be reduced to improve the signal-to-noise ratio, or a filter link can be added to the buffer amplification stage to improve it

(3) although the compensation circuit is set in ADXL50, it is impossible to fully compensate. When the acceleration is small and the temperature changes greatly, the temperature drift and time drift will also have a greater impact. It can be improved by measuring the AC component of acceleration signal or periodically reading 0 g value for correction

(4) ADXL50 works based on the capacitance measurement mechanism, so do not add capacitive load to the Vpr pin. Adding a 10 pf capacitor to this pin will cause a 1 MV signal error. It is best to output the sensor signal from the Vout pin

4. ADXL50 self calibration

if there is no standard acceleration calibration equipment at hand, the 0 g potential and magnification of ADXL50 can be easily calibrated to a higher accuracy by using the 1 g gravitational acceleration of the earth. Figure 4 shows the relationship between the output on the ADXL50 (to100 package) Vpr foot and its orientation relative to the ground under the action of gravity. Excellent and perfect after-sales service will be provided in accordance with the following terms: when its sensitive axis points vertically to the ground, ADXL50 displays -1 g reading, and when its sensitive axis points vertically upward, it displays +1 G. When its sensitive axis is parallel to the ground surface, ADXL50 displays 0 g, and the reading is shown in Figure 4. As shown in Figure 4, it can be used even for the odor resistance of shoes and socks. The orientation can be used to calibrate the 0 g potential of ADXL50 and the buffer stage magnification of ADXL50. First, place the ADXL50 horizontally, select two positions with a difference of 180 °, and preliminarily determine the 0 g potential. As shown in Figure 4, take two output potentials in the vertical direction with a difference of 180 ° to obtain the acceleration change of 2G and the corresponding sensor output change, so as to adjust the magnification of the amplification stage. This can be repeated twoorthree times to achieve better results

Figure 4 ADXL50 orientation

Figure 5 shielding ring of ADXL50

5. Several factors affecting ADXL50 measurement accuracy

if the acceleration sensor ADXL50 is used improperly, it will produce large measurement error and reduce the accuracy. Pay special attention to the following points during use:

(1) the ADXL50 has a sensitive axis, and its direction is from the ⑤ pin to the 10th pin (that is, to the shell positioning piece). The sensor measures the acceleration along this axis. Let this axis be the exergy X axis, and the direction perpendicular to the X axis in the bottom plane of the sensor shell is the Y axis. If the included angle between the direction of acceleration a and the XY plane at the bottom of the sensor is α, Then the component of acceleration a on the XY plane is ax=a × cos α。 Let the clamping foot between the acceleration component ax and the sensitive axis of the sensor be α, Then the component of ax on the sensitive axis is ax=ax × cos β= a × cos α× cos β。 Therefore, when installing the acceleration sensor, try to make its sensitive axis consistent with the direction of acceleration, otherwise the measurement results must be modified to avoid excessive error

(2) ideally, any lateral acceleration at right angles to the sensitive axis will not produce an output signal. However, if the direction of the lateral acceleration does not intersect the sensitive axis at right angles, an output signal will be generated. Suppose that the transverse acceleration axis is in the XY plane and the included angle with the acceleration sensitive axis is α, His component on the x-axis is ax=ax × cos α。 set up α At 89 °, axis will generate an error equivalent to 1.7% axis on the sensor

(3) ADXL50 shall be installed on the circuit board for use. When the resonant frequency of the circuit board is close to the frequency of the acceleration to be measured, if the circuit board is slightly loose, resonance will occur, which will make the measurement result larger and cause errors. Therefore, when installing the circuit board, it is necessary to select a suitable position and fix it firmly. When testing the system, it is better to test at different frequencies in order to find possible resonance problems

(4) the demodulator filter capacitor C1 is used to determine the bandwidth of ADXL50 and filter the demodulator signal. The bandwidth is determined by the following formula:

fdb= (28.6/c1) 40%

where: the unit of C1 is μ F。 Generally, c1=0.0022 is selected μ F. Minimum not less than 0.0015 μ F. To avoid system instability or oscillation

reducing C1 can improve the response speed of the system to rapid acceleration changes, but it will increase the output noise of the demodulator, because the noise of the system is proportional to the square root of the bandwidth. A low-pass filter function can be added to the buffer amplifier stage to improve the signal-to-noise ratio. In circuit board design, attention shall be paid to reducing or eliminating C1 pairs

Copyright © 2011 JIN SHI