Ad522 integrated data acquisition amplifier can carry out high-precision data acquisition under harsh working conditions. It has the advantages of good linearity, high common mode rejection ratio, low voltage drift and low noise. It is suitable for most 12 bit data acquisition systems. Ad522 is usually used for bridge sensor amplifier composed of resistance sensors (electric heating regulator, strain gauge, etc.), process control, instruments, information processing and medical instruments.
Ad522 has the following characteristics:
● low drift: 2.0 μ V/℃（AD522B）；
● low nonlinearity: 0.005% (g = 100);
● high common mode rejection ratio: 110dB (g = 1000);
● low noise: 1.5 μ Vp-p（0.1～100Hz）；
● programmable gain of single resistance: 1 ≤ g ≤ 1000;
● it has output reference terminal and remote compensation terminal;
● internal compensation is available;
● no other peripheral devices are required except gain resistance;
● adjustable offset, gain and common mode rejection ratio.
Ad511 adopts 14 pin dip package, and its structure and appearance are similar to the commonly used ad521. Figure 1 shows the pin arrangement of ad522. Table 1 is the functional description of each pin.
Table 1 pin function description
2 main features of ad522
Ad522 can provide high-precision signal conditioning. Its output offset voltage drift is less than 1V / ℃, and the input offset voltage drift is less than 2.0 μ V / ℃, the common mode rejection ratio is higher than 80dB (110dB when g = 1000), the maximum nonlinear gain is 0.001% when g = 1, and the typical input impedance is 109 Ω.
Ad522 uses automatic laser adjusted film resistance, so it has small tolerance, low loss, small volume and reliable performance. At the same time, ad522 also has the best characteristics of monolithic circuit and standard component amplifier. It is a high cost-effective amplifier.
To adapt to different accuracy requirements and operating temperature ranges, the ad522 is available in three levels. “A” and “B” are industrial grade, which can be used for – 25 ~ + 85 ℃. “S” is military grade, used for – 55 ~ + 125 ℃. The ad522 provides four drift options. The maximum drift of output offset voltage increases with the increase of gain. The voltage error caused by offset current drift is equal to the product of offset current drift and asymmetric source resistance. In addition, the nonlinear gain of ad522 will increase with the decrease of closed-loop gain.
The common mode rejection ratio of ad522 amplifier is measured under ± 10V ambient conditions, and an asymmetric resistance with a resistance of 1K Ω is used. In the case of low gain, the common mode rejection ratio mainly depends on the stability of thin film resistance, but due to the influence of gain bandwidth, ad522 is relatively constant at frequencies below 60Hz. However, in the limited bandwidth, the phase shift of ad522 will increase with the increase of DC common mode rejection ratio.
In terms of dynamic performance, the stability time, unit gain bandwidth and gain of ad522 are directly proportional.
3.1 typical applications
Fig. 2 is a typical circuit diagram of ad522 applied to bridge amplifier circuit. The circuit can obtain the best performance in the environment of low voltage, high impedance and large noise. Of course, this requires proper shielding and grounding. In the circuit of Fig. 2, the signal ground is directly connected with the ad522 to form the bias current loop of the input amplifier. When designing, users can connect directly like the circuit shown in Figure 2 or indirectly through a resistance less than 1m Ω.
In order to reduce noise, the input pin and gain resistance should be shielded. Using bootstrap circuit, passive data protection can be realized to improve AC common mode rejection ratio. This method can reduce the differential phase shift and restrain the decrease of system bandwidth.
Using this balance design in Figure 2, ideal performance can be obtained without using external bypass capacitor. However, if the signal source is placed at a distance (10 feet or more) or carries more than thousands of millivolts of noise, it is necessary to use bypass capacitors to obtain better performance.
The reference terminal and the compensation terminal can compensate the long-distance load, and can also be used to adjust the common mode rejection ratio, increase the bootstrap of the output circuit and adjust the output drift.
When in use, RG should be as close to ad522 as possible. Too long conductor will increase parasitic capacitance and produce phase shift, resulting in reduced common mode rejection ratio of high frequency part.
When the frequency is lower than 10Hz, RG connected at a distance will not cause stability problems. When g = 1, the 200m Ω leakage impedance between RG leads will cause a gain error of 0.1%.
3.2 special applications
When adjusting the drift and gain, the gain accuracy is determined by RG. RG recommends the use of precision resistors with a temperature coefficient of 10ppm / ℃. The drift and gain are adjusted by the laser, so it can be suitable for most applications. If further adjustment is required, it can be adjusted by a potentiometer, but a high-quality potentiometer (25ppm) must be used to ensure voltage drift characteristics.
The common mode rejection ratio can be adjusted according to the circuit shown in Fig. 3. Using this circuit, a small common mode rejection ratio can be increased to 10dB at low gain. The input signal with a peak to peak value of 20V at low frequency needs to pass through two equivalent source resistors rs. adjusting the potentiometer in Figure 3 can make the AC output of the amplifier zero.
The compensation output can be realized by long-distance load compensation or output current bootstrap. Using the circuit of Fig. 4, the bootstrap drift error can be reduced in the closed loop. When not in use, the compensation output terminal shall be connected to the output terminal.
3.3 error adjustment
The 0 ~ 1V output voltage zero static error frequency converter provided by ad522 is a 1K Ω unbalanced source. The accuracy of ad522b is high when the operating temperature range is 0 ~ 50 ℃. Table 2 lists the error sources and their influence on the system accuracy.
Table 2 Influence of error source on system accuracy
8-bit calibration operation is allowed when the influence on absolute accuracy does not exceed ± 0.2%. In the data sampling system controlled by computer or microprocessor, the automatic readjustment operation can make the noise value generated by gain and offset drift zero. At this time, distortion and common mode rejection ratio are the only error sources. In this case, the intended purpose can be achieved by using 12 bit operation.
Three common errors of ad522 are described below:
(1) Gain error
By adjusting RG, the absolute gain error can be 0. When the operating temperature range is exceeded, the change of internal resistance will cause gain drift. However, because the effect of gain drift is linear, the resolution will not be reduced. An “intelligent” system can generally correct these errors in the process of automatic cyclic correction. When g = 10, the nonlinear gain of ad522 will not exceed 0.002%.
(2) Offset drift and lead current error
The input part of ad522 takes special measures to reduce offset drift. When the unbalanced impedance of the sensor is less than 2K Ω, the error caused by offset current drift is much less than that caused by offset voltage drift, so it can be ignored. In most applications, although the initial compensation voltage is set to zero by the laser, other measures still need to be taken when further correcting the initial system bias voltage. In the above example, the total compensation bias drift is only ± 0.014%, so it will not affect the resolution (which can be realized by automatic cyclic correction adjustment).
(3) Common mode rejection ratio and noise error
Because the error caused by common mode rejection ratio and system noise cannot be corrected, there must be strict requirements for common mode rejection ratio and system noise characteristics. The common mode resistance of ad522 is a thin film resistance adjusted by automatic laser, and its stability is very good. However, regular recalibration is necessary to maintain performance. In general, the error caused by unadjusted common mode rejection ratio and noise accounts for ± 0.0065% of the whole gain, so this is also one of the main causes of resolution error.
Responsible editor: GT