In DC to low-frequency sensor signal conditioning applications, relying solely on the common-mode rejection ratio (CMRR) of the instrumentation amplifier is not enough to ensure effective noise suppression in harsh industrial environments. To prevent unwanted noise from affecting the system, the components in the low-pass filter at the input of the instrumentation amplifier must be carefully matched and adjusted. This ensures that internal EMI/RFI filtering and CMRR can work effectively, reducing noise to an acceptable signal-to-noise ratio (SNR).
As an example, consider the low-pass filter shown in Figure 1. A resistive sensor is connected to a high-impedance instrumentation amplifier through a network consisting of RSX and CCM. Ideally, if the CCM capacitors on each input pin are perfectly matched, any common-mode noise will be significantly reduced before it reaches the INA input.
When the common-mode filter capacitor (Ccm) is perfectly matched, the noise is almost completely attenuated. Figure 2 shows this result from a TINA SPICE simulation where a 100mVpp, 100kHz common-mode error signal is injected into the INA333 input.
However, off-the-shelf capacitors typically have a tolerance of 5% to 10%. If the Ccm values on each pin do not match, the differential mismatch can be as high as 20%, which can lead to significant noise issues. Figure 3 illustrates a case where there is a capacitor mismatch, showing the common-mode noise (eN) at the output of the resistive sensor.
This input mismatch (?C) causes a cutoff frequency error, allowing common-mode noise to differentially enter the INA input and then pass through the gain stage, resulting in an error voltage. Equations 1 through 3 demonstrate how much common-mode noise reaches the input.
Assuming the sensor signal frequency (Vsensor) is much lower than the noise-cut frequency of all common-mode filters (i.e., fC ≥ 100 × fsensor), and RS1 = RS2, the noise becomes differential. The resulting signal (eIN) becomes a common-mode noise signal (eN) of VIN, as shown in Equation 4.
Equation 4 further illustrates that when a 100mVpp, 100kHz common-mode error signal is applied to the INA333, and there's a 10% RC mismatch at a 1.6kHz cutoff frequency, the resulting errors are as shown in Figure 5.
Figure 5 presents a more efficient and commonly used method of input filtering, where a differential capacitor (Cdiff) is added between the instrumentation amplifier inputs. Although this approach doesn't completely eliminate the problem, Cdiff must be adjusted according to two key criteria: the cutoff frequency should be high enough to avoid the signal bandwidth and stabilize the filtering, while also being low enough to reduce common-mode noise to an acceptable level, allowing the instrumentation amplifier’s CMRR to handle residual noise and achieve an acceptable SNR.
Equation 5 outlines the general principle for making this adjustment. Figure 6 compares the VinP and VinN plots with and without Cdiff. It's important to note that the output of the INA333 changes without differential capacitance, and this difference is amplified, contributing to noise that reduces the SNR. When Cdiff = 1 μF, the difference between VinP and VinN is minimized.
Figure 7 demonstrates the overall improvement in noise performance of the INA333 output when Cdiff = 1 μF.
In summary, the low-pass filter at the front end of the instrumentation amplifier should include a differential capacitor that is at least 10 times larger than the common-mode capacitor. This helps minimize the impact of Ccm mismatches, converting common-mode noise into differential noise and significantly improving the filter’s efficiency.
electric linear actuator,linear electric motor,precision linear actuator,electro hydraulic actuator,high speed linear actuator
Suzhou Johnson Automation Technology Co., Ltd. , https://www.cn-johnson.com