Introducción: We use a lot of pressure sensors, and we often find that pressure sensors will drift after a period of use. What causes the pressure sensor to drift? How can we eliminate the pressure sensor drift during design?
Causes of sensor drift
Sensor drift refers to the phenomenon that the output value of the sensor changes over time. This drift may cause inaccurate sensor measurement results, affecting its reliability and stability in practical applications. There are many reasons for sensor drift, which will be introduced one by one below.
Temperature change: Temperature change is one of the common causes of sensor drift. Temperature changes can cause the expansion and contraction of the material inside the sensor element, which in turn affects the mechanical structure and electrical characteristics of the sensor, causing the output value to drift. For example, an increase in temperature will increase the resistance value of a resistor sensor, resulting in a higher output value.
Power supply changes: The output value of the sensor is affected by the supply voltage. When the supply voltage changes, the output value of the sensor will also change. This is because changes in the supply voltage will cause the working state of the internal circuit of the sensor to change, which in turn affects the amplitude and stability of the output signal.
Long-term use: Long-term use is also an important cause of sensor drift. During use, the sensor may be affected by mechanical, chemical or thermal expansion and contraction factors, causing changes in its internal structure, which in turn causes the output value to drift. Además, the sensor may also be affected by external environmental factors such as vibration and impact, further exacerbating the drift phenomenon.
Sensor aging: Over time, the performance of the sensor may gradually decline and drift may occur. This is because the materials and components inside the sensor will age with the increase in use time, causing its physical properties to change. For example, the electrolyte inside the sensor will gradually drain away, causing its sensitivity and stability to decrease, which in turn causes the output value to drift.
Environmental influence: The drift of the sensor may also be affected by environmental factors. For example, changes in environmental factors such as air pressure, humidity, and light may cause the sensor output value to drift. This is because changes in environmental factors will change the interaction between the sensor and the object to be measured, thereby affecting the measurement accuracy and stability of the sensor.
In the early days of the development of pressure sensors, glass powder was used to seal the diffused silicon chip and the metal base. The disadvantage was that there was a large stress around the pressure chip, and even after annealing, the stress could not be completely eliminated. When the temperature changes, due to the different thermal expansion coefficients of metal, glass and diffused silicon chips, thermal stress will be generated, causing the zero point of the sensor to drift. This is why the zero point thermal drift of the sensor is much larger than the zero point thermal drift of the chip. If silver paste and terminal welding are not handled properly, it is easy to cause unstable contact resistance. Especially when the temperature changes, the contact resistance is more likely to change. These factors are the reasons for the large zero point drift and temperature drift of the sensor.
Semiconductor theory analysis of the cause of zero-point thermal drift: Only when the doping concentration and resistance value of the resistor are consistent can the zero-point output voltage of the bridge be small and the zero-point thermal drift is also small, which is very beneficial to improving the performance of the sensor. However, it is not easy to achieve uniform doping distribution during diffusion, so the varistor strips are required to be as close as possible and as short as possible.
Circuit analysis of the cause of zero point thermal drift: Ideally, the resistance values of the four diffused resistors that make up the Wheatstone bridge should be equal. Zero point temperature drift is caused by the change of diffused resistor value with temperature. Within a certain temperature range, the resistance value increases with the increase of temperature, eso es, the temperature coefficient R of the diffused resistor is positive.
Solutions to sensor drift problems
En general, the zero drift compensation of pressure sensors can be divided into two directions: hardware compensation and software compensation.
Hardware zero compensation method: Appropriate constant resistance method in series and parallel on the bridge arm: bridge arm thermistor compensation method, bridge external series and parallel thermistor compensation method, dual bridge compensation technology, transistor compensation technology, etc..
Optimize circuit design: Reasonable circuit design can reduce the impact of sensor drift. For example, the use of temperature compensation circuit can correct the impact of temperature changes on sensor output values and improve measurement accuracy and stability. Además, circuit design methods such as filtering and amplification can also be used to eliminate the impact of power supply changes and environmental interference on sensors.
Software compensation zero drift method:In the signal acquisition process, from the time when the trigger signal does not occur to the time when the acquisition is triggered and after the acquisition is completed, the input signal is zero and the output signal is not zero. This collected output data exists in the form of random noise, which is meaningless for data calculation and processing. We define the signal value collected during this period as zero drift.
The software methods adopted are:
Polynomial fitting specification method. Since in actual measurement, the temperature, pressure and other physical quantities measured by the pressure sensor will not have a strict linear relationship with the output value, the functional relationship is often in the form of a polynomial. Polynomials can be used to fit nonlinear signals, and the key is to solve their coefficients.
RBF neural network method. Basic principle: Usually the formula method in the zero-point temperature compensation software algorithm is relatively complex, and the fitting accuracy is often limited. The artificial neural network method has the advantages of a small number of samples, a simple algorithm, the ability to approximate arbitrary functions, and good application prospects.
Además, the software method also includes table lookup method, interpolation method, etc..
To reduce the impact of drift, the following measures can be taken:
Stabilize the temperature: Keep the sensor in a constant temperature state as much as possible to avoid the impact of temperature fluctuations.
Use temperature compensation measures: Add a temperature sensor inside the sensor to perform correction compensation by sensing temperature changes.
Choose a suitable substrate bonding method: A suitable substrate bonding method can reduce the impact of mechanical stress.
Choose an independent amplifier: Use an independent amplifier to amplify the signal, which is not affected by other external factors and can reduce drift problems.
Use automatic calibration technology: Through automatic calibration, the sensor can maintain a stable output under different temperature, humidity and other environments.
Choose a high-precision sensor: The drift of a high-precision sensor is small, which can reduce the impact.
Process the drift data: By collecting data over a period of time and averaging the drift data, the impact of drift on the measurement results can be reduced.
