Data Acquisition Web

Sensor’s output signals can vary widely. Considering only electric output signals, they can vary from microvolts to volts, from picoamperes to milliamperes. Some need DC or AC polarizing currents or voltages. Some are linear while others are non-linear and need linearizing circuits.

Many output signals are of analog type, but some can be digital as many encoders.

Sensors are not perfect, nor strictly linear, many are affected by temperature changes, but noise, mechanical stress and fumes or contaminants can alter and modify sensors response. Some sensors are more linear than others as for example LVDT (Linear Voltage Displacement Transformer) for an almost perfect linear sensor. Others are very non linear and temperature sensitive as for example ph electrodes, that need frequent calibration around the measurement point. Many compensation schemes are commonly used. Bridges are used on strain gauges, thermistors, capacitive and Hall effect sensors among others. Many amplifier configurations are used to magnify and condition sensors output, most of them based on the instrumentation amplifier concept.

Modern trend is to amplify and condition sensor’s output signal just enough to convert it to a digital signal using. Then on the digital world software linearizing and conditioning is done. To have a rough comparative analog linearizing circuits are mostly parabolic of 2nd or 3rd order but software can achieve easily 10th order corrections. Moreover DSP techniques can be applied to digital signals, filtering and modifying them on ways not possible on the analog world. Some times digital sensor’s output is in the form of a frequency modulated signal. As frequency and time are perhaps the most precise measurements we can do, that response is desirable and very accurate.