Data Acquisition¶
In most apparatus we need to record something about the physical system in a format to analyze later, commonly digital data. In this section we will examine how the data acquisition chain works and discuss each part and its role in making sure you get the best quality and most valid data. Doing data acquisition incorrectly can result in poor quality data, or data that is incorrectly sampled, often leading to mis-interpretation.
Physical System¶
The physical system is the actual thing we care about measuring and recording. It generally does not lend itself to direct recording. For example measuring the displacement of an object or the mass of an object is not natively a digital process. The most important thing when considering the physical system is considering how you can attach instrumentation without modifying the behavior of the system. It is also important to ensure that the system is functioning correctly and safely.
Transducer/Sensor¶
The transducer or sensor is the device that will turn the physical quantity (temperature, acceleration, mass, etc) into an electrical signal. Transducers transform one quantity into another through a well defined transfer function or calibration. Transducers often use the electrical quantities of voltage, resistance, or current to encode information about the physical system. We will go over many transducers in detail later in the course.
Signal Conditioning¶
Most transducers do not produce a signal that is immediately ready to be recorded. The act of modifying the signal to make it useful and recordable is known as signal conditioning. Common steps in signal conditioning include conversion to an easy to measure quantity (i.e. resistance to voltage), amplification (increasing the amplitude or output range of the signal), and filtering (cleaning up the signal). Each of these steps has a devoted section that we will discuss during the course.
Analog-to-Digital Conversion¶
Once the signal is in an easy to measure quantity (most commonly voltage) and in the ideal range for your hardware, it is time to convert the information to a digital quantity that we can store and analyze later. This process is known as analog-to-digital conversion (A2D or ADC). The process of converting an continuous quantity (an analog voltage that can have any value) to a digital representation that much have a discrete number of steps of values, results in a loss of some information. The resolution of the ADC is generally described in the number of bits used to represent the analog quantity, with higher bit ADCs having a finer resolution. The time resolution required should also be considered. Sampling must be a minimum of twice the maximum frequency of interest (the Nyquist criteria). In practice it is wise to always sample at least 5-10 times the maximum frequency of interest to avoid any aliasing issues and make filtering/decimation an easier process.
Computer/Recording¶
The ADC system is generally connected to a computer (embedded or otherwise) to store the data to a storage media. Often the incoming readings are visualized in real-time to help the equipment operator and/or used to control the system being monitored.