3.8 KiB
#Reading #Sensors #ADC The quick brown fox jumps over the lazy dog. The dog stays blissfully asleep. :)
4.1 Input Characteristics of Interface Circuits
What is an interface circuit?
Interface circuits are almost always required to make use of a sensor. Sensor signals are usually, very small currents or voltages, have a significant amount of noise in them, and may be very sensitive to circuit conditions. This is a lot of what this first section talks about: why do sensors need interface circuits, and what pieces make up these circuits?
A primary reason interface circuits exist is because sensor dynamics often use different physical phenomena than may be desired. For example, most thermistors use resistance to measure temperature, but resistance is not a quantity that can be directly observed by a microcontroller. Instead, this resistance must be converted into a voltage signal that a microcontroller can read. This, in a way that maintains a non-interfering level of current or 'load' into the sensor, is what an interface circuit must do.
Interface circuits can be either extremely technical and detailed, or simple and crude. Extremely well designed interface circuits will carefully orient traces of a board s.t. signals between PCB layers will not interfere with sensor dynamics.
4.2 Amplifiers
This chapter covers operational amplifiers, which the author calls an OPAM as opposed to op-amp name I have been taught. OPAMs have a few key characteristics:
- They have extremely high input impedance
- The input bias current is very low
- They are stable to a large range of supply conditions
- They are stable to a large range of environmental conditions
- And more things that are discussed in the book...
The author then talks about a series of different OPAM configurations and how they are used in practice.
4.3 Excitation Circuits
I have not read this 3/24
4.4 Analog to Digital Converters
Analog to digital converters (ADC) are a critical piece of an interface circuit when the continuous time sensors interact with a discrete controller or storage device, such as a modern computer. They convert a reference signal (usually a voltage), and turn it into a binary number that a computer can read. These ADC converters usually output a value between 0 and using fractional binary numbers, where 0 indicates the minimum value, and 1 the saturation / reference / maximum value.
4.4.4 Successive Approximation Converter
This type of ADC is very common in a monolithic form. These ADC work by using a comparator with precise reference voltages to evaluate the closest digital equivalent of the measured signal. These ADC start with the most significant bit (MSB), and iterate through all possible test bits to determine the final binary number. These ADC take time to do these comparisons, and as such have two distinct features: 1) They use a sample-and-hold architecture, and 2) they take several clock cycles to obtain a measurement.
The SAC ADC has to perform its approximation over several time steps by its nature of being a digital system. The sensor signal however, may change in this time, and if a high-frequency noise is included, may wildly vary between successive approximations. Because of this, the SAC ADC includes a circuit that holds the measurement for a certain sample while the SAC is performed. At the end of the sampling, the sample-and-hold is then cleared. This memory is analog.
Clock cycles determine the time steps for each successive approximation in the ADC, and because these are often built into an existing microcontroller or chip, they often share the clock cycle of the larger device. This creates an issue that the larger system has to wait for the ADC to finish before it can access the measurement. This means that the speed of the control system is ultimately determined by the speed of the ADC.