



Inelectronics, asample and hold (also known as sample and follow) circuit is ananalog device that samples (captures, takes) thevoltage of a continuously varyinganalog signal and holds (locks, freezes) its value at a constant level for a specified minimum period of time. Sample and hold circuits and relatedpeak detectors are the elementary analogmemory devices. They are typically used inanalog-to-digital converters to eliminate variations in input signal that can corrupt the conversion process.[1] They are also used in electronic music, for instance to impart a random quality to successively-played notes.
A typical sample and hold circuit storeselectric charge in acapacitor and contains at least one switching device such as aFET (field effect transistor) switch and normally oneoperational amplifier.[2] To sample the input signal, the switch connects the capacitor to the output of abuffer amplifier. The buffer amplifier charges or discharges the capacitor so that the voltage across the capacitor is practically equal, or proportional to, input voltage. In hold mode, the switch disconnects the capacitor from the buffer. The capacitor is invariably discharged by its ownleakage currents and useful load currents, which makes the circuit inherentlyvolatile, but the loss of voltage (voltage drop) within a specifiedhold time remains within an acceptable error margin for all but the most demanding applications.
Sample and hold circuits are used in linear systems. In some kinds ofanalog-to-digital converters (ADCs), the input is compared to a voltage generated internally from adigital-to-analog converter (DAC). The circuit tries a series of values and stops converting once the voltages are equal, within some defined error margin. If the input value was permitted to change during this comparison process, the resulting conversion would be inaccurate and possibly unrelated to the true input value. Suchsuccessive approximation converters will often incorporate internal sample and hold circuitry. In addition, sample and hold circuits are often used when multiple samples need to be measured at the same time. Each value is sampled and held, using a common sample clock.
For practically all commercialliquid crystalactive matrix displays based on TN, IPS or VA electro-optic LC cells (excluding bi-stable phenomena), eachpixel represents a small capacitor, which has to be periodically charged to a level corresponding to thegreyscale value (contrast) desired for a picture element. In order to maintain the level during a scanning cycle (frame period), an additional electric capacitor is attached in parallel to each LC pixel to better hold the voltage. Athin-film FET switch is addressed to select a particular LC pixel and charge the picture information for it. In contrast to an S/H in general electronics, there is no output operational amplifier and no electrical signal AO. Instead, the charge on the hold capacitors controls the deformation of the LC molecules and thereby the optical effect as its output. The invention of this concept and its implementation in thin-film technology have been honored with theIEEE Jun-ichi Nishizawa Medal.[3]

During a scanning cycle, the picture doesn't follow the input signal. This does not allow the eye to refresh and can lead to blurring during motion sequences, also the transition is visible between frames because the backlight is constantly illuminated, adding todisplay motion blur.[4][5]
Sample and hold circuits are also frequently found onsynthesizers, either as adiscrete module or as an integral component. They are used to take periodic samples of an incoming signal, typically as a source of modulation for other components of the synthesizer. When a sample and hold circuit is plugged into awhite noise generator the result is a sequence of random values, which - depending on the amplitude of modulation - can be used to provide subtle variations in a signal or wildly varying random tones.[6]
To keep the input voltage as stable as possible, it is essential that the capacitor have very lowleakage, and that it not be loaded to any significant degree which calls for a very highinput impedance.