Introduction of 55 ADC and DAC Commonly Used Terms

Acquisition time

The acquisition time is the time it takes for the sampling capacitor voltage to stabilize to within 1 LSB of the new input value when the hold state is released (by the sample-and-hold input circuit). The acquisition time (Tacq) formula is as follows:

Aliasing

The "aliasing" frequency is the frequency of the input signal that exceeds the Nyquist frequency, according to the sampling theorem. This means that these frequencies are "folded" or copied to spectral areas near the Nyquist frequency.

To avoid aliasing, all damaging signals must be attenuated to the point where the ADC does not digitize them. Aliasing can be exploited as a beneficial circumstance while undersampling.

Aperture delay

The aperture delay (tAD) in an ADC is the time gap between the sampling edge of the clock signal (the rising edge of the clock signal in the diagram below) and the sampling time. Sampling is conducted when the ADC track-and-hold is switched to the hold state.

Aperture jitter

The variation in aperture delay between samples is referred to as aperture jitter (tAJ), as illustrated in the figure. The aperture jitter of a typical ADC is substantially smaller than the aperture delay.

Binary encoding (unipolar)

For unipolar signals, standard binary is a common coding approach. The positive full-scale value from all 0s (00...000) to all 1s (zero to full scale) is the binary coding range (zero to full scale) (11...111). A 1 (MSB) represents the middle value, which is followed by all 0s (10...000). The encoding is comparable to offset binary encoding, which allows for both positive and negative bipolar transfer functions.

Bipolar input

The term "bipolar" refers to a signal that oscillates between high and low levels at a set reference level. Because the analog ground is frequently used as the input in single-ended systems, the bipolar signal is a signal that swings above and below the ground level.

The signal is not linked to ground in a differential system, but the positive input is to the negative input. The positive input signal might be higher or lower than the negative input signal is a bipolar signal.

Common mode rejection

The ability of a device to suppress the common-mode signals of two inputs is referred to as common-mode rejection. An AC or DC signal, or a combination of the two, can be used as the common-mode signal. The ratio of differential signal gain to common-mode signal gain is known as the common-mode rejection ratio (CMRR). CMRR is typically measured in decibels (dB).

Crosstalk

The degree of isolation between each analog input and other analog inputs is measured by crosstalk. Crosstalk in ADCs with multiple input channels refers to the total amount of signals coupled from one analog input signal to another analog input, and this value is usually expressed in decibels (dB); crosstalk in DACs with multiple output channels refers to the total amount of signals coupled from one analog output signal to another analog input, and this value is usually expressed in decibels (dB). When the output of one DAC is updated, the total quantity of noise generated at the output of another DAC is referred to.

Differential nonlinearity error

The difference between the analog input levels that trigger any two successive output codes should be 1 LSB (DNL = 0) for ADC, and DNL represents the deviation of the actual level difference from 1 LSB.

The DNL error of a continuous DAC encoding is the difference between the ideal and measured output response. The ideal DAC response's analog output value should be one code (LSB) apart (DNL = 0).

(To achieve monotonicity, the DNL index must be greater than or equal to 1LSB.)

Digital feedthrough

The noise generated at the DAC's output when the DAC's digital control signal changes is referred to as digital feedthrough. The feedthrough at the DAC's output is the result of serial clock signal noise, as seen in the diagram below.

Dynamic Range

The dynamic range of a device is defined as the distance between its noise floor and its specified maximum output level, commonly represented in decibels (dB). The signal amplitude range that the ADC can differentiate is defined by its dynamic range; for example, if the dynamic range of the ADC is 60 dB, the signal amplitude range that can be discriminated is x to 1000x.

The dynamic range is critical in communication applications since the range of signal strength variations is so wide. The ADC input will over-range if the signal is too large; if the signal is too little, it will be submerged in the converter's quantization noise.

Effective digits

ENOB is a measure of an ADC's dynamic performance at a given input frequency and sampling rate. Only quantization noise is present in the error of a perfect ADC. Because the overall noise (particularly the distortion component) increases as the input frequency increases, ENOB and SINAD are reduced (see "Signal to Noise + Distortion Ratio (SINAD)").

Load-sensing output

A measurement method in which a voltage (or current) is applied to a circuit's remote point, and the current (or voltage) produced is subsequently measured (detected). Load-sensing outputs are occasionally included in DACs with integrated output amplifiers. The feedback circuit must create a closed loop externally, and the output amplifier can give an inverted input for external connection.

Full power bandwidth

During ADC operation, the analog input signal is equal to or close to the converter's preset full-scale voltage. Then, to reduce the amplitude of the digital conversion result by 3dB, increase the input frequency to a specified frequency. The complete power bandwidth is represented by this input frequency.

Full-scale error

The difference between the actual value of the trigger leap to the full-scale code and the desired analog full-scale jump value is the full-scale error. As seen in the diagram below, the full-scale error equals "offset error + gain error."

FS gain error

The difference between the actual and ideal output span is the full-scale gain error of a digital-to-analog converter (DAC). The difference between the output when the input is set to all 1s and when the input is set to all 0s is the real span. The reference used to measure gain error affects the full-scale gain error of all data converters.

Gain error

The gain error of an ADC or DAC reflects how closely the actual transfer function slope matches the desired transfer function slope. Gain error is commonly represented in LSBs or as a percentage of full-scale range (percent FSR), and it can be reduced or eliminated through hardware or software calibration. The gain error is the offset error minus the full-scale error.

Gain error drift

The change in gain error induced by ambient temperature, commonly reported in ppm/°C, is referred to as gain error drift.

Gain consistency

The gain consistency of a multi-channel ADC reflects how well the gains of all channels are aligned. Apply the same input signal to all channels to determine gain consistency, and then record the highest gain deviation, which is commonly reported in dB.

Spike

The voltage transient oscillation created at the DAC output when the MSB jumps is known as a spike pulse and is commonly expressed in nVs, which is equal to the area under the voltage-time curve.

harmonic

Periodic signals' harmonics are sine components that are integer multiples of the signal's fundamental frequency.

Integral nonlinearity error

Integral nonlinearity (INL) is the deviation of the actual transfer function from the transfer function's straight line in data converters. The straight line is the best-fit straight line or the straight line between the transfer function's end points after the offset and gain errors have been eliminated. "Relative accuracy" is a term used to describe INL.

Intermodulation distortion

Due to the nonlinearity of the circuit or device, IMD refers to the phenomena of new frequency components that do not exist in the original signal. Harmonic and two-tone distortion are included in IMD. Use the ratio of the total power of the specified intermodulation products (ie IM2 to IM5) to the total power of the two input signals when measuring (f1 and f2).

Least significant bit

The LSB is the lowest weighted bit in binary integers. The rightmost bit is usually LSB. The weight of LSB in an ADC or DAC is equal to the converter's full-scale voltage range divided by 2N, where N is the converter's resolution. If the full-scale voltage of a 12-bit ADC is 2.5V, 1LSB = (2.5V/212) = 610V.

MSB transition

When MSB jumps (middle scale point), MSB goes from low to high, and all other data bits go from high to low; or MSB goes from high to low, and all other data bits go from high to low. From low to high, the data bit changes.

It is an MSB jump, for example, if 01111111 becomes 10000000. MSB transitions are notorious for generating the most severe switching noise (see spikes).

Most significant bit

In binary numbers, MSB is the highest weighted bit. Usually, the MSB is the leftmost bit.

Multiplying DAC

An AC signal can be applied to the reference input using the multiplying DAC. By connecting the signal of interest to the reference input and utilizing the DAC to encode and scale the signal, the DAC can be utilized as a digital attenuator.

Lossless coding

If the ADC creates all feasible digital codes when the ramp signal is applied to the analog input terminal, the ADC has no lost codes.

Nyquist frequency

According to the Nyquist theorem, the ADC's sampling rate must be at least twice the signal's maximum bandwidth in order to fully restore the analog signal without distortion. The Nyquist frequency is the maximum bandwidth.

Offset binary encoding

Offset binary is a popular coding method for bipolar signals. In offset binary coding, all 0s (00...000) represent the negative maximum value (negative full amplitude), and all 1s (11...111) represent the positive maximum value (positive full amplitude)Express.

A 1 (MSB) precedes all 0s to denote the zero width (10...000). Standard binary, which is commonly used for unipolar signals, is similar to this approach (see binary coding, unipolar).

Offset error (bipolar)

The offset error of a bipolar converter is measured in the same way that the offset error of a unipolar converter is measured, except that the error at zero amplitude is recorded at the midpoint of the bipolar transfer function (see Offset Error Unipolar)

Offset error (unipolar)

The discrepancy between the actual transfer function and the ideal transfer function at a given operating point is referred to as offset error or "zero amplitude" error. The first transition happens at 0.5LSB above zero for an ideal data converter. Apply a zero-amplitude voltage to the analog input terminal of an ADC and gradually boost it until the first transition occurs; the offset error of a DAC is the analog output when the input code is all 0s.

Offset error drift

The change in offset error induced by ambient temperature, commonly stated in ppm/°C, is known as offset error drift.

Oversampling

Oversampling occurs when the frequency of sampling the analog input is substantially higher than the Nyquist frequency in an ADC. Oversampling effectively lowers the noise floor, increasing the ADC's dynamic range. The resolution improves as the dynamic range is increased. The basis of - ADCs is oversampling.

Phase matching

The phase matching of the identical signals applied to all channels of the multi-channel ADC reflects the degree of phase matching. The maximum phase shift in all channels, usually given in degrees, is referred to as phase matching.

Power supply rejection ratio

The PSRR (Power Supply Rejection Ratio) is the dB ratio of power supply voltage variation to full-scale error change.

Quantization error

The discrepancy between the real analog input and the digital code that represents that value is characterized as quantization error in ADCs (see "Quantization").

Ratio measurement

The voltage applied to the ADC voltage reference input is proportional to the signal applied to the transmitter, rather than being a constant value (ie, load cell or bridge). Proportional measurement is a method of measurement that eliminates any mistakes caused by variations in the reference voltage. Ratio measuring is demonstrated in the diagram below using a resistance bridge.

Resolution

The number of bits used to describe the analog input signal is referred to as ADC resolution. The resolution must be enhanced in order to duplicate the analog signal more accurately. Quantization errors are also reduced when a higher resolution ADC is used. The resolution of the DAC is similar: the higher the DAC's resolution, the smaller the step produced at the analog output as the encoding is raised.

Effective value

The effective DC value or comparable DC signal of the signal is the RMS value of the AC waveform. When determining the RMS value of an AC waveform, square and time average it first, then calculate the square root of it. The RMS value of a sine wave is 2/2 (or 0.707) times the peak, which is 0.354 times the peak-to-peak value.

Sampling rate/frequency

The sampling rate, also known as sampling frequency, is measured in "samples per second" (sps), which refers to how quickly the ADC collects (samples) the analog input. The sampling rate also refers to the throughput rate for ADCs that conduct one sampling per conversion (such as SAR, Flash ADC, or pipelined ADC). The sample rate of - ADCs is typically substantially higher than the data output frequency.

Establishment time

The settling time for a DAC is the time period between when it receives the command to update (alter) its output value and when it reaches the final value (within a specified percentage). The output amplifier's slew rate, as well as the total amount of amplifier ringing and signal overshoot, influence the settling time. It's critical for ADCs that the time it takes for the sampling capacitor voltage to stabilize to 1 LSB is shorter than the converter's capture time.

Beloved

SINAD is the ratio of the RMS value of the sine wave (the ADC's input or the DAC's output) to the RMS value of the noise + distortion of the converter (no sine wave). Except for the fundamental wave and DC offset, RMS noise plus distortion comprises all spectral components below the Nyquist frequency. SINAD is commonly represented in decibels (dB).

Signal-to-noise ratio

At any given time, the signal-to-noise ratio (SNR) is the ratio of the useable signal amplitude to the noise amplitude. The higher the monetary value, the better. The theoretical maximum SNR for a waveform perfectly reconstructed from digital sampling is the ratio of the full-scale analog input (RMS value) to the RMS quantization error (residual error).

The theoretical minimal ADC noise, in theory, solely incorporates quantization error and is defined directly by the ADC's resolution (N bits). (ADCs produce thermal noise, reference noise, clock jitter, and other types of noise in addition to quantization noise.)

Signed binary code

The MSB represents the binary number's sign (positive or negative) in the signed binary encoding method. As a result, -2 has an 8-bit representation of 10000010, while +2 has an 8-bit value of 00000010.

Slew rate

The maximum rate at which the DAC output changes, or the maximum rate at which the input changes without causing an error in the ADC's digital output, is known as the slew rate. The required slew rate for DACs with output amplifiers is usually the amplifier's slew rate.

Small signal bandwidth

An analog input signal with a sufficiently tiny amplitude is applied to the ADC to measure the small signal bandwidth, and the ADC's slew rate does not limit the ADC's performance. The input frequency is then scanned until the digital conversion result's amplitude is lowered by 3dB. The performance of the correlated sample-and-hold amplifier is frequently limited by the short signal bandwidth.

Spurious free dynamic range

The ratio of the RMS amplitude of the fundamental wave (the biggest signal component) to the RMS value of the second largest spurious component is known as spurious-free dynamic range (SFDR) (excluding DC offset). The signal-to-noise ratio (SFDR) is measured in decibels (dBc) in relation to the carrier.

Total harmonic distortion

THD is a measurement of the signal's distortion component, expressed in decibels (dB) in relation to the fundamental wave. The ratio of the RMS sum of the selected input signal harmonics to the fundamental is the total harmonic distortion (THD) for ADCs. Only harmonics within the Nyquist limit are included in the measurement.

Track-and-hold

The ADC's input sampling circuit is referred to as "sample-and-hold" in track-and-hold. Analog switches and capacitors are the most basic version of track-and-hold input (see figure). The circuit is in "tracking" mode when the switch is closed; while the switch is open, the sampling capacitor keeps the input's last transient value, and the circuit is in "hold" mode.

Transition noise

The range of input voltage change that causes the ADC output to fluctuate between adjacent output codes is referred to as conversion noise. Because of the associated transient noise, the voltage that triggers each code jump (code edge) is unknown when the analog input voltage is increased.

Two's complement encoding

The twos complement coding method is used for positive and negative coding to make addition and subtraction calculations easier. The 8-bit representation of -2 in this encoding method is 11111110, and the representation of +2 is 00000010.

Under-sampling

The ADC sampling rate is lower than the analog input frequency in the under-sampling technique, resulting in aliasing. Under-sampling will result in the loss of signal information, according to Nyquist's theorem.

The aliasing components that contain signal information can be transferred from a higher frequency to a lower frequency and subsequently converted if the input signal is appropriately filtered and the analog input and sampling frequency are suitably adjusted.

The ADC is effectively used as a downconverter in this manner, converting the larger bandwidth signal to the ADC's effective bandwidth. The bandwidth of the ADC track-and-hold circuit must be adequate to accommodate the highest frequency signal expected for this technology to succeed.

Unipolar

The unipolar signal input range for single-ended analog input ADCs is zero amplitude (typically ground) to full amplitude (usually reference voltage); the signal input range for differential input ADCs is zero amplitude to full amplitude, which is equivalent to the positive input range. At the negative input, the input range is measured.