Types of Analog-to-Digital Converters
Analog-to-Digital Converters (ADCs) change analog voltage to a binary number -- this is a series of 1’s and 0’s -- and then ultimately to a base-10 digital number which may be displayed on a meter, laptop/PC screen, or chart. The number of binary digits (aka bits) that represents the digital number determines the ADC resolution. But, the digital number is only an approximation of the true value of the analog voltage at any instant of time because the voltage can only be represented (digitally) in discrete steps. How precisely the digital number approximates the analog value also depends on the ADC resolution.
A mathematical relationship notes how the number of bits an ADC handles determines its specific theoretical resolution: An n-bit ADC has a resolution of one part in 2n. For instance, a 12-bit ADC has a resolution of one part in 4,096, where 212 = 4,096. Therefore, a 12-bit ADC with a maximum input of 10 Vdc can resolve the measurement into 10 Vdc/4096 = 0.00244 Vdc = 2.44 mV. Similarly, for the same 0 to 10-Vdc range, a 16-bit ADC resolution is 10/216 = 10/65,536 = 0.153 mV. The resolution is normally specified with respect to the full-range reading of the ADC, not with respect to the measured value at any particular instant.
Successive-Approximation ADC
Above: Successive-Approximation
ADC.
Interestingly, this ADC uses a digital-to- analog converter and a comparator. The logic sets the DAC to zero and starts counting
up, setting each following bit until it reaches the value of the measured
input voltage. The conversion is then fin ished and the final number
is stored in the register.
Voltage-to-Frequency ADCs
:
Above: Voltage-to-frequency converters
reject noise well and frequently are used for measuring slow signals
or those in noisy environments.
Dual-Slope ADC Integration and Discharge Time
Above. Dual-slope integrating ADCs provide high-resolution
measureinent with excellent noise rejection. They integrate upward from
an unknown voltage and then integrate downward with a known source voltage.
They are more accurate than single slope ADCs because component errors
are washed out during the dc-integration period.
Sigma-Delta ADC
Above: Integrating converters such as the sigma delta ADC have
both high resolution and exceptional noise rejection. They work
particularly well for low-bandwidth measurements and reject high-frequency
noise as well as 50/60 Hz interference.
Sigma-Delta ADC With Digital Filter and Decimator Stage
Above: Sigma-delta ADCs are well suited to high-resolution acquisition
because they use over- sampling and often combine an analog modulator,
a digital filter, and a decimator stage. The low-pass digital filter
converts the analog modulator output to a digital signal for processing
by the decimator.
Above: Table of ADC attributes. * sps = samples per second;
** With line cycle rejection.
Accuracy and Resolution
Common ADC Errors (such as Quantization Errors)
Above: The straight line in each graph represents the analog input
voltage and the perfect output voltage read ing from an ADC with infinite
resolution. The step function in Graph A shows the ideal response for
a 3-bit ADC. Graphs B, C, D, and E show the effect on ADC output from
the various identified errors.
Above: The histogram illustrates how 12-bit ADC samples in a set
were distributed among the various codes for a 2.5-V measurement in a
FSR (full-scale range) of 10- V. Most codes intended for the 1024 bin
representing 2.5 V actually ended up there, but others fell under a gaussian
distribution due to white noise content. Histogram shape
approximates a Gaussian
distribution caused by
white noise mixed
with dc input voltage. Bin with most samples
represents the input signal
of 2.50 Vdc. Bins with samples
caused by noise.
