Data Acquisition

Dr. Abhishek Kumar Ghosh

RME 1201, RMEDU
Digital signals
• Where a microprocessor is used as part of the measurement or control system:
• the analogue output from the sensors has to be converted into a digital form before it can
be used as an input to the microprocessor.
• Likewise, the digital output from a microprocessor has to be converted into an analogue
form before it can be used as input to an actuator.
• The binary system is based on just the two symbols or states 0 and 1, these possibly being 0 V and
5 V signals. These are termed binary digits or bits.

• In a binary number, the bit 0 is termed the least significant bit (LSB) and the highest bit the most
significant bit (MSB).
• The combination of bits to represent a number is termed a word. Thus 1111 is a 4-bit word.

Abhishek K Ghosh, RME 1201, RMEDU 2

Analog to Digital Conversion

Sample and hold circuit

The sample and hold unit is

Three main stages are required for analog to digital conversion: necessary because the ADC
1. Sampling: requires a finite amount of time,
termed the conversion time, to
The process of converting continuous time signal into discrete convert the analogue signal into
values is called sampling. a digital one.
2. Quantization:
The process of converting continuous sample values into discrete
values is called quantization.
3. Coding:
It is the assignment of a unique digital code (usually a binary
number) to each output state.

Abhishek K Ghosh, RME 1201, RMEDU 3

Analog to Digital Conversion
1. Sampling
• A clock supplies regular time signal pulses to the analogue-to-digital converter (ADC)
and every time it receives a pulse it samples the analogue signal.
• The result of the sampling is a series of narrow pulses.

Abhishek K Ghosh, RME 1201, RMEDU 4

Analog to Digital Conversion
1. Sampling
• How often should an analogue signal be sampled to obtain an accurate representation?
• It is only when the sampling rate is at least twice that of the highest frequency in the
analogue signal that the sample gives the original form of signal.
• This criterion is known as the Nyquist criterion or Shannon’s sampling theorem.
• When the sampling rate is less than twice the highest frequency, the reconstruction can
represent some other analogue signal and we obtain a false image of the real signal. This is
termed aliasing.

Abhishek K Ghosh, RME 1201, RMEDU 5

Analog to Digital Conversion
2. Quantization
• The process of converting continuous sample values into discrete values is
called quantization.
• With 3 bits, there are 23 = 8 possible output levels, which are termed
quantization levels.
• The difference in analogue voltage between two adjacent levels is termed
the quantization interval.
• Because of the step-like nature of the relationship, the digital output is
not always proportional to the analogue input and thus there will be an
error termed the quantization error.
• The difference between an instantaneous analogue sample and its closest
decision point is known as quantization error:

• VH = Max voltage value, VL = Min voltage value, n = number of bits in ADC.

Abhishek K Ghosh, RME 1201, RMEDU 6

Analog to Digital Conversion
2. Quantization
• Resolution
• The term resolution in this context refers to the smallest change
in input which will result in a change in the digital output.
• Resolution can be described in several different ways, which
can be used interchangeably.

Converter
range
• For the case shown in the figure (4-bit ADC), resolution can be
simply described as the number of bits in the converter – in this
example 4 (i.e. “4-bit resolution”).
• Additionally, resolution can be described as the ratio between
the range and the number of output states available in the
converter.
• Thus, if we have an ADC with a word length of 4 bits and the
converter range is 15 V, then the number of levels with a 4-bit
word is 24 = 16, and thus the resolution is 15/16 = 0.93 V.

Abhishek K Ghosh, RME 1201, RMEDU 7

Analog to Digital Conversion
2. Quantization
• Resolution
• Consider a thermocouple giving an output of 0.5 mV/°C. What will be the word length required
when its output passes through an ADC if temperatures from 0 to 200°C are to be measured
with a resolution of 0.5°C?

The full-scale output from the sensor is 200*0.5 = 100 mV.

With a word length n, this voltage will be divided into 100/2n mV steps.
For a resolution of 0.5°C, we must be able to detect a signal from the sensor of
0.5 * 0.5 = 0.25 mV.
Thus we require
0.25 = 100/2n
⇒ n = 8.6.
Thus a 9-bit word length is required.

Abhishek K Ghosh, RME 1201, RMEDU 8

Analog to Digital Conversion
2. Quantization
• Input range vs. Converter range

Converter
range
Abhishek K Ghosh, RME 1201, RMEDU 9
Analog to Digital Conversion
2. Quantization
• Example problem: A 12-bit ADC has a range of 0 - 10V. Determine its resolution (in volts),
and hence the effective dynamic range, in dB, if the ADC is used to record a signal that
varies only from 0.2 to 0.5 V

Abhishek K Ghosh, RME 1201, RMEDU 10

Analog to Digital Conversion
3. Encoding or Coding
• It is the assignment of a unique digital code (usually
a binary number) to each output state.
• During the “coding” process, each discretised
sample (after quantization) is assigned a unique
code. For instance, in the Figure, a 3-bit device is
used to convert an analogue voltage signal, then a
binary 3-bit code is used to label each output state.
• In our example, eight quantizing levels are used.
These levels can be coded using 3 bits if the binary
system is used, so each sample is represented by 3
bits.
• The analog signal is represented digitally by the
following series of binary numbers: 001, 011, 100,
100, 010, 001, 011, 110, and 111.

Abhishek K Ghosh, RME 1201, RMEDU 11

Digital to analog conversion
• It is the process to convert digital values to analog
signals
• Consider the situation where a microprocessor gives
an output of an 8-bit word. This is fed through an 8-
bit DAC to a control valve.
• The control valve requires 6.0 V to be fully open.
• If the fully open state is indicated by 11111111 what
will be the output to the valve for a change of 1 bit?
• The full-scale output voltage of 6.0 V will be divided
into 28 intervals.
• A change of 1 bit is thus a change in the output
voltage of 6.0/28 = 0.023 V.

Abhishek K Ghosh, RME 1201, RMEDU 12

Digital to analog converter (DAC)
• Weighted-resistor DAC
• A simple form of DAC uses a summing amplifier to form the weighted
sum of all the non-zero bits in the input word.
• The reference voltage is connected to the resistors by means of
electronic switches which respond to binary 1.
• The values of the input resistances depend on which bit in the word a
switch is responding to, the value of the resistor for successive bits
from the LSB being halved.
• Hence the sum of the voltages is a weighted sum of the digits in the
word.
• Such a system is referred to as a weighted-resistor network.
• The function of the op-amp circuit is to act as a buffer to ensure that
the current out of the resistor network is not affected by the output
load and also so that the gain can be adjusted to give an output range
of voltages appropriate to a particular application.
• A problem with the weighted-resistor network is that accurate Weighted-resistor DAC
resistances have to be used for each of the resistors and it is difficult
to obtain such resistors over the wide range needed. As a result, this
form of DAC tends to be limited to 4-bit conversions.
Abhishek K Ghosh, RME 1201, RMEDU 13
Digital to analog converter (DAC)
• Weighted-resistor DAC

Inverting amplifier Summing amplifier

Abhishek K Ghosh, RME 1201, RMEDU 14

Digital to analog converter (DAC)
• Weighted-resistor DAC

4-bit Binary Weighted Digital-to-Analogue Converter

Abhishek K Ghosh, RME 1201, RMEDU 15

Digital to analog converter (DAC)
• Weighted-resistor DAC

Abhishek K Ghosh, RME 1201, RMEDU 16

Analog-to-Digital Converters (ADC)
• A/D converters are designed based on a number
of different principles: successive approximation,
flash or parallel encoding, single-slope and dual-
slope integration etc.
• The successive approximation A/D converter is
very widely used because it is relatively fast and
cheap.

17
Analog-to-Digital Converters (ADC)
Successive approximation ADC
• When the start signal is applied, the sample
and hold (S&H) amplifier latches the analog
input.
• Then the control unit begins an iterative
process, where the digital value is
approximated, converted to an analog value
with the D/A converter, and compared to the
analog input with the comparator.
• When the D/A output equals the analog input,
the end signal is set by the control unit, and
the correct digital output is available at the
output.
18
Analog-to-Digital Converters (ADC)
Successive approximation ADC
• A 4-bit successive approximation procedure is illustrated
graphically in Figure 8.13.
• The MSB is 1/2 FS, which in this case is greater than the
signal value; therefore, the bit is turned off.
• The second bit is 1/4 FS and is less than the signal, so it is
left on.
• The third bit gives 1/4 + 1/8 of FS, which is still less than the
analog signal, so the third bit is left on.
• The fourth provides 1/4 + 1/8 + 1/16 of FS and is greater
than the signal, so the fourth bit is turned off and the
conversion is complete.
• The digital result is 0110.
19
Multiplexers
• A multiplexer is a circuit that can have inputs
of data from several sources and then, by
selecting an input channel, give an output
from just one of them.
• In applications where there is a need for
measurements to be made at a number of
different locations, rather than using a
separate ADC and microprocessor for each
measurement, a multiplexer can be used to
select each input in turn and switch it through
a single ADC and microprocessor.
• The multiplexer is essentially an electronic
switching device that enables each of the
inputs to be sampled in turn.

Abhishek K Ghosh, RME 1201, RMEDU 20

THE END

21