Introduction to AD and DA

 What are AD/DA Converters

 What are they used for
 Why do you need to know how they work

 Digital coding methods

 Waveform digitising
 What are AD/DA
Converters (1)

 An Analog to Digital converter [AD or ADC]

is an electronic circuit which accepts an
analog input signal (usually a voltage) and
produces a corresponding digital number at
the output
 An Digital to Analog converter [DA or DAC]
is an electronic circuit which accepts a
digital number at its input and produces a
corresponding analog signal (usually a
voltage) at the output
 They exist as modules, ICs, or fully
integrated inside other parts, e.g. µCs
 What are AD/DA Converters (2)

Analog Digital Analog

continuous time world discrete time world continuous time world

+/-10v
The Real World

The Real World

+/-10v
ADC 1 DAC 1
12 16
COMPUTER
or µP/µC
+/-5v
ADC 2
16

Typical AD & DA Application

 What are they used for

 Any time a real world analog signal is

connected to a digital system
 CD players, GSMs, DVMs, Digital Camcorders
etc, etc
 CERN control systems & instruments
 HOWEVER, each application has particular
needs
 Resolution - number of bits
 Speed and Accuracy
 Level of input/output waveforms
 Cost etc
 Why do you need to know
how they work
 Because the theoretical course you will
shortly undertake assumes perfect
converter products - BUT
 Practical converters have :
 Many conversion methods - why
 Trade-offs between resolution and speeds +
delays
 Different methods of “sampling” the
waveforms
 A large number of basic and method-dependent
error sources
 Manufacturers specifications which ‘differ’ -
AND
 Almost all converters need some analog ‘signal
conditioning’ which is application dependent
 Digital coding methods (1)
AD/DA Transfer Characteristic
 8,10,12,14,16,18, 20-24bits?
 Most/Least significant bit +10v
MSB/LSB
 Uni-polar, bipolar, straight
binary, 2’s complement -
invert MSB
 Parallel I/O or serial [delay]
0v
 Bytes or words
 Double buffering
 Digital ‘breakthrough’
 Digital correction methods
 Time skewing & jitter
-10v
0000 8000 FFFF
0000 7FFF
8000 FFFF
 Digital coding methods (2)

 Resolution = 2n-1 [n = number of bits]

n 2n 1bit ppm [1x10-6]

 8bits 256 3906
 10bits 1024 976
 12bits 4096 244
 14bits 16384 61
 16bits 65536 15
 18bits 262144 3.8
 20bits 1,048576 0.95
 22bits 4,194304 0.24
 24bits 16,777216 0.06
Digital value
Waveform digitising (1)

time

 A waveform is ‘digitised’ (sampled) at a constant

rate D t
 Each such sample represents the instantaneous
amplitude at the instant of sampling
 Between samples the value remains constant [zero
order hold]
 What errors can occur in this process ?
 Waveform digitising (2)

A C

D
B

 A & B show aliasing in the time domain

 C & D show a different case in the frequency
domain
- it is important to understand these effects
 Waveform digitising errors

 For a DAC
 output waveform is a ‘distorted’ version of original
 higher frequencies not reproduced - aliasing ?
 ‘average shape’ displaced in time
 ‘sharp’ edges need filtering
 For an ADC
 converter sampling errors
 with a ‘sample & hold’ circuit ahead of the converter?
 integrating action during part, or all of the sample-time
?
 conversion time
 data ‘available’ delay
 aliasing - [ is multiplication of input spectrum and fs]
…[must ‘remove’ all spectrum > fs/2 before
sampling]
 Sampling rate

 Nyquist rate = 2x highest frequency of

interest
 Practically, - always sample at least 5x, or
higher
 Ensure ADCs have input filtering [anti-alias]
where necessary [large hf signals]
 Filter DAC outputs to remove higher
frequencies and switching ‘glitches’
 ‘Over-sampling’ converters sample x4 to
x500 - this may reduce above problems
and/or extend resolution
Understanding Conversion
Methods
 AD/DA Methods

 Some very simple ideas

 DAC circuits
 Basic ADC circuits
 Successive approximation, flash - S&H
 Integrating - single/dual/multi slope
 Charge balance, 
 Some very simple ideas

‘Digitally set’
Comparator
potentiometer

DAC ADC
Vref
dial Unknown
voltage
Vdac equal
=
 ADC =
 precise reference voltage
 comparison of divider value with unknown [analog input]
 “digitally adjustable” divider or potentiometer [output
value]
 DAC =
 precise reference voltage ……. {multiplying dac}
 “digitally adjustable” divider or potentiometer [input
value]
 optional output amplifier of pot. value [analog output]
 DAC circuits (1)

Simplified binary weighted resistor DAC R - 2R ladder DAC

8.75V

9.375
max.

 Summation of binary weighted currents

 Modern DACs use the ‘R-2R ladder’
 DAC circuits (2)
 Important circuit concepts
 Resistor tracking - temp. & time > ratios
 Switch is part of R [on & off resistance]
 Limits for tracking and adjustment
 Switch transition times - glitches
 Switched current sources are faster
 Other DAC methods
 DC performance not needed for all uses
 Different ladders, Caps. as well as Resistors
 PWM, F>V
 Sigma-Delta
 Performance cannot be better than the Reference
- {multiplying DAC concept}
 ADC methods

 Ramp-compare ADC
 Flash ADC (direct-conversion)
 Successive approximation
 Integrating ADC
 Delta-encoded
 Pipelined
 Sigma-delta
 …
Ramp-compare ADC
 Basic ADC circuits (1)

Simple ramp-compare ADC

Unknown
analog
input

start Binary output

 Digitising begins with a ‘start’ pulse

 DAC is ramped up from zero
 counter stopped by comparator when Vin = DAC out
 ADC output is counter value
 Tracking ADC
 Basic ADC circuits (2)

 This ADC circuit is limited and rarely used

WHY -
 slow
 variable time to give result
 input signal can vary during digitising

 Successive Approximation ADC solves these

problems - using
 complex logic to test and retain each DAC bit
 a sample and hold circuit ahead of the
comparator
Successive Approximation ADC
 Successive Approximation ADC

 Fast process - 1 -
100µsecs

 Result always n clocks

after start

 Used extensively for

12-16bit DAQ systems
Flash ADC
 Vref
Flash
Flash ADC

Vref
Half-Flash
analog
input

analog
 The fastest process <50nsecs input
 Limited resolution typically 8 -
10bits
 Half-flash technique is cheaper