498 Chapter 8. Signal Processing
8.7 AnalogtoDigitalConversion
Converting an analog pulse into digital form is very convenient in terms of analyzing
and storing the relevant information contained in the signal. The so calledanalog
to digital converters(ADC) are extensively used in detection systems to convert the
information contained in the analog pulse into an equivalent digital form. Before
we go on to discuss different types of ADCs, let us have a look at some factors that
need to be considered while deciding on the method of conversion.
8.7.A A/D-ConversionRelatedParameters..............
A.1 ConversionTime........................
This is the time it takes the ADC to perform one complete conversion. It depends
not only on the A/D conversion method but also on the pulse height.
A.2 DeadTime
The total time it takes the ADC to acquire a signal, complete the conversion, and
become available for the next acquisition is called dead time because during this
time it can not accept a new signal. The dead time typically consists of:
signal acquisition time,
conversion time,
data transfer to buffers, and
reset time.In well designed systems, the transfer of converted signal to the memory is per-
formed during the reset time. How that memory is subsequently read out and
handled is not part of ADC performance. However since such a memory is generally
of limited capacity (such as 1kB FIFOs) , it must be read out continuously to avoid
overflow, which leads to loss of information.
Additionally, the operations on ADCs are controlled through digital signals,
which have their own response times and uncertainties. Such uncertainties are re-
ferred to astime jittersand should be given proper consideration in high resolution
systems.
If the event rate is such that the pulses arrive during dead time of ADC, the
information gets lost unless the charge is dynamically integrated on some capacitor
for later acquisition by the ADC. Another possibility is to determine the average
dead time of the system and then correct for it in the final analysis. This strat-
egy works well for random signals where all events experience the same dead time.
Determination of the dead time is a straight forward process in which the ADC is
fed with a known stream of pulses and the output is recorded. The comparison of
output to the input gives quantitative measure of the dead time.
A.3 Resolution............................
The ADC output is in the form of the so calleddigital word, which is simply a
number. This number should be directly proportional to the analog input at the